Microbial Ecology in Growing Animals

Microbial Ecologyin Growing Animals

Edited by

W.H. HolzapfelInstitute of Hygiene and Toxicology BFE, Karlsruhe, Germany

P.J. NaughtonNorthern Ireland Centre for Food and Health, School of Biomedical Sciences,

University of Ulster, Coleraine, Co. Londonderry, United Kingdom

in Series

Biology of Growing AnimalsSeries Editors

S.G. PierzynowskiDepartment of Cell and Organism Biology, Lund University, Lund, Sweden

R. ZabielskiDepartment of Physiological Sciences, Warsaw Agricultural University

Warsaw, PolandThe Kielanowski Institute of Animal Physiology and Nutrition PAS

Jablonna n/Warsaw, Poland

Technical Editor

E. SalekThe Kielanowski Institute of Animal Physiology and Nutrition PAS

Jablonna n/Warsaw, Poland

Edinburgh • London • New York • Oxford • Philadelphia • St Louis • Sydney • Toronto

2005

Elsevier Limited© 2005, Elsevier Limited. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior per-mission of the publishers or a licence permitting restricted copying in the United Kingdom issued by theCopyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Permissions may be soughtdirectly from Elsevier’s Health Sciences Rights Department in Philadelphia, USA: phone: (+1) 215 2387869, fax: (+1) 215 238 2239, e-mail: [email protected]. You may also complete yourrequest on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’and then ‘Obtaining Permissions’.

First published 2005

ISBN 0 444 509 267

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data

A catalog record for this book is available from the Library of Congress

Notice

Veterinary knowledge and best practice in this field are constantly changing. As new research and experi-ence broaden our knowledge, changes in practice, treatment and drug therapy may become necessary orappropriate. Readers are advised to check the most current information provided (i) on proceduresfeatured or (ii) by the manufacturer of each product to be administered, to verify the recommended doseor formula, the method and duration of administration, and contraindications. It is the responsibility ofthe practitioner, relying on their own experience and knowledge of the patient, to make diagnoses, todetermine dosages and the best treatment for each individual patient, and to take all appropriate safetyprecautions. To the fullest extent of the law, neither the publisher nor the editors assumes any liability forany injury and/or damage.

The Publisher

The publisher’s

policy is to usepaper manufactured

from sustainable forests

Printed in China

vi

Keynotes I

Genomic and proteomix projects have a profound impact on biological sciences andthe biotech-tech world. Nonetheless, we should be aware that probably nothing “new”has happened in biology since Watson and Crick’s discovery—prions are somethingnew, but they are controversial. We are thus, in essence, constantly exploiting the dis-covery of the structure of DNA, but with unbelievable speed and effectiveness. Itappears that the development of biological sciences coupled with genomic/proteomixfields is occuring at a disproportionately fast rate in comparison with other biologicaldisciplines, provided that these disciplines are still in existence. In addition to strictscientific problems, genomic/proteomix biology raises questions of an existential andphilosophical origin. Does nature produce more genes and more protein than needed,or does it only take the chance to produce new proteins when it’s necessary to producethem? Do we produce proteins before needing them? Or, more theologically, is biol-ogy predestined to produce exclusively programmed proteins and nothing more, andthe genome is only waiting for the right signals? Or, maybe more cynically, we simplydo not know what these proteins are for? Definitely, they are for something and weneed to explore it. This “for something” leads us to another question. Do the particularmolecules/atoms taking part in these life mysteries get a different value?

Stefan G. Pierzynowski, prof.

vii

Keynotes II

The volume “Microbial Ecology in Growing Animals” is mostly, but not solely,about the ecology of microorganisms living in the gastrointestinal tract of younggrowing animals. This book is released two years after the corresponding volumeregarding the development of gut function, and entitled “Biology of the Intestine inGrowing Animals”. The Editors of the present volume faced a very ambitious task toprovide the reader with a new and comprehensive knowledge on the biology of thegastrointestinal microorganisms simultaneously with describing a labyrinth of inter-actions between them and the host. Furthermore in the early postnatal life the colo-nization of the gut is just initiated, thus the changes are more complex and dramaticthan in the adults with balanced ecosystem. This of course makes all the story evenmore complicated and difficult to put in plain words. Therefore we would like todeeply thank the Volume Editors, William H. Holzapfel and Patrick J. Naughton,and all Authors for their efforts since in our opinion they made a great job, and supplied the academic society with a valuable and “must have” book.

Series Editors

ix

Preface

MICROBIAL ECOLOGY IN GROWING ANIMALS

This book “Microbial Ecology in Growing Animals”, is the second volume in theElsevier book series entitled Biology of Growing Animals. In individual chapters,recent developments in our knowledge of the role of microorganisms in the gastro-intestinal tract are reflected, whilst new approaches towards improving and stabilisinganimal health are also addressed. The book discusses the interactions between theanimal host and the microbial population associated with the gastrointestinal tract. No one publication can adequately describe the numerous interactions which occur inthe gastrointestinal tract of the growing animal and the myriad differences which existbetween these in different animals. This book attempts to draw together in one volume acollection of work representing different areas of scientific research which, while distinctin their own right, are presented here under the unifying theme of microbial ecology inits relation to and interaction with the animal gastrointestinal tract.

Even though the complexity of the intestinal micro-ecology was recognised longago, investigations have thus far been limited to a few major bacterial groups, considered to be dominating, and to pathogens, both in relation to concomitant finan-cial losses in the production animal, and with regard to the food infection chain.Thanks to recent developments, including improved microbiological detection andsampling techniques, and the application of molecular tools to monitor the presenceof specific strains in the intestine, our knowledge has increased rapidly in recentyears. This book reflects on these developments, and addresses new approachestowards improving and/or stabilising animal health. Special emphasis is also placedon probiotics and the use of selected bacterial strains as vehicles for delivery ofbiologically active compounds to the mucosa. Colonisation, development andsuccession, as well as the normal microbial population of the mucosal surface in thehealthy animal, are addressed. Extensive information is provided on diverse and

dominating bacterial populations of different animal types. Reference is also madeto those microbial groups considered to be of special benefit to the health andimmune protection of the (young) animal. The development and application of modelsof the gastrointestinal tract provide a solid basis for studying gut microbial inter-actions, whilst molecular approaches and the use of molecular tools to monitor thepresence of specific strains in the intestine is treated in a comprehensive manner.

W.H. Holzapfel and P.J. NaughtonVolume Editors

x Preface

xi

Acknowledgements

The editors wish to thank all of the authors for their outstanding contributions to thebook. We also thank Ewa Salek for her assistance with technical editing. Thanksalso go to the Series Editors, Stefan G. Pierzynowski and Romuald Zabielski, for theinvitation and opportunity to put together this book. We sincerely thank the institu-tions providing patronage and financial support, including Federal Research Centrefor Nutrition, Institute of Hygiene and Toxicology BFE (Germany), Northern IrelandCentre for Food and Health, School of Biomedical Sciences, University of Ulster(United Kingdom), Lund University (Sweden), The State Committee for ScientificResearch (KBN) – International Network Project, SPUB-M-MSN (Poland), TheKïelanowski Institute of Animal Physiology and Nutrition, Polish Academy ofSciences (Poland) and SGPlus (Sweden).

Volume Editors

xiii

Contributors

Aschfalk A. – Section of Arctic Veterinary Medicine, Department of Food Safetyand Infection Biology, The Norwegian School of Veterinary Science, NO-9292Tromsø, Norway

Beeckmans S. – Laboratory of Protein Chemistry, Institute of Molecular Biologyand Biotechnology, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels,Belgium

Birkbeck T.H. – Division of Infection and Immunity, Institute of Biomedical and Life Sciences, University of Glasgow, Joseph Black Building, Glasgow G12 8QQ, UK

Blaut M. – German Institute of Human Nutrition, Gastrointestinal Microbiology,Arthur-Scheunert-Allee 114-116, D-14558 Bergholz-Rehbrücke, Germany

Bomba A. – University of Veterinary Medicine, Komenského 73, 041 81 Kossice,Slovak Republic

Ellis A.E. – Marine Laboratory, Victoria Road, Aberdeen AB11 9DB, UKFekete P.Zs. – Veterinary Medical Research Institute of the Hungarian Academy of

Sciences, 1143 Budapest, Hungária krt. 21, HungaryFranz C.M.A.P. – Federal Research Centre for Nutrition, Institute of

Biotechnology and Molecular Biology, D-76131 Karlsruhe, GermanyFuller R. – 59 Ryeish Green, Three Mile Cross, Reading RG7 1ES, UKGancarcíková S. – University of Veterinary Medicine, Komenského 73, 041 81

Kosice, Slovak RepublicGram L. – Danish Institute for Fisheries Research, Department of Seafood

Research, Søltofts Plads, c/o Technical University of Denmark bldg. 221, DK-2800 Kgs. Lyngby, Denmark

Grant G. – Rowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB21 9SB, UK

Havenith C.E.G. – Toegepast Natuurkundig Onderzoek (TNO) Prevention andHealth, Department of Infection and Immunology, Special Programme InfectiousDiseases, Post Box 2215, NL-2301 CE Leiden, The Netherlands

xiv Contributors

Holzapfel W.H. – Institute of Hygiene and Toxicology BFE, D-76131 Karlsruhe,Germany

Katouli M. – Faculty of Science, University of the Sunshine Coast, Maroochydore,Queensland 4558, Australia

Klein G. – Institute for Food Science, School of Veterinary Medicine Hannover,Bischofsholer Damm 15, D-30173 Hannover, Germany

Klucinski W. – Department of Clinical Sciences, Faculty of Veterinary Medicine,Warsaw Agricultural University, Ciszewskiego 8, 02-786 Warsaw, Poland

Kremer S.H.A. – Toegepast Natuurkundig Onderzoek (TNO) Prevention andHealth, Department of Infection and Immunology, Special Programme InfectiousDiseases, Post Box 2215, NL-2301 CE Leiden, The Netherlands

La Ragione R.M. – Department of Bacterial Diseases, Veterinary Laboratories Agency(Weybridge), Woodham Lane, Addlestone, New Haw, Surrey KT15 3NB, UK

Mackie R.I. – Department of Animal Sciences, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, USA

Mathiesen S.D. – Section of Arctic Veterinary Medicine, Department of FoodSafety and Infection Biology, The Norwegian School of Veterinary Science, NO-9292 Tromsø, Norway

Medina M. – Research Centre for Lactobacilli (CERELA), Chacabuco 145,RA-4000 San Miguel de Tucumán, Argentina

Michalowski T. – The Kielanowski Institute of Animal Physiology andNutrition, Polish Academy of Sciences, Instytucka 3, 05-110 Jablonna nearWarsaw, Poland

Minekus M. – TNO Nutrition and Food Research, P.O. Box 360, 3700 AJ, Zeist,The Netherlands

Mudronová D. – University of Veterinary Medicine, Komenského 73, 041 81Kossice, Slovak Republic

Nagy B. – Veterinary Medical Research Institute of the Hungarian Academy ofSciences, 1143 Budapest, Hungária krt. 21, Hungary

Naughton P.J. – Northern Ireland Centre for Food and Health, School ofBiomedical Sciences, University of Ulster, Cromore Road, Coleraine, Co.Londonderry BT52 1SA, UK

Nemcová R. – University of Veterinary Medicine, Komenského 73, 041 81 Kossice,Slovak Republic

Newell D.G. – Department of Bacterial Diseases, Veterinary Laboratories Agency(Weybridge), Woodham Lane, Addlestone, New Haw, Surrey KT15 3NB, UK

Niemialtowski M. – Immunology Laboratory, Division of Virology, Mycology andImmunology, Department of Preclinical Sciences, Warsaw AgriculturalUniversity, Grochowska 272, PL-03-849 Warsaw, Poland

Perdigón G. – Research Centre for Lactobacilli (CERELA), Chacabuco 145,RA-4000 San Miguel de Tucumán, Argentina; Immunology Department Facultyof Biochemistry, National Tucumán University, Argentina

xv Contributors

Pouwels P.H. – Toegepast Natuurkundig Onderzoek (TNO) Prevention and Health,Department of Infection and Immunology, Special Programme InfectiousDiseases, Post Box 2215, NL-2301 CE Leiden, The Netherlands

Ringø E. – Section of Arctic Veterinary Medicine, Department of Food Safety andInfection Biology, The Norwegian School of Veterinary Science, NO-9292Tromsø, Norway

Schillinger U. – Institute of Hygiene and Toxicology BFE, D-76131 Karlsruhe,Germany

Schollenberger A. – Immunology Laboratory, Division of Virology, Mycology andImmunology, Department of Preclinical Sciences, Warsaw AgriculturalUniversity, Grochowska 272, PL-03-849 Warsaw, Poland

Seegers J.F.M.L. – Toegepast Natuurkundig Onderzoek (TNO) Prevention andHealth, Department of Infection and Immunology, Special Programme InfectiousDiseases, Post Box 2215, NL-2301 CE Leiden, The Netherlands

Sundset M.A. – Department of Arctic Biology and Institute of Medical Biology,University of Tromsø, NO-9037 Tromsø, Norway

Schwiertz A. – German Institute of Human Nutrition, GastrointestinalMicrobiology, Arthur-Scheunert-Allee 114-116, D-14558 Bergholz-Rehbrücke,Germany

Tóth I. – Veterinary Medical Research Institute of the Hungarian Academy ofSciences, 1143 Budapest, Hungária krt. 21, Hungary

Van Driessche E. – Laboratory of Protein Chemistry, Institute of MolecularBiology and Biotechnology, Vrije Universiteit Brussel, Pleinlaan 2, B-1050Brussels, Belgium

Wallgren P. – Department of Ruminant and Porcine Diseases, National VeterinaryInstitute, Uppsala S-751 89, Sweden

Woodward M.J. – Department of Bacterial Diseases, Veterinary LaboratoriesAgency (Weybridge), Woodham Lane, Addlestone, New Haw, Surrey KT15 3NB, UK

Zentek J. – Institute of Nutrition, University of Veterinary Medicine, Vienna A-1210, Vienna, Veterinärplatz 1, Austria

3

Gnotobiology is the science of gnotobiotic animals. Such animals have a preciselydefined microflora, and have proved to be very useful models in studying the physi-ology of the digestive tract. They mainly enable observation of the role of micro-organisms in the process of functional and morphological development of thedigestive tract. Experiments on gnotobiotic lambs demonstrate that the functions ofthe rumen, and the stability of the ecosystem, depend on the complexity and diver-sity of the microflora. Gnotobiotic lambs have considerably shorter rumen papillaethan in conventional lambs. In germ-free animals, the normal structure and morphol-ogy of the gut are altered in ways that emphasize the importance of an animal’s inter-action with its indigenous microbial flora in establishing its defences againstmicrobial invasion, while, at the same time, adequate host nutrition is maintained bynormal alimentary tract function. The overall mass of the small intestine in germ-freepiglets is decreased, and its surface area is smaller, whereas the villi of the small intes-tine are unusually uniform in shape and are slender, with crypts that are shorter andless populated than in the respective conventional control animals. Further studies withgnotobiotic animals should clarify the role of the host microecosystem in the physiol-ogy of the alimentary tract, and the pathophysiology of gastrointestinal diseases.

1. INTRODUCTION

Quality nutrition and optimum development of the digestive tract are essential for proper growth, high production and a good state of health of livestock.Underdevelopment of the digestive tract of the young is a predisposing factor fordiseases and disturbances which negatively influence the economic effectivenessof livestock husbandry. Diseases of the gastrointestinal tract can be considered to be

1 Development of the digestive tract of gnotobiotic animals

A. Bomba, R. Nemcová and S. Gancarcíková

University of Veterinary Medicine,Komenského 73, 041 81 Kosice, Slovak Republic

Microbial Ecology in Growing AnimalsW.H. Holzapfel and P.J. Naughton (Eds.)

© 2005 Elsevier Limited. All rights reserved.

the most important health and economic problem when rearing young livestock,since they may cause extremely high losses as a consequence of morbidity, mortal-ity, costs of treatment and weight loss. At an early age, diseases debilitate the animal organism and cause delays in development, which can subsequently becomeevident in further health problems and productivity decrease. For this reason, itis extremely important to ensure the optimum development of the digestive tract of young animals. Recent research provides extensive possibilities to carry out thor-ough studies and to acquire new knowledge on the physiological and functionaldevelopment of the gastrointestinal tract of animals. Management of gnotobiotictechniques and the use of gnotobiotic animals for experimental purposes have substantially influenced the methodologic approach of scientists to the topic.Microflora is of great importance in the development of the digestive tract. The useof gnotobiotic animals in experiments has enabled the study of the role of micro-organisms in the process of morphological and functional development of the diges-tive tract. Gnotobiology has enabled scientists to gain new information that hasenriched the theoretical knowledge of developmental physiology of the digestivetract and at the same time supported targeted manipulation of the development ofthe gastrointestinal tract in young livestock.

2. CONTRIBUTION OF GNOTOBIOTIC TECHNIQUES AND GNOTOBIOTIC ANIMALS TO RESEARCH INTO THEPHYSIOLOGY AND PATHOPHYSIOLOGY OF ANIMALS

Gnotobiology is the science of gnotobiotic animals. Such animals possess a preciselydefined microflora (Duskin et al., 1983; Coates and Gustafsson, 1984). As a science,gnotobiology emerged from the need to study the role of microflora in the livingprocesses of macroorganisms. After scientists revealed that microorganisms coloniz-ing the macroorganism take an active part in many important processes of life, theconception arose that the life of the microorganism was impossible without the activeinvolvement of microorganisms. Initial experiments proved that organisms can alsolive in germ-free conditions. These experiments also showed the prospect of a ratherextensive use of germ-free animals in studies of microorganism interactions, so rela-tions between microflora and the macroorganism are investigated to clarify the roleof microflora in the physiological and pathological processes of the macroorganism.

Gnotobiology has passed several stages of development. Initially, mainly the technology of obtaining and rearing germ-free animals developed. Managing thetechnology of obtaining germ-free animals and their biological characterizationestablished the basis for gnotobiotic animals to be widely used in scientific studies.The possibility of standardizing gnotobiotic experimental models from the micro-biological viewpoint presents an extraordinary benefit to experiments with gnotobioticanimals since exact and comparable results can be achieved. Finally, gnotobioticmethods have found use in medical and agricultural practice as well.

A. Bomba, R. Nemcová and S. Gancarcíková4

Gnotobiotic animals typically display remarkable morphological and physiologi-cal properties resulting from a total or partial absence of microflora. In the first phase,changes occur in those organ systems which come into direct contact with themicroflora. Primary morphological deviations develop in the digestive tract and thelymph organs (Kruml et al., 1969); later, secondary changes occur in the blood-makingsystem, the liver and other organs. In gnotobiotic animals, morphological changes areaccompanied by physiological changes of which digestive processes and immunereaction changes are the most typical (Abrams and Bishop, 1967; Havell et al., 1970).

The possibility of exact control of the microflora in gnotobiotic animals laid thefoundations for using the latter in research in several branches of science and instudies into the importance of microflora in the physiology and pathology of livingorganisms. Gnotobiotic animals are an important contribution to studies into thephysiology and pathology of the digestive tract.

Gnotobiotic animals are extremely suitable for immunological research. Germ-free animals that have not come into contact with any antigen, present an optimummodel for studies into the primary immune response. Germ-free animals are ofinvaluable importance in research into the role of antigens in the ontogenesis of theimmune system and the role of the thyroid gland in the immune response (Wilsonet al., 1964). The use of gnotobiotic animals has enabled scientists to gain valuableknowledge on cell and humoral immunity (Talafantová et al., 1989) and on theimmune response to various pathogens (Saif et al., 1996; Herich et al., 1999).

Gnotobiotic animals have also enabled scientists to gain new knowledge of thephysiology of the cardiovascular system (Gordon et al., 1963), the liver (Wostmanet al., 1983), the kidneys (Lev et al., 1970) and the endocrine system (Ukai andMitsuma, 1978). Germ-free animals have become a suitable experimental model forstudies into nitrogen and carbohydrate metabolism (Combe, 1973), and the metab-olism of vitamins, minerals, fatty and bile acids and cholesterol (Coates et al., 1965).

Gnotobiotic animals are frequently employed in medical research. Experimentsin such animals help to clarify the role of gut microflora in the process of carcino-genesis (Drasar and Hill, 1974; Narushima et al., 1998). They have also proved suitable in studies into the role of microflora in the metabolism of different substances and their pharmacological or toxic effects. Experiments with germ-free rats helped to explain the toxicity of cycasine. For cycasine toxicity, the presence of bacteria and bacterial glycosidases metabolizing cycasine to toxic aglycon-methylazoxymethanol proved to be requisite (Laguer and Spatz, 1975).Gnotobiotic techniques are also employed in radiopathology (Mandel et al., 1980),where they were used to confirm the role of microorganisms in the pathogenesis ofradiation disease and the gastrointestinal syndrome, and in the mechanism of actionof radioprotective substances. These studies showed that microorganisms wereinvolved in the postradiation syndrome, both directly by damaging the anatomicalstructure and physiology of the organism, and indirectly through the effects ofmicrobial metabolites. In comparison to conventional animals, germ-free ones

The digestive tract of gnotobiotic animals 5

display increased resistance to irradiation (Mandel et al., 1979). In infectiology,gnotobiotic animals help to disclose the role of microorganisms and their interrela-tions in the etiology and pathogenesis of infectious diseases (Wilson et al., 1986;Rogers et al., 1987a,b; Hodgson et al., 1989; Saif et al., 1996; Ellis et al., 1999).

3. THE USE OF GNOTOBIOTIC ANIMALS IN STUDIES INTO THEFUNCTIONAL AND MORPHOLOGICAL DEVELOPMENT OFTHE DIGESTIVE TRACT

Gnotobiotic animals are a very useful model in studying the physiology of the diges-tive tract, and enable the observation of the role of microorganisms during func-tional and morphological development of the digestive tract. In ruminants, theypresent the optimum model of developmental physiology of the rumen. The gnoto-biotic young of ruminants can be used to observe the development of the rumenecosystem and its interrelations, to study the relations between rumen microflora,microfauna and the macroorganism as well as to determine the effects of rumenmetabolism upon intermediary metabolism. The rumen wall is an important elementof rumen and intermediary metabolism, its epithelium being the connection sitebetween rumen and intermediary metabolism (Kalachnyuk et al., 1987) as well asbeing capable of absorption, synthesis and secretion. The length of rumen papillaeis positively correlated with body growth. Rumen fermentation microflora catabo-lizes soluble carbohydrates; in this process, volatile fatty acids (VFAs) are the majorfinal products of rumen fermentation. VFAs present the chemical stimulus of devel-opment of the rumen epithelium. They stimulate the epithelial metabolism of therumen and support the structural development and resorption activity of the rumen.The growth rate of rumen papillae depends on the amount of VFA produced(Ørskov, 1985). In this way, the rumen microflora directly affects the developmentof the rumen epithelium and the level of intermediary metabolism through the actionof rumen fermentation and its final metabolites – the volatile fatty acids. Fonty et al.(1983a,b, 1988) used meroxenic lambs to assess whether the complexity and originof rumen microflora influenced VFA concentrations and composition. They alsostrived to determine the minimum quantitative and qualitative composition of rumenmicroflora that was required to enable rumen colonization by cellulolytic bacteriaand protozoa. Fonty et al. (1991) also studied the role of rumen microflora in thedevelopment of the rumen ecosystem and the functional development of the rumenat an early age. Bomba et al. (1995) used gnotobiotic equipment to study the devel-opment of rumen fermentation in lambs from birth up to 7 weeks of age in relationto the complexity of the digestive tract ecosystem. In gnotobiotic lambs, coloniza-tion of the individual gut segments by lactobacilli and the inhibitory effects ofLactobacillus casei on the adhesion of enterotoxigenic Escherichia coli K 99 to theintestinal wall were also subjected to examination (Bomba et al., 1994, 1997).Soares et al. (1970) and Lysons et al. (1976a,b) compared several parameters of


morphological and functional development in germ-free, gnotobiotic and conven-tional lambs.

Monogastric gnotobiotic animals were also used to study the functional and mor-phological development of the digestive tract. Nemcová et al. (1997) and Bomba et al.(1994) studied the colonization ability of selected strains of lactobacilli in the smallintestine of gnotobiotic piglets. Studies also focused on the effects of lactobacilli onintestinal metabolism during the first 3 weeks of life (Bomba et al., 1998), and uponorganic acid levels in the mucosal film and the contents of the small intestine (Bombaet al., 1996a). Gnotobiotic animals present the ideal model to determine bacterial inter-actions in the digestive tract. Bomba et al. (1996b, 1999) observed the interactions oflactobacilli and enterotoxinogenic E. coli in the intestinal tract of gnotobiotic piglets.

In experiments on gnotobiotic animals, studies focused on the effects of microfloraupon morphology (Gordon and Pesti, 1971), motility (Gustafsson and Norman, 1969)and secretion and absorption in the digestive tract (Yokota and Coates, 1982).Gnotobiotic animals were also used to clarify the role of intestinal mucosa (Loesche,1968) and pancreatic enzymes (Genell et al., 1977).

4. DEVELOPMENT OF THE DIGESTIVE TRACT IN THE GNOTOBIOTIC RUMINANT YOUNG

In ruminants, the rumen is of major importance from the viewpoint of alimentarytract development. The importance of the rumen increases with the age of the indi-vidual, the growing intake of dry fodder, weaning, and transition to plant nutrition.Weaning is conditioned by full functional development of the rumen. Digestion inthe rumen is a complex system of interactive processes, which include microflora,feed and the animal (Demeyer et al., 1986). In the course of the anatomical devel-opment of the forestomach, striking morphological changes become manifest in theincreased volume of the individual parts of the forestomach, and in the typicalphases of wall formation. In this process, the muscle tissue of the wall, the mucosaand villi and the resorptive tissues are formed. This phase of development is basi-cally influenced by mechanical and chemical stimuli, which arise from the ingestedfeed. Functional and morphological development are closely connected. With theonset of rumination and functioning of the abomaso-ruminal cycle, colonizationof the forestomach parts by bacteria and protozoa, increased metabolic activity andthe resorptive capacity of the forestomach are the criteria of functional development. In addition to the aforementioned endogenous and exogenous factors of food intakeregulation, blood composition and increasing enzyme production in the digestivetract exert their influence upon the functional development of the forestomach andspleen system (Bergner and Ketz, 1975). Simultaneously with the ongoing morpho-logical and functional development, the forestomach is colonized by bacteria andprotozoa, which are of decisive importance in the biochemical processes in this partof the digestive tract. In this way, bacteria and protozoa influence the development


of the forestomach. Gnotobiotic animals present the optimum model for studies intothe role of microflora in the development of the digestive tract in young ruminants.As the intake of dry feeds increases and the anatomical and functional developmentof the rumen proceeds, the level of rumen metabolism increases as well.

4.1. Morphological development of the digestive tract in gnotobiotic lambs

Soares et al. (1970) studied the morphological development of gnotobiotic lambsobtained by hysterotomy or hysterectomy of ewes. The gnotobiotic lambs in theexperiment were not inoculated at all; the control group was reared under conven-tional conditions.

At the age of 8 weeks, the body weight of a conventional and a gnotobiotic lambwas 10.82 and 9.09 kg, respectively. There was a marked reduction in the total stom-ach and reticulorumen weights in gnotobiotic lambs as compared to conventionallambs. At the age of 8 weeks, reticulorumen weight in conventional and gnotobioticlambs receiving identical diets was 322 and only 93 g, respectively. Reticulorumenweight in conventional lambs amounted to 74.5% of forestomach weight and 2.97%of body weight, whereas in gnotobiotic lambs the respective proportions reached58.1 and 1.02%.

Conventional lambs developed reticulorumens which seemed to be normal withregard to age and size. In gnotobiotic lambs, however, reticuloruminal developmentat 8 weeks of age, only approximated that of 2- to 3-week-old conventional animals.Inspection of the papillary development revealed virtually no growth of the papillaein the rumens of gnotobiotic lambs. The ruminal lining was thinner and pink incolour, and the rudimentary papillae were not more than 1 mm high and wererounded in appearance. In comparison, the ruminal lining of conventional lambs ona sterile diet was black, thus indicating a parakeratotic condition, while that of conventional lambs on a conventional diet revealed the normal greyish-green colour.In conventional lambs, papillary development was normal. The papillae measuredabout 5 mm in height and were finger-like in shape. Alexander and Lysons (1971)compared the relative thinness of the walls of the gastrointestinal tract in gnotobioticlambs to that in conventional lambs and found the size of the rumen, reticulum andomasum to be similar at similar ages. Lysons et al. (1971) reported gnotobiotic lambsinoculated with a culture of 8 anaerobic rumen bacteria to grow more intensively thangerm-free lambs; their rumen papillae were better developed, too.

Lysons et al. (1976a) studied the morphological differences in the alimentary tractof gnotobiotic and conventional lambs. The main gross differences were: a) the thick-ness of the wall of the reticulorumen and the intestines, b) the consistency of the con-tents, and c) the development of the papillae in the forestomachs. No differences wereobserved between the individual categories of lambs with respect to wall thickness of


the oesophagus, abomasum, caecum or intestines, nor in the consistency of the contents, except for the contents of the large intestines, which were softer in gnoto-biotic animals. The overall size of the reticulorumen including the contents relativeto body weight was similar in gnotobiotic and conventional lambs. However, macro-scopically, the muscles of the rumen in gnotobiotic lambs seemed to be poorly devel-oped. The position of the rumen pillars in gnotobotic lambs was less obvious fromthe outside and they had less effect in maintaining the shape of the rumen.

Histologically there was a marked hypoplasia of the external muscular layersin the rumen of gnotobiotic lambs, that is, a reduction in the number and size ofthe smooth muscle fibrils. The muscle bundles were small, shrunken and widelyseparated from each other. The individual cells were small; they had dark pyknoticnuclei and did not stain well with eosin.

Macroscopically, the inside of the rumen wall of uninoculated gnotobiotic lambswas pinkish or greyish brown in colour and covered with low rugae rather thanpapillae though some short papillae with bulbous extremities were present in theanterior sac. There seemed to have been little or no development from the neonatalstage on. Uninoculated gnotobiotic lambs and gnotobiotic lambs inoculated withone bacterial species (Bacteroides ruminicola) had considerably shorter rumenpapillae than in conventional lambs.

Microscopically, the thinness of the reticular wall in uninoculated gnotobioticlambs was not as marked as that of the rumen. Histologically, however, the outermuscular layers were hypoplastic. The fibrils of the muscularis mucosae, althoughnot apparently reduced in number, were poorly stained and had shrunken nuclei. Thelamina propria was much reduced in thickness in comparison with that of a normalreticulum. Macroscopically, the papillae in the reticulum were poorly developed andthe mucosal folds less developed in the uninoculated gnotobiotic lambs. There wasno marked difference between gnotobiotic and conventional lambs concerning theconsistency of the ruminal and reticular contents.

The large papillae in the omasal groove near the reticulo-omasal orifice wereapparently well developed in gnotobiotic lambs but the papillae on the leaves of theomasum were smaller in uninoculated gnotobiotic lambs than in inoculated andconventional lambs. The consistency of the contents in gnotobiotic lambs wassimilar to that in conventional lambs.

Macroscopically, the walls of the small intestines of gnotobiotic lambs appearedto be thinner than those of conventional lambs. In the large intestine the differencewas less striking. Histologically, the longitudinal and circular muscle layers ofthe small intestine in gnotobiotic lambs were not obviously thinner than those in conventional lambs but there seemed to be some hypoplasia of the mucosal layer(Lysons et al., 1976a). As figs 1 and 2 indicate, histological examination revealedbetter development of rumen mucosa in conventional lamb in comparison to gnoto-biotic lamb at 7 weeks of age (Zitnan, 2001, personal communication).


4.2. Physiological development of the digestive tract in gnotobiotic lambs

The functional development of the rumen also depends on the complexity of itsmicroflora. In conventional animals it is difficult to study the specific role ofmicroorganisms and their interactions because of the complexity of the microbialecosystem of the rumen. In order to understand the digestive mechanisms involvedin the rumen, microbial components need to be simplified by using animals with areduced number of bacterial and protozoan species (Fonty et al., 1983a).

Rumination was observed in gnotobiotic lambs, but it occurred only occasionallyand much less frequently than in conventional animals. Inoculated gnotobioticlambs did not appear to ruminate more frequently than their uninoculated gnotobi-otic fellows, but the conventionalized lambs ruminated normally within 6 weeksafter inoculation (Lysons et al., 1976a).

Bomba et al. (1995) observed the development of rumen fermentation in conventional and gnotobiotic lambs from birth to 7 weeks of age. Conventionallambs with a complex microflora did not receive any inoculum. The inoculumof gnotobiotic lambs contained Streptococcus bovis, Prevoxella ruminicola,


Fig. 1. Histological section of rumen mucosa in conventional lamb at 7 weeks of age (Zitnan, 2001, personal communication).

Fig. 2. Histological section of rumen mucosa in gnotobiotic lamb at 7 weeks of age (Zitnan, 2001, personalcommunication).

Butyrivibrio fibrisolvens and Selenomonas ruminantium at a concentration of 106

each. In both groups of lambs rumen fluid pH proved to be rather stable throughoutthe observation period. The values of pH ranged within 6.5–6.8 and 7.1–7.4 in theconventional and gnotobiotic groups, respectively. When compared to conventionallambs, the pH of the rumen contents in gnotobiotic lambs was increased throughoutthe investigated period, the differences being significant (P < 0.01) at 7 weeks of age(conventional lambs 6.7, gnotobiotic lambs 7.4).

Comparison to gnotobiotic lambs revealed total volatile fatty acid (VFA) concentrations in conventional lambs to be higher throughout the observationperiod, the differences being significant at 4 and 5 weeks of age (P < 0.05 and P <0.001, respectively). In conventional lambs, total VFA levels manifested an increas-ing tendency between weeks 4 and 7 of age and reached their maximum at 7 weeks (57 mmol l−1) whereas in gnotobiotic lambs the range was narrow (24.3–30.1 mmol l−1) and the peak occurred at 6 weeks of age. In gnotobiotic lambs signi-ficantly increased molar proportions of acetic acid were observed whereas in con-ventional lambs the molar proportions of propionic acid proved to be significantlyincreased. The molar proportions of butyric and valeric acids were increased in conventional lambs but the group differences were not significant. In gnotobioticlambs no isoacids were found. Alpha amylase (E.C.3.2.1.1.) activity of the rumencontents was significantly increased in gnotobiotic lambs between weeks 2 and 6 of age whereas cellulase (endoglucanase E.C.3.2.1.4. and cellobiohydrolaseE.C.3.2.1.91.) activity was significantly increased in 4-week-old conventionallambs. Over the whole period of milk nutrition no significant differences in urease(E.C.3.5.1.5.) activity of the rumen contents were observed in the groups examined.

The above study compared the level of rumen fermentation in conventionallyreared lambs and in lambs with an extremely reduced and defined microflora, thelatter enabled demonstration of the role of the complexity of the rumen ecosystemin the functional development of the rumen at an early age. The results obtainedindicated that the complexity of rumen microflora significantly influenced thedevelopment of rumen fermentation both from the quantitative and the qualitativeviewpoint. Fonty et al. (1988) observed the level of rumen fermentation in gnotobi-otic lambs inoculated with 182, 106, 32 and 16 non-cellulolytic strains isolated from the rumen. Volatile fatty acid levels rather differed from one lamb group to theother. The more complex the inoculum administered to the animals was, the higherwere the VFA levels observed. In animals inoculated with 182 strains the VFA con-centration was similar to that measured in conventional lambs fed the same diet(approximately 80 mmol l−1 after feeding). In lambs inoculated with only 16 strainsthere was almost no fermentation (30 mmol l−1 of VFA). These results demonstratethat the functions of the rumen and the stability of the ecosystem depend on thecomplexity and diversity of the microflora. In the light of present knowledge it is notpossible to determine accurately the composition of the minimum flora enablingrumen development and function.


Soares et al. (1970) reported ruminal fluids from gnotobiotic lambs to containmarkedly less total VFA than those from conventional lambs. Lysons et al. (1976b)dosed five gnotobiotic lambs with different combinations of 11 species of rumen bacteria. Two of the species could not be reisolated but the remainder established readily in the rumen and the viable counts of most of the individual species were comparable to those in normal sheep, however, VFA levels were decreased and in fourof the lambs the proportion of butyric and propionic acids was higher and lower thanin normal sheep, respectively. Cellulolytic, ureolytic and methanogenic activitiesappeared to be taking place and lactate-utilizing bacteria appeared to reverse the accu-mulation of lactate, which resulted from the activity of lactate-producing bacteria.

Fonty et al. (1991) studied the development of rumen digestive functions inlambs placed in sterile isolators at 1, 4, 8 or 9 days of age in order to define the roleof the bacterial species that colonize the rumen just after birth. The values of themain rumen digestive parameters (pH, VFA levels, ammonia, lactic acid) in theselambs were close to those observed in the conventional controls. Likewise, digestiveutilization of dry matter and starch was comparable in the isolated and control animals but digestibility of crude cellulose was higher in isolated lambs which harboured Fibrobacter succinogenes as the major cellulolytic bacterial species.These results suggest that rumen flora of the very young lamb play an essential rolein the establishment of the rumen ecosystem and in the setting up of the digestivefunctions. Those bacterial species that colonize the rumen immediately after birthwhen this organ is not yet active, contribute a biotype favouring the establishment ofcellulolytic strains and the set-up of digestive processes that affect both degradationof the lignocellulose-rich feeds and fermentation of the resulting soluble compounds.

Ecological factors controlling the establishment of cellulolytic bacteria and ciliateprotozoa in the lamb rumen were studied in meroxenic lambs (Fonty et al., 1983a).The results obtained in this study suggest that establishment of cellulolytic bacteriaand protozoa requires an abundant and complex flora and is favoured by early inocu-lation of the animals. The difficulty in establishing cellulolytic bacteria in the rumenof animals with a limited flora is probably linked to the very high and strict nutritionalrequirements of such organisms (Bryant, 1973). Creation of conditions necessary forthe establishment of cellulolytic bacteria probably depends on a number of very complex requirements. These requirements are not necessarily provided by the dominant bacteria of the flora, which are nonetheless generally thought to play themain role in the rumen.

All the above-mentioned results point to the extremely important role microfloraplays in the development of the rumen. There is a good relationship between the devel-opment of rumen function and flora complexity. The presence of a simple flora cannot assure the digestive function as properly as a complex flora can (Fonty et al.,1983b). The fact that early inoculation of animals is a factor favouring fermentationand digestive activities in the rumen is probably related to the action of bacteria on thedevelopment of papillae, rumen mucosa and the digestive tract (Lysons et al., 1976a).


A complex microflora presents a requisite condition of optimum development of thealimentary tract in ruminants.

5. DEVELOPMENT OF THE GASTROINTESTINAL TRACT OF GNOTOBIOTIC PIGLETS

The period immediately after birth is probably the most critical one in the whole lifeof the animal. In this period significant growth, morphological changes and matura-tion of the gastrointestinal tract take place. Prior to birth the alimentary tract isexposed to substances from the ingested amniotic fluid, which seems to be of impor-tance to its development (Trahair and Harding, 1992). The colostrum, however, differs from the amniotic fluid by the density of nutrients, and having highimmunoglobulin, enzyme, hormone, growth factor and neuroendocrine peptide levels (Koldovsky and Thornburg, 1989). Widdowson and Crabb (1976) were thefirst to demonstrate the effect of the colostrum upon alimentary tract developmentby comparing piglets suckling colostrum to watered animals. In this way high levels of several hormones and growth promoting peptides like insulin, cortisol, epidermal growth factor (EGF) and insulin-type growth factor I (IGF-I) were statedin the maternal colostrum. It was proved that colostral growth factors play an impor-tant role in the postnatal development of the digestive tract of newborn young. Fromthis point of view, gnotobiotic piglets are a suitable model for studies into the devel-opment of the digestive tract.

5.1. Morphological development of the digestive tract in gnotobiotic suckling pigs

In germ-free animals, the normal structure and morphology of the gut are altered inways that emphasize the importance of an animal’s interaction with its indigenousmicrobial flora in establishing its defences against microbial invasion while, at thesame time, adequate host nutrition is maintained by normal alimentary tract func-tion (Heneghan, 1965). However, the overall mass of the small intestine in eachgerm-free species is decreased, and its surface area is smaller, whereas the villi ofthe small intestine are unusually uniform in shape and slender, with crypts whichare shorter and less populated than in the respective conventional control animals(Meslin et al., 1973).

The structure of the small intestine is a very sensitive indicator of the shift fromthe germ-free state into a state in which contact of the mucosa with pathogenic ornon-pathogenic microorganisms occurs (Kruml et al., 1969). The lamina propria ismuch thinner in germ-free animals (Abrams, 1969). The mucosal villi of the smallintestine of germ-free piglets are very fine and they have a small amount of axialstroma with low cellularity. The vilus/crypt cell ratio is always higher in germ-freepigs than in conventional pigs which indicates that less proliferating tissue is


required to keep the germ-free mucosa intact (Heneghan, 1979). In general, manualstereological morphometric techniques tended to confirm these morphologicaltrends (Heneghan et al., 1979). The lymph follicles, which are already present atbirth, increase only very slowly in size during postnatal ontogenesis. The increasein the number of large pyroninophilic cells in the follicles is likewise only moder-ate. No germinal centres were found within a 68-day observation period. In con-ventional piglets, in addition to the large primary follicles, germinal centres ofvarying size were already found on the 12th day. Small quantities of cells of theplasmocyte series were also demonstrated. The mucosal villi of such animals weremore cellular and contained numerous lymphocytes, which in many places formedlarge aggregates expanding the villous space (Kruml et al., 1969).

During the first 5–6 weeks of postnatal life, the epithelium of the small intestinein germ-free pigs has a particular appearance. In epithelial cells great vacuoles arefound, thus forming the “water transparent” cells. As a rule, enterocytes of this typeare seen in pig fetuses and in newborn piglets but in conventional environment theychange within a few days after birth. The “ageing” or “senescent” enterocytes ingerm-free pigs are obviously a consequence of the prolonged life span of epithelialcells which is caused by the lower mitotic rate of stem cells in Lieberkuhn’s crypts.The life span of epithelial cells was found to be 96 h in conventional pigs but 200 hin germ-free pigs. However, the “senescent” enterocytes were stated to be workingwell in the transport of nutrients across the mucous membrane. It is of advantage in rearing germ-free pigs that these animals do not exhibit an enlarged caecum(megacaecum) and colon, as can be seen in many species of germ-free mammals,particularly rodents (Mandel and Trávnicek, 1987).

5.2. Intestinal metabolism in gnotobiotic pigs

Bomba et al. (1998) studied the intestinal metabolism in two groups of gnotobiotic pigs(one non-inoculated and one inoculated only with Lactobacillus casei subsp. casei)during the first 3 weeks of life. The Lactobacillus casei subsp. casei counts in the jejunal and ileal contents of inoculated gnotobiotic piglets ranged from 8.37 to 9.87 log 10 ml−1 during the entire period of investigation whereas the numbers ofLactobacillus casei subsp. casei adhering to the jejunal and ileal mucous membranewere significantly lower (P < 0.05) ranging from 5.63 to 6.06 log 10 cm−2. The num-bers of lactobacilli adhering to the jejunal and ileal mucosa and found in the jejunal andileal contents were comparable to the data obtained in conventional and gnotobioticpiglets by other authors (Pollmann et al., 1980; Sarra et al., 1991; Tortuero et al., 1995).

At the age of 1 and 3 weeks, the actual acidity of the jejunal contents of gnoto-biotic piglets inoculated with Lactobacillus casei subsp. casei was significantlylower (P < 0.05 and P < 0.01) in comparison with that in non-inoculated animals (seetables 1 and 2, respectively). The pH value of the ileal contents of inoculated pigletswas also lower, however, the differences were not significant. Zitnan et al. (2001)


observed the pH of the jejunal and ileal contents in conventional suckling piglets atan identical age. When comparing the actual acidity of the individual small intestinalsegments it can be stated that the pH of the jejunal contents in non-inoculated gnotobiotic piglets aged 1 and 3 weeks (7.49 and 7.12, respectively) was significantlyincreased when compared to values recorded in conventional piglets of the same age(6.23 and 6.19, respectively). In contrast, the pH of the jejunal contents in gnotobioticpiglets inoculated with Lactobacillus casei subsp. casei was moderately lower (5.63and 5.84, respectively). The actual acidity of the ileum in non-inoculated gnotobioticpiglets (8.63 and 8.35, respectively) as compared to the conventional ones (6.27 and6.79, respectively) was even more significantly increased than that of the jejunum. In gnotobiotic piglets inoculated with lactobacilli, ileal pH (6.43 and 7.30, respec-tively) was slightly increased when compared to that in conventional piglets.

Morphological changes in the digestive tract are also influenced by volatile fattyacids (Goodlad et al., 1989), which play an important role in bacterial interactions ofthe alimentary tract. The ability to generate organic acids, particularly lactic andacetic acids, presents one of the mechanisms by which lactobacilli perform theirinhibitory effect upon pathogens (Piard and Desmazeaud, 1991). With decreasing pH values, the inhibitory activity of the above acids increases (Daly et al., 1972), their molecular form being toxic for bacteria. The increased toxicity of acetic acid is


Table 1. The influence of continuous application of Lactobacillus casei on colonization andactual acidity in the jejunum and ileum in 1-week-old gnotobiotic piglets

Lactobacilli (content) Lactobacilli (mucosa)Intestine Group (log 10 ml−1) (log 10 cm−2) pH

Jejunum O 0 0 7.49 ± 0.22*L 8.39 ± 0.07 5.63 ± 0.41 5.63 ± 0.31

Ileum O 0 0 8.63 ± 0.10L 8.38 ± 0.07 5.69 ± 0.65 6.43 ± 0.83

* P < 0.05.Group L: inoculated with Lactobacillus casei.Group O: non-inoculated.

Table 2. The influence of continuous application of Lactobacillus casei on colonization andactual acidity in the jejunum and ileum in 3-week-old gnotobiotic piglets

Lactobacilli (content) Lactobacilli (mucosa)Intestine Group (log 10 ml−1) (log 10 cm−2) pH

Jejunum O 0 0 7.12 ± 0.06**L 8.81 ± 0.33 5.75 ± 0.52 5.84 ± 0.06

Ileum O 0 0 8.35 ± 0.69L 9.87 ± 0.59 6.06 ± 0.36 7.30 ± 0.44

** P < 0.01.Group L: inoculated with Lactobacillus casei.Group O: non-inoculated.

attributed to its higher pKa in comparison to lactic acid. Increased lactic acid levelsintensify the toxicity of acetic acid (Adams and Hall, 1988).

Comparison of lactic acid levels in the jejunal and ileal contents of gnotobioticpiglets (Bomba et al., 1998) and conventional suckling piglets (Zitnan et al., 2001) at 1 week of age, revealed the highest levels in conventional animals (27.50 and 26.90 mmol l−1, respectively) and in Lactobacillus casei inoculated gnotobioticpiglets (26.60 and 14.20 mmol l−1, respectively). The lowest levels of lactic acid in thejejunal and ileal contents were seen in non-inoculated gnotobiotic piglets (4.40 and6.45 mmol l−1, respectively). At the age of 3 weeks, lactic acid levels in the jejunum andileum reached maximum values in Lactobacillus casei inoculated gnotobiotic piglets(33.15 mmol l−1) and in conventional piglets (15.52 mmol l−1), respectively. At 1 weekof age, maximum acetic acid levels in the jejunal and ileal contents were stated in con-ventional piglets (33.05 and 21.81 mmol l−1, respectively), the respective levels inLactobacillus casei inoculated and non-inoculated gnotobiotic piglets were somewhatlower (11.80 and 11.85 mmol l−1 vs 13.15 and 3.90 mmol l−1). At 3 weeks of age, max-imum acetic acid levels were observed in the jejunal and ileal contents of conventionalpiglets (10.17 and 25.9 mmol l−1, respectively) and Lactobacillus casei inoculatedgnotobiotic piglets (3.9 and 11.00 mmol l−1, respectively). These results show that thecomplexity of the intestinal microflora affects the production of the investigated organicacids in the alimentary tract of piglets.

Bomba et al. (1996a) investigated the effect of the inoculation of threeLactobacillus strains upon lactic, acetic, acetoacetic and propionic acid levels in themucosal film and ileal contents of gnotobiotic pigs. In the jejunum of inoculatedanimals, the mucosal film revealed significantly increased levels of lactic, propionicand acetoacetic acids when compared to the contents (25.3 vs 10.8 mmol l−1, 18.5 vs5 mmol l−1, and 29.7 vs 11.2 mmol l−1, respectively) as well as non-significantlyincreased acetic acid levels (11.0 vs 5.8 mmol l−1). In the ileum of gnotobiotic pigs,propionic acid levels in the mucosal film were significantly higher than those in thecontents (21.2 vs 9.5 mmol l−1). In comparison to the contents, the increased lactic,acetic and acetoacetic acid levels in the film proved to be non-significant. The aboveresults suggest that significantly increased levels of the lactobacilli-producedorganic acids in the intestinal mucosal film may present an efficient barrier to inhibitthe adherence of digestive tract pathogens to intestinal mucosa.

6. FUTURE PERSPECTIVES

In the future, gnotobiotic research will be involved in many different fields of biol-ogy, nutrition and medicine. It can be suggested that gnotobiology will be part ofmainstream modern ecology, environmental toxicology, molecular immunology,modern clinical medicine, investigations of genetically modified microorganismsand modern food research. In the field of digestive tract physiology, gnotobioticresearch will be aimed at gastrointestinal ecosystem interactions. Despite a lot of


knowledge obtained, the mode of action of probiotic microorganisms upon digestivetract pathogens has not yet been explained. In order to enhance the efficacy of pro-biotics it is necessary to obtain additional important knowledge on the mechanismsmediating their effect in the digestive tract. Gathering knowledge in the given fieldswill support the development of more effective probiotic products that will con-tribute to increased health and a more effective prevention of alimentary tractdiseases in both humans and animals.

REFERENCES

Abrams, G.D., Bishop, J.E., 1967. Effect of the normal microbial flora on gastrointestinal motility. Proc.Soc. Exp. Biol. Med. 126, 301–304.

Abrams, G.D., 1969. Effects of normal flora on the host defences against microbial invasion. Adv. Exp.Med. Biol. 3, 197–206.

Adams, M.R., Hall, C.J., 1988. Growth inhibition of food borne pathogens by lactic and acetic acids andtheir mixtures. Int. J. Food Sci. Tech. 23, 287–292.

Alexander, T.J.L., Lysons, R.J., 1971. Observations on reaching gnotobiotic lambs. Brit. Vet. J. 127, 347–357.

Bergner, H., Ketz, A., 1975. Digestion, Resorption and Intermedial Metabolism in Farm Animals(in Slovak). Príroda, Bratislava.

Bomba, A., Kravjansky, I., Kastel’, R., Cízek, M., Kapitancik, B., Juhásová, Z., Herich, R., Zitnan, R.,Bucko, V., 1994. Colonization of the digestive tract in germ-free and conventional lambs by a definedlactoflora (in Slovak, with English abstract). Vet. Med. Czech. 39, 701–710.

Bomba, A., Zitnan, R., Koniarova, I., Lauková, A., Sommer, A., Posivák, J., Bucko, V., Pataky, J., 1995.Rumen fermentation and metabolic profile in conventional and gnotobiotic lambs. Arch. Anim. Nutr.48, 231–243.

Bomba, A., Kastel’, R., Gancarcíková, S., Nemcová, R., Herich, R., Cízek, M., 1996a. The effect ofLactobacilli inoculation on organic acid levels in the mucosal film and the small intestine contents ingnotobiotic pigs. Berl. Munch. Tierärztl. Wschr. 109, 11/12, 428–430.

Bomba, A., Nemcová, R., Kastel’, R., Herich, R., Pataky, J., Cízek, M., 1996b. Interactions ofLactobacillus sp. and enteropathogenic Escherichia coli under in vitro and in vivo conditions. Vet. Med. Czech. 41, 155–158.

Bomba, A., Kravjansky, I., Kastel’, R., Herich, R., Juhásová, Z., Cízek, M., Kapitancik, B., 1997.Inhibitory effect of Lactobacillus casei upon the adhesion of enterotoxigenic Escherichia coli K 99 tothe intestinal mucosa in gnotobiotic lambs. Small Ruminant Res. 23, 199–206.

Bomba, A., Gancarcíková, S., Nemcová, R., Herich, R., Kastel’, R., Depta, A., Demeterová, M., Ledecky, V.,Zitnan, R., 1998. The effect of lactic acid bacteria on intestinal metabolism and metabolic profile ingnotobiotic pigs. Dtsch. Tierärztl. Wschr. 105, 365–396.

Bomba, A., Nemcová, R., Gancarcíková, S., Herich, R., Kastel’, R., 1999. Potentiation of the effective-ness of Lactobacillus casei in the prevention of E. coli induced diarrhoea in conventional and gnoto-biotic pigs. In: Paul, P.S., Francis, D.H. (Eds.), Mechanisms in the Pathogenesis of Enteric Diseases2. Kluwer Academic, Plenum Publishers, New York, pp. 185–190.

Bryant, M.P., 1973. Nutritional requirements of the predominant rumen cellulolytic bacteria. FederationProceedings 32, 1809–1813.

Coates, M.E., Gustafsson, B.E., 1984. The Germ-free Animal in Biomedical Research. LaboratoryAnimals Ltd, London.

Coates, M.E., Harrison, G.F., Moore, J.H., 1965. Cholesterol metabolism in germ-free and conventionalchicks. Ernährungsforschung 10, 251–256.

Combe, E., 1973. Utilization of nitrogen and amino acids by the germfree rat. Ann. Biol. Anim. Biochem.Biophys. 13, 738–739.

Daly, C., Sandine, W.E., Elliker, P.R., 1972. Interaction of food starter cultures and food borne pathogens:Streptococcus diacetylactis versus food pathogens. J. Milk Food Technol. 35, 349–357.


Demeyer, D., Van Nevel, C., Teller, E., Godeau, J.M., 1986. Manipulation of rumen digestion in relationto the level of production in ruminants. Arch. Anim. Nutr. 36, 132–143.

Drasar, B.S., Hill, M.J., 1974. Human Intestinal Flora. Academic Press, New York, p. 263.Duskin, V.A., Intuzarov, M.M., Petracev, D.A., Podonpusova, G.U., Ponov, B.I., Sorokin, B.B.,

Siskov, B.P., 1983. Theoretical and Practical Principles of Gnotobiology (in Russian). Kolos, Moskva.Ellis, J., Krakowka, S., Laimore, M., Haines, D., Bratanich, A., Clark, E., Allan, G., Konoly, C.,

Hassard, L., Meehan, B., Martin, K., Harding, J., Kennedy, S., McNeilly, F., 1999. Reproduction oflesions of postweaning multisystemic wasting syndrome in gnotobiotic piglets. J. Vet. Diagn. Invest.11, 3–14.

Fonty, G., Gouet, P., Jouany, J.P., Senaud, J., 1983a. Ecological factors determining the establishment ofcellulolytic bacteria and protozoa in the rumen of meroxenic lambs. J. Gen. Microbiol. 129, 213.

Fonty, G., Jouany, J.P., Thivend, P., Gouet, P. Senaud, J., 1983b. A descriptive study of rumen digestionin meroxenic lambs according to the nature and complexity of the microflora. Reprod. Nutr. Develop.23, 857.

Fonty, G., Gouet, P., Ratefiarivelo, H., Jouany, J.P., 1988. Establishment of Bacteroides succinogenes andmeasurement of the main digestive parameters in the rumen of gnotoxenic lambs. Can. J. Microbiol.34, 39–46.

Fonty, G., Jouany, J.P., Chavarot, M., Bonnemoy, F., Gouet, P., 1991. Development of the rumen diges-tive functions in lambs placed in a sterile isolator a few days after birth. Reprod. Nutr. Develop. 31,521–528.

Genell, S., Gustafsson, B.E., Ohlsson, K., 1977. Imunochemical quantitation of pancreatic endopeptidasesin the intestinal contents of germfree and conventional rats. Scand. J. Gastroenterol. 12, 811–820.

Goodlad, R.A., Ratcliffe, B., Fordham, J.P., Wright, N.A., 1989. Does dietary fibre stimulate intestinalepithelial cell proliferation in germ-free rats? Gut 30, 820–825.

Gordon, H.A., Pesti, L., 1971. The gnotobiotic animal as a tool in the study of host–parasite relationships.Bacteriol. Rev. 35, 390–429.

Gordon, H.A., Wostman, B.S., Bruckner-Kardoss, E., 1963. Effects of microbial flora on cardiac outputand other elements of blood circulation. Proc. Soc. Exp. Biol. Med. 114, 301–304.

Gustafsson, B.E., Norman, A., 1969. Influence of the diet on the turnover of bile acids in germ-free andconventional rats. Brit. J. Nutr. 23, 429–442.

Havell, E.A., Holtermann, O.A., Starr, T.J., 1970. The influence of the indigenous bacterial flora uponthe interferon response of mice. J. Gen. Vir. 9, 93–95.

Heneghan, J.B., 1965. Imbalance of the normal microbial flora: The germ-free alimentary tract. Amer. J.Dig. Dis. 10, 864–869.

Heneghan, J.B., 1979. Enterocyte kinetics mucosal surface area and mucus in gnotobiotes. In: Fliedner, T.,Heit, H., Niethammer, D., Pflieger, H. (Eds.), Clinical and Experimental Gnotobiotic. Gustav FischerVerlag, Stuttgart, pp. 19–27.

Heneghan, J.B., Gordon, H.A., Miniats, O.P., 1979. Intestinal mucosal surface area and goblet cells ingerm-free and conventional piglets. In: Fliedner, T., Heit, H., Niethammer, D., Pflieger, H. (Eds.),Clinical and Experimental Gnotobiotic. Gustav Fischer Verlag, Stuttgart, pp. 107–111.

Herich, R., Bomba, A., Nemcová, R., Gancarcíková, S., 1999. The influence of short-term and continu-ous administrations of Lactobacillus casei on basic hematological and immunological parameters ingnotobiotic piglets. Food Agr. Immunol. 11, 287–295.

Hodgson, J.C., King, T., Moon, G., Donacke, W., Quirie, M., 1989. Host responses during infection innewborn lambs. FEMS Microbiol. Inmunol. 47, 311–312.

Kalachnyuk, G.I., Savka, O.G., Leskovich, B.M., Baran, M., Kmet’, V., Várady, J., Marounek, M.,Kopecny, J., Simunek, I., 1987. Metabolism status of young cattle at prolonged feeding of non/traditional feedstuffs. In: Bod’a, K. (Ed.), Proceedings of IV International Symposium on Physiologyof Ruminant Nutrition, Strbské Pleso. Institute of Animal Physiology, Slovak Academy of Sciences,Kosice, pp. 43–56.

Koldovsky, O., Thornburg, W., 1989. Peptide hormones and hormone-like substances in milk. In:Atkinson, S.A., Lonnerdal, B. (Eds.), Protein and Non-Protein Nitrogen in Human Milk. CRC Press,New York, pp. 53–65.


Kruml, J., Ludvík, J., Trebichovsky, I., Mandel, L., Kovaru, F., 1969. Morphology of germ-free piglets.Folia Microbiol. Prague 14, 441–446.

Laguer, G.L., Spatz, M., 1975. Oncogenicity of cycasin and methylazoxymethanol. GANN 17, 189–204.Lev, M., Alexander, R.H., Levenson, S., 1970. Impaired water metabolism in germ-free rats. Proc. Soc.

Exp. Biol. Med. 135, 700–705.Loesche, W.J., 1968. Protein and carbohydrate composition of cecal contents of gnotobiotic rats and

mice. Proc. Soc. Exp. Biol. Med. 128, 195–199.Lysons, R.J., Alexander, T.J.L., Hobson, P.N., Mann, S.O., Stewart, C.S., 1971. Establishment of a lim-

ited rumen microflora in gnotobiotic lambs. Res. Vet. Sci. 12, 486–487.Lysons, R.J., Alexander, T.J.L., Wellstead, D., Jennings, W., 1976a. Observations on the alimentary tract

of gnotobiotic lambs. Res. Vet. Sci. 20, 70–76.Lysons, R.J., Alexander, T.J.L., Wellstead, D., Hobson, P.N., Mann, S.O., Stewart, C.S., 1976b. Defined

bacterial populations in the rumens of gnotobiotic lambs. J. Gen. Microbiol. 94, 257–269.Mandel, L., Trávnicek, J., 1987. The minipig as a model in gnotobiology. Die Nahrung 31, 613–618.Mandel, L., Trebichavsky, I., Morávek, F., Trávnicek, J., 1980. Changes in the intestinal epithelial cells

in abdominally irradiated germ-free piglets. Strahlentherapie 156, 284–289.Meslin, J.C., Sacquet, E., Guenet, J.L., 1973. Effects of the bacterial flora upon the morphology and sur-

face of mucosa surface in the small intestine of rats (in French). Ann. Biol. Anim. Biochem. Biophys.11, 334–335.

Narushima, S., Kukuji, I., Mitsuoka, T., Nakayama, H., Itoh, T., Hioki, K., Nomura, T., 1998. Effect of mouse intestinal bacteria on incidence of colorectal tumors induced by 1,2-dimethylhydrazine injection in gnotobiotic transgenic mice harbouring human prototype c-Ha-ras genes. Exp. Anim. 47,111–117.

Nemcová, R., Bomba, A., Herich, R., Gancarcíková, S., 1997. Colonization capability of orally adminis-tered lactobacillus strains in the gut of gnotobiotic piglets. Dtsch. Tierärztl. Wschr. 105, 199–200.

Ørskov, E.R., 1985. Protein Nutrition of Ruminants (in Russian). Agropromizdat, Moskva.Piard, J.C., Desmazeaud, M., 1991. Inhibiting factors produced by lactic acid bacteria. 1. Oxygen

metabolites and catabolism end-products. Lait 71, 525–541.Pollmann, D.S., Danielson, D.M., Wren, W.B., Peo, E.R., Shahani, K.M., 1980. Influence of

Lactobacillus acidophillus inoculum on gnotobiotic and conventional pigs. J. Anim. Sci. 51, 629–637.Rogers, D.G., Cheville, N.F., Pugh, G.W., 1987a. Pathogenesis of corneal lesions caused by Moraxella

bovis in gnotobiotic calves. Vet. Pathol. 24, 287–295.Rogers, D.G., Cheville, N.F., Pugh, G.W., 1987b. Conjunctival lesions caused by Moraxella bovis in gno-

tobiotic calves. Vet. Pathol. 24, 554–559.Saif, L.J., Ward, L.A., Yuan, L., Rosen, B.I., To, T.L., 1996. The gnotobiotic piglets as a model for stud-

ies of disease pathogenesis and immunity to human rotaviruses. Arch. Virol. 12, 153–161.Sarra, P.G., Cantalupo, R., Massa, S., Trovatelli, L.D., 1991. Colonization of the gastrointestinal tracts of

conventional piglets by Lactobacillus strains. J. Gen. Appl. Microbiol. 37, 219–223.Soares, J.H., Leffel, E.C., Larsen, R.K., 1970. Neonatal lambs in a gnotobiotic environment. J. Anim. Sci.

31, 733–740.Talafantová, M., Mandel, L., Trebichavsky, I., 1989. The occurrence of intestinal bacteria in the lungs of

gnotobiotic piglets. Microbiol. Ther. 18, 311–320.Tortuero, F., Rioperex, J., Fernandez, E., Rodriguez, M.L., 1995. Response of piglets to oral administra-

tion of lactic acid bacteria. J. Food Protect. 58, 1369–1374.Trahair, J.F., Harding, R., 1992. Ultrastructural anomalies in the fetal small intestine indicate that fetal

swallowing is important for normal development: An experimental study. Virchows Archiv. 420 A,302–312.

Ukai, M., Mitsuma, T., 1978. Plasma triiodothyromine, thyroxine and thyrotropin levels in germfree rats.Experientia 34, 1095–1096.

Widdowson, E.M., Crabb, D.E., 1976. Changes in the organs of pigs in response to feeding for the first24 h. after birth. Biol. Neonate 28, 261–271.

Wilson, K.H., Sheagren, J.N., Freter, R., Weatherbee, L., Lyerly, D., 1986. Gnotobiotic models for studyof the microbial ecology of Clostridium difficile and Escherichia coli. J. Infect. Dis. 153, 3, 547–551.


Wilson, R., Sjodin, K., Bealmear, M., 1964. Thymus studies in germfree mice. In: Defendi, V., Metealf, D.(Eds.), The Thymus. Wistar Institute Press, Philadelphia, pp. 89–93.

Wostmann, B.S., Larkin, C., Moriarty, A., Bruchner-Kardoss, E., 1983. Dietary intake, energy metabo-lism and excretory losses of adult male germfree Wistar rats. Lab. Anim. Sci. 33, 46–50.

Yokota, H., Coates, M.E., 1982. The uptake of nutrients from the small intestine of gnotobiotic and con-ventional chicks. Brit. J. Nutr. 47, 349–356.

Zitnan, R., Sommer, A., Gancarcíková, S., Nemcová, R., Bomba, A., Guba, P., Mudronová, D., Lukacko, M., Zupcanová, M., 2001. Some aspects of the morphological and functional developmentof the digestive tract in piglets during milk feeding and weaning. Proc. Soc. Nutr. Physiol. 10, 116.


The intestinal flora of pigs contains several hundred microbial species, mostly strictanaerobes. A great amount of these bacteria reside in the large intestine, which inadult pigs consists of mainly Gram-positive bacteria such as cocci, lactobacilli,eubacteria and clostridia. The composition of the intestinal microflora is a result ofthe interaction between the microorganisms that colonize the gut and the intestinalphysiology of the pigs. The initial inoculum is usually derived from the sow at thetime of birth and the climax flora is developed through a gradual process in whichthere is a shift in relative abundance of various microorganisms, especially through-out the first month of the pig’s life. While a part of this microflora is constantly present in the gut (resident flora), some microorganisms have a short residence anddynamically change the composition of the microflora. The turnover of this flora(also known as the transient flora) in the gut depends on both the composition of theresident flora and the degree of contamination of ingested food and other sourceswhich in traditional indoor farming include the sow’s skin and the pen’s environ-ment. The stability and diversity of this flora has a tremendous role in maintainingthe health status of the pigs, especially during the suckling and post-weaning period.Most investigations of the intestinal flora in pigs focus on classical and/or molecu-lar methods, aiming to isolate, enumerate and/or qualitatively identify different bacterial groups. Other recent studies that measure the metabolic capability andfunctional status of the intestinal microflora in pigs have added knowledge about thecomposition and dynamics of the gut flora, especially in pre- and post-weaning pigs.

1. INTRODUCTION

The intestinal microflora of pigs comprises hundreds of bacterial species most ofwhich are residing in the lower part of the gastrointestinal tract. This flora develops

2 Metabolism and population dynamics ofthe intestinal microflora in the growing pig

M. Katoulia and P. Wallgrenb

aFaculty of Science, University of the Sunshine Coast, Maroochydore, Queensland4558, AustraliabDepartment of Ruminant and Porcine Diseases, National Veterinary Institute,Uppsala S-751 89, Sweden

21



through a process of ecological succession and plays a tremendous role in the stateof health and disease of pigs, especially during suckling and post-weaning periods.Among the important factors in this process is the influence of the interactionbetween the microorganisms that contaminate the animal, diet regime and foodcomposition, immunological status of the pig and environmental factors on theintestinal physiology. Several studies have tried to identify the types of bacteria colonizing the intestine of growing pigs. Most of these studies utilize selectivemedia, which lead to enumeration of few particular bacterial groups. The problemsassociated with culture-based techniques are yet exacerbated in anaerobic habitats.Using conventional techniques of culturing and identification, only about 30−40separate species have been generally recovered from any individual animal. Recentdevelopment of more refined molecular techniques has opened new windows ofopportunity to study unculturable bacterial components of the alimentary tract or ofmembers of the intestinal microflora that cannot tolerate exposure to oxygen.

In addition to this new and promising approach, recent in vitro methods focuson measuring the metabolic activities of the major intestinal flora. These methods,alone or in combination, have been extensively used to investigate the populationstructure and the functional status of the intestinal flora in growing pigs.

In this chapter we discuss the available information on population dynamics ofthe intestinal flora in growing pigs, and address factors involved in changes of thisflora during different stages of the animal’s life and in health and disease.

2. PIG HUSBANDRY

Despite the fact that adult pigs may weigh over 300 kg, they only weigh between 1 and2 kg at birth. A sow normally gives birth to a litter of around 10 piglets and, in accor-dance with modern agricultural systems, piglets are allowed to suckle their dam for acomparably short period. Many countries practise weaning when piglets are around3 weeks old. However, the length of the suckling period varies somewhat betweencountries and rearing systems. Thus, it could be summarized that piglets in modern sys-tems have access to their dam and her milk for a period ranging from 2 to 7 weeks.From the time of weaning until the weight of approximately 25 kg, piglets are referredto as weaners or weaning pigs. From then, and until slaughter, pigs are denoted fatten-ers or finishing pigs. Market weight varies over the world but is commonly around100 kg live weight. The age at slaughter varies with health status, breed, feed intensityand rearing system, but fatteners generally are around half a year when slaughtered.

The breeding stock is often primarily selected soon after birth in terms ofprospective gilts and boars. After a secondary selection at puberty, the selected giltsare mated at approximately 7 months of age. After a pregnancy period of 116 daysthey deliver their first litter at the age of 11 months. After that time they may deliverslightly more than two litters annually. In modern husbandry, sows could give birthto up to 10 litters, but they are generally replaced far earlier.

M. Katouli and P. Wallgren22

3. PHYSIOLOGY AND ANATOMY OF THE GASTROINTESTINALTRACT OF PIGS

The digestive tract of the newborn piglet is specialized on a diet comprising milk.Therefore, a high lactase activity is present in the small intestine (Hampson andKidder, 1986), at least for as long as milk comprises the dominant feed source. Otherorgans important for the digestive system, such as the pancreas (Kelly et al., 1991a),are not quite active during intensive suckling. Owing to the abrupt change from milkto cereal consumption at weaning, the lactase activity rapidly declines during thepost-weaning period (Hampson and Kidder, 1986). Instead, α-amylase is increas-ingly produced by saliva, stimulated by chewing. Further, the pancreas becomesactive at weaning, and starts to excrete pancreatic juice (Kelly et al., 1991a,b).

The sow’s milk constitutes a compact food, and the intestine (especially the largeintestine) is comparably small during the suckling period (Kelly et al., 1991a). Atweaning, the domesticated piglets are offered cereals instead of milk. As a conse-quence, the stomach and the intestine rapidly increase in size (Kelly et al., 1991a).Despite this, the ability to absorb nutrients might decrease due to a reduction in theheight of the intestinal villi during the post-weaning period (Hampson, 1986), result-ing in a decreased total area of the surface of the intestinal lumen. The diet might alsoinfluence the size of the intestine during the subsequent rearing of pigs. Fibre-rich feedsources are correlated to an enlargement of both stomach and large intestine (Anugwaet al., 1989). The latter is rather expected, because fermentation as well as absorptionof electrolytes and fluids takes place in the large intestine. However, a lower capacityto absorb water during the first 2 weeks following weaning makes recently weanedpiglets vulnerable to loss of fluid from the intestine (van Beers-Schreurs et al., 1998),and may possibly contribute to outbreaks of post-weaning diarrhoea (see below).

4. DIET REGIMES AND ALTERATIONS OF FOOD COMPOSITIONS DURING THE GROWTH OF PIGS

It is of decisive importance that the newborn piglet consumes colostrum, not only toget energy, but also to obtain immunoglobulins, since the porcine placenta does notallow transfer of passive immunity from the sow. Therefore, the intestine of thepiglet allows digestion of macromolecules during the first 24−36 h of life. At far-rowing, the colostrum comprises around 160 g (16%) protein per litre, which rap-idly declines. Twenty-four hours later the protein content of the milk is around 6%.During the first day post-farrowing the lactose content increases from 3 to 5%. Incontrast, the fat content is rather stable around 5.5−6.5% (Klobasa et al., 1987).

The young piglet is continuously dependent on milk until weaning. The compo-sition of the milk varies somewhat over the suckling period, but generally comprises5.0−6.5% protein, 5.5−6.5% lactose and 5.5−6.5% fat (Klobasa et al., 1987). Themilk also contains IgA, which may protect the piglet from enteric diseases by act-ing locally in the gut.

Intestinal microflora in the growing pig 23

The sow eagerly offers her milk to the litter during their first week of life.However, sucking milk is generally initiated by the constantly hungry offspringfrom the second week of life onwards (Algers, 1993). To protect herself from catab-olism, the wild sow copes with this situation by avoiding contact with her litter during a great part of the day. Thereby, wild piglets are weaned in a gradual process.The access to milk will continuously be reduced in comparison to the energyrequired, and the piglets are forced to successively search for alternative energysources. The final weaning takes place at around 16 weeks of age (Jensen andRecen, 1985), and at that age the piglets are well adapted to other foodstuffs thanmilk. Further, they are well developed with respect to immune functions at that age(Joling et al., 1994).

A domesticated sow shares pen with her offspring during the suckling period. Sheis thereby denied the ability to shun the litter and, as piglets prefer milk as a sourcefor energy, she will be intensively suckled. In order to protect domesticated sowsfrom their hungry brood, piglets in modern agricultural systems are weaned between2 and 7 weeks of age. However, the early weaning system is chiefly employed toimprove production, i.e. to increase the number of piglets produced per sow per year.Generally, this weaning is effected by removal of the dam from the offspring. As aconsequence, the domesticated piglet will experience weaning at an unexpected pointof time. There is a potential risk to develop disease at weaning due to:1. an abrupt change of the feed composition where the diet is switched from

milk-based to solid-based feed, mainly cereals. This change also includes asudden withdrawal of the protective IgA that is also present in the milk;

2. a poorly developed immune system. In this context it is relevant to point outthat piglets aged 5−6 weeks are not fully developed with respect to immunefunctions, and that piglets aged 2−3 weeks are even more immature (Wallgren et al., 1998);

3. the social alterations at weaning, which contribute to a long-lasting unpleasantsituation for the piglet at weaning.To prevent disturbances at weaning (and at other occasions), so-called growth

promoters have generally been added to the feed of growing pigs for decades. The term “growth promoter” in this context refers to low dose administrations of antimicrobials (i.e. antibiotics or chemotherapeutics). Recently such a routineadministration of antimicrobials to animal feed has been questioned, both from ethicaland from ecological and medical perspectives. For instance, a ban for routine in-feedmedication was effected in Sweden during 1986 (Swedish statute-book; SFS 1985:295, Stockholm, Sweden). According to that act, antimicrobials may only be incor-porated in animal feed for the purpose of preventing, alleviating or curing disease, i.e.not for growth or yield promoting purposes. The European Communities (EC) have followed this example regarding 8 out of 12 permitted substances during 1999 (Councildirective 70/524/EEC on Feed additives, EC, Brussels, Belgium), and the remainingsubstances are to be discussed (COM 2002, 153: final, EC, Brussels, Belgium).


In this chapter it is assumed that antibiotics are not added to the food for growthpromotion.

The pigs will never again experience such a dramatic alteration of feeding habitsas they do at weaning. From that time they are offered feed based on cereals. Thecereal-based feed may be supplemented with protein-rich sources, such as fishmealand soybeans. Also bone meal and meat meal have been used as a protein source.However, the present discussion concerning transmissible spongiforme encephalitis(TSE) makes the future of the abattoir waste as protein source for meat producinganimals less clear. High protein levels in the feedstuff are known to stimulate growth.However, protein may also provoke the enteric flora, which might lead to diarrhoea(Newport, 1980; Shone et al., 1988; van der Peet-Schwering and van der Binnendijk,2000). Predigestion of proteins, for instance casein, is proven to decrease the risk ofdeveloping diarrhoea (Miller et al., 1984). Aiming to reduce feed provocation with-out reducing the growth, feed proteins can therefore, to some extent, be substitutedwith pure amino acids (Inborr and Suomi, 1988). In spite of this, proteins will alwaysbe an important source of nitrogen because purified amino acids are expensive.

Historically the pigs’ feed has been served as a dry feed, and this is still the mostprevalent feed for weaners. Liquid feeds on the other hand, are becoming more popular. They were initially introduced aiming to get rid of whey at cheese produc-tion. However, as the access to whey is limited, liquid feeds based on water arebeing used extensively. To avoid uncontrolled growth of bacteria in liquid feeds,their pH should be below 4.

5. INDIGENOUS INTESTINAL MICROFLORA OF PIGS AND ITS IMPORTANCE

The gastrointestinal (GI) tract of pigs is a dynamic ecosystem consisting ofmicrobes that colonize the gut and become established in the intestine (indigenousor autochthonous) and those that are simply passing through (transient or allochtho-nous). The normal microflora (also known as normal microbiota) develops as aresult of the influence of the intestinal ecophysiology, and the interaction betweenthe microorganisms that colonize the gut (Drasar and Barrow, 1985). It is believedthat the initial inoculum is usually derived from the sow at the time of birth (Drasarand Hill, 1974; Savage, 1977). The climax flora is different in different animalspecies and alters as the host ages.

The predominant microorganisms are anaerobes, which require special cultivationtechniques, involving rigorous exclusion of oxygen. In this habitat, the anaerobicbacteria outnumber the aerobes by a factor of at least 3 to 5 log10. The obligate andfacultative anaerobic bacteria are of diverse genera and range over a wide spectrumof taxonomic species. Implantation of bacteria in the GI-tract occurs by an elaborateprocess of ecological succession in which the composition of microflora constantlychanges (table 1). Organisms which dominate the intestine early in this process, are


suppressed by other groups of microorganisms, which are in turn suppressed by newgroups and so forth until a balanced ecosystem is established dominated by anaero-bic species (Lee and Gemmell, 1972; Savage, 1977; Varel and Pond, 1985). Thisorderly ecological succession makes the pig’s intestine a complex milieu of mixedbacterial populations, which is suggested to contain between 400 and 500 species ofbacteria (Moore and Holdeman, 1974; Drasar and Barrow, 1985). The populationsize of each bacterial species is regulated by the whole ecosystem. Various micro-organisms are completely eliminated from the digestive tract as a result of this effect.

5.1. Interference and protection

The indigenous intestinal microflora is known to be of substantial benefit to the host(Hentages et al., 1985; Wilson and Freter, 1986; Wilson et al., 1988). It is generallyagreed that this flora serves as one of the major defence mechanisms that protects thehost’s body against colonization by invading bacteria, an effect which is referred toas “colonization resistance” (van der Waaij, 1979; Finegold et al., 1983; Tancrede,1992; Rolfe, 1997). This effect is postulated to be due to the competition for attach-ment sites, nutrients and production of antimicrobial substances such as bacteriocins,defensins and volatile fatty acids. This function of the flora, although of great impor-tance to the host, has yet to be fully exploited in veterinary practice.

The pathogenic bacteria may colonize the host either by expressing specific bacterial virulence factors which may overcome the colonization resistance, or bytaking advantage of an already reduced colonization resistance, such as that inducedby antibiotic treatment. For instance, it has been shown that the normal flora is suppressed during antibiotic treatment and that this suppression is often correlatedto simultaneous colonization and overgrowth of potentially pathogenic bacteria(Gorbach et al., 1987; Tannock, 1995). Both bacterial interactions and host defencemechanisms are important weapons against colonization by pathogenic bacteria.

Data from experimental models reinforce conclusions about the efficacy of usingeven some members of normal flora as biotherapeutic agents (probiotics) (Axelsson


Table 1. The density* (log10 /g fresh weight contents) of microorganisms in various sections of thegastrointestinal tract of pigs

Stomach Duodenum Ileum Caecum Rectum

Lactobacilli 7−8 6–7 7−8 8−9 6−9Coliforms 5−6 4−5 6−7 7−8 6−8Enterococci 0−7 0−6 3−8 4−8 5−8Cl. perfringens Nil Nil 0−7 5−6 0−6Bacteroides Nil Nil 0−7 5−8 5−10Total anaerobes 5−6 5−8 7−9 9−11 9−10Yeast 0−7 0−7 0−7 5−7 5−7

*The density of bacteria may alter with age. For details, see the references.

et al., 1989; Gorbach, 1990; Fuller, 1992). For instance, some members of lacto-bacilli have been successfully used as a preventive measure against colonization ofpathogens (Gorbach et al., 1987; Lidbeck and Nord, 1993; Salminen and Arvilommi,2001). It should be noted, however, that not all members of the intestinal microfloraare beneficial to the host.

6. METHODS OF ANALYSING INTESTINAL MICROFLORA

Under normal conditions, with an intact immune system and normal ecology of thegut, a high diversity of bacterial species is observed in the intestine. Most of thesebacteria are permanent inhabitants of the GI-tract. They are mostly strict anaerobesand difficult to cultivate. Traditional cultivation techniques separate about 30−40species from any individual (Moore and Holdeman, 1974; Drasar and Barrow,1985). Yet, owing to the complexity of this flora, not all of them can be fully inves-tigated. In this section we will briefly evaluate some of the available quantitative andqualitative methods for sampling in order to understand the present limitations foranalysing the intestinal microfloras in animals.

6.1. Quantitative analyses

Of the various techniques used to study the intestinal microflora, most deal with theinvestigation of physiological capabilities of specific species of bacteria (Lee andGemmell, 1972; Moore and Holdeman, 1974; Daniel et al., 1987). These techniquesrequire initial isolation steps, and do not represent the entire intestinal bacterialflora. The total microscopic count of faecal samples together with viable counts ofthe numerically most important groups of microorganisms, may suffice for somesamples of the intestinal contents. Such procedures, apart from problems of dilutingfaecal samples under a reduced condition (Meynell and Meynell, 1970), require anumber of selective media. A wide range of selective media has been used to estimate the number of easily recognized groups of intestinal microflora such as coliforms, staphylococci, streptococci, yeast and lactobacilli. For strict anaerobessuch as Bacteroides, Fusobacterium, Clostridium, Eubacterium, etc., highly enrichedmedia containing antibiotics such as neomycin, kanamycin and/or vancomycin toprevent growth of Gram-negative facultative anaerobes are used. It should benoticed, however, that these media are not always highly selective and lactobacillifrequently grow well on them, especially when strict anaerobes are present in smallnumbers (Drasar and Barrow, 1985). Some of these anaerobes are extremely sensi-tive to oxygen, dying within 10 min after exposure to air, which adds to the techni-cal problems in culturing the intestinal microflora. In addition, the pure culturecondition is not a natural state of bacteria in a community, and characterization ofbacteria chosen for study under these circumstances may not be of great ecologicalimportance.


6.2. Qualitative analyses

Qualitative analysis of the intestinal flora has also been used to estimate microbialactivity of the GI-tract. For instance, in vitro systems have been used to assess the fer-mentation capacity of colonic microflora by measuring the ability of this complexecosystem to metabolize specific carbohydrate(s) (Ehle et al., 1982; Edwards et al.,1985; McBurney et al., 1985; Wyatt and Horn, 1988), as well as the metabolitesevolved by the sugar fermentation including the gas (Clarke, 1977; Smith and Bryant,1979; Cummings, 1984; Ross and Shaffer, 1989). Since studying all microbial typespresent in the GI-tract is virtually impossible, a convenient way would be to study themetabolic activities of all or selected groups of bacteria by assessing, in vitro, theircapacity to metabolize a number of substrates (Clarke, 1977). Again, owing to thecomplexity of the intestinal flora, study of the metabolic activities would be facili-tated if rapid and multiple assay methods were used (Katouli et al., 1997a).

Still, one complementary way to study a complex flora is to investigate what themicrobes have done during their presence in the gut. Over the years, long series ofbiochemical and microbial transformation processes have been studied in materialsfrom germ-free animals and their conventional counterparts (Midtvedt, 1999; Norinand Midtvedt, 2000). As a result, a complementary way to study the metaboliccapacity of the intestinal microflora has been established to evaluate what themicrobes can do and/or what the microbes have done. With a slight travesty of theterms initially used by Claude Bernhard, the mammalian organisms, or the host sideof the ecosystem, can be defined as Milieu interior (MI), and the non-host side asthe Milieu exterior (ME). MI plus ME together are referred to as Milieu total (MT)(Midtvedt, 1985). A simple equation of MT minus MI gives ME or “what themicrobes have done”. The approach for such studies is investigating mammals with-out any normal microflora, i.e. germ-free animals, thereby establishing the functionsof the microorganisms per se. When various microorganisms are associated withthese animals, their influence on host-derived structures and functions can easilybe studied. These findings have been described as germ-free animal characteristics(GAC) and microflora-associated characteristics (MAC) (Norin and Midtvedt,2000). MAC is defined as the recording of any anatomical structure, or physiologi-cal or biochemical function in an animal that has been influenced by the microflora.When the microbes that actually influence the parameter under study are absent, asin germ-free animals, this particular recording is defined as GAC.

6.3. New methods to analyse the intestinal microflora

6.3.1. Application of nucleotide probes

The use of molecular probes to characterize the intestinal microflora has recentlybeen the centre of attention by many investigators. Using the most refined molecu-lar methods together with the cultural-based methods to describe the natural


communities of the gut, have clearly shown the extent of the unknown microbialdiversity of the gut (Raskin et al., 1995). These methods are mainly based on the useof oligonucleotide probes complementary to conserved tracts of the 16S rRNA ofphylogenetically defined groups of bacteria.

Using 11 DNA oligonucleotide probes targeting the small sub-unit rRNA of majormicrobial groups, Lin and co-workers (Lin et al., 1997) have successfully quantifiedseveral phylogenetically defined groups of methanogens and sulphate-reducing bacteria of the GI-tracts of various domestic animals. This technique has also beenused to assess and analyse fibre-digesting bacteria of the gut (Stahl et al., 1988; Lin et al., 1994; Lin and Stahl, 1995).

Apart from the high specificity and accuracy, another advantage of these methodsfor detecting defined groups of bacteria is that the faecal samples can be frozen ondry ice and stored at −80°C immediately after sampling until they are processed. Oneshould, however, realize that not all laboratories have equipment to utilize rRNAtechniques since the probes should be synthesized, purified by high-performance liquid chromatography (HPLC) and labelled. Besides, high numbers of probes arerequired to fully quantify different microbial groups of the gut.

6.3.2. Metabolic fingerprinting

Competition for nutrients in a mixed bacterial population depends, to a great extent,on the population size and degree of affinities of each bacterial species to the avail-able substrates. Two similar bacterial populations normally yield similar patterns ofmetabolic activities upon utilization of similar substrates. Any changes in the popu-lation size or type of bacteria in a sample would be reflected in the overall metabolicfingerprint of that population. Therefore, measuring the metabolic activities of abacterial population will not only yield the metabolic potential of that flora, but willalso help to identify changes in functional status of that flora. Characterization ofcertain microbial populations of the gut on the basis of their metabolic activities hasalso been used to define the effect of environmental factors or nutritional status onthe natural structure of microbial populations (Rowe et al., 1979; Edwards et al.,1985; McBurney et al., 1985; MacFarlande et al., 1992). Changes in the pattern ofsubstrate utilization have then been correlated to the environmental parameters thatregulate microbial populations/communities.

Katouli et al. (1997a) evaluated a microplate-based fingerprinting system (PhPlatesystem) for characterizing and measuring the metabolic capacity of mixed bacterialpopulations. This system is based on interval measurements of the colour changesgenerated by an indicator caused by bacterial utilization of different sole carbonsources and production or consumption of acids in microtitre plates (Möllby et al.,1993). The bacterial strains chosen for this evaluation and their concentration in thesynthesized mixtures represented those commonly found in the colon of man andanimals (Katouli et al., 1997a). This simple approach successfully yielded metabolic


fingerprints that varied among samples. The results, however, should be interpretedwith care since these workers found that exclusion or addition of different bacterialtypes did not cause a change in the resultant function of a microbial community onsome occasions. They also examined the suitability of the PhPlates system to detectchanges in the composition and function of the intestinal microflora in pigs (Katouliet al., 1997b). The system proved to be efficient in detecting changes in the compo-sition and metabolic function of the intestinal flora of the animals during differentnutritional and pathophysiological statuses. Among the useful information that theyobtained from such biochemical fingerprints, was the capacity of a given flora to fer-ment different carbohydrates, an ability that they referred to as fermentative capacity(FC) since most tests used in their system were carbohydrates. These workers alsoconcluded that several factors might contribute to the FC-value of a given microflora.A flora with numerous but similar bacterial strains normally yields higher FC-valuesthan a flora with fewer strains. On the other hand, a flora with a few but metaboli-cally more active strains, capable of fermenting a vast number of carbon sources, alsoyields a high FC-value. The differences in types and numbers of the utilized carbo-hydrates can also be used to compare different microflora (Katouli et al., 1992).

7. IN VIVO MODELS AND SAMPLE COLLECTION STRATEGIES

Most of our present knowledge about the composition of the intestinal flora hascome from animal studies. Faeces comprise the final phase of the intestinal flora. Asthe consistency of the faeces reflects the status of the intestine, faecal samples areassumed to represent the intestinal flora. Indeed, a major problem when studying theintestinal microflora, is obtaining samples which truly represent parts of the GI-tractthat are normally inaccessible. For instance, samples from gastric, small intestinaland colonic contents can only be obtained through a peroral or nasal tube, abdomi-nal surgery, excised appendices, or the use of open-ended tubes. Withdrawal of thecontents at various levels by a magnetically guided tube has also been used (Wilson,1974), but this method only affords a sample of the organisms that are free in thelumen. Therefore, microorganisms that are attached to the villi or other parts of thesurface, which often are present in large numbers may be left out.

Samples from different parts of the intestinal content can be obtained by remov-ing the relative portion of the alimentary tract while the animal is anaesthetized orin abattoirs and immediately after the animal is slaughtered. The latter has been usedextensively for analysis of the caecum and colon contents of the rumen (Stewart andBryant, 1988; Lin et al., 1997). A way to scrutinize the enteric bacterial populationsin vivo would be to surgically insert cannulas at strategic spots of the intestine, andto collect samples via these fistulas. Surgical insertions of cannulas have previouslybeen used in pigs, mainly to study the utilization of feed (Sauer et al., 1983;Rainbard et al., 1984; Johansen and Bach Knudsen, 1994). One location often usedhas been the ileo-caecal ostium, representing the transition from the small to thelarge intestine (van Leeuwen et al., 1991). An advantage of using this method is that


a fistula can remain in place for a long period, and courses of events can be followedin vivo at correct spots of the intestine. The surgical insertion of such a fistula at theileo-caecal ostium has recently been shown not to affect the intestinal coliform floraby itself, not even close to the surgery (Högberg et al., 2001).

For obvious ethical reasons associated with these routes of sampling describedabove, many investigators prefer to analyse bacterial flora found in faecal samples orto collect rectal samples. Although the rectal bacterial flora differ from those locatedin anterior parts of the intestine (Zoric et al., 2001), they share certain properties. Forinstance, the diversity among coliform populations collected from different sites ofthe intestinal tract in healthy pigs, has been shown to be equally high (Zoric et al.,2001). In the pig, the ampulla of the rectum generally contains ingesta that can easilybe collected by swabs or specula. However, in newborn piglets, collection of rectalsamples may be obstructed because the anus is small and the ampulla may be emptyfor long periods during the first week of life.

8. IN VITRO MODELS

Development of multistage reactors to simulate the gastrointestinal microbial ecosys-tems has opened a new window to investigate the fermentation fluxes and products(e.g. volatile fatty acids, enzymatic activities and head space gases) of this complexsystem. These reactors are normally designed to simulate both the small and largeintestine. For instance, Molly and co-workers (1994) have developed a reactor, inwhich the small intestine is simulated by a two-step “fill and draw” chamber and the large intestine by a three-step reactor. These workers have used this systemto compare the composition and activity of microbial flora grown under various concentrations and combinations of carbon sources such as arabinogalacton, xylan,pectin, dextrin and starch with those described in the literature. The supply of differ-ent media or enzymes at each stage of the reactor, to support microbial communitiesresembling those of the GI-tract, is an additional advantage of such a system.

Construction of such bioreactors to simulate the GI-tract may be of high valuefor monitoring microbial community structures during biological processes. Thesein vitro models may be used for comparisons of microbial population changes over time, and for assessing the diversity of microbial communities under certainconditions. However, the input of host-derived substances and osmotic conditionsand redox-potential differences are very difficult to mimic within these systems.

9. MICROFLORA OF DIFFERENT REGIONS OF THE ALIMENTARY TRACT OF PIGS

Development of the intestinal flora in pigs takes place through an ecologicalprocess. During this process of succession, organisms which are dominant at theearly stage of life, are suppressed by other groups of microorganisms, which arein turn also suppressed and so forth. This process will continue until a stable and


complex flora dominated by anaerobic species is established in the gut.Unfortunately, due to the complexity of this flora, it is impossible to measure bothquantitatively and qualitatively all types of microorganisms present and this imposesa great restriction in assessing changes in the composition of the intestinal flora ofthe animal at any given time.

Since the beginning of the 20th century, an increasing number of investigatorshave been engaged in studying the intestinal microorganisms in pigs. Many of thesestudies comprise extensive quantitative and qualitative analyses of the intestinal floraof conventional pigs at varying ages. The normal flora most studied includeEscherichia coli and other coliform bacteria, streptococci, and lactobacilli(Komarew, 1940; Quinn et al., 1953a,b; Briggs et al., 1954) and, in some cases, thetotal number of aerobic and anaerobic bacteria and yeasts (Willingale and Briggs,1955; Horvath, 1957; Wilbur, 1959). Some later studies also report findings ofBacteroides and Veillonella (Smith and Crabb, 1961; Smith and Jones, 1963; Smith,1965a) and Clostridium perfringens (Månsson and Olsson, 1961; Van der Heyde andHenderickx, 1964). Some of these workers even compared the bacterial flora in fae-ces with that in the caecum (Briggs et al., 1954) or observed variation in the totalfaecal counts between individual pigs and between days for one animal. In the caseof caecal samples, it appeared that the variations in the total count were small and ofthe same order as in faeces. Using more selective media and rigorous techniques toexclude oxygen, Kovacs et al. (1972) investigated variation in microflora of differentgut segments of pigs. These workers found that the bacteriological status of the stom-ach, and small and large intestines, is strongly contrasted, as would be expected fromthe anatomical and physiological differences of these functional units. Among thefour segments of the small intestine studied, the duodenum contained reduced num-bers of all bacterial flora studied (except coliforms) compared to the stomach. Theseworkers suggested that the inhibitory factors operative in the duodenum affectcoliforms1 the least, compared to other groups such as streptococci, lactobacilli andclostridia. On the basis of these and many other detailed studies it has been concludedthat certain groups of bacteria such as E. coli, streptococci and clostridia are amongthe early groups of microorganisms that colonize the stomach within few hours afterbirth; all obtained from the dam and the immediate surroundings (Savage, 1977;Ducluzeau, 1983; Drasar and Barrow, 1985). Using a biochemical fingerprintingmethod to measure the stability and diversity of coliforms and enterococcal flora ofrearing pigs, Kühn et al. (1995) established a population similarity model and usedthis to measure similarity among the coliform and enterococcal populations of pigletsin a litter and their dam. These workers found that all studied piglets acquired adiverse coliform and enterococci bacterial flora during the first day of life, which,although common among the piglets, was different from that of their sow. Both theenterococcal and coliform floras from different piglets were more similar to each


1 The term “coliforms” is used to avoid incorrect citation of the literature (see box).

other during the suckling period than after weaning. These workers also found thatthe enterococcal flora in the piglets was more persistent than the coliforms.

Pigs are continuously exposed to microorganisms from their surrounding environ-ment. Microorganisms that pass through the GI-tract (transient microflora) may befound in the luminal contents or in faeces (Savage, 1977). However, it is highly likelythat most of them are eradicated by host factors such as acids in the stomach, or bilein the upper small intestine. In contrast, resident microflora represent microorganismsthat colonize different regions of the GI-tract. The type and number of these organismsin these regions are highly variable. For instance, Lactobacillus was shown to repre-sent 67% of the bacterial population of the non-secreting stomach region in healthyunweaned pigs (McGillivery and Cranwell, 1992). Their level ranges between 107

and 108 CFU per gram caecal content in suckling piglets (Jonsson, 1986). TheLactobacillus species that are most frequently isolated from the stomach, intestine andfaeces of healthy piglets are L. fermentum (Fuller et al., 1978), L. acidophilus and L. delbrueckii (Mäyrä-Mäkinen et al., 1983). The majority of these isolates can attachto epithelial cells of the small intestine. In addition to lactobacilli, other bacterialgroups have been isolated from the non-secreting part of the stomach. These includeStreptococcus (Fuller et al., 1978), Eubacterium, E. coli, Bifidobacterium,Staphylococcus, Clostridium and Bacteroides (McGillivery and Cranwell, 1992).However, their population size in the stomach is much smaller than in the large intes-tine (McAllister et al., 1979). The viable cells of lactobacilli are continuously beingreleased from the non-secreting region, together with desquamated epithelial cellsand thereby inoculate the stomach and intestinal luminal content (Fuller et al., 1978;Jonsson and Conway, 1992). Using a combination of protein profile analysis andcolony morphology, Henriksson et al. (1995) characterized the lactobacilli colonizingvarious regions of the porcine GI-tract and detected several different groups of lacto-bacillus. Specific groups of lactobacilli were associated with, and often unique for thestomach, jejunal, caecal, and colonic regions of the GI-tract. These workers also foundthat there were major differences between population densities of the gastric mucosalLactobacillus population of individual pigs.


The term coliforms has been traditionally used to refer to Escherichia coli (E. coli)-like bacteria(coli-form) since these bacteria could be readily isolated from faecal materials of warm-bloodedanimals. During the early 1900s, the technology was not available to easily distinguish E. coli fromother coliforms and therefore most of the coliforms recovered from human and animal faeceswere assumed to reflect the presence of E. coli. As a result, the term “coliforms” was considered tobe equivalent to E. coli. It is now known that coliform bacteria comprise of at least four generaof the family Enterobacteriaceae that can all ferment lactose. These genera are Escherichia,Klebsiella, Enterobacter and Citrobacter and collectively they represent only 1% of the total bacterial populations in human and animal faeces. Among coliforms, however, E. coli representsthe majority of the population (90−95%). During the early 1950s, although more specific tests weredeveloped to easily identify E. coli from the rest of coliforms, the use of “faecal coliforms” was socommonplace that the term was not dropped in favour of E. coli.

Digesta transferred from the stomach to the duodenum are subjected to a dramaticenvironmental change, mainly due to the introduction of host factors such as bile,enzymes and bicarbonate (Drasar and Barrow, 1985). Compared to the large intestine,this region is less densely populated by microorganisms, partly due to the high flowrate of the luminal content through the small intestine. Lactobacilli are the dominantspecies in the piglet’s small intestine, while E. coli, Clostridium, Bacteroides and oxygen tolerant anaerobes are also present in large numbers. In addition low levels(100−1000-folds) of Streptococcus, Enterococcus and Staphylococcus have beendetected on the mucosa of the small intestine (McAllister et al., 1979). The densitiesof the small intestinal microflora tend to increase in the distal part. In piglets, this is amajor site for colonization by certain diarrhoegenic E. coli strains. The large intestine,on the other hand, is the major site of microbial activities in the digestive tract of thehealthy pig, and the slowly moving digesta contains the most dense microbial popu-lation of the entire GI-tract. The large intestine includes the caecum and the colon. The pig is a monogastric herbivore with a relatively large caecum and colon.Consequently, the transient time through this region is considerable, allowing largepopulations of bacteria to be accumulated in the large intestine. Obligate anaerobesdominate the microflora of this region and increase in number from the ileum to thespiral colon (Cranwell, 1990). It is generally acknowledged that between 1011 and 1012

CFU bacteria are present per gram dry weight of the colon contents (Onderdonk,1999; Borriello, 2002). This figure for facultative anaerobes such as Bacteroides andclostridia can be as high as 109 CFU per gram (Smith, 1965b; Drasar, 1974; Salanitroet al., 1977; McAllister et al., 1979).

10. DYNAMICS OF THE INTESTINAL MICROFLORA IN HEALTHY CONVENTIONAL PIGS

Detailed identification of different bacterial species in the pig’s intestinal tract is anextremely laborious and lengthy process. For this reason, studies investigating thedynamics of the intestinal flora focus on methods that can yield information on the functional status of the intestinal flora. These methods include measuring the fermentative capacity (FC) of the microflora, which evaluates the amount of sugarmetabolized by the faecal microflora and the metabolites evolved (McBurney et al.,1985), and testing the reaction of the whole or part of the gut flora against a relativelyhigh number of substrates (Clarke, 1977). Using a combination of 48 substrates,Katouli and co-workers examined the pattern of metabolic response of the intestinalflora of healthy pigs and compared that with diseased pigs (Katouli et al., 1997b).They found that the response of the animal’s intestinal microflora to different carbo-hydrates varied among individual piglets at different sampling occasions. Similarresults have also been reported by others (Edwards et al., 1985). Despite these indi-vidual differences, the overall FC-values in most stages of animal life were similaramong piglets (Katouli et al., 1997b). These workers also showed that piglets receive


a high proportion of their intestinal microflora from their dams during the first fewdays of their life. However, despite the close contact with their dams, they developintestinal floras that are very different from the sow’s flora. This suggested that themilk-based diet in piglets would yield a flora that is different from the initial “birthflora” derived from sows. This finding is supported by the fact that microfloras ofpiglets during the suckling period show more similarity to each other than during thepost-weaning and fattening period.

The post-weaning period in piglets is associated with a dietary shift from milk tosolid food, which will be replaced with a high-energy fattening diet when piglets areallocated into fattening stables. This dietary shift will result in substitution and/orestablishment of new microflora in the pig intestine. It has been shown that theintestinal flora of pigs during the fattening period is more diverse than that of thesuckling period. This might be due to the fact that in fattening stables, pigs fromdifferent pens may be mixed together and a direct contact between adjacent pens isestablished. As a result, pigs are exposed to more diverse bacterial species (Katouliet al., 1997b). The fermentative capacity of the intestinal flora, which is normallyhigh during the suckling period, decreases during post-weaning and fattening periods, indicating that organisms dominating the pigs’ intestine very early in lifeare able to utilize more diverse carbon sources than those dominating the animalduring post-weaning and fattening periods (fig. 1).

10.1. Intestinal microflora of sows

Several studies have attempted to determine the type of bacteria that are present in theintestinal tract of sows. Such efforts, using selective plate media, have led to the enumeration of only bacterial groups such as lactobacilli, streptococci, Bacteroides, E. coli and C. perfringens (Rall et al., 1970; Terada et al., 1976; Salnitro et al., 1977).These studies have clearly shown that Gram-negative anaerobic species of Bacteroides,


Fig. 1. Fermentative capacity (FC) - values of the whole intestinal flora of conventional pigs during thesuckling, post-weaning and fattening periods. FC - values are the mean of four pigs and their standard errors.W = Weaning. F = Pigs were transferred to the fattening stable.

Veillonella, Fusobacterium and Peptostreptococcus can be isolated from differentsegments of the intestinal tract (Aalbaek, 1972; Mitsuoka et al., 1974). Other groupsusing strict anaerobic methods have isolated streptococci, Eubacterium species,Clostridium species and Propionibacterium acnes as the predominant flora in adultpigs (Salnitro et al., 1977). Kühn and co-workers measured phenotypic diversity andstability of the intestinal coliforms (Kühn et al., 1993) and enterococcal floras (Kühnet al., 1995) in piglets during their first 3 and 5 months of age, respectively, and compared the results with those of their sows. They found that the diversity of bacterial flora of the sows was higher than in most of the piglets during the first weekof the pig’s life. In a more comprehensive study, Katouli et al. (1997b) investigatedsimilarities between biochemical fingerprints of the whole intestinal microflora ofsows and their offspring. They found that the sow’s flora had a considerably lowerFC-value than those of the piglets at the time of birth, which remained so over theentire suckling period. The fact that the bacterial floras of sows had lower FC-values(even lower than those of pigs at the end of the fattening period) suggests that the lossof fermentative capacity will continue as the animal ages (see fig. 1).

10.2. Intestinal microflora of piglets during the suckling period

As mentioned before, the piglet intestine is sterile at birth (Kenworthy and Crabb,1963). Piglets receive their initial microflora from the sow’s teats and skin as wellas maternal faeces (Arbuckle, 1968; Berschinger et al., 1988). In fact, it has beenreported that piglets eat considerable amounts of their sow’s faeces during the suck-ling period (Sansom and Gleed, 1981). Studies carried out during the early 1960s bySmith and Crabb (1961) and Kenworthy and Crabb (1963), and more recently byMelin et al. (1997) and Katouli et al. (1995), have shown that despite differences ingenetics and feeding strategies, healthy piglets reared in different environmentsdevelop a very comparable intestinal flora. During the early life when diet consistsof mainly milk and the species-specific differentiation of the intestinal tract is low,several bacterial groups increase and decrease in a similar way. For instance, Melinet al. (1997) have shown that the number of faecal coliforms, E. coli, enterococciand C. perfringens decrease over the first 9 weeks of the piglet’s life. C. perfringenseven reaches undetectable levels after 3 weeks.

During the first week of life the coliforms and enterococci in the piglets’ intestinesmay differ considerably from that of the dam, suggesting that these floras are comingfrom sources other than sows (Katouli et al., 1995; Kühn et al., 1995). However, thismay contribute very little to the overall similarity between the gut microflora of thesows and their offspring. The diversity of these floras in piglets is very high alreadyfrom the first week and remains so during the suckling period. The fact that pigletshoused together develop highly similar floras during the first week and onwards, alsoconfirms the environmental nature of these floras among litter and pen-mates (Katouliet al., 1999). The early differences among coliform and enterococci floras between


piglets and sows will gradually decrease during the suckling period so that beforeweaning these specific floras of littermates are fairly similar to each other and to thatof their dams (Melin et al., 1997; Katouli et al., 1999).

The intestinal colonization of E. coli in piglets comprises successive waves ofdifferent strains. Most of these strains are transient bacteria, the tenure of whichvaries between a few days to 2 weeks. On the other hand, most resident E. colistrains colonize the intestine of young piglets during the suckling period. Katouli et al.(1995) have shown that during this period each piglet may carry more than one typeof resident strain in their gut. In a herd or stable, several piglets may be colonizedby the same resident strains, which indicates that both strain and host specificity,especially during the suckling period, are important for colonization and persistenceof E. coli in piglets.

Comparison of metabolic fingerprints of the faecal samples from piglets and theirdams has shown that despite the difference in the E. coli floras, most members of theintestinal flora in pigs are similar to those of their dams, suggesting that sows are theinitial source of most microflora for piglets. However, it seems that despite the closecontact of piglets with their dams during the suckling period, they will eventuallydevelop floras that might not be very similar to that of their sow (Katouli et al., 1997b).

10.3. Intestinal microflora of piglets following weaning including effects of regroupings and movements

In modern agricultural systems, weaning is generally achieved by abruptly removingthe sow. However, these circumstances expose the piglets to a considerable amount ofstress that affects the immune system negatively (Blecha et al., 1985; Bailey et al.,1992; Hessing et al., 1995; Wattrang et al., 1998). The stress and the sudden alterationof diet also contribute to a disturbed enteric flora of the piglets during the post-weaning period (Kühn et al., 1993, 1995; Katouli et al., 1995, 1997b; Melin et al.,1997, 2000a). The diversity of the intestinal flora may decrease dramatically duringthe first 3 days post-weaning among apparently healthy piglets. Since a high micro-bial diversity of the gut is believed to protect the animals not only from intrinsicmicrobes but also from microorganisms of external origin (Pielou, 1975; Kühn et al.,1993), this points to a situation of potential danger due to a decreased colonizationresistance. It has been shown that the population size of bacteria is not altered duringthis period (Melin et al., 1997), indicating that the decreased microbial diversity mustbe achieved by proliferation of some strains. Melin et al. (1997) also showed thatthe similarity of the intestinal flora among pen-mates decreases during this period. Thisin turn points to the fact that while some strains proliferate in one pig, other strainsproliferate in other pigs. Thus, if a pig develops diarrhoea due to proliferation of apathogenic clone, there is an increased risk that also pen-mates will be diseased astheir colonization resistance is decreased due to the low diversity of the intestinal floraat the actual time. In apparently healthy pigs the enteric microflora will again stabilize


between 2 to 3 weeks after weaning. However, if invaded by pathogenic strains ofE. coli, the intestinal flora may be disturbed for an even longer period (Melin et al.,2000a).

Some pig herds practise mixing and moving of piglets at weaning. This willpotentially increase the risk of developing diarrhoea not only due to an increasedlevel of stress imposed on the piglets, but also because a larger number of pathogenicstrains (from more than one pen) have the chance of proliferating and invading thevulnerable piglets (Katouli et al., 1999). It should be noted, however, that theincreased number of strains might also increase the piglets’ colonization resistance.Therefore, the mixing practice may not always be detrimental to pigs especially if itis done under proper hygienic management. Under high hygienic and managementstandards, the disturbed normal flora will be restored in around 2−3 weeks and pigsregain a high microbial diversity and a high similarity between microbial populationswithin groups (Katouli et al., 1999). Analysis of the intestinal microflora of pigs afterweaning has shown that the post-weaning coliform populations differ from thoseof the suckling period (Katouli et al., 1997b). Despite this, the overall similaritybetween intestinal populations of pen-mates may remain high (Katouli et al., 1997b).

10.4. Intestinal microflora of pigs during the fattening period

Pigs are generally transferred from weaning facilities to the fattening enterprises atthe weight of approximately 25 kg, corresponding to an age of 10−14 weeks. Thistransfer may provoke the pigs in a similar way as weaning (Wallgren et al., 1993),and the provocations increase if the animals are transported and regrouped (Lundet al., 1998). However, pigs are more immunologically (Wallgren et al., 1998) andphysiologically mature during this transfer than at weaning. Consequently, the effecton the enteric flora because of this transfer is much less evident than at weaning(Katouli et al., 1995). If healthy, the fattening pigs show a high diversity of the entericflora throughout the fattening period (Kühn et al., 1995). However, transientmicrobes are continuously present, and the similarity of the intestinal flora betweenpigs at this stage may be considerably lower than at the suckling or post-weaningperiods (Katouli et al., 1995; Kühn et al., 1995).

As mentioned before, an alteration of the composition of the intestinal popula-tions takes place when pigs are allocated into fattening units and mixed with otherpigs. The intestinal populations may differ considerably between pigs, mainly dueto the fact that pigs are exposed to the diverse bacterial species, a situation which isnormally expected in stables of mixed pigs. The overall fermentative capacity of theflora of the animals during this period is far less than during the suckling period.This loss of fermentative capacity is a gradual process but will be accelerated dur-ing the late post-weaning period (Katouli et al., 1997b). Changes in the compositionand fermentative capacity of the intestinal flora of pigs after weaning and after allo-cation of pigs into the fattening stables, coincide with the dietary shift from milk tosolid food and further to a high-energy fattening diet (see fig. 1).


11. INTESTINAL MICROFLORA OF SPECIFICPATHOGEN FREE PIGS

Specific pathogen free (SPF) pigs are declared free from a defined number ofmicroorganisms pathogenic to pigs. However, it should be noticed that these pigs arenot reared under germ-free conditions. Diseases such as salmonellosis and swinedysentery (induced by Brachyspira hyodysenteriae) are not present, but microor-ganisms such as E. coli are in reality impossible to avoid. As feed and straw areoften of similar sources as those offered to conventional pigs, the intestinalmicroflora is virtually the same for SPF pigs as for conventional pigs. Indeed, whencomparing intestinal microflora obtained from SPF pigs with those obtained fromconventional pigs they basically share a similar composition throughout their life.On the other hand, owing to the absence of certain pathogenic microbes and pre-cautions undertaken to avoid introduction of infections, development of clinicaldiarrhoea is rarely seen in SPF herds.

We have recently studied the biochemical fingerprints and fermentative capacityof the whole and/or selected intestinal microflora of SPF pigs during weaning, post-weaning and the fattening period (Katouli et al., unpublished data). We found that, asin conventional pigs, the fermentative capacity of the SPF pigs also decreased as thepigs grew older and that there was a decrease in the fermentative capacity values ofthe intestinal flora immediately after weaning and after the pigs were transferred tothe fattening stable (fig. 2). However, we also found that both SPF piglets and theirsows had much higher FC-values than their conventional counterparts during the firstweek of life. These values, however, dropped to a level close to what we have normally obtained from conventional pigs during this period (see fig. 2).Interestingly, the FC-values of sows reached the same level as those of piglets at thetime of weaning. Similar patterns were basically observed among selected groups ofnormal flora in SPF piglets except for Lactobacillus flora. The fermentative capacity


Fig. 2. FC - values of the whole intestinal microflora of specific pathogen free (SPF) pigs during suckling,post-weaning and fattening periods. FC - values are the mean of four pigs and their standard errors. W = Weaning. F = Pigs were transferred to the fattening stable.

of this flora, which was high during the first 4 weeks of weaning, showed a dramaticdecrease just before weaning, reaching its minimum level at the time of weaning (fig. 3).

12. ALTERATIONS OF THE INTESTINAL MICROFLORA OWING TO STRESS AND DISEASE

Physiological stresses and disease especially during suckling and early post-weaning,are a major concern within piglet production, and disturbances in the composition of the intestinal microflora constitute the greatest problem (Cutler et al., 1999). It is believed that changes in the stability of the intestinal flora will result in the development of a low diversity of the flora making the animal susceptible to


Fig. 3. FC - values of streptococcal (a), coliform (b) and lactobacilli (c) populations of SPF pigs during thesuckling, post-weaning and fattening periods. W = Weaning.

gastrointestinal diseases. These factors and their effect on the health status of grow-ing pigs are discussed below.

12.1. Intestinal coliforms during the suckling and post-weaning periods

As described earlier, piglets rapidly develop a highly diverse intestinal coliform flora.Unless diarrhoea develops, this flora will remain stable until weaning and the intestinalcoliform populations of pen-mates are fairly similar, indicating a high colonizationresistance of the piglets within the pen.

Introduction of weaning, more or less, leads to a collapse of the intestinal coliformpopulation. At this time the piglets are highly vulnerable to disease and if they areexposed to a low pathogen load, because of good herd management, they may be ableto resist developing diarrhoea before the disturbed flora is completely recovered. In healthy pigs, the coliform flora will remain stable throughout the weaning period.At transfer to the fattening enterprises, a situation similar to weaning may occur.However, as the pigs are growing older and their feed composition changes less dra-matically, the intestinal coliform floras are restored faster following this allocation.

12.2. At-risk situations

The intestinal bacterial populations may be influenced by changes in the life of a pig.Consequently, all adjustments should be defined as situations that may threaten thestability of the enteric microflora. The younger the pig and the more dramatic thealteration(s), the larger the risk will be. The influence of alterations of the intestinalflora on the pig’s life has been thoroughly described earlier in this chapter. Examplesof induced at-risk situations are weaning, regrouping, transportation and alterationsin feed regiments. In addition, non-optimized management may provide some addi-tional risk situations, such as high pathogen load, chill, draught and moisture.

12.3. Diarrhoea pre-weaning

Pre-weaned piglets are frequently infected with enteropathogens at two stages: asnewborn and at the age of 2−3 weeks. In systems effectuating weaning at the age of 2−3 weeks, the latter stage coincides with the weaning and thereby could poss-ibly be referred to as post-weaning diarrhoea. Factors such as immune defence,indigenous flora, pH, food composition and environmental errors may influence thedefensive capacity of the animals at these occasions and therefore exposure to enteropathogens may be hazardous. A number of host mechanisms have evolved which protect the GI-tract from invading pathogens. E. coli, Salmonella andB. hyodysenteriae are among the globally most economically important causes ofbacterial induced diarrhoea in piglets (Bergeland and Henry, 1982; Edfors-Lilja and Wallgren, 1999; Straw et al., 1999).


The two most important pathogenic microorganisms that affect newborn pigletsare E. coli (summarized by Fairbrother, 1999) and C. perfringens (summarized byTaylor, 1999). Both these microorganisms may induce neonatal diarrhoea in largenumbers of piglets and may become fatal. However, owing to the unifactorial causeof these diseases, vaccination of sows and the subsequent transfer of their protectiveimmunity to the offspring via colostrum have effectively prevented outbreaks of dis-eases caused by these species in newborn pigs. Other microorganisms associated withdiarrhoea in neonatal animals include Bacteroides fragilis, Campylobacter spp. andYersinia enterocolitica (Holland, 1990) and a number of viruses (Straw et al., 1999).

It should be noted, however, that the maternal immunity declines with increasingage of the piglets (Saito et al., 1986; Fu et al., 1990; Wallgren et al., 1998). Therefore,diarrhoea induced by the species mentioned above may occur at the age of 2−3 weeksif the pathogen load of the environment is high enough. Furthermore, these speciesmay be found in association with large numbers of other potentially pathogenicmicrobes (summarized by Straw et al., 1999). Virulent organisms such as the proto-zoan Isospora suis as well as rotavirus and coronavirus, have frequently been corre-lated to diarrhoea in suckling piglets (Glock, 1981). The number of microbes thatcould potentially contribute to development of gastrointestinal disturbances increasesas the piglets grow. Consequently, diarrhoea among somewhat older piglets may wellreflect mixed infections, and can certainly be influenced by environmental conditionsand hygiene.

When diarrhoea is observed during the first days of life, the causative agent canoften be re-isolated in pure culture from faecal samples, and under such conditions thecorrelation between infection and signs of disease is obvious. Somewhat older pigletswill receive a rather diverse enteric flora prior to infection. Kühn and co-workers (1993)have shown that during an E. coli associated outbreak of diarrhoea, the diseased pigletshad a lower diversity of the intestinal coliforms, indicating that the pathogenic strainhad outgrown the others. While studying the diversity of coliform populations in a group of pigs, we also noticed that pigs that received antibiotic during an outbreak of diarrhoea showed a lower diversity of coliforms. This effect, however, was not seen among all piglets. We also noticed that the piglets affected with diarrhoea did not recover from the low coliform diversity until long after weaning (Katouli et al.,unpublished data) (fig. 4).

12.4. Diarrhoea post-weaning

As described above, the enteric microflora is severely disturbed following weaning,thereby paving the way for potentially pathogenic microbes. Toxin-producingstrains of E. coli (mainly serogroups O138, O139 and O141) associated withoedema disease, may act as the main sources of post-weaning diarrhoea, a diseasethat is often fatal for newly weaned piglets (Berschinger, 1999). On the other hand,it should be mentioned that experimental challenge of healthy, newly weaned piglets


with any of the above pathogens, in most cases might not result in a state of diar-rhoea. This may be due to the lack of environmental factors necessary for completedisturbance of the flora. For instance, Melin et al. (2000a) used a highly virulentstrain of E. coli to challenge a set of weaned piglets. The challenge strain belongedto serogroup O149 and carried surface antigen K88, which confer adhesion to theF4 receptor on the epithelial cells of the pigs. Furthermore, the strain was a potenttoxin producer (STa, STb and LT) and the challenged piglets all had receptors forF4 (Edfors-Lilja et al., 1995) and the pigs were proven truly infected. Still, post-weaning diarrhoea (PWD) was not achieved, indicating that although PWD isstrongly associated with E. coli, it should be considered as a multifactorial syndromerather than a specific infection. Indeed, these workers succeeded in inducing anexperimental PWD by exposing piglets to a cascade of different pathogenic strainsof E. coli (Melin et al., 2000b,c), possibly imitating conditions that could occurpost-weaning in a herd (see above).

Microbes should not be regarded as the sole cause of PWD in practical pig production. The influence of pathogenic microorganisms can be amplified by environmental stress such as chill, draught, moisture, etc. Further, insufficientmanagement may contribute to the development of PWD. The newly weaned pigis poorly developed with respect to immune functions (Blecha et al., 1985; Bailey et al., 1992; Wallgren et al., 1998; Wattrang et al., 1998) and therefore vulnerable toinfections.

Pigs affected by PWD will express a decreased diversity of the intestinal floraduring the course of the disease owing to the overgrowth of one or several bacterialstrains (Melin et al., 2000c). Because of the influence of the strain(s) causing disease, the similarity between intestinal coliform populations of diseased pigs maybe larger than between apparently healthy pigs. However, this type of similarity does


Fig. 4. A representative figure showing the effect of sulfametoxasol/trimethoprim (administered during an outbreak of diarrhoea) on the diversity of coliforms in conventional pigs. Four pigs (P1 to P4) from fourlitters were studied. Diversity of coliforms was measured as Simpson’s index of diversity after testing randomly 40 coliforms from MacConkey plates. W = Weaning. DO = Onset of the diarrhoeal outbreak in the herd. F = Pigs were transferred to a fattening stable.

not indicate any colonization resistance, as seen among suckling pigs (Katouli et al.,1999). Instead, it is achieved by infection of several individuals by the same patho-genic strain(s).

The balance of the intestinal microflora may be severely affected for as long as4 weeks following an infection with coliforms at weaning, regardless of whetherclinical PWD has been developed or not (Melin et al., 2000a). Of course, this mayfacilitate infections with other pathogenic microorganisms, such as Brachyspiraspecies or Salmonella.

Taken together, the weaning is a critical physiological period for the pig. It isoften accompanied by an abrupt multiplication of some strains of E. coli in thedigestive tract and may result in development of diarrhoea and/or oedema disease.To avoid PWD, good management should be applied.

13. PRECAUTIONS AIMING AT STABILIZING THE INTESTINAL MICROFLORA

The negative influence of environmental disturbances and infections on the intestinalflora could possibly be reduced. Different strategies are discussed briefly below.

13.1. Feed composition

The food itself can be a provoking factor in causing disease and/or disturbance of thegut flora of the pigs. As an optimized growth of pigs is of economical importance, feedconsumption and feed utilization are of great importance in modern pig husbandry.Pig feed is often processed, i.e. pre-heated, and the digestive ability of the food is facilitated. However, this normally leads to a reduced chewing and a shorter residenceperiod of food in the stomach. As a consequence, the digestive and bactericidal effectsof saliva and hydrochloric acid will be reduced and disturbances in the digestive tractmay be facilitated. By avoiding pre-heating of dry feed, the natural protective effect ofintestinal flora to infection will increase. An increased amount of fibre in the diet willfurther stimulate the natural protection towards disease owing to a slower passagethrough the gut (Heidelberg et al., 1984; Hampson and Kidder, 1986).

In a situation of increased risk, the food composition may have a big impact onthe clinical outcome with respect to enteric health in a herd. For instance, the provo-cation of the abrupt change of food at weaning may be minimized by adding lactoseto the food, thereby resembling the milk from the sow to some extent. In this context it should be mentioned that commercially available milk substitutes gener-ally emanate from either cow milk, that will include proteins from foreign species,or even from soya.

Protein, which is required to stimulate the growth of pigs, may also affect thecomposition of the enteric flora, leading to diarrhoea (Newport, 1980; Shone et al.,1988; van der Peet-Schwering and van der Binnendijk, 2000). In fact, some protein


sources, such as soya have actually been linked to outbreaks of diarrhoea (Jager et al., 1986; Nabuurs, 1986). Predigestion of proteins is proven to decrease the riskof developing diarrhoea (Miller et al., 1984) and feed proteins can therefore to someextent be substituted with pure amino acids (Innborr and Suomi, 1988). However, aspurified amino acids are expensive, the proteins themselves form the most importantsource of nitrogen in pig feed.

13.2. Antibiotics and feed additives

Antibiotics have commonly been used to control the enteric health and improve thegrowth rate. This is certainly achieved by suppression of the intestinal bacterialactivity. Animals given antibiotics in the feed generally perform better than thoseoffered probiotics (Eidelsburger et al., 1992). However, it should be rememberedthat intestinal treatments with antibiotics may affect the normal flora severely(Barza et al., 1987; Thijm and van der Waaij, 1979; Hashimoto et al., 1996; Wilsonet al., 1996). Further, a continuous use of antibiotics will increase the risk of devel-opment of bacterial resistance to antibiotics used (Linton et al., 1988; Aarestrup,2000). Consequently, a permanent use of antimicrobial agents in feed ought to beavoided and replaced with proper management systems and well-designed feed thatmaintain and stabilize the intestinal flora. As a result, development of diseaseswould be minimized and the number of medical treatments would be reduced.

13.3. Zinc oxide supplementation of the feed

Zinc is an essential component of several enzyme systems and plays an importantrole in stabilizing membrane integrity. As epithelial cells are the first line of defenceagainst microbial invasion, zinc has a special role in resistance to infections. Zinc in the form of zinc oxide (ZnO) has been successfully used to prevent outbreaks ofPWD (Holm, 1988; Holmgren, 1994). By adding a high concentration of the ZnOto the feed, it has been possible to preserve the integrity of the coliform populationin weaned pigs (Melin et al., 1996; Katouli et al., 1999). This may partly explain theprotective effect of the ZnO against post-weaning diarrhoea as the colonizationresistance of the gut flora is preserved. No similar effect can be achieved if an equalamount of zinc is given parenterally (Shell and Korneay, 1994). This calls for a localeffect of the ZnO in the intestine and, since a high concentration of zinc is required,it is possibly toxic. Piglets given ZnO-supplemented feed may grow faster than non-treated piglets close to weaning. However, a continuous feeding of high amounts ofZnO in the food should be avoided, because pigs that were offered a feed with2500 ppm ZnO for 4 weeks expressed signs of intoxication (Jensen-Waern et al.,1998). Further, as most of the zinc oxide will pass through the pig’s intestine, the envi-ronmental aspects must be considered. Katouli et al. (1999) found that loss of diver-sity and disruption of the integrity of coliform flora in weaned piglets supplemented


with ZnO in their feed could be restored within 14 days post-weaning. On the basisof this finding, these workers concluded that feeds supplemented with ZnO shouldbe restricted to only 2 weeks post-weaning in veterinary practice.

13.4. Probiotics

Using probiotics to stimulate the intestinal flora has been tempting, mainly due tothe facts that probiotics can be used to improve colonization resistance of the intes-tinal flora and thereby potentially reduce the dependence on antibiotics in order to prevent and/or treat bacterial infection of the gut in animals (Kyriakis, 1989;Kyriakis et al., 1999). Studies on the suitability of the members of the intestinal florahave suggested potential candidates such as Lactobacillus species (Toit et al., 1998),and Bacillus cereus (Kirchgessner et al., 1993) as probiotics of microbial origin.Acidifiers have also been suggested and used as probiotics. These include fumaricacid, hydrochloric acid and sodium formate (Eidelsburger et al., 1992).

The existing reports on the use of probiotics in pigs range from positive effectson enteric health and weight gain (Eidelsburger et al., 1992) to no effects at all(McLeese et al., 1992). Also increased weight gains have been reported without anyvisible positive effects with respect to intestinal health (Eidelsburger et al., 1992;Kirchgessner et al., 1993). Presently, much work is focused on probiotics. However,the variations in the results obtained may indicate an influence of the managementand environmental conditions. Therefore, great efforts should be made to scrutinizethe effects of probiotics under unbiased conditions. The effect of probiotics on thehealth and well being of animals is discussed elsewhere in this book.


The control of diarrhoeal diseases still presents a challenge in pig husbandry. Recentdevelopments in management and production facilities, as well as availability ofpotential vaccines, has reduced mortality associated with diarrhoea in piglets.Changes in the composition and stability of the intestinal flora of piglets have beenshown to play an important role in the development of diarrhoea during the sucklingand early post-weaning periods. Factors such as stress, especially at weaning andearly post-weaning periods, are among the main causes of disruption to the integrityof the intestinal flora. Approaches to challenge enteric diseases should includeestablishing diverse intestinal floras in piglets during the suckling period and main-taining the stability and diversity of this flora after weaning. While several methods,including molecular-based techniques, are available to detect and identify uncultur-able bacterial flora of the gut, there is a need for more advanced techniques tomeasure the functional status of the normal flora in response to dietary feed andenvironmental stress. The recent practice of withdrawing growth promoters in pigsin some countries should be monitored with respect to the composition of the gut


flora and the development of enteric disease. Furthermore, there is a growing interest in ecological agricultural practice and the farming of pigs outdoors. Thismay have a significant impact on the composition of the intestinal flora, which maydiffer from that of the traditional indoor microflora. Application of new methodsalone or in combination with classical methods can be used to identify the stabilityand the impact of the outdoor flora on the general health of pigs.

REFERENCES

Aalbaek, B., 1972. Gram-negative anaerobes in the intestinal flora of pigs. Acta Vet. Scand. 13, 228−237.Aarestrup, F.M., 2000. Occurrence, selection and spread of resistance to antimicrobial agents used for

growth promotion for food animals in Denmark. APMIS, Suppl. 101, 1−48.Algers, B., 1993. Nursing in pigs: communicating needs and distribution resources. J. Anim. Sci. 71,

2826−2831.Anugwa, F.O.I., Varel, V.H., Dickson, J.S., Pond, W.G., Krook, L.P., 1989. Effects of dietary fiber and

protein concentration on growth, feed efficiency, visceral organ weights and large intestine microbialpopulations of swine. J. Nutr. 6, 879−886.

Arbuckle, J.B.R., 1968. The distribution of certain Escherichia coli strains in pigs and their environment.J. Med. Microbiol. 3, 333−340.

Axelsson, L.T., Chung, T.C., Dobrogosz, D.J., Lindgren, S.E., 1989. Production of a broad spectrumantimicrobial substance by Lactobacillus reuteri. Microbial Ecol. Health Dis. 2, 131−136.

Bailey, M., Clarke, C.J., Wilson, A.D., Williams, N.A., Stokes, C.R., 1992. Depressed potential for inter-leukin-2 production following early weaning of piglets. Vet. Immunol. Immunopath. 34, 197−207.

Barza, M., Giuliano, M., Jacobus, N.V., Gorbach, S.L., 1987. Effect of broad spectrum antibiotics on col-onization resistance of intestinal microflora of humans. Antimicrob. Agents Chemother. 31, 723−727.

Benini, A., Bertazzoni Minelli, E., Vesetini, S., Marchiri, L., Nardo, G., 1992. Intestinal ecosystem andmetabolic capacity in women with breast cancer. Pharmacol. Res. 25, 184−185.

Bergeland, M.E., Henry, S.C., 1982. Infectious diarrheas of young piglets. Vet. Clin. North Amer. LargeAnim. Pract. 4, 389−399.

Berschinger, H.W., Eng, V., Wegmann, P., 1988. Relationship between coliform contamination of floorand teats and the incidence of puerperal mastitis in two types of farrowing accommodations. In:Proceeding of the 6th International Congress on Animal Hygiene. Swedish University of AgriculturalScience, Skara, Sweden, pp. 86–88.

Berschinger, H.U., 1999. Postweaning Escherichia coli diarrhea and edema disease. In: Straw, B.E.,D’Allaire, S., Mengelin, W.L., Taylor, D.J. (Eds.), Diseases of Swine, 8th edition. Iowa StateUniversity Press, Ames, IA, pp. 441−454.

Blecha, F., Pollman, D.S., Nichols, D.A., 1985. Weaning pigs at an early age decreases cellular immu-nity. J. Anim. Sci. 52, 396−400.

Borriello, S.P., 2002. The normal flora of the gastrointestinal tract. In: Hart, A.L., Stagg, A.J., Graffner, H.,Glise, H., Falk, P., Kamm, M.A. (Eds.), Gut Ecology. Martin Dunitz Ltd, London, pp. 3−12.

Briggs, C.A.E., Willingale, J.M., Braude, R., Mitchell, K.G., 1954. The normal intestinal flora of the pig.I. Bacteriological methods for quantitative studies. Vet. Rec. 66, 241−242.

Clarke, R.T.J., 1977. Methods for studying gut microbes. In: Clarke, R.T.J., Bauchop, T. (Eds.),Microbial Ecology of the Gut. New York, Academic Press, pp. 1−33.

Cranwell, P.D., 1990. Development of the gastrointestinal microflora in the milk-fed human infant.Thesis, LaTrobe University, School of Agriculture, Melbourne, Australia.

Cummings, J.H., 1984. Microbial digestion of complex carbohydrates in man. Proc. Nutr. Soc. 43, 35−44.Cutler, R.S., Fahy, E.M., Spicey, E.M., Gronin, G.M., 1999. Preweaning mortality. In: Straw, B.E.,

D’Allaire, S., Mengelin, W.I., Taylor, D.J. (Eds.), Diseases of Swine, 8th edition. Iowa StateUniversity Press, Ames, IA, pp. 985−1001.

Daniel, S.L., Hartman, P.A., Allison, M.J., 1987. Microbial degradation of oxalate in gastrointestinal tractof rats. Appl. Environ. Micobiol. 53, 1793−1797.


Drasar, B.S., 1974. Some factors associated with geographical variations in the intestinal microflora. In: Skinner, F.A., Carr, J.G. (Eds.), The Normal Microbial Flora of Man. London, Academic Press, pp. 187−196.

Drasar, B.S., Barrow, P.A., 1985. Intestinal Microbiology. Van Nostrand Reinhold (UK).Drasar, B.S., Hill, M.J., 1974. Human Intestinal Flora, 1st edition. London, Academic Press.Ducluzeau, R., 1983. Implantation and development of the gut flora in the newborn animal. Ann. Vet.

Res. 14, 354−359.Edfors-Lilja, I., Wallgren, P., 1999. Escherichia coli and Salmonella diarrhoea in pigs. In: Axford, R.F.E.,

Bishop, S., Nicholas, F., Owen, J.B. (Eds.), Breeding for Disease Resistance in Farm Animals, 2ndedition. CAB International, Wallingford, pp. 253−267.

Edfors-Lilja, I., Gustafsson, U., Duval Iflah, Y., Ellergren, H., Johansson, M., Jeneja, R.K., Marklund, L.,Andersson, L., 1995. The porcine intestinal receptor for E. coli K88ab, K88ac: regional localisationon chromosome 13 and influence of IgG response to the K88 antigen. Anim. Genetics 26, 237−242.

Edwards, C.A., Duerden, B.I., Read, N.W., 1985. Metabolism of mixed human colonic bacteria in con-tinuous culture mimicking the human cecal content. Gastroenterology 88, 1903−1909.

Ehle, F.R., Robertson, J.B., van Soest, P.J., 1982. Influence of dietary fibers on fermentation in the humanlarge intestine. J. Nutr. 112, 158−166.

Eidelsburger, U., Kirgessner, M., Roth, F.X., 1992. Influence of fumaric acid, hydrochloric acid, sodiumformate, tylosin and toyocerin on daily weight gain, feed intake, feed conversion rate and digestibility.II. Investigations about the nutritive efficacy of organic acids in rearing of piglets. J. Anim. Physiol.Anim. Nutr. 68, 82−92.

Fairbrother, J.M., 1999. Neonatal Escherichia coli diarrhea. In: Straw, B.E., D’Allaire, S., Mengelin, W.L.,Taylor, D.J. (Eds.), Diseases of Swine, 8th edition. Iowa State University Press, Ames, IA, pp. 433−441.

Finegold, S.M., Sutter, V.L., Mathisen, G.E., 1983. In: Hentges, D.J. (Ed.), Human Microflora in Healthand Disease. Academic Press, London, p. 3.

Fu, Z.F., Hampson, D.J., Wilks, C.R., 1990. Transfer of maternal antibody against group A rotavirus fromsows to piglets and serological responses following natural infection. Res. Vet. Sci. 48, 365−373.

Fuller, R., 1992. The effects of probiotics on the gut micro-ecology of farm animals. In: Wood, B.J.B.(Ed.), The Lactic Acid Bacteria, Vol. 1: The Lactic Acid Bacteria in Health and Disease. New York,Elsevier Applied Science, pp. 171−192.

Fuller, R., Barrow, P.A., Brooker, B.E., 1978. Bacteria associated with the gastric epithelium of neonatalpigs. Appl. Environ. Microbiol. 35, 582−591.

Glock, R.D., 1981. Disease of swine. In: Leman, A.D. (Ed.), The Iowa State University Press, Ames, IA,pp. 130−137.

Gorbach, S.L., 1990. Lactic acid bacteria and human health. Ann. Med. 22, 37−41.Gorbach, S.L., Chang, T.W., Goldin, B., 1987. Successful treatment of relapsing Clostridium difficile

colitis with Lactobacillus GG. Lancet ii, 1519.Hampson, D.J., 1986. Alterations in piglet small intestinal structure at weaning. Res. Vet. Sci. 40, 32−40.Hampson, D.J., Kidder, D.E., 1986. Influence of creep feeding and weaning on brush border enzyme

activities in the piglet small intestine. Rev. Vet. Sci. 40, 24−31.Hashimoto, S., Igimi, H., Uchida, K., Satoh, T., Benno, Y., Takeuchi, N., 1996. Effects of beta-lactam

antibiotics on intestinal microflora and bile acid metabolism in rats. Lipids 31, 601−609.Heidelberg Miller, B.G., Newby, T.J,. Stokes, C.R., Bourne, F.J., 1984. Influence of diet on preventing

post weaning malabsorbtion and diarrhoea in the pig. Res. Vet. Sci. 36, 187−193.Henriksson, A., Andre, L., Conway, P.L., 1995. Distribution of lactobacilli in the porcine gastrointestinal

tract. FEMS Microbial Ecol. 16, 55−60.Hentages, D.J., Stein, A.J., Casey, S.W., Que, J.U., 1985. Protective role of intestinal flora against infec-

tion with Pseudomonas aeruginosa in mice: influence of antibiotics on colonization resistance.Infect. Immun. 47, 118−122.

Hessing, M.J.C., Coenen, G.J., Vaiman, M., Renard, C., 1995. Individual differences in cell-mediated andhumoral immunity in pigs. Vet. Immunol. Immunopath. 45, 97−113.

Holland, R.E., 1990. Some infectious causes of diarrhoea in young farm animals. Clin. Microbiol. Rev.3, 345−375.


Holm, A., 1988. Escherichia coli induced post-weaning diarrhoea in pigs. Dietary supplementation as anantimicrobial method? Dansk Veterinærtidskrift 71, 1118−1126.

Holmgren, N., 1994. Prophylactic effects of zinc oxide or olaquindox against post-weaning diarrhoea inswine. Svensk Veterinärtidning 46, 217−222.

Horvath, D.J., 1957. The intestinal flora of simple-stomached animals, with special reference to the pig.Thesis, Cornell University, Ithaca, New York, pp. 147.

Inborr, J., Suomi, K., 1988. Industrial amino acids in diets for piglets and growing pigs. J. Agr. Sci. Fin.60, 673−683.

Jager, L.P., Ziljstra, F.J., Hoogendoorn, A., Nabuurs, M.J.A., 1986. Enteropooling in piglets induced by soya-peptone mediated via an increased biosynthesis of prostanoids. Vet. Res. Com. 10, 407−412.

Jensen, P., Recen, B., 1985. Maternal behaviour of free-ranging domestic pigs. Proc. Int. Eth. Congr. 4, 5.Jensen-Waern, M., Melin, L., Lindberg, R., Johannisson, A., Petersson, L., Wallgren, P., 1998. Dietary

zinc oxide in weaned pigs – effects on performance, tissue concentrations, morphology, neutrophilfunctions and faecal microflora. Res. Vet. Sci. 64, 225−231.

Johansen, H.N., Bach Knudsen, K.E., 1994. Effects of wheat-flour meal and oat mill fractions on jejunalflow, starch degradation and absorption of glucose over an isolated loop of jejunum in pigs. Brit. J. Nutr.72, 299−313.

Joling, P., Bianchi, A.T.J., Kappe, A.L., Zwart, R.J., 1994. Distribution of lymphocyte subpopulations in thymus, spleen and peripheral blood of specific pathogen free pigs from 1 to 40 weeks of life. Vet. Immunol. Immunopath. 40, 105−117.

Jonsson, E., 1986. Persistence of Lactobacillus strain in the gut of suckling piglets and its influence onperformance and health. Swedish J. Agr. Res. 16, 43−47.

Jonsson, E., Conway, P.L., 1992. Probiotics for pigs. In: Fuller, R. (Ed.), Probiotics, the Scientific Basis.Chapman & Hall, London, pp. 87−110.

Katouli, M., Erhart-Bennet, A.S., Kühn, I., Kollberg, B., Möllby, R., 1992. Metabolic capacity andpathogen properties of the intestinal coliforms in patients with ulcerative colitis. Microbial. Ecol.Health Dis. 5, 245−255.

Katouli, M., Lund, A., Wallgren, P., Kühn, I., Söderlind, O., Möllby, R., 1995. Phenotypic characteriza-tion of intestinal Escherichia coli of pigs during suckling, postweaning and fattening periods. Appl.Environ. Microbiol. 61, 778−783.

Katouli, M., Foo, E.L., Kühn, I., Möllby, R., 1997a. Evaluation of the Phene Plate generalized microplatefor metabolic fingerprinting and for measuring fermentative capacity of mixed bacterial populations.J. Appl. Microbiol. 82, 511−518.

Katouli, M., Lund, A., Wallgren, P., Kühn, I., Söderlind, O., Möllby, R., 1997b. Metabolic fingerprintingand fermentative capacity of the intestinal flora of pigs during pre- and post weaning periods. J. Appl.Bacteriol. 83, 147−154.

Katouli, M., Melin, L., Jensen-Waern, M., Wallgren, P., Möllby, R., 1999. The effect of zinc oxide sup-plementation on the stability of the intestinal flora with special reference to composition of coliformsin weaned pigs. J. Appl. Microbiol. 87, 564−573.

Kelly, D., Smyth, J.A., McCracken, K.J., 1991a. Digestive development of the early weaned pig. 1. Effectof continuous nutrient supply on the development of the digestive tract and on changes in digestiveenzyme activity during the first week post weaning. Brit. J. Nutr. 65, 169−180.

Kelly, D., Smyth, J.A., McCracken, K.J., 1991b. Digestive development of the early weaned pig. 2. Effectof level of food intake on digestive enzyme activity during the immediate post-weaning period. Brit.J. Nutr. 65, 169−180.

Kenworthy, R., Crabb, W.E., 1963. The intestinal flora of young pigs, with reference to early weaning,Escherichia coli and scours. J. Comp. Path. 73, 215−228.

Kirchgessner, M., Roth, F.X., Eidelsburger, U., Gedek, B., 1993. Nutritive effects of Bacillus cereus as aprobiotic on piglet rearing. I. Influence of growth variables and gastrointestinal tract. Arch. Anim.Nutr. 44, 111−121.

Klobasa, F., Werhahn, E., Butler, J.E., 1987. Composition of sow milk during lactation. J. Anim. Sci. 64,1458−1466.


Komarew, M.G., 1940. Differences in the normal microflora of the intestines of piglets at different ages(in Russian). Soviet Vet. 1, 27.

Kovacs, F., Nagy, B., Sinkvics, G., 1972. The gut bacterial flora of healthy early weaned piglets, withspecial regard to factors influencing its composition. Acta Vet. Acad. Sci. Hung. Tomus 22, 327−338.

Kyriakis, S.C., 1989. New aspects of the prevention and/or treatment of the major stress induced diseasesof the early-weaned piglet. Pig News Info. 2, 177−181.

Kyriakis, S.C., Tsiloylannis, V.K., Vlemmas, J., Sarris, K., Tsinas, C., Alexopolous, C., Jansegers, L.,1999. The effect of probiotic LPS 122 on the control of post-weaning diarrhoea syndrome of piglets.Res. Vet. Sci. 67, 223−228.

Kühn, I., Katouli, M., Lund, A., Wallgren, P., Möllby, R., 1993. Phenotype diversity and stability of intes-tinal coliform flora in piglets during the first three months of age. Microbial. Ecol. Health Dis. 6,101−107.

Kühn, I., Katouli, M., Wallgren, P. Söderlind, O., Möllby, R., 1995. Biochemical fingerprinting as a toolto study the diversity and stability of intestinal microfloras. Microecol. Ther. 23, 140−148.

Lee, A., Gemmell, E., 1972. Changes in the mouse intestinal microflora during weaning: role of volatilefatty acids. Infect. Immun. 5, 1−7.

Lidbeck, A., Nord, C.E., 1993. Lactobacilli and normal human anaerobic microflora. Clin. Infect. Dis.16 (Suppl 4), S181−S187.

Lin, C., Stahl, D.A., 1995. Toxin-specific probes for the cellulolytic genus Fibrobacter reveal abundantand novel equine-associated populations. Appl. Environ. Microbiol. 61, 1348−1351.

Lin, C., Flesher, B., Capman, W.C., Amann, R.I., Stahl, D.A., 1994. Taxon specific hybridisation probesfor fiberdigesting bacteria suggest novel gut-associated Fibrobacter. System Appl. Microbiol. 17,418−424.

Lin, C., Raskin L., Stahl, D.A., 1997. Microbial community structure in gastrointestinal tracts of domes-tic animals: comparative analyses using rRNA-targeted oligonucleotide probes. FEMS MicrobialEcol. 22, 281−294.

Linton, A.H., Hedges, A.J., Bennet, P.M., 1988. Monitoring for the development of antimicrobial resist-ance during the use of olaquindox as feed additive on commercial pig farms. J. Appl. Bacteriol. 64,311−327.

Lund, A., Wallgren, P., Rundgren, M., Artursson, K., Thomke, S., Fossum, C., 1998. Performance, behav-iour and immune capacity of domestic pigs reared for slaughter as siblings or transported and rearedin mixed groups. Acta Agr. Scand. 48, 103−112.

MacFarlande, G.T., Gibson, G.R., Cummings, J.H., 1992. Comparison of fermentation reaction in dif-ferent regions of the human colon. J. Appl. Bacteriol. 72, 57−64.

Månsson, I., Olsson, B., 1961. The intestinal flora of pigs. I. Quantitative studies of coliforms, entero-cocci and clostridia in the faeces of pigs self-fed a high protein and high calcium diet. Acta Agr.Scand. 11, 197−210.

Mäyrä-Mäkinen, A.A., Manninen, A., Gyllenberg, H., 1983. The adherence of lactic acid bacteria to thecolumnar cells of pigs and calves. J. Appl. Bacteriol. 55, 241−245.

McAllister, J.S., Kurtz, H.J., Short, E.C., 1979. Changes in the intestinal flora of young pigs with post-weaning diarrhea or edema disease. J. Anim. Sci. 49, 868−879.

McBurney, M.I., Horvath, P.J., van Soest, P.J., 1985. Effect of in vitro fermentation using human faecalinoculum in the water-holding capacity of dietary fibre. Brit. J. Nutr. 53, 17−24.

McGillivery, D.J., Cranwell, P.D., 1992. Anaerobic microflora associated with the pars oesophagea of thepig. Res. Vet. Sci. 53, 110−115.

McLeese, J.M., Tremblay, M.L., Patience, J.F., Christison, G.I., 1992. Water intake and patterns in theweanling pig: effects of water quality, antibiotics and probiotics. Anim. Prod. 54, 35−142.

Melin, L., Katouli, M., Jensen-Waern, M., Wallgren, P., 1996. The influence of zinc oxide on the intes-tinal microflora of piglets at weaning. Proc. IPVS 14, 465.

Melin, L., Jensen-Waern, M., Johannisson, A., Ederoth, M., Katouli, M., Wallgren, P., 1997.Development of selected faecal microfloras and phagocytic killing capacity of neutrophils in youngpigs. Vet. Microbiol. 54, 287−300.

Melin, L., Katouli, M., Lindberg, Å., Fossum, C., Wallgren, P., 2000a. Weaning of piglets. Effects of anexposure to a pathogenic strain of Escherichia coli. J. Vet. Med. B. 47, 663−675.


Melin, L., Mattsson, S., Wallgren, P., 2000b. Challenge with pathogenic strains of Escherichia coli atweaning. I. Clinical signs and reisolation of the challenge strains. Proc. IPVS 16, 22.

Melin, L., Mattsson, S., Wallgren, P., 2000c. Challenge with pathogenic strains of Escherichia coli atweaning. II. Monitoring of the faecal coliform flora. Proc. IPVS 16, 23.

Meynell, G.G., Meynell, E., 1970. Theory and Practice in Experimental Bacteriology. CambridgeUniversity Press, Cambridge.

Midtvedt, T., 1985. Microflora associated characteristics (MACs) and germ-free animal characteristics(GACs) in man and animal. Microecol. Ther. 15, 295−302.

Midtvedt, T., 1999. Microbial functional activities. Probiotics, other nutritional factors, and intestinalflora. Nestle Nutrition Workshop Series 42, 79−96.

Miller, B.G., Newby, T.J., Stokes, C.R., Bourne, F.J., 1984. Influence of diet on post-weaning malab-sorbtion and diarrhoea in pigs. Res. Vet. Sci. 36, 187−193.

Mitsuoka, T., Terada, A., Watanabe, K., Uchida, K., 1974. Bacteroides multicidus, a new species from thefaeces of humans and pigs. Int. J. Sys. Bacteriol. 24, 35−41.

Möllby, R., Kühn, I., Katouli, M., 1993. Computerised biochemical fingerprinting – a new tool for typ-ing of bacteria. Rev. Med. Microbiol. 4, 231−241.

Molly, K., Vandewoestyne, M., Desmet, I., Verstraete, W., 1994. Validation of the simulator of the humanintestinal microbial ecosystem (SHIME) reactor using microorganism-associated activities. MicrobialEcol. Health Dis. 7, 191−200.

Moore, W.E., Holdeman, L.V., 1974. Special problems associated with the isolation and identification ofintestinal bacteria in fecal flora studies. Amer. J. Clin. Nutr. 27, 1450−1455.

Nabuurs, M.J.A., 1986. Thermostable factor(s) in soya producing a net excess of secretion in the ligatedgut test in pigs. Vet. Res. Com. 10, 399−405.

Newport, M.J., 1980. Artificial rearing of pigs. II. Effect of replacement of dried skim-milk by an iso-lated soya-bean protein on the performance of the pigs’ digestion of protein. Brit. J. Nutr. 44, 171−178.

Norin, E., Midtvedt, T., 2000. Interactions of bacteria with the host. Alteration of microflora-associatedcharacteristics of the host; non-immune functions. Microbial Ecol. Health Dis. Suppl. 2, 186−193.

Pielou, E.C., 1975. Ecological Diversity. Wiley Interscience. New York.Onderdonk, A.B., 1999. The intestinal microflora and intra-abdominal sepsis. In: Tannock, G.W. (Ed.),

Medical Importance of the Normal Microflora. Kluwer Academic Publishers, Dordrecht, pp. 164−176.Quinn, L.Y., Lane, M.D., Ashton, G.C., Maddock, H.M., Carlton, D.V., 1953a. Mode of action of antibi-

otics on intestinal flora. Antibiotics Chemother. 3, 622−628.Quinn, L.Y., Story, C.D., Carlton, D.V., Jensen, A.H., Whalen, W.M., 1953b. Effect of antibiotics on the

growth rate and intestinal flora of swine. Antibiotics Chemother. 3, 527−532.Rainbard, A.L., Low, A., Z

.ebrowska, T., 1984. Effect of guar gum on glucose and water absorption from

isolated loops of jejunum in conscious growing pigs. Brit. J. Nutr. 52, 489−498.Rall, G.D., Wood, A.J., Wescott, R.B., Dommert, A.R., 1970. Distribution of bacteria in faeces of swine.

Appl. Microbiol. 20, 789−792.Raskin, L., Capman, W.C., Sharp, R., Stahl, D.A., 1995. Molecular ecology of gastrointestinal systems: In:

Mackie, R.I., White, B.A., Issacson, R.E. (Eds.), Ecology and Physiology of Gastrointestinal Microbes,Vol 2: Gastrointestinal Microbiology and Host Interactions. Chapman and Hall, Inc, New York.

Risatti, J.B., Capman, W.C., Stahl, D.A., 1994. Community structure of a microbial mat: the phyloge-netic dimension. Proc. Natl. Acad. Sci. USA 91, 10 173−10 177.

Rolfe, R.D., 1997. Colonization resistance. In: Mackie, R.I., White, B.A., Isacson, R.E. (Eds.),Gastrointestinal Microecology, Vol. 2. Chapman and Hall, Inc, New York, pp. 501−536.

Ross, L.F., Shaffer, G.P., 1989. Fermentation of carbohydrates under aerobic and anaerobic conditions byintestinal microflora from infants. J. Clin. Microbiol. 27, 2529−2534.

Rotimi, V.O., Duerden, B.I., 1981. The development of the bacterial flora in normal neonates. J. Med.Microbiol. 14, 51−62.

Rowe, J.B., Loughnan, M.L., Nolan, J.V., Leng, R.A., 1979. Secondary fermentation in the rumen of asheep given a diet based on molasses. Brit. J. Nutr. 41, 393−397.

Saito, Y., Sanada, T., Iwasaka, K., Mita, T., Ohba, S., Tsumagari, S., Takeishi, M., 1986. Changes in IgG,IgM and IgA content in the milk of sows and blood serum of suckling pigs. Jpn. J. Zootechnol. Sci.57, 349−351.


Salminen, S., Arvilommi, H., 2001. Probiotics demonstrating efficacy in clinical settings. Clin. Infect.Dis. 32, 1577−1578.

Salnitro, J.P., Blake, I.G., Muirhead, P.A., 1977. Isolation and identification of fecal bacteria from adultswine. Appl. Environ. Microbiol. 33, 79−84.

Sansom, B.F., Gleed, P.T., 1981. The ingestion of sow’s faeces by suckling piglets. Brit. J. Nutr. 46, 451−456.Sauer, W.C., Jörgensen, H., Berzins, R., 1983. A modified nylon bag technique for determining apparent

digestibilities of protein in feedstuffs for pigs. Can. J. Anim. Sci. 63, 233−237.Savage, D.C., 1977. Microbial ecology of the gastrointestinal tract. Annual Rev. Microbiol. 31, 107−133.Shone, F., Ludke, H., Brys, J., Bruckner, H., 1988. The performance of pigs weaned at 5 weeks related to

weaning weight, feed composition and the use of effective ergotropics. Arch. Anim. Nutr. 38, 861−877.Shell, T.C., Korneay, E.T., 1994. Effectiveness of zinc acetate injection in alleviating post weaning per-

formance lag in pigs. J. Anim. Sci. 72, 3037−3042.Smith, C.J., Bryant, M.P., 1979. Introduction to metabolic activities of intestinal bacteria. Amer. J. Clin.

Nutr. 32, 149−157.Smith, H.W., 1965a. Observations on the flora of the alimentary tract of animals and factors affecting its

composition. J. Path. Bacteriol. 89, 95−102.Smith, H.W., 1965b. The development of the flora of the alimentary tract in young animals. J. Path.

Bacteriol. 90, 459−513.Smith, H.W., Crabb, W.E., 1961. The faecal bacterial flora of animals and man: its development in the

young. J. Path. Bact. 82, 53−66.Smith, H.W., Jones, J.E.T., 1963. Observation on the alimentary tract and its bacterial flora in healthy and

diseased pigs. J. Path. Bact. 86, 387−412.Stahl, D.A., Flesher, B., Mansfield, H.R., Montgomery, L., 1988. Use of phylogenetically based hybridi-

sation probes for studies of ruminal microbial ecology. Appl. Environ. Microbiol. 54, 1079−1084.Stewart, C.S., Bryant, M.P., 1988. The rumen microbial ecosystem. In: Hobson, P.N. (Ed.), Elsevier

Applied Science, New York, p. 21.Straw, B.E., Dewey, C.E., Wilsson, M.R., 1999. Differential diagnosis of swine diseases. In: Straw, B.E.,

D’Allaire, S., Mengelin, W.L., Taylor, D.J. (Eds.), Diseases of Swine, 8th edition. Iowa StateUniversity Press, Ames, IA, pp. 41−50.

Tancerede, C., 1992. Role of human microflora in health and disease. Eur. J. Clin. Microbiol. Infect. Dis.11, 1012−1015.

Tannock, G.V., 1995. Normal Microflora. An Introduction to Microbes Inhabiting the Human Body.Chapman and Hall, London.

Taylor, D.J., 1999. Clostridial infection. In: Straw, B.E., D’Allaire, S., Mengelin, W.L., Taylor, D.J.(Eds.), Diseases of Swine, 8th edition. Iowa State University Press, Ames, IA, pp. 395−412.

Terada, A., Uchida, K., Mitsouka, T., 1976. The Bacteriodaceae flora in faeces of pigs. Zbl. Bakt. A. 234,362−370.

Thijm, H.A., van der Waaij, D., 1979. The effect of three frequently applied antibiotics on the colonisa-tion resistance of the digestive tract of mice. J. Hyg. (London) 82, 397−405.

Toit, M. du, Franz, C.M.A.P., Dicks, L.M.T., Schillinger, U., Haberer, P., Warlies, B., Ahrens, F.,Holzapfel, W.H., 1998. Characterisation and selection of probiotic lactobacilli for a preliminaryminipig feeding trial and their effect on serum cholesterol levels, faeces pH and faeces moisture con-tent. Int. J. Food Microbiol. 40, 93−104.

Van Beers-Schreurs, H.M.G., Nabuurs, M.J.A., Vellenga, L., Kalsbeck-van der Valk, H.J., Wensing, T.,Breukink, H.J., 1998. Weaning piglets, microbial fermentation, short chain fatty acid and diarrhoea.Vet. Quart. 20, s64−s68.

Van der Heyde, H., Henderickx, H., 1964. Quantitative studies of the intestinal flora of suckling piglets.Zbl. Bakt. I. Abt. Orig. 1195, 215−226.

Van der Peet-Schwering, C.M.C., van der Binnendijk, G.P., 2000. Influence of protein sources inprestarter diets on performance and health of weaned piglets. Proefverslag PraktijkonderzoekVarkenshouderij No P1. 234, p. 16.

Van der Waaij, D., 1979. New Criteria for Antimicrobial Therapy. Maintenance of Digestive TractColonisation Resistance. Excerpta Medica, Amsterdam.


Van Leeuwen, P., van der Kleef, D.J., van Kempen, G.J.M., Huisman, J., Verstegen, M.W.A., 1991. Thepost valve T-caecum cannulation technique to determine the digestibility of amino acids in maize,groundnut and sunflower meal. J. Anim. Physiol. Anim. Nutr. 65, 183−193.

Varel, V.H., Pond, W.G., 1985. Enumeration and activity of cellulolytic bacteria from gestating swine fedvarious levels of dietary fiber. Appl. Environ. Microbiol. 49, 858−862.

Wallgren, P., Artursson, K., Fossum, C., Alm, G.V., 1993. Incidences of infections in pigs bred for slaughterrevealed by elevated serum levels of interferon and development of antibodies to Mycoplasmahyopneumoniae and Actinobacillus pleuropneumoniae. J. Vet. Med. B. 40, 1−12.

Wallgren, P., Bölske, G., Gustafsson, S., Mattsson, S., Fossum, C., 1998. Humoral immune response toMycoplasma hyopneumoniae in sows and offspring following an outbreak of mycoplasmosis. Vet.Microbiol. 60, 193−205.

Wattrang, E., Wallgren, P., Lindberg Å., Fossum, C., 1998. Signs of infections and reduced immune func-tions at weaning of conventionally reared and specific pathogen free pigs. J. Vet. Med. B. 45, 7−17.

Wilbur, R.D., 1959. The intestinal flora of the pigs as influenced by diet and age. PhD thesis, Iowa StateUniversity, Ames, Iowa.

Willingale, J.M., Briggs, C.A.E., 1955. The normal intestinal flora of the pig. II. Quantitative bacterio-logical studies. Appl. Bact. 18, 284−292.

Wilson, D.A., MacFadden, K.E., Green, E.M., Crabill, M., Frankeny, R.L., Thorne, J.G., 1996. Case con-trol and historical cohort study of diarrhoea associated with administration of trimethoprim-potenti-ated sulphonamides to horses and ponies. J. Vet. Intern. Med. 10, 258−264.

Wilson, G., 1974. In: Skinner, F.A., Carr, J.G. (Eds.), The Normal Microbial Flora of Man. AcademicPress, London and New York, p. 1.

Wilson, K., Moore, L., Patel, M., Permoad, P., 1988. Suppression of potential pathogens by a definedcolonic microflora. Microbial Ecol. Health Dis. 1, 237−243.

Wilson, K.H., Freter, R., 1986. Interaction of Clostridium difficile and Escherichia coli with microflorasin continuous-flow cultures and gnobiotic mice. Infect. Immun. 54, 354−358.

Wyatt, G.M., Horn, H., 1988. Fermentation of resistant food starches by human and rat intestinal bacteria.J. Sci. Food Agr. 44, 281−288.

Zoric, M., Arvidsson, A., Melin, L., Kühn, I., Lindberg, J.E., Wallgren, P., 2001. The correlation betweencoliform populations collected from different sites of the intestinal tract of pigs. In: Lindberg, J.E.,Ogle, B. (Eds.), Digestive Physiology of Pigs. CABI Publishing, Wallingford, pp. 283−285.


54

This review summarizes the development and activity of ciliate protozoa in therumen of domestic ruminants during the first months of postnatal life. The appear-ance and establishment of ciliate fauna in the rumen following both natural mouth-to-mouth contact of newborn animals with their dams and/or other members of theflock as well as by experimental introduction of the rumen fluid or digesta inoculumto the rumen of ciliate-free lambs, calves and kids are considered in relation to diet,age of the animal, type of management and the pH of the rumen digesta. Ciliates asa factor affecting fermentation in the rumen and the growth of young ruminants arediscussed. The growth and functional maturation of the rumen as a fermentingchamber is briefly described. The presented information also concerns the taxon-omy of ciliates and their role in the digestion and fermentation of nutrients in therumen, in particular with respect to the degradation of structural carbohydrates inplant cell walls.

1. INTRODUCTION

Although the rumen microbial ecosystem plays a crucial role in ruminant nutrition, theabomasum is the only well-developed and functioning portion of the complex stomachin newborn animals (McGilliard et al., 1965; Warner and Flatt, 1965; Hofmann, 1988;Lyford, 1988). Thus, during the first days of postnatal life ruminants do not differ frommonogastric mammals when digestive processes are considered.

It is known that the growing domestic ruminant becomes dependent on fermen-tation products by 8 weeks of age (Lyford, 1988). Thus a 2-month-long period isnecessary for the rumen to become a functional fermentation chamber. The growthand functional maturation of the rumen tissues is accompanied by the development

3 Rumen protozoa in the growing domesticruminant

T. Michalowski

The Kielanowski Institute of Animal Physiology and Nutrition, Polish Academy ofSciences, Instytucka 3, 05-110 Jablonna near Warsaw, Poland

Microbial Ecology in Growing AnimalsW.H. Holzapfel and P.J. Naughton (Eds.)© 2005 Elsevier Limited. All rights reserved.

Rumen protozoa in the domestic ruminant 55

of a microbial population comprising bacteria, fungi and protozoa, which appear inthe rumen in a defined sequence (Cheng et al., 1991; Dehority and Orpin, 1997).The aim of this chapter is to present the findings of studies on the development ofciliate fauna in domestic ruminants during the first months of postnatal life withrespect to the role of protozoa in the rumen metabolism and performance of the hostanimals.

2. ROLE OF MICROORGANISMS IN RUMINANT NUTRITION

Despite their inability to manufacture fibrolytic enzymes, mammalian herbivoreshave for over 35 million years been eating plant material containing large quantitiesof cellulose and hemicellulose (Langer, 1988). All of these animals harbour symbi-otic microbes in the alimentary tract and depend on microbial digestion and fermentation to release the energy stored in the structural carbohydrates of plant cellwalls (Hungate, 1966; Janis, 1976; McBee, 1977; Van Soest, 1982; Owens andGoetsch, 1988; Flint, 1997; Russell and Rychlik, 2001). Symbiotic microbes colonize either the pregastric (foregut fermentors) or lower gut (hindgut fermentors)portions of the digestive tract, which become capacious chambers (Janis, 1976;McBee, 1977; Hume and Sakaguchi, 1991). Independently of their location, thesechambers are filled with large quantities of plant organic matter characterized bylong residence time (up to 47 h), high but constant temperature (~+40°C) and low(−350 mV) redox potential (Clarke, 1977; Van Soest, 1982). Such conditions allowanaerobic microorganisms to densely colonize this habitat.

Ruminants are the most specialized mammalian herbivores (McBee, 1977) andthe most dependent on the symbionts harboured in the rumen. The rumen is the largestportion of the complex stomach and also of the digestive tract in adult ruminants(Warner and Flatt, 1965; Lyford, 1988). Its primary function in ruminant ancestorswas, perhaps, to store the ingested plant feed (Janis, 1976). However, it has evolvedinto the almost ideal fermentation chamber (Russell and Rychlik, 2001). The largequantities of plant material filling the rumen and the very dense (over 1010 cells/g)microbial population colonizing this habitat allow ruminants to meet up to 80% oftheir energy requirements from the end products of the carbohydrate fermentation(Owens and Goetsch, 1988; Russell and Rychlik, 2001).

The microbial pathways involved in the fermentation of structural carbohydratesare similar in all herbivore mammals (Janis, 1976). In ruminants, fibrolytic microbesdigest cellulose and hemicellulose to soluble oligosaccharides that can also be utilized by non-fibrolytic species (Flint and Forsberg, 1995). End products of sugarfermentation are released from the cells of microbes mainly as acetic acid, propionicacid and butyric acid. They are absorbed into the blood via the fermenting chamberepithelium, transported to the tissues and cells of the animal body and used there inthe ATP generation processes, gluconeogenesis and milk fat synthesis (Van Soest,1982; Fahey and Berger, 1988).

56 T. Michalowski

The beneficial role of the ruminal microbes results not only from providing thehost animals with energy yielding products. It has also been reported that over 90%of the amino acids reaching the duodenum can originate from the microbial proteinsynthesized in the rumen (Russell and Rychlik, 2001). Microbial protein synthesisand utilization by the host is, however, a complex problem affected by different factors. Two processes should at least be distinguished, gross and net synthesis. Thelatter can be interpreted as a microbial protein or N incorporated into the cellularprotein of microbes reaching the duodenum and is the true measure of microbes asa source of amino acids to ruminants (Demeyer and Van Nevel, 1986).

3. POSTNATAL GROWTH AND FUNCTIONAL MATURATION OF THE RUMEN

Ruminants are characterized by the presence of a complex stomach consisting offour chambers: the reticulum, rumen, omasum and abomasum (Hungate, 1966).Only the last chamber, the abomasum, is lined with glandular mucosa and is capable of producing digestive enzymes. The three other chambers are lined withnon-glandular mucosa (Hofmann, 1988) and fail to produce hydrolases. All three ofthese compartments are termed forestomachs. In adult animals, the rumen is thelargest chamber of the complex stomach. It is functionally integrated with a muchsmaller reticulum and they are often considered as a common compartment, i.e. the reticulorumen (Hungate, 1966). The weight of the reticulorumen content canexceed 100 kg in cattle and 15 kg in sheep. It contributes to 9–13% of total body weight and to over 70% of the digesta weight in the digestive tract (Van Soest, 1982).

In newborn calves and lambs, the abomasum is the largest and functionally mostdeveloped organ whereas the rumen is small and flaccid and the papillae are only rudimentary (Hofmann, 1988; Lyford, 1988). Rapid growth of the rumen isobserved by 8 weeks of age with a maximum between 4 and 8 weeks. During thisperiod the rumen approaches adult proportions and becomes a capacious chamber(Lyford, 1988). The relative weight of content filling the rumen of 8-week-old lambsand calves is still less than in adult ruminants (6.8% vs 9–13%) but at this age theybecome dependent on fermentation products for maintenance and growth (Warnerand Flatt, 1965; Van Soest, 1982; Lyford, 1988). This suggests that the rumenecosystem can already function in 2-month-old domestic ruminants. In newbornanimals, the rumen mucosa is also poorly developed and its absorptive and meta-bolic abilities are very low (McGilliard et al., 1965). These abilities increase duringthe first weeks of postnatal life. Solid food and end products of carbohydrate fermentation are necessary to accelerate the growth and functional development ofthe rumen (Warner and Flatt, 1965). It was found that hay is a stimulating factor ofrumen tissue growth, while concentrates stimulate the growth of rumen papillae(Lyford, 1988; Swan and Groenewald, 2000). The presence of volatile fatty acids


(VFA) is necessary to accelerate the functional development of the rumen mucosaas measured by VFA absorption rate, rate of glucose and butyrate oxidation, or rateof acetoacetate and lactate production (McGilliard et al., 1965; Lane and Jesse,1997). More recently published results suggest that exposure of the rumen mucosato VFA at an early stage of development is required to induce the genes responsiblefor metabolic development of the rumen (Lane et al., 2000). It is obvious thatmicroorganisms are the only VFA producers in the rumen (Hungate, 1966; VanSoest, 1982). This shows the significant role of microbes in the functional matura-tion of the rumen.

4. RUMEN PROTOZOA

The findings presented above show that ruminants depend on the rumen microbiotastarting from the early hours of their postnatal life. Ruminal microorganisms belongto three different groups, i.e. bacteria, fungi and protozoa. Bacteria are most numer-ous in the rumen. They belong also to the most intensively studied organisms and there are excellent articles summarizing progress in the rumen bacteriology (see Stewart et al., 1997, for review). Ruminal fungi belong to the classChytridiomycetales and family Neocallimastigaceae. Five genera and more than 20species have been identified to date (Theodorou et al., 1996; Orpin and Joblin,1997). The importance of fungi to the host results from their ability to degrade lignocellulosic plant material. The relation between this group of microbes andyoung ruminants is not well known.

Rumen protozoa are principally ciliates representing two morphologically andphysiologically different groups, i.e. Entodiniomorphids (according to older litera-ture the Oligotrich or Spirotrich protozoa) and holotrichs. According to Williamsand Coleman (1992) and references therein both groups belong to the same subclassTrichostomatia (Bütschli, 1889) and two different orders: Entodiniomorphida(Reichenow in Doflein and Reichenow, 1929) and Vestibuliferida (Puytorac et al.,1974). Ruminal entodiniomorphs (fig. 1) belong to the family Ophryoscolecidae.Dogiel (1927) distinguished five genera among the ophryoscolecid ciliates:Entodinium, Diplodinium, Epidinium, Ophryoscolex and Opistotrichum.Additionally to this, the genus Diplodinium has been divided into four subgenera:Anoplodinium, Eudiplodinium, Polyplastron and Ostracodinium. In fact, the generaCunhaja and Caloscolex belong also to the family Ophryoscolecidae (Dogiel,1927). However, they have not been found in ruminants. Following subsequent revisions summarized by Williams and Coleman (1992) particular genera have been raised to the rank of subfamilies, i.e. Entodiniinae, Diplodiniinae, Epidiniinae,Opistotrichinae and Ophryoscolecinae, and the subfamily Diplodiniinae has beendivided into 10 genera: Eodinium, Diplodinium, Eudiplodinium, Eremoplastron,Ostracodinium, Metadinium, Diploplastron, Elytroplastron, Enoploplastronand Polyplastron. According to the same revisions the subfamily Epidiniinae

58 T. Michalowski

comprises the genera Epidinium and Epiplastron, whereas the other subfamilies arerepresented by single genera, i.e. Entodinium, Ophryoscolex and Opistotrichum. Incontrast to this Imai (1998) distinguished only three subfamilies: Entodiniinae,Diplodiniinae and Ophryoscolecinae which have been already postulated earlier byLubinsky (1957b). The first two families do not differ from those mentioned above,while the last comprises all five genera from the subfamilies Epidiniinae,Opisthotrichinae, Ophryoscolcinae and Caloscolecinae.

The taxonomy of the second group is also not quite clear. According to Levineet al. (1980) the ciliates routinely termed holotrichs belong to the subclassVestibuliferia and Gymnostomata, orders Trichostomatida and Prostomatida.However, Lee et al. (1985) included them in the subclass Trichostomatia, ordersVestibuliferida and Entodiniomorphida. The species most often present in the rumenand also the best known belong to the family Isotrichidae, order Vestibuloferida (fig. 2).

Taxonomy of rumen ciliates is based on the morphology of their cells. Over 250species belong to the family Ophryoscolecidae. They are the most numerous of allprotozoa in the rumen. Their numbers often exceed 106 cells/g rumen digesta. Theruminal ophryoscolecids are characterized by a rigid pellicle and ciliature reducedto only adoral (Entodinium spp.) or adoral and dorsal ciliary bands (the remaininggenera). Other organella of taxonomic significance are skeletal plates (not present in

Fig. 1. Microphotographs of some commonEntodiniomorphid ciliates. 1. Entodinium caudatum; 2. Eudiplodinium maggii; 3. Epidinium ecaudatum f. caudatum; 4. Ophryoscolex caudatus.


Entodinium, Eodinium and Diplodinium spp.), the number and positions of vacuoles,the shape of the macronucleus and position of both the macro- and micronucleus aswell as the presence of spines and lobes and even the ciliate cell dimensions (to studythis problem in detail see Dogiel, 1927; Kofoid and MacLennan, 1930, 1932, 1933;Lubinsky, 1957a,b; Noirot-Timothée, 1960; Latteur, 1969, 1970). It is noteworthythat many taxonomists believe that ciliates classified as different species are in factsimply different forms of the same species. This concern is especially relevant to thesmall Entodinia (Latteur, 1968). The author of this chapter agrees with this opinionbecause considerable differences in cell morphology can be observed between indi-viduals of the same clone (unpublished). These observations need to be confirmed by molecular studies such as 18S ribosomal gene sequencing. Initial sequences have been reported from some species of ruminal entodiniomorphs, i.e. Entodinium simplex, Entodinium caudatum, Eudiplodinium maggii, Polyplastron multivesiculatum,Epidinium ecaudatum and Ophryoscolex purkynjei (Embley et al., 1995; Wright and Lynn, 1997; Wright et al., 1997) and suggest that rumen protozoa represent a monophyletic group.

In contrast to the entodiniomorphs, the cells of holotrich ciliates are flexible andthe ciliature of the most common species is almost complete (fig. 2). The concen-tration of holotrichs in the rumen is in the range of 104/ml rumen fluid.

Fig. 2. Microphotographs of the most common species of the ruminal holotrichs. 1. Dasytricha ruminantium; 2. Isotricha prostoma; 3. Isotricha intestinalis.

5. ROLE OF PROTOZOA IN FOOD CONVERSION

Ciliates from the family Ophryoscolecidae prefer particulate matter as food, while sol-uble substances are rather poorly utilized (Michalowski, 1989; Williams, 1989). Theciliates Eudiplodinium maggii, Polyplastron multivesiculatum, Epidinium ecaudatumand some other large ophryoscolecids are able to digest and metabolize cellulose(Coleman, 1985, 1986; Bonhomme et al., 1986; Dehority, 1993; Michalowski, 1997).It was also found that the ciliates Eudiplodinium maggii and Epidinium ecaudatum syn-thesize both β-endoglucanases and β-endoxylanases (Michalowski, 1997; Michalowskiet al., 2001b). The genes encoding for β-endoglucanase and xylanase were also clonedfrom Polyplastron multivesiculatum and Epidinium ecaudatum (Selinger et al., 1996;Devillard et al., 2000). These findings confirm the direct participation of some largeophryoscolecids in fibre degradation in the rumen. Thus it is not surprising that 50–70%of the fibrolytic activity in the rumen results from the presence of ciliates (Coleman,1986; Michalowski and Harmeyer, 1998). It is also well known that Entodinia andother small entodiniomorphid species prefer starch (Williams, 1989). Engulfed carbo-hydrates are digested and fermented while acetic acid, propionic acid and butyric acidare released as end products (Williams and Coleman, 1992; Michalowski, 1997). It wasfound that 50% of the VFA produced in the rumen of sheep fed a hay-concentrate dietwas of ciliate origin and that ophryoscolecids were dominating in the rumen of exam-ined animals (Michalowski, 1987). Thus entodiniomorphid protozoa can play animportant role in providing the host with energy stored in carbohydrates includingstructural polysaccharides. In contrast with entodiniomorphs, rumen holotrichs do notingest fibrous material and readily utilize soluble compounds (Williams, 1989). Lacticacid is an important end product of carbohydrate metabolism in these protozoa (Prinsand Van Hoven, 1977; Van Hoven and Prins, 1977).

The ciliates engulf and digest ruminal bacteria (Coleman, 1989). Owing to this,the protozoa negatively influence the microbial protein supply at the duodenum ofthe host. They also enhance ammonia production in the rumen. This sometimesresults in diminishing weight gain in ruminants maintained on a low quality diet(Jouany et al., 1988). The reduction in bacterial N supply can only partially be com-pensated by protozoal protein because ciliates are selectively retained in the rumen(Weller and Pigrim, 1974; Michalowski et al., 1986). Thus, the importance of pro-tozoa to ruminants seems to result from their digesting and fermenting activities inthe rumen rather than from their role as an amino acid source (John and Ulyatt,1984; Michalowski, 1990).

6. ESTABLISHMENT OF CILIATES IN THE RUMEN OF YOUNG RUMINANTS

Ruminants are born with a sterile digestive tract but invasion of microorganismsbegins immediately after birth. Colonization of the rumen by bacteria begins fromassociation of microaerophilic and ureolytic species with the rumen epithelium and

60 T. Michalowski


this is followed by initial development of the cellulolytic consortia as early as 2–4 daysafter birth (Cheng et al., 1991). However, Fonty et al. (1984) observed very low concentration of cellulolytic bacteria in 8-day-old lambs. According to Cheng et al.(1991) rumen fungi were observed in the rumen on day 8–10 after birth and were followed by protozoa. This shows that ciliates appear in the rumen as the last member of the microbial ecosystem.

Rumen ciliates produce neither resistance forms nor cysts (Dehority and Orpin,1997) and Becker and Hsiung demonstrated as early as 1929 that direct contact isnecessary to transfer these organisms between the host animals. It is obvious thatmouth-to-mouth contact is more probably between newborn ruminants and theirdams. However, successful transmission of protozoa to the rumen of lambs and/orcalves is rarely possible during the first 1–2 days of their postnatal life. Thus they canremain ciliate-free for a long period if they become separated from their dams withinthe mentioned period (Bryant et al., 1958; Eadie and Gill, 1971). Development ofmicrofauna in the rumen of young domestic ruminants as a result of natural animalto animal transmission has been examined by several authors during the past 50 years(table 1). Lengemann and Allen (1959) observed protozoa in the rumen of calveswithin the first week after birth, whereas in flock reared lambs they appeared by 7–20days (Fonty et al., 1984, 1988). On the other hand Eadie (1962) found quite definitefauna in 21-day-old lambs, while Naga et al. (1969) and Fonty et al. (1984, 1988)have reported that a 60- to 150-day period was necessary to establish a mixed type ofrumen fauna at the adult animal level in lambs as well as buffalo and cow calves.

Few studies have investigated the development of populations of particular genera (table 2). The results obtained showed that Entodinia colonized the rumenat first, while the establishment of higher ophryoscolcesids varied in relation to thehost and management system. In general, the ciliates from the genera Diplodinium,Eudiplodinium, Ostracodinium and Polypalstron followed Entodinia whileElytroplastron, Enoploplastron and Ophryoscolex became established distinctly later.The establishment of holotrichs varied also in relation to the mentioned factors.

Table 1. Appearance of ciliates and establishment of mixed type fauna in the rumen of growingdomestic ruminants

Days, weeks or months after birth

Animals Appearance Establishment References

Cow calves Up to 7 days 4–11 weeks Lengemann and Allen, 1959

29–132 days 80–150 days Naga et al., 1969Buffalo calves 13–46 days 40–60 daysLambs 9–14 days 21 days Eadie, 1962

15–20 days 2 months Fonty et al., 19847–50 days up to 100 days Fonty et al., 1988

62 T. Michalowski

For example, Isotricha spp. appeared as the last ciliate in flock reared lambs but theyfollowed Entodinia in early-weaned cow calves. It was also reported that an estab-lished population of Polyplastron multivesiculatum disappeared after existence inthe rumen for several months owing to undefined causes (Fonty et al., 1984, 1988).Such a phenomenon was also observed in the case of the same species as well as ciliates from the genus Ophryoscolex inoculated into the rumen of ciliate-free buffalo calves and lambs (Eadie, 1967; Naga et al., 1969).

Development of the ciliate fauna was also examined following the experimentalinoculation of young ruminants. Pounden and Hibbs (1948, 1949) observed numerousciliates in the rumen of 3- and 6-week-old calves. The animals were inoculated by pass-ing pieces of cuds taken from cows into the mouths of calves on the 5th, 10th, 15th and21st day after birth. Bryant et al. (1958) kept ciliate-free calves 13–15 weeks of agetogether with faunated mature animals and this resulted in transmission of ciliates fromadult ruminants to the young. Entodinia were observed in 17–20-week-old calves andwere followed by ciliates from the genus Diplodinium and family Isotrichae, whichwere found in 27- and 33–37-week-old animals, respectively. A similar sequence in theappearance of ciliates was found in calves inoculated with the samples of rumen content from a mature cow (Bryant and Small, 1960). Abou Akkada and El-Shazly(1964) inoculated 5-week-old lambs with the rumen fluid from a mature sheep.Entodinium and Isotricha spp. were the first ciliates to develop in almost all of 11 lambs. Ophryoscolex, Polyplastron and Diplodinium spp. were observed in small orappreciable numbers not earlier than on day 11 after inoculation. The authors noted difficulties in developing a population of Dasytricha ruminantium, which became

Table 2. Establishment of ciliates from different genera in the rumen of calves and lambs (daysafter birth)

Naga et al., 1969 Fonty et al., 1988

Cow calves Buffalo calves Lambs

Ciliate genera A B A B Flock reared Dam reared

Entodinium 29 46 13 18 7–10 13–20Diplodinium 30 70 18 37 – –Eudiplodinium 30 60 22 18 10–20 30Ostracodinium 30 70 27 35 – –Elytroplastron 68 132 39 46 – –Enoploplastron – – – – – 100Polyplastron – – – – 10–20 30Epidinium – – – – 20 –Ophryoscolex – – – – – 50Isotricha 30 46 28 33 50 –Dasytricha 47 130 22 27 – –

A = early weaned calves. B = late weaned calves.


established in only three of 11 animals examined. Borhami et al. (1967) inoculated 2–3-week-old buffalo calves by introducing samples of fresh rumen content taken frommature faunated animals. The authors observed Entodinia and Diplodinia as quickly as4 days after inoculation. Eudiplodinium and Isotricha spp. were found 2 days later,whereas small or moderate numbers of Metadinium spp. were identified as late as 7 months after inoculation and only in three of 11 calves tested. Similarly Eadie (1967)observed a very long lag phase between the inoculation and appearance of ciliates fromthe genus Ophryoscolex in the rumen of lambs and kids. Naga et al. (1969) inoculatednewly born buffalo and cow calves with rumen contents of adult buffalo, cow and sheep. Ciliates from the genus Entodinium were found first (4–40 days after inoculation), while Diplodinium, Ostracodinium and Dasytricha last (68–87 days aftertransferring). However, the time of appearance of protozoa depended on the origin ofthe inoculum (results are summarized in table 3).

It can be concluded on the basis of the cited findings that independent of the formof inoculation Entodinia becomes established first, often followed by other ento-diniomorphids and/or large holotrichs. It is noteworthy that Entodinia appear to bethe first of the ciliates that appeared in ruminant ancestors (Lubinsky, 1957a,b).

7. FACTORS AFFECTING ESTABLISHMENT OF CILIATES IN THE RUMEN

Different factors are considered to affect colonization of the rumen by protozoa.Williams and Dinusson (1972) showed that the number of ciliates transferred fromone animal to another influences the length of the period between the inoculationand establishment of the ciliate fauna in calves. Diet is also a factor affecting thetime taken for a population to develop. For example, no increase in ciliate numberswas observed for a period of over 11 weeks when calves were fed a milk-containingdiet (Lengemann and Allen, 1959). On the other hand, a milk diet inhibited theanatomical growth and functional development of the rumen (McGilliard et al.,1965; Warner and Flatt, 1965). In contrast to that, Eadie (1962) observed a positiveeffect of milk on the development of ciliate fauna even in the rumen of the early-weaned calves. According to the same author, the acidity of the rumen digesta is acrucial factor affecting the establishment of ciliates in young ruminants. At a pH below 6.0 only a small number of ciliates could be detected. At a little above pH6.0 species from the genus Entodinium were most frequent while at pH above 6.5,mixed fauna including populations of large ophryoscolecids became established. Ifthe pH in the rumen of calves with developed mixed ciliate fauna dropped, largeophryoscolecids disappeared first whereas Entodinia and Isotricha spp. were the last. It is well documented that feeding concentrate resulted in a drop in pH, whichis accompanied by the disappearance of ciliates from the rumen. In contrast to concentrate, roughage feed favoured a pH increase and thus stimulated both the development and establishment of adult-type ciliate fauna in calves and

Tabl

e 3.

App

eara

nce

of c

iliat

es in

the

rum

en o

f gr

owin

g ci

liate

-fre

e an

imal

s fo

llow

ing

expe

rim

enta

l ino

cula

tion

Ref

eren

ces

Item

56

23

14

47

7

Ani

mal

sC

ow c

alve

sC

ow c

alve

sC

ow c

alve

sC

ow c

alve

sB

uffa

lo c

alve

sL

ambs

Kid

sB

uffa

lo c

alve

sC

ow c

alve

s

Inoc

ulat

ion

age

5,10

,15,

21d

1,3,

6w13

–15w

2–3w

18,2

6,49

w13

–18w

New

born

App

eara

nce

of:

Und

efin

ed p

roto

zoa

3,6w

Ent

odin

ium

3–6w

17–2

0w>

4d7a

,7b ,

7cd

40a ,4

b ,4c

dD

iplo

dini

um3–

9w27

w>

4–6d

14,1

9,7d

87,3

8,8d

Eud

iplo

dini

um>

6d14

,×,1

7d49

,6d

Met

adin

ium

>28

wO

stra

codi

nium

39,×

,20d

87d

Poly

plas

tron

7d*

8d*

Ely

trop

last

ron

58d

Oph

ryos

cole

x51

,61,

104w

34,5

1wIs

otri

cha

>6d

28,4

2,20

d42

,22,

8dD

asyt

rich

a6–

9w33

–37w

59,7

d68

,8d

d =

day

s. w

= w

eeks

.a,

b,c

= in

ocul

ated

with

rum

en c

onte

nt o

f co

w, b

uffa

lo a

nd s

heep

, res

pect

ivel

y.*

= d

isap

pear

ed a

fter

a f

ew d

ays.

×=

not

pre

sent

in in

ocul

um.

Ref

eren

ces:

1 =

Bor

ham

i et a

l. (1

967)

, 2 =

Bry

ant a

nd S

mal

l (19

60),

3 =

Bry

ant e

t al.

(195

8), 4

= E

adie

(19

67),

5 =

Pou

nden

and

Hib

bs (

1948

), 6

= P

ound

en a

nd

Hib

bs (

1949

), 7

= N

aga

et a

l. (1

969)

.


lambs (Eadie, 1962). Other factors are the weaning systems and to some extent hostspecificity (Naga et al., 1969). The development of ruminal bacteria cannot there-fore be ruled out. Some relations between establishment of the bacterial flora and cil-iate fauna in meroxenic as well as conventional and conventionalized lambs werestudied by Fonty et al. (1983, 1984, 1988). The authors found that a period longerthan 30 days was necessary to establish protozoa in germ-free reared lambs.Conversely, ciliates required only 4 days to become established when the rumen wascolonized by bacteria prior to the isolation of the lambs. They concluded that a well-developed and complex bacterial flora is necessary to establish protozoa in therumen. Indeed bacteria are necessary in the rumen at least to produce the environ-mental conditions such as appropriate acidity and redox potential (Fonty et al.,1983). Bacteria seem also to be the main source of amino acids for protozoa(Coleman, 1989). On the other hand, Lengemann and Allen (1959) found that addition of aureomycin to the diet favoured the establishment of protozoal fauna in calves.

8. EFFECT OF PROTOZOA ON RUMEN METABOLISM IN CALVES AND LAMBS

The presence of ciliates in the rumen of adult ruminants enhances the deaminationprocesses and results in an increase in ammonia concentrations in the rumen fluid(Jouany et al., 1988). A similar effect was observed in lambs and calves independ-ently of age or body weight (table 4). However, a tendency to reverse the relationshipwas found by Demeyer et al. (1982) in lambs fed a diet based on alkali-treated straw.The author is of the opinion that large doses of urea supplemented to the diets wereresponsible for the observed enhancement in the level of ammonia in the rumen ofexperimental animals.

In the majority of performed experiments the presence of ciliates in the rumen ofcalves and lambs resulted in higher concentration of volatile fatty acids (VFA) compared with ciliate-free animals (table 5). This suggests that ciliates positively affectthe metabolism of dietary carbohydrates. Establishment of protozoa also resulted in adecrease in or at least a tendency to decrease the proportion of acetate with a simulta-neous increase in the proportion of propionate. Conversely, the proportion of butyricacid decreased or increased in relation to the authors and experiments. For example,butyrate decreased to an undetectable level in lambs faunated with mixed-type ciliatefauna (Abou Akkada and El-Shazly, 1964), whereas it increased by over 40% follow-ing colonization of the rumen of ciliate-free lambs by Diplodinium sp. (Christiansen et al., 1965). A similar reaction followed the establishment of Eudiplodinium maggii inthe rumen of defaunated adult sheep (Michalowski et al., 2001a).

The cited results show that independently of diet and age colonization of therumen habitat in growing calves and lambs by protozoa led to both an increase inVFA concentration and a shift in the pattern of carbohydrate metabolism. It has beenalready described earlier that volatile fatty acids have an important role in accelerating

Tabl

e 4.

The

eff

ect o

f ab

senc

e (–

P) o

r pr

esen

ce (

+P)

of

cilia

tes

on a

mm

onia

con

cent

ratio

n in

the

rum

en f

luid

of

grow

ing

dom

estic

rum

inan

ts

Am

mon

ia m

g/10

0 m

l

Ani

mal

Age

or

wei

ght

Die

t–P

+P

Sign

ific

ance

Ref

eren

ces

Lam

bs4–

5 m

onth

sC

otto

n se

ed, r

ice

bran

mix

ture

2–3

3.5–

7n.

dA

bou

Akk

ada

and

El-

Shaz

ly, 1

964

5–8

mon

ths

Con

cent

rate

mix

ture

2–6.

51.

5–11

by 2

4 w

eeks

Not

giv

en4.

58.

2n.

dA

bou

Akk

ada,

196

529

–38

kgA

lfal

fa h

ay, c

once

ntra

te m

ixtu

re6.

49.

2–12

.2n.

dC

hris

tians

en e

t al.,

196

5A

lfal

fa h

ay 8

0%, c

once

ntra

tes

20%

6.5

9.6

n.d

Lut

her

et a

l., 1

966

25 k

gA

lfal

fa h

ay 2

0%, c

once

ntra

tes

80%

6.0

14.4

Alf

alfa

hay

, con

cent

rate

3.5–

8.3

8–16

0.01

Klo

pfen

stei

n et

al.,

196

625

–27

kgC

once

ntra

te, u

rea

2–8

4–15

0.01

26 w

eeks

Dri

ed g

rass

8.4–

9.6

3.3–

19.7

0.00

1E

adie

and

Gill

, 197

1B

eet p

ulp,

mol

asse

s, u

rea

22.1

29.6

0.01

20 k

gA

lkal

i-tr

eate

d st

raw

, mol

asse

s, u

rea

27.0

22.7

nsD

emey

er e

t al.,

198

2A

s ab

ove

+ ta

pioc

a17

.715

.2ns

14.4

–21.

9 kg

Mol

asse

s, b

eef

prom

ol10

.913

.3ns

Van

Nev

el e

t al.,

198

5B

uffa

lo c

alve

sM

ilk10

–28

10–1

62–

18 w

eeks

Milk

, con

cent

rate

, mol

asse

s, r

ice

bran

11–2

010

–28

n.d

Bor

ham

i et a

l., 1

967

Milk

, con

cent

rate

, mol

asse

s7–

1811

–29

n.d

= n

ot d

eter

min

ed. n

s =

not

sig

nifi

cant

.


the metabolic maturation of the rumen epithelium. The effect of protozoa on VFAproduction (Michalowski, 1987) suggests that ciliates can be considered as a factorpositively affecting the functional maturation of the mucosa of the rumen.

9. INFLUENCE OF PROTOZOA ON GROWTH OF YOUNG RUMINANTS

The effect of the presence of the ciliates in the rumen on the growth of calves andlambs was examined by several authors (table 6). A number of the investigationsobserved either a positive or a neutral influence of ciliates on weight gain of ani-mals. However, a negative effect was also found. For example Bird et al. (1979)found a negative growth rate in faunated lambs fed a diet composed of oaten chaff,sugar, urea and minerals. It was also found that a fish meal supplement abolishedthis effect independently of the supplementation level. Moreover, the growth rate offaunated animals tended to be higher than that of ciliate-free when fish meal wasadded at a proportion of 102 g/kg diet. Similarly Demeyer et al. (1982) and Van Nevelet al. (1985) observed that the effect of ciliates on the growth rate of lambs dependedon the diet. Ciliates diminished the growth rate of young animals when their dietwas based on molasses and straw. Partial replacement of alkali-treated straw withtapioca was the factor in increasing the daily gain of the faunated animals andimproved the food conversion efficiency.

The nutritional behaviour of ciliates and the chemical composition of feed appear tobe very significant factors when the effect of protozoa on the growth rate of ruminants

Table 5. Total concentration (mmol/100 ml) of volatile fatty acids and molar proportions ofindividual acids in the rumen of ciliate-free (–P) and faunated (+P) growing domestic ruminants

Total VFA Acetate Propionate ButyrateAnimals

–P +P –P +P –P +P –P +P

References

Lambs 4–5 3–6.5 77.5 67.5 14.9 32.5 7.6 0 Abou Akkada and 3–8 5–9.5 75.9 68.5 21.1 24.9 1.2 0 El-Shazly, 19647.1 8.5 not determined Abou Akkada, 19656.3 7.6 62.3 54.8 21.4 24.6 15.2 17.4 Christiansen et al., 19657.0 7.9 50.8 47.8 30.8 30.7 15.5 18.1 Luther et al., 19667.4 7.4 38.7 38.7 26.0 34.0 30.5 23.85.7 7.3 77.3 71.0 14.6 19.9 7.8 9.1 Eadie and Gill, 197110.6 9.7 66.1 67.6 22.9 24.9 10.2 7.0 Demeyer et al., 19828.5 9.0 76.4 73.4 17.1 19.1 6.5 7.56.3 8.0 58.8 60.5 29.4 31.8 10.8 8.07.5 7.8 58.0 56.8 18.9 25.8 22.0 17.2 Van Nevel et al., 1985

Buffalo 2.5–5.5 1.9–6.5calves 2.8–4.7 3.6–7.2 not determined Borhami et al., 1967

2.4–5.5 2.5–6.53.5–6.0 2.7–8.3

Diet and age or weight of animals are given in table 4.

Tabl

e 6.

The

eff

ect o

f ab

senc

e (–

P) o

r pr

esen

ce (

+P)

of

cilia

tes

in th

e ru

men

of

grow

ing

dom

estic

rum

inan

ts o

n liv

e w

eigh

t gai

n (g

/day

), d

aily

fee

d in

take

(g)

and

fee

dco

nver

sion

(g/

g ga

in)

Wei

ght g

ain

Feed

inta

keFe

ed c

onve

rsio

n

Ani

mal

sA

ge o

r w

eigh

tD

iet

–P+

P–P

+P

–P+

PR

efer

ence

s

Lam

bs48

–51

days

Ber

esee

m h

ay, c

once

ntra

te m

ixtu

re92

122.

8no

t det

erm

ined

Abo

u A

kkad

a an

d E

l-Sh

azly

, 196

4up

to 2

4 w

eeks

Con

cent

rate

?91

116

not d

eter

min

ed12

.3a

128.

3A

bou

Akk

ada,

196

531

kg

Alf

alfa

hay

, con

cent

rate

mix

ture

191

136.

6no

t det

erm

ined

1.9

73C

hris

tians

en e

t al.,

196

57–

13 w

eeks

221

95.7

270

100

not d

eter

min

ed14

–21

wee

ksD

ried

gra

ss (

ad l

ibit

um)

227

121.

884

310

6.8

not d

eter

min

ed22

–29

wee

ks14

498

.011

7110

3.7

not d

eter

min

edE

adie

and

Gill

, 197

1

36–5

9 w

eeks

Dri

ed g

rass

, con

cent

rate

s (r

estr

icte

d)83

86.2

1014

98.6

not d

eter

min

ed21

–21.

7 kg

Oat

en c

haff

, sug

ar, u

rea

37–3

1.1

454

82.2

37.3

–21

.2–2

2.4

kgA

s ab

ove

+ f

ish

mea

l (3

%)

133

56.4

693

93.1

5.3

160.

4B

ird

et a

l., 1

979

21.8

kg

As

abov

e +

fis

h m

eal

(7%

)15

991

.868

596

.44.

410

6.8

22.5

kg

As

abov

e +

fis

h m

eal (

10.2

%)

154

116.

273

510

1.4

4985

.7B

eet p

ulp,

mol

asse

s, u

rea

213

84.9

857

101.

64.

111

9.8

20 k

gA

lkal

i-tr

eate

d st

raw

, mol

asse

s, u

rea

140

72.9

964

91.1

7.2

123.

8D

emey

er e

t al.,

198

2A

s ab

ove

+ ta

pioc

a19

212

4.5

1895

93.2

10.2

74.2

16 k

gO

aten

cha

ff, s

ugar

, ure

a13

3.5

91.4

not d

eter

min

ed6.

810

7.4

Bir

d an

d L

eng,

198

414

.4–2

1.9

kgM

olas

ses,

bee

f pr

omol

(0–

5 w

eeks

)12

570

.410

8593

.58.

513

1.5

As

abov

e (5

–9 w

eeks

)99

110.

111

8998

.111

.783

.8V

an N

evel

et a

l., 1

985

Cow

3–

6 w

eeks

Col

ostr

um u

p to

day

4, t

hen

hay

170

100

not d

eter

min

edPo

unde

n an

d H

ibbs

, 194

8ca

lves

66 d

ays

Milk

+ p

astu

re26

%M

29%

Mno

t det

erm

ined

Poun

den

and

Hib

bs, 1

949

up to

8 m

onth

sM

ilk, a

lfal

fa h

ay, g

rain

s10

410

2.8

not d

eter

min

edPo

unde

n an

d H

ibbs

, 195

0up

to 1

7 w

eeks

Hay

, gra

ins

520

6616

4861

.5no

t det

erm

ined

Bry

ant a

nd S

mal

l, 19

60B

uffa

lo

2–18

wee

ksM

ilk33

011

5.2

1017

100

3.1

87.1

Bor

ham

i et a

l., 1

967

calv

esM

ilk, c

once

ntra

te, m

olas

ses,

ric

e br

an31

012

9.0

140.

810

3.6

4.5

82.2

As

abov

e w

ith d

iffe

rent

fee

ding

per

iods

240

150.

012

5010

8.6

5.2

71.2

Milk

, con

cent

rate

, mol

asse

s32

012

8.1

1450

91.4

4.6

69.6

Val

ues

conc

erni

ng f

auna

ted

anim

als

are

expr

esse

d as

a p

erce

ntag

e of

thos

e of

the

cilia

te-f

ree.

a=

N r

eten

tion

(g/d

ay).

%M

= g

ain

per

mon

th.


is considered. Ophryoscolecid ciliates dominate in the rumen. They readily ingeststarch and grow well in vitro on insoluble protein (Michalowski, 1989; Williams, 1989)whereas straw, soluble sugars and urea are purely utilized. Engulfment and digestion ofbacteria presumably increases when the quantity of preferred nutrients is insufficient tosatisfy the nutritional requirement of protozoa. This results in diminishing of both themicrobial protein flow to the duodenum and the growth rate of the host. Conversely,diets supporting the development of the ciliate population improve the gain of rumi-nants and their food conversion efficiency (Abou Akkada and El-Shazly, 1964;Christiansen et al., 1965; Borhami et al., 1967; Jouany et al., 1988).

Another factor affecting the role of ciliates appears to be the age of the younggrowing ruminants. Eadie and Gill (1971) observed a positive effect of ciliates onthe growth of only 14–21-week-old lambs fed dried grass. No significant influencewas found in either younger or older animals obtaining the same feed.

10. EFFECT OF CILIATES ON BLOOD COMPONENTS

There are few studies on the blood components in faunated and ciliate-free lambsand/or calves. Abou Akkada (1965) found that reducing sugars, non-protein N,ammonia N and urea N were lower while haemoglobin and protein N were higherin faunated lambs when compared with ciliate-free (table 7). Borhami et al. (1967)measured the glucose concentration as well as the ammonia and urea nitrogen inblood samples of buffalo calves starting from the second week after birth. Theyobserved a continuous decrease in blood sugar from 95–99 to 49–83 mg/100 ml during the next 6 weeks in ciliate-free and 16 weeks in faunated calves. The values

Table 7. Some of the blood components in ciliate-free (–P) and faunated (+P) growing domesticruminants

Animals

Item –P +P References

Reducing sugars, mg/100 ml 84.2 60.8 Abou Akkada, 196568–73 46–67 Borhami et al., 1967

Haemoglobin, g/100 ml 11.2 13.3 Abou Akkada, 1965Protein, mg N/100 ml 2.32 2.69 Abou Akkada, 1965Free amino acids, mg/100 ml 14.2 10.2 Klopfenstein et al., 1966Ammonia, mg N/100 ml 0.9 0.35 Abou Akkada, 1965Urea, mg N/100 ml 15.6 10.1 Abou Akkada, 1965

6–8.5a 9–12.5a Klopfenstein et al., 19663.5–5.5b 3–4.5b Klopfenstein et al., 1966

Oleic acid, % 18.5 23.1 Klopfenstein et al., 1966Linoleic acid, % 32.4 29.6 Klopfenstein et al., 1966Other long chain acids, % 49.1 47.3 Klopfenstein et al., 1966

a wethers fed hay-concentrate diet.b wethers fed concentrate diet.

70 T. Michalowski

noted at the end of the experiment showed, however, a similar relationship to thosecited above. On the other hand, no differences in ammonia and urea N levels wereobserved there in relation to the presence or absence of ciliates. Blood plasma lipidswere examined by Klopfenstein et al. (1966). The authors found that faunation ofthe rumen of 5-month-old wethers resulted in an increase in concentration of oleicacid and in a decrease in that of linoleic acid. On the other hand, the plasma concentration of urea was related to the diet rather than to the protozoa.

11. CONCLUSIONS

The rumen in newborn ruminants is anatomically and functionally immature. Itspostnatal growth is accompanied by development of the microbial ecosystem insidethis organ. Colonization of the rumen by protozoa can begin within the first week oflife of the host, but establishment of the adult type ciliate fauna followed rather itsanatomical and functional maturation. Development and maturation of ruminalfauna depends on several factors among which the age of calves and lambs, directcontact with faunated animals, access to solid food and numerous and complex bacterial flora, responsible for appropriate pH and redox potential of the rumendigesta, are presumably the most important. The role of ciliates in rumen metabo-lism and in the performance of growing ruminants seems to depend on diet and age.


The literature concerning the appearance and establishment of the ciliate fauna inthe rumen of young domestic ruminants is not comprehensive and in the majority ofcases relatively old. On the other hand the objectives of these studies and methodsused by the authors were different. Thus the results were sometimes hardly compa-rable. Because of this further studies of a comparative character would be of value.Of importance would be experiments similar to the excellent studies of Dr MargaretEadie (Eadie, 1962, 1967) from the Rowett Research Institute (Aberdeen, UK) andDr Gerard Fonty with co-workers (Fonty et al., 1984, 1988) from INRA (Clermont-Ferrand, France) to determine the sequence of the appearance and establishment ofciliates in the rumen of calves, lambs and kids faunated by natural transmission fromtheir dams. Studies on the role of ciliates in development and functional maturationof the rumen would also be of value. Jouany et al. (2002) have lately postulated apossible role of undefined factor(s) of host animal origin which could affect theestablishment of some species of ciliates in the rumen of adult ruminants. Similarstudies on calves and lambs would be of special interest. Rumen protozoa seem alsoto be a natural barrier against the pathogens (Newbold et al., 2001) and a factordiminishing diarrhoea. These findings suggest that studies on the involvement of ciliates in both the development of the immune system and the function of the lowerportions of gut during early postnatal life of domestic ruminants seem to be anintriguing future challenge for physiologists and microbiologists.


REFERENCES

Abou Akkada, A.R., 1965. The metabolism of ciliate protozoa in relation to rumen function. In:Dougherty, R.W., Allen, R.S., Burroughs, W., Jacobson, N.L., McGilliard, A.D. (Eds.), Physiology ofDigestion in the Ruminant. Butterworths, London, pp. 335–345.

Abou Akkada, A.R., El-Shazly, K., 1964. Effect of absence of ciliate protozoa from the rumen on micro-bial activity and growth of lambs. Appl. Microbiol. 12, 384–390.

Becker, E.R., Hsiung, T.S., 1929. The method by which ruminants acquire their fauna of infusoria, andremarks concerning experiments on the host specificity of these protozoa. Proc. Natl. Acad. Sci. USA15, 684–690.

Bird, S.H., Leng, R.A., 1984. Further studies on the effects of the presence or absence of protozoa in therumen on live-weight gain and wool growth of sheep. Brit. J. Nutr. 52, 607–611.

Bird, S.H., Hill, M.K., Leng, R.A., 1979. The effect of defaunation of the rumen on the growth of lambson low-protein-high-energy diets. Brit. J. Nutr. 42, 81–87.

Bonhomme, A., Fonty, G., Foglietti, M.J., Weber, M., 1986. Endo-1,4-glucanase and β-glucosidase of theciliate Polyplastron multivesiculatum free of cellulolytic bacteria. Can. J. Microbiol. 32, 219–225.

Borhami, B.E.A., El-Shazly, K., Abou Akkada, A.R., Ahmed, I.A., 1967. Effect of early establishment ofciliate protozoa in the rumen on microbial activity and growth of early weaned buffalo calves. J. DairySci. 50, 1654–1660.

Bryant, M.P., Small, N., 1960. Observation on the ruminal microorganisms of isolated and inoculatedcalves. J. Dairy Sci. 43, 654–667.

Bryant, M.P., Small, N., Bouma, C., Robinson, I., 1958. Studies on the composition of the ruminal floraand fauna of young calves. J. Dairy Sci. 41, 1747–1767.

Cheng, K.-J., Forsberg, C.W., Minato, H., Costerton, J.W., 1991. Microbial ecology and physiology offeed degradation within the rumen. In: Tsuda, T., Sasaki, Y., Kawashima, R. (Eds.), PhysiologicalAspects of Digestion and Metabolism in Ruminants. Academic Press, San Diego, pp. 595–624.

Christiansen, W.C., Kawashima, R., Burroughs, W., 1965. Influence of protozoa upon rumen acid production and live weight gains in lambs. J. Anim. Sci. 24, 730–734.

Clarke, R.T.J., 1977. Methods for studying gut microbes. In: Clarke, R.T.J., Bauchop, T. (Eds.),Microbial Ecology of the Gut. Academic Press, New York, pp. 1–33.

Coleman, G.S., 1985. The cellulase content of 15 species of entodiniomorphid protozoa, mixed bacteriaand plant debris isolated from the ovine rumen. J. Agr. Sci. 104, 349–360.

Coleman, G.S., 1986. The distribution of carboxymethylcellulase between fractions taken from therumen of sheep containing no protozoa or one of five different protozoal populations. J. Agr. Sci. 106,121–127.

Coleman, G.S., 1989. Protozoal-bacterial interaction in the rumen. In: Nolan, J.V., Leng, R.A., Demeyer,D.I. (Eds.), The Roles of Protozoa and Fungi in Ruminant Digestion. Penambul Books, Armidale, pp. 13–27.

Dehority, B.A., 1993. Microbial ecology of cell wall fermentation in forage. In: Jung, H.G., Buxton, D.R.,Hatfield, R.D., Ralph, J. (Eds.), Cell Wall Structure and Digestibility. American Society of Agronomy(ASA), Inc., Crop Science Society of America (CSSA), Inc., Soil Science Society of America(SSSA), Inc., Madison, WI, pp. 425–453.

Dehority, B.A., Orpin, C.G., 1997. Development of, and natural fluctuations in, rumen microbial popu-lations. In: Hobson, P.N., Stewart, C.S. (Eds.), The Rumen Microbial Ecosystem. Blackie Academicand Professional, London, pp. 196–245.

Demeyer, D.I., Van Nevel, C.J., 1986. Influence of substrate and microbial interaction on efficiency ofrumen microbial growth. Reprod. Nutr. Dev. 26, 161–179.

Demeyer, D.I., Van Nevel, C.J. and Van de Voorde, G., 1982. The effect of defaunation on the growth oflambs fed three urea containing diets. Arch. Tierernähr. 32, 595–604.

Devillard, E., Flint, H.J., Scott, K.P., Newbold, C.J., Jouany, J.-P., Wallace, R.J., Forano, E., 2000. Fiber-degrading enzymes from a ruminal protozoan Polyplastron multivesiculatum. Reprod. Nutr.Dev. 40, 193.

Dogiel, V.A., 1927. Monographie der Famile Opryoscolecidae. Arch. Protistkde. 59, 1–288.Eadie, J.M., 1962. The development of rumen microbial population in lamb and calves under various

conditions of management. J. Gen. Microbiol. 29, 563–578.

72 T. Michalowski

Eadie, J.M., 1967. Studies on the ecology of certain rumen ciliate protozoa. J. Gen. Microbiol. 49, 175–194.Eadie, J.M., Gill, J.C., 1971. Effect of the absence of rumen ciliate protozoa on growing lambs fed on a

roughage-concentrate diet. Brit. J. Nutr. 26, 155–167.Embley, T.M., Finlay, B.J., Dyal, P.L., Hirt, R.P., Wilkinson, M., Williams, A.G., 1995. Multiple origins

of anaerobic ciliates with hydrogenosomes within the radiation of aerobic ciliates. Proc. R. Soc.Lond., B. Biol. Sci. 262, 87–93.

Fahey, Jr, G.C., Berger, L.L., 1988. Carbohydrate nutrition of ruminants. In: Church, D.C. (Ed.), TheRuminant Animal Digestive Physiology and Nutrition. Reston, Englewood Cliffs, NJ, pp. 269–297.

Flint, H.J., 1997. The rumen microbial ecosystem – some recent developments. Trends Microbiol. 5, 483–488.Flint, H.J., Forsberg, C.W., 1995. Polysaccharide degradation in the rumen: biochemistry and genetics.

In: Von Engelhardt, W., Leonard-Marek, S., Breves, G., Giesecke, D. (Eds.), Ruminant Physiology:Digestion, Metabolism, Growth and Reproduction. Ferdinand Enke Verlag, Stuttgart, pp. 43–70.

Fonty, G., Gouet, Ph., Jouany, J.-P., Senaud, J., 1983. Ecological factors determining establishment ofcellulolytic bacteria and protozoa in the rumen of meroxenic lambs. J. Gen. Microbiol. 129, 213–223.

Fonty, G., Jouany, J.-P., Senaud, J., Gouet, Ph., Grain, J., 1984. The evolution of microflora, microfaunaand digestion in the rumen of lambs from birth to 4 months. Can. J. Anim. Sci. 64 (Suppl.), 165–166.

Fonty, G., Senaud, J., Jouany, J.-P., Gouet, Ph., 1988. Establishment of ciliate protozoa in the rumen ofconventional and conventionalized lambs: influence of diet and management conditions. Can. J.Microbiol. 34, 235–241.

Hofmann, R.R., 1988. Anatomy of the gastro-intestinal tract. In: Church, D.C. (Ed.), The RuminantAnimal Digestive Physiology and Nutrition. Reston, Englewood Cliffs, NJ, pp. 14–43.

Hume, I.D., Sakaguchi, E., 1991. Patterns of digesta flow and digestion in foregut and hindgut fermen-tors. In: Tsuda, T., Sasaki, Y., Kawashima, R. (Eds.), Physiological Aspects of Digestion andMetabolism in Ruminants. Academic Press, San Diego, pp. 427–451.

Hungate, R.E., 1966. The Rumen and its Microbes. Academic Press, New York.Imai, S., 1998. Phylogenetic taxonomy of rumen ciliate protozoa based on their morphology and distri-

bution. J. Appl. Anim. Res. 13, 17–36.Janis, C., 1976. The evolutionary strategy of the Equidae and the origins of the rumen and cecal diges-

tion. Evolution 30, 757–774.John, A., Ulyatt, M.J., 1984. Measurement of protozoa, using phosphatidyl choline, and bacteria using

nucleic acids, in the duodenal digesta of sheep fed chaffed lucerne (Medicago sativa L.) diets. J. Agr.Sci. 102, 33–44.

Jouany, J.-P., Demeyer, D.I., Grain, J., 1988. Effect of defaunating the rumen. Anim. Feed. Sci. Technol.21, 229–265.

Jouany, J.-P., Nsabimona, E., Kisidayova, E., Michalowski, T., Macheboeuf, D., 2002. Why cannot somespecies of protozoa grow in the rumen. Reprod. Nutr. Dev. 42 (Suppl. 1), S79.

Klopfenstein, T.J., Purser, D.B., Tyznik, W.J., 1966. Effects of defaunation on feed digestibility, rumenmetabolism and blood metabolites. J. Anim. Sci. 25, 765–773.

Kofoid, C.A., MacLennan, R.F., 1930. Ciliates from Bos indicus Linn. I. The genus Entodinium Stein.Univ. Calif. Publ. Zool. 33, 471–544.

Kofoid, C.A., MacLennan, R.F., 1932. Ciliates from Bos indicus Linn. II. A revision of DiplodiniumSchuberg. Univ. Calif. Publ. Zool. 37, 53–152.

Kofoid, C.A., MacLennan, R.F., 1933. Ciliates from Bos indicus Linn. III. Epidinium Cravley,Epiplastron gen. nov., and Ophryoscolex Stein. Univ. Calif. Publ. Zool. 39, 1–34.

Lane, M.A., Jesse, B.W., 1997. Effect of volatile fatty acid infusion on development of neonatal sheeprumen epithelium. J. Dairy Sci. 80, 740–746.

Lane, M.A., Baldwin, R.L., Jesse, B.W., 2000. Sheep rumen metabolic development in response to ageand dietary treatment. J. Anim. Sci. 78, 1990–1996.

Langer, P., 1988. The Mammalian Herbivore Stomach, Comparative Anatomy, Function and Evolution.Gustav Fischer, Stuttgart.

Latteur, B., 1968. Revision systématique de la Famille des Ophryoscolecidae Stein, 1858: sous-Famille desEntodiniinae Lubinsky, 1957, Genere Entodinium Stein, 1858. Ann. Soc. Royal Zool. Belgique 98, 1–41.

Latteur, B., 1969. Revision systématique de la Famille Ophryoscolecidae Stein, 1858: sous-Famille desEntodiniinae Lubinsky, 1957, Genere Entodinium Stein, 1858. Ann. Soc. Royal Zool. Belgique 99, 3–25.


Latteur, B., 1970. Revision systématique de la Famille des Ophryoscolecidae Stein, 1858. Sous-FamilleDiplodiniinae Lubinsky, 1957. Genere Diplodinium (Schuberg, 1888) sensu novo. Ann. Soc. RoyalZool. Belgique 100, 275–312.

Lee, J.J., Hutner, S., Bovee, E.C., 1985. An Illustrated Guide to the Protozoa. Society of Protozoologists,Lawrence, KS, p. 629.

Lengemann, F.W., Allen, N.N., 1959. Development of rumen function in the dairy calf. II. Effect of dietupon characteristics of the rumen flora and fauna of young calves. J. Dairy Sci. 42, 1171–1181.

Levine, N.D., Corlis, J.O., Cox, F.E.G., Deroux, G., Grain, J., Honigberg, B.M., Leedale, G.F., Loebich, A.R.,Lom, J., Lynn, D., Merinfeld, E.G., Page, F.C., Poljansky, G., Sprague, V., Vavra, J., Wallace, F.G., 1980.A newly revised classification of the protozoa. J. Protozool. 27, 37–58.

Lubinsky, G., 1957a. Studies on the evolution of Ophryoscolecidae (Ciliata; Oligotricha). I. A newspecies of Entodinium with “caudatum”, “loboso-spinosum” and “dubardi” forms and some evolu-tionary trends in the genus Entodinium. Can. J. Zool. 35, 111–133.

Lubinsky, G., 1957b. Studies on the evolution of the Ophryoscolecidae (Ciliata: Oligotricha). III. Phylogenyof the Ophryoscolecidae based on their comparative morphology. Can. J. Zool. 35, 141–159.

Luther, R., Trenkle, A., Burroughs, W., 1966. Influence of rumen protozoa on volatile fatty acid produc-tion and ration digestibility in lambs. J. Anim. Sci. 25, 1116–1122.

Lyford, Jr, S.J., 1988. Growth and Development of the ruminant digestive system. In: Church, D.C. (Ed.),The Ruminant Animal Digestive Physiology and Nutrition. Reston, Englewood Cliffs, NJ, pp. 44–63.

McBee, R.H., 1977. Fermentation in the hindgut. In: Clarke, R.T.J., Bauchop, T. (Eds.), MicrobialEcology of the Gut. Academic Press, New York, pp. 185–222.

McGilliard, A.D., Jacobson, N.L., Sutton, J.D., 1965. Physiological development of the ruminant stom-ach. In: Dougherty, R.W., Allen, R.S., Burroughs, W., Jacobson, N.L., McGilliard, A.D. (Eds.),Physiology of Digestion in the Ruminant. Butterworths, London, pp. 39–50.

Michalowski, T., 1987. The volatile fatty acid production by ciliate protozoa in the rumen of sheep. ActaProtozool. 26, 335–345.

Michalowski, T., 1989. Importance of protein solubility and nature of dietary protein for the growth ofrumen ciliates in vitro. In: Nolan, J.V., Leng, R.A., Demeyer, D.I. (Eds.), The Roles of Protozoa andFungi in Ruminant Digestion. Penambul Books, Armidale, pp. 223–231.

Michalowski, T., 1990. The synthesis and turnover of the cellular matter of ciliates in the rumen. ActaProtozool. 29, 47–72.

Michalowski, T., 1997. Digestion and fermentation of the microcrystalline cellulose by the rumen ciliateprotozoon Eudiplodinium maggii. Acta Protozool. 36, 181–185.

Michalowski, T., Harmeyer, J., 1998. The effect of the rumen ciliates Eudiplodinium maggii on thexylanase and CMC-ase activity in sheep. In: Van Arendonk, J.A.M., Ducrocq, V., Van der Honing, Y.,Madec, F., Van der Lende, T., Pullar, D., Folch, J., Fernandez, J.A., Bruns, E.W. (Eds.), Book ofAbstracts of the 49th Annual Meeting of the European Association for Animal Production.Wageningen Pers, Wageningen, p. 86.

Michalowski, T., Harmeyer, J., Breves, G., 1986. The passage of protozoa from the reticulo-rumenthrough the omasum of sheep. Brit. J. Nutr. 56, 625–634.

Michalowski, T., Kwiatkowska, E., Belzecki, G., Pajak, J.J., 2001a. Effect of the microfauna composi-tion on fermentation pattern in the rumen of sheep. J. Anim. Feed. Sci. 10 (Suppl. 2), 135–140.

Michalowski, T., Rybicka, K., Wereszka, K., Kasperowicz, A., 2001b. Ability of the rumen ciliateEpidinium ecaudatum to digest and use crystalline cellulose and xylan for in vitro growth. ActaProtozool. 40, 203–210.

Naga, M.A., Abou Akkada, A.R., El-Shazly, K., 1969. Establishment of rumen ciliate protozoa in cows andwater buffalo (Bos bubalus L.) calves under late and early weaning system. J. Dairy Sci. 52, 110–112.

Newbold, C.J., Stewart, C.S., Wallace, R.J., 2001. Developments in rumen fermentation – the scientificview. In: Gornsworthy, P.C., Wiseman, J. (Eds.), Recent Advances in Animal Nutrition. UniversityPress, Nottingham.

Noirot-Timothée, C., 1960. Etude d’une famille de ciliés: les Ophryoscolecidae. Structures et ultrastruc-tures. Ann. Sci. Nat. Zool. Biol. Anim. 2, 527–718.

Orpin, C.G., Joblin, K.N., 1997. The rumen fungi. In: Hobson, P.N., Stewart, C.S. (Eds.), The RumenMicrobial Ecosystem. Blackie Academic and Professional, London, pp. 140–195.

74 T. Michalowski

Owens, F.N., Goetsch, A.L., 1988. Ruminal fermentation. In: Church, D.C. (Ed.), The Ruminant AnimalDigestive Physiology and Nutrition. Reston, Englewood Cliffs, NJ, pp. 145–171.

Pounden, W.D., Hibbs, J.W., 1948. The influence of the ration and rumen inoculation and the establish-ment of certain microorganisms in the rumen of young calves. J. Dairy Sci. 31, 1041–1050.

Pounden, W.D., Hibbs, J.W., 1949. The influence of pasture and rumen inoculation on the establishmentof certain microorganisms in the rumen of young dairy calves. J. Dairy Sci. 32, 1025–1031.

Pounden, W.D., Hibbs, J.W., 1950. The development of calves raised without protozoa and certain othercharacteristic rumen microorganisms. J. Dairy Sci. 33, 639–644.

Prins, R.A., Van Hoven, W., 1977. Carbohydrate fermentation by the rumen ciliate Isotricha prostoma.Protistologica 13, 549–556.

Russell, J.B., Rychlik, J.L., 2001. Factors that alter rumen microbial ecology. Science 292, 1119–1122.Selinger, L.B., Forsberg, C.W., Cheng, K.-J., 1996. The rumen: a unique source of enzymes for enhanc-

ing livestock production. Anaerobe 2, 263–284.Stewart, C.S., Flint, H.J., Bryant, M.P., 1997. The rumen bacteria. In: Hobson, P.N., Stewart, C.S. (Eds.),

The Rumen Microbial Ecosystem. Blackie Academic and Professional, London, pp. 10–72.Swan, G.E., Groenewald, H.B., 2000. Morphological changes associated with the development of the

rumino-reticulum in growing lambs fed different rations. Onderstpoort J. Vet. Res. 67, 105–114.Theodorou, M.K., Mennim, G., Davies, D.R., Zhu, W.-Y., Trinci, A.P.J., Brookman, J.L., 1996.

Anaerobic fungi in the digestive tract of mammalian herbivores and their potential for exploitation.Proc. Nutr. Soc. 55, 913–926.

Van Hoven, W., Prins, R.A., 1977. Carbohydrate fermentation by the rumen ciliate Dasytricha ruminan-tium. Protistologica 13, 599–606.

Van Nevel, C.J., Demeyer, D.I., Van de Voorde, G., 1985. Effect of defaunating the rumen on growth andcarcass composition of lambs. Arch. Tierernähr. 35, 331–337.

Van Soest, P.J., 1982. Nutritional Ecology of the Ruminant. O&B Books, Corvallis.Warner, R.G., Flatt, W.P., 1965. Anatomic development of the ruminant stomach. In: Dougherty, R.R.,

Allen, R.S., Burroughs, W., Jacobson, N.L., McGilliard, A.D. (Eds.), Physiology of Digestion in theRuminant. Butterworths, London, pp. 24–38.

Weller, R.A., Pilgrim, A.F., 1974. Passage of protozoa and volatile fatty acids from the rumen of thesheep and from a continuous in vitro fermentation system. Brit. J. Nutr. 32, 341–351.

Williams, A.G., 1989. Metabolic activity of rumen protozoa. In: Nolan, J.V., Leng, R.A., Demeyer, D.I.(Eds.), The Roles of Protozoa and Fungi in Ruminant Digestion. Penambul Books, Armidale, pp. 97–126.

Williams, A.G., Coleman, G.S., 1992. The Rumen Protozoa. Springer-Verlag, New York.Williams, P.P., Dinusson, W.E., 1972. Composition of the ruminal flora and establishment of ruminal cil-

iated protozoal species in isolated calves. J. Anim. Sci. 34, 469–474.Wright, A.-D.G., Lynn D.H., 1997. Phylogenetic analysis of the rumen ciliate family Ophryoscolecidae

based on 18S ribosomal RNA sequence, with new sequences from Diplodinium, Eudiplodinium, andOphryoscolex. Can. J. Zool. 75, 963–970.

Wright, A.-D.G., Dehority, B.A., Lynn, D.H., 1997. Phylogeny of the rumen ciliates Entodinium,Epidinium and Polyplastron (Litostomata: Entodiniomorphida) inferred from small subunit ribosomalRNA sequences. J. Euk. Microbiol. 44, 61–67.

Reindeer populations in the Arctic region are limited by the availability and the utilization of the diet they eat. Data on seasonal changes in the gastrointestinal micro-biota were reviewed in reindeer in northern Norway and in Svalbard reindeer.Digestion in reindeer depends on a highly active anaerobic symbiotic rumen bacterialpopulation, ciliated protozoa and anaerobic fungi, compartmentalized in the rumenfluid, plant solid digesta and epithelial mucosa. The distal fermentation chamber alsoharbours anaerobic fermentative bacteria. Cultivation studies indicate that the numberand composition of the microorganisms change with the season and chemistry of theforage consumed. The Svalbard reindeer have large populations of cellulolytic bacte-ria in their rumen in winter making their digestive system highly suitable for energyutilization of poor quality food through slow rumen breakdown and fermentation;rumen cellulolysis, however, is more rapid in Norwegian reindeer in winter than inSvalbard reindeer. Digestion of plant polysaccharides may be limited by the availabil-ity of easily digestible energy and non-protein nitrogen available for microbial synthesis. Reindeer, unlike domestic ruminants, are highly adaptable mixed feeders.

1. INTRODUCTION

This chapter reviews the existing literature on the gastrointestinal microbiota in twodifferent sub-species of reindeer: Svalbard reindeer (Rangifer tarandus plathyrynchus)living on the high-Arctic archipelago of Svalbard (74–81°N) and Norwegian reindeer

4 Microbial ecology of the digestive tract inreindeer: seasonal changes

S.D. Mathiesena, R.I. Mackieb, A. Aschfalka, E. Ringøa and M.A. Sundsetc

aSection of Arctic Veterinary Medicine, Department of Food Safety and InfectionBiology, The Norwegian School of Veterinary Science, NO-9292 Tromsø, NorwaybDepartment of Animal Sciences, University of Illinois at Urbana-Champaign,Urbana, Illinois 61801, USAcDepartment of Arctic Biology and Institute of Medical Biology, University ofTromsø, NO-9037 Tromsø, Norway

75



(Rangifer tarandus tarandus) semi-domesticated by the Saami people on mainlandNorway (69°N) (figs 1 and 2). Reindeer have a four-chambered stomach system con-sisting of the reticulum, rumen, omasum and abomasum (fig. 3), just like other rumi-nants. No major differences are observed in the gross anatomy of the gastrointestinaltract in Svalbard reindeer and Norwegian reindeer (Westerling, 1975b; Staaland et al.,1979; Sørmo, 1998; Utsi, 1998; Sørmo et al., 1999). However, Staaland et al. (1979),emphasized the importance of the large distal fermentation chamber in mineralabsorption in Svalbard reindeer compared to Norwegian reindeer, as well as their shortintestines (fig. 1).

76 S. D. Mathiesen et al.

Fig. 1. Reindeer appeared 15 million years ago and their circumpolar distribution in the northern hemisphere currently comprises about 5 million individuals divided between seven sub-subspecies: Eurasian tundra reindeer/Norwegian reindeer Rangifer tarandus tarandus (A); Eurasian forest reindeer R. t. fennicus (B); Alaskan caribou R. t. granti (C); Woodland caribou R. t. caribou (D); barren-ground caribou R. t. groenlandicus (E); Peary caribou R. t. pearyi (F); Svalbard reindeer R. t. platyrhynchus (G).(Courtesy of Dr Nicholas J.C. Tyler, University of Tromsø, Norway.)

Our knowledge of the rumen metabolism has increased greatly since Aristotlefirst described the multiple ruminant stomach (Mason, 1962). However, today’sunderstanding of the ruminant gastrointestinal microbial ecosystem is primarilybased on studies of artificially fed domestic sheep and cattle (Hespell et al., 1997),and only a few limited studies have been conducted on wild ruminants such as muledeer (Pearson, 1969), elk (McBee et al., 1969), buffalo (Pearson, 1967) and camel

Microbial ecology of reindeer digestive tract 77

Fig. 2. Reindeer experience large seasonal changes in forage availability and quality. These photographsshow Norwegian reindeer in their natural environment in northern Norway and Svalbard reindeer feeding on the high-Arctic desert of Svalbard winter and summer. (Photos: S.D. Mathiesen and M.A. Sundset.)

Fig. 3. Both Norwegian reindeer and Svalbard reindeer have been classified as intermediate feeders and nomajor differences are observed in the gross anatomy of their gastrointestinal tract, although Svalbard reindeerhave a larger distal fermentation chamber. This figure shows a photograph of the gastrointestinal tract ofNorwegian reindeer. Scale bar = 10 cm. (Photo: S. D. Mathiesen.)

(Hungate et al., 1959). Hence, relatively little is known on wild ruminants and howtheir gastrointestinal microbiota has developed through a long history of feeding onnatural pastures. Ruminant evolution began more than 30 million years ago duringthe Eocene period under very different climatic conditions. Ruminants such asmoose (Alces alces) and roe deer (Capreolus carpools) belong to an older type ofruminant, while bovine species first evolved when cellulosic grasses became abun-dant later in the Miocene (Hofmann, 1989, 1999). Reindeer are classified as highlyadaptable feeders intermediate to roe deer and the bovine species and are one of 180different ruminant species in the world. According to Randi et al. (1998) reindeerdeveloped about 15 million years ago. Their circumpolar distribution currently comprises about 5 million individuals, divided between seven different living sub-species (fig. 1) and spanning a latitudinal range of 50−82°N (Banfield, 1961).Throughout history, global environmental changes have had a powerful influence onclimatic conditions in the Arctic region and have thus limited plant availability andgrowth. The last glaciation in northern Europe came to an end 10 000−20 000 yearsago or even as long as 40 000 years ago on Svalbard which was later occupied byreindeer (Salvigsen, 1979). Persistent climatic instabilities have always affectedindigenous herbivores, the availability of their forage plants and their ability to utilize plant carbohydrates and proteins for maintenance energy, growth and repro-duction. Likewise, growth and survival of reindeer in the circumpolar region isstrongly dependent on seasonal climatic factors. The Svalbard reindeer experienceextreme variations in daylength through the year. From mid-November until mid-February the light intensity remains below civil twilight, while from April untilmid-August the sun never sets. Norwegian reindeer living in northern Norway expe-rience two hours twilight daily for two months in mid-winter and an equally longperiod of midnight sun in summer. It is assumed that these seasonal changes indaylength have a substantial influence on the intrinsic seasonal physiology of theseanimals, including appetite and reproduction (Stokkan et al., 1994). All northernruminants investigated so far show pronounced seasonal changes in appetite andgrowth; voluntary food intake and rate of growth being high in summer and low inwinter (Ryg and Jacobsen, 1982; Larsen et al., 1985; Tyler, 1987; Nilssen et al.,1994; Tyler et al., 1999).

Norwegian reindeer select and eat a variety of plants (table 1), including shrubs(e.g. Empetrum spp., Loiseleuira procumbers, Vaccinium spp.), birch (Betula spp.),willow (Salix spp.), grasses (e.g. Poa spp., Deschampsia spp.) and sedges (Carexspp.) from 37 different plant families. In early spring reindeer also eat differentherbs (Rumex acetosa, Ranuculus repens and Alchemilla subcrenata) and grasses(Festuca rubra, Poa pratensis, Agrostis capillaris, Deschampsia caespitosa andPhleum alpinum) along the coast of northern Norway when snow still covers the mountain vegetation. The metabolic demands of Arctic ruminants are high insummer when body growth and appetite are high and much reduced in winter(Nilssen et al., 1984, 1994; Fancy and White, 1985). This seasonal fluctuation in


growth represents a major adaptation to seasonal variation in the quality and avail-ability of forage, which strongly influences the gastrointestinal microbiota andmetabolism (Mathiesen, 1999). Svalbard reindeer forage on the tundra or on polardesert vegetation all year round (table 1). Their diet is generally dominated bySaxifraga oppositifolia, but grasses are also eaten in both seasons. Also sedges (e.g.Deschampsia spp., Dupontia spp., Poa spp., Carex spp., Luzula spp.), shrubs (e.g.Dryas octopetala and Salix polaris), herbs (e.g. Saxifraga spp.) and mosses arefound in the rumen of Svalbard reindeer (Sørmo et al., 1999). Over thousands ofyears these reindeer have removed the dietary lichens by grazing and trampling andthese therefore no longer form a significant part of their winter diet. Svalbard rein-deer mobilize a large proportion of their energy and protein reserves in winter result-ing in a substantial decline in carcass mass from 72 kg in summer to 46 kg in winter(Tyler, 1987). However, in free-living Norwegian reindeer the live body massdecreased from summer (69 kg) to winter (59 kg) but they suffered no net declinein carcass mass in winter (Mathiesen et al., 2000).

2. RUMEN FERMENTATION AND MICROBIOLOGY

In contrast to domestic ruminants living in more stable nutritional and climatic envi-ronments, the understanding of the basic digestive physiology of Arctic ungulateshas to be considered in the light of the large seasonal variations outlined above. Verydifferent gastrointestinal microfloras have evolved in the two reindeer populationsstudied − most probably due to differences in environmental conditions betweenSvalbard and northern Norway.

The gastrointestinal tract of newborn ruminants is colonized from birth. Therumen in reindeer contains a complex consortia of microorganisms that live in amutualistic relationship with the host (Utsi, 1998; Sørmo, 1998; Mathiesen, 1999;Olsen, 2000). The dominant populations of microorganisms consist of anaerobicbacteria (Bacteria), methanogens (Archaea), ciliates and anaerobic fungi (Eucarya),compartmentalized into different populations associated with the rumen fluid, feed


Table 1. Mean ruminal (% of total) composition of main groups of dietary plants in Norwegianreindeer and Svalbard reindeer in different seasons (Mathiesen, 1999)

Grasses Woody plants Lichens Mosses

Norwegian reindeerSummer pasture 65 17 11 4Winter pasture 21 36 35 7

Svalbard reindeerSummer pasture 40 44 0 16Winter pasture 35 55 0 10

particles and the rumen wall. Microbial enzymatic breakdown of complex plantcomponents in the rumen results in fermentation products such as short chain fattyacids (SCFA), CO2 and CH4.

Energy-rich SCFA including acetate, butyrate and propionate are absorbedacross the rumen wall and may support up to 70% of the daily energy requirementsof the host (Hungate, 1966; Annison and Armstrong, 1970). In domestic ruminantsthe concentration and the size of the SCFA pool in the rumen is correlated to the rateof production of SCFA and reflects forage quality (Leng and Brett, 1966; Leng et al., 1968; Weston and Hogan, 1968). Few measurements of rumen fermentationrates have been conducted in wild ruminants (White and Grau, 1975; White andStaaland, 1983; Lechner-Doll et al., 1991; Boomker, 1995; Sørmo et al., 1997).Lechner-Doll et al. (1990) observed that the rate of dilution, rumen fluid volume andrate of absorption influenced concentration of SCFA in the rumen fluid far morethan the rate of production in African domestic ruminants. Likewise, ruminal SCFAconcentration only partly reflects the forage quality eaten in Arctic ruminants, aslarge seasonal differences in food intake probably also influence ruminal retentionand rate of absorption in these animals (Sørmo et al., 1997). The SCFA productionrate does, however, seem to reflect the quality of the pasture. On Svalbard the for-age quality is very high in summer, but almost negligible amounts of SCFA wereproduced in the rumens in winter, reflecting the poor quality of the winter diet. Incontrast, the Norwegian reindeer maintain a high SCFA production in winter whenthey eat significant amounts of lichens (Mathiesen et al., 1999). Furthermore, thelate summer pasture in northern Norway was regarded as moderate in terms of car-bohydrate fermentation in the rumen compared to Svalbard reindeer. The seasonaland sub-species differences in body composition in Svalbard and Norwegian rein-deer were also reflected in differences in their rumen metabolism (Mathiesen et al.,1999). Given the low rate of rumen SCFA production, and the poor in vitro dry matter digestibility recorded in Svalbard reindeer, these animals seem to survive the winter by optimizing the ruminal utilization of the low quality food and byreducing their energy expenditure to a minimum.

When ruminants are exposed to starvation or abrupt dietary changes, the micro-biota colonizing the different parts of the digestive tract changes. Climatic condi-tions in the Arctic often result in natural periods of starvation for reindeer duringwinter, as pastures may be blocked by ice or crusted snow for shorter or longer periods of time. Starvation for 3−4 days reduces the numbers of culturable anaerobicrumen bacteria in reindeer by as much as 99.7%, and also changes the compositionof the population and its ability to digest cellulose (Mathiesen et al., 1984b; Aagneset al., 1995; Olsen, 2000). Likewise, starvation has a pronounced effect on therumen ciliate population, which is decreased by 75% after 3 days starvation(Mathiesen et al., 1984b). Furthermore, the ruminal dry matter content and the concentration of SCFA decrease, while the ruminal pH increases to as much as 7.8after 4 days starvation. An increased intake of snow and a maintained rumen volume


and turnover time (Aagnes and Mathiesen, 1994), may, however, contribute to sustain the microbial inoculum during starvation.

As much as 13% of the daily digestible food intake in domestic ruminants maybe lost in the form of methane (Hungate, 1966). The density of viable methanogensin Svalbard reindeer, however, seems to be low regardless of season (Mathiesen,1999). Likewise, Øritsland (personal communication) was unable to detect methanefrom the rumen gas phase in live Svalbard reindeer. However, in Norwegian reindeer fed pelleted reindeer concentrate large methane production has beenrecorded (Gotaas and Tyler, 1994). Microbial competition for hydrogen could behigh in the rumen of these reindeer and future investigations should include studiesof the relationship between acetogenic and methanogenic bacteria in Svalbard reindeer. Reduced loss of methane can be of significant importance for the growthof Svalbard reindeer if hydrogen produced by the rumen microorganisms is used toproduce more acetate, rather than methane.

Savage (1989) defined bacteria isolated from the digestive tract of endothermicanimals as either autochthonous (indigenous) or allochthonous (transient), and presented a list of criteria for autochthony. The composition of the microflora isgreatly influenced by the passage rates of fluid and particles through the digestivetract (Van Soest, 1994). But the diet and hence the substrate available for fermenta-tion is also a very important factor influencing the composition and the numbers ofmicroorganisms in the rumen. The rumen microbiology of Arctic ruminants has onlybeen partly investigated due to the considerable time required to return rumen samples to the laboratory from the remote areas in which these animals live.Furthermore, specialized techniques are required for cultivation of these strict anaer-obes and for many years only direct microscopic observations were made of fixedsamples of bacteria and protozoa from Arctic ungulates (Dehority, 1975a; Hobsonet al., 1976). Live bacteria were first isolated from the rumen of Alaskan reindeer byDehority (1975b). During the past 15 years anaerobic techniques for field use havebeen developed to investigate viable anaerobic bacteria in wild sheep, reindeer andmuskoxen (Orpin et al., 1985, Aagnes and Mathiesen, 1995; Aagnes et al., 1995)making it possible to investigate the microbial ecology of the gastrointestinal tractof animals living in areas remote to the laboratory.

3. BACTERIA

3.1. Numbers and composition of the bacterial population in differentniches of the rumen

Seasonal changes in forage quality and availability on Svalbard affect both numbersof viable bacteria in rumen fluid (table 2) and their composition. Bacterial speciesknown to utilize soluble carbohydrates dominated in summer and those that utilizefibrous polysaccharides dominated in winter (table 3). The viable bacterial popula-tion decreased by about 75% from summer to winter but the winter population was



Table 2. Mean reticulo-rumen bacterial numbers in rumen fluid from Norwegian and Svalbardreindeer in different seasons (Orpin et al., 1985; Aagnes et al., 1995; Olsen et al., 1997; Mathiesenand Orpin, unpublished data)

Bacterial cells × 108/ml fluid

Norwegian reindeerSummer pasture 6.4–9.9Winter pasture 5.2–25.0Starved for 4 days 0.1Fed pure lichen 39.0

Svalbard reindeerSummer pasture 209Summer fed concentrate 350Winter pasture 36

Table 3. Cellulolytic bacteria as per cent of total viable bacteria in the rumen fluid of differentruminants (Hungate, 1966; Orpin et al., 1985; Aagnes et al., 1995; Olsen et al., 1997; Mathiesen and Orpin, unpublished data)

Cellulolytic bacteria (% of total population)

Norwegian reindeerSummer pasture 3.4Winter pasture 2.5

Svalbard reindeerSummer pasture 15Summer fed pellets 21Winter pasture 35

MuskoxRyøy, summer pasture 18

SheepDomestic 4–10Seaweed-eating 0

CattleDomestic 15

still regarded as high, compared to domestic ruminants and was probably influenced by the increased ruminal retention of fibrous plant particles inwinter. Cellulolytic bacteria might be important in the large rumen of Svalbard rein-deer in winter, since as much as 35% of all bacteria isolated in winter were cellulolytic(table 3), and, although not very efficient, were slowly degrading the fibrous foodeaten. The bacterial population in Norwegian reindeer seems to increase from summerto winter, a finding which might be related to the increased lichen intake in these ani-mals in winter increasing the availability of digestible energy in the rumen (table 2).

No major differences have been observed in the morphology of bacteria thatadhere to grass particles and lichens (fig. 4). Aagnes et al. (1995) reported that therumen bacteria seem to break down the lichen from the inside and that the densityof bacteria close to the plant particles was ten times higher than in the rumen fluid.Microorganisms that utilize plant structural polysaccharides as their energy sourceachieve preferential access to their substrate by associating closely with plant parti-cles entering the rumen, and at the same time extend their residence time in therumen to that of the least digestible part of the animals’ diet. The ability of bacteriato adhere to plant material and break down cell walls is of primary importance andappears to be an essential first stage in the digestive process in the rumen.

Bacteria adhering to the rumen wall are taxonomically distinct from bacteria inthe rumen fluid (Cheng et al., 1979). Ureolytic species adhering to the rumen wallmay play an important role in nitrogen recycling in ruminants such as reindeer byproviding ammonia to the bacteria in the rumen fluid used for protein synthesis.Direct examination by scanning electron microscopy (SEM) of sites on the rumenepithelium of Svalbard reindeer showed that only 30% of their epithelial surface wascovered in adherent bacteria in summer and only 10% in winter, compared to 75% infed cattle (K.-J. Cheng, personal communication). Many epithelial cells were partially sloughed, particularly in the summer, which might be explained by the highrate of rumen fermentation in summer and high rate of metabolism in rumen epithe-lial cells where the SCFA are absorbed (White and Staaland, 1983). The adherentbacterial population consisted largely of curved cells and cocci which resembledRuminococcus spp., by their possession of a condensed glycocalyx on the cell surface (fig. 5). The low level of adhering bacteria in winter in Svalbard reindeer corresponds with observations in starving cattle (K.-J. Cheng, personal communica-tion). We have therefore to assume that the low population of adhering bacteria in the


Fig. 4. Electron microscopic ultrathin sections of grass and lichen particles from the rumen of Norwegianreindeer in summer on South Georgia (A) (scale bar = 500 nm) and in winter in Norway (B) (scale bar = 1 mm).Morphological types resembling Fibrobacter are shown associated with both grass and lichen, while bacteriaresembling B. fibrisolvens are shown between the lichen particles. (Transmission electron micrographs: S. D. Mathiesen.)

rumen of Svalbard reindeer in winter is caused by a low rate of urea diffusion acrossthe rumen wall and reduced rumen carbohydrate fermentation rate in winter.

3.2. Cellulose degradation

Ruminants are highly specialized users of plant polysaccharides as their source ofenergy for growth. The microbial breakdown of structural polysaccharides is a slowchemical process under anaerobic conditions and therefore a large delay chambersuch as the rumen is necessary to enable enzymatic breakdown. The cellulase complexes secreted usually consist of three major types of enzymes which functionsynergistically in the hydrolysis of cellulose. They are: endo-1,4-β-glucanase, cellobiohydrolase, and β-glucosidase (Forsberg et al., 1997, 2000). The ruminaldigestion of plant cell wall polysaccharides such as cellulose, however, is limited bythe availability of non-protein nitrogen in the form of ammonia and also amino acidsand easily digestible carbohydrates in the rumen (Ørskov, 1992). Available nitrogenand carbon, rather than the cellulose content of the plants or the number of cellulolytic bacteria, may limit the rates of rumen cellulolysis and fermentation in reindeer (Aagnes and Mathiesen, 1994, 1995; Aagnes et al., 1996; Norberg and


Fig. 5. Rumen papilla in Svalbard reindeer on summer pasture (A, scale bar = 10 μm) andits epithelial cells with facultatively anaerobicadhering bacteria (B, scale bar = 5 μm).(Scanning electron micrographs: S.D.Mathiesen.)

Mathiesen, 1998; Sørmo, 1998; Utsi, 1998). This is probably one reason for the seasonal changes in the density of the viable bacterial population and the cellulolyticbacterial population in the rumen of Svalbard reindeer (Sørmo et al., 1997, 1999).This might also explain the efficient ruminal breakdown of cellulose in summercompared to winter in Svalbard reindeer (Sørmo et al., 1998).

Strains of B. fibrisolvens from Svalbard reindeer and Norwegian reindeer onSouth Georgia have been shown to solubilize cellulose, but it has not been possibleto isolate dominant bacteria with strong cellulolytic activity in the rumen of reindeerthat eat pure lichen (Aagnes et al., 1995; Olsen and Mathiesen, 1998). Likewise,Orpin et al. (1985) in their study were unable to isolate cellulolytic bacteria from therumen of seaweed-eating sheep on Ronaldsay. On the other hand the in vitro break-down of pure cellulose in rumen fluid from Norwegian reindeer in winter and inrumen fluid from lichen-fed reindeer was high (Olsen et al., 1997). Recently, Olsenand Mathiesen (1999) isolated a cellulolytic bacterial strain of B. fibrisolvens fromthe rumen of lichen-fed reindeer on a specially developed lichen medium and this maypartly explain the high rate of cellulose breakdown observed in lichen-fed animals.

Likewise, Deutsch et al. (1998) showed that roe deer eat highly cellulosic forageplants in winter but are almost incapable of digesting cellulose because their popu-lations of cellulase-producing bacteria are reduced in winter. Roe deer are concen-trate selector ruminant feeding type with short ruminal retention of plant fibres.Bacteria that require a long time for cell division are easily washed out of the rumen.However, the low ruminal cellulolysis in this species in winter might also beexplained by the low availability of metabolizable energy and of non-protein nitro-gen to microbial synthesis in winter. In contrast, ruminal cellulolysis in Norwegianreindeer was maintained at a high level in winter, which could be explained by supported bacterial growth owing to the high intake of highly digestible lichens(Olsen et al., 1997). Intake of digestible lichens might also influence the diffusiongradient of urea across the rumen wall and the recycling of nitrogen supplying themicrobial environment in the rumen in winter with ammonia (Hove and Jacobsen,1975). The high numbers of cellulolytic bacteria in the rumen of Svalbard reindeerin winter represent an important digestive adaptation to the fibrous food eaten, but due to reduced availability of non-protein nitrogen and digestible energy thedegradation is not as efficient as in rumen fluid from Norwegian reindeer offeredlichens in winter. The large rumen, high population of cellulolytic bacteria inSvalbard reindeer and perhaps a long ruminal retention time, however, allow for ahigher degree of digestion of fibrous plants eaten in winter in these animals and represent an adaptation to a winter diet without lichens.

3.3. Genetic studies of the rumen bacteria from reindeer

Rumen bacteria essential for the utilization of plant materials such as Butyrivibriofibrisolvens, Selenomonas spp., Streptococcus bovis and Lactobacillus spp. have all


been isolated from reindeer (Dehority, 1975a, 1986). Dehority (1986) concludedthat bacterial species isolated from the Alaskan reindeer were similar to thosewidely found in domestic ruminants and no unique physiological or biochemicalcharacteristics were observed in the species studied. It remains unclear whetherArctic ruminants have unique species of bacteria in their rumen. However, the composition of rumen bacterial populations in Svalbard reindeer is very specializedin relation to fibre digestion and nitrogen metabolism (Mathiesen et al., 1984a;Orpin and Mathiesen, 1990). As much as 30% of the culturable rumen bacterial population in Svalbard reindeer in winter consists of Butyrivibrio-like bacteria(Orpin et al., 1985), some of which have been studied in more detail to describegenes and expressed enzymes responsible for the symbiotic fibre degradation inthese animals. Hazelwood et al. (1990) first successfully cloned and expressed anovel endo-β-1-4-glucanase gene from B. fibrisolvens A46 in Escherichia coli.Other cellulase enzymes induced by substrate availability may be produced by B. fibrisolvens A46 that are multifunctional (Orpin and Mathiesen, 1990). Suchenzymes have been described for Chytridiomycetes (Orpin and Joblin, 1997) andwould be of great benefit in the rumen of these animals, which feed on diets thatvary greatly in quality. A cellulolytic strain (S-89) of Bacillus spp. from Svalbardreindeer was later transformed to kanamycin resistance with plasmid puB110, andwas inoculated into the rumen of Norwegian cattle. This bacterium did establish in the rumen of cattle, but at very low levels and unfortunately without significancefor the rumen cellulose breakdown (Mathiesen and Orpin, unpublished data). Some attempts to return laboratory bacteria to the rumen have shown them to have difficulties readapting to the rumen condition. Likewise, cellulolytic B. fibrisolvens E14from the Svalbard reindeer were inoculated into the rumen of sheep. Their survival wasinvestigated using a combined plasmid DNA probe (Orpin et al., 1987; Mathiesen,Orpin and Blix, unpublished data). Up to 105 cells/ml rumen fluid could be detected 30 days after the inoculation of the E14. However, it is doubtful that bacterial strains ofabout 105−106 have a significant effect on the overall ruminal metabolic activities.

The genus Butyrivibrio represents a diverse group of obligate anaerobic, curvedrod-shaped bacteria that produce large amounts of butyrate when fermenting carbo-hydrates, and B. fibrisolvens is the major culturable cellulolytic bacterium in therumen of Svalbard reindeer in both summer and winter (Orpin et al., 1985). Two B. fibrisolvens strains (ARD-22a and ARD-23c) isolated from reindeer in Alaska byDehority in 1975, were later examined for DNA relatedness with a total of 37 different bacterial isolates (all resembling B. fibrisolvens) from the rumen or caecumof sheep, steer, goats and pigs (Mannarelli, 1988). The guanine-plus-cytosine (G + C)base content of strains ARD-22a and ARD-23c was 42 and 43 mol%, respectively,but a large variability was observed in the G − C base content (39−49 mol%)between the different strains, indicating that the various isolates were in fact differentspecies. In a later study Mannarelli et al. (1990a,b) determined taxonomic relatednessbetween different strains of gastrointestinal bacteria using DNA−DNA hybridization


stating that two other strains of B. fibrisolvens (ARD-27b and ARD-31a) from therumen of Alaskan reindeer do not exhibit DNA relatedness to any of the previouslystudied strains (from other animals), although strains ARD-27b and ARD-31a arerelated to each other. However, a more recent study by Forster et al. (1996) thatsequenced the complete 16S rDNA of four strains of B. fibrisolvens isolated fromthe rumen of white-tailed deer (Odocoileus virginianus) in Canada showed a highsimilarity (96.5−99.7%) with the 16S rDNA sequence of B. fibrisolvens 49 from steer(Bryant and Small, 1956). Unlike B. fibrisolvens H17c (isolated from a domesticruminant), B. fibrisolvens E14 isolated from the rumen of Svalbard reindeer (Orpinet al., 1985) is unable to synthesize methionine in vivo and therefore needs methionineadded to the medium (Nili and Brooker, 1995). Methionine is, metabolically, themost expensive amino acid to produce (Old et al., 1991). Tests with intermediates ofthe methionine biosynthetic pathway indicate that in strain E14 the final step fromhomocysteine to methionine is blocked, probably due to the lack of methionine synthetase (Nili and Brooker, 1997).

4. CILIATE PROTOZOA

The rumen ciliate protozoa population in reindeer was first described by Ebenlein(1895). He compared reindeer from a zoological garden with domestic ruminants inGermany and concluded that they were the same. Similar results were obtained byLubinsky (1958) who worked with caribou in northern Canada. Reindeer werereported to have a unique ciliate fauna as demonstrated in Finnish reindeer(Westerling, 1970). Svalbard reindeer and Finnish reindeer both show clear seasonalchanges in numbers and composition of the ciliate population (Westerling, 1970;Orpin and Mathiesen, 1988, 1990). Dehority (1975b) reported that the rumen cili-ates of semi-domesticated reindeer in Alaska were similar to those of the domesticruminants in the area, while a typical reindeer fauna was observed in wild caribouand reindeer. In the rumen of Svalbard reindeer the density of rumen ciliates variedfrom 105 cells per ml rumen fluid in summer to 104 cells per ml in winter and onlyentodiniomorphid ciliates were present, while no holotrich ciliates were observed incontrast to Norwegian and Finnish reindeer (Westerling, 1970; Orpin andMathiesen, 1988, 1990). Their absence in the Svalbard reindeer could be due tostarvation resulting from poor quality of the forage eaten in winter, since theymetabolize only soluble carbohydrates and the rumen content of Svalbard reindeerwas highly fibrous in winter (Williams and Coleman, 1988). Short ruminal retentionin summer is known from ruminants of the CS ruminant type and this would makethe establishment of holotrichs difficult in Svalbard reindeer because of the longtime needed for cell division. The major species of entodiniomorphs identified inSvalbard reindeer were Entodinium simplex, E. triacum triacum, E. longinucleatum,Polyplastron multivesiculatum, Eremoplastron bovis and Eudiplodinium maggi.There is little evidence that the Entodinium spp. is important in fibre digestion in


ruminants utilizing principally starch and bacteria as their carbon source (Williamsand Coleman, 1988). E. maggi, Polyplastron multivesiculatum and Eremoplastronbovis are all known to ingest plant particles and to contain cellulase (Coleman,1985). Holotrich and entodiniomorphic ciliates were detected in Norwegian rein-deer that had survived on South Georgia for almost 100 years without lichens.Diplodinium rangiferi, Epidiumun ecaudatum gigas and Polyplastron arcticumwere identified in population densities of 105−106 cells per ml rumen fluid. Speciessuch as D. rangiferi were originally associated with reindeer eating lichens(Westerling, 1970). Although there are differences between the protozoal fauna in var-ious Arctic ruminants, currently available information does not suggest that the faunaare unique to a given animal species or that they are essential for the digestion ofArctic plants; rather, it appears that diet and isolation of the host are likely to be theprimary factors that have determined faunal composition. Dehority (1986) empha-sized that several protozoal species are unique to Arctic ruminants, i.e. the rangifer-type fauna. Ciliate protozoa, however, seem not to be essential for rumen digestion in


Fig. 6. Ciliate protozoa in the reindeer rumen.A micrograph of a rumen entodininomorph ciliate, Entodinium triacum triacum, from therumen of a Svalbard reindeer in summer (A)(phase-contrast micrograph by S.D. Mathiesen)and a ciliate Entodinium sp. in Norwegian reindeer on natural winter pasture (B) showing a ciliate that engulfs bacteria associatedwith lichen particles (transmission electron micrograph by M.A. Sundset; scale bar = 2 mm).

reindeer since rumen contents without ciliates have been observed in healthy reindeerand muskoxen (Mathiesen, unpublished data) (fig. 6).

5. ANAEROBIC FUNGI

The anaerobic fungi commonly found in the rumen of large ruminants were first dis-covered by Orpin (1974). Several species are known to exist but only one type hasbeen found in the rumen of Svalbard reindeer. This was a monocentric species withpolyflagellated zoospores, endogenous development and a branching rhizoidal systemcharacteristic of Neocallimastics spp. Multiflagellated zoospores of the rumen fungiNeocallimastix frontalis (Chytridiomycetes) have been reported in Svalbard reindeerby Orpin et al. (1985). This species utilizes a range of different polysaccharidesincluding cellulose and their hyphae may penetrate plant vascular tissue normally notaccessible to bacteria (Bauchop, 1979; Orpin and Letcher, 1979). Relatively low num-bers of zoospores were demonstrated in Norwegian reindeer calves on winter pasture(1.5 × 103 zoospores/ml rumen fluid) as compared to summer pasture (3.1 × 102

zoospores/ml) (Olsen, 2000), while higher numbers of zoospores were found in adultSvalbard reindeer on winter pasture (up to 1 × 105 zoospores/ml) (Mathiesen, 1999).Large populations of anaerobic fungi are normally associated with a high intake ofrough, fibrous diets in domestic ruminants (Bauchop, 1979) and the higher concen-trations of zoospores in Svalbard reindeer on winter pasture may support this. Electronmicroscopic ultrathin sections of plant particles from the rumen revealed fungi pene-trating the plant cell wall (Mathiesen and Aagnes, 1990; Mathiesen and Utsi, unpub-lished data) (fig. 7). Rumen anaerobic fungi were isolated from free-living andartificially fed reindeer in northern Norway (fig. 7). These fungi are able to invade theplant tissue to a greater extent than bacteria and ciliates and some fungi express very


Fig. 7. Scanning electron micrograph (A) of anaerobic rumen fungi in Norwegian reindeer on South Georgiain summer, with sporangium (s) and rhizomes (r) invading a grass particle; rumen bacteria (b) are shown in the background. The transmission electron micrograph (B) shows grass particles from the same animals withchytridiomycetes hypha (arrow) invading the vascular bundle sheet (v) and penetrating the plant cell walls (p)(scale bar = 2 μm). (Photographs: S.D. Mathiesen.)

efficient multifunctional polysaccharide degrading enzymes in reindeer (Orpin andJoblin, 1997) and could contribute substantially to ruminal fibre breakdown. We havenot yet been able to associate anaerobic fungi with ruminal lichen particles in reindeer.

6. THE DISTAL FERMENTATION CHAMBER

Many species of ruminants evolved before the spread of grasses during the Mioceneepoch of the Tertiary and earlier ruminant types seem to have a limited capacity todigest cellulose. According to Hofmann (1999) they selected mainly dicot plants andof these primarily their plant cell content. Fermentation of such forage plants in thedistal fermentation chamber (DFC) might have been important in these early rumi-nant types. The caecum and colon of “modern” ruminants contain microorganismscapable of producing SCFA from plant material not digested in the rumen (Ulyatt et al., 1975). Svalbard reindeer have a DFC similar in size to that of concentrate selec-tor ruminant feeders, while DFCs in reindeer offered lichens in northern Norway aresmall. In concentrate selectors or in intermediate feeders with short ruminal retentiontime, plant material escapes from ruminal digestion, and digesta may be retained inan enlarged DFC for a second fermentation. Under these circumstances a consider-able proportion of potentially digestible fibre may escape microbial fermentation inthe rumen and enter the DFC (Hofmann, 1989). The ruminal particle size distributionin Svalbard reindeer indicates a rapid release of small particles from their reticulo-rumen which could influence the large DFC in these animals. The large DFC ofSvalbard reindeer was, however, correlated with a high hemicellulose content of thereticulo-rumen (Sørmo et al., 1999). The acidic environment in the abomasum andthe presence of pepsin might change the configuration of plant hemicelluloses andfacilitate their digestion in the DFC (Van Soest, 1994).

The total culturable bacterial population isolated from caecal contents fromSvalbard reindeer was 8.9 × 108 cells/ml in summer and 1.5 × 108 cells/ml in winter(Mathiesen et al., 1987). Of the dominant species of culturable bacteria, B. fibrisol-vens represented 23% in summer and 18% in winter, while S. bovis represented 17%in summer and 5% in winter. Prevotella ruminicola represented 10% of the dominantculturable species in summer and as much as 26% in winter. As much as 10% of theviable bacterial population was cellulolytic in summer, compared to 6% in winter, themost abundant cellulolytic species being B. fibrisolvens representing 62% of the totalcellulolytic population in summer, and R. albus representing 80% of the cellulolyticpopulation in winter. The hindgut bacterial population has the ability to digest starchand major structural carbohydrates that escape digestion in the rumen.

The very low pH recorded in this organ in reindeer eating lichen indicates that thedigestive function of the DFC is not yet understood. In contrast, in Svalbard reindeerfeeding in an area on Svalbard characterized as an Arctic desert, the relative size of theDFC was small but SCFA from the DFC contributed 17% of the total SCFA produced,indicating that fermentation in the DFC might be important when the substrate is avail-able (Sørmo et al., 1997). In comparison, in sheep fed chopped alfalfa hay SCFA


production in the hindgut contributed 7% of the maintenance requirement (Hume,1977), while in black-tailed deer on natural pasture it contributed only 1% (Allo et al.,1973). Fibre-degrading bacteria have been isolated from the caecum of Svalbard rein-deer in both winter and summer indicating that it is important for fibre digestion.Furthermore, the caecum of Norwegian reindeer on South Georgia in summer contained low concentrations of anaerobic bacteria with cellulolytic activity (Mathiesenand Aagnes, 1990). The breakdown of cellulose in the DFC is a slow reaction and themicrobial population might therefore need additional energy and nitrogen to contributesubstantially to the daily energy requirements in these animals.

7. THE SMALL INTESTINE

The small intestine is about 22 m long in Norwegian reindeer and 18 m in Svalbardreindeer of similar sex and age (Westerling, 1975a; Staaland et al., 1979). It is coveredwith villi and microvilli, the epithelial cells being rapidly replaced with new cells(Myklebust and Mayhew, 1997), providing a large surface area for the absorption ofmicrobial protein, water and minerals. The small intestinal pH increases from 5.9 inthe duodenum, to 6.1 in the jejunum and 7.5 in the ileum (Sørmo and Mathiesen,1993; Sørmo et al., 1994).

Electron microscopic examinations of the gut represent important tools for inves-tigating the microbial ecology of the gastrointestinal tract ecosystem and for deter-mining the presence of autochthonous or allochthonous microbiota. However, veryfew microscopic examinations of the small intestine have been conducted so far, andSørmo and Mathiesen (1993) failed to reveal any bacteria associated to the tip of themicrovilli or between the microvilli. This may be explained by the low numbers of bac-teria residing in the small intestine of reindeer. Table 4 shows the anaerobic bacterial population level in the small intestine of free-living reindeer on natural winter pasture in Finnmark and in captive lichen-fed reindeer. In two of the animals fednatural winter pasture, no detectable viable bacteria were found colonizing the proximal part of the small intestine (Sørmo et al., 1994). The authors suggested thatthis probably was due to antibacterial substances activated in the acidic environmentin the abomasum, inhibiting the establishment and growth of bacteria in the proximal


Table 4. Numbers of viable anaerobic and aerobic bacteria per gram mucosa in the proximal and distal part of the small intestine of Norwegian reindeer (Sørmo and Mathiesen, 1993; Sørmo et al., 1994; Sørmo, unpublished data)

Anaerobes Aerobes

Proximal part Distal part Proximal part Distal part

Norwegian reindeerWinter pasture 0–2 600 70–40 000 0–2 000 20–74 000Captive lichen-fed 700–42 000 650–72 000 nd ndFed RF-71, after 117 000–140 000 500 000–600 000 nd nd3 days starvation

nd = not detected

small intestine. Organic acids, lipids, antibiotics and usnic acid known to occur inlichens are potential agents inhibiting bacterial growth (Vartia, 1949, 1973; Lauterweinet al., 1995; Huneck, 1999; Müller, 2001). Autochthonous microorganisms associatedclosely with the small intestinal epithelium could be important for the health and welfare of the host by limiting direct attachment or interaction of pathogenic bacteriato the mucosa (Slomiany et al., 1994; Henderson et al., 1996). It is well known that theindigenous microbiota plays a protective role against pathogenic bacteria colonizing themucosal surface or the microvilli (Hentges, 1992; Salmonen, 1996). Microorganismscolonizing the small intestinal mucosa could by various mechanisms inhibit the absorp-tion of water and salt from the intestine, causing diarrhoea and dehydration of the host.

Sørmo and Mathiesen (1993) showed that bacteria colonizing the small intestine ofreindeer were dominated by the lactic acid bacteria Streptococcus spp., 66.3% of thetotal bacterial population when the animals were fed lichens ad libitum. The most common of the streptococci resembled S. bovis and S. durans, but the authors did notpresent any molecular data confirming this statement. In a later study, variable popula-tion densities of streptococci, from nil to 50%, were reported to colonize the proximaland distal parts of the small intestine of reindeer grazing on a natural winter pasture(Sørmo et al., 1994). However, the importance of streptococci in the small intestine ofreindeer remains unknown as great variations between individuals seem to occur. In 36out of 40 faecal samples from clinically healthy reindeer calves (90%), Enterococcusspp. were isolated (Kemper et al., 2002). These organisms are common inhabitants andopportunistic pathogens in the intestinal tract of animals (Carter and Cole, 1990).

Strains of lactobacilli have been isolated from the small intestine of lichen-fedreindeer, but they seem to represent only a minor part of the microbiota as they constitute only 0.9% of the bacterial population (Sørmo and Mathiesen, 1993).Lactobacillus spp. have also been isolated from the small intestine of free-livingreindeer grazing on natural winter pasture (Sørmo et al., 1994), but this study alsoconcluded that lactobacilli represent only a minor part of the small intestinal micro-biota. The authors showed that these lactobacilli were more tolerant to low pH thanthe rumen bacteria described by Kandler and Weis (1986).

Great variations in the population level of Bacterioidaceae, from nil to 68.4%, colonizing the proximal part of the small intestine were found in four reindeer graz-ing on natural winter pasture (Sørmo et al., 1994). The proportion of Bacterioidaceaefound associated to the distal part of the small intestine was 2.4, 6.5, 11.5 and 21.4%of the total bacterial population. Propionibacterium has been isolated from the prox-imal part and the distal part of the small intestine of free-living reindeer, but only atlow population levels, as it accounts for only 2% of the viable bacteria (Sørmo et al.,1994). Likewise, Eubacterium spp. has only been isolated from the distal part of thesmall intestine of free-living reindeer (Sørmo et al., 1994). Also bacteria belongingto the genus Ruminococcus have only sporadically been isolated from the small intes-tine of reindeer, and then they have only been isolated from the distal part of the smallintestine (Sørmo et al., 1994).


8. ENTERIC PATHOGENIC BACTERIA

Similar to the situation in other wild or free-ranging animals, little information is avail-able on the occurrence of pathogenic faecal bacteria and on the impact of enteric diseases on production in growing reindeer and in reindeer in general. The few avail-able reports describe outbreaks of disease in single affected individuals only, as possibilities to do systematic studies on bacterial diseases in free-ranging reindeercalves are restricted. One must consider that the predominant extensive form of rein-deer husbandry, on vast areas that are covered for a long period of the year by ice andsnow, does not necessarily favour outbreaks of infectious disease (Skjenneberg andSlagsvold, 1968), even though potential pathogenic bacteria may be found in the nat-ural environment (Kapperud, 1981) and also in the intestinal tract of healthy reindeer(Aschfalk et al., 1998; Kobayashi et al., 1999). Outbreaks of disease are often depend-ent not only on the presence of the bacterial agent, but also on other factors that mayalter the balance in the intestinal tract. In an unpublished study by Dr Wenche Sørmo,it was demonstrated that when reindeer were eating lichens, then starved for three days and thereafter fed RF-71 (a pelleted concentrate diet), simulating an emergencyfeeding situation, the animals developed diarrhoea. At that time, a population level of 5 × 105 anaerobic bacteria per gram of small intestinal mucosa was found. Diarrhoea-induced diseases in the small intestines of reindeer frequently occur during emergencyfeeding with commercial available pellets in winter in Norway. Replacement of newepithelial cells, and the numbers of lymphocytes in the intestinal cells could be important factors in the protection against pathogenic bacteria, the development of a beneficial small intestinal microbiota and the ability to absorb nutrients.

Once established in a herd, an infection may spread easily among individuals,especially if reindeer are kept intensively, e.g. for artificial feeding during winter orcalf marking. Numerous infectious agents including bacteria, virus and protozoahave been related to diarrhoea in young ruminants (Tzipori, 1981; Munoz et al.,1996; Busato et al., 1998). Bacteria such as Clostridium perfringens, Escherichiacoli and Salmonella spp. are among the most important bacterial agents in causingenteric and other diseases, as is known from domestic, young ruminants (Dubourguieret al., 1978; Lintermans and Pohl, 1983; De Rycke et al., 1986; Alexander andBuxton, 1994; Munoz et al., 1996; Steiner et al., 1997; Busato et al., 1998, 1999).De Rycke et al. (1986) classified some of the bacterial agents as primary calfenteropathogens, e.g. enterotoxigenic E. coli and Salmonella spp., other bacteria ashaving a less expressed enteropathogenicity, such as enterotoxigenic Clostridiumperfringens, and others as agents that are not directly associated to diarrhoea, such asYersinia enterocolitica. Obviously, a certain impact of these enteric pathogens onreindeer production cannot be excluded even though there are only a very few reportson diseases and mortality caused by these bacteria in reindeer.

The presence of Clostridium spp. colonizing the small intestine of reindeer hasbeen reported in lichen-fed animals (Sørmo and Mathiesen, 1993) where 6.4%


of the bacterial population belonged to genus Clostridium, and in one individualfrom free-living reindeer grazing on a natural winter pasture (Sørmo et al., 1994).C. perfringens was reported in the intestine and in faecal samples associated to diseased reindeer (Kummeneje and Bakken, 1973), as well as healthy, adult reindeer(Aschfalk et al., 1998, 2001). In all these examinations, C. perfringens toxin type A(alpha-toxin) was the only or most dominant one diagnosed. In addition, the genefor the novel described β2-toxin and the gene encoding for enterotoxin were foundby Aschfalk et al. (2001, 2002a). Outbreaks of disease in reindeer were also reportedby Rehbinder and Nikander (1999), caused by C. perfringens type C (alpha- andepsilon-toxin) and type D (alpha- and beta-toxin). Mortality in reindeer caused byC. perfringens enterotoxemia was reported by Kummeneje and Bakken (1973) andby Rehbinder and Nikander (1999). C. perfringens is known to cause important gastrointestinal and enterotoxemic diseases in young ruminants, sheep and deer(Alexander and Buxton, 1994; Songer, 1998). The virulence and pathogenicity ofthis organism are closely related to the expression of different toxins (Petit et al.,1999), and outbreaks of clostridial enteric diseases are combined with further factors, such as type of nutrition and seasonal changes.

The first report on the occurrence of E. coli (O-group 55) in the intestine of rein-deer associated to calf mortality was by Clausen et al. (1980). In their study onlichen-fed reindeer, Sørmo and Mathiesen (1993) reported that E. coli contributed25.5% of the viable bacterial population colonizing the small intestine. Recently,this bacterium was isolated in all faeces samples from clinically healthy, youngreindeer calves (n = 40). By polymerase chain reaction (PCR) analysis, the genes foreae and hly could be detected in two and seven of these isolates, respectively(Kemper et al., 2002). STEC, shigatoxin-producing bacteria were detected byKobayashi et al. (1999) and Aschfalk et al. (2001) in reindeer. However, as PCRanalysis was done on mixed cultures, the evidence of shigatoxin-producing E. coliwas not given. Escherichia coli is recovered from a wide variety of diseases, such as colibacillosis and colisepticaemia in all domesticated animals as a primaryor secondary pathogenic agent (Carter and Cole, 1990) and is of special importancein causing diarrhoea in young ruminants (Alexander and Buxton, 1994; Munoz et al., 1996). STEC have been isolated frequently from cattle (Busato et al., 1998), lamb and kid faeces (Beutin et al., 1993; Munoz et al., 1996), but there are only a few reports on the association between occurrence of STEC and disease in rumi-nants. However, Busato et al. (1998) assume that STEC, or rather the free faecalverotoxin in the faecal matter, possibly may be a most significant cause of calfdiarrhoea.

Salmonella spp. associated to mortality of reindeer calves was reported byKuronen et al. (1998). The screening of serum samples from 2000 clinically healthy,slaughter reindeer from Norway, revealed a seroprevalence of 0.6% for Salmonellaspp. (Aschfalk et al., 2002b), originating assumingly from an infection following thefaecal–oral route. In Finland, a prevalence of 2.8% was detected, as also sera from


animals showing intestinal disorders were examined (Aschfalk and Denzin, 2000).Salmonella serotypes are known to cause enteric diseases and septicaemia, in youngruminants (Carter and Cole, 1990; Alexander and Buxton, 1994).

Yersinia enterocolitica was found in one faecal sample out of 40 (2.5%) young,clinically healthy calves (Kemper et al., 2002). Disease in reindeer caused byYersinia sp. was reported by Rehbinder and Nikander (1999). Disease caused byYersinia sp. is considered one of the more important diseases of corralled deer andis also reported to affect young individuals (Alexander and Buxton, 1994).

Other than being the primary cause of diseases, several entopathogenic agentsgenerally not considered to be major aetiologic agents of calf diarrhoea may, however, play a role as preceding or synergistic infections in animals with inade-quate passive immunity. These organisms can also gain importance in situations ofepidemic outbreaks of diarrhoea in herds with specific management problems(Busato et al., 1998). As corralling of reindeer is getting more and more common inreindeer husbandry, this intensification of reindeer production eventually leads to anincreased putative risk of outbreaks of infectious diseases, by different bacterialagents, as is known from domesticated ruminants brought into intensive farmingconditions (Mackintosh, 1998).

9. FUTURE PERSPECTIVES: THE IMPLEMENTATION OF NEW MOLECULAR TOOLS FOR STUDIES OF THE MICROBIAL ECOLOGY IN THE GASTROINTESTINAL TRACT OF REINDEER

Extensive studies of the rumen ecosystem using conventional anaerobic cultivationmethods have provided us with a rich knowledge of diverse types of gastrointestinalbacteria. The cultivation-based techniques are, however, hampered with some obvi-ous limitations. Direct counts of fixed, filtered rumen fluid from the Svalbard rein-deer showed total population densities of 5.5 × 1010 in summer and 1.1 × 1010 inwinter, as compared to viable counts of 2.1 × 1010 and 0.36 × 1010, respectively(Orpin et al., 1985). Hence, 62−67% of the total population was unable to grow onthe culture media employed. Typically, several of the large bacteria, such asOscillospira guilliermondii, Magnoovum eadii and Quinella ovalis, have never beencultivated. Synergistic microorganisms depend on others and this may also limittheir ability to grow in pure cultures. Furthermore, culture methods not only are verylaborious and time consuming, but they also identify the bacterial isolates from theirphenotypic pattern – which will vary depending on a range of factors such as, e.g.the expression of their genes and the artificial conditions in the laboratory. In fact, avast majority of the rumen microorganisms have not been isolated and still remainto be characterized (Whitford et al., 1998; Tajima et al., 1999, 2000; Ramsak et al.,2000; Kocherginskaya et al., 2001). The microflora of the rumen may be far morediverse than earlier believed (Avgustin et al., 1997; Forster et al., 1997). It has


proved impossible to identify several of the strains of the viable bacteria isolated inSvalbard reindeer, and Norwegian reindeer on South Georgia and in Norway, usingstandard anaerobic microbiological techniques (Aagnes et al., 1993; Mathiesen andUtsi, unpublished). To what extent unfamiliar bacterial strains contribute to therumen ecosystem is unknown. Therefore it is still possible that some bacterialspecies in reindeer could be unique. Likewise, a range of unidentified bacterialstrains has been isolated from the small intestine of reindeer, most of which arestrictly anaerobic, motile Gram-positive large rods, single or pairs of irregular rods,capable of utilizing glucose, maltose, sucrose, cellobiose and starch, which have notyet been characterized (Sørmo et al., 1998). The population level of these bacteriaseems to dominate in the small intestine of some animals, while their populationlevel in other animals is lower than the detection level.

In recent years, molecular methods have been developed that in combinationwith the traditional cultivation-based methods can be used to give a completedescription of the gastrointestinal ecosystem (e.g. Lin et al., 1997; Whitford et al.,1998; Simpson et al., 1999, 2000; Tajima et al., 1999, 2001a,b; White et al., 1999).The molecular methods are based on comparative analysis of rRNA sequences andits encoding genes, retrieving the sequences directly from the gastrointestinal tractsamples by the help of PCR using highly conserved primer binding sites on the 16S rRNA genes. Oligonucleotide hybridization probes can be designed that targetand discriminate between broad phylogenetic groups such as Archaea, Bacteriaand Eucarya, or even specific strains, allowing studies of population structure and dynamics in the gastrointestinal microbial ecosystem (Amann and Kühl, 1998;Mackie et al., 2000).

We are currently working on a project to construct 16S rDNA clone libraries toexamine rumen bacterial diversity in reindeer on natural pastures in northernNorway and on Svalbard (Olsen et al., 2002). The diversity of bacterial, archaealand eucaryal populations in each compartment of the digestive tract of reindeer willalso be determined for the first time using molecular microbial ecology techniques.

REFERENCES

Aagnes, T.H., Mathiesen, S.D., 1993. Food and snow intake, body mass and rumen function in reindeerfed lichen and subsequently starved for 4 days. Rangifer 14, 33–37.

Aagnes, T.H., Mathiesen, S.D., 1995. Roundbaled grass silage as food for reindeer in winter. Rangifer15, 27–35.

Aagnes, T.H., Sørmo, W., Mathiesen, S.D., 1995. Ruminal microbial digestion in free-living, in captive lichen-fed and in starved reindeer (Rangifer tarandus tarandus) in winter. Appl. Environ. Microbiol. 61, 583–591.

Aagnes, T.H., Blix, A.S., Mathiesen, S.D., 1996. Food intake, digestibility and rumen fermentation inreindeer fed baled timothy silage in summer and winter. J. Agr. Sci. 127, 517–523.

Alexander, T.L., Buxton, D., 1994. Management and Diseases of Deer. Macdonald Lindsay Pindar,Loanhead, Scotland.

Allo, A.A., Oh, J.H., Longurst, W.M., Connoly, G.E., 1973. VFA production in the digestive systems ofdeer and sheep. J. Wildl. Manage. 37, 202–211.

Amann, R., Kühl, M., 1998. In situ methods for assessment of microorganisms and their activities. Curr.Opin. Microbiol. 1, 352–358.


Annison, E.F., Armstrong, D.G., 1970. Volatile fatty acid metabolism and energy supply. In: Phillipson, A.(Ed.), Physiology of Digestion and Metabolism in the Ruminant. Oriel Press, Newcastle upon Tyne,pp. 422–437.

Aschfalk, A., Denzin, N., 2000. Prevalence of antibodies to Salmonella subsp. in reindeer in Finland.Oral presentation in the course: Advanced Methods for Test Validation and Interpretation inVeterinary Medicine, Berlin, Germany, 21–23 June 2000. Course Manual, p. 102.

Aschfalk, A., Nieminen, M., Müller, W., 1998. Clostridium perfringens in reindeer – nutrition as a factorin causing enterotoxemia. 2nd International Symposium on Physiology and Ethology of Wild and ZooAnimals, Berlin, 7–10 October 1998. Advances in Ethology (1998) 33, Supplements to Ethology, 97.

Aschfalk, A., Ryeng, K., Höller, C., 2001. Occurrence of Campylobacter spp. Clostridium perfringens,Salmonella spp, Yersinia spp and shigatoxin-1,2-producing bacteria in semi-domesticated reindeercadavers in Northern Norway. Nordic Conference on Reindeer Research, June 2001, Inari, Finland.Abstract in Proceedings, pp. 42–43.

Aschfalk, A., Valentin-Weigand, P., Müller, W., Goethe, R., 2002a. Toxin types of Clostridium perfrin-gens from free-ranging, semi-domesticated reindeer in Norway. Vet. Rec. 151, 210–213.

Aschfalk, A., Laude, S., Denzin, N., 2002b. Seroprevalence of antibodies to Salmonella spp. in semido-mesticated reindeer in Norway, determined by enzyme-linked immunosorbent assay. Berl. Munch.Tierarztl. Wochenschr. 115, 351–354.

Avgustin, G., Ramsak, A., Tererka, M., Nekrep, F.V., Flint, H.J., 1997. Evolutionary relationship and thediversity of the rumen bacteria belonging to the Cytophaga-Flexibacter-Bacteroides phyllum. Reprod. Nutr. Dev., Suppl., pp. 27–28.

Banfield, A.W.F., 1961. A revision of the reindeer and caribou genus. Nat. Museum Can. Bull. 177, Biol. Ser. 66, 1–137.

Bauchop, T., 1979. The rumen anaerobic fungi: colonizers of plant fibre. Ann. Rech. Vet. 10, 246–248.Beutin, L., Geier, D., Steinruck, H., Zimmermann, S., Scheutz, F.J., 1993. Prevalence and some proper-

ties of verotoxin (Shiga-like toxin)-producing Escherichia coli in seven different species of healthydomestic animals. Clin. Microbiol. 31, 2483–2488.

Boomker, E.A., 1995. Aspects of fermentative digestion in the Kudu, Pragelaphus strepsiceros. In:Schwartz, H.J., Hofmann, R.R. (Eds.), Wild and Domestic Ruminants in Extensive Land UseSystems. Eokologische Hefte der Landwirshaflich-Geartinersichen Fakulteat, Berlin, Germany.

Bryant, M.P., Small, N., 1956. The anaerobic monotrichous butyric acid-producing curved rod-shapedbacteria of the rumen. J. Dairy Sci. 72, 16–21.

Busato, A., Lentze, T., Hofer, D., Burnens, A., Hentrich, B., Gaillard, C., 1998. A case control study ofpotential enteric pathogens for calves raised in cow-calf herds. Zbl. Veterinarmedizin (B) 45, 519–528.

Busato, A., Hofer, D., Lentze, T., Gaillard, C., Burnens, A., 1999. Prevalence and infection risks ofzoonotic enteropathogenic bacteria in Swiss cow-calf farms. Vet. Microbiol. 69, 251–263.

Carter, G.R., Cole, J.R., 1990. Diagnostic Procedures in Veterinary Bacteriology and Mycology.Academic Press, San Diego, California.

Cheng, K.-J., McCowan, R.P., Costerton, J.W., 1979. Adherent epithelial bacteria in ruminants and theirrole in digestive tract function. Amer. J. Clin. Nutr. 32, 139–148.

Christiansen, H.R., Tyler, N.J.C., Johansen, O., 1996. Do female reindeer in Finnmark use their bodyreserves in winter? (in Norwegian). Reindriftsnytt 3, 15–19.

Clausen, B., Dam, A., Elvestad, K., Krogh, H.V., Thing, H., 1980. Summer mortality among cariboucalves in West Greenland. Nord. Vet. Med. 32, 291–300.

Coleman, G.G., 1985. The cellulase content of 15 species of entodiniomorphid protozoa, mixed bacteriaand plant debris isolated from the ovine rumen. J. Agr. Sci. 104, 349–360.

De Rycke, J., Bernard, S., Laporte, J., Naciri, M., Popoff, M.R., Rodolakis, A., 1986. Prevalence of various enteropathogens in the feces of diarrheic and healthy calves. Ann. Rech. Vet. 17, 159–168.

Dehority, B.A., 1975a. Characterisation studies on rumen bacteria isolated from Alaskan reindeer(Rangifer tarandus L.). In: Luick, J.R., Lent, P.C., Klein, D.R., White, R.G. (Eds.), Proceedings of the 1st International Reindeer and Caribou Symposium, University of Alaska, Fairbanks, pp. 228–240.

Dehority, B.A., 1975b. Rumen ciliate protozoa of Alaskan reindeer and caribou (Rangifer tarandus L.)in reindeer. In: Luick, J.R., Lent, P.C., Klein, D.R., White, R.G. (Eds.), Proceedings of the 1stInternational Reindeer and Caribou Symposium, University of Alaska, Fairbanks, pp. 241–250.


Dehority, B.A., 1986. Microbes in the foregut of Arctic ruminants. In: Milligan, L.P., Grovum, W.L.,Dobson, A. (Eds.), Control of Digestion and Metabolism in Ruminants. Prentice-Hall, EnglewoodCliffs, New Jersey, pp. 307–325.

Deutsch, A., Lechner-Doll, M., Wolf, G.A., 1998. Activity of cellulolytic enzymes in the contents of retinulo-rumen and caecocolon of roe deer (Cabreolus capreolus). Comp. Biochem. Physiol. Part A 119, 925–930.

Dubourguier, H.C., Gouet, P., Contrepois, M., Girardeau, J.P., 1978. Diarrhea of the newborn animal:nature and mechanism of action of enteropathogenic Escherichia coli in the calf and piglet (inFrench). Ann. Rech. Vet. 9, 129–152.

Ebenlein, R., 1895. On rumen ciliates (in German). Zeitsschr. Wiss Zool. 59, 233–303.Fancy, S.G., White., R.G., 1985. Energy expenditures by caribou while erating snow. J. Wild. Manag. 49,

987–993.Forsberg, C.W., Cheng, K.-J., White, B.A., 1997. Polysaccharide degradation in the rumen and large

intestine. In: Mackie, R.I., White, B.A. (Eds.), Gastrointestinal Microbiology. Vol. 1, GastrointestinalEcosystems and Fermentation. Chapman & Hall, New York, pp. 319–379.

Forsberg, C.W., Forano, E., Chesson, A., 2000. Microbial adherence to the plant cell wall and enzymatichydrolysis. In: Cronjé, P.B. (Ed.), Ruminant Physiology: Digestion, Metabolism, Growth andReproduction. CABI Publishing, New York, pp. 79–98.

Forster, R.F., Whitford, M.F., Beard, C.E. 1997. An investigation of microbial diversity in the Rumen ofdairy cattle using comparative sequence analysis of clones 16r RNA genes. Reprod. Nutr. Dev. Suppl.pp. 28–29.

Forster, R.J., Teather, R.M., Gong, J., Deng, S.-J., 1996. 16S rDNA analysis of Butyrivibrio fibrisolvens:phylogenetic position and relation to butyrate-producing anaerobic bacteria from the rumen of white-tailed deer. Lett. Appl. Microbiol. 23, 218–222.

Gotaas, G., Tyler, N.J.C., 1994. Methane production and doubled labelled water method in reindeer(Rangifer tarandus). Proc. Soc. Nutr. Physiol. 3, 181.

Hazlewood, G.P., Davidson, K., Laurie, J.I., Romaniec, M.P.M., Gilbert, H.J., 1990. Cloning andsequencing of the celA gene encoding endoglucanase A of Butyrivibrio fibrisolvens strain A46. J.Gen. Microbiol. 136, 2089–2097.

Henderson, B., Pool, S., Wilson, M., 1996. Microbial host interactions in health and disease: Who con-trols cytokinine network? Immunopharmacology 35, 1–21.

Hentges, D.J., 1992. Gut flora and disease resistance. In: Fuller, R. (Ed.), Probiotics, The Scientific Basis.Chapman & Hall, London, pp. 87–100.

Hespel, R.B., Akin, D.E., Dehority, B.A., 1997. Bacteria, fungi, and protozoa of the rumen. In: Mackie, R.A.,White, B.A., Isaacson, R.E. (Eds.), Gastrointestinal Microbiology. Vol. 2, Gastrointestinal Microbesand Host Interactions. Chapman & Hall, New York, pp. 59–141.

Hobsen, P.N., Mann, S.O., Summers, R., 1976. Rumen micro-organisms in red deer, hill sheep and rein-deer in the Scottish highlands. Proc. Roy. Soc. Edinburgh 75, 171–180.

Hofmann, R.R., 1989. Evolutionary steps of ecophysiological adaptation and diversification of rumi-nants: a comparative view of their digestive system. Oecologica 78, 443–457.

Hofmann, R.R., 1999. Oryx antelopes – selective dry region grazers (GR) with extreme adaptation oftheir digestive system; a comparative view. Proceedings of the 1st Abu Dhai International ArabianOryx Conference.

Hofmann, R.R., 2000. Functional and comparative digestive system anatomy of Arctic Ungulates.Rangifer 20, 71–82.

Hove, K., Jacobsen, E., 1975. Renal excretion of urea in reindeer: effects of nutrition. Acta Vet. Scand.16, 513–519.

Hume, I.D., 1977. Production of volatile fatty acids in two species of wallaby and in sheep. Comp.Biochem. Physiol. 56A, 299–304.

Huneck, S., 1999. The significance of lichens and their metabolites. Naturwissenschaften 86, 559–570.Hungate, R.G., 1966. Rumen and its Microbes. Academic Press, New York.Hungate, R.E., Phillips, G.D., McGregor, A., Hungate, D.P., Buechner, H.K., 1959. Microbial fermenta-

tion in certain mammals. Science 130, 1192–1194.Kandler, O., Weiss, N., 1986. Regular, nonsporing Gram-positive rods, genus Lactobacillus Beijerinck

1901. In: Sneath, P.H.A., Mair, N.S., Sharpe, M.E., Holt, J.G. (Eds.), Williams Bergey’s Manual ofSystematic Bacteriology, Volume 2. Wilkins, Baltimore, pp. 1208–1234.


Kapperud, G., 1981. Survey on the reservoirs of Yersinia enterocolitica and Yersinia enterocolitica-likebacteria in Scandinavia. Acta Path. Microbiol. Scand. Sec. B 89, 29–35.

Kemper, N., Aschfalk, A., Nieminen, M., Höller, C., 2002. Examination for occurrence of importantenteropathogenic bacteria and parasites (Cryptosporidia) in faeces from reindeer calves. Berl. Munch.Tierarztl. Wochenschr. 115, 340.

Kobayashi, H., Pohjanvitra, T., Hirvelä-Koski, V., Pelkonen, S., 1999. Prevalence and characteristics ofstx and eae positive Escherichia coli in reindeer (Rangifer tarandus) and gulls (Larus ridibundus andL. argentatus). Poster. Pathogenicity and Virulence: Verotoxigenic E. coli in Europe. An EUConcerted Action Project, Liège, Belgium. 8–10 November.

Kocherginskaya, S.A., Aminov, R.I., White, B.A., 2001. Analysis of the rumen bacterial diversity undertwo different diet conditions using denaturing gradient gel electrophoresis, random sequencing, andstatistical ecology approaches. Anaerobe 7, 119–134.

Kummeneje, K., Bakken, G., 1973. Clostridium perfringens enterotoxaemia in reindeer. Nord. Vet. Med.25, 196–202.

Kuronen, H., Hirvelä-Koski, V., Nylund, M., 1998. Salmonellaa poronvasoissa. Poromies 4–5, 54.Larsen, T.S., Nilson, N.O., Blix, A.S., 1985. Seasonal changes in lipogensis and lipolysis in

isolated adipocytes from Svalbard reindeer and Norwegian reindeer. Acta Physiol. Scand. 123,97–104.

Lauterwein, M., Oethinger, M., Belsner, K., Peters, T., Marre, R., 1995. In vitro activities of the lichensecondary metabolites vulpinic acid, (+) usnic acid, and (−) usnic acid against aerobic and anaerobicmicroorganisms. Antimicrob. Agent. Chemoth. 39, 2541–2543.

Lechner-Doll, M., Rutagwenda, T., Schwartz, H.J., Schultka, W., Engelhardt, W., 1990. Seasonal changesof ingesta, mean retention time and forestomach volume in indigenous camels, cattle, sheep and goatsgrazing in thornbush savannah pasture in Kenya. J. Agr. Sci. 115, 409–420.

Lechner-Doll, M., Becker, G., Engelhardt, W., 1991. Short chain fatty acids in the forestomach of camelsand indigenous cattle, sheep and goats and in caecum of donkeys grazing a thornbush savannah pas-ture. In: International Symposium on Nuclear and Related Techniques in Animal Production andHealth. International Atomic Energy Agency, Vienna, pp. 253–268.

Leng, R.A., Brett, D.J., 1966. Simultaneous measurements of the rates of production of acetic, propionicand butyric acids in the rumen of sheep on different diets and correlated between production rates andconcentrations of these acids in the rumen. Brit. J. Nutr. 20, 541–552.

Leng, R.A., Corbett, J.L., Brett, D.J., 1968. Rates of production of volatile fatty acids in the rumen ofgrazing sheep and their relation to ruminal concentrations. Brit. J. Nutr. 22, 57–68.

Lin, C., Raskin, L., Stahl, D.A., 1997. Microbial community structure in gastrointestinal tracts of domes-tic animals: comparative analyses using rRNA-targeted oligonucleotide probes. FEMS Microbiol.Ecol. 22, 281–294.

Lintermans, P., Pohl, P., 1983. Salmonella infections in calves and piglets. Ann. Rech. Vet. 14, 412–419.Lubinski, G., 1958. Ophryoschlecidae (Ciliata: Entodinimorphida) of reindeer (Rangifer tarandus L.)

from the Canadian Arctic. I: Entodiniinae. Can. J. Zool. 36, 819–835.Mackie, R.I., Aminov, R.I., White, B.A., McSweeney, C.S., 2000. Molecular ecology and diversity in gut

microbial ecosystems. In: Cronjé, P.B. (Ed.), Ruminant Physiology: Digestion, Metabolism, Growthand Reproduction. CAB International, New York, pp. 61–77.

Mackintosh, C.G., 1998. Deer health and disease. Acta Vet. Hung. 46, 381–394.Manarelli, B.M., 1988. Deoxyribonucleic acid relatedness among strains of the species Butyrivibrio

fibrisolvens. Int. J. Syst. Bact. 38, 340–347.Mannarelli, B.M., Evans, S., Lee, D., 1990a. Cloning, sequencing, and expression of a xylanase gene

from the anaerobic ruminal bacterium Butyrivibrio fibrisolvens. J. Bacteriol. 172, 4247–4254.Mannarelli, B.M., Robert, J.S., Lee, D., Ericsson, L., 1990b. Taxonomic relatedness of Butyrivibrio,

Lachnospira, Roseburia, and Eubacterium species as determined by DNA hybridization and extracellular-polysaccharide analysis. Int. J. Syst. Bact. 40, 370–378.

Mason, S.F., 1962. A History of the Sciences. Collier Books, New York.Mathiesen, S.D., 1999. Comparative aspects of digestion in reindeer. D.Phil. Thesis, University of

Tromsø, Norway, p. 202.Mathiesen, S.D., Aagnes, T.H., 1990. Microbial digestion in Norwegian reindeer on South Georgia.

Norwegian Polar Institute, Meddelelser 113, 27–35.


Mathiesen, S.D., Orpin, C.G., Blix, A.S., 1984a. Rumen microbial adaptation to fiber digestion inSvalbard reindeer. Can. J. Anim. Sci. 64, 261–262.

Mathiesen, S.D., Rognmo, A., Blix, A.S., 1984b. A test of the usefulness of a commercially available mill“waste product” (AB–84) as feed for starving reindeer. Rangifer 4, 28–34.

Mathiesen, S.D., Orpin, C.G., Greenwood, Y., Blix, A.S., 1987. Seasonal changes in the cecal microfloraof the high Arctic Svalbard reindeer (Rangifer tarandus platyrhynchus). Appl. Environ. Microbiol. 53,114–118.

Mathiesen, S.D., Sørmo, W., Aagnes-Utsi, T.H., 1999. Comparative aspects of volatile fatty acids in rein-deer (Rangifer tarandus tarandus) in northern Norway and on South Georgia. Rangifer 20, 201–210.

Mathiesen, S.D., Haga, Ø.E., Kaino, T., Tyler, N.J.C., 2000. Diet composition, rumen papillation andmaintenance of carcass mass in female Norwegian reindeer (Rangifer tarandus tarandus). J. Zool.Lond. 251, 129–138.

McBee, R.H., Johnson, J.L., Bryant, M.P., 1969. Ruminal microorganisms from elk. J. Wildlife Mgmt.33, 181–186.

Müller, K., 2001. Pharmaceutically relevant metabolites from lichens. Appl. Microbiol. Biotechnol. 56, 9–16.Munoz, M., Alvarez, M., Lanza, I., Carmenes, P., 1996. Role of enteric pathogens in the aetiology of

neonatal diarrhoea in lambs and goat kids in Spain. Epidemiol. Infect. 117, 203–211.Myklebust, R., Mayhew, T.M., 1997. Further evidence of species variation in mechanisms of epithelial

cell loss in mammalian small intestine: ultrastructural studies on the reindeer (Rangifer tarandus) andseal (Phoca groenlandica). Cell Tiss. Res. 291, 513–523.

Nili, N., Brooker, J.D., 1995. A defined medium for rumen bacteria and identification of strains impairedin de novo biosynthesis of certain amino acids. Lett. Appl. Microbiol. 21, 69–74.

Nili, N., Brooker, J.D., 1997. Lack of methionine biosynthesis de novo in Butyrivibrio fibrisolvens strainE14. Lett. Appl. Microbiol. 25, 85–90.

Nilssen, K.J., Sundsfjord, J.A., Blix, A.S., 1984. Regulation of metabolic rate in Svalbard and Norwegianreindeer. Amer. J. Physiol. 147, R963–R967.

Nilssen, K.J., Mathiesen, S.D., Blix, A.S., 1994. Metabolic rate in muskoxen. Rangifer 14, 79–81.Norberg, H.J., Mathiesen, S.D., 1998. Emergency feeding of reindeer with high quality first cut timothy

silage. Rangifer 18, 65–72.Old, I.G., Phillips, S.E.V., Stockley, P.G., Girons, I.S., 1991. Regulation of methionine biosynthesis in

the Enterotacteriaceae. Progr. Biophys. Molec. Biol. 56, 145–185.Olsen, M.A., 2000. Microbial digestion in reindeer and minke whales. D.Phil. Thesis, University of

Tromsø, Norway, p. 114.Olsen, M.A., Mathiesen, S.D., 1998. The bacterial population adherent to plant particles in the rumen of

reindeer (Rangifer tarandus tarandus) fed lichen, timothy hay or silage. Rangifer 18, 55–64.Olsen, M.A., Mathiesen, S.D., 1999. Fibre digestion in the rumen of lichen-fed reindeer. S. Afr. J. Anim.

Sci. 29, 183–184.Olsen, M.A., Aagnes, T.H., Mathiesen, S.D., 1997. The effect of timothy silage on the rumen bacterial pop-

ulation in rumen fluid of reindeer (Rangifer tarandus tarandus). FEMS Microbiol. Ecol. 24, 127–136.Olsen, M.A., Mathiesen, S.D., Mackie, R.I., 2002. Ecological analysis of the rumen bacteria in two indige-

nous Arctic reindeer. Amer. Soc. Microbiol. 102nd General Meeting, Salt Lake City, Abstr. N-80, p. 319.Orpin, C.G., 1974. The rumen flagellate Callimastix frontalis: does sequestration occur? J. Gen.

Microbiol. 84, 395–398.Orpin, C.G., Joblin, K.N., 1997. The rumen anaerobic fungi. In: Hobson, P.N., Stewart, C.S. (Eds.), The

Rumen Microbial Ecosystem. Blackie Academic and Professional, London, pp. 140–195.Orpin, C.G., Letcher, A.J., 1979. Utilization of cellulose, starch, xylan, and other hemicelluloses for

growth by the rumen phycomycete Neocallimastix frontalis. Curr. Microbiol. 3, 121–124.Orpin, C.G., Mathiesen, S.D., 1988. Microcetus lappus gen. nov. sp. nov.: new species of ciliated proto-

zoon from bovine rumen. Appl. Environ. Microbiol. 52, 527–530.Orpin, C.G., Mathiesen, S.D., 1990. Microbiology of digestion in Svalbard reindeer (Rangifer tarandus

platyrhynchus). Rangifer (Special issue) 3, 187–199.Orpin, C.G., Greenwood, Y., Hall, F.J., Pareson, I.W., 1985. The rumen microbiology of seaweed diges-

tion of Orkney sheep. J. Appl. Bacteriol. 59, 585–596.


Orpin, C.G., Jordan, J.D., Mathiesen, S.D., Yeal, N.L., Hazlewood, G.P., Mann, S.P., 1987. Plasmid pro-files in the rumen bacteria Selenomonas ruminantium and Butyrivibrio fibrisolvens. Appl. Bacteriol.XIII (Abstr.).

Ørskov, R.G., 1992. Protein Nutrition in Ruminants. Academic Press, London, p. 175.Pearson, H.A., 1967. Rumen microorganisms in buffalo from southern Utah. Appl. Environ. Microbiol.

15, 1250–1451.Pearson, H.A., 1969. Rumen ecology in mule deer. Appl. Microbiol. 17, 819–824.Petit, L., Gibert, M., Popoff, R., 1999. Clostridium perfringens: toxin type and genotype. Trends

Microbiol. 3, 104–110.Ramsak, A., Pereka, M., Tajima, K., Martin, J.C., Wood, J., Johnston, M.E.A., Aminov, R.I., Flint, H.J.,

Avgustin, G., 2000. Unravelling the genetic diversity of ruminal bacteria belonging to the CFB phylum. FEMS Microbiol. Ecol. 33, 69–79.

Randi, E., Mucci, N., Pierpaoli, M., Douzery, E., 1998. New phylogenic perspectives on the Cervidae (Artio-dactyla) are provided by the mitochondrial cytochrome b gene. Proc. Roy. Soc. London, B 265, 793–801.

Rehbinder, C., Nikander, S., 1999. Reindeer and Reindeer Diseases (in Swedish). Studentlitteratur, Lund,Sweden.

Ryg, M., Jacobsen, E., 1982. Seasonal changes in growth rate, feed intake, growth hormone, and thyroidhormones in young male reindeer (Rangifer tarandus tarandus). Can. J. Zool. 60, 15–23.

Salivigsen, O., 1979. The last deglaciation of Svalbard. Boreas 8, 229–231.Salmonen, E., 1996. Clinical uses of probiotics for stabilizing the gut mucosal barrier: successful strains

for future challenges. Anton. van Leewenh. 70, 347–358.Savage, D.B., 1989. The normal human microflora composition. In: Grubb, T., Midtvedt, T., Norin, E. (Eds.),

The Regulatory and Protective Role of the Normal Microflora. Macmillan, Houndsmills, pp. 1–18.Simpson, J.M., McCracken, V.J., White, B.A., Gaskins, H.R., Mackie, R.I., 1999. Application of denat-

urant gradient gel electrophoresis for the analysis of the porcine gastrointestinal microbiota. J. Microbiol. Meth. 36, 167–179.

Simpson, J.M., McCracken, V.J., Gaskins, H.R., Mackie, R.I., 2000. Denaturing gradient gel elec-trophoresis analyses of 16S ribosomal DNA amplicons to monitor changes in fecal bacterial popula-tions of weaning pigs after introduction of Lactobacillus reuteri Strain MM53. Appl. Environ.Microbiol. 66, 4705–4714.

Skjenneberg, S., Slagsvold, L., 1968. Reindeer husbandry and its natural basis (in Norwegian).Universitetsforlaget Oslo, Bergen, Tromsø, p. 332.

Slomiany, B.L., Murty, V.L.N., Piotrowski, J., Slomiany, A., 1994. Gastroprotective agents in mucosaldefense against Helicobacter pylori. Gen. Pharmacol. 5, 833–841.

Songer, J.G., 1998. Clostridial diseases of small ruminants. Vet. Res. 29, 219–232.Sørmo, W., 1998. Interactions between the function of the digestive system and the pasture plants in rein-

deer. D.Sci. Thesis, Department of Arctic Biology and Institute of Medical Biology, University ofTromsø, Norway, p. 94.

Sørmo, W., Mathiesen, S.D., 1993. Bacteria in the small intestine of lichen-fed Norwegian reindeer(Rangifer tarandus tarandus). Lett. Appl. Microbiol. 16, 170–172.

Sørmo, W., Aagnes, T.H., Olsen, M.A., Mathiesen, S.D., 1994. The bacteriology of the small intestinalmucosa of free-living reindeer. Rangifer 14, 65–78.

Sørmo, W., Haga, Ø.E., White, R.G., Mathiesen, S.D. 1997. Comparative aspects of volatile fatty acidproduction in the rumen and distal chamber in Svalbard reindeer. Rangifer 17(2):81–95.

Sørmo, W., Haga, Ø.E., Mathiesen, S.D., 1998. Cellulolysis in the rumen and distal fermentation cham-ber in Svalbard reindeer (Rangifer tarandus platyrhynchus). Rangifer 18, 47–50.

Sørmo, W., Haga, Ø.E., Gaare, E., Langvatn, R., Mathiesen, S.D., 1999. Forage chemistry and digestivesystem in Svalbard reindeer. J. Zool. Lond. 247, 247–256.

Staaland, H., Jacobsen, E., White, R.G., 1979. Comparison of the digestive tract in Svalbard andNorwegian reindeer. Arctic Alpine Res. 11, 457–466.

Steiner, L., Busato, A., Burnens, A., Gaillard, C., 1997. Frequency and etiology of calf losses and calfdiseases before weaning in cow-calf farms. II. Microbiological and parasitological diagnoses in diar-rheic calves (in German). Dtsch Tierarztl. Wochenschr. 104, 169–173.


Stokkan, K.-A., Tyler, N.J.C., Reiter, R., 1994. The pineal gland signals autumn to reindeer (Rangifertarandus tarandus) exposed to the continuous daylight of Arctic summer. Can. J. Zool. 72, 904–909.

Tajima, K., Aminov, R.I., Nagamine, T., Ogata, K., Nakamura, M., Matsui, H., Benno, Y., 1999. Rumenbacterial diversity as determined by sequence analysis of 16S rDNA library. FEMS Microbiol. Ecol.29, 159–169.

Tajima, K., Arai, S., Ogata, K., Nagamine, T., Matsui., N., Aminov, R.I., Benno, Y., 2000. Rumen bacte-rial community transition during adaptation to high-grain diet. Anaerobe 6, 273–284.

Tajima, K., Nagamine, T., Matsui, H., Nakamura, M., Aminov, R.I., 2001a. Phylogenetic analysis ofarchaeal 16S rRNA libraries from the rumen suggests the existence of a novel group of archaea notassociated with known methanogens. FEMS Microbiol. Lett. 200, 67–72.

Tajima, K., Aminov, R.I., Nagamine, T., Matsui, H., Nakamura, M., Benno, Y., 2001b. Diet-dependentshifts in the bacterial population of the rumen revealed with real-time PCR. Appl. Environ. Microbiol.67, 2766–2774.

Tyler, N.J.C., 1987. Body composition and energy balance of pregnant and non pregnant Svalbard rein-deer during winter. Zool. Soc. Lond. Symp. 57, 203–229.

Tyler, N.J.C., Fauchald, P., Johansen, O., Christiansen, H.R., 1999. Seasonal inappetance and weight lossin female reindeer in winter. In: Hofgaard, A., Ball, J.P., Danell, K., Callaghan, T.V. (Eds). Animalresponse to global change in the north. Ecological Bulletins 47:47–116.

Tzipori, S., 1981. The aetiology and diagnosis of calf diarrhoea. Vet. Rec. 108, 510–515.Ulyatt, M.J., Dellow, D.W., Reid, C.S.W., Bauchop, T., 1975. Structure and function of the large intes-

tine of ruminants. In: McDonald, I.W., Warner, A.V.I. (Eds.), Digestion and Metabolism inRuminants. Adelaide. University of New Eng. Publ. Unit. Armidale, Australia.

Utsi, T.H.A., 1998. Digestive strategies in reindeer in winter. D.Sci. Thesis, University of Tromsø,Norway, p.112.

Van Soest, P.J., 1994. Nutritional Ecology of the Ruminant. Cornell University Press, Ithaca.Vartia, K.O., 1949. Antibiotics in lichens. Ann. Med. Exp. Biol. Fenn. 27, 46–54.Vartia, K.O., 1973. Antibiotics in lichens. In: Ahmadijan, N., Hale, M.F. (Eds.), The Lichens. Academic

Press, New York, pp. 547–561.Westerling, B., 1970. Rumen ciliate fauna of semi-domestic reindeer (Rangifer tarandus t) in Finland:

Composition, volume and some seasonal variations. Acta Zool. Fennica 127, 1–76.Westerling, B., 1975a. A comparative study of the intestinal anatomy of deer. Anatomischer Anzeiger.

137, 178–186.Westerling, B., 1975b. Effect of changes in diet on the reindeer mucosa. In: Luick, J.R., Lent, P.C., Klein,

D.R., White, R.G. (Eds.), Proceedings of the 1st International Reindeer and Caribou Symposium,University of Alaska, Fairbanks, pp. 278–283.

Weston, R.H., Hogan, J.P., 1968. The digestion of pasture plants by sheep. I: Ruminal production ofvolatile fatty acids by sheep offered diets of rye grass and forage oats. Austr. J. Agr. Res. 19, 419–432.

White, B.A., Cann, I.K.O., Kocherginskaya, S.A., Aminov, R.I., Thill, L.A., Mackie, R.I., Onodera, R.,1999. Molecular analysis of Archaea, Bacteria and Eucarya communities in the rumen. AJAS 12,129–138.

White, R.G., Grau, A.M., 1975. Volatile fatty acid (VFA) production in the rumen and cecum of reindeer.In: Luick, J.R., Lent, P.C., Klein., D.R., White, R.G. (Eds.), Proceedings of the 1st InternationalReindeer and Caribou Symposium, University of Alaska, Fairbanks, pp. 284–289.

White, R.G., Staaland, H., 1983. Ruminal volatile fatty acid production as an indicator of forage qualityin Svalbard reindeer. Acta Zool. Fennica. 175, 61–63.

White, R.G., Holleman, D.F., Hubbert, M.E., Staaland, H., 1987. Herbivores in cold climate. In: Nutritionof Herbivores, Academic Press, Australia, pp. 465–486.

Whitford, M.F., Forster, R.J., Beard, C.E., Gong, J., Teather, R.M., 1998. Phylogenetic analysis of rumenbacteria by comparative sequence analysis of cloned 16S rRNA genes. Anaerobe 4, 153–163.

Williams, A.G., Coleman, G.S., 1988. The rumen protozoa. In: Hobson, P.N (Ed.), The Rumen MicrobialEcosystem. Elsevier, London, pp. 77–128.


Microbial colonization of the gastrointestinal tract of puppies and kittens starts afterbirth, and the composition of the intestinal microflora approaches the spectrum inadult dogs and cats during the first weeks of life. The colonization of the gastrointestinal tract starts with an aerobic type of flora, mainly streptococci,Escherichia coli and in puppies, staphylococci. Lactobacilli have been isolated afterthe first week of life and a similar time schedule was observed for Bacteroides, whichcolonize the lower intestinal tract from the second week of life. In contrast to otherspecies, the composition of the gut flora is characterized by relatively high numbersof Clostridium perfringens and also lecithinase negative clostridia, probably reflect-ing the carnivorous type of diet. Puppies may acquire gastric Helicobacter infectionfrom dams or can infect each other in early life. Clostridium difficile is neither fordogs nor for cats an important enteric agent. The possibility of occasional humaninfections by household dogs and cats needs further investigation. Young puppies andkittens can be regarded as potential transmitters of Campylobacter spp., whereas salmonellosis seems to occur rarely and there is no unequivocal link between theoccurrence of these potentially harmful bacteria and the occurrence of diarrhoea.

1. INTRODUCTION

Historically, dogs and cats were useful models for studying the intestinal flora inhumans. Some researchers were driven by their specific interests into the potentialhealth implications of zoonotic disease transmission. Extensive work was done fromthe beginning of the 20th century, regarding the physiological and pathologicalmicrobial colonization of the oral cavity, the stomach and the intestinal tract(Oppmann, 2001). In contrast to that in other experimental animals, the microbial

5 Microbial ecology of the gastrointestinaltract of the growing dog and cat

J. Zentek

Institute of Nutrition, University of Veterinary Medicine, Vienna A-1210 Vienna,Veterinärplatz 1, Austria

103



colonization of the feline and canine digestive tract has been studied less intensivelyin the later years of the 20th century. This can be explained by the decreasing significance of dogs and cats as models for human research and by the limited inter-est of researchers in specific particularities of the intestinal flora in these species.

The microbial colonization of the gastrointestinal tract in newborn dogs and catsbegins immediately after the delivery from the sterile uterine environment. It is influ-enced by maternal, environmental, and nutritional factors. After birth, pups are normally fed exclusively on dam’s milk until the age of 3−4 weeks. At that time, theenergy requirements of the growing young are beginning to exceed the dam’s capac-ity for milk production. The first diet introduced is normally milk-based, liquid foodthat can be easily ingested and which is characterized by similar nutritional traits tocanine or feline milk. This type of feed is gradually changed to a more carnivoroustype of diet, based on such ingredients as animal proteins and fats, starch and lowamounts of dietary fibre. This shift of dietary habits is normally coinciding with environmental changes and both may affect the composition of the intestinalmicroflora, but specific data for this period are lacking.

2. MICROBIAL COLONIZATION OF THE ORAL CAVITY

The development of the microbial colonization of the oral cavity has not been stud-ied in dogs and cats and no data are available for newborn kittens and puppies. A comparison of human, canine and feline mouth flora (Rayan et al., 1991) demon-strated that human oral flora contained the smallest number of bacteria followed by dog and cat oral flora. In cats between 6 and 12 months of age the mean numberof viable bacteria from samples taken from gingival margins was log 10 (10.7) witha range of 7 to 16 different species (Love et al., 1990). Of a total number of 150 isolates studied, 73% were obligate anaerobes. Of the facultatively anaerobic species,Actinomyces (including Actinomyces viscosus, Actinomyces hordeovulneris andActinomyces denticolens) comprised 12%, Pasteurella multocida 9.3% andPropionibacterium species 6%. Gram-negative bacilli belonging to the generaBacteroides and Fusobacterium represented 77% of the obligate anaerobes isolated.Clostridium villosum comprised 10.1% of the obligatory anaerobic isolates,Wolinella species made up 6.4%, while 4.6% were Peptostreptococcus anaerobius.The most commonly isolated obligatory anaerobic species was Clostridium villosumand the most commonly isolated facultatively anaerobic species was Pasteurellamultocida. Pasteurella spp. have been described early in the oral cavity of dogs andcats (Bisgaard and Mutters, 1986). Bacteroides species were isolated from the oral cavity of dogs and were also demonstrated in cats, in part associated with diseases (Love et al., 1989, 1990, 1992b). By dot-blot hybridization assay pig-mented asaccharolytic Bacteroides/Porphyromonas species were investigated.Bacteroides salivosus was distinguished from other anaerobic species isolated fromnormal and diseased mouths of cats (Love et al., 1992a). Bacteroides tectum andBacteroides fragilis were cultured from the oral cavity of cats (Love and Bailey, 1993).

104 J. Zentek

Bacteroides species constituted, as a proportion of all anaerobic isolates examined,37.5% from normal gingiva and 27.7% from diseased gingiva in cats (Love et al.,1989). Gingival scrapings of dogs were examined for the presence of CDC GroupsEF−4 bacteria. Fifty-nine EF−4 strains were isolated from 92% of 49 dogs. Amongthe Group EF−4 bacteria, the majority of isolates belonged to the arginine-negative (biovar “b”) Group EF−4 (42 strains recovered in 82% of dogs). Seventeen arginine-positive strains (biovar “a”) were recovered from only 35% of dogs (Ganiere et al.,1995). Occurrence of Gram-positive, catalase-negative, facultatively anaerobic cocciwas reported. Different sublines belonging to the genus Gemella were cultured,among these Gemella haemolysans, Gemella bergeri, Gemella morbillorum andGemella sanguinis (Collins et al., 1999). Ureaplasma spp. have been isolated from the oral cavities of cats and dogs and genomic relatedness was shown (Harasawa et al., 1990a,b, 1993). Filamentous bacteria could be cultured from oral eosinophilicgranulomas of a cat (Russell et al., 1988). Capnocytophaga spp. have been isolatedfrom the oral flora of dogs and cats (Blanche et al., 1998) and are of importance as totheir infectious potential in dog bite wounds.

3. MICROBIAL COLONIZATION OF THE STOMACH

The stomach seems to be colonized by few bacteria in newborn dogs and cats, com-pared to calves, lambs or piglets. Clostridium perfringens and Staphylococcusaureus were the predominant bacteria in the stomach contents of puppies fedmother’s milk, while Bacteroides spp. were not detectable over the suckling period.Lactobacilli were found only in puppies older than 10 days (Smith, 1965).Staphylococcus aureus may have originated from the bitches’ breast skin, for thisorganism was isolated from this location and from the stomach contents of puppiesbut it was not found in the faeces of the dams (table 1). In clinically healthy

Microbial ecology of the dog and cat 105

Table 1. Microflora (log10/g chyme) in the stomach of puppies receiving dam’s milk (Smith, 1965)

post-parturient bitches staphylococci were isolated from 30.3% of milk samples inpure cultures and 6.8% in anacultures. Small numbers of bacteria were isolated inmost of the samples, but 67.4% showed moderate bacterial growth. According tothis study, there was no direct influence of lactiferous gland colonization on themortality of puppies (Kuhn et al., 1991).

Advance in age and the potential influence on the luminal gastrointestinal microfloraof beagle dogs was investigated by comparing dogs less than 12 months of age withdogs more than 11 years of age (Benno et al., 1992). There was no clear tendency forthe bacterial counts to be higher in the older beagles compared to dogs below 1 year.Lactobacilli, enterobacteria and streptococci were found in the gastric chyme of mostdogs, while other bacteria and also yeasts were determined only irregularly (table 2).

Gram-negative bacilli were found adhering to the gastric surface as well as withinepithelial cells in the stomach of a puppy (Wada et al., 1996). Puppies may acquiregastric Helicobacter infection, as proven for Helicobacter salomonis, from dams during lactation or puppies can infect each other in early life (Hanninen et al., 1998).

In kittens, Clostridium perfringens, Escherichia coli and streptococci were foundin the stomach contents of newborns in reasonably high numbers (table 3). In kittens over 5 days old Gram-positive anaerobic rod-like bacteria formed the majorcomponent of the chyme in the stomach. Lactobacilli were only found in 10-day-oldkittens in higher numbers. In contrast to puppies, Staphylococcus aureus was not

106 J. Zentek

Table 2. Microflora of the stomach (log10/g chyme) in beagles less than 1 year old compared tobeagles more than 11 years old (Benno et al., 1992)

cultured from the chyme of any location of the gastrointestinal tract, and breastswabs taken from the queens were negative, too.

The gastric microflora of felines was investigated in three unweaned kittens and inadult cats fed a conventional or chemically defined, elemental ration (Osbaldiston andStowe, 1971). According to this study, enterococci and lactobacilli were the predominantmicroflora. Streptococci, Escherichia coli, Clostridium spp. and Bacteroides spp. wereisolated from one out of three individuals (table 4). Dietary changes were not associated


Table 3. Microflora (log10/g chyme) in the stomach of kittens receiving dam’s milk (Smith, 1965)

Table 4. Microflora (log10 /g chyme) in the stomach chyme of kittens receiving dam’s milk andadult cats fed a conventional or a chemically defined diet (Osbaldiston and Stowe, 1971)

with significant changes in the bacterial populations or patterns of distribution within theintestinal tract nor was there a clear difference between the kittens and the adult cats.

4. MICROBIAL COLONIZATION OF THE SMALL INTESTINE

In newborns the small intestine is sterile, but first colonization with bacteria occursduring the few hours after birth. Within 12 hours, the upper (table 5) and the lower(table 6) parts of the small intestine of canine puppies were colonized by strepto-cocci, staphylococci and Clostridium perfringens and after a few days also byEscherichia coli and lactobacilli. Bacteroides species were not identified as part ofthe small intestinal microflora in this study (Smith, 1965).

The small intestinal microflora may contribute to infectious diseases in puppies.Gram-positive bacilli in association with focal to diffuse necrosis of the superficialportions of the villi, were observed in histological sections of specimens of smallintestine from all except one of 57 dogs from which parvovirus and Clostridium perfringens had been identified. These findings indicate that Clostridium perfringensfrequently proliferates in dogs with parvovirus infection (Turk et al., 1992). Entericinfection with an attaching and effacing Escherichia coli was diagnosed in a puppywith diarrhoea. Characteristic lesions of bacterial attachment of the brush border ofthe enterocytes were demonstrated by transmission electron microscopy. TheEscherichia coli strain isolated from the small intestine belonged to serotypeO49:H10, and a positive immunoperoxidase reaction was obtained on the bacteriaattached to the enterocytes with an anti-Escherichia coli O49 antiserum (Broes et al., 1988). In puppies with a clinical history of gastrointestinal disease attachingand effacing Escherichia coli (AEEC) or enterotoxigenic Escherichia coli (ETEC)

108 J. Zentek

Table 5. Microflora (log10/g chyme) in the upper small intestine (duodenum) of puppies receivingdam’s milk (Smith, 1965)

infection has to be considered often with coinfection with other enteric pathogens(Drolet et al., 1994). Campylobacter jejuni was inoculated experimentally into theduodenum of 13 puppies (2 to 5 weeks old). All had positive faecal cultures for 1 to10 days without clinical signs of disease (Boosinger and Dillon, 1992).

Different and in relation to the current standards inadequate methods have to betaken into account when comparing the results of the studies in puppies and in olderdogs. Compared to the findings in puppies, a broader spectrum and higher counts ofintestinal bacteria were demonstrated in young dogs less than 1 year old and in dogsat the age of over 11 years (table 7). The total counts and the numbers of lactobacilli,Bacteroides, peptostreptococci, and bifidobacteria in the elderly animals were sig-nificantly lower than those in younger dogs. The numbers of lecithinase-positiveclostridia (mainly Clostridium perfringens) and bacilli were significantly higher inolder dogs compared to dogs below 1 year.

Clostridium perfringens, Escherichia coli and streptococci were the organismsthat first colonized the alimentary tract of kittens (tables 8 and 9). Gram-positiveanaerobic rods formed the major component of the flora in kittens that were over 5 days old. Individual variations have to be taken into account, as there were highnumbers of lactobacilli in one 120-day-old kitten. Lactobacilli were only rarelyfound in the other kittens and in lower numbers. Uchida et al. (1971) investigatedthe intestinal microflora of kittens after weaning at the age of 12 weeks, when a dietbased on horse meat and milk was fed. They found Bacteroides, bifidobacteria, lac-tobacilli, eubacteria, clostridia, streptococci, staphylococci, Enterobacteriaceae andmoulds as part of the duodenal and ileal microflora, indicating the shift in the com-position of the small intestinal microflora depending on age and diet and theexpected establishment of anaerobic conditions in the gut.


Table 6. Microflora (log10/g chyme) in the lower small intestine (ileum) of puppies receivingdam’s milk (Smith, 1965)

Table 7. Duodenal and ileal microflora (log10 /g chyme) in beagles (n = 8) of different ages(Benno et al., 1992)

Table 8. Microflora (log10/g chyme) in the upper small intestine (duodenum) of kittens receivingdam’s milk (Smith, 1965)

5. MICROBIAL COLONIZATION OF THE COLON AND THE RECTUM

Bacteria are much more numerous in the colon and rectum compared to the smallintestine in newborn puppies and kittens. The flora in puppies consists ofClostridium perfringens, streptococci and staphylococci shortly after birth and thisis subsequently followed by Escherichia coli, lactobacilli and Bacteroides (tables 10and 11). Bacteroides have been found from day 9 and are the dominating part of thecolonic microflora in older puppies (Smith, 1965). In another study in newborn pup-pies, rectal swabs were investigated periodically from nine puppies from threebitches over a period of 55 days after birth (Matsumoto et al., 1976). The bacterial


Table 9. Microflora (log10/g chyme) in the lower small intestine (ileum) of kittens receiving dam’smilk (Smith, 1965)

Table 10. Microflora (log10/g chyme) in the colon of puppies receiving dam’s milk (Smith, 1965)

groups most frequently encountered after 6 hours post natum were staphylococci,streptococci, Enterobacteriaceae and clostridia; lactobacilli, Bacteroides and bifi-dobacteria were found later, although their time of appearance varied considerablywith individuals. After their appearance these organisms showed sharp fluctuationsin number. The total count of viable bacteria in the faecal samples was log 109/g or more 24 h after birth. Streptococci and Enterobacteriaceae were predominant up to 7 days of age. After that no definite groups were prevalent. Lactobacilli andbifidobacteria, however, were predominant at 42 days of age and later. The bacter-ial flora in puppies at this stage was identical with the one in adult dogs (Matsumotoet al., 1976). The composition of the normal staphylococcal flora of bitches and theirlitters in a breeding unit showed that Staphylococcus intermedius formed the predominant staphylococcal isolate. Staphylococcus intermedius counts at the vagi-nal vestibulum of the pregnant bitches were higher than at any other site sampledand did not alter markedly until whelping when a decrease was observed.Staphylococcus intermedius was not found at the anal site in any of the six bitchesand only transiently colonized some of the puppies (Allaker et al., 1992).

The impact of age on the gastrointestinal microflora of beagle dogs seems to bemore pronounced in the large intestine compared to the stomach or the small intestine(Benno et al., 1992). The numbers of peptostreptococci and bifidobacteria in the largebowel of the younger dogs were significantly higher than those in elderly dogs. Largerpopulations of lactobacilli were found in the caecum and colon of dogs less than 1 yearold, whereas the numbers of Clostridium perfringens and streptococci increased in theolder population (table 12). The numbers of Bacteroides in the caecum and rectum andeubacteria in the caecum and colon of the elderly dogs were lower compared to theyounger individuals. The incidence of lecithinase-negative clostridia increased in theolder dogs only in the rectum but not in the other locations of the large intestine.

112 J. Zentek

Table 11. Microflora (log10 /g chyme) in the rectum of puppies receiving dam’s milk (Smith, 1965)

Table 12. Large intestinal microflora (log10/g chyme) in beagles (n = 8) of different ages (Benno et al., 1992)

Clostridium difficile was monitored during the first 10 weeks after birth in puppies(Perrin et al., 1993). More than 90% of the puppies and 43% of the dams harbouredClostridium difficile at least once in their faeces and 58% of the puppies carried toxigenicClostridium difficile at least once during the survey. In the puppies, Clostridium difficilecarriage rates ranging from 3.1 to 67.1% were observed. In comparison, the Clostridiumdifficile carriage rate was 1.4% in a control group of healthy dogs more than 3 monthsold. Discrepancies in the toxigenic phenotype of the Clostridium difficile strains isolatedin the same litter, showed that the newborn dogs were transiently infected with differentstrains, and that the dam is often not the source of infection with Clostridium difficile. Theincidence of Clostridium difficile was 46% in faecal samples from healthy puppies withtoxigenic strains found in 61.5% of the healthy neonate dogs (Buogo et al., 1995). Itseems that, in contrast to the significance for man, Clostridium difficile is neither for dogsnor for cats an important enteric agent. The possibility of occasional human infections byhousehold dogs and cats needs further investigation (Weber et al., 1989).

Dogs in a closed breeding unit were shown to be asymptomatic excretors ofCampylobacter at the age of 8 weeks. Increasing serum antibody levels, which werecorrelated with the excretion of organisms, were demonstrated in the puppies andserum antibodies were also demonstrated in adult dogs (Newton et al., 1988). In across-sectional study carried out in Denmark, 72 healthy puppies and kittens were sampled for faecal Campylobacter shedding by culture of rectal swab specimens. 29%of the puppies were positive for Campylobacter spp., with a species distribution of 76%Campylobacter jejuni, 5% Campylobacter coli and 19% Campylobacter upsaliensis.Of the kittens examined, only two (5%) excreted Campylobacter spp.; both strains wereCampylobacter upsaliensis. Young puppies and kittens can be regarded as potentialtransmitters of human-pathogenic Campylobacter spp., including Campylobacterupsaliensis (Hald and Madsen, 1997). In another study, 32% of healthy puppies werepositive for Campylobacter spp. with a peak at the age of 8 weeks (Buogo et al., 1995).Salmonellosis seems to occur very rarely in puppies. Infection with Salmonellatyphimurium induces haemorrhagic enteritis with discrete fibronecrotic areas (King,1988). In bacteriological studies of 159 faecal and intestinal content samples from dogswith diarrhoea, Escherichia coli (73 non-haemolytic) was found in 157 samples,Klebsiella and Staphylococcus aureus were found in nine cases and Salmonella sp. inone (Zschock et al., 1989). Other investigators determined Salmonella in 6.5% of faecal samples of puppies (Buogo et al., 1995), but could not establish a link betweenthe incidence of Salmonella, Campylobacter or Clostridium difficile with episodes ofdiarrhoea. In 11-day-old puppies with diarrhoea, a mixed growth of non-haemolyticEscherichia coli and Enterococcus durans serotype 2 was isolated from the jejunumwith lesions that resembled those reported in natural and experimental Enterococcusdurans infections in foals and gnotobiotic pigs (Collins et al., 1988).

The development of the colonic microflora in kittens has been shown to be com-parable to the findings in puppies. Clostridium perfringens, Escherichia coli and strep-tococci were the first organisms to colonize the alimentary tract of kittens (table 13).

114 J. Zentek

Bacteroides, lactobacilli and Gram-positive obligatory anaerobic rods could bedetected within 2 days after birth (Smith, 1965). Bacteroides were found in thecolon but not in the anterior parts of the alimentary tract in kittens.

Enterococci, Enterobacter, Catenabacterium and, in one individual, yeasts, weredetermined in the midcolon of kittens (table 14). The concentrations were compar-able to the findings in adult cats (Osbaldiston and Stowe, 1971). The concentrations


Table 13. Microflora (log10/g chyme) in the colon of kittens receiving dam’s milk (Smith, 1965)

Table 14. Microflora (log10/g chyme) in the midcolon chyme of kittens receiving dam’s milk andadult cats fed a conventional or a chemically defined diet (Osbaldiston and Stowe, 1971)

of enterococci were greater than in the stomach or in the jejunum. Clostridia werenot detected in these three kittens. The caecal and faecal microflora of 3-month-oldkittens fed on horse meat and milk was dominated by Bacteroides, clostridia, eubac-teria and streptococci and in lower numbers staphylococci, Enterobacteriaceae andmoulds. The total bacterial counts were 9.1−9.2 log10/g and lactobacilli and bifido-bacteria were found only infrequently (Uchida et al., 1971).

Dual infection by Clostridium piliforme and feline panleukopenia virus (FPLV)was found in three kittens. Pathology was characterized by focal necrosis anddesquamation of epithelial cells with occasional neutrophile infiltration in the largeintestine. Large filamentous bacilli and spores were observed in the epithelium(Ikegami et al., 1999). Salmonella typhimurium was isolated in kittens with intestinalcrypt necrosis, hepatic, splenic and lymph node inflammation and necrosis. All hadbeen vaccinated previously with a modified-live virus vaccine. Salmonellosis wasinterpreted as a consequence of a mild immunosuppression induced by vaccination(Foley et al., 1999). Enterococcus hirae was isolated from a 2-month-old femalePersian cat that had been showing episodes of anorexia and diarrhoea. Cocci werelocated along the brush border of small intestinal villi, without significant inflamma-tory infiltration. Similar bacteria were present within hepatic bile ducts and pancre-atic ducts and were associated with suppurative inflammation and exfoliation ofepithelial cells. Faecal culture from an asymptomatic adult female from the same cattery also yielded large numbers of Enterococcus hirae (Lapointe et al., 2000).


The current knowledge of the intestinal microecology of puppies and kittens and itsdevelopment is limited and further investigations are needed. Questions to beaddressed include the significance of the intestinal microflora for kitten and puppylosses and their potential significance as carriers for human diseases. The character-ization of the intestinal microecology should be studied by molecular biologicalmethods and will offer new insights into the significance of the gut bacteria for current topics of high scientific interest, such as the development and the function ofthe intestinal immune system, the aetiology and pathogenesis of chronic inflamma-tory bowel disease, allergies and the potential effects of manipulation of themicroflora by probiotics or dietary means. Age effects on the intestinal microflora areof increasing interest in dogs and cats and both species could be useful models for thesituation in humans, as to the age distribution with an increasing geriatric population.

REFERENCES

Allaker, R.P., Jensen, L., Lloyd, D.H., Lamport, A.I., 1992. Colonization of neonatal puppies by staphy-lococci. Brit. Vet. J. 148, 523−528.

Benno, Y., Nakao, H., Uchida, K., Mitsuoka, T., 1992. Impact of the advances in age on the gastro-intestinal microflora of beagle dogs. J. Vet. Med. Sci. 54, 703−706.

116 J. Zentek

Bisgaard, M., Mutters, R., 1986. Characterization of some previously unclassified “Pasteurella” spp.obtained from the oral cavity of dogs and cats and description of a new species tentatively classifiedwith the family Pasteurellaceae Pohl 1981 and provisionally called taxon 16. Acta Pathol. Microbiol.Immunol. Scand. B 94, 177−184.

Blanche, P., Bloch, E., Sicard, D., 1998. Capnocytophaga canimorsus in the oral flora of dogs and cats.J. Infect. 36, 134.

Boosinger, T.R., Dillon, A.R., 1992. Campylobacter jejuni infections in dogs and the effect of erythro-mycin and tetracycline therapy on fecal shedding. J. Amer. Anim. Hosp. Ass. 28, 33−38.

Broes, A., Drolet, R., Jacques, M., Fairbrother, J.M., Johnson, W.M., 1988. Natural infection with anattaching and effacing Escherichia coli in a diarrheic puppy. Can. J. Vet. Res. 52, 280−282.

Buogo, C., Burnens, A.P., Perrin, J., Nicolet, J., 1995. Presence de Campylobacter spp., Clostridiumdifficile, C. perfringens et salmonelles dans des nichees de chiots et chez des chiens adultes d’unrefuge. Schweizer Arch. Tierheilkde. 137, 165−171.

Collins, J.E., Bergeland, M.E., Lindeman, C.J., Duimstra, J.R., 1988. Enterococcus (Streptococcus)durans adherence in the small intestine of a diarrheic pup. Vet. Pathol. 25, 396−398.

Collins, M.D., Rodriguez, J.M., Foster, G., Sjoden, B., Falsen, E., 1999. Characterization of a Gemella-like organism from the oral cavity of a dog: description of Gemella palaticanis sp. nov. Int. J. Syst.Bacteriol. 49, 1523−1526.

Drolet, R., Fairbrother, J.M., Harel, J., Helie, P., 1994. Attaching and effacing and enterotoxigenicEscherichia coli associated with enteric colibacillosis in the dog. Can. J. Vet. Res. 58, 87−92.

Foley, J.E., Orgad, U., Hirsh, D.C., Poland, A., Pedersen, N.C., 1999. Outbreak of fatal salmonellosis incats following use of a high-titer modified-live panleukopenia virus vaccine. J. Amer. Vet. Med. Ass.214, 67−70.

Ganiere, J.P., Escande, F., Andre-Fontaine, G., Larrat, M., Filloneau, C., 1995. Characterization of groupEF-4 bacteria from the oral cavity of dogs. Vet. Microbiol. 44, 1−9.

Hald, B., Madsen, M., 1997. Healthy puppies and kittens as carriers of Campylobacter spp., with specialreference to Campylobacter upsaliensis. J. Clin. Microbiol. 35, 3351−3352.

Hanninen, M.L., Happonen, I., Jalava, K., 1998. Transmission of canine gastric Helicobacter salomonisinfection from dam to offspring and between puppies. Vet. Microbiol. 62, 47−58.

Harasawa, R., Imada, Y., Ito, M., Koshimizu, K., Cassell, G.H., Barile, M.F., 1990a. Ureaplasma felinum sp.nov. and Ureaplasma cati sp. nov. isolated from the oral cavities of cats. Int. J. Syst. Bacteriol. 40,45−51.

Harasawa, R., Stephens, E.B., Koshimizu, K., Pan, I.J., Barile, M.F., 1990b. DNA relatedness amongestablished Ureaplasma species and unidentified feline and canine serogroups. Int. J. Syst. Bacteriol.40, 52−55.

Harasawa, R., Imada, Y., Kotani, H., Koshimizu, K., Barile, M.F., 1993. Ureaplasma canigenitalium sp.nov., isolated from dogs. Int. J. Syst. Bacteriol. 43, 640−644.

Ikegami, T., Shirota, K., Goto, K., Takakura, A., Itoh, T., Kawamura, S., Une, Y., Nomura, Y., Fujiwara, K.,1999. Enterocolitis associated with dual infection by Clostridium piliforme and feline panleukopeniavirus in three kittens. Vet. Pathol. 36, 613−615.

King, J.M., 1988. Intestinal salmonellosis. Vet. Med. 83, 765.Kuhn, G., Pohl, S., Hingst, V., 1991. Elevation of the bacteriological content of milk of clinically

unaffected lactating bitches of a canine research stock. Berl. Münch. Tierarztl. Wochenschr. 104,130−133.

Lapointe, J.M., Higgins, R., Barrette, N., Milette, S., 2000. Enterococcus hirae enteropathy with ascend-ing cholangitis and pancreatitis in a kitten. Vet. Pathol. 37, 282−284.

Love, D.N., Bailey, G.D., 1993. Chromosomal DNA probes for the identification of Bacteroides tectumand Bacteroides fragilis from the oral cavity of cats. Vet. Microbiol. 34, 89−95.

Love, D.N., Johnson, J.L., Moore, L.V.H., 1989. Bacteroides species from the oral cavity and oral-associated diseases of cats. Vet. Microbiol. 19, 275−281.

Love, D.N., Vekselstein, R., Collings, S., 1990. The obligate and facultatively anaerobic bacterial floraof the normal feline gingival margin. Vet. Microbiol. 22, 267−275.

Love, D.N., Bailey, G.D., Bastin, D., 1992a. Chromosomal DNA probes for the identification of asaccha-rolytic anaerobic pigmented bacterial rods from the oral cavity of cats. Vet. Microbiol. 31, 287−295.


Love, D.N., Bailey, G.D., Collings, S., Briscoe, D.A., 1992b. Description of Porphyromonas circumden-taria sp. nov. and reassignment of Bacteroides salivosus (Love, Johnson, Jones, and Calverley, 1987)as Porphyromonas (Shah and Collins, 1988) salivosa comb. nov. Int. J. Syst. Bacteriol. 42, 434−438.

Matsumoto, H., Baba, E., Ishikawa, H., Hodate, Y., 1976. Bacterial flora of the alimentary canal of dogs. II.Development of the faecal bacterial flora in puppies. Jap. J. Vet. Sci. 38, 485−494.

Newton, C.M., Newell, D.G., Wood, M., Baskerville, M., 1988. Campylobacter infection in a closed dogbreeding colony. Vet. Rec. 123, 152−154.

Oppmann, H., 2001. Studies in nutritional research in dogs (digestion, energy and protein metabolism)between the years 1900 and 1950. Vet. Med., Thesis, Tierärztliche Hochschule, Hannover.

Osbaldiston, G.W., Stowe, E.C., 1971. Microflora of alimentary tract of cats. Amer. J. Vet. Res. 32, 1399−1405.

Perrin, J., Buogo, C., Gallusser, A., Burnens, A.P., Nicolet, J., 1993. Intestinal carriage of Clostridiumdifficile in neonate dogs. J. Vet. Med. B 40, 222−226.

Rayan, G.M., Downard, D., Cahill, S., Flournoy, D.J., 1991. A comparison of human and animal mouthflora. J. Okla. State Med. Assoc. 84, 510−515.

Russell, R.G., Slattum, M.M., Abkowitz, J., 1988. Filamentous bacteria in oral eosinophilic granulomasof a cat. Vet. Pathol. 25, 249−250.

Smith, H.W., 1965. The development of the flora of the alimentary tract in young animals. J. Pathol.Bacteriol. 90, 495−513.

Turk, J., Fales, W., Miller, M., Pace, L., Fischer, J., Johnson, G., Kreeger, J., Turnquist, S., Pittman, L.,Rottinghaus, A., Gosser, H., 1992. Enteric Clostridium-perfringens infection associated with parvo-viral enteritis in dogs – 74 cases (1987−1990). J. Amer. Vet. Med. Ass. 200, 991−994.

Uchida, K., Terada, A., Tanaka, S., Ichiman, Y., 1971. Intestinal microflora of cats. Bull. Nippon Vet.Zootech. Coll., 11-16.

Wada, Y., Kondo, H., Nakaoka, Y., Kubo, M., 1996. Gastric attaching and effacing Escherichia colilesions in a puppy with naturally occurring enteric colibacillosis and concurrent canine distempervirus infection. Vet. Pathol. 33, 717−720.

Weber, A., Kroth, P., Heil, G., 1989. Occurrence of Clostridium difficile in faeces of dogs and cats. J. Vet.Med. B 36, 568−576.

Zschock, M., Herbst, W., Lange, H., Hamann, H.P., Schliesser, T., 1989. Results from microbiologicalstudies (bacteriology and electron microscopy) of diarrhoea in puppies. Tierärztl. Praxis 17, 93−95.

118 J. Zentek

Advances in molecular biology have led to the development of a variety of culture-independent approaches to describe bacterial communities. Most of the new strate-gies, based on the analysis of DNA or RNA allow direct investigation of communitydiversity, structure and phylogeny of microorganisms in the gastrointestinal tract.This chapter will cover molecular approaches for studying the microbial flora, andthe molecular tools to monitor the presence of specific strains in the intestine.Special emphasis will be on the advantages and disadvantages, respectively, of various DNA- or RNA-based methods for the study of the microbiota in the gastrointestinal tract of humans and animals.

1. INTRODUCTION

The bacterial flora of the gastrointestinal (GI) tract of humans and animals has beenstudied more extensively than that of any other anatomical site. This may be due tothe high number of bacteria encountered in the intestine. The total number ofbacteria resident in the human gastrointestinal tract has been estimated to reach 1014

bacterial cells, thus outnumbering the total number of body cells (Savage, 1977).The highest bacterial density is found in the distal colon. The composition differsbetween and within animal species. In humans, there are differences in the compo-sition of the flora which are influenced by age, diet, cultural conditions, and the useof antibiotics. Any estimations on the number of bacterial species are mere guesses.It can be undoubtedly stated that the microorganisms described so far, represent onlya small fraction of the species making up the natural microbial community. In ruminants, the microbial community of the rumen consists of 1010 bacteria/ml, 106 protozoa/ml and 103 to 107 fungi/ml (Hespell et al., 1997). A wealth of data on

6 Molecular approaches in the study of gut microecology

A. Schwiertz and M. Blaut

German Institute of Human Nutrition, Gastrointestinal Microbiology,Arthur-Scheunert-Allee 114-116, D-14558 Bergholz-Rehbrücke, Germany

119



the identity and metabolic potential of the bacteria have been accumulated, but it hasbecome clear that a considerable proportion of bacteria has eluded cultivation anddescription. This is due to the fact that, until recently, the study of the gut diversitywas restricted to the use of classical microbiological techniques, such as selectiveenrichments, pure culture isolation, and most-probable-number estimates. However,many bacteria have eluded cultivation because their specific growth requirementsare not known and media are often not truly selective or specific. Hence, it has beendifficult to obtain a realistic view of the microbial community in the gastrointestinaltract. It has been estimated that only 15−58% of the human intestinal bacteria havebeen cultured or detected yet (Langendijk et al., 1995; Wilson and Blitchington,1996; Suau et al., 1999). The classical culture methods may lead to over- or under-estimations of bacterial species and groups (Langendijk et al., 1995; Doré et al., 1998).Therefore, it is not surprising that the lack of exact classification schemes and biasesintroduced by culture-based techniques have resulted in an inaccurate description andunderstanding of the microbial community of the GI-tract. The classification and enumeration of the genus Eubacterium may serve as an example. In the human intes-tinal tract, Eubacterium is the second most common genus (Finegold et al., 1974,1983). Since the identification of Eubacterium species based on phenotypic traitsrequires experience and is time-consuming, many studies involving the analysis ofhuman faecal flora composition have refrained from looking at this genus. Recentstudies indicate that the numerical importance of several Eubacterium species hasbeen overestimated (Schwiertz et al., 2000).

The detection of organisms or groups has to be based on targets which allow theirunequivocal identification. The morphological and physiological characteristics ofprokaryotes are simpler than those of eukaryotes. Therefore, an identification basedon phenotypic features is relatively difficult, time-consuming and requires experi-enced personnel. To overcome these difficulties, the information contained in themolecular sequences of their DNA, RNAs and proteins is increasingly used to inferthe relationships of microorganisms. Phylogenetic investigations targeting phyloge-netic markers such as large subunit rRNA, elongation factors, and ATPases haveshown that 16S rRNA-based trees reflect the history of the corresponding organismsglobally (Woese, 1987; Gutell et al., 1994; Amann et al., 1995).

Molecular sequence analysis, particularly of rRNA, reflects the phylogeneticinterrelationships of microorganisms (Woese, 1987). It is possible to identifymicrobes based solely on their ribosomal RNAs. Taxonomists have become inde-pendent of culturing for identifying a microbial species. The numerous techniquesemployed in the description of gut microbial diversity are depicted in fig. 1.

2. MOLECULAR TECHNIQUES

Nucleic acid-based approaches for the detection and characterization of microbialpopulations and their function within the GI-tract allow the microbiologist to deduce

120 A. Schwiertz and M. Blaut

evolutionary linkages between the involved populations. Various culture-independentmethods have been developed to identify microorganisms in samples from the GI-tract without prior cultivation. These include direct sequence analysis, sequencing ofextracted 5S, 16S and 23S rRNAs, and analysis of rRNA using reverse transcriptaseor cloned rRNA genes obtained by amplification using the polymerase chain reaction(PCR). Further techniques such as denaturing gradient gel electrophoresis (DGGE),temperature gradient gel electrophoresis (TGGE), dot-blot or slot-blot hybridizationand whole cell-in situ-hybridization, better known as fluorescence-in situ-hybridization(FISH), have been applied to analyse the complex microbial flora of the gut.

We begin with a brief description of the molecular-based techniques used inmicrobiology. Emphasis will be on the RNA-based methods, simply because RNA(especially rRNA) is the most commonly employed target nucleic acid in environ-mental microbiology.

2.1. RNA versus DNA

Over the past decade, important advances in molecular biology have led to the devel-opment of culture-independent approaches in describing bacterial communities.These new strategies, based on the analysis of DNA or RNA directly extracted fromenvironmental samples, circumvent the steps of isolation and culturing of bacteria. It is important to distinguish between identification, quantification, and monitoringfor function and activity.

Molecular approaches in gut microecology 121

Fig. 1. Strategies for the characterization of microbial communities. Arrows show the interconnections ofmethods, material and information in the study of microbial ecosystems. PCR, polymerase chain reaction; RT-PCR, reverse transcriptase PCR; FISH, fluorescence-in situ-hybridization; STARFISH, substrate-trackingaudioradiographic FISH; DGGE, denaturing gradient gel electrophoresis; RFLP, restriction fragment lengthpolymorphism; RAPD, randomly amplified polymorphic DNA. See text for further discussion.

Fundamental to the use of the various techniques is the ability to define suitablenucleic-acid sequences that identify a particular microorganism or gene. For thispurpose various oligonucleotide probes have been developed and used successfully.In general, target nucleic acid sequences fall into the following categories:1. DNA sequences that code for proteinaceous toxins2. DNA sequences that code for antigens3. Unique plasmid-borne DNA sequences4. Intergenic spacer regions (ISR)5. Ribosomal RNA (rRNA) sequences6. Messenger RNA (mRNA) sequences

All these strategies employ the unique physical properties of DNA and RNA toreassociate or hybridize. Hybridization is a term used to describe the specific comple-mentary association due to hydrogen bonding, under experimental conditions, of single-stranded nucleic acids. It should more exactly be referred to as “annealing”, asthis is the physical process responsible for the association: two complementarysequences will form hydrogen bonds between their complementary bases (G to C, andA to T or U) and form a stable double-stranded, anti-parallel “hybrid” helical molecule.The various hybridization techniques (DNA–DNA, DNA–RNA) are similar in so far asthey are simple, fast, inexpensive and universally applicable. Essential to most is thedetermination of an optimal hybridization temperature. For a theoretical background onthe development and application of nucleic acid probes see Stahl and Amann (1991).

DNA-based technologies are aimed at the specific detection of genes. However,the copy number of a given gene on the bacterial genome is usually low. In contrast,the copy number of the various RNAs, in particular of rRNA, is considerably higher.Depending on the type of RNA (mRNA, rRNA) it can vary from 1000 to more than50 000 copies in any living cell. Moreover, RNA levels may be an indirect reflec-tion of the cell’s metabolic activity.

2.2. Isolation of nucleic acids

Most molecular techniques require the prior isolation of nucleic acids from a faecalsample. Depending on the method to be used, e.g. hybridization, cloning, or amplifica-tion, the purity of the nucleic acid is crucial. Several techniques require high-qualityRNA or DNA in high yield, free of the respective other nucleic acid. Since the amountof DNA or RNA isolated from a given bacterial cell may be very small, many investi-gators concentrate on combinations of direct lysis methods with subsequent PCRamplification (Wang et al., 1996; Hengstler et al., 1998). However, extraction of RNAfrom faeces requires special attention as RNAs are highly susceptible to degradation by RNases during the extraction procedures.

Several methods are now in use for the extraction from faeces of either RNA(Stahl et al., 1988; Doré et al., 1998) or DNA (Marmur, 1961; Wang et al., 1996;Hengstler et al., 1998). All of the methods have in common that the cell wall has tobe disrupted to allow the extraction of the nucleic acids. The cell wall is disrupted by


mechanic, enzymatic or chemical methods followed by methods involving extractionwith phenol and chloroform and/or further preparations using caesium chloride gradients or spin columns to isolate the nucleic acids. A critical step in the procedureis the purification of nucleic acids in such a way that they can be used for furtheranalysis. This is not a trivial task since the humic compounds and complex polysac-charides (Monteiro et al., 1997) may inhibit enzymes involved in further analyticalsteps. Especially humic substances have been shown to interfere with enzymaticdigestion of DNA, PCR amplification of DNA (Steffan et al., 1988; Rochelle et al.,1992; Porteous and Armstrong, 1991; Tebbe and Vahjen, 1993) and dot-blothybridization of DNA (Tijssen, 1993). In a recent study Alm and co-workers showedthat humic substances may affect RNA hybridization (Alm et al., 2000).

Considering these effects, it becomes clear why recent developments in nucleicacid extraction have focused on strategies to remove humic compounds (Tsai andOlson, 1992; Herrick et al., 1993; Moran et al., 1993).

2.3. DNA-based detection

The application of DNA-based techniques for the detection and identification ofmicroorganisms is well established. Whole genomic DNA of an organism is employedin the pulse-field gel electrophoresis which extends the size range of resolution of DNAmolecules to many megabases. Treatment of plasmid-DNA or fragments of genomicDNA with restriction enzymes may yield DNA molecules larger than 25 kb which arepoorly resolved by standard agarose gel electrophoresis. A method resolving largerDNA molecules is the pulse-field gel electrophoresis (PFGE) developed by Schwartzet al. (1982). In this method, the DNA molecules are applied to an agarose gel in whichthe direction of the electric field keeps changing constantly. While the molecules fol-low the electric field, they become trapped in the gel matrix every time the direction ofthe electric field is altered. They cannot make any further progress through the gel untilthey have reorientated themselves along the new axis of the electric field. The larger the DNA molecules, the longer the time that is required for the reorientation. DNA molecules whose reorientation times are less than the period of the electric pulse, willtherefore be separated according to size (fig. 2). The protocols for PFGE are now routinely used in many laboratories working with DNA. In gut microbial ecology, however, this technique has rarely been used. Nevertheless, a recent study described thedistribution of Salmonella in swine herds using PFGE (Letellier et al., 1999).

In gut microbial ecology, various hybridization and PCR techniques have founda wide acceptance. These techniques will be discussed in more detail.

2.3.1. DNA hybridization

The first attempts to differentiate biochemically indistinguishable bacterial speciesof the GI-tract by using cloned genomic fragments were performed by Kuritza andSalyers (1985). Fragments of genomic DNA isolated from faecal samples were



Fig. 2. Scheme depicting various molecular methods for the differentiation of microorganisms. A: PFGE ofλ-DNA cut with HindIII. B: RAPD profiles were obtained with genomic DNA of Eubacterium rectale (1),Bacteroides fragilis (2), Escherichia coli (3) and various strains of Eubacterium ramulus using the M13-coreas random primer (Simmering et al., 1999). C: RFLP/ARDRA profiles of amplified 16S rDNA fromEubacterium dolichum (lanes 2, 4, and 6) and Fusobacterium mortiferum (lanes 1, 3, and 5) digested withEcoRI, BamHI and XmnI (Schneider et al., 1999). M depicts the used marker lanes.

bound to filter supports and subsequently hybridized with an oligonucleotide probespecific for Bacteroides vulgatus, to detect this organism. Differences in the appli-cation of the DNA to the filter gave rise to the following designations: dot blotting,slot blotting and touch blotting.

In dot blotting, DNA in solution is applied to the filter by applying small volumes of this solution to small circular wells of a manifold. Slot blotting is identical except that the wells are elongated to form a slot. In touch blotting, DNAor whole organisms are applied manually to the filter. In all these methods it isessential to denature the DNA prior to hybridization. Denaturation can be doneeither before or after blotting to the filter. The detection of the target is done byhybridization of a probe, which is complementary to the target sequence. Theseprobes are single strands of nucleic acid with the potential of carrying detectablemarker molecules (32P, DIG, Biotin). Radioactively labelled probes are often preferred because of their sensitivity, but non-radioactive systems (DIG, Biotin)have the advantage of being more convenient in field or clinical assays (Yamamotoet al., 1992; Kaneko and Kurihara, 1997; Miyamoto and Itoh, 1999). Their longershelf life compared to 32P-labelled probes is another advantage. Furthermore, non-radioactive probes eliminate radioactive hazards.

2.3.2. PCR-based methods

With the advent of the PCR, it became possible to amplify DNA molecules fromvery low quantities. The ability to analyse PCR amplification products is a prereq-uisite for rapid data acquisition on genomic organization and regulation. Essentialto all this is the nucleotide sequencing of the PCR products to confirm the specificityof the amplicon, identify genetic variations (e.g., polymorphisms), identifyunknown genes and map these genes within the bacterial genome. Thus, the PCRhas opened a huge variety of new methodologies for the study of environmentalsamples including the GI-tract. The rapid amplification of DNA with PCR has alsoaccelerated the sequencing of several bacterial genomes (http://www.ebi.ac.uk/genomes/). Most of the hitherto completely sequenced bacterial genomes are frompathogens. Of the bacteria for which the complete genome sequence is available,only Escherichia coli is relevant for the GI-tract. Additional sequencing of bacterialgenomes of the more abundant species of the GI-tract, will help to redefine our viewof this ecosystem and help in fast and accurate descriptions.

One of the first PCR applications to faecal DNA, aimed to identify enterotoxi-genic Escherichia coli in clinical specimens (Olive, 1989). Since then the PCR technology has been used repeatedly for the detection of bacteria in faecal samples(Kreader, 1995; Wang et al., 1996; Hengstler et al., 1998).

2.3.3. DNA fingerprinting

DNA-fingerprinting methods have been introduced for the characterization of bacterial species or communities (Kimura et al., 1997; Bateup et al., 1998;McBurney et al., 1999). The DNA-fingerprinting methods are based on the amplifi-cation of one or several stretches of genomic DNA. The first such method was arbi-trarily primed PCR (AP-PCR), better known as randomly amplified polymorphicDNA (RAPD; Welsh and McClelland, 1990). With this method it is possible to cre-ate a genomic fingerprint of a given species. Since random primers are used, it is notknown to which sites of the genome they bind. The generated amplification prod-ucts often show size polymorphisms even within species. RAPD/AP-PCR analysisoffers the possibility of creating polymorphisms without any prior knowledge of the DNA sequences of the organism investigated. The patterns produced are highlypolymorphic, allowing even discrimination between isolates of a species, if suffi-cient numbers of primers are used. The method is fast and economic for screeninglarge numbers of samples. Strain-specific arrays of DNA fragments (fingerprints)are generated by PCR amplification using arbitrary oligonucleotides to prime DNAsynthesis from genomic sites which they fortuitously match or almost match (fig. 2).

DNA amplified in this manner can be used to determine the relatedness ofspecies (Fanedl et al., 1998; Simmering et al., 1999) or for analysis of restrictionfragment length polymorphisms (RFLPs). An RFLP may be the result of lengthmutation, and/or point mutation at a restriction enzyme cleavage site at a given


chromosomal location. RFLPs can be detected by analysing restriction digests ofgenomic DNA through Southern hybridization. The probes used in RFLP analysiscan be generated from cloned genomic DNA, cDNA, or from specific DNA segments amplified by PCR. Depending on the probe used, RFLPs can be used toanalyse variations in the ribosomal rDNA region or in repetitive and single-copysequences. DNA hybridization-based RFLP analysis requires the isolation of largeamounts of purified DNA. With PCR it becomes possible to analyse specificsequences from small amounts of cells. The advantages of PCR-RFLP lie in itsspeed, sensitivity and specificity. PCR can be performed on crude DNA extractswith a pair of region-specific primers. Variation of the amplified fragment can befurther analysed by restriction enzyme digestion and electrophoretic separation.This technique is widely accepted and used in the characterization of GI-tract iso-lates (McIntosh et al., 1999; Ohkuma et al., 1999; Barcenilla et al., 2000). There areseveral variations of this technique but not all have been applied to the analysis ofGI-tract organisms. The regions most commonly examined by PCR-RFLP are therDNA sequences. This method is termed amplified ribosomal DNA restrictionanalysis (ARDRA) and has been used among others for the identification of lacto-bacilli and Enterobacteriaceae (respectively, Giraffa et al., 1998; Di Giovanni et al.,1999). The main limitation of this method lies in the choice of restriction enzymes,which is crucial for obtaining a good resolution. An example of ARDRA/RFLP isgiven in fig. 2.

T-RFLP or “terminal-RFLP” and length heterogeneity PCR (LH-PCR) havebeen introduced recently (Suzuki et al., 1998). Bernhard and Field (2000) used T-RFLP and LH-PCR to analyse faecal samples from cows and humans with respectto the Bacteroides-Prevotella group and the genus Bifidobacterium. They reportedthat host-specific patterns suggested a species composition difference in theanalysed bacterial populations. The patterns were highly reproducible within cowsand humans, respectively. Both methods recognize differences in the length of genefragments owing to insertions and deletions and the relative abundance of each fragment. In order to track a bacterial species or group, it is essential to identify a sequence (marker sequence) which is common to the bacterial species or group.Such a marker sequence, e.g. 16S rDNA, can then be amplified by specific primersfor PCR and cut by appropriate restriction enzymes which will give a unique patternof the amplified marker sequence. Once a reliable identification pattern of thismarker sequence has been obtained, the sequence can be monitored easily andquickly with fluorochrome-labelled primers. In addition, a database of identificationpatterns can be generated for a given restriction enzyme, thus making it possible toidentify the bands by a profile to profile comparison.

Amplified fragment length polymorphism (AFLP) analysis (Vos et al., 1995) isa modification of RFLP. RFLPs can be converted to AFLPs by ligating fluorescentlylabelled adapters to the primers used for PCR amplification. AFLP has the potentialto detect large numbers of amplification products although AFLP, just like RFLP,


does not target specific areas of the genome. AFLP has been successfully employedin the detection of Chlamydia psittaci strains (Boumedine and Rodolakis, 1998),Salmonella enterica subsp. enterica (Lindstedt et al., 2000a), and Clostridium perfringens (McLauchlin et al., 2000). By combining the detection of fluorescentbands with a laser detection system, it is possible to obtain much more accurate andfaster results (Lindstedt et al., 2000b).

Ribosomal intergenic spacer analysis (RISA) uses the internal transcribed spacers (ITS), which are non-coding regions of DNA sequence separating genescoding for the ribosomal RNAs (Jensen et al., 1993). These ribosomal RNA (rRNA)genes are highly conserved across taxa while the spacers in between them can be subspecies-specific (Barry et al., 1991; Narayanan et al., 2001). The conservation ofthe rRNA genes allows for easy access to the ITS regions with “versatile” primers forPCR amplification. The variation in the spacers has proven useful for distinguishingamong a wide range of microorganisms. Using the ITS regions, Brookman et al.(2000) examined the relationships within and between two genera of monocentric gutfungi gathered from various geographical locations and host animals. Tannock et al.(1999) demonstrated the usefulness of this technique for the identification of intestinalLactobacillus spp.

Fluorescent primers can be used in the amplification reactions to result in fluorescent amplification products that, when separated by electrophoresis, yieldhighly discriminative profiles. However, all these techniques have limited resolutionin identifying a specific phylogenetic group within a complex microbial community,since they do not take advantage of the sequence information but only of restrictionsites. The drawback of spacer polymorphism analysis is that more than one PCRproduct can result from a single organism because of different spacers. For exam-ple, there are two kinds of spacers in Escherichia coli. The E. coli genome is knownto contain seven loci coding for ribosomal RNA (Kiss et al., 1977). In four of them,the spacer region contains a single tRNAGlu gene. The other three loci have twotRNA genes in this spacer region: tRNAIle and tRNAAla.

Another method that takes advantage of PCR amplification for identification ofbacteria, targets repetitive extragenic palindromic sequences (REP-PCR). REP-PCRgenomic fingerprinting makes use of DNA primers complementary to naturallyoccurring, highly conserved, repetitive DNA sequences, present in multiple copiesin the genomes of most Gram-negative and several Gram-positive bacteria (Lupskiand Weinstock, 1992).

The PCR-based method of single-strand-conformation polymorphism (SSCP)analysis which has not yet been employed for the description of gastrointestinalcommunities, has been first used to generate genetic profiles of a freshwater community (Lee et al., 1996). The method is based on the fact that under non-denaturing conditions, single-stranded DNA has a folded structure which resultsfrom intramolecular interactions and its nucleotide sequence. Prior to the separationby electrophoresis, the amplified DNA is denaturated by heat and subsequently


separated by electrophoresis. The electrophoretic mobility of the single-strandedDNA in a matrix is dependent on its length, molecular weight, and shape (Yap andMcGee, 1994). Therefore, in SSCP analysis, DNA fragments with the same size but different sequences can be distinguished by electrophoresis, because of the different mobilities due to their structure (Hayashi, 1991).

The fact that the described DNA-fingerprint techniques require prior cultivationof the bacteria to be studied, is a major drawback of the methods described in thissection. They may therefore be regarded as less suitable for use in general popula-tion descriptions.

2.3.4. DGGE/TGGE

In recent years denaturing gradient gel electrophoresis (DGGE) and temperaturegradient gel electrophoresis (TGGE) have gained increasing popularity amongmicrobiologists as molecular tools to compare the diversity of microbial communi-ties and to monitor population dynamics (fig. 3). Both techniques take advantage ofthe fact that the electrophoretic mobility of fragments increases, as part of the DNAdouble helix unravels. This allows in principle the separation of DNA on the basisof a difference in one single nucleotide (Sheffield et al., 1989). In brief, thesequences of interest are amplified with PCR using an appropriate pair of primers.One of these has a so-called “G + C-clamp” attached to the 5′ end. This “G + C-clamp” prevents the two DNA strands from dissociating completely even underhighly denaturing conditions. Strand separation can be induced by an increase intemperature (TGGE) or in the concentration of chemical denaturants such as for-mamide or urea (DGGE). The vertical orientation of the denaturing temperaturefacilitates the simultaneous screening of many samples. The obtained fragments can


Fig. 3. Principle of denaturing gradient gel electrophoresis (DGGE). DGGE patterns of PCR productsobtained using universal primers for the detection of bacteria species on total nucleic acids isolated from faecal samples.

be recovered from the gels and used in further analyses such as sequencing. Anotherpossibility is the direct blotting of fragments to a filter followed by hybridizationexperiments. However, some practical disadvantages have to be mentioned. In orderto generate the GC-clamp, which is indispensable for the stability of transitionalmolecules, a relatively long primer must be used, and this may cause artefacts in theannealing step of the PCR. In addition, the results of TGGE/DGGE may be affectedby the heteroduplex molecules formed during PCR. Those mismatched base pairsare less stable under TGGE/DGGE conditions and can lead to unspecific bands(Ruano and Kidd, 1992).

Several studies employed DGGE and TGGE, respectively, for the description of gut diversity (Millar et al., 1996; Simpson et al., 1999; McCracken et al., 2001;Walter et al., 2001). DGGE analysis performed on pigs’ faeces by Simpson et al.(1999) revealed changes in bacterial populations with age, and differences betweenindividual animals and gut compartments. Furthermore, the comparison of amplifiedfaecal DNA from pigs of different age revealed several unique bands indicating thepresence of unique bacterial populations. Comparison of different gut compartmentsdemonstrated that bacterial populations within a single compartment showed thehighest similarity followed by adjacent compartments. For a review of TGGE/DGGEapplication in microbial ecology, see Muyzer and Smalla (1998) and Muyzer (1999).

The majority of the above mentioned DNA-based approaches suffer from the dis-advantage of not providing any phylogenetic information on the examined species.

2.4. RNA-based detections

RNA-based detection has become increasingly important in microbial ecology.Although technically similar, the RNA approach has major advantages over the DNAapproach. The content of RNA molecules in any living cell is usually more than 1000(Amann et al., 1995; Langendijk et al., 1995) thus making detection of an RNA target much easier because no amplification procedure is necessary. Moreover, ribo-somal RNA sequences, especially 16S rRNA, have been deposited in databases for alarge fraction of bacterial species (Van de Peer et al., 1996; Maidak et al., 2001).

The first analysis of microbial diversity in an ecosystem based on RNA techniques,was done by Stahl and co-workers (1985). Since the information content of the 5SrRNA examined in those studies is rather limited, Olsen et al. (1986) suggested anapproach based on the larger rRNA molecules. The 16S rRNA of a bacterium has anaverage length of 1500 nucleotides, and the 23S rRNA of about 3000 nucleotides. TherRNA molecules comprise highly conserved sequence domains interspersed withmore variable regions (Gutell et al., 1994; Van de Peer et al., 1996). The latter, alsocalled signature-sequence motifs, may be used for bacterial identification (Woese,1987; Amman et al., 1995).

The principles of the method are based on sequence comparison of 16S rRNAs.Since Woese and co-workers introduced the concept of comparative small subunit


rRNA sequence analysis for the elucidation of bacterial phylogeny (Woese, 1987),this technique has found worldwide acceptance and application. In the meantime anever increasing data set of ribosomal RNA sequences is available. The phylogeneticanalysis of these data provides the basis for ongoing research and reconstruction ofbacterial systematics. Initially, discovering bacterial evolution had been the majorgoal. Meanwhile, however, the applied aspects have gained more and more impor-tance. Besides comparative sequencing of rRNA genes, specific rRNA targetedprobes and diagnostic PCR in combination with a variety of experimental tech-niques are at the centre of interest.

Ribosomal RNA sequences are generally obtained either directly from rRNA or from the encoding genes located at various positions in the bacterial genome, i.e. rDNA. In practice, sequences of 16S rRNAs are generated by amplification of bacterial 16S rRNA genes (16S rDNA) using universal primers and PCR. PCR products are then integrated into vectors, and rDNA clone libraries are created. Withthese techniques Suau et al. (1999) obtained 284 clones from human faecal DNAand classified them into 82 molecular species. Three phylogenetic groups contained95% of the clones: the Bacteroides group, the Clostridium coccoides group, and theClostridium leptum subgroup. The remaining clones were distributed among a variety of phylogenetic clusters. Suau et al. (1999) reported that only 24% of therecovered molecular species corresponded to described organisms. All of these wereestablished members of the dominant human faecal flora (e.g., Bacteroides theta-iotaomicron, Fusobacterium prausnitzii, and Eubacterium rectale). However, themajority of generated rDNA sequences (76%) did not correspond to known organ-isms but to hitherto unknown species within human gut microflora. It can thereforebe stated that owing to the limitations of culture-based techniques, our knowledgeof the gut microflora composition is far from complete. In particular, it is believedthat a significant portion of the gut microbial diversity has not been cultivated yet.This view is corroborated by several studies (Snel et al., 1995; Lawson et al., 1998;Barcenilla et al., 2000).

A number of computer programs for the analysis of sequences and phylogeneticrelationships are available at various internet sites. All databases provide ribosomalRNA-related data services, including online data analysis, rRNA derived phylo-genetic trees, and aligned and annotated rRNA sequences. The most up-to-datesequence datasets are currently available at:

Genbank (http://www.ncbi.nlm.nih.gov)EMBL (http://www.ebi.ac.uk)the Antwerp database (http://www.psb.ugent.be/rRNA/index.html)the Ribosomal Database project (RDP; http://rdp.cme.msu.edu/html)ARB (Strunk et al., 1998; http://www.mikro.biologie.tu-muenchen.de/)In the last two, there are currently more than 20 000 aligned sequence entries,

thus making them the largest databases on bacterial phylogeny. The user is able to integrate new sequences into already existing databases and to subsequently


generate trees. For further information on the theory of bacterial phylogeny, basedon comparative sequence analysis, see Ludwig et al. (1998).

Tools for the design of diagnostic oligonucleotide probes, integrated in these programs, make it possible to construct a variety of cluster, group- or species-specificoligonucleotide probes for the detection of microorganisms. Hitherto unculturedspecies can be detected and even be visualized. Thus, the designed probes enable thescientist to monitor the spatial arrangements and interactions of unknown species intheir natural environment. Using oligonucleotide probes ranging from the group orgenus level down to species-specific probes (nested approach) on different culturemedia, those species can even be isolated and subsequently be tested for their biochemical properties. Considerable effort has been made in the past few years togenerate new probes for the detection of gut microorganisms (Langendijk et al.,1995; Kaneko and Kurihara, 1997; Doré et al., 1998; Simmering et al., 1999).Furthermore, the project FAIR-CT97–3035, financed by the European Union, wassolely devoted to the development and application of molecular approaches assess-ing the human gut flora in diet and health.

2.4.1. Hybridizations targeting ribosomal RNA

Both quantitative dot-blot hybridization and whole-cell hybridization have beenused for the analysis of intestinal bacteria. For quantitative dot-blot hybridization,total RNA is extracted from the sample and bound to a filter support. The RNA issubsequently labelled with oligonucleotide probes. Universal probes hybridize totargets in the rRNA that have been conserved during evolution and therefore recog-nize all bacteria. In contrast, specific oligonucleotides are designed in such a waythat they recognize, depending on the specificity, bacteria at various levels of thephylogenetic hierarchy. The relative abundance of a bacterial population group isthen calculated by dividing the amount of a specific probe bound to a given sampleby the amount of hybridized universal probe referred to as rRNA index. With thistechnique Sghir et al. (2000) were able to quantify several bacterial groups withinthe human faecal flora. By applying six probes to faecal samples from human adults,70% of the total 16S rRNA could be accounted for by the Bacteroides (37%), theClostridium leptum subgroup (16%) and the Clostridium coccoides group (14%).Bifidobacterium and Lactobacillus groups made up less than 2% and the entericbacteria accounted for less than 1%. This study indicated that cultural-based estimations of the main bacterial groups may lead to overestimations and under-estimations, respectively. The ribosomal RNA-targeted hybridization probes appliedin this study were also applied in studying the gut microbial ecology of pigs (Boyeet al., 1998; Sghir et al., 1998), cattle (Forster et al., 1997; Kocan et al., 1998) andsheep (Forster et al., 1997).

The analysis at the single-cell level provides a more detailed picture than dot-blothybridization because not only can the cell morphology of bacteria be determined,


but also their spatial distribution in situ. The first applications of in situ nucleic acidhybridization used isotopically labelled probes that bind to the rRNAs of fixed andintact cells. Organisms recognized by the probes were identified by audioradiography(Giovannoni et al., 1988). However, the major drawback of microaudioradiography isthe requirement for a relatively long exposure time and the low resolution. Theintroduction of fluorescently labelled probes was an important step forward to amuch faster and more accurate detection of target cells. Furthermore, the use of several oligonucleotide probes, each labelled with a different fluorescent dye, allowsthe simultaneous detection of several different organisms. In recent years, wholecell-in situ-hybridization, better known as fluorescence-in situ-hybridization (FISH)has become one of the most widely used tools in microbial gut ecology (Franks et al., 1998; Simmering et al., 1999; Harmsen et al., 2000). The major advantage ofwhole-cell hybridization for the detection of bacteria in contrast to other molecularmethods (see above), lies in their microscopic visualization. Quantification of positively hybridized cells in faecal preparations is performed by means of visualcounting procedures. It is a time-consuming process which depends highly on the skills and experience of the person performing the counting. Therefore, only moderate levels of accuracy are reached (Langendijk et al., 1995). In order to over-come this problem, automated microscopic counting has been introduced (Jansen et al., 1999). This methodology is particularly useful when large numbers of faecalsamples need to be processed. This is the case in studies that investigate the influ-ence of diet on the faecal flora. The principle of FISH is depicted in fig. 4.

It has to be emphasized, however, that the application of whole-cell hybridiza-tion to faecal samples also has its limitations, the most significant one being the lack


Fig. 4. Principle of fluorescence-in situ-hybridization (FISH).

of sensitivity. This may be partly due to the fact that the number of rRNA targets is lower in cells in their natural environment than in cells growing in pure culturesunder optimal conditions. Nutritional limitation and other competitive factors influence the cellular ribosome content (Amann et al., 1995; Langendijk et al., 1995).It is therefore not surprising that the fluorescence signal of cells in faecal samples islower than in pure cultures (Langendijk et al., 1995). In addition, unspecific fluo-rescence may hamper the visualization of bacteria. Another limitation is the possibleinaccessibility of the target sequence, which may be due to the structure of the ribo-some, which includes rRNA–rRNA and rRNA–ribosomal protein interactions(Binder and Liu, 1998; Clemons et al., 1999). In a recent paper, Fuchs et al. (2000)described the use of unlabelled helper oligonucleotides to improve the in situ acces-sibility to the 16S rRNA of E. coli. The in situ identification of individual cells mayalso be hindered by a limited cell wall permeability (Binder and Liu, 1998; Fegatellaet al., 1998). Some researchers employed lysozyme to improve the permeability ofthe cell wall (Simmering et al., 1999; Schwiertz et al., 2000). However, the intensityof the signal depends on the penetration of the probes across the cell periphery. In this respect, carbocyanine dye-labelled (Cy3) probes are superior to biotinylatedprobes (Alfreider et al., 1996; Simmering et al., 1999; Schwiertz et al., 2000). It hasbeen reported that treatment with chloramphenicol, an inhibitor of protein synthesisand RNA degradation, can lead to an increase in the percentage of detectable cells(Ouverney and Fuhrman, 1999).

Several technical developments have aimed to increase the sensitivity of FISH: oneof these is the Tyramide System Amplification (TSA®; NEN Research Products). Itcombines horseradish peroxidase (HRP)-labelled, rRNA-targeted oligonucleotideprobes and the TSA® system. The TSA® amplification is based on the covalent bindingof radicalized fluorochrome-tyramide substrate molecules to electron-rich moieties,such as tyrosines or tryptophans. HRP-containing cells give a very bright fluorescentsignal. Schonhuber et al. (1997) observed an increase of fluorescence with the TSA®

system. Although the TSA® yields bright fluorescent signals, the disadvantages shouldbe noted. Owing to the relatively large molecular size of the HRP-oligonucleotideprobe-complex, a penetration of the fixed bacterial cells is rather difficult. This is moreeasily achieved with the smaller fluorescently labelled oligonucleotides.

Developments of new molecular approaches in the study of bacterial ecosystemson the basis of RNA detection are numerous. Recently a new technique for the analy-sis of aquatic samples has been introduced by Ouverney and Fuhrman (1999). Thistechnique can be applied easily to gastrointestinal ecology. This technique is termedsubstrate-tracking autoradiographic fluorescent-in situ-hybridization (STARFISH)which affords the detection of cells that take up a radioactively labelled substance,and its distribution. When combined with 16S rRNA-targeted in situ hybridization,bacteria metabolizing the labelled substrate can be identified. Further development ofthis technique will help to get a better look at the gastrointestinal ecology and tostudy for example the influence of diet on the microbial community.


2.4.2. Other RNA-based detection methods

Numerous techniques have been developed to monitor gene expression in bacterialcells. These include coupled reverse transcription PCR amplification (RT-PCR) andribonuclease protection assays (RPA). Of these methods, RT-PCR is the most sensitiveand versatile. It can be used to determine the presence or absence of a transcript, estimate expression levels and clone cDNA products without the necessity of constructing and screening a cDNA library (Noda et al., 1999). In contrast, RPA is ahighly sensitive and specific method for the detection and quantification of specificmRNAs. The assay was made possible by the discovery and characterization of DNA-dependent RNA polymerases from the bacteriophages SP6, T7 and T3, and the elucidation of their cognate promoter sequences. These polymerases are ideal for theselective and specific synthesis of RNA probes from DNA templates, because thesepolymerases exhibit a high degree of fidelity to their promoters, polymerize RNA at avery high rate, efficiently transcribe long segments, and do not require high concentra-tions of rNTPs. Thus, a cDNA fragment of interest can be subcloned into a plasmid thatcontains bacteriophage promoters, and the construct can then be used as a template forthe synthesis of radiolabelled anti-sense RNA probes (fig. 5).


Fig. 5. Principle of theribosomal protectionassay (RPA).


Advances in molecular technology have led to improvements in the methods avail-able for studies of the gut microbial ecology. One of the most recent developments,which may have a future impact on microbial ecology, is the so-called molecularbeacons. Molecular beacons are oligonucleotide probes that report the presence ofspecific nucleic acids in homogeneous solutions (Tyagi and Kramer, 1996). They areuseful in situations where it is either not possible or not desirable to isolate theprobe-target hybrids from an excess of the hybridization probes, such as in real-timemonitoring of PCRs in sealed tubes or in detection of RNAs within living cells.Molecular beacons are hairpin-shaped molecules with an internally quenched fluo-rophore, whose fluorescence is restored when they bind to a target nucleic acid (fig. 6).They are designed in such a way that the loop portion of the molecule is a probesequence complementary to a target nucleic acid molecule. The stem is formed bythe annealing of the complementary arm sequences located at the ends of theoligonucleotide. A fluorescent moiety is attached to the end of one arm and aquenching moiety is attached to the end of the other arm. The stem keeps these twomoieties in close proximity to each other, causing the fluorescence of the fluorophore to be quenched by energy transfer. Since the quencher moiety is a non-fluorescent chromophore and emits the energy that it receives from the fluorophoreas heat, the probe does not fluoresce. When the probe encounters a target molecule,it forms a hybrid that is longer and more stable than the stem and its rigidity andlength preclude the formation of the stem hybrid. Thus, the molecular beacon under-goes a spontaneous conformational reorganization that forces the stem apart, andcauses the fluorophore and the quencher to move away from each other, leading tothe restoration of fluorescence, which can be detected. In order to detect multipletargets in the same solution, molecular beacons can be made in many differentcolours utilizing a broad range of fluorophores (Tyagi et al., 1998). DABCYL, a


Fig. 6. Operation of molecular beacons.

non-fluorescent chromophore, serves as the universal quencher for any fluorophorein molecular beacons. Owing to their loop sequence, the recognition of targets bymolecular beacons is so specific that single-nucleotide differences can be readilydetected. In combination with the biochip technology, molecular beacons mightbecome a powerful tool in molecular ecology.

Biochips enable DNA or RNA sequences to be quickly analysed. They allowrapid and precise information on the genetic composition of a given sample. Theprinciple of analysing material with a biochip is simple: to the substrate material,genetic material with a known structure is applied in periodic patterns over an areaabout the size of a thumbnail. These prepared measuring points contain for examplespecific 16S rRNA oligonucleotide probes for the detection of different bacterialgenera, groups and species. Treated faecal samples in an aqueous solution are thenapplied onto the chip and complementary 16S rRNA sequences are allowed tohybridize. A positive hybridization result can then be proven with the aid of markerspreviously bound to the specific probe. The attached molecules of dye are irradiatedwith laser light, emitting a characteristic wavelength. A camera detects this light andan analysis software shows where the marked points are located, and arranges them.

For a fast and more reliable detection of bacterial groups, genera or species fromenvironmental samples, specific molecular beacons can be developed and put onto achip. Owing to the nature of the molecular beacons, a high specificity and a rather fastexecution of samples is permitted. The application of these concepts can bring similarbenefits of speed and reduced costs to research in gastrointestinal microbiology.

In conclusion, the opportunities for the discovery of new organisms and thedevelopment of techniques based on microbial diversity are greater than ever before.Utilizing all those tools in conjunction with the study of mechanisms and interac-tions between the microorganism and the eukaryotic cell, will lead to a better insightinto the microbiology of the gastrointestinal tract of humans and animals.

REFERENCES

Alfreider, A., Pernthaler, J., Amann, R., Sattler, B., Glöckner, F.O., Wille, A., Psenner, R., 1996.Community analysis of the bacterial assemblages in the winter cover and pelagic layers of a highmountain lake by in situ hybridization. Appl. Environ. Microbiol. 62, 2138−2144.

Alm, E.W., Zheng, D., Raskin, L., 2000. The presence of humic substances and DNA in RNA extractsaffects hybridization results. Appl. Environ. Microbiol. 66, 4547−4554.

Amann, R.I., Ludwig, W., Schleifer, K.H., 1995. Phylogenetic identification and in situ detection of indi-vidual microbial cells without cultivation. Microbiol. Rev. 59, 143−169.

Barcenilla, A., Pryde, S.E., Martin, J.C., Duncan, S.H., Stewart, C.S., Henderson, C., Flint, H.J., 2000.Phylogenetic relationships of butyrate-producing bacteria from the human gut. Appl. Environ.Microbiol. 66, 1654−1661.

Barry, T., Colleran, G., Glennon, M., Dunican, L.K., Gannon, F., 1991. The 16S/23S ribosomal spacerregion as a target for DNA probes identify eubacteria. PCR Methods. Applicat. 1, 51−56.

Bateup, J., Dobbinson, S., Munro, K., McConnell, M.A., Tannock, G.W., 1998. Molecular analysis of thecomposition of Lactobacillus populations inhabiting the stomach and caecum of pigs. Microb. Ecol.Health. Dis. 10, 95−102.


Bernhard, A.E., Field, K.G., 2000. A PCR assay to discriminate human and ruminant feces on the basisof host differences in bacteroides-prevotella genes encoding 16S rRNA. Appl. Environ. Microbiol. 66,4571−4574.

Binder, B.J., Liu, Y.C., 1998. Growth rate regulation of rRNA content of a marine Synechococcus(Cyanobacterium) strain. Appl. Environ. Microbiol. 64, 3346−3351.

Boumedine, K.S., Rodolakis, A., 1998. AFLP allows the identification of genomic markers of ruminantChlamydia psittaci strains useful for typing and epidemiological studies. Res. Microbiol. 149, 735−744.

Boye, M., Jensen, T.K., Moller, K., Leser, T.D., Jorsal, S.E., 1998. Specific detection of the genusSerpulina, S. hyodysenteriae and S. pilosicoliin porcine intestines by fluorescent rRNA in situhybridization. Mol. Cell Probes. 12, 323−330.

Brookman, J.L., Mennim, G., Trinci, A.P., Theodorou, M.K., Tuckwell, D.S., 2000. Identification andcharacterization of anaerobic gut fungi using molecular methodologies based on ribosomal ITS1 and185 rRNA. Microbiology 146, 393−403.

Clemons, W.M., Jr, May, J.L., Wimberly, B.T., McCutcheon, J.P., Capel, M.S., Ramakrishnan, V., 1999.Structure of a bacterial 30S ribosomal subunit at 5.5 Å resolution. Nature 400, 833−840.

Di Giovanni, G.D., Watrud, L.S., Seidler, R.J., Widmer, F., 1999. Fingerprinting of mixed bacterial strainsand BIOLOG gram-negative (GN) substrate communities by enterobacterial repetitive intergenic con-sensus sequence-PCR (ERIC-PCR). Curr. Microbiol. 38, 217−223.

Doré, J., Sghir, A., Hannequart-Garmet, G., Corthier, G., Pochart, P., 1998. Design and evaluation of a16S rRNA-targeted oligonucleotide probe for specific detection and quantification of human faecalbacteroides populations. System. Appl. Microbiol. 21, 65−71.

Fanedl, L., Nekrep, F.V., Avgustin, G., 1998. Random amplified polymorphic DNA analysis and demon-stration of genetic variability among bifidobacteria isolated from rats fed with raw kidney beans. Can.J. Microbiol. 44, 1094−1101.

Fegatella, F., Lim, J., Kjelleberg, S., Cavicchioli, R., 1998. Implications of rRNA operon copy numberand ribosome content in the marine oligotrophic ultramicrobacterium Sphingomonas sp. strainRB2256. Appl. Environ. Microbiol. 64, 4433−4438.

Finegold, S.M., Attebery, H.R., Sutter, V.L., 1974. Effect of diet on human fecal flora: comparison ofJapanese and American diets. Amer. J. Clin. Nutr. 27, 1456−1469.

Finegold, S.M., Sutter, V.L., Mathisen, G.E., 1983. Normal indigenous intestinal flora. In: Hentges, D.J.(Ed.), Human Intestinal Microflora in Health and Disease. Academic Press, New York, pp. 3−31.

Forster, R.J., Gong, J., Teather, R.M., 1997. Group-specific 16S rRNA hybridization probes for determi-native and community structure studies of Butyrivibrio fibrisolvens in the rumen. Appl. Environ.Microbiol. 63, 1256−1260.

Franks, A.H., Harmsen, H.J.M., Raangs, G.C., Jansen, G.J., Schut, F., Welling, G.W., 1998. Variations of bacterial populations in human feces measured by fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 64, 3336−3345.

Fuchs, B.M., Glockner, F.O., Wulf, J., Amann, R.I., 2000. Unlabeled helper oligonucleotides increase thein situ accessibility to 16S rRNA of fluorescently labeled oligonucleotide probes. Appl. Environ.Microbiol. 66, 3603−3607.

Giovannoni, S.J., DeLong, E.F., Olsen, G.J., Pace, N.R., 1988. Phylogenetic group-specific oligonu-cleotide probes for identification of single microbial cells. J. Bacteriol. 170, 720−726.

Giraffa, G., De Vecchi, P., Rossetti, L., 1998. Note: identification of Lactobacillus delbrueckii subspeciesbulgaricus and subspecies lactis dairy isolates by amplified rDNA restriction analysis. J. Appl.Microbiol. 85, 918−924.

Gutell, R.R., Larsen, N., Woese, C.R., 1994. Lessons from an evolving rRNA. 16S and 23S rRNA struc-tures from a comparative perspective. Microbiol. Rev. 58, 10−26.

Harmsen, H.J., Wildeboer-Veloo, A.C., Grijpstra, J., Knol, J., Degener, J.E., Welling, G.W., 2000.Development of 16S rRNA-based probes for the Coriobacterium group and the Atopobium clusterand their application for enumeration of Coriobacteriaceae in human feces from volunteers of differ-ent age groups. Appl. Environ. Microbiol. 66, 4523−4527.

Hayashi, K., 1991. PCR-SSCP: a simple and sensitive method for detection of mutations in the genomicDNA. PCR Methods Applicat. 1, 34−38.

Hengstler, M., Kronberg, U., Fahr, A.M., 1998. A reliable method for the removal of PCR inhibitors dur-ing the preparation of bacterial DNA from stool specimens. Clin. Lab. 44, 447−450.


Herrick, J.B., Madsen, E.L., Batt, C.A., Ghiorse, W.C., 1993. Polymerase chain reaction amplification ofnaphthalene-catabolic and 16S rRNA gene sequences from indigenous sediment bacteria. Appl.Environ. Microbiol. 59, 687−694.

Hespell, R.B., Akin, D.E., Dehority, B.A., 1997. Bacteria, fungi, and protozoa of the rumen. In: Machie, R.I.,White, B.A., Isaacson, R.E. (Eds.), Gastrointestinal Microbiology. Vol. 2: Gastrointestinal Microbesand Host Interaction. Chapman & Hall, New York, USA, pp. 59−141.

Jansen, G.J., Wildeboer-Veloo, A.C., Tonk, R.H., Franks, A.H., Welling, G.W., 1999. Development andvalidation of an automated, microscopy-based method for enumeration of groups of intestinal bacte-ria. J. Microbiol. Meth. 37, 215−221.

Jensen, M.A., Webster, J.A., Strauss, N., 1993. Rapid identification of bacteria on the basis of polymerasechain reaction-amplified ribosomal DNA spacer polymorphisms. Appl. Environ. Microbiol. 59, 945−952.

Kaneko, T., Kurihara, H., 1997. Digoxigenin-labeled deoxyribonucleic acid probes for the enumerationof bifidobacteria in fecal samples. J. Dairy Sci. 80, 1254−1259.

Kimura, K., McCartney, A.L., McConnell, M.A., Tannock, G.W., 1997. Analysis of fecal populations ofbifidobacteria and lactobacilli and investigation of immunological responses of their human hosts tothe predominant strains. Appl. Environ. Microbiol. 63, 3394−3398.

Kiss, A., Sain, B., Venetianer, P., 1977. The number of rRNA genes in Escherichia coli. FEBS Lett. 79,77−79.

Kocan, K.M., Ge, N.L., Blouin, E.F., Murphy, G.L., 1998. Development of a non-radioactive DNA probeand in situ hybridization for detection of Anaplasma marginale in ticks and cattle. Ann. NY Acad. Sci.(Review) 849, 137−145.

Kreader, C.A., 1995. Design and evaluation of bacteroides DNA probes for the specific detection ofhuman fecal pollution. Appl. Environ. Microbiol. 61, 1171−1179.

Kuritza, A.P., Salyers, A.A., 1985. Use of a species-specific DNA hybridization probe for enumeratingBacteroides vulgatus in human feces. Appl. Environ. Microbiol. 50, 958−964.

Langendijk, P.S., Schut, F., Jansen, G.J., Raangs, G.C., Kamphuis, G.R., Wilkinson, M.H.F., Welling, G.W.,1995. Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus-specific 16S rRNA-targeted probes and its application in fecal samples. Appl. Environ. Microbiol.6, 3069−3075.

Lawson, A.J., Linton, D., Stanley, J., 1998. 16S rRNA gene sequences of “Candidatus Campylobacterhominis”, a novel uncultivated species, are found in the gastrointestinal tract of healthy humans.Microbiology 144, 2063−2071.

Lee, D.H., Zo, Y.G., Kim, S.J., 1996. Nonradioactive method to study genetic profiles of natural bacter-ial communities by PCR-single-strand-conformation polymorphism. Appl. Environ. Microbiol. 62,3112−3120.

Letellier, A., Messier, S., Pare, J., Menard, J., Quessy, S., 1999. Distribution of Salmonella in swine herdsin Quebec. Vet. Microbiol. 67, 299−306.

Lindstedt, B.A., Heir, E., Vardund, T., Kapperud, G.A., 2000a. A variation of the amplified-fragmentlength polymorphism (AFLP) technique using three restriction endonucleases, and assessment of theenzyme combination BglII-MfeI for AFLP analysis of Salmonella enterica subsp. enterica isolates.FEMS Microbiol. Lett. 189, 19−24.

Lindstedt, B.A., Heir, E., Vardund, T., Kapperud, G., 2000b. Fluorescent amplified-fragment length poly-morphism genotyping of Salmonella enterica subsp. enterica serovars and comparison with pulsed-field gel electrophoresis typing. J. Clin. Microbiol. 38, 1623−1627.

Ludwig, W., Strunk, O., Klugbauer, S., Klugbauer, N., Weizenegger, M., Neumaier, J., Bachleitner, M.,Schleifer, K.H., 1998. Bacterial phylogeny based on comparative sequence analysis. Electrophoresis19, 554−568.

Lupski, J.R., Weinstock, G.M., 1992. Short, interspersed repetitive DNA sequences in prokaryoticgenomes. J. Bacteriol. (Review) 174, 4525−4529.

Maidak, B.L., Cole, J.R., Lilburn, T.G., Parker, C.T., Jr, Saxman, P.R., Farris, R.J., Garrity, G.M., Olsen, G.J.,Schmidt, T.M., Tiedje, J.M., 2001. The RDP-II (Ribosomal Database Project). Nucleic Acids Res. 29,173−174.

Marmur, J., 1961. A procedure for the isolation of deoxyribonucleic acids from microorganisms. J. Mol.Biol. 3, 208−218.


McBurney, W.T., McCartney, A., Apun, A., McConnell, M., Tannock, G.W., 1999. Perturbation of theenterobacterial microflora detected by molecular analysis. Microb. Ecol. Health. Dis. 11, 75−79.

McCracken, V.J., Simpson, J.M., Mackie, R.I., Gaskins, H.R., 2001. Molecular ecological analysis ofdietary and antibiotic-induced alterations of the mouse intestinal microbiota. J. Nutr. 131, 1862−1870.

McIntosh, G.H., Royle, P.J., Playne, M.J., 1999. A probiotic strain of L. acidophilus reduces DMH-induced large intestinal tumors in male Sprague-Dawley rats. Nutr. Cancer. 35, 153−159.

McLauchlin, J., Ripabelli, G., Brett, M.M., Threlfall, E.J., 2000. Amplified fragment length polymor-phism (AFLP) analysis of Clostridium perfringens for epidemiological typing. Int. J. Food Microbiol.56, 21−28.

Millar, M.R., Linton, C.J., Cade, A., Glancy, D., Hall, M., Jalal, H., 1996. Application of 16S rRNA genePCR to study bowel flora of preterm infants with and without necrotizing enterocolitis. J. Clin.Microbiol. 34, 2506−2510.

Miyamoto, Y., Itoh, K., 1999. Design of cluster-specific 16S rDNA oligonucleotide probes to identifybacteria of the Bacteroides subgroup harbored in human feces. FEMS Microbiol. Lett. 177, 143−149.

Monteiro, L., Bonnemaison, D., Vekris, A., Petry, K.G., Bonnet, J., Vidal, R., Cabrita, J., Mégraud, F.,1997. Complex polysaccharides as PCR inhibitors in feces: Helicobacter pylori model. J. Clin.Microbiol. 35, 995−998.

Moran, M., Torsvik, V.L., Torsvik, T., Hodsen, R.E., 1993. Direct extraction and purification of rRNA forecological studies. Appl. Environ. Microbiol. 59, 915−918.

Muyzer, G., 1999. DGGE/TGGE a method for identifying genes from natural ecosystems. Curr. Opin.Microbiol. (Review) 2, 317−322.

Muyzer, G., Smalla, K., 1998. Application of denaturing gradient gel electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology. Antonie Van Leeuwenhoek(Review) 73, 127−141.

Narayanan, S.K., Nagaraja, T.G., Chengappa, M.M., Stewart, G.C., 2001. Electrophoretic mobilityanomalies associated with PCR amplification of the intergenic spacer region between 16S and 23Sribosomal RNA genes of Fusobacterium necrophorum. J. Microbiol. Meth. 46, 165−169.

Noda, S., Ohkuma, M., Usami, R., Horikoshi, K., Kudo, T., 1999. Culture-independent characterizationof a gene responsible for nitrogen fixation in the symbiotic microbial community in the gut of the termite Neotermes koshunensis. Appl. Environ. Microbiol. 65, 4935−4942.

Ohkuma, M., Noda, S., Kudo, T., 1999. Phylogenetic diversity of nitrogen fixation genes in the symbi-otic microbial community in the gut of diverse termites. Appl. Environ. Microbiol. 65, 4926−4934.

Olive, D.M., 1989. Detection of enterotoxigenic Escherichia coli after polymerase chain reaction ampli-fication with a thermostable DNA polymerase. J. Clin. Microbiol. 27, 261−265.

Olsen, G.J., Lane, D.J., Giovannoni, S.J., Pace, N.R., Stahl, D.A., 1986. Microbial ecology and evolution:a ribosomal RNA approach. Annu. Rev. Microbiol. 40, 337−365.

Ouverney, C.C., Fuhrman, J.A., 1999. Combined microautoradiography-16S rRNA probe technique fordetermination of radioisotope uptake by specific microbial cell types in situ. Appl. Environ.Microbiol. 65, 1746−1752.

Porteous, L.A., Armstrong, J.L., 1991. Recovery of bulk DNA from soil by rapid, small-scale extractionmethod. Curr. Microbiol. 22, 345−348.

Rochelle, P.A., Weightman, A.J., Fry, J.C., 1992. DNase I treatment of Taq DNA polymerase for com-plete PCR decontamination. Biotechniques 13, 520.

Ruano, G., Kidd, K.K., 1992. Modeling of heteroduplex formation during PCR from mixtures of DNAtemplates. PCR Methods Applicat. 2, 112−116.

Savage, D.C., 1977. Microbial ecology of the gastrointestinal tract. Annu. Rev. Microbiol. 31, 107−133.Schneider, H., Schwiertz, A., Collins, M.D., Blaut, M., 1999. Anaerobic transformation of quercetin-3-

glucoside by bacteria from the human intestinal tract. Arch. Microbiol. 171, 81−91.Schonhuber, W., Fuchs, B., Juretschko, S., Amann, R., 1997. Improved sensitivity of whole-cell

hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramidesignal amplification. Appl. Environ. Microbiol. 63, 3268−3273.

Schwartz, D.C., Saffran, W., Welsh, J., Haas, R., Goldberg, M., Cantor, C., 1982. New techniques forpurifying large DNAs and studying their properties and packaging. Cold Spring Harbor Symp. Quant.Biol. 47, 189−195.


Schwiertz, A., Le Blay, G., Blaut, M., 2000. Quantification of different Eubacterium spp. in human fecalsamples with species-specific 16S rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol.66, 375−382.

Sghir, A., Antonopoulos, D., Mackie, R.I., 1998. Design and evaluation of a Lactobacillus group-specificribosomal RNA-targeted hybridization probe and its application to the study of intestinal microecol-ogy in pigs. Syst. Appl. Microbiol. 21, 291−296.

Sghir, A., Gramet, G., Suau, A., Rochet, V., Pochart, P., Dore, J., 2000. Quantification of bacterial groupswithin human fecal flora by oligonucleotide probe hybridization. Appl. Environ. Microbiol. 66,2263–2266.

Sheffield, V.C., Cox, D.R., Lerman, L.S., Myers, R.M., 1989. Attachment of a 40-base-pair G + C-richsequence (GC-clamp) to genomic DNA fragments by the polymerase chain reaction results inimproved detection of single-base changes. Proc. Natl. Acad. Sci. USA 86, 232−236.

Simmering, R., Kleessen, B., Blaut, M., 1999. Quantification of the flavonoid-degrading bacteriumEubacterium ramulus in human fecal samples with a species-specific oligonucleotide hybridizationprobe. Appl. Environ. Microbiol. 65, 3705−3709.

Simpson, J.M., McCracken, V.J., White, B.A., Gaskins, H.R., Mackie, R.L., 1999. Application of denat-urant gradient gel electrophoresis for the analysis of the porcine gastrointestinal microbiota. J. Microbiol. Meth. 36, 167−179.

Snel, J., Heinen, P.P., Blok, H.J., Carman, R.J., Duncan, A.J., Allen, P.C., Collins, M.D., 1995.Comparison of 16S rRNA sequences of segmented filamentous bacteria isolated from mice, rats, andchickens and proposal of “Candidatus Arthromitus”. Int. J. Syst. Bacteriol. 45, 780−782.

Stahl, D.A., Amann, R., 1991. Development and application of nucleic acid probes. In: Stackebrandt, E.,Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics. John Wiley & Sons,Chichester, pp. 205−248.

Stahl, D.A., Flesher, B., Mansfield, H.R., Montgomery, L., 1988. Use of phylogenetically based hybridiza-tion probes for studies of ruminal microbial ecology. Appl. Environ. Microbiol. 54, 1079−1084.

Stahl, D.A., Lane, D.J., Olden, G.J., Pace, N.R., 1985. Characterisation of a Yellowstone hot springmicrobial community by 5S rRNA sequence. Appl. Environ. Microbiol. 49, 1379−1384.

Steffan, R.J., Goksoyr, J., Bej, A.K., Atlas, R.M., 1988. Recovery of DNA from soils and sediments.Appl. Environ. Microbiol. 54, 2908−2915.

Strunk, O., Gross, O., Reichel, B., May, M., Hermann, S., Stuckmann, N., Nonhoff, B., Lenke, M.,Ginhart, A., Vilbig, A., Ludwig, T., Bode, A., Schleifer, K.H., Ludwig, W., 1998. ARB: A SoftwareEnvironment for Sequence Data. (Online. http://www.mikro.biologie.tu-muenchen.de.) Department ofMicrobiology, Technische Universität München, Munich, Germany.

Suau, A., Bonnet, R., Sutren, M., Godon, J.J., Gibson, G.R., Collins, M.D., Doré, J., 1999. Direct analy-sis of genes encoding 16S rRNA from complex communities reveals many novel molecular specieswithin the human gut. Appl. Environ. Microbiol. 65, 4799−4807.

Suzuki, M., Rappe, M.S., Giovannoni, S.J., 1998. Kinetic bias in estimates of coastal picoplankton community structure obtained by measurements of small-subunit rRNA gene PCR amplicon lengthheterogeneity. Appl. Environ. Microbiol. 64, 4522−4529.

Tannock, G.W., Tilsala-Timisjarvi, A., Rodtong, S., Ng, J., Munro, K., Alatossava, T., 1999. Identificationof Lactobacillus isolates from the gastrointestinal tract, silage, and yoghurt by 16S-23S rRNA geneintergenic spacer region sequence comparisons. Appl. Environ. Microbiol. 65, 4264−4267.

Tebbe, C.C., Vahjen, W., 1993. Interference of humic acids and DNA extracted directly from soil indetection and transformation of recombinant DNA from bacteria and a yeast. Appl. Environ.Microbiol. 59, 2657−2665.

Tijssen, P., 1993. Hybridization with Nucleic Acid Probes, Part I: Theory and Nucleic Acid Precipitation.Elsevier, Amsterdam.

Tsai, Y.L., Olson, B.H., 1992. Rapid method for separation of bacterial DNA from humic substances insediments for polymerase chain reaction. Appl. Environ. Microbiol. 58, 2292−2295.

Tyagi, S., Kramer, F.R., 1996. Molecular beacons: probes that fluoresce upon hybridization. Nat.Biotechnol. 14, 303−308.

Tyagi, S., Bratu, D., Kramer, F.R., 1998. Multicolor molecular beacons for allele discrimination. Nat.Biotechnol. 16, 49−53.


Van de Peer, Y., Chapelle, S., De Wachter, R., 1996. A quantitative map of nucleotide substitution ratesin bacterial rRNA. Nucl. Acids. Res. 24, 3381−3391.

Vos, P., Hogers, R., Bleeker, M., Reijans, M., Vandelee, T., Hornes, M., Frijters, A., Pot, J., Peleman, J.,Kuiper, M., Zabeau, M., 1995. AFLP: a new technique of DNA fingerprinting. Nucl. Acids Res. 23,4407−4414.

Walter, J., Hertel, C., Tannock, G.W., Lis, C.M., Munro, K., Hammes, W.P., 2001. Detection ofLactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-specific PCR primers and denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 67,2578−2585.

Wang, R.F., Cao, W.W., Cerniglia, C.E., 1996. PCR detection and quantitation of predominant anaerobicbacteria in human and animal fecal samples. Appl. Environ. Microbiol. 62, 1242−1247.

Welsh, J., McClelland, M., 1990. Fingerprinting genomes using PCR with arbitrary primers. NucleicAcids Res. 18, 7213−7218.

Wilson, K.H., Blitchington, R., 1996. Human colonic biota studied by ribosomal DNA sequence analysis.Appl. Environ. Microbiol. 62, 2273−2278.

Woese, C.R., 1987. Bacterial evolution. Microbiol. Rev. 51, 221−271.Yamamoto, T., Morotomi, M., Tanaka, R., 1992. Species-specific oligonucleotide probes for five

Bifidobacterium species detected in human intestinal microflora. Appl. Environ. Microbiol. 58,4076−4079.

Yap, E.P.H., McGee, J.O., 1994. Non-isotopic single-strand conformation polymorphism (SCCP) analy-sis of PCR products. In: Griffin, H.G., Griffin, A.M. (Eds.), PCR Technology: Current Innovations.CRC Press, Boca Raton, Florida, pp. 165−177.


142

In vitro models can be a useful tool to study microbial interactions. However, it isimportant to include sufficient parameters to obtain the desired predictability, whilekeeping the model as simple as possible. When using an in vitro model of the gas-trointestinal tract (GIT), it is necessary to have information on the relevant para-meters in vivo, such as GIT morphology, residence times, secretion rates andcomposition of digestive juices, pH profiles and microflora composition. Theseparameters may vary between species and between the growing and adult animal.Different types of models are used to study the intestinal microflora. Batch modelsare the simplest systems where microflora and substrate are incubated in a vessel fora certain period. Continuous models are systems where the culture is regularly fedwith medium, while the pH is kept at a set point. Several of these vessels can be con-nected to mimic different parts of the GIT. Dynamic systems include the interactionsbetween gastric emptying, gastric pH profiles, rates of secretion, water absorption,removal of digestive products and microbial metabolites, and transit of the mealthrough a separate part of the gut. The use of in vitro systems is limited by theabsence of interactions between microflora and the host, the availability of relevantin vivo data, and analytical methods to analyse the microflora composition and itsmetabolites.

1. INTRODUCTION

The feed, petfood and pharmaceutical industries need to develop and improve products continuously in order to meet the increasing demands of the market. The feed industry aims at knowledge about the efficacy of feed ingredients, improv-ing the bio-availability of nutrients, introducing alternative protein, carbohydrate

7 Models of the gastrointestinal tract tostudy microbial interactions

M. Minekus

TNO Nutrition and Food Research, P.O. Box 360, 3700 AJ, Zeist,The Netherlands


Models of GIT for microbial interactions 143

and fat sources, and reducing the faecal output of environmental pollutants such asphosphate. The petfood industry is following the trend in human nutrition to offerfunctional foods with additional health-promoting characteristics. New feed addi-tives and pharmaceutical products are developed to improve and maintain the healthstatus of the animals. These developments require studies to test the behaviour ofcompounds in the gastrointestinal tract (GIT) in relation to their efficacy, digestibil-ity, fermentation and their effect on the microflora. These studies are often per-formed in laboratory animals. However, animal studies are hampered by costs,ethical constraints, sampling problems and variation between individual animals.

As an alternative to animal studies, experiments are performed on in vitro models.These models can be used to perform simplified experiments under uniform andwell-controlled conditions. However, simulating such a complex system as the GITcarries the risk of oversimplification. The predictive value of a model generallyincreases when more parameters are included. However, the inclusion of moreparameters leads to more complex model systems. Thus, a model system shouldtake into account all relevant parameters, while being as simple as possible.Therefore, each type of in vitro model has its specific use and limitations.

2. THE DIGESTIVE SYSTEM: VARIATION BETWEEN SPECIES AND AGE

Although the digestive system of vertebrates has a general layout (table 1), eachspecies has adapted its GIT to its specific eating behaviour and nutrients, resultingin variation of anatomy and physiology (Chivers and Langer, 1994).

The diversity in GIT morphology is expressed by different shapes, relative sizesand functions of the digestive compartments, leading to different residence times forthe digesta in the successive parts of the GIT. Even compartments with specializedfunctions have been developed such as the crop and gizzard in birds and theforestomachs in ruminants. This anatomical and physiological variation has resultedin intestinal microbial populations that are specific for each species. Besides varia-tion between species, also age related differences exist.

Table 1. General layout of the digestive tract

Compartment Function

Oral cavity Ingestion and pre-treatment of the mealStomach (crop in birds) Storage, particle reduction, peptic digestion,

acidification or fermentationSmall intestine Digestion and absorption of nutrients and waterLarge intestine (cloaca in birds, some fish,

amphibians and reptiles) Fermentation of undigested materials, absorption of water, fermentation products and electrolytes

Young animals may differ from adult animals with respect to concentrations ofsecreted digestive compounds such as bile and enzymes. In addition, some younganimals, such as the preruminant calf, have a different GIT morphology in comparison to the adult animal.

To study microbial interactions in the GIT with an in vitro model, one shouldtake into account the relevant conditions prevailing in the digestive system of theanimal of interest. These conditions might be different between species, but alsobetween the young and the adult animal. Important gastrointestinal aspects and theireffect on digestive processes are shown in table 2.

3. MODELS TO STUDY MICROBIAL INTERACTIONS

The behaviour of intestinal microflora is predominantly studied in models simu-lating the rumen of ruminants (Davies, 1979; Sudweeks et al., 1979; Dong et al.,1997), and the large intestine of monogastric animals (Rumney and Rowland, 1992;Minekus et al., 1999). Only a few models take account of the microecology of thestomach and the small intestine (Coutts et al., 1987; Molly et al., 1993).

Three different types of model systems have been described to perform experi-ments with complex microbial populations, viz.: a) batch culture systems, b) (semi-)continuous culture systems, and c) dynamic systems.

3.1. Batch cultures

Batch cultures are mainly used for simple, short-term experiments (1 to 2 days) andonly involve incubation of test material with faeces or colonic contents (Vince et al.,1990; Barry et al., 1995; Van Hoeij et al., 1997). The experiments are generally performed in closed vessels under anaerobic conditions, while the pH is maintainedat a preset level with a buffer or by a pH-stat. Mixing of the content is absent or donewith a stirrer (Batch, fig. 1). Their predictability is limited by the fact that themicroflora is not continuously fed and that there is no removal of metabolites during incubation. Accumulation of metabolites might eventually influence themetabolic activity of the microflora.

144 M. Minekus

Table 2. Gastrointestinal aspects and their effect on digestive processes

Aspects Effect

Morphology Transit of the meal, specific compartmentsMotilitySecretion and concentration of Digestion and uptake of nutrients

digestive compoundspHGut wall propertiesMicroflora composition Fermentation and bioconversion of compounds

3.2. (Semi-)continuous cultures

Semi-continuous cultures need a more sophisticated set-up. The microflora is fedwhile part of the content is removed intermittently from the fermentor to mimic thepassage of chyme through the simulated part of the GIT. Continuous cultures havethe same set-up as the semi-continuous cultures, except that they have a regularfeeding of the microflora and removal of contents. Mixing is achieved by an impel-lor, thus resulting in a continuously stirred tank reactor (CSTR, fig. 1). Incubationin a CSTR results in a steady-state situation where the growth rate of the micro-organisms is determined by the dilution rate. To simulate consecutive parts of thegut, different systems have been designed with two, three or five vessels in series(fig. 2) (Miller and Wolin, 1981; Manning et al., 1987; Gibson et al., 1988; Allisonet al., 1989; Molly et al., 1993). These systems generally allow for growth of strictlyanaerobic microorganisms by flushing with anaerobic gas. The pH is measuredand controlled within the physiological range through addition of acid or alkali.Both continuous and semi-continuous cultures have been used to study the micro-ecology of the flora, degradation of undigested materials, enzyme activities andproduction of interesting metabolites such as short chain fatty acids, gases andtoxic compounds.

A drawback of (semi)-continuous culture systems is that they operate understeady-state conditions which means that the concentrations of metabolites are keptwithin the physiological range by a limited amount of substrates in the influent andthe dilution rate (Edwards and Rowland, 1992). Substrate limitation, dilution andproduct inhibition limit the amounts of microorganisms in these systems. Realisticfeeding, and removal of the metabolites and water without the microorganisms, are


Fig. 1. Schematic presentation of differentreactor types used to simulate gastrointestinaltransit.

prerequisites for maintaining the amounts of microorganisms as well as theirmetabolites at physiological levels. Feeding and mixing of dense fibrous and vis-cous materials is a common problem in large intestinal systems (Edwards andRowland, 1992).

3.3. Dynamic GIT systems

3.3.1. Dynamic conditions in the GIT

The model systems described so far are static models that do not simulate the dynamic conditions to which a compound or a microorganism is exposed whentravelling through (transiting microflora) or colonizing (resident microflora) thestomach and intestines. The importance of including dynamic interactions of para-meters is demonstrated when studying the survival of microorganisms during transitthrough the GIT. Gastric pH and bile concentration are the main determinants of survival (Marteau et al., 1997). However, both gastric pH and bile concentration arenot constant in time. Gastric pH increases during ingestion of a meal and thendecreases, depending on the rate of gastric acid secretion and the buffer capacity ofthe meal. The duodenal bile concentration shows an increasing and later decreasingpattern due to emptying of the bile bladder after the start of a meal. Along the smallintestine, the bile concentration first increases due to water absorption and laterdecreases due to absorption of bile in the distal part of the small intestine. Not only

146 M. Minekus

Fig. 2. Schematic presentation of the SHIME model. M, feed; P, pancreatic enzymes; I, stomach; II, duodenum and jejunum; III, ileum; IV, caecum and ascending colon: V, transverse colon; VI, descendingcolon; E, effluent.

pH and bile, but also secretion of other digestive compounds are affected by themeal and therefore are not constant in time.

Another aspect that determines the fate of microorganisms during gastrointestinalpassage is the time that they are exposed to unfavourable gastric pH values and bileconcentrations, which is determined by the pattern of gastric emptying and the smallintestinal transit time of the meal. Gastrointestinal passage is often mimicked bysequential static exposure to gastric and small intestinal conditions (fig. 3A).However, this is not a realistic situation since the meal is gradually emptied from thestomach after which it travels through the intestines (Decuypere et al., 1986; fig. 3B).

The gastric pH and the gastric emptying of milk in a preruminant calf are shown in fig. 4, as an example of the dynamic interaction between pH and gastricemptying. The figure shows that, in this case, more than 50 per cent of the meal hasleft the stomach while the pH is above 4.


Fig. 3. Exposure of a meal to successive steps of digestion in a static system (A) and in a dynamic system (B).

Fig. 4. pH (solid line) and gastric emptying (dotted line) during the digestion of milk in a preruminant calf(adapted from Caugant et al., 1992).

The considerations described above, have led to the development of a multi-compartmental dynamic computer controlled model (nicknamed TIM) that simu-lates the dynamically changing conditions in the different parts of the digestive tract (Minekus et al., 1995, 1999). The complete system consists of a gastric–smallintestinal system (TIM-1) (fig. 5), usually used for digestive studies and a largeintestinal system (TIM-2) (fig. 6), used for microbiological studies. The two systemsare – for practical reasons – constructed and used as separate systems.

3.3.2. The gastro–small intestinal system

The gastro–small intestinal model consists of four successive compartments (fig. 5),simulating the stomach (a), duodenum (b), jejunum (c) and ileum (d). A meal canbe fed to the gastric compartment during a preset time. In the gastric compartmentgastric juice is added (e), while the pH is measured (f) and adjusted to follow a

148 M. Minekus

Fig. 5. Schematic diagram of the multi-compartmental model of the stomach and small intestine: a, gastric compartment; b, duodenal compartment; c, jejunal compart-ment; d, ileal compartment; e, gastric secretionpumps; f, pH electrode; g, peristaltic valvepump; h, duodenal secretion pumps; i, dialysisfluid; j, hollow-fibre device; k, collecting vessel for ileal delivery.

Fig. 6. Schematic representation of TIM-2. a, Peristaltic compartments; b, pH-electrode; c, alkali pump; d, dialysis circuit with hollowfibres; e, level sensor; f, “ileal effluent” container; g, peristaltic valve pump for influent;h, peristaltic value pump for effluent; i, sampling-port; j, N2 gas inlet; k, gas collection bag.

predetermined curve. The compartments consist of connected glass units with flexible walls inside. The walls can be squeezed by varying the pressure on thewater. The chyme in each compartment is mixed by alternately squeezing the flexible walls. To control the transit of the meal, the compartments are separatedby computer-regulated peristaltic valve pumps (g). Bile and pancreatic juice aresecreted into the duodenal compartment (h). The pH in each small intestinal compartment is measured (f) and controlled through the addition of sodium hydro-gen carbonate. Products of digestion and water are absorbed from the jejunal andileal compartments by pumping dialysis liquid (i) through hollow-fibre membraneunits with a molecular weight cut off of approximately 5000 (j). The chyme deliv-ered from the ileal compartment (ileal delivery) is collected on ice in a vessel (k).

The system is controlled by a computer that can be programmed with a protocolthat contains formalized data on the dynamic digestive conditions of a specificspecies having a specific meal. The protocol for the TIM-1 contains data on: i) thegradual gastric emptying of the meal and the transit of the meal through the smallintestine; ii) the gastric pH profile which increases first due to the buffer capacityof a meal and then decreases due to acid secretion; iii) the different pH values inthe small intestinal compartments; iv) the rate of gastric and duodenal secretions;and v) the absorption of water.

The small intestinal system is used to study the digestion and availability forabsorption of nutrients and pharmaceuticals. It has also been used to study the survival of transiting microorganisms (Marteau et al., 1993, 1997), the transfer ofgenetic material from genetically modified foods to microorganisms (van der Vossenand Havenaar, 1997; Gänzle et al., 1999), and as a pre-treatment for studies in thelarge intestinal model.

3.3.3. The large intestinal system

The large intestinal system (fig. 6) consists of connected glass units (a), each with aflexible wall inside. Peristaltic movements are achieved by changing the pressure onthe water. The computer controls the sequential squeezing of these walls, thus caus-ing a peristaltic wave which forces the chyme to circulate through the loop-shapedsystem. The tubular-shaped lumen of the system prevents “constipation”. The pH ismeasured with a pH electrode (b) and controlled by the addition of a sodiumhydroxide solution (c). The dialysis liquid is pumped through hollow-fibre mem-branes positioned in the lumen of the reactor (d), to maintain the appropriate con-centrations of electrolytes and metabolites. The amount of chyme in the reactor ismonitored with a level sensor (e) and kept on a preset level by the absorption ofwater with a pump in the dialysis circuit. The feeding medium (f) is mixed and keptanaerobic with nitrogen, and is introduced into the reactor with the peristaltic valvesystem (g) as described for the gastro-small intestinal system. A peristaltic valvepump is also used to remove chyme from the reactor (h). Samples can be taken from


a sampling port (i). The system is kept anaerobic by flushing with nitrogen (j).Produced gas can be collected in a bag (k) or is allowed to leave through a water lock.

The large intestinal model is controlled to simulate the conditions for the largeintestine with a dense microflora of human or animal origin (Minekus et al., 1999).This is achieved by combining the feeding of concentrated ileal effluent with the removal of water and metabolites through a semi-permeable membrane. Theperistaltic movements allow adequate mixing and transport of the chyme. Severalunits can be connected to mimic successive parts of the large intestine. The largeintestinal system is used to study the fermentation of carbohydrates, production oftoxic metabolites, transfer of genetic material from food or microorganisms to themicroflora and the efficacy of probiotic strains.

All the models described, including the TIM systems, have the general limitationthat it is not possible to maintain a microflora that is completely identical to the in vivosituation. The conditions that affect the microflora in vivo are much too complicated to be completely simulated in an in vitro system. The first problem is tohave an inoculum that is an exact copy of the site of interest in vivo. A faecal inocu-lum is easy to obtain, is considered representative of the large intestinal microflora andtherefore is widely used (Rumney and Rowland, 1992). A better inoculum, but moredifficult to obtain, would be the intestinal content itself, e.g. from the proximal colon.

The second problem is the composition of the substrate medium. Ideally, thestandard medium composition should allow the growth and maintenance of amicroflora that is identical to the inoculum. However, in practice some shift inmicroflora composition cannot be avoided.

The most important drawback of in vitro models is that they do not include theconsiderable interactions between microflora and the host (Umesaki et al., 1997).First, the immunological response of the host is a major determinant of the micro-flora’s composition and cannot be included in any in vitro model. Second, there isno gut wall to include gut wall mechanisms. Under normal conditions only thosemicroorganisms colonize the GIT which grow fast enough to resist the flow or thosethat are associated with the mucus layer of the gut wall. Association to the gut wallis regarded as a prerequisite for invading microorganisms to overcome their lagphase in the new environment and to colonize the intestine (Beachey, 1980; Freteret al., 1983). Recent research has revealed that indigenous microorganisms are ableto modify the specific attachment receptor and thus prevent colonization of thepathogen (Bry et al., 1997; Umesaki et al., 1997). The body can react to pathogensby increasing cell turnover to dispose of the pathogens attached to the cells.Experiments in fermentors have shown that attachment to the glass wall of the vessel is also necessary for microorganisms to overcome the colonization barrier.Although the ecological principles might be similar, the mechanisms of adherencein a fermentor are clearly different from those in vivo.

Although it is not feasible to study adherence directly in the culture vessel, someinteraction of microorganisms can be studied with cell cultures. The adherence of

150 M. Minekus

microorganisms to enterocytes can be studied using a cultured monolayer of Caco-2cells on a semi-permeable membrane (fig. 7A) (Naaber et al., 1996). The Caco-2cells are cultivated in cups (fig. 7B) and incubated with a bacterial suspension. Afterincubation, the non-adhered bacteria are washed from the cells. The adhered bacteriaare removed from the cells by sonification and enumerated on selective agar plates.

4. ANALYTICAL METHODS TO STUDY THE MICROFLORA

An advantage of in vitro models is that they allow easy sampling, which can only be exploited when adequate analytical methods are available. Apart from test compounds, analysis is generally focused on the microflora composition, enzymeactivities and the products of microbial activity.

The composition of the microflora is traditionally determined by enumeration on(s)elective culture media. In addition to traditional culturing techniques, variousmolecular techniques are possible to collect information on the composition of themicroflora of the gastrointestinal tract (Tannock, 2001). These techniques includepolymerase chain reaction (PCR), fluorescent in situ hybridization (FISH) (Langendijket al., 1995; Harmsen et al., 2000) and denaturing gradient gel electrophoresis (DGGE;Walter et al., 2001). These molecular techniques are directed towards the ribosomalRNA encoding regions of the bacterium as the scientific community recognizes theseregions as perfect markers for the taxonomic position of an organism.

By using the PCR method, individual species can be detected. The specificity ofthis detection approach depends on the specificity of the DNA synthesis primingsequences. These sequences can be selected in such a way that they recognize onlya single species. PCR is particularly useful for showing the presence or absence ofa specific organism. Presently, it is also possible to quantify the presence of thespecific organism by using, among others, so-called “real-time-PCR”.

FISH offers a good way of detection and quantification of specific bacteria forwhich in situ probes are available. These probes target the specific sequence in theribosomal RNA that is present in each ribosome of the specific microorganism.Since more than 100 copies of the ribosomal RNA are present in an active cell,


Fig. 7. Caco-2 cells on a semi-permeable membrane (A) cultivated in cups (B).

enough fluorescent material is bound to allow fluorescence under a fluorescencemicroscope. Specific bacteria can be counted in this way. The only disadvantage ofthe technique is that it can only provide information on those bacteria for whichprobes are defined.

To get a more holistic view of the microbial flora DGGE is a more suitedapproach. By using DGGE ribosomal sequences of the various species are separatedby their difference in DNA sequence. In this DGGE system there exists an increas-ing gradient of denaturing agent. When a segment of DNA enters a region wherepart of the duplex is unstable, the conformation changes and the mobility decreasesdramatically. This technique allows the visualization of the heterogeneity of micro-bial floras and as a consequence changes in the flora by a banding pattern andchanging pattern, respectively. Each band in the DGGE pattern represents an indi-vidual species. Since the technique is targeting ribosomal RNA or DNA sequences,individual bands can be further analysed by nucleotide sequence analysis in orderto get information on their species affiliation. There exists a large database withnucleotide sequence information that is extremely useful for taxonomists and phy-logeneticists. In proceeding this way, it can be analysed as to which organism isinvolved in a change of flora composition.

Although microflora composition is an interesting aspect to study microbial inter-actions, probably more relevant when studying health aspects are the compounds thatare produced by the microflora. These compounds can be products of normal fer-mentation, such as short chain fatty acids, lactic acid and gases (mainly H2, CH4 andCO2). Microbial enzyme activities such as those involving β-glucuronidase andazoreductase may lead to the production of toxic compounds (Rowland et al., 1985;Bingham et al., 1996). Analyses of specific metabolites are necessary to study muta-genicity, bioconversion, antimicrobial activity, gene stability and gene transfer.


In vitro models can be a useful tool to study the microbial interactions in the gut.However, their use is limited by the complexity of the microflora and their interac-tions with the host. More insight in the physiological conditions of the target speciesis necessary to determine the right conditions during the experiments. The amountof distinguishable species present in the gut microflora has increased rapidlythrough molecular techniques and will increase further through the use of tech-niques such as DNA chip technology. In spite of the enormous metabolic activity ofthe microflora, little is still known about the microbial metabolites and their effecton the interactions between the species and the host. Analytical techniques such asnuclear magnetic resonance (NMR) and gas or liquid chromatography coupled tomass spectrometry (LC-MC, GC-MS) can be used to detect relevant microbialproducts. With pattern recognition techniques, specific metabolites can be linked tospecific conditions, substrates, or microorganisms.

152 M. Minekus

REFERENCES

Allison, C., Macfarlane, C., Macfarlane, G.T., 1989. Studies in mixed populations of human intestinalbacteria grown in single-stage and multistage continuous culture systems. Appl. Environ. Microbiol.55, 672–678.

Barry, J.L., Hoebler, C., Macfarlane, G.T., Macfarlane, S., Mathers, J.C., Reed, K.A., Mortensen, P.B.,Nordgaard, I., Rowland, I.R., Rumney, C.J.W., 1995. Estimation of the fermentability of dietary fibrein vitro: a European interlaboratory study. Brit. J. Nutr. 74, 303–322.

Beachey, E.H., 1980. Bacterial Adherence, Chapman and Hall, London.Bingham, S.A., Pignatelli, B., Pollock, J.R.A., Ellul, A., Malaveille, C., Gross, G., Runswick, S.,

Cummings, J.H., O’Niell, I.K.O., 1996. Does increased endogenous formation of N-nitroso com-pounds in the human colon explain the association between red meat and colon cancer? Carcinogenesis17 (3), 515–523.

Bry, L., Falk, P.G., Midtvedt, T., Gordon, J.I., 1997. A model of host-microbial interactions in an openmammalian ecosystem. Science 273, 1380–1383.

Caugant, I., Petit, H.V., Carbonneau, R., Savoie, L., Toullec, R., Thirouin, S., Yvon, M., 1992. In vivoand in vitro gastric emptying of protein fractions of milk replacers containing whey proteins. J. DairySci. 75, 847–856.

Chivers, D.J., Langer, P. (Eds.), 1994. The Digestive System in Mammals. Cambridge University Press,Cambridge.

Coutts, T.M., Aldrick, A.J., Rowland, I.R., 1987. Use of continuous culture to study the gastric microfloraof a hypochlorhydric patient. Toxicity in Vitro 1, 17–21.

Davies, M.E., 1979. Studies on the microbial flora of the large intestine of the horse by continuousculture in an artificial colon. Vet. Sci. Commun. 3, 39–44.

Decuypere, J.A., Denhooven, R.M., Henderickx, H.K., 1986. Stomach emptying of milk diets in pigs. A mathematical model allowing description and comparison of the emptying pattern. Arch. Anim.Nutr. 36, 679–696.

Dong, Y., Bae, H.D., McAllister, T.A., Mathison, G.W., Cheng, K.J., 1997. Lipid-induced depressionof methane production and digestibility in the artificial rumen system (RUSITEC). Can. J. Anim.Sci. 77 (2), 269–278.

Edwards, C.A., Rowland, I.R., 1992. Bacterial fermentation in the colon and its measurement. In:Schweizer, T.F., Edwards, C.A. (Eds.), Dietary Fibre – a Component of Food. Springer Verlag, Berlin.

Freter, R., 1992. Factors affecting the microecology of the gut. In: Fuller, R. (Ed.), Probiotics. Chapmanand Hall, London, pp. 111–144.

Freter, R., Stauffer, E., Cleven, E., Holdeman, L.V., Moore, W.E.C., 1983. Continuous-flow cultures asin vitro models of the ecology of large intestinal flora. Infect. Immun. 39, 666–675.

Gänzle, M.G., Hertel, C., Van der Vossen, J.M.B.M., Hammes, W.P., 1999. Effect of bacteriocin-produc-ing lactobacilli on the survival of Escherichia coli and Listeria in a dynamic model of the stomachand the small intestine. Int. J. Food Microbiol. 48, 21–35.

Gibson, G.R., Cummings, J.H., Macfarlane, G.T., 1988. Use of three-stage continuous culture system tostudy the effect of mucin on dissimilatory sulfate reduction and methanogenesis by mixed populationsof human gut bacteria. Appl. Environ. Microbiol. 54, 2750–2755.

Harmsen, H.J.M., Raangs, G.C., Wildeboer-Veloo, A.C.M., Tonk, R.H.J., Degener, J.E., Welling, G.W.,2000. Characterisation and dynamics of the bacterial populations in human feces studied with 16SrRNA-based fluorescent probes. Reprod. Nutr. Dev. 40, 175.

Langendijk, P.S., Schut, F., Jansen, G.J., Raangs, G.C., Kamphuis, G.R., Wilkinson, M.H.F., Welling,G.W., 1995. Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genusspecific 16S rRNA-targeted probes and its application in fecal samples. Appl. Environ. Microbiol.61, 3069–3075.

Manning, B.W., Federle, T.W., Cerniglia, C.E., 1987. Use of a semicontinuous culture system as a modelfor determining the role of human intestinal microflora in the metabolism of xenobiotics. J. Microbiol. Meth. 6, 81–94.

Marteau, P., Pochart, P., Bouhnik, Y., Rambaud, J.C., 1993. Fate and effects of some transiting microor-ganisms in the human gastrointestinal tract. World Rev. Nutr. Diet 74, 1–21.


Marteau, P., Minekus, M., Havenaar, R., Huis in ’t Veld, J.H.J., 1997. Survival of lactic acid bacteria ina dynamic model of the stomach and small intestine: Validation and the effects of bile. J. Dairy Sci.80, 1031–1037.

Miller, T.L., Wolin, M.J., 1981. Fermentation by the human large intestine microbial community in an invitro semi-continuous culture system. Appl. Environ. Microbiol. 42, 400–407.

Minekus, M., Marteau, P., Havenaar, R., Huis in ’t Veld, J.H.J., 1995. A multi compartmental dynamiccomputer-controlled model simulating the stomach and small intestine. Alternativ. Laborat. Anim.(ATLA) 23, 197–209.

Minekus, M., Smeets-Peeters, M., Bernalier, A., Marol-Bonnin, S., Havenaar, R., Marteau, P., Alric, M.,Fonty, G., Huis in ’t Veld, J.H.J., 1999. A computer-controlled system to simulate conditions of thelarge intestine with peristaltic mixing, water absorption and absorption of fermentation products.Appl. Microbiol. Biotechnol. 53, 108–114.

Molly, K., Vande Woestyne, M., Verstraete, W., 1993. Development of a 5-step multi-chamber reactor asa simulation of the human intestinal microbial ecosystem. Appl. Microbiol. Biotechnol. 39, 254–258.

Naaber, P., Lehto, E., Salminen, S., Mikelsaar, M., 1996. Inhibition of adhesion of Clostridium difficileto Caco-2 cells. FEMS Immunol. Med. Microbiol. 14, 205–209.

Rowland, I., Mallett, A.K., Wise, A., 1985. The effect of diet on the mammalian gut flora and its meta-bolic activities. Crit. Rev. Toxicol. 16, 31–103.

Rumney, C.J., Rowland, I.A., 1992. In vivo and in vitro models of the human colonic flora. Crit. Rev.Food Sci. Nutr. 31, 299–331.

Sudweeks, E.M., Ely, L.O., Moon, N.J., Sisk, L.R., 1979. A continuous culture artificial rumen. J. DairySci. 62 (1), 210.

Tannock, G.W., 2001. Molecular assessment of intestinal microflora. Amer. J. Clin. Nutr. 73 (2),410S–414S.

Umesaki, Y., Okada, Y., Imaoka, A., Setoyama, H., Matsumoto, S., 1997. Interactions between epithelialcells and bacteria, normal and pathogenic. Science 276, 964–965.

Van der Vossen, J., Havenaar, R., 1997. Gene stability and gene transfer of genetically modified foods inthe TNO gastro-intestinal model. In: van der Kamp, J.W., Havenaar, R. (Eds.), Safety EvaluationMethods of Novel and Transgenic Food Crops. Proceedings of the 6th SABAF Workshop AIRConcerted Action CT 94 2342, Vienna, Austria. TNO, Zeist, The Netherlands, pp. 22–23.

Van Hoeij, K.A., Green, C.J., Pijnen, A., Speckmann, A., Bindels, J.G., 1997. A novel in vitro method toassess colonic short chain fatty acid (SCFA) and gas production of indigestible carbohydrates.Proceedings of the International Symposium “Non-digestible Oligosaccharides: Healthy Food for theColon?”, Wageningen, The Netherlands, p. 131.

Vantrappen, G., 1997. Small intestinal motility and bacteria. In: Scheidt, P.J., Rush, V., Van der Waaij, D.(Eds.), Gastro Intestinal Motility. Old Herborn University Seminar Monographs 9. Herborn Litterae,Herborn-Dill, Germany, pp. 53–67.

Vince, A.J., Mcniel, N.I., Wager, J.D., Wrong, O.M., 1990. The effect of lactulose, pectin, arabinogalac-tan and cellulose on the production of organic acids and metabolism of ammonia by intestinal bacte-ria in a faecal incubation system. Brit. J. Nutr. 63, 17–26.

Walter, J., Hertel, C., Tannock, G.W., Lis, C.M., Munro, K., Hammes, W.P., 2001. Detection ofLactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by using group-specific PCR primers and denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 67(6), 2578–2585.

154 M. Minekus

157

Enteric diseases of pigs and calves owing to Escherichia coli typically appear during the first few days (and weeks) of life. The so far recognized pathotypes of E. coli involved are the enterotoxic E. coli (ETEC), verotoxic E. coli (VTEC),enteropathogenic E. coli (EPEC), and necrotoxic E. coli (NTEC). The first step inthe pathogenesis of all these types is to adhere to the intestinal microvilli with orwithout inducing morphological lesions and produce specific toxins acting locallyon enterocytes and/or absorbed into the bloodstream. This action is assured by specific ligand (adhesins) and receptor interactions which are characteristic of the pathotypes involved and may vary according to the animal species. The host-adapted adhesin/receptor systems are targets of several preventive measuresincluding vaccines, receptor blocking and breeding for genetic resistance. They alsoprovide the basis for cross-species infections including zoonoses. Therefore theydeserve the attention of epidemiologists, research scientists and technologists.Besides, they provide tools for increased understanding of molecular pathogenesis,and offer excellent models for comparative studies of different disease entities ofenteric colibacillosis in humans.

1. INTRODUCTION

Enteric colibacillosis of pigs and calves has decreased as a devastating problem ofintensive animal farming during the past two decades but even today it represents

8 Adhesins and receptors for colonization bydifferent pathotypes of Escherichia coli incalves and young pigs1

B. Nagy, I. Tóth and P.Zs. Fekete

Veterinary Medical Research Institute of the Hungarian Academy of Sciences,1143 Budapest, Hungária krt. 21, Hungary

1For their research data in this chapter, the authors acknowledge support from the following grants: OTKA T034970(to B. Nagy), OTKA T026150 (to I. Tóth) and OTKA A312 (to the VMRI), as well as FAIR3-CT96-1335 (NTEC infarm animals).



158 B. Nagy, I. Tóth and P.Zs. Fekete

one of the major health issues in the pig or cattle industry of several developed andof most less developed countries. The reasons for the decreased losses are primarilythe new diagnostic tools and vaccines that have been developed and widely used asa result of intensive research efforts on enteric E. coli infections of animals duringthe 1970s and 1980s. The major breakthroughs in the diagnosis and prevention ofenteric colibacillosis of growing pigs and calves were mainly due to our increasedunderstanding of colonization and adhesion mechanisms of these enteric pathogensto the intestinal mucosae. In spite of these positive developments there is still roomfor research in this area partly because of the obvious human implications (severalpathotypes of E. coli are also prevalent in humans), and partly because our knowl-edge is still quite limited. This is especially true for the area of intestinal receptorsof bacterial adhesions.

The main pathotypes involved in enteric colibacillosis of pigs and calves are theenterotoxigenic E. coli (ETEC), verotoxigenic E. coli (VTEC), enteropathogenic E. coli (EPEC), and necrotoxigenic E. coli (NTEC).

This review aims to give a general overview of the virulence factors and theirgenetic regulators in E. coli. Furthermore it aims to describe the most importantadhesins and their receptors playing a role in the pathogenesis of different pathotypesof enteric E. coli. It also points out some of the areas where future research is needed.

As there is a lot of analogy in the regulation of virulence factors of the abovepathotypes and as the most abundant information is available on ETEC, we will useETEC as the “veterinarians’s horse” to describe the basic organization of virulencegenes. Therefore we will start with the description of adhesions and receptors withthe ETEC pathotype and will refer to them in the case of analogies at appropriatesections of other pathotypes. We will be able to describe the practical applicationsof present knowledge essentially also on ETEC. We have to admit that little infor-mation is available on that aspect for EPEC or NTEC.

2. ENTEROTOXIGENIC E. COLI

In enteric E. coli infections, especially in enterotoxic E. coli (ETEC) infections of different species, bacteria adhere to the small intestinal epithelial cells (overwhelm-ingly in newborn or very young animals), thereby colonizing the gut. They also secreteproteins or peptides (enterotoxins) which stimulate the small intestine for increasedwater and electrolyte secretion and/or decreased fluid absorption. The ability of adhe-sion of ETEC to intestinal epithelial cells is mainly due to the production of thin(3–7 nm) proteinaceous surface appendages (fimbriae or pili) which can be morpho-logically, biologically and antigenically different on various strains. Some of themmorphologically resemble the common fimbriae (“Type 1” fimbriae or pili) of E. coli(Duguid et al., 1955). With the help of these adhesins (fimbriae), the bacteria are ableto attach themselves to the microvilli of small intestinal epithelial cells, thereby more intensively transferring the enterotoxins to the target cells. There is no characteristic

Adhesins and receptors for E. coli 159

histological or ultrastructural morphology of adhesion or colonization by ETEC.The microvilli and the epithelial cells remain intact (fig. 1a).

2.1. Adhesins and other virulence factors in the pathogenesis of ETEC

According to our present understanding, the pathogenesis of enterotoxic colibacillosisstarts with the adhesin–ligand interaction on the small intestinal microvilli, resulting ina strong but morphologically non-destructive attachment of bacteria to the microvilli.Therefore the virulence characteristics of ETEC are strongly dependent on the produc-tion of adhesins (fimbriae) and enterotoxins. In addition to adhesive and enterotoxicvirulence factors, pathogenesis due to ETEC infection also involves host factors amongwhich the most important ones are the receptors for adhesin (and/or enterotoxin).Species specificity − which is a general characteristic of ETEC infections − is largely due to the presence of specific receptors in only one (or in a limited spectrum of) animal species. Several of these adhesive virulence factors of ETEC and some of theirreceptors are known and will be discussed in detail below, but some of them are stillunknown. Future research in this area is clearly needed and could bring further under-standing of pathogenesis, thereby it would contribute to more successful strategies inthe prevention and treatment of enteric enterotoxic colibacillosis due to ETEC.

The most common adhesive fimbriae of animal ETEC strains can be differenti-ated as surface antigens such as K88 or K99, 987P or F41 or F107 and 2134P in pigsand calves, also designated as F4, F5, F6, F41, or F18ab and F18ac, respectively(Ørskov and Ørskov, 1983; Moon, 1990; Rippinger et al., 1995) (table 1).

Fig. 1. Schematic presentation of cellular changes due to interaction of E. coli bacteria pathotypes with intestinal epithelial cells: (a) ETEC/VTEC: no obvious change in cellular microvillus morphology. (b) EPEC/EHEC: attaching to cell membrane and effacing of microvilli with pedestal formation. (c) NTEC: multinucleation, induced by CNF; distension of the cell and nucleus induced by CDT.


These fimbriae are characterized as straight, bent or kinked proteinaceousappendages originating from the outer membrane of the bacterial cells. They havevarious molecular weights (from 15 to 25 kDa). In general, fimbriae are composed of“major” and “minor” subunit structures governed and assembled under the directionof structural and accessory genes respectively. For adhesive fimbriae, the adhesivefunction is often represented by molecules at the tip of the filaments. The ability offimbriae (pili) to agglutinate red blood cells of different species was recognized veryearly (Elsinghorst and Weitz, 1994) and it has been used for classification along withthe effect of 0.5% D-mannose: MS = mannose sensitive (adhesion blocked by mannose)or MR = mannose resistant adhesion. Among fimbriae of animal ETEC bacteria wecan recognize the following categories: MS haemagglutinating fimbriae (Type 1),MR haemagglutinating fimbriae (K88, K99, F41), and MR non-haemagglutinatingfimbriae (987P, F18ab, F18ac). Adhesive fimbriae provide the necessary first step forthe enterotoxins to act efficiently.

Enterotoxins can be described as extracellular proteins or peptides (exotoxins)which are able to exert their actions on the intestinal epithelium. ETEC strains are characterized by the production of one or both of the following enterotoxin categories (Sherman et al., 1983), all of which are plasmid regulated:

large molecular weight (88 kDa) heat-labile enterotoxins (LT);small molecular weight (11–48 amino acid containing) heat-stable peptide

toxins (ST) resistant to 100°C for at least 15 min.LT enterotoxins are produced predominantly by human and porcine ETEC, while

ST enterotoxins are produced by ETEC of human, porcine and bovine origin.

Table 1. Adhesins and their receptors of different pathotypes of enteric E. coli in calves andin young pigs

Pathotype Adhesin Gene/Operon Location Receptor

ETEC F4 (K88) fae plasmid glycoprotein / mucinF5 (K99) fan plasmid glycolipidF6 (987P) fas plasmid glycoproteinF18 (F107) fed plasmid ?F41 fimf41 chromosome ?

EPEC Bfp pili bfp plasmid ? PE (phosphatidylethanolamine)(EHECa) Intimin eae chromosome Tir (bacterial protein)

NTEC P (Pap) pap chromosome α-dGal(1-4)-β-Galactosefimbria sfa chromosome (glycolipids)S fimbria α-sialyl(2,3) β-Galactose

Afimbrial AFA afa chromosome/ Dr blood group antigenAdhesions plasmid

a EHEC does not have Bfp.


LT toxins have good antigenicity while ST toxins do not. LT toxins can bedivided into two antigenically and biologically distinct but structurally similargroups: LTI and LTII. Within LTI are the LTh−I (human) and LTp−I (porcine) strains,while within LTII two antigenic variants (LTIIa and LTIIb) can be distinguished(O’Brien and Holmes, 1996).

ST toxins fall into two classes: STa and STb (also referred to as STI and STII,respectively). STa toxins have variants which are STaH and STaP [indicating human(H) or porcine (P) type of the STa enterotoxins]. STa toxins are further characterizedby method solubility and by the ability to induce small intestinal fluid secretion inbaby mice and to a lesser extent in weaned pigs. STb is not soluble in methanol anddoes not react in baby mice, however it can induce small intestinal fluid secretion innewborn and weaned pigs.

2.2 Expression and regulation of adhesive virulence factors of ETEC

Genetic regulation and byosynthesis of fimbrial adhesins is of course differentaccording to different fimbriae. There is, however, a general scheme under whichthese operons are constructed and function. There are regulator elements that codefor transacting polypeptides involved in the biogenesis of the whole fimbria andthere are several structural genes encoding polypeptides that partly form majorstructural units ensuring fimbrial (pilus) formation or minor fimbrial units ensuringadhesive capacity and variant specificity (de Graaf, 1990). Changes in the geneexpression can be the result of a random genetic event (stochastic process), butexpression of virulence factors is usually linked more to environmental signals, suchas temperature, ion concentration, osmolarity, carbon source, Fe++, pH, O2 etc.These signals can also be sensed by ETEC bacteria in order to more appropriatelyaccommodate the in vitro and in vivo environment (stereotypic response). Under in vivo conditions some of the above factors can induce a whole cascade of virulencefunctions, turning on different genes while turning off others at different steps of theinfectious process (for instance: invasion genes are turned on early in the infectionbut are repressed once bacteria are within the host cell) (Finlay and Falkow, 1997).For ETEC, and for some other pathotypes mentioned below much less is knownabout regulation. Virulence factors are influenced by the above signals through the“regulator elements”. Some of these control the fimbrial synthesis only, some otherscontrol the expression of many unrelated genes and are therefore called “global regulators”. Virulence genes of enteropathogenic strains of E. coli are mainly genes“foreign” to E. coli and they can be controlled by several regulators. These regulatorsare therefore a possible exciting area of research for ETEC in terms of pathogenesis(in vivo functions) and diagnosis.

Expression and regulation of virulence determinants are also dependent on secre-tion mechanisms: there are three general secretion pathways recognized in Gram-negative bacteria that export virulence factors (I–III). Another group of bacterial


proteins (IV) mediate their own transport and are therefore called autotransportersystems (Finlay and Falkow, 1997). It is also known that the secretion of STa andSTb involves an energy and secA-dependent (type II) conversion of the performedtoxins to the extracellular toxins (Kupersztoch et al., 1990; Yamanaka et al., 1997).However, several steps of enterotoxin and adhesin production as related to secretionsystems have yet to be clarified. It should be noted that the maturation of virulenceproteins is also part of the different secretion and expression mechanisms, i.e. formation of disulphide bonds within the periplasm (for cholera toxin and for LT).

2.3. Adhesins and receptors for ETEC of calves

Most of the ETEC strains responsible for diarrhoea of newborn calves are charac-terized by K99 (F5) and F41 and by STaP enterotoxins. They usually belong to theO8, O9, O20 and O101 serogroups and often produce an acidic polysaccharide typeof K(A) antigen (K25, K28, K30, K35), making the colonies of such strains morecompact and less transparent. It seems that such capsular polysaccharide antigensenhance colonization induced by K99 (Isaacson et al., 1977; Hadad and Gyles,1982). K99 and other fimbrial adhesins mediate attachment of the ETEC to thesmall intestinal (mainly ileal) microvilli, thereby resisting removal and facilitatingcolonization. Thus bacteria are able to efficiently transmit the STa that they typicallyproduce, which in turn induces extensive excretion and loss of water and electrolytes, rapidly leading to dehydration. Other, less frequently occurringadhesins are the so-called F17 (earlier known as FY and Att25) (Lintermans et al.,1988). Adhesions mediated by these surface proteins are generally dependent on thepresence of glycoprotein or glycolipid receptors, which are abundantly present innewborn calves and lambs. In the case of K99, for instance, the receptors are acidicglycolipids (gangliosides) like N-glycolyl-GM3, which gradually decrease with age(Runnels et al., 1980; Willemsen and de Graaf, 1993; Teneberg et al., 1994).Although K99 and F41 are frequently produced simultaneously by bacteria of thesame ETEC strain, there are different receptors for K99 (sheep and horse haemag-glutinin) and for F41 (guinea pig and human-A haemagglutinin). K99 and F41 also differ in their genetic regulation (K99 is regulated by a plasmid while F41 isregulated by a chromosome). Both K99 and F41 as well as F17 can, however, alsoadhere to the porcine small intestinal brush border and can induce porcine entero-toxic colibacillosis. Receptors for these adhesins are of course different. K99 receptors are certain glycolipids (as mentioned above), F41 receptors are glyco-proteins (i.e. glycophorin) (Brooks et al., 1989), while the receptors for F17(FY/Att25) are on the sialyated mucus (Mouricout and Julien, 1987). It must be mentioned that association of F17 (FY/Att25) with ETEC is not quite clear. Originaldescriptions of F17+ E. coli reported enterotoxic activities (Pohl et al., 1986;Lintermans et al., 1988). Studies in recent years revealed that F17 fimbrial adhesinsare somewhat heterogeneous and they form a so-called F17 family of fimbriae


(F17a, F17b, F17c, F17d and G fimbriae) based on their receptor specificities (Le Bougouénec and Bertin, 1999). Another (non-fimbrial) surface protein (CS31A)has also been associated with calf diarrhoea (Girardeau et al., 1988) but it is alsodetected on septicaemic E. coli from calves, in contrast to K99 or F41. Interestingly,CS31A is genetically related to K88 fimbria (known as a typical porcine adhesin)(Girardeau et al., 1988). In fact, the N-terminal sequence of purified CS31A showsa homologous protein of 26.77 kDa between CS31A, F41 and K88, indicating anevolutionary relationship between these fimbrial and afimbrial adhesions (Girardeauet al., 1991). More on the F17 fimbriae will be mentioned in the section on necro-toxigenic E. coli (NTEC).

In connection with ETEC of calves there should also be a few words on ETECof goat kids and lambs. As mentioned above, lambs have a very similar clinical diarrhoeal disease and similar strains of ETEC as calves. However, this seems muchless certain in goat kids. In general, it is true for both animal species that we havemuch more limited information about their ETEC infections as compared to thoseof calves. For instance, the adhesins F17 (FY/Att25) and CS31A detected on calfdiarrhoea strains have not been described so far for E. coli bacteria from lamb orgoat diarrhoea, but such isolates can be prevalent among septicaemic strains oflambs and goat kids (Le Bougouénec and Bertin, 1999). Information about ETECinfection in goats is even more limited. According to our earlier studies (Nagy et al.,1984), infection by K99 + ETEC may also cause diarrhoea of young goat kids in some herds but cryptosporidiosis and rotavirus infections seem to be the mainaetiological agents. This observation is supported by the experimental infection ofgoat kids with K99 + ETEC strains and by successful prevention of diarrhoea bythe K99 vaccine (Contrepois et al., 1993). In contrast to ETEC, verotoxic E. coli(VTEC) strains have been isolated more frequently from 1–2-month-old goat kidswith diarrhoea and they seem to be the major diarrhoeal agent of this age group(Duhamel et al., 1992). More information is needed, however, about ETEC (and ingeneral about enteric E. coli) infection of goat kids and lambs.

2.4. Adhesins and receptors for ETEC of pigs

Enteric enterotoxic colibacillosis produces significant losses in two different agegroups of pigs: first among newborn pigs and later at the post-weaning age.Aetiology, pathogenesis and epidemiology should be discussed separately for thetwo age groups, but diagnosis, treatment and prevention have enough in common tobe described under one separate heading for pigs and calves.

E. coli strains of enterotoxic colibacillosis in suckling piglets are characterizedby one or the other of the K88 adhesins (in variants K88ab, K88ac, and K88ad) alsoknown as the (F4), by K99 (F5) or 987P (F6) adhesins and occasionally by the F41(Vazquez et al., 1996), F165 (Fairbrother et al., 1986) or F42 adhesins (Sperandioand da Silveira, 1993). Among these adhesins K88 (F4) and 987P (F6) are specific


for pigs, while K99 (F5) and F41 seem to have receptors in both pigs and calves.ETEC strains possessing K88 (especially K88ac) are the most common cause of diar-rhoea and they usually produce LT in addition to STaP or STb. K88 + ETEC are alsocharacterized by haemolysin production in vitro. ETEC strains carrying K99 and/or F41or 987P produce only STaP and are non-haemolytic. While K88 + ETEC may representabout 40 – 60% of the E. coli strains causing diarrhoea in piglets, the above non-K88strains make up between 20 –30% (Woodward and Wray, 1990; Nagy, 1993). The typical O serogroups for neonatal porcine ETEC infections are O8, O9, O20, O101,O141, O147 and O157 representing both K88+ and non-K88 ETEC. In our experience,the two groups (K88+ and non-K88) of ETEC have a somewhat different clinical pic-ture: K88 strains cause more severe diarrhoea at a younger age (1–5 days) while non-K88 strains give rise to milder diarrhoea with a later onset (approximately 4–14 days ofage). It should also be noted that the rotavirus infection often complicates neonatal coli-bacillosis of pigs, especially in non-K88 ETEC infections at the second week of age.

Small intestinal receptors of adhesins are the other essential element of intestinalcolonization by ETEC. It has been shown that the receptors for K88 are glycoproteinsand the lack of production is a recessive trait. Thus, homozygous piglets are resistantto K88 mediated adhesions, to colonization and to disease (Sellwood et al., 1975).The genes responsible for production of intestinal receptors belong to the TF bloodgroup linkage group (Gibbons et al., 1977). Receptor functions seem to be dependenton the “b”, “c” and “d” components, and in genetically resistant piglets the receptorsare usually absent for both of these components of the K88 variants (Hohmann andWilson, 1975; Bijlsma et al., 1982). In vitro adhesion tests have revealed a polymor-phism of intestinal receptors for K88 and indicated that there are 5–6 different adhe-sion patterns (A–F) among piglets according to the K88ab, K88ac and K88advariants (Bijlsma et al., 1982; Rapacz and Hasler-Rapacz, 1986; Billey et al., 1998).Unfortunately, this phenomenon of genetically determined resistance could not gaina wide practical application. It may, however, complicate epidemiological pictures,by partially producing non-diarrhoeal homozygous recessive (ss) litters, and by partially leaving heterozygous (Ss) piglets (which are born to resistant sows and sensitive boars) without colostral immunity (such sows would not have acquired theinfection and could not produce specific antibodies in their colostrum). The practicalapplication of this knowledge is further complicated by the fact that the correlationof the adhesion of K88 variants to the small intestinal brush borders with suscepti-bility to colonization and diarrhoea may be lacking. This can be explained by thefindings of Francis et al. (1998), suggesting that the intestinal mucin-type glyco-protein (IMTGP) is a biologically more relevant receptor for K88ab and K88ac ascompared to the so far widely accepted enterocyte brush border glycoprotein.

So far, no information is available about the genetic determination of receptorsfor K99, F41 or 987P in pigs, but there are mice that are genetically resistant to col-onization by K99 (Duchet-Suchaux et al., 1990). Future research on these areas ofmammalian genetics would clearly be needed.


The production of receptors also influences age-related resistance to the disease.This is, however, manifested in different ways for different adhesins. Receptors forK88 are abundant in newborn pigs and will decrease with age but remain relativelystable throughout the weaning and post-weaning periods. Receptors for K99 gradu-ally decrease with age (Runnels et al., 1980). In contrast, production of receptors for987P, does in fact increase with age (Dean et al., 1989). This invariably leads to alower intensity of adhesion and colonization because the receptors are shed intothe lumen and block bacterial adhesion before contacting intestinal epithelial cells.The ageing nature of receptors for F41 is unknown but data indicate that they maybe produced all through the weaning age in pigs (Nagy, unpublished results).

Post-weaning diarrhoea (PWD) is usually the most constant disease problem oflarge-scale farms, especially of those that wean around 3–4 weeks of age. PWD startsa few days after lacteal protection completely ceases, and pigs are placed in an envi-ronment that is completely new from a technical, social and microbiological pointof view. It is widely accepted that specific serotypes and pathotypes of ETEC areresponsible for the major part of PWD. It is also without debate that the disease is ahighly complex one in which ETEC only plays a part (although an essential one). Itis frequently seen in almost all large-scale piggeries but it is one of the most difficultdiseases to reproduce experimentally. Diarrhoea and reduction of weight are onlypart of the losses. Retarded growth, which usually follows diarrhoeal episodes in weaned pigs, makes the losses even worse. The main cause of post-weaning diarrhoea is the weaning itself. Only on this basis can we understand the aetiologyand pathogenesis more realistically and can we be more humble about our capacitiesto bring real (economically feasible) improvement to this enigma. The ETEC strainsinvolved are most frequently of the O serogroup: O8, O141, O138, O147, O149,O157, of which O149:K88 seems to be the predominant serotype in most countries(Hampson, 1994). So far, all the typical PWD strains of ETEC are haemolytic,although haemolysin does not play an essential part in the virulence of porcine ETEC(Smith and Linggood, 1971).

The most frequent adhesive virulence factors of ETEC strains in the case ofPWD are K88 (mainly K88ac) fimbria. Furthermore, K99, 987P and F41 have alsobeen described on some PWD strains (Nakazawa et al., 1987; Nagy et al., 1990a,1996a) but they seem to be rarely involved in diarrhoea at that age. Recently, a newfimbrial adhesin has been recognized under the F18 designation.

The F18 fimbriae have been described under different names, and misunder-standings are frequent in the use of the earlier names and new designations. Duringthe past few years, three new colonization factors or adhesive fimbriae have beendescribed for groups of E. coli involved in PWD or oedema disease: F107 onoedema strains (Bertschinger et al., 1990), 2134P on ETEC strains (Nagy et al.,1992b), and “8813” also on ETEC strains (Salajka et al., 1992). Additionally, fimbriae of two ETEC strains of serogroup O141 have also been described (Kennanand Monckton, 1990), although no data have been given on their adhesive or


pathogenetic significances. As a first attempt to clarify the relationships betweenthese factors, pili 2134P were compared to fimbriae F107 by means of polyclonaland monoclonal antibodies. It was provisionally concluded that these two adhesinswere morphologically similar and shared a common antigenic determinant in addi-tion to a type-specific one (Nagy et al., 1992a). These findings were confirmed(Wittig et al., 1994) and it was suggested that the symbol “a” should be used for thecommon determinant and the symbols “b” and “c” for the specific determinants ofF107 and 2134P respectively. Furthermore, Rippinger et al. (1995) investigated the morphological, immunological, genetic and receptor-binding relatedness of fimbriae F107 and 2134P, together with the colonization factor “8813”. Based onearlier suggestions made by Ida and Ørskov (International Escherichia coli Centre,Copenhagen, 1992) for a new F18 fimbria, it was shown that two serological variants were determined and should be designated as follows: F18ab (for F107) andF18ac (for 2134P and 8813) (Rippinger et al., 1995). The genetic relatedness of theabove family of F18 fimbriae was described by Imberechts et al. (1994), supportingthe above grouping, and adding the fimbriae of Kennan’s O141 strains (Kennan andMonckton, 1990) to the group of F18ac. In a recent study, it was pointed out thatF18ab and F18ac fimbriae are biologically distinct: F18ab fimbriae are poorlyexpressed both in vitro and in vivo. They are frequently linked with the productionof SLT-IIv (VTEC strains), while F18ac are more efficiently expressed both in vitroand in vivo and they are more characteristic of ETEC strains (Nagy et al., 1997). It should also be mentioned that some ETEC strains may produce multiple adhesinssuch as K88, F18ac or K88, F41 or even K88, F18ac, and F41 (Nagy et al., 1996a).It remains to be shown if such strains have a pathogenetic advantage over strainswith one kind of adhesin. It may also be questioned under what conditions there arereceptors for these rarely occurring adhesins (K99, 987P, F41) available in the rightamount on the small intestinal mucosae.

In weaned pigs receptors for K88 are produced, although to a somewhat reducedextent, all through the weaning age, while receptors for the variants of F18 (F18aband F18ac) are increasingly produced up to the weaning age (Nagy et al., 1992a,1997) and the fimbriae F18ac seem to have more receptors around the ileal Peyer’spatches (Nagy et al., 1992a). The lack of receptors for F18ab and F18ac in newbornpigs offers an explanation why these VTEC and ETEC strains (and why the oedemadisease itself) are only prevalent in weaned pigs.

Inherited resistance to PWD owing to production of intestinal receptors of fimbria F18ab has also been investigated by oral inoculation of weaned pigs and byin vitro adhesion tests (Bertschinger et al., 1993), and it seems that phenotypes susceptible or resistant to F18 adhesion can be differentiated. Pigs with at least onecopy of a dominant allele for receptors are susceptible to colonization and in vitroadhesion (which is similar to the K88 receptors). Additional genetic marker studieslocalized the receptor gene on the porcine chromosome 6, closely linked to the gene


encoding for halothane sensitivity (Vogeli et al., 1994). It seems that the lack ofreceptors will coincide with halothane (stress) sensitivity, making it difficult toselect and raise pigs without small intestinal receptors for F18 fimbria. Small intestinalreceptors for K88 and for F18 seem to be different, on the basis of comparative in vitro studies (Nagy et al., 1997) and the different localization of their regulationon the porcine chromosome (Gibbons et al., 1977; Vogeli et al., 1994).

Breeding pigs resistant to ETEC adhesion seems to be very difficult. First of allit is difficult to select subdominant alleles of two different, independently inheritedtraits (lack of receptors for K88 and F18) but we should also consider that the E. colibacteria are genetically much more flexible than their host. This would ultimatelylead to the emergence and proliferation of new ETEC pathotypes. Furthermore, weshould take into account the possible co-selection of unwanted traits (such ashalothane sensitivity).

3. VEROTOXIGENIC E. COLI IN PIGS AND CALVES

Verotoxigenic E. coli (VTEC) is one of the alternative names for E. coli bacteriaproducing a cytotoxin detectable on Vero (African green monkey kidney) cell culture (Konowalchuk et al., 1977) and sharing a number of properties with Shigatoxin (O’Brien et al., 1977). Therefore they are also called “Shiga-like” toxin producing E. coli (SLTEC). The toxins of this group are generally characterized bythe same or similar structure and a pathomechanism like that of the toxin of Shigelladysenteriae (composed of one A and five B subunits). As a result of enteric infectionwith VTEC strains, these toxins (VT1 and VT2) produce enteric (haemorrhagic colitis)and systematic disease (haemolytic uraemic syndrome) in humans, practically onlyenteric disease in calves (calf dysentery) and systemic disease in pigs (oedema disease).The pathomechanisms of these toxins are characterized by a receptor-mediatedendocytosis of the A subunit of the VT, followed by a fusion in lysosomes andrelease of the enzymatically active fragment A1, leading to inhibition of protein synthesis and cell death.

Most of the VTEC (SLTEC) bacteria of ruminants and humans produce a characteristic attachment and effacement (AE) type of microvillous degenerationand bacterial adhesion (fig. 1b) (as will be described later in the section on EPECand EHEC). However, there are verotoxin-producing E. coli strains which do nothave the AE phenotype, and therefore do not produce such characteristic lesions butpossibly adhere to the brush border, leaving the microvilli intact. In order to differ-entiate these verotoxic bacteria from those which also produce characteristic lesions(and haemorrhagic colitis) in this chapter we refer to them as verotoxigenic E. coli(VTEC). Such VTEC bacteria produce oedema disease in pigs and there are some others that produce milder diarrhoea in humans and in calves (Wieler, 1996;Mainil, 1999).


3.1. Adhesins of VTEC for calves and pigs

In calves most of the VTEC strains also have the attachment effacement capacity(Gyles, 1994; Wieler, 1996) and could therefore be designated as enterohaemor-rhagic E. coli (EHEC) (see below). The non-AE strains of VTEC of calves have notbeen thoroughly studied for adhesins yet. Wieler (1996) demonstrated that bovineVTEC strains without AE genes were relatively frequent among VT2 strains ofcalves. Wieler has also demonstrated that many of these strains also adhered to cultured Hep2 cells, and they were all negative for bundle-forming pili (Bfp) characteristic of human EPEC. At this point it remains an interesting question howthese strains could colonize the bovine intestine and continue being shed into theenvironment. It can be speculated whether such VTEC strains of calves have losttheir earlier capacity to produce AE lesions or they have not gained it yet, or theyhave developed as a clonally different lineage.

In pigs at present the only VTEC known are the ones producing oedema disease.They produce a variant of VT2 (VT2e). This toxin leaves the intestinal epithelialcells without damage but enters the bloodstream, using receptors on red blood cells,and damages the endothelial cells of the small blood vessels by inhibiting proteinsynthesis. This leads to perivascular oedema and hyalinization in several organs andto death. The only known adhesins of the porcine VTEC are the F18 fimbria (asdescribed above). Adhesion and colonization mediated by F18 do not cause charac-teristic damage to the morphology of the intestinal cells, resembling the morphol-ogy of adhesion by ETEC (Bertschinger et al., 1990; Nagy et al., 1997).

4. ENTEROPATHOGENIC E. COLI AND ENTEROHAEMORRHAGIC E. COLI

Enteropathogenic Escherichia coli (EPEC) were first described in the 1940s and 1950sas the causative agents of infantile diarrhoea, and are still a major cause of infant diar-rhoea in the developing world. EPEC do not produce enterotoxins and are not invasive;instead their virulence depends on causing characteristic intestinal histopathologycalled attaching and effacing (AE), which can be observed in intestinal biopsy and in vitro (Moon et al., 1983; Knutton et al., 1987). The AE phenotype is characterizedby effacement of microvilli and intimate adherence between the bacterium and theepithelial cell membrane. The AE phenotype develops due to a specific signallingpathway and the AE lesions are characterized by localized effacement of the brushborder of enterocytes with intimate bacterial attachment and pedestal formationbeneath the adherent bacteria. EPEC have a set of adhesins (reviewed by Nataro andKaper, 1998). Intimin is essential, but not enough for the pathogenesis of EPEC, and itis encoded by a chromosome (Jerse et al., 1990). Most of the EPEC strains possessa plasmid of about 60 Mda which promotes the adherence to cultured epithelial cellsin a localized adherence (LA) pattern. Early studies proved the importance of this


plasmid, named EAF (EPEC adherence factor) (Baldini et al., 1983). The EAF plas-mid encodes a fimbrial adhesin called bundle-forming pilus (Bfp, Giron et al., 1991).

4.1. Intimin and translocated intimin receptor (Tir)

The first gene to be associated with the AE phenotype was the eae (for E. coliattachment effacement) encoding intimin, a large molecular weight outer membrane protein (Jerse et al., 1990; Jerse and Kaper, 1991). Subsequently, the eae gene wasshown to be part of a large chromosomal region of DNA that encodes all the neces-sary determinants for the AE phenotype (McDaniel et al., 1995). This chromosomalregion is named the locus of the enterocyte effacement (LEE) pathogenicity island.The LEE is responsible for the AE intestinal histopathological changes caused byEPEC and enterohaemorrhagic E. coli (EHEC) and related animal pathogens, first ofall rabbit EPEC. The LEE is organized into three main parts (gene clusters).

The middle part of the LEE contains the eae and tir genes as well as the cesTgene. The right side encodes for the proteins secreted via the type III secretory system (espA, espB, espD, espF genes), while the left side encodes for the genes ofthe type III secretory system itself (partly functioning like a molecular syringe).Details of the functions of the three areas of the LEE follow.

On the middle part the eae codes for intimin (Eae), a 94–97 kDa outer membraneprotein that is an intestinal adherence factor to epithelial cells, and tir (translocatedintimin receptor) encodes the Tir, the intimin receptor protein (Kenny et al., 1997;Deibel et al., 1998). E. coli eae genes have been cloned and sequenced from differ-ent EPEC and EHEC strains isolated from humans and animals including calf(Goffaux et al., 1997). Sequence comparisons of different eae genes revealed thatthe N-terminal regions show high conservation but the C-terminal regions encodingthe last 280 amino acids are heterogeneous. The cell-binding activity of intimin islocalized at the C-terminal 280 amino acids of polypeptide “Int280” (Frankel et al.,1995; Liu et al., 1999).

Immunological and genetic studies revealed the existence of pathotype-specificintimin subtypes. Agin and Wolf (1997) identified three intimin types, α, β, γ, Adu-Bobie et al. (1998a) detected four distinct subtypes of intimin, α, β, γ, δ, andrecently Oswald et al. (2000) characterized an additional new intimin variant,intimin ε. Molecular studies revealed that these intimin types are pathotype (andspecies) specific. Intimin α was specifically expressed by human EPEC strainsbelonging to classical EPEC (clone 1) serotypes of O55:H6, O125:H, O127:H6,O142:H6 and O142:H34 (Adu-Bobie et al., 1998b). Intimin β appears to be themost ubiquitous type: it is associated with EPEC strains belonging to clone 2(O26:H−, O111:H−, O111:H2, O142:H2, O119:H2, O1219:H6, and O128:H2) andEHEC O26:H11; intimin β was detected in rabbit O15:H−, O26:H11, and O103:H2strains; and this subtype was present in O26:H11 bovine strains as well (Oswald et al., 2000). Intimin γ is associated mainly with human and cattle Shiga-like toxin


producing E. coli (SLTEC) strains including sorbitol-fermenting and sorbitol non-fermenting EHEC O157:H7, O157:H− strains, and SLTEC strains of serotypesO111:H8, O111:H−, O86:H40, O145:H−, and EPEC O55:H− and O55:H7 strainsalso harbour intimin γ. Intimin δ was associated with human EPEC O86:H34 (Adu-Bobie et al., 1998a). Intimin ε was present in human and bovine EHEC strainsof serogroups O8, O11, O45, O103, O121 and O165 (Oswald et al., 2000).

The observation that different intimin subtypes are associated with differentpathogenic clones can explain why these strains colonize different segments ofthe intestine in different host species. Tzipori et al. (1995) infected pigs with humanstrains having different types of intimin and demonstrated that the intimin α-producing strain caused AE lesions in both the large and the small intestine, whilethe intimin γ-producing EHEC strain caused AE lesions only in the large intestine.When the pigs were infected with an eaeγ − but eaeα+ EHEC recombinant strain, AElesions were observed in both the small and the large intestine (Tzipori et al., 1995).

Interestingly, in the case of EPEC, the receptor for bacterial adhesin (intimin) isanother protein of the same bacteria called translocated intimin receptor (Tir). Tir isa bacterial protein that is translocated into the host cell via a type III secretion system and upon entry into the eukaryotic cell it serves as the receptor for theintimin. Initially it was believed that the intimin receptor protein is a mammalianmembrane protein that was originally called Hp90 (“host protein”) and that wastyrosine phosphorylated in response to EPEC infection (Rosenshine et al., 1996).The combined interactions between host kinases and EPEC proteins result in additional host signalling events such as actin aggregation and polymerization leading to the characteristic cellular pathology. In E. coli O157:H7 infection the Tirprotein has an analogous function, but it is not phosphorylated after translocating tothe eukaryotic cell (DeVinney et al., 1999).

As mentioned above, the LEE contains two additional main functional clusters:on the right side of the LEE are the espA, espB, espD, espF genes, of which the firstthree genes are necessary for the AE phenotype. The EspA is a structural protein anda major component of a large organelle; it is transiently expressed on the bacterialsurface and interacts with the host cell during the early stage of AE lesion formation.EspA forms a physical bridge between the bacterium and the infected eukaryotic cellsurface and is required for the translocation of EspB into infected epithelial cells, andmay contribute to bacterial adhesion as well (Knutton et al., 1998). EspB protein istranslocated into the host cell membrane by EspA and cytoplasm and serves as thedistal end of EspA filament, and it might have a function in the host signal transduc-tion events. EspB promotes tyrosine phosphorylation of Tir and induction of inositolphosphates and calcium fluxes. The increased calcium levels can induce cytoskeletalrearrangements and activate calcium-dependent kinases resulting in morphologicalchanges including microvillus effacement and pedestal formation.

McNally et al. (2001) observed clear differences in the expression of LEE-encodedfactors between O157 strains, with the same stx+ eae+ genotype, isolated from

human disease cases and those isolated from asymptomatic cattle. All strains produced a detectable amount of EspD when grown in tissue culture medium, but in the case of the human O157 strains that amount was on average 90-fold higherthan for the bovine O157 strains. The level of secretion also correlated with the abil-ity to form AE lesions on HeLa cells, and with only high-level protein secretors intissue culture medium exhibiting a localized adherence phenotype (McNally et al.,2001). These data correlate with earlier findings based on the results of a compre-hensive molecular analysis (Kim et al., 1999). That analysis revealed the existenceof two distinct lineages of E. coli O15:H7 in the United States. Human and bovineisolates are non-randomly distributed among the lineages, suggesting that lineage IIstrains may not readily cause disease or may not be transmitted efficiently tohumans from bovine sources. Alternatively, the distribution may reflect a loss ofcharacteristics in lineage II that are necessary for virulence in humans, perhaps as aconsequence of adaptation to the bovine environment.

On the left side of the LEE is a set of genes coding for the type III secretion system itself. These genes share sequence homology with the type III secretion systems of Yersinia enterocolitica, Shigella flexneri and Salmonella typhimurium(reviewed by Mecsas and Strauss, 1996). The type III secretion systems are respon-sible for secretion and translocation of different virulence determinant proteins suchas the espA, -B, -D, -F encoded proteins and the Tir protein.

4.2. EPEC adherence factor plasmid

The majority of EPEC strains possess a plasmid 50–70 MDa in size, named the EAF(EPEC adherence factor) plasmid. These plasmids share extensive homology amongvarious EPEC strains. The typical EPEC strains associated with diarrhoea possessEAF, while EPEC strains that do not have the EAF plasmid are referred to as atypical EPEC (Nataro and Kaper, 1998). The importance of EAF was demonstrated in vivo by Baldini et al. (1983) and a volunteer study revealed that EAF is essentialfor the full virulence (Levine et al., 1985). Giron et al. (1991) identified an EAFencoded adhesin, called bundle-forming pilus (BFP), which is a member of the type IV pilus family. The expression of BFP was associated with localized adherence to HEp-2 cells and the presence of the EPEC adherence factor plasmid(Giron et al., 1991).

Barnett-Foster et al. (1999) demonstrated that phosphatidylethanolamine (PE)serves as a receptor for EPEC and EHEC. These bacteria bind to PE specifically andin a dose-dependent manner, and this binding was consistently observed whether thelipid was immobilized on a thin-layer chromatography plate, in a microtitre well orincorporated into a unilamellar vesicle suspended in aqueous solution. Bacterialbinding to two epithelial cell lines also correlated with the level of outer leaflet PEand it was reduced following preincubation with anti-PE. The PE-binding pheno-type of EPEC correlated with the bfp genotype of a number of clinical isolates.



4.3. EPEC (and EHEC) in pigs and calves

In calves the first description of attachment effacement (AE) lesions due to “atypi-cal E. coli” was given in the UK by Chanter et al. (1984), in relation to natural casesof “calf dysentery”. The E. coli O5 strain produced bloody diarrhoea and typical AElesions in gnotobiotic piglets (Chanter et al., 1986) and it turned out to be verotox-igenic as well. Further observations in the United States indicated that verotoxigenicand AE lesion-producing strains of E. coli O26 are a relatively frequent cause of calfdiarrhoea (Janke et al., 1990) and such strains have also been detected to cause nat-ural infections in the UK (Gunning et al., 2001). On the basis of the fact that most ofthe bovine AE lesion-producing strains also produce verotoxins, they could also benamed enterohaemorrhagic-like E. coli (or EHEC-like strains). As is well known,the classical human EHEC strains O157:H:H7 and O157:NM are frequently carriedasymptomatically by calves and older cattle as well as by small ruminants and it isusually among the least frequent serotype occurring in cattle, in contrast to humanswhere O157 is the leading serogroup of EHEC (reviewed by Dean-Nystrom et al.,1998). The EHEC strains causing bovine diseases (O26, O103, O111, O118 andO157) can also be transmitted to humans and, thus, these strains have a seriouszoonotic potential. So far it seems that several bovine and human strains have betaintimin (supporting the zoonotic significance of bovine EHEC).

Non-verotoxigenic AE E.coli seem to be relatively rare in calves, although theycan also produce watery diarrhoea (Pearson et al., 1989), and can be regarded as thebovine EPEC. The intimin type of these bovine EPEC strains is usually also the betaintimin (Oswald et al., 2000). At present there is no solid information available onany additional adhesive factor of bovine EPEC or EHEC strains, although it seemsquite likely that there are some peculiarities in the adhesins of these strains as well.

EPEC strains of porcine origin were first detected by Janke et al. (1989) andporcine EPEC infection was studied on newborn pigs by Helie et al. (1991) whohave shown colonization and typical AE lesions in the ileum and jejunum as earlyas 12–24 h after infection with a porcine O45:K“E65” E. coli, while the caecum andcolon were colonized at 24–48 h post infection. This group demonstrated thatporcine EPEC have virulence characteristics similar to those of human strains (Zhu et al., 1994, 1995) and, by using transposon mutagenesis, identified a porcineattaching-effacing-associated (paa) factor associated with the presence of the eaegene. Interestingly this paa was found in EHEC O157:H7 and in O26 strains and astrong association was with the heat-labile enterotoxin (LT) gene (An et al., 1999).Further studies have proven that the eae gene of porcine EPEC prototype E. coli1930 (O45) strain was a member of the beta intimin group and showed the highestsimilarity with the rabbit EPEC strains (An et al., 2000). Such strains may be pres-ent in small numbers in the pig population not only in North America but also inEurope as well (Osek, 2001). However, the overall significance of porcine EPECstrains cannot be judged on the basis of the available data. It seems that further epi-demiologic studies are needed to establish their significance in porcine enteric disease.

It is interesting to see that in pigs two genetic lineages have diverged: one is VTEC(verotoxin production without AE lesion) and the other is EPEC (AE lesion withoutverotoxin production). The first is responsible for oedema disease of weaned pigs(with well-identified fimbrial adhesin) while the other may induce diarrhoea in pigs(with beta intimin and paa antigen).

5. NECROTOXIGENIC E. COLI

Necrotoxigenic Escherichia coli (NTEC) are defined as E. coli strains producing alarge molecular weight toxin named cytotoxic necrotizing factor (CNF). NTEC areassociated with intestinal and extraintestinal diseases in animals and human beings(DeRycke et al., 1999). CNF was first identified from children with enteritis byCaprioli et al. (1983). The large monomeric protein toxin causes necrosis in rabbitskin and induces formation of multinucleation and thick bundles of actin stressfibres in HeLa, CHO and Vero cells (Caprioli et al., 1984) (fig. 1c). Two types ofCNF (CNF1 and CNF2) have been identified, each of them being genetically linkedto several other specific virulence markers (DeRycke and Plassiart, 1990). TheCNFs covalently modify Rho proteins (small GTPases) that regulate the physiologyof the cell cytoskeleton of mammalian cells, and lead to polymerization of actinfibres (Oswald et al., 1994; Fiorentini et al., 1995). CNFs are encoded by a singlestructural gene. The CNF1 operon is located on the chromosome (Falbo et al., 1992)and CNF2 is determined by a conjugative plasmid (Oswald et al., 1994).

NTEC1 strains can be found in humans and in all species of domestic mammals(DeRycke et al., 1999). The CNF1 operon is frequently associated with other virulencefactor genes and these genes constitute large chromosomal regions called pathogenic-ity islands (PAIs, Hacker et al., 1997). One of these PAIs (PAI II) encodes CNF1, alpha-haemolysin and P-fimbriae and it was first identified in a human uropathogenic E. coli(UPEC) strain. This virulence gene pattern was reported in intestinal strains isolatedfrom suckling (Garabal et al., 1996; Dozois et al., 1997) and from weaned pigs (Tóthet al., 2000), which may be explained by the unusual mobility of the PAIs.

NTEC2 strains have only been reported in ruminants (DeRycke et al., 1999). In NTEC-2 strains, CNF2 is encoded by a virulence plasmid (pVir, Oswald et al.,1994). pVir also codes for a new member of the cytolethal distending toxin family(CDTIII, Peres et al., 1997) and for the F17b or F17c fimbrial adhesin that confers theability to adhere to calf intestinal villi (Oswald et al., 1994) and enter the bloodstream(Van Bost et al., 2001). It is tempting to speculate that the large conjugative plasmid(pVir) is also carrying a PAI containing the operons for CNF2, F17b, and CDTIII.

5.1. Adhesins and receptors of NTEC isolated from animals

Molecular epidemiological studies revealed that most of the human and animalNTEC strains have different fimbrial (pap, sfa, f17) and afimbrial adhesin (afa)genes. Mainil et al. (1999) reported that most NTEC1 extraintestinal calf isolates



hybridized with the PAP probe and additionally either with the SFA probe (37%) orwith the AFA probe (49%). In contrast, the NTEC2 isolates hybridized with the F17probe (45%), with the AFA probe (19%), or with the F17 and AFA probes simulta-neously (22%). In correlation with the CNF2 prototype strains E. coli S5 (Smith,1974) and E. coli 1404 (Peres et al., 1997) all the 19 NTEC2 cattle isolates had avirulence plasmid coding for CNF2 and most of them coded for fimbrial (F17) orafimbrial (AFA) adhesins as well, for which the PCR results suggested the existenceof a new variant of AFA (Mainil et al., 1999).

Examining 32 herds for NTEC, it was found that CNF2 was more frequentlydetected than CNF1 in the faecal samples of healthy cattle. CNF2-producing NTECstrains were significantly more frequently isolated from calves (24%; 17 of 71) than from cows (4%; 11 of 257). Reports confirmed that healthy calves are a reservoirof NTEC producing CNF2 (Blanco et al., 1998), and NTEC that produced CNF2 may be part of the normal intestinal flora of cattle (Blanco et al., 1993). A sero-epidemiological study revealed that the O groups of CNF2+ strains isolated from cows(O2, O8, and O14) were different from those found in calves (O8-O75, O15, O55,O86, O88, O115 and O147) (Blanco et al., 1998). Depending on the serotypes theCNF1-producing strains isolated from human extraintestinal infection had differentadhesins (Blanco et al., 1994). These latter authors also suggested that extraintestinalinfections are caused by a limited number of virulent clones. CNF1 strains of serotypesO2:K7:H− and O4:K12:H1 express P fimbriae, whereas CNF1 strains of serotypesO2:K?:H1, O2:K1:H6 and O75:K95:H5 possess the adhesin responsible for the so-called MRHA type III. In the following section the above mentioned P and S fim-briae and the afimbrial adhesions (AFA) will be discussed in some detail. Informationon the F17 fimbrial family has partly been provided in the bovine ETEC section.

5.2. P (pap) fimbriae

Type P fimbriae are also named pyelonephritis-associated pili (pap) and have been recognized as P blood-group-specific adhesins (Kallenius et al., 1981). They are composed of a thin fibrillum (carrying the adhesin) at the proximal end of a more rigidpilus rod 7 nm in diameter (Kuehn et al., 1992). P fimbriae are part of a family of adhesive organelles that are characterized by an assembly machinery consisting of aperiplasmatic chaperone (PapD) and a pore-forming outer membrane (PapC) usher protein (Hultgren et al., 1996). The 11 genes coding for functional P fimbrial adhesinare clustered in an operon encoding the main component of the pilus rod (PapA) andseveral minor fimbrial subunits (PapH; K; E; F), the PapG which is the adhesin and theassembly machinery (PapC; D; J), and the two regulatory proteins (Pap J; B) (reviewedby Hultgren et al., 1996). PapG adhesin located at the tip of the fimbriae binds to the alpha-D-galactopyranosyl-(1–4)-beta-D-galactopyranose or Gal alpha (1–4)Galdisaccharides (Kuehn et al., 1992), while the receptor for the P-related sequences (prs)is the GalNAc-α-(1–3)-GalNAc which is related to fimbriae of serotype F13.


There are known alleles of PapG, referred to as classes I, II, and III. Theseclasses have different haemagglutination patterns. PapG I agglutinates only humanerythrocytes, PapG II agglutinates human erythrocytes very well and sheep erythro-cytes only poorly, and PapG III agglutinates only sheep erythrocytes (reviewed byHultgren et al., 1996).

The pap operon is located on the bacterial chromosome mostly associated withother virulence factor genes forming PAIs in uropathogenic E. coli strains. At leastfour PAIs are present in the genome of UPEC 536 of O6:K15:H31 prototype, andthree of them encode different adhesions: PAI I and II carry genes for P fimbriae andhaemolysin, while PAI III encodes the S fimbrial adhesin. UPEC J96 of serotypeO4:K6 has two PAIs. One PAI carries virulence determinants pap and hlyI. The second PAI encodes CNF1, hlyII and harbours prs (pap-related sequence) genes.Because these islands represent a mechanism for spreading the pathogenicity factorsbetween strains belonging to the same and different species, the presence of classi-cal UPEC specific adhesins in intestinal isolates such as NTEC1 strains is under-standable (reviewed by Hacker et al., 1997).

5.3. S fimbriae

S fimbrial adhesins I and II (SfaI and SfaII), produced by extraintestinal Escherichiacoli pathogens that cause urinary tract infections (UTI, Hacker et al., 1985) andnewborn meningitis (NBM, Hacker et al., 1993), respectively, mediate bacterialadherence to sialic acid-containing glycoprotein receptors (Moch et al., 1987) pres-ent on host epithelial cells and on extracellular matrix. The S fimbrial adhesin (sfa)determinant of E. coli comprises nine genes (Schmoll et al., 1990). Both SfaI andSfaII adhesin complexes consist of four proteins: SfaA (16 kDa) is the major sub-unit protein and the minor subunit proteins are SfaG (17 kDa), SfaS (15 kDa), andSfaH (29 kDa).

Genetic and functional analysis of the sfa I complex conducted by Khan et al. (2000) revealed that sialic acid-specific binding is mediated by the minor subunit protein SfaI-S, which is located at the tip of the fimbriae. The SfaI-S was the only minor protein gene which increased the degree of fimbriation and pro-vided adhesion properties for a non-adhesive derivative K-12 strain which had the sfaI-A major subunit gene but had neither the sfaI-G nor the sfaI-H gene. sfaEF genes are part of the assembly and transport apparatus, while sfaC and sfaBgenes are regulators. The receptor of the S fimbrial adhesions is α-sialyl (2,3)-β-galactose.

Although both the P and S fimbrial families are recognized as typical extra-intestinal (mainly uropathogenic) adhesive virulence factors, the fact that they can be detected relatively frequently on intestinal isolates, indicates that they mayhave a role in the intestinal colonization of animals (including pigs and calves) and humans.


5.4. Afimbrial adhesins

Afimbrial adhesins (AFA) are the first adhesin structures that are not associated withfimbriae. They were observed for the first time on a uropathogenic E. coli strain(Labigne-Roussel et al., 1984). At present at least eight different afa gene clusters areknown. The afaI gene cluster identified first from E. coli strains associated with urinaryand intestinal infections encodes AfaABCDE proteins and it is involved in adhesion toepithelial cells and haemagglutination (Labigne-Roussel and Falkow, 1988). Three Afaproteins, AfaB, AfaC and AfaE, are required for the mannose-resistant haemagglutina-tion (MRHA) and for adherence to uroepithel cells (Labigne-Roussel et al., 1984;Labigne-Roussel and Falkow, 1988). Among these three proteins AfaA and AfaF aretranscriptional regulators, AfaB functions as a chaperone, AfaC is an outer membraneusher, AfaD is an invasin, and AfaE is the adhesin protein (Walz et al., 1985).

Immunological and DNA hybridization studies revealed the existence of at leastfour afa operons encoding different adhesins in which the afaB, afaC, and afaD genesare highly conserved but the afaE genes (encoding the adhesin proteins) are variable(Labigne-Roussel and Falkow, 1988). All these Afa I–IV variants were identified inhuman UPEC strains but later a hybridization and PCR analysis based study revealedthe existence of related sequences in pathogenic E. coli isolates of bovine and porcineorigin (Harel et al., 1991). Further studies suggested that these operons are differentfrom the afa operons of human isolates (Maiti et al., 1993; Mainil et al., 1997).

Lalioui et al. (1999) cloned and characterized afa-7 and afa-8 gene clustersencoding afimbrial adhesins from diarrhoeagenic and septicaemic E. coli strains ofbovine origin. The AfaE-VII and AfaE-VIII adhesin proteins are genetically differ-ent from the AfaE adhesins produced by human pathogenic strains, and they alsohave different binding specificity. The AfaE adhesins of human pathogenic strainsmediate the MRHA of human erythrocytes and specific attachment to HeLa, uroepithelcells and Caco-2 cells via recognition of the so-called decay-accelerating factor(DAF) molecule as a receptor. AfaE-VII mediates MRHA of human, bovine andporcine erythrocytes and the adhesion of bacteria to HeLa, Caco-2 and uroepithelcell lines, and to MBDK bovine kidney cell line and does not bind to canine kidney.AfaE-VII does not recognize the SCR-3 domain of DAF, which is the receptor ofthe human AfaE adhesins (Nowicki et al., 1993). AfaE-VIII binds to different stillunidentified receptors. In vitro assays showed that it binds to uroepithel cells and tocanine kidney cell line, but does not bind to HeLa and Caco-2 cell lines. AfaE-VIIis slightly similar to fimbrial adhesin AAF/I produced by enteroaggregative E. coliisolates and AfaE-VIII is very similar to the M agglutinin (Lalioui et al., 1999).Further, the afaE-VIII gene is frequent and highly conserved among E. coli strainsisolated from calves, particularly in NTEC strains in association either with the cnf1or the cnf2 gene. The fact that the afa-VIII gene cluster is located on the chromo-some or on the plasmid suggests that it could be carried by a mobile element, facil-itating its dissemination among bovine pathogenic E. coli strains.


6. PRACTICAL APPLICATIONS

The above information on basic mechanisms of pathogenesis of enteric E. coli infec-tions of calves and pigs has led to practical applications mainly in the area of diag-nosis and prevention of these diseases. Unfortunately, specific preventive measureshave only been worked out against ETEC infection of newborn calves and pigs asdiscussed below.

6.1. Diagnosis of enteric E. coli infections

Diagnosis of ETEC infections requires the phenotypic detection of virulence factors(adhesins, enterotoxins) using in vitro tests (slide or latex agglutination or ELISA)in most cases (Thorns et al., 1989). Adhesive fimbriae can, however, be most effi-ciently detected in vivo, by an immunofluorescent method using absorbed poly-clonal or monoclonal antifimbrial antibodies (Isaacson et al., 1978). In contrast tofimbriae, enterotoxins produced in vivo are much more difficult to detect. Therefore,in early ETEC studies in vitro produced toxins could only be tested by biologicalassays: ligated small intestinal segments (for all enterotoxins) or baby mouse assay(for STa), followed by cell cultures (for LT), and later on by ELISA assays (for LTand ST) (Czirok et al., 1992). Now, with the advent of molecular methods in thediagnostic laboratories, the cumbersome biological assays can be replaced by so-called gene probes: DNA hybridization and PCR (recently in a complex form) fordetecting the genes of different virulence characters (Mainil et al., 1990; Francket al., 1998; Tsen and Jian, 1998). The question can be raised, however, of whetherour chances to discover new adhesive and other virulence attributes will not belimited if we disregard classical biological assays in the long run.

6.1.1. Diagnosis in calves

According to our present knowledge, the diagnosis of ETEC infection in calves isgreatly facilitated by the high frequency of K99 antigens on bovine ETEC. The pres-ence of K99 can, however, be covered by the K(A) antigens. Besides, the produc-tion of K99 may also be repressed by the presence of glucose, while for other strainsglucose may even enhance K99 production (Girardeau et al., 1982). Therefore, spe-cial media such as Minimal Casein Agar with Isovitalex® added (MINCA-Is) arerequired (Guinee et al., 1977) for the detection of K99 in vitro. Alternatively, theimmunostaining of small intestinal segments from calves that died as a result ofdiarrhoea proved to be more efficient (Isaacson et al., 1978; Nagy and Nagy, 1982).Monoclonal based latex reagents (Thorns et al., 1989, 1992) and DNA probes(hybridization and PCR) that detect the above fimbrial genes are available for moreefficient diagnosis (Mainil et al., 1990) not only for ETEC but for other pathotypesas well.


6.1.2. Diagnosis in pigs

Piglet diarrhoea is almost always accompanied by some type of non-commensal E. coli infection at the suckling age and within the first 2 weeks after weaning.Today, we already know of several types of porcine ETEC (although it seems thatother pathotypes can also complicate and partly induce diarrhoea in newborn andespecially in weaned pigs). Furthermore, it should be remembered that on the herdlevel, diarrhoeal episodes are infrequently monocausal. The presence of one or more types of ETEC (for example) can often be accompanied by rotaviruses, caliciviruses, coccidia, or by the coronavirus of porcine epidemic diarrhoea (PED)in both age groups but especially in weaned pigs (Hampson, 1994; Nagy et al.,1996b). In this chapter only the diagnosis of infections due to known and establishedtypes of ETEC will be discussed, which are in most cases the dominant elements ofsporadic diarrhoeal diseases on the herd level. Diagnosis of ETEC infection is basedon the detection of known virulence factors (and of the serogroup) of the suspectedETEC. This would not necessarily require culturing of bacteria (see below), but theneed to determine antibiotic resistance patterns simultaneously makes culture andtest of bacterial attributes in vitro an accepted routine for diagnostic laboratories.For cultures, usually small intestinal or faecal samples are available, from which it is advisable to inoculate specific media (besides classical media) required for preferential growth of some adhesins (such as MINCA-Is for K99, or Difco Bloodagar Base with sheep blood for 987P) (Guinee et al., 1977; Nagy et al., 1977). Totest if the isolates are ETEC, the fimbrial antigens K88, K99, F41 and 987P can bedetected by slide agglutination using specific absorbed sera or by latex agglutinationfor which there are monoclonal antibody based kits available (Thorns et al., 1989,1992). Adhesive fimbriae produced in vivo can be more efficiently detected by testing small intestinal smears of diarrhoeal pigs using fluorescence antibodyassays. As there may be ETEC strains without known (or detectable) adhesive virulence factors, it is advisable to perform tests for enterotoxins as well. LT andSTa toxins can be identified by ELISA or by latex agglutination; unfortunately nosuch tests are available for STb. DNA probes (hybridization and PCR) are also inuse for in vitro detection of almost all known virulence genes of porcine ETEC(Mainil et al., 1990; Nagy et al., 1990a; Franck et al., 1998).

Besides bacteriological results, there is almost always a need for differentialdiagnostic investigations (such as virus detection) as well. Therefore, in the case ofweaning pigs it is strongly advised not to be content with a possible bacteriologicalresult detecting some types of ETEC (carrying K88 or F18 surface antigens), but itis also necessary to consider other physiological, environmental, dietary and viralfactors that may sometimes be as important as the given ETEC bacteria themselves.Therefore, differential diagnosis should frequently include the detection of rota- andcoronaviruses as well as spirochaetes and Salmonella (Hampson, 1994; Nagy et al.,1996b). Culturing and/or immunofluorescent in vivo identification of ETEC strainsfrom the ilea of diarrhoeal pigs is the most effective and simplest way of making a

bacteriological diagnosis (as described for diarrhoea of newborns). The bacterio-logical analysis of faecal samples for ETEC is more difficult because the bacteriapresent in the faeces may not reflect the microbial status of the small intestine. Thereare a variety of in vitro techniques that detect virulence factors (adhesins and toxins) of ETEC including immunological and biological assays, molecular probes(DNA hybridization and PCR) as mentioned above for newborn diarrhoea.

6.2. Prevention by vaccination (using adhesins as protective antigens)

ETEC infections can, and should, be prevented by several hygienic and managementtechniques which are outside the scope of this chapter. Among these, the mostimportant factor, in the case of newborn animals, remains the early and sufficientcolostral supply. The protective value of colostrum against diarrhoeal diseases of theoffspring caused by ETEC can be increased essentially by maternal immunization.For that purpose several vaccines are used mainly by parenteral application (whichcan be adjuvanted by oral immunization). These vaccines contain the so-called protective antigens (virulence factors – fimbrial adhesins with or without LT entero-toxins). Vaccinations should usually take place in late pregnancy and can berepeated as “reminder” vaccinations before each subsequent farrowing. As a result,colostral antibodies would block virulence factors and propagation of bacteria in theintestine. Similar effects can be expected in the case of passive immunization, i.e.the oral application of polyclonal or monoclonal antibodies (Sherman et al., 1983).Immune colostrum or specific antibodies can also be applied metaphylactically,however, with much less success. Amongst the mechanisms of action describedabove, the success of colostral vaccines depends largely upon matching the rightprotective antigens with the pathogens present in a given animal population. Ourknowledge about the possible existing virulence factors is, however, still limited andfurther improvements in this area are to be expected.

Vaccines against enterotoxic colibacillosis of calves or small ruminants containboth K99 and F41 (Contrepois et al., 1978; Acres et al., 1979; Nagy, 1980). In coun-tries where F17(FY/Att25) fimbriae are prevalent, vaccines should also contain theF17(FY/Att25) antigens (Contrepois and Girardeau, 1985; Lintermans et al., 1988).As ETEC infections of calves and small ruminants frequently occur simultaneouslywith rotavirus infection, most of the vaccines used today contain bovine rotavirusantigens as well (Bachmann et al., 1984; Köves et al., 1987). So far, no informationis available about a possible shift in fimbrial characteristics of ETEC in herds or areaswhere K99 and/or F41 containing vaccines are used. There is evidence, however, suggesting that the strongly reduced incidence of K99 and F17 may be explainedby the use of vaccines containing these antigens (Contrepois and Guillimin, 1984).During the past decade, no new adhesins or toxins of calf or ruminant ETEC strainswere discovered, although it seems almost impossible that the adhesin (and toxin)spectrum in these animal species is that limited all over the world.



Vaccinations against neonatal diarrhoea of pigs caused by ETEC have been verysuccessful especially since the most prevalent adhesins (K88, K99, 987P) and toxin(LT) became standard components of the vaccines (Moon and Bunn, 1993). It seemsthat LT could act not only as a protective antigen, but also as an oral adjuvant (Ahrenet al., 1998). Such vaccines are almost always used to provide maternal immunitythrough immune colostrum to the offspring. This requires parenteral (or oral) appli-cation of the above antigens well before farrowing. As a result, passively acquiredantibodies through colostrum will protect piglets for about a week against mosttypes of ETEC under normal farming conditions, provided that the piglets ingestimmune colostrum early enough and in an adequate quantity during the first 12 h oflife (before the sharp decline of their absorptive capacity for colostral immuno-globulins). There have been several ways to improve the efficiency of maternal parenteral vaccines against ETEC (Morein et al., 1984; Nagy et al., 1990b).

Some companies advise the use of “in-feed” vaccines (containing killed or livebacteria) for sows or to combine them with parenteral vaccines. The results of Moonet al. (1988) suggest that effective presentation of the protective antigens wouldrequire the use of live oral vaccines for such purposes. Such oral vaccines, iflicensed, could efficiently stimulate the mucosa-associated lymphoid system(GALT) so that secretory antibodies (especially SIgA) – which are protected fromdigestion – could be produced and provide the firmest protection. Strong lactogenicimmunity mediated in this way lasts for about the first 10−14 days of life. It shouldbe noted that first farrowing gilts are less able to produce high levels of antibodieswhatever the route of immunization. The combination of “in-feed” and parenteralvaccines can be recommended for first and second pregnant gilts as well (Moon et al., 1988). It should be remembered, however, that licensing of live oral bacterialvaccines for use in veterinary medicine, especially those produced by genetic engi-neering, is difficult in most countries. Killed oral vaccines are, however, of limitedvalue. Live oral vaccines still represent a more controlled and more effective wayof specific immune prevention of neonatal diarrhoea as compared to the so-called“feed back” (feeding of diarrhoeal faecal material to pregnant sows, as practised onsome farms). The use of recombinant Salmonella-vector vaccines expressing thenecessary adhesive epitopes could also come into question (Attridge et al., 1988;Morona et al., 1994). Finally, it is hoped that more progress in the area of geneti-cally engineered plants (containing the required antigens produced for feeding) willbe made in the future.

Vaccinations against post-weaning diarrhoea of pigs have not shown muchprogress lately, although the theoretical basis is clear and the need is unquestionable.In-feed vaccines containing heat-treated ETEC bacteria have not been consistentlyeffective and most have been removed from the market. Parenteral vaccination ofpiglets before weaning is advised by some companies but its efficacy against PWDhas not been convincingly demonstrated. At present the most promising experimentsare in the area of live oral vaccines applied before weaning. Bertschinger et al.


(1979) demonstrated the efficacy of such a vaccine when a low-energy diet was alsogiven. Further experiments of this group provided evidence about the protection ofpigs against PWD and oedema disease by a live oral vaccine containing F18 fimbria.A combined (live oral plus killed parenteral) vaccine against PWD also seems to besuccessful in preventing losses (Alexa et al., 1995).

7. CONCLUDING REMARKS

Adhesion and colonization are the first (but not the only) functional prerequisites fora mucosal bacterium to be pathogenic. The previous sections have shown the vastgenetic and phenotypic arsenal of adhesins of E. coli bacteria for successful colo-nization of small intestinal mucosal surfaces in calves and young pigs. Theseadhesins represent surface proteins, governed by specific operons and constructedin ways according to the particular adhesin (with fimbrial or afimbrial structures).Beside their structure, these adhesins can also be grouped according to their recep-tors usually present on the intestinal mucosal epithelium (but also on red blood cellsof different animal species and humans) and on the urinary epithelium. Our knowl-edge of the genetics and function of these adhesins has helped so far to reduce lossesdue to enterotoxic colibacillosis of calves and young pigs and may bring further success in the prevention of diseases due to other pathotypes (EHEC, EPEC andNTEC), which at present seem to be a greater threat to human health. Our tools incombating these losses are better and more specific diagnostic reagents (includingDNA-based diagnostic tests) and vaccinations (mainly using the proteinaceousadhesins as protective antigens). The knowledge on genetics of receptors foradhesins of different E. coli pathotypes and subtypes has raised great hopes forbreeding genetically resistant animals – in the case of newborn piglet diarrhoea(receptors for K88) and in the case of weaned pig diarrhoea or oedema (receptorsfor F18). As the classical selection in breeding would not be practical (disadvanta-geous linkage groups with other important genes), it seems that the utilization ofthese genes will have to await further technological developments.


Because E. coli is a highly flexible organism (acquiring new virulence characters ormasking the ones that may be disadvantageous for survival) (Mainil et al., 1987),and because there are several kinds of infections (due to viruses and protozoa asdescribed above) and conditions that may predispose the host to colonization byETEC, thereby enhancing the chances for E. coli to utilize its pathogenic potential,the protection of pigs and calves from pathogenic E. coli is a constant challenge forfarmers and veterinarians alike. As described in the previous sections, the knowl-edge on adhesins and receptors for colonization by different pathotypes of E. colihas been utilized quite extensively for diagnostic purposes (antifimbrial diagnostic


sera and reagents) and for the prevention of diarrhoeal diseases (mainly in the formof killed maternal vaccines containing fimbrial antigens). In the future, furtherapplications can be expected in the development of live oral vaccines (to establishmore efficient local immunity in the intestine). This would imply using non-virulentbut adherent E. coli strains with apropriate adhesins for the species and age of thetarget animal population. Furthermore – in spite of the difficulties described above –progress may also be expected in the area of application of genetic resistance againstenterotoxic colibacillosis of pigs. Apart from direct practical applications, there arefurther significant scientific developments and applications expected in the area of neonatal biology and comparative human pathobacteriology. The most likelyareas for further advancements will be (and in some cases are) the applications of real-time PCR and DNA chip technology in studying quantitative aspects of geneexpression and functional analysis of the genes discussed above. The results of thesestudies will reveal more complex interactions between the pathogenic bacteria andthe host on the gene expression level.

REFERENCES

Acres, S.D., Isaacson, R.E., Babiuk, L.A., Kapitany, R.A., 1979. Immunization of calves against entero-toxigenic colibacillosis by vaccinating dams with purified K99 antigen and whole cell bacterins.Infect. Immun. 25, 121–126.

Adu-Bobie, J., Frankel, G., Bain, C., Goncalves, A.G., Trabulsi, L.R., Douce, G., Knutton, S., Dougan, G.,1998a. Detection of intimins alpha, beta, gamma, and delta, four intimin derivatives expressed byattaching and effacing microbial pathogens. J. Clin. Microbiol. 36, 662–668.

Adu-Bobie, J., Trabulsi, L.R., Carneiro-Sampaio, M.M., Dougan, G., Frankel, G., 1998b. Identificationof immunodominant regions within the C-terminal cell binding domain of intimin alpha and intiminbeta from enteropathogenic Escherichia coli. Infect. Immun. 66, 5643–5649.

Agin, T.S., Wolf, M.K., 1997. Identification of a family of intimins common to Escherichia coli causingattaching-effacing lesions in rabbits, humans, and swine. Infect. Immun. 65, 320–326.

Ahren, C., Jertborn, M., Svennerholm, A., 1998. Intestinal immune responses to an inactivated oralenterotoxigenic Escherichia coli vaccine and associated immunoglobulin A responses in blood.Infect. Immun. 66, 3311–3316.

Alexa, P., Salajka, E., Salajkova, Z., Machova, A., 1995. Combined parenteral and oral immunizationagainst enterotoxigenic Escherichia coli diarrhea in weaned piglets. Vet. Med. Praha 12, 365–370.

An, H., Fairbrother, J.M., Desautels, C., Harel, J., 1999. Distribution of a novel locus called Paa (porcineattaching and effacing associated) among enteric Escherichia coli. Adv. Exp. Med. Biol. 473, 179–184.

An, H., Fairbrother, J.M., Desautels, C., Mabrouk, T., Dugourd, D., Dezfulian, H., Harel, J., 2000.Presence of the LEE (locus of enterocyte effacement) in pig attaching and effacing Escherichia coliand characterization of eae, espA, espB and espD genes of PEPEC (pig EPEC) strain 1390. Microb.Pathog. 28, 291–300.

Attridge, S.R., Hackett, J., Morona, R., Whyte, P., 1988. Towards a live oral vaccine against enterotoxi-genic Escherichia coli of swine. Vaccine 6, 387–389.

Bachmann, P.A., Baljer, G., Gmelch, X., Eichhorn, W., Plank, P., Mayr, A., 1984. Vaccination of cowswith K99 and rotavirus antigen: Potency of K99 antigen combinated with different adjuvants in stim-ulation milk antibody secretion. Zentralbl. Veterinärm. B 31, 660–668.

Baldini, M.M., Kaper, J.B., Levine, M.M., Candy, D.C., Moon, H.W., 1983. Plasmid-mediated adhesionin enteropathogenic Escherichia coli. J. Pediat. Gastroenterol. Nutr. 2, 534–538.

Barnett-Foster, D., Philpott, D., Abul-Milh, M., Huesca, M., Sherman, P.M., Lingwood, C. A., 1999.Phosphatidylethanolamine recognition promotes enteropathogenic E. coli and enterohemorrhagicE. coli host cell attachment. Microb. Pathog. 27, 289–301.

Bertschinger, H.U., Tucker, H., Pfirter, H.P., 1979. Control of Escherichia coli in weaned pigs by use oforal immunization combined with a diet low in nutrients. Fortschr. Veterinärmed. 29, 73–81.

Bertschinger, H.U., Bachmann, M., Mettler, C., Pospischil, A., Schraner, E., Stamm, M., Sadler, T., Wild, P.,1990. Adhesive fimbriae produced in vivo by Escherichia coli O139:K12(B):H1 associated withenterotoxaemia in pigs. Vet. Microbiol. 15, 267–281.

Bertschinger, H.U., Stamm, M., Vogeli, P., 1993. Inheritance of resistance to oedema disease in the pig:experiments with an Escherichia coli strain expressing fimbriae 107. Vet. Microbiol. 35, 79–89.

Bijlsma, I.G., de Nijs, A., van der Meer, C., Frik, J.F., 1982. Different pig phenotypes affect adherence of Escherichia coli to jejunal brush borders by K88ab, K88ac, or K88ad antigen. Infect. Immun. 3, 891–894.

Billey, L.O., Erickson, A.K., Francis, D.H., 1998. Multiple receptors on porcine intestinal epithelial cellsfor the three variants of Escherichia coli K88 fimbrial adhesin. Vet. Microbiol. 59, 203–212.

Blanco, J.E., Blanco, J., Blanco, M., Alonso, M.P., Jansen, W.H., 1994. Serotypes of CNF1-producingEscherichia coli strains that cause extraintestinal infections in humans. Eur. J. Epidemiol. 10, 707–711.

Blanco, M., Blanco, J., Blanco, J.E., Ramos, J., 1993. Enterotoxigenic, verotoxigenic, and necrotoxigenicEscherichia coli isolated from cattle in Spain. Amer. J. Vet. Res. 54, 1446–1451.

Blanco, M., Blanco, J.E., Mora, A., Blanco, J., 1998. Distribution and characterization of faecal necro-toxigenic Escherichia coli CNF1+ and CNF2+ isolated from healthy cows and calves. Vet. Microbiol.59, 183–192.

Brooks, D.E., Cavanagh, J., Jayroe, D., Janzen, J., Snoek, R., Trust, T.J., 1989. Involvement of the MNblood group antigen in shear-enhanced hemagglutination induced by the Escherichia coli F41adhesin. Infect. Immun. 57, 377–383.

Caprioli, A., Falbo, V., Roda, L.G., Ruggeri, F.M., Zona, C., 1983. Partial purification and characteriza-tion of an Escherichia coli toxic factor that induces morphological cell alterations. Infect. Immun. 39,1300–1306.

Caprioli, A., Donelli, G., Falbo, V., Possenti, R., Roda, L.G., Roscetti, G., Ruggeri, F.M., 1984. A celldivision-active protein from E. coli. Biochem. Biophys. Res. Commun. 118, 587–593.

Chanter, N., Morgan, J.H., Bridger, J.C., Hall, G.A., Reynolds, D.J., 1984. Dysentery in gnotobioticcalves caused by atypical Escherichia coli. Vet. Rec. 114, 71.

Chanter, N., Hall, G.A., Bland, A.P., Hayle, A.J., Parsons, K.R., 1986. Dysentery in calves caused by anatypical strain of Escherichia coli (S102-9). Vet. Microbiol. 12, 241–253.

Contrepois, M., Girardeau, J.P., 1985. Additive protective effects of colostral antipili antibodies in calvesexperimentally infected with enterotoxigenic Escherichia coli. Infect. Immun. 50, 947–949.

Contrepois, M., Guillimin, P., 1984. Vaccination anti Escherichia coli K99 chez la chévre. In: Yvore, P.,Perrin, G. (Eds.), Les maladies de la chévre. INRA, Paris, pp. 473–476.

Contrepois, M., Girardeau, J.P., Dubourguier, H.C., Gouet, P., Levieux, D., 1978. Specific protection bycolostrum from cows vaccinated with the K99 antigen in newborn calves experimentally infected withE. coli Ent+ K99+. Ann. Rech. Vet. 9, 385–388.

Contrepois, M., Bertin, Y., Girardeau, J.P., Picard, B., Goullet, P., 1993. Clonal relationships among bovinepathogenic Escherichia coli producing surface antigen CS31A. FEMS Microbiol. Lett. 106, 217–222.

Czirok, E., Semjen, G., Steinruck, H., Herpay, M., Milch, H., Nyomarkay, I., Stverteczky, Z., Szeness, A.,1992. Comparison of rapid methods for detection of heat-labile (LT) and heat-stable (ST) enterotoxinin Escherichia coli. J. Med. Microbiol. 36, 398–402.

Dean, E.A., Whipp, S.C., Moon, H.W., 1989. Age-specific colonization of porcine intestinal epitheliumby 987P-piliated enterotoxigenic Escherichia coli. Infect. Immun. 57, 82–87.

Dean-Nystrom, E.A., Bosworth, B.T., Moon, H.W., O’Brien, A.D., 1998. Bovine infection with Shigatoxin-producing Escherichia coli. In: Kaper, J.B., O’Brien, A.D. (Eds.), Escherichia coli O157:H7and other Shiga Toxin-Producing E. coli Strains. ASM Press, Washington DC.

Deibel, C., Kramer, S., Chakraborty, T., Ebel, F., 1998. EspE, a novel secreted protein of attaching andeffacing bacteria, is directly translocated into infected host cells, where it appears as a tyrosine-phos-phorylated 90 kDa protein. Mol. Microbiol. 28, 463–474.

De Graaf, F.K., 1990. Genetics of adhesive fimbriae of intestinal Escherichia coli. In: Jann, K., Jann, B.(Eds.), Bacterial Adhesins. Curr. Topics Microbiol. Immunol. 151, 29–53.

DeRycke, J., Plassiart, G., 1990. Toxic effects for lambs of cytotoxic necrotising factor from Escherichiacoli. Res. Vet. Sci. 49, 349–354.



DeRycke, J., Milon, A., Oswald, E., 1999. Necrotoxic Escherichia coli (NTEC): two emerging categoriesof human and animal pathogens. Vet. Res. 30, 221–233.

DeVinney, R., Stein, M., Reinscheid, D., Abe, A., Ruschkowski, S., Finlay, B.B., 1999.Enterohemorrhagic Escherichia coli O157:H7 produces Tir, which is translocated to the host cellmembrane but is not tyrosine phosphorylated. Infect. Immun. 67, 2389–2398.

Dozois, C.M., Clement, S., Desautels, C., Oswald, E., Fairbrother, J., 1997. Expression of P, S, and F1Cadhesins by cytotoxic necrotizing factor 1-producing Escherichia coli from septicemic and diarrheicpigs. FEMS Microbiol. Lett. 152, 307–312.

Duchet-Suchaux, M., Le Maitre, C., Bertin, A., 1990. Differences in susceptibility of inbred and outbredinfant mice to enterotoxigenic Escherichia coli of bovine, porcine and human origin. J. Med.Microbiol. 31, 185–190.

Duguid, J.P., Smith, I.W., Dempster, G., 1955. Nonflagellar filamentous appendages (fimbriae) andhaemagglutinating activity in Bacterium coli. J. Pathol. Bacteriol. 53, 335–355.

Duhamel, G.E., Moxley, R.A., Maddox, C.W., Erickson, E.D., 1992. Enteric infection of a goat withenterohemorrhagic Escherichia coli (O103:H2). J. Vet. Diagn. Invest. 4, 197–200.

Elsinghorst, E.A., Weitz, J.A., 1994. Epithelial cell invasion and adherence directed by the enterotoxi-genic Escherichia coli tib locus is associated with a 104-kilodalton outer membrane protein. Infect.Immun. 62, 3463–3471.

Fairbrother, J.M., Lariviere, S., Lallier, R., 1986. New fimbrial antigen F165 from Escherichia coliserogroup O115 strains isolated from piglets with diarrhea. Infect. Immun. 51, 10–15.

Falbo, V., Famiglietti, M., Caprioli, A., 1992. Gene block encoding production of cytotoxic necrotizingfactor 1 and hemolysin in Escherichia coli isolates from extraintestinal infections. Infect. Immun. 60,2182–2187.

Finlay, B.B., Falkow, S., 1997. Common themes in microbial pathogenicity revisited. Microbiol. Mol.Biol. Rev. 61, 136–169.

Fiorentini, C.G., Donelli, P., Matarrese, A., Fabbri, S., Paradisi, P., Boquet, P., 1995. Escherichia colicytotoxic necrotizing factor 1: evidence for induction of actin assembly by constitutive activation ofthe p21 Rho GTPase. Infect. Immun. 63, 3936–3944.

Francis, D.H., Grange, P.A., Zeman, D.H., Baker, D.R., Sun, R., Erickson, A.K., 1998. Expression ofmucin-type glycoprotein K88 receptors strongly correlates with piglet susceptibility to K88(+)enterotoxigenic Escherichia coli, but adhesion of this bacterium to brush borders does not. Infect.Immun. 66, 4050–4055.

Franck, S.M., Bosworth, B.T., Moon, H.W., 1998. Multiplex PCR for enterotoxigenic, attaching and effac-ing, and Shiga toxin-producing Escherichia coli strains from calves. J. Clin. Microbiol. 36, 1795–1797.

Frankel, G., Candy, D.C., Fabiani, E., Adu-Bobie, J., Gil, S., Novakova, M., Phillips, A.D., Dougan, G.,1995. Molecular characterization of a carboxy-terminal eukaryotic-cell-binding domain of intiminfrom enteropathogenic Escherichia coli. Infect. Immun. 63, 4323–4328.

Garabal, J.I., Gonzalez, E.A., Vazquez, F., Blanco, J., Blanco, J.E., 1996. Serogroups of Escherichia coliisolated from piglets in Spain. Vet. Microbiol. 48, 113–123.

Gibbons, R.A., Sellwood, R., Burrows, M., Hunter, P.A., 1977. Inheritance of resistance to neonatal E.coli diarrhoea in the pig. Theor. Appl. Gen. 51, 65–70.

Girardeau, J.P., Dubourguier, H.C., Gouet, P., 1982. Effect of glucose and amino acids on expression ofK99 antigen in Escherichia coli. J. Gen. Microbiol. 128, 2243–2249.

Girardeau, J.P., Der Vartanian, M., Ollier, J.L., Contrepois, M., 1988. CS31A, a new K88-related fimbrialantigen on bovine enterotoxigenic and septicemic Escherichia coli strains. Infect. Immun. 56,2180–2188.

Girardeau, J.P., Bertin, Y., Martin, C., Der Vartanian, M., Boeuf, C., 1991. Sequence analysis of the clpGgene, which codes for surface antigen CS31A subunit: evidence of an evolutionary relationshipbetween CS31A, K88, and F41 subunit genes. J. Bacteriol. 173, 7673–7683.

Giron, J.A., Ho, A.S., Schoolnik, G.K., 1991. An inducible bundle-forming pilus of enteropathogenicEscherichia coli. Science 1254, 710–713.

Goffaux, F., Mainil, J., Pirson, V., Charlier, G., Pohl, P., Jacquemin, E., China, B., 1997. Bovine attach-ing and effacing Escherichia coli possess a pathogenesis island related to the LEE of the humanenteropathogenic Escherichia coli strain E2348/69. FEMS Microbiol. Lett. 154, 415–421.


Guinee, P.A.M., Veldkamp, M.J., Jansen, W.H., 1977. Improved Minca medium for the detection of K99antigen in calf enterotoxigenic strains of Escherichia coli. Infect. Immun. 15, 676–678.

Gunning, R.F., Wales, A.D., Pearson, G.R., Done, E., Cookson, A.L., Woodward, M.J., 2001. Attachingand effacing lesions in the intestines of two calves associated with natural infection with Escherichiacoli O26:H11. Vet. Rec. 148, 780–782.

Gyles, C.E., 1994. Escherichia coli verotoxins and other cytotoxins. In: Gyles, C.E. (Ed.), Escherichiacoli in Domestic Animals and Humans. CAB International, Wallingford, pp. 365–398.

Hacker, J., Schmidt, G., Hughes, C., Knapp, S., Marget, M., Goebel, W., 1985. Cloning and characteri-zation of genes involved in production of mannose-resistant, neuraminidase-susceptible (X) fimbriaefrom a uropathogenic O6:K15:H31 Escherichia coli strain. Infect. Immun. 47, 434–440.

Hacker, J., Kestler H., Hoschutzky, H., Jann, K., Lottspeich, F., Korhonen, T.K., 1993. Cloning and char-acterization of the S fimbrial adhesin II complex of an Escherichia coli O18:K1 meningitis isolate.Infect. Immun. 61, 544–550.

Hacker, J., Blum-Oehler, G., Muhldorfer, I., Tschape, H., 1997. Pathogenicity islands of virulent bacte-ria: structure, function and impact on microbial evolution. Mol. Microbiol. 23, 1089–1097.

Hadad, J.J., Gyles, C.L., 1982. Scanning and transmission electron microscopic study of the small intestine of colostrum-fed calves infected with selected strains of Escherichia coli. Amer. J. Vet. Res.43, 41–49.

Hampson, D.J., 1994. Postweaning E. coli in pigs. In: Gyles, C.E. (Ed.), Escherichia coli in DomesticAnimals and Humans. CAB International, Wallingford, pp. 178–181.

Harel, J., Daigle, F., Maiti, S., Désautels, C., Labigne, A., Fairbrother, J.M., 1991. Occurrence of pap-, sfa-, and afa-related sequences among F165-positive Escherichia coli from diseased animals.FEMS Microbiol. Lett. 66, 177–182.

Helie, P., Morin, M., Jacques, M., Fairbrother, J.M., 1991. Experimental infection of newborn pigs withan attaching and effacing Escherichia coli O45:K“E65” strain. Infect. Immun. 59, 814–821.

Hohmann, A., Wilson, M.R., 1975. Adherence of enteropathogenic Escherichia coli to intestinal epithe-lium in vivo. Infect. Immun. 12, 866–880.

Hultgren, S.J., Jones, C.H., Normark, S., 1996. Bacterial adhesins and their assembly. In: Neidhardt, F.C.(Ed.), Escherichia coli and Salmonella cellular and molecular biology. ASM Press, Washington DC,pp. 2730–2756.

Imberechts, H., Van Pelt, N., De Greve, H., Lintermans, P., 1994. Sequences related to the major subunitgene fedA of F107 fimbriae in porcine Escherichia coli strains that express adhesive fimbriae. FEMSMicrobiol. Lett. 3, 309–314.

Isaacson, R.E., Nagy, B., Moon, H.W., 1977. Colonization of porcine small intestine by Escherichia coli:colonization and adhesion factors of pig enteropathogens that lack K88. J. Infect. Dis. 135, 531–539.

Isaacson, R.E., Moon, H.W., Schneider, R.A., 1978. Distribution and virulence of Escherichia coli in thesmall intestines of calves with and without diarrhea. Amer. J. Vet. Res. 39, 1750–1755.

Janke, B.H., Francis, D.H., Collins, J.E., Libal, M.C., Zeman, D.H., Johnson, D.D., 1989. Attaching andeffacing Escherichia coli infections in calves, pigs, lambs, and dogs. J. Vet. Diagn. Invest. 1, 6–11.

Janke, B.H., Francis, D.H., Collins, J.E., Libal, M.C., Zeman, D.H., Johnson, D.D., Neiger, R.D., 1990.Attaching and effacing Escherichia coli infection as a cause of diarrhoea in young calves. J. Amer.Vet. Med. Assoc. 196, 897–901.

Jerse, A.E., Kaper, J.B., 1991. The eae gene of enteropathogenic Escherichia coli encodes a 94-kilodaltonmembrane protein, the expression of which is influenced by the EAF plasmid. Infect. Immun. 59,4302–4309.

Jerse, A.E., Yu, J., Tall, B.D., Kaper, J.B., 1990. A genetic locus of enteropathogenic Escherichia colinecessary for the production of attaching and effacing lesions on tissue culture cells. Proc. Natl. Acad.Sci. USA 87, 7839–7843.

Kallenius, G., Mollby, R., Svenson, S.B., Helin, I., Hultberg, H., Cedergren, B., Winberg, J., 1981.Occurrence of P-fimbriated Escherichia coli in urinary tract infections. Lancet 2, 1369–1372.

Kennan, R.M., Monckton, R.P., 1990. Adhesive fimbriae associated with porcine enterotoxigenicEscherichia coli of the O141 serotype. J. Clin. Microbiol. 28, 2006–2011.

Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.J., Frey, E.A., Finlay, B.B., 1997. EnteropathogenicE. coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell 91, 511–520.


Khan, A.S., Muhldorfer, I., Demuth, V., Wallner, U., Korhonen, T.K., Hacker, J., 2000. Functional analy-sis of the minor subunits of S fimbrial adhesion (SfaI) in pathogenic Escherichia coli. Mol. Gen.Genet. 263, 96–105.

Kim, J., Nietfeldt, J., Benson, A.K., 1999. Octamer-based genome scanning distinguishes a unique sub-population of Escherichia coli O157:H7 strains in cattle. Proc. Natl. Acad. Sci. USA 96, 13 288–13 293.

Knutton, S., Lloyd, D.R., McNeish, A.S., 1987. Adhesion of enteropathogenic Escherichia coli to humanintestinal enterocytes and cultured human intestinal mucosa. Infect. Immun. 55, 69–77.

Knutton, S., Rosenshine, I., Pallen, M.J., Nisan, I., Neves, B.C., Bain, C., Wolff, C., Dougan, G., Frankel, G.,1998. A novel EspA-associated surface organelle of enteropathogenic Escherichia coli involved inprotein translocation into epithelial cells. EMBO J. 17, 2166–2176.

Konowalchuk, J., Speirs, J.I., Stavric, S., 1977. Vero response to a cytotoxin of Escherichia coli. Infect.Immun. 18, 775–779.

Köves, B., Nagy, B., Kulcsar, A., 1987. Development of a bovine combined Rotavirus-Escherichia colivaccine (in German). Arch. Exp. Veterinärmed. 41, 904–907.

Kuehn, M.J., Heuser, J., Normark, S., Hultgren, S.J., 1992. P pili in uropathogenic E. coli are compositefibres with distinct fibrillar adhesive tips. Nature 356, 252–255.

Kupersztoch, Y.M., Tachias, K., Moomaw, C.R., Dreyfus, L.A., Urban, R., Slaughter, C., Whipp, S., 1990.Secretion of methanol-insoluble heat-stable enterotoxin (STb): energy- and secA-dependent conver-sion of pre-STb to an intermediate indistinguishable from the extracellular toxin. J. Bacteriol. 172,2427–2432.

Labigne-Roussel, A.F., Falkow, S., 1988. Distribution and degree of heterogeneity of the afimbrial-adhesin-encoding operon (afa) among uropathogenic Escherichia coli isolates. Infect. Immun. 56, 640–648.

Labigne-Roussel, A.F., Lark, D., Schoolnik, G., Falkow, S., 1984. Cloning and expression of an afimbr-ial adhesin (AFA-I) responsible for P blood group-independent, mannose-resistant hemagglutinationfrom a pyelonephritic Escherichia coli strain. Infect. Immun. 46, 251–259.

Lalioui, L., Jouve, M., Gounon, P., Le Bouguenec, C., 1999. Molecular cloning and characterization ofthe afa-7 and afa-8 gene clusters encoding afimbrial adhesins in Escherichia coli strains associatedwith diarrhea or septicemia in calves. Infect. Immun. 67, 5048–5059.

Le Bougouénec, C., Bertin, Y., 1999. AFA and F17 adhesins produced by pathogenic Escherichia colistrains in domestic animals. Vet. Res. 30, 317–342.

Levine, M.M., Nataro, J.P., Karch, H., Baldini, M.M., Kaper, J.B., Black, R.E., Clements, M.L., O’Brien,A.D., 1985. The diarrheal response of humans to some classic serotypes of enteropathogenic Escherichiacoli is dependent on a plasmid encoding an enteroadhesiveness factor. J. Infect. Dis. 152, 550–559.

Lintermans, P.F., Pohl, P., Bertels, A., Charlier, G., Vandereckerchove, J., Van Damme, J., Schoup, J.,Schlickere, C., Korhouen, T., De Greve, H., Van Montagu, M., 1988. Characterisation and purifica-tion of the FY Adhesin present on the surface of bovine enteropathogenic and septicemic Escherichiacoli. Amer. J. Vet. Res. 49, 1794–1799.

Liu, H., Magoun, L., Luperchio, S., Schauer, D.B., Leong, J.M., 1999. The Tir-binding region of entero-haemorrhagic Escherichia coli intimin is sufficient to trigger actin condensation after bacterial-induced host cell signalling. Mol. Microbiol. 34, 67–81.

Mainil, J., 1999. Shiga/verocytotoxins and Shiga/verotoxigenic Escherichia coli in animals. Vet. Res.30, 235–257.

Mainil, J.G., Sadowski, P.L., Tarsio, M., Moon, H.W., 1987. In vivo emergence of enterotoxigenicEscherichia coli variants lacking genes for K99 fimbriae and heat-stable enterotoxin. Infect. Immun.55, 3111–3116.

Mainil, J.G., Bex, F., Jacquemin, E., Pohl, P., Couturier, M., Kaeckenbeeck, A., 1990. Prevalence of fourenterotoxin (STaP, STaH, STb, and LT) and four adhesin subunit (K99, K88, 987P, and F41) genesamong Escherichia coli isolates from cattle. Amer. J. Vet. Res. 51, 187–190.

Mainil, J.G., Jacquemin, E., Herault, F., Oswald, E., 1997. Presence of pap-, sfa-, and afa-relatedsequences in necrotoxigenic Escherichia coli isolates from cattle: evidence for new variants of theAFA family. Can. J. Vet. Res. 61, 193–199.

Mainil, J.G., Jacquemin, E., Pohl, P., Fairbrother, J.M., Ansuini, A., Le Bouguenec, C., Ball, H.J.,De Rycke, J., Oswald, E., 1999. Comparison of necrotoxigenic Escherichia coli isolates from farmanimals and from humans. Vet. Microbiol. 70, 123–135.


Maiti, S.N., Harel, J., Fairbrother, J.M., 1993. Structure and copy number analyses of pap-, sfa-, and afa-related gene clusters in F165-positive bovine and porcine Escherichia coli isolates. Infect. Immun. 61,2453–2461.

McDaniel, T.K., Jarvis, K.G., Donnenberg, M.S., Kaper, J.B., 1995. A genetic locus of enterocyte efface-ment conserved among diverse enterobacterial pathogens. Proc. Natl. Acad. Sci. USA 92, 1664–1668.

McNally, A., Roe, A.J., Simpson, S., Thomson-Carter, F.M., Hoey, D.E., Currie, C., Chakraborty, T.,Smith, D.G., Gally, D.L., 2001. Differences in levels of secreted locus of enterocyte effacement proteins between human disease-associated and bovine Escherichia coli O157. Infect. Immun. 69,5107–5114.

Mecsas, J.J., Strauss, E.J., 1996. Molecular mechanisms of bacterial virulence: type III secretion andpathogenicity islands. Emerg. Infect. Dis. 2, 270–288.

Moch, T., Hoschutzky, H., Hacker, J., Kroncke, K.D., Jann, K., 1987. Isolation and characterization ofthe alpha-sialyl-beta-2,3-galactosyl-specific adhesin from fimbriated Escherichia coli. Proc. Natl.Acad. Sci. USA 84, 3462–3466.

Moon, H.W., 1990. Colonization factor antigens of enterotoxigenic Escherichia coli in animals. In: Jann, K.,Jann, B. (Eds.), Bacterial Adhesins. Curr. Topics Microbiol. Immunol. 151, 147–165.

Moon, H.W., Bunn, T.O., 1993. Vaccines for preventing enterotoxigenic Escherichia coli infections infarm animals. Vaccine 11, 213–219.

Moon, H.W., Whipp, S.C., Argenzio, R.A., Levine, M.M., Giannella, R.A., 1983. Attaching and effacingactivities of rabbit and human enteropathogenic Escherichia coli in pig and rabbit intestines. Infect.Immun. 41, 1340–1351.

Moon, H.W., Rogers, D.G., Rose, R., 1988. Effects of an orally administered live Escherichia coli pilusvaccine on duration of lacteal immunity to enterotoxigenic Escherichia coli in swine. Amer. J. Vet.Res. 49, 2068–2071.

Morein, B., Sundquist, B., Hoglund, S., Dalsgaard, K., Osterhaus, A., 1984. ISCOM, a novel structurefor antigenic presentation of membrane proteins from enveloped viruses. Nature 308, 457–460.

Morona, R., Morona, J.K., Considine, A., Hackett, J.A., van den Bosch, L., Beyer, L., Attridge, S.R.,1994. Construction of K88- and K99-expressing clones of Salmonella typhimurium G30: immuno-genicity following oral administration to pigs. Vaccine 12, 513–517.

Mouricout, M.A., Julien, R.A., 1987. Pilus-mediated binding of bovine enterotoxigenic Escherichia colito calf small intestinal mucins. Infect. Immun. 55, 1216–1223.

Nagy, B., 1980. Vaccination of cows with a K99 extract to protect newborn calves against experimentalenterotoxic colibacillosis. Infect. Immun. 27, 21–24.

Nagy, B., 1993. Pathogenesis and prevention of enteric infection of calves and pigs (in Hungarian).Unpublished D.Sc. Thesis, Hungarian Academy of Sciences, Budapest.

Nagy, B., Nagy, Gy, 1982. Detection of virulence factors of enteropathogenic E. coli strains in the diag-nosis of calf diarrhoea (in Hungarian). Magyar Állatorvosok Lapja 37, 512–517.

Nagy, B., Moon, H.W., Isaacson, R.E., 1977. Colonization of porcine intestine by enterotoxigenicEscherichia coli: selection of piliated forms in vivo, adhesion of piliated forms to epithelial cells in vitro, and incidence of a pilus antigen among porcine enteropathogenic E. coli. Infect. Immun. 16,344–352.

Nagy, B., Bozso, M., Palfi, V., Nagy, Gy., Sahiby, M.A., 1984. Studies on cryptosporidial infection ofgoat kids. In: Yvore, P., Perrin, G. (Eds.), Les maladies de la chévre. INRA, Paris, pp. 443–451.

Nagy, B., Casey, T.A., Moon, H.W., 1990a. Phenotype and genotype of Escherichia coli isolated frompigs with postweaning diarrhea in Hungary. J. Clin. Microbiol. 28, 651–653.

Nagy, B., Hoglund, S., Morein, B., 1990b. Iscom (immunostimulating complex) vaccines containingmono- or polyvalent pili of enterotoxigenic E. coli; immune response of rabbit and swine. Zentralbl.Veterinärmed. 37, 728–738.

Nagy, B., Casey, T.A., Whipp, S.C., Moon, H.W., Dean, E.A., 1992a. Pili and adhesiveness of porcinepostweaning enterotoxigenic and verotoxigenic Escherichia coli. In: Proceedings of the 12thCongress Int. Pig Vet. Soc., The Hague, The Netherlands, p. 240.

Nagy, B., Casey, T.A., Whipp, S.C., Moon, H.W., 1992b. Susceptibility of porcine intestine to pilus-medi-ated adhesion by some isolates of piliated enterotoxigenic Escherichia coli increases with age. Infect.Immun. 60, 1285–1294.


Nagy, B., Awad-Masalmeh, M., Bodoky, T., Munch, P., Szekrenyi, M.T., 1996a. Association of shiga-liketoxin type II (SLTII) and heat stable enterotoxins with F18ab, F18ac, K88 and F41 fimbriae ofEscherichia coli from weaned pigs. In: Proceedings of the 14th Congress Int. Pig Vet. Soc. Bologna,Italy, p. 264.

Nagy, B., Nagy, G., Meder, M., Mocsári, E., 1996b. Enterotoxigenic Escherichia coli, rotavirus, porcineepidemic diarrhoea virus, adenovirus and calici-like virus in porcine postweaning diarrhoea inHungary. Acta Vet. Hung. 44, 9–19.

Nagy, B., Whipp, S.C., Imberechts, H., Bertschinger, H.U., Dean-Nystrom, E.A., Casey, T.A., Salajka, E.,1997. Biological relationship between F18ab and F18ac fimbriae of enterotoxigenic and verotoxigenicEscherichia coli from weaned pigs with oedema disease or diarrhoea. Microb. Pathog. 22, 1–11.

Nakazawa, M., Sugimeto, C., Isayama, Y., Kashiwazaki, M., 1987. Virulence factors of Escherichia coliisolated from piglets with neonatal and postweaning diarrhea. Jpn. Vet. Microbiol. 13, 291–300.

Nataro, J.P., Kaper, J.B., 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11, 142–201.Nowicki, B., Hart, A., Coyne, K.E., Lublin, D.M., Nowicki, S., 1993. Short consensus repeat-3 domain

of recombinant decay-accelerating factor is recognized by Escherichia coli recombinant Dr adhesinin a model of a cell-cell interaction. J. Exp. Med. 178, 2115–2121.

O’Brien, A.D., Holmes, R.K., 1996. Protein toxins of Escherichia coli and Salmonella. In: Neidhart, F.C.,Curtiss, III R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Reznikoff, W.S., Riley, M.,Schaechter, M., Umbarger, H.E. (Eds.), Escherichia coli and Salmonella: Cellular and MolecularBiology. ASM Press, Washington D.C., pp. 2788–2802.

O’Brien, A.D., Thompson, M.R., Formal, S.B., 1977. Production of Shigella dysenteriae-like toxin bypathogenic Escherichia coli. In: Abstracts of the Annual Meeting of the American Society forMicrobiology. ASM, Washington DC (Abstr.) B103.

Ørskov, I., Ørskov, F., 1983. Serology of Escherichia coli fimbriae. Prog. Allergy 33, 80–105.Osek, J., 2001. Identification of eae genes in Escherichia coli strains isolated from pigs with postwean-

ing diarrhoea. Vet. Rec. 148, 241–243.Oswald, E., Sugai, M., Labigne, A., Wu, H.C., Fiorentini, C., Boquet, P., O’Brien, A.D., 1994. Cytotoxic

necrotizing factor type 2 produced by virulent Escherichia coli modifies the small GTP-binding pro-teins Rho involved in assembly of actin stress fibers. Proc. Natl. Acad. Sci. USA 91, 3814–3818.

Oswald, E., Schmidt, H., Morabito, S., Karch, H., Marches, O., Caprioli, A., 2000. Typing of intimingenes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characteriza-tion of a new intimin variant. Infect. Immun. 68, 64–71.

Pearson, G.R., Watson, C.A., Hall, G.A., Wray, C., 1989. Natural infection with an attaching and effac-ing Escherichia coli in the small and large intestines of a calf with diarrhoea. Vet. Rec. 124, 297–299.

Peres, S.Y., Marches, O., Daigle, F., Nougayrede, J.P., Herault, F., Tasca, C., DeRycke, J., Oswald, E.,1997. A new cytolethal distending toxin (CDT) from Escherichia coli producing CNF2 blocks HeLacell division in G2/M phase. Mol. Microbiol. 24, 1095–1107.

Pohl, P., Lintermans, P., Moury, J., 1986. Virulence factors of septicaemic and commensal E. coli incalves (in French). Ann. Méd. Vét. 130, 515–520.

Rapacz, J., Hasler-Rapacz, J., 1986. Polymorphism and inheritance of swine small intestinal receptorsmediating adhesion of three serological variants of Escherichia coli-producing K88 pilus antigen.Anim. Genet. 17, 305–321.

Rippinger, P., Bertschinger, H.U., Imberechts, H., Nagy, B., Sorg, I., Stamm, M., Wild, P., Wittig, W.,1995. Designations F18ab and F18ac for the related fimbrial types F107, 2134P and 8813 ofEscherichia coli isolated from porcine postweaning diarrhoea and from oedema disease. Vet.Microbiol. 45, 281–295.

Rosenshine, I., Ruschkowski, S., Finlay, B.B., 1996. Expression of attaching/effacing activity byenteropathogenic Escherichia coli depends on growth phase, temperature, and protein synthesis uponcontact with epithelial cells. Infect. Immun. 64, 966–973.

Runnels, P.L., Moon, H.W., Schneider, R.A., 1980. Development of resistance with host age to adhesionof K99+ Escherichia coli to isolated intestinal epithelial cells. Infect. Immun. 28, 298–300.

Salajka, E., Salajkova, Z., Alexa, P., Hornich, M., 1992. Colonization factor different from K88, K99, F41and 987P in enterotoxigenic Escherichia coli strains isolated from postweaning diarrhoea in pigs. Vet.Microbiol. 32, 163–175.


Schmoll, T., Morschhauser, J., Ott, M., Ludwig, B., van Die, I., Hacker, J., 1990. Complete genetic organ-ization and functional aspects of the Escherichia coli S fimbrial adhesion determinant: nucleotidesequence of the genes sfa B, C, D, E, F. Microb. Pathog. 9, 331–343.

Sellwood, R., Gibbons, R.A., Jones, G.W., Rutter, J.M., 1975. Adhesion of enteropathogenic Escherichiacoli to pig intestinal brush borders: the existence of two pig phenotypes. J. Med. Microbiol. 8, 405–411.

Sherman, D.M., Acres, S.D., Sadowski, P.L., Springer, J.A., Bray, B., Raybould, T.J., Muscoplat, C.C.,1983. Protection of calves against fatal enteric colibacillosis by orally administered Escherichia coliK99-specific monoclonal antibody. Infect. Immun. 42, 653–658.

Smith, H.W., 1974. Search for transmissible pathogenic characters in invasive strains of Escherichia coli:the discovery of a plasmid toxin and a plasmid lethal character closely associated, or identical, withColicine V. J. Gen. Microbiol. 83, 95–111.

Smith, H.W., Linggood, M.A., 1971. Observations on the pathogenic properties of the K88, Hly and Ent plasmids of Escherichia coli with particular reference to porcine diarrhoea. J. Med. Microbiol. 4, 467–485.

Sperandio, V., da Silveira, W.D., 1993. Comparison between enterotoxigenic Escherichia coli strainsexpressing “F42,” F41 and K99 colonization factors. Microbiol. Immunol. 37, 869–875.

Teneberg, S., Willemsen, P.T., de Graaf, F.K., Stenhagen, G., Pimlott, W., Jovall, P.A., Angstrom, J.,Karlsson, K.A., 1994. Characterization of gangliosides of epithelial cells of calf small intestine, withspecial reference to receptor-active sequences for enteropathogenic Escherichia coli K99. J. Biochem.(Tokyo) 116, 560–574.

Thorns, C.J., Sojka, M.G., Roeder, P.L., 1989. Detection of fimbrial adhesins of ETEC using monoclonalantibody-based latex reagents. Vet. Rec. 125, 91–92.

Thorns, C.J., Bell, M.M., Chasey, D., Chesham, J., Roeder, P.L., 1992. Development of monoclonal anti-body ELISA for simultaneous detection of bovine coronavirus, rotavirus serogroup A, andEscherichia coli K99 antigen in feces of calves. Amer. J. Vet. Res. 53, 36–43.

Tóth, I., Oswald, E., Mainil, J.G., Awad-Masalmeh, M., Nagy, B., 2000. Characterization of intestinalcnf1+ Escherichia coli from weaned pigs. Int. J. Med. Microbiol. 290, 539–542.

Tsen, H.Y., Jian, L.Z., 1998. Development and use of a multiplex PCR system for the rapid screening ofheat labile toxin I, heat stable toxin II and shiga-like toxin I and II genes of Escherichia coli in water.J. Appl. Microbiol. 84, 585–592.

Tzipori, S., Gunzer, F., Donnenberg, M.S., de Montigny, L., Kaper, J.B., Donohue-Rolfe, A., 1995. Therole of the eaeA gene in diarrhea and neurological complications in a gnotobiotic piglet model ofenterohemorrhagic Escherichia coli infection. Infect. Immun. 63, 3621–3627.

Van Bost, S., Roels, S., Mainil, J., 2001. Necrotoxigenic Escherichia coli type-2 invade and cause diarrhoeaduring experimental infection in colostrum-restricted newborn calves. Vet. Microbiol. 81, 315–329.

Vazquez, F., Gonzalez, E.A., Garabal, J.I., Blanco, J., 1996. Fimbriae extracts from enterotoxigenicEscherichia coli strains of bovine and porcine origin with K99 and/or F41 antigens. Vet. Microbiol.48, 231–241.

Vogeli, P., Bolt, R., Fries, R., Stranzinger, G., 1994. Co-segregation of the malignant hyperthermia andthe Arg615-Cys615 mutation in the skeletal muscle calcium release channel protein in five EuropeanLandrace and Pietrain pig breeds. Anim. Genet. 1, 59–66.

Walz, W., Schmidt, M.A., Labigne-Roussel, A.F., Falkow, S., Schoolnik, G., 1985. AFA-I, a cloned afim-brial X-type adhesin from a human pyelonephritic Escherichia coli strain. Purification and chemical,functional and serologic characterization. Eur. J. Biochem. 152, 315–321.

Wieler, H.L., 1996. Determination of virulence factors in bovine Shiga-toxin-producing Escherichia coli-(STEC-) strains and a critical review of their clinical relevance to bovines and man. Dissertation (inGerman), Justus Liebig Universität, Giessen. ISBN 3-8265-2307-5.

Willemsen, P.T., de Graaf, F.K., 1993. Multivalent binding of K99 fimbriae to the N-glycolyl-GM3 gan-glioside receptor. Infect. Immun. 61, 4518–4522.

Wittig, W., Prager, R., Stamm, M., Streckel, W., Tschape, H., 1994. Expression and plasmid transfer ofgenes coding for the fimbrial antigen F107 in porcine Escherichia coli strains. Int. J. Med. Microbiol.Virol. Parasitol. Infect. Dis. 281, 130–139.

Woodward, M.J., Wray, C., 1990. Nine DNA probes for detection of toxin and adhesin genes inEscherichia coli isolated from diarrhoeal disease in animals. Vet. Microbiol. 25, 55–65.


Yamanaka, H., Nomura, T., Fujii, Y., Okamoto, K., 1997. Extracellular secretion of Escherichia coli heat-stable enterotoxin I across the outer membrane. J. Bacteriol. 179, 3383–3390.

Zhu, C., Harel, J., Jacques, M., Desautels, C., Donnenberg, M.S., Beaudry, M., Fairbrother, J.M., 1994.Virulence properties and attaching-effacing activity of Escherichia coli O45 from swine postweaningdiarrhea. Infect. Immun. 62, 4153–4159.

Zhu, C., Harel, J., Dumas, F., Fairbrother, J.M., 1995. Identification of EaeA protein in the outer mem-brane of attaching and effacing Escherichia coli O45 from pigs. FEMS Microbiol. Lett. 129,237–242.

191

Food and food animals are often associated with the transfer of antibiotic resistancesto humans. The cause of this is frequently ascribed to the use of antibiotics in ani-mal husbandry. However, not only the rearing of animals, but also the slaughteringand processing as well as the further preparation are important factors for the spreadof resistances via microorganisms. There are three basic routes for the transfer ofantibiotic resistances via food or food animals. The first is a direct contribution viaantibiotic residues or chemotherapeutics in foods. However, based on actual data,this route is of negligible importance. The second route is by ingestion of commen-sal bacteria that can transfer their antibiotic resistance genes to pathogens in thehuman gastrointestinal tract. One example of these is the enterococci that are con-sidered to be part of the normal flora of a variety of foodstuffs. Glycopeptide resist-ant enterococci can transfer this resistance and thus are potential causative agents forcomplications in human infections. Enterococci in foods can be glycopeptide resist-ant but differ from clinical isolates in phenotypic as well as genotypic properties.There is no direct correlation with glycopeptide resistant enterococci from food andhuman disease but one should keep a close eye on their ability to transfer this kindof resistance to other bacteria. The third route is the ingestion of already resistant,real pathogens such as Campylobacter spp. and salmonellae. For this route, fluoro-quinolone resistance is becoming increasingly important as this antibiotic is alsoused in animal husbandry at therapeutic levels and a connection between the two

9 The farm animal as potential reservoir of antibiotic resistant bacteria in thefood chain

G. Kleina and C.M.A.P. Franzb

aInstitute for Food Quality and Food Safety, School of Veterinary MedicineHannover, Bischofsholer Damm 15, D-30173 Hannover, GermanybFederal Research Centre for Nutrition, Institute of Biotechnology andMolecular Biology, D-76131 Karlsruhe, Germany



is possible. Of further importance is the detection of new resistance variants and theoccurrence of new resistant clones. Molecular biological techniques are well suitedfor this. Possible preventative measures include the ban of antibiotics as growthpromoters and the success of such policies are currently being evaluated. In addition,a ban or strong reduction of therapeutic antibiotics is also being considered; however,this raises questions of therapeutic emergencies in animal husbandry (Bogaard andStobberingh, 2001) and a balanced solution must be found. Some possible avenues forreaching such a balance include the classification of therapeutic agents and a represen-tative monitoring system to determine the status quo, as well as new resistance devel-opments. Finally, for selection of technologically used strains (e.g., enterococci) it isimportant to determine their resistance profile as well as their potential for transfer.

1. INTRODUCTION

One of the most important and, ironically, accidental discoveries in medical historywas that of penicillin by Alexander Fleming in 1928. Penicillin and subsequentlydiscovered naturally occurring substances that killed bacteria were termed anti-biotics, a definition that was later broadened to include chemically derived, syn-thetic antibacterial drugs (Walsh, 2003). The antibiotic era started in the 1940s,when penicillin saved countless lives of soldiers in World War II. The success ofpenicillin in saving human life soon led to the discovery (natural antibiotics) anddevelopment (synthetic antibiotics) of other antibiotics (Walsh, 2003).

It has been estimated that since the late 1940s mankind has released 109 – 1010 kgof antimicrobial agents into the environment (Davies, 1996). We know all too wellthat bacteria have survived this onslaught because of their ability to mutate rapidlyand, more importantly, to inherit, express and disseminate genetic material encodingantibiotic resistance (Davies, 1996; Khachatourians, 1998; Teuber et al., 1999;SCOPE, 2002; Walsh, 2003). The past two decades have seen a dramatic develop-ment in infections of bacterial aetiology, i.e., the rise of antibiotic resistant bacteriathat once were susceptible to treatment and now have developed resistance to thesemedications. This has led to an alarming increase in fatalities from gonorrhoea, pneumonia, tuberculosis, meningitis, dysentery, septicaemia, endocarditis and otherinfections. The reason for this is considered to be two-fold: 1) the remarkable geneticplasticity of bacteria to develop resistance to antibiotics, and 2) the abuse and misuseof antibiotics.

2. DEVELOPMENT OF BACTERIAL ANTIBIOTIC RESISTANCE

Essentially, antibiotic resistance is the result of genetic change in the microorgan-ism, either by mutation (a chromosomal change) or by genetic transfer (acquisitionof extrachromosomal genes) (Walsh, 2003). The resulting resistances are oftenreferred to as intrinsic or acquired, respectively. New phenotypic traits of bacteria

192 G. Klein and C.M.A.P. Franz

result from mutations occurring in their genetic material. Mutations result from achange in one or more nucleotide basepairs in the chromosomal DNA. In the caseof spontaneous mutations, they occur naturally at a rate of about 10−7–10−11 per generation and result from errors in DNA replication. Although this rate appears tobe low, it actually is quite the contrary, when considering that a bacterium such asEscherichia (E.) coli can produce 20 generations in about 7 h. Induced mutations,on the other hand, occur at a far greater frequency than spontaneous mutations andare caused by external factors such as chemical agents (e.g., antibiotics), heat, orirradiation (Madigan, 1997). Although induced mutations are considered randomevents, the likelihood of mutation increases when the challenging agent is of limitedstrength and is applied over a prolonged period. Once developed, a mutant trait issubsequently passed on to all succeeding progeny.

It is now well known that extrachromosomal genetic material containing geneswhich encode, for example, antibiotic resistances can be exchanged by bacteria.Such extrachromosomal genetic material exists in the form of plasmids (covalentlyclosed, circular DNA molecules that reside in the chromosome and replicate independently of the chromosome) or in the form of transposons. Transposons arealso mobile genetic elements that can translocate within or between larger piecesof DNA such as plasmids or the chromosome. Transposons themselves often carryall possible combinations of antibiotic resistance genes (Clewell, 1990; Teuber et al.,1999). Plasmid mediated resistance is a tremendous clinical problem because it concerns most bacterial species and they often mediate multidrug resistance and canhave a high rate of transfer. Thus, these days it is generally accepted that there is wide-spread transfer of genes (also antibiotic resistance genes), i.e., either vertical transfer(to progeny) as well as horizontal transfer (to other genera, species or strains).

Antibiotic usage creates a phenomenon known as selective pressure, in whichsusceptible bacteria are killed and the more resistant bacteria survive. Because of theability of bacteria to multiply rapidly, those that survive can give rise to millions ofprogeny, all containing the genes for resistance to the antibiotic. These resistancegenes may confer cross-resistance to other antibiotics and more resistance genesmay also be acquired. Bacterial strains that are resistant to three or more antibioticagents with different mechanisms of inhibition are defined as multidrug resistant.

3. ANTIBIOTIC USE AND RESISTANCE IN ANIMALS

The production of meat, milk and eggs has since World War II been characterizedby greater intensity (i.e., fewer but larger farms) and high scales of production(McEwen and Fedorka-Cray, 2002) so that in modern agriculture it has attainedindustrial dimensions (Teuber, 1999, 2001). Antimicrobials, including antibiotics,are used in food animals to prevent or treat disease or to promote growth (table 1).Therapeutic use is when animals are diseased and antibiotics are administered tocure the infection. In food animal production, the afflicted individual may be

Antibiotic resistant bacteria in the food chain 193

treated, but it is often more efficient to treat entire groups by medication of feed andwater (McEwen and Fedorka-Cray, 2002). For farming of some animals (i.e., poul-try and fish), mass medication is the only feasible means of treatment. Thus certainmass-medication procedures, called “metaphylaxis”, aim to treat sick animals whilemedicating others in the group to prevent disease (McEwen and Fedorka-Cray,2002) (table 1).

Antimicrobials approved for use in the United States in food animals either fortreatment of various infections or for growth promotion are shown in table 2.


Table 1. Types of antimicrobials used in animals for food production

Type of Route or vehicle of Administration to antimicrobial use Purpose administration individuals or groups

Therapeutic Therapy Injection, feed, water Individual or group“Metaphylactic” Disease prophylaxis, Injection, feed, water Group

therapyProphylactic Disease prevention Feed GroupSubtherapeutic Growth promotion Feed Group

feed efficiency, disease prophylaxis

Adapted from McEwen and Fedorka-Cray (1999).

Table 2. Examples of antimicrobials used in food animals in the United States

Purpose Cattle Swine Poultry Fish

Treatment of Amoxicillin Amoxicillin Erythromycin Ormetopriminfections Cephapirin Ampicillin Fluoroquinolone Sulfonamides

Erythromycin Chlortetracycline Gentamicin OxytetracyclineFluoroquinolone Gentamicin NeomycinGentamicin Lincomycin PenicillinNovobiocin Sulfamethazine SpectinomycinPenicillin Tiamulin TetracyclinesSulfanomides Tylosin TylosinTilmicosin VirginiamycinTylosin

Growth and Bacitracin Asanilic acid Bambermycinfeed efficiency Chlortetracycline Bacitracin Bacitracin

Lasalocid Bambermycin ChlortetracyclineMonensin Chlortetracycline PenicillinOxytetracycline Erythromycin Tylosin

Penicillin VirginiamycinTiamulinTylosinVirginiamycin

Adapted from McEwen and Fedorka-Cray (1999).

Growth promoters are usually administered in relatively low concentrations, rang-ing from 2.5 to 125 mg/kg (ppm), depending on the antibiotic used and the speciestreated (Visek, 1978; Jukes, 1986; McEwen and Fedorka-Cray, 2002). High energyfeed for meat and dairy cattle, sheep and goats may be supplemented with 35–100 mgof bacitracin, chlortetracycline or erythromycin per head per day, or 7–140 g oftylosin or neomycin per ton of feed. For swine, 2–500 g of bacitracin, chlortetra-cycline, erythromycin, lincomycin, neomycin, oxytetracycline, penicillin, strepto-mycin, tylosin or virginiamycin may be added to each ton of feed; the same agentsare used for poultry, but at 1–400 g per ton of feed (Khachatourians, 1998). InEurope (i.e., the EU) and for meat intended for import to the EU these substancesare not allowed as growth promoters. Most substances have been banned in the EUrecently, with the exception of salinomycin (swine), avilamycin (swine), flavomycin(laying hens, turkeys, swine, calves, cattle for meat production), and monensin (cat-tle for meat production). There are efforts to prohibit the use of these substances intotal in the EU.

The scale of agricultural use of antibiotics in the US is about 100 to 1000 timesgreater than that for human use (Feinman, 1998; Khachatourians, 1998; Levy, 1998;Witte, 1998). Overall, the annual estimates for application of antibiotics in the USagrifood industry are over 8 million kg for animals (Khachatourians, 1998). In 1997in the EU 5400 tons of antimicrobials were used in human medicine, while 3494 tonsand 1599 tons were used in animal medicine and as growth promoters, respectively(Ungemach, 1999). A breakdown of antimicrobials used for animals in the EU in1997 is shown in table 3, from which it is clear that tetracycline was the mostintensely used antibiotic for animals in the EU at 2294 tons.

As a consequence of the agricultural use of antimicrobials, antimicrobial resist-ance is widespread in the bacteria of farm animals (Rosdahl and Pedersen, 1998).Especially the growing animal is at risk of harbouring resistant bacteria because ofthe admission of antimicrobial feed additives and the therapeutic use of antibioticsduring the breeding period (Kamphues, 1999). Antibiotics fed particularly to younganimals to promote their growth have physiological effects on the intestinal wall, pas-sage, intestinal flora and absorption, together resulting in a better absorption of feed.


Table 3. Antimicrobials used in animals in the EU in 1997

Antimicrobial Tons

Beta-lactams 322Tetracycline 2294Macrolides 424Aminoglycosides 154Fluoroquinolones 43Trimethoprim/sulfonamide 75Other 182

Adapted from Ungemach (1999).

Used on a subtherapeutic scale, they also have a preventative function with respectto infections, resulting in less disease and diarrhoea (Hoogkamp-Korstanje, 1999).In recent years most of the antimicrobial feed additives have been banned withinthe EU to reduce the percentage of resistant bacteria in farm animals (EuropeanCommission, 1999). The effect of this ban on antibiotic resistance within farm animals and for human medicine is still under discussion (Boerlin et al., 2001).In part, the effect has been undermined by the increased use of therapeutic agents(Kamphues, 1999). The amount of antibiotic resistance is dependent not only fromthe application of antimicrobials, but also from the bacteria present in the farm animal. The most important bacteria comprise Salmonella, Campylobacter, entero-cocci, E. coli and various specific animal pathogens (including streptococci, staphylo-cocci and Pasteurella). The distribution of these species is dependent on the animalspecies and the kind of animal husbandry system (extensive, intensive, etc.). Thenature of the antibiotic resistance is highly variable, i.e. resistance can occur againstall known antimicrobial substances and groups. The most important substancegroups for human medicine, which will be mentioned in more detail, are the fluoroquinolones, the glycopeptides, aminoglycosides, macrolides, tetracycline and beta-lactams.

Resistance mechanisms are manifold and include destruction of the antimicro-bial inside or outside of the cell, efflux mechanisms, target alteration and alternativemetabolic pathways. The genetic information of the resistance can be located chro-mosomally or on extrachromosomal elements (plasmids). Aspects of bacterial andanimal species, acquisition, spread and mechanisms of resistance, as well as thesources of resistance, will be discussed with respect to the special situation in thegrowing animal.

4. RESISTANCE IN FARM ANIMALS

The growing animal is of great importance with respect to antibiotic resistance,especially as a production animal. Most animals used for meat production will beslaughtered within their growing phase (cattle within approximately 18 months, pigswithin 6–7 months, poultry within 30–42 days). Therefore, the focus of this chapteris on the main production animals as mentioned before (cattle, pig, poultry).

Owing to the treatment of neonatal E. coli infections and pneumonia in calves,antimicrobial therapy is quite common in cattle (Anonymous, 1998). Antibioticgrowth promoters were also in use until the recent ban of the antimicrobial feed addi-tives. Especially Salmonella and E. coli are reported to be resistant in cattle, with focuson S. Typhimurium DT 104, a serovar with multiple resistances (ampicillin, tetracy-cline, sulfanomides, streptomycin, chloramphenicol, fluoroquinolones) (Anonymous,1998). Major antibiotic classes which are currently allowed for therapy in cattle andcalves in Europe comprise beta-lactams including cefazolines, quinolones, tetracy-clines, sulfonamides, aminoglycosides and macrolides (Kluge and Ungemach, 2000).


In the US, various antimicrobials are fed to cattle. Monensin and lasalocid were commonly used for growth promotion, whereas some producers used neomycin orvirginiamycin. Chlortetracycline, chlortetracycline-sulfamethazine, oxytetracyclineand tylosin were fed on feedlots for group therapy of animals. For individual animaltherapy about 50% of feedlots used tilmicosin, florfenicol and tetracyclines, whileantibiotics used less frequently by feedlots for individual animal therapy includedcephalosporins, penicillins, macrolides and fluoroquinolones. Approximately 41% offeedlots administered antimicrobials such as tilmicosin, florfenicol and oxytetra-cyclines for metaphylaxis (McEwen and Fedorka-Cray, 2002). On US dairy farms, penicillins, cephalosporins, erythromycin and oxytetracycline are mainlyadministered through intramammary infusion to treat mastitis and are routinelyadministered to herds to prevent mastitis during non-lactating periods (McEwen andFedorka-Cray, 2002).

Campylobacter coli, C. jejuni, E. coli and Salmonella are the main targets forantibiotic use in pigs. Substance classes in use for antibiotic therapy in Europe forweaning pigs are the same as summarized for cattle. In the US, several antibiotics(table 2) are used for growth promotion or disease prophylaxis in swine.Antimicrobials such as ceftiofur, sulfonamides, tetracyclines and tiamulin are givento treat and prevent pneumonia, an important problem among swine. Gentamicin,apramicin, and neomycin are used to treat bacterial diarrhoea, caused by pathogenssuch as E. coli and C. perfringens. Swine dysentery (Serpulina hyodysenteriae) andileitis (Lawsonia intracellularis) are other important diseases that may be treatedwith antimicrobials such as lincomycin, tiamulin and macrolides (McEwen andFedorka-Cray, 2002).

The main concerns for bacterial infections in poultry production are C. jejuniand Salmonella. In Europe, a variety of fluoroquinolones are currently in use forantimicrobial therapy (e.g., enrofloxacin, difloxacin). Consequently, the increaseof fluoroquinolone resistant bacteria in poultry is an emerging problem. In the US,broiler feed usually contains coccidiostats (e.g., ionophores, sulfonamides) whileother antimicrobials (e.g., bacitracin, bambermycin, chlortetracycline, penicillin,virginiamycin and arsenic compounds) are approved for growth promotion and feedefficiency (McEwen and Fedorka-Cray, 2002).

5. BACTERIAL SPECIES AND RESISTANCE

Monitoring systems for antibacterial resistance often focus on zoonotic bacteria,commensals and animal pathogens. Within the European Union no system exists forEU-wide monitoring (OIE, 2001); however, effort exists to establish such a systemfor the main zoonotic pathogens (Salmonella, Campylobacter), commensals (ente-rococci, E. coli) and animal pathogens (e.g., enteropathogenic E. coli, streptococcietc.) (Caprioli et al., 2000). In the following discussion on the antibiotic resistanceof zoonotic pathogens and commensals special focus is on the situation in Germany.


However, the regional resistance rates can be used as an example for the main trendsin Europe as well as in industrialized countries.

5.1. Salmonella

Numerous research studies report on the antibiotic resistance of salmonellae.However, coordinated monitoring programmes for the EU have not been establishedso far. In Germany, a total of 65.9% of isolates from animals and food were resist-ant or multiple-resistant against antibiotics (BfR, 2003). Especially, pigs and calvesare responsible for the overall high resistance rates. Up to 40% of the isolates are multiply resistant, often with resistances against five antibiotics (ampicillin,chloramphenicol, streptomycin-spectinomycin, sulfanomides, tetracycline) (BfR,2003). Salmonella Typhimurium definite type DT 104 is the most often found typeamong resistant isolates. Especially, the multiple-resistant strains (96%) possessintegrons and can thus be able to transfer resistance (BfR, 2003). S. TyphimuriumDT 104 initially emerged in cattle in 1988 in England and Wales and was subse-quently found in meat and meat products. Human illness occurred as a result of contact of humans with farm animals, or from consumption of meat (Khachatourians,1998). The number of DT 104 isolates from humans in Britain increased from 259 to3837 between 1990 and 1995 (Lee et al., 1994). The proportion of antibiotic resistantSalmonella associated with human infections rose from 17 to 31% of isolates between1979/80 and 1989/90 in the US, and the proportion of Salmonella isolates exhibitingantibiotic and multidrug resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides and tetracyclines increased from 39 to 97% in the same period (Lee et al., 1994). In 1990, 90% of all DT 104 isolates obtained from humans were multi-drug resistant, including resistance to fluoroquinolones (Khachatourians, 1998). In1997, an interagency workshop with representatives from Canada, the US, the UK andthe Netherlands reported a rise in the number of multidrug-resistant DT 104 isolatesand resistance to trimethoprim and fluoroquinolone was also reported (Angulo, 1997).

However, some countries, e.g. Sweden, show very low prevalence of resistantSalmonella isolates (SVARM, 2000). These countries, especially the Scandinaviancountries, have traditionally low rates of Salmonella contamination in animal husbandry and apply a strict regime to establish Salmonella free livestock.

5.2. Campylobacter

In human medicine, the antimicrobials used for treatment of severe Campylobacterinfections are fluoroquinolones and macrolides (Skirrow and Blaser, 2000).However, the use of fluoroquinolone antibiotics in veterinary medicine has led tothe emergence of antibiotic resistant C. jejuni in human and chicken populations.The prevalence for the instances of enrofloxacin resistant strains of Campylobacterin poultry and in humans increased from 0 to 14% and from 0 to 11%, respectively


(Khachatourians, 1998; Endtz et al., 1991). Erythromycin resistant campylobactershave been reported, but erythromycin resistance is found in E. coli isolates ratherthan in C. jejuni isolates (Smith et al., 2000). Resistance to fluoroquinolones in isolates from human clinical specimens has been reported to occur worldwide(Talsma et al., 1999). In some regions, the increase in antimicrobial resistance hasbeen very rapid, e.g. in Spain the development of resistance in the past decade wasremarkable (Sáenz et al., 2000). In coincidence with the increasing occurrence ofresistance to fluoroquinolones in human clinical isolates, isolates from food animalsalso show increasing resistance rates (Sáenz et al., 2000).

In Germany origin-specific resistance rates for Campylobacter spp. from pigs,broilers and cattle could be demonstrated (Luber et al., 2003). Whereas isolates frompigs were significantly more often resistant to erythromycin (37.9%) and tetra-cycline (60.8%) than those from cattle or broilers, the latter were significantly moreoften resistant to ampicillin (37.9%), nalidixic acid and ciprofloxacin (each 55.2%)(Luber et al., 2003). Multiresistant strains occurred also among poultry isolates.These high resistance rates indicate the importance of food animals as potentialsources for resistant strains. The origin-specific resistance rates shown above reflectthe differences in the use of antibiotics in different animal species.

5.3. Enterococci

The focus of research concerning antibiotic resistant enterococci in animals andfood is on glycopeptides, especially vancomycin and teicoplanin. These substancesare reserve antibiotics in human medicine for severe nosocomial infections withenterococci (E. faecium and more often E. faecalis) and are therefore of primaryimportance. Vancomycin has been used since the 1950s in human medicine and thestructurally similar glycopeptide antibiotic avoparcin has been used as a growthpromoter in farm animals in Europe since the 1980s. Depending on the livestock,4–50 mg/kg of avoparcin was added to animal feed (Feed Additive Directive 70/524of the EC) (Witte, 1997). In Europe, ergotropic (growth stimulatory) use ofavoparcin has been suspected to contribute to the rise of vancomycin resistant ente-rococci (VRE) in hospitals, as VRE isolates can be transmitted to humans via thefood chain (Bates et al., 1994; Klare et al., 1995a,b; McDonald et al., 1997; Witte1997). The problem of VRE in European hospitals and the supposed food transmis-sion route for infection led to a ban of avoparcin as a growth promoter in the EU in1997. The prevalence of VRE was reported in recent years to be up to 17% (Peters,2003). After the ban of avoparcin, studies from Germany indicated that VRE couldnot be isolated from cattle and food from cattle or pig meat (Peters, 2003; Peterset al., 2003). Klare et al. (1999) showed that the incidence of VRE from frozen andfresh poultry meats decreased two years after the avoparcin ban. Also in other coun-tries, a decrease of resistance rates could be demonstrated. However, the effect of the ban of the growth promoter avoparcin (a glycopeptide) is still under discussion


(Boerlin et al., 2001). In the US, the situation regarding development of VRE differsfrom Europe, in that ergotropic use of avoparcin did not occur as this antibiotic wasnot licensed for use in animal husbandry. The development of VRE in the US mayhave occurred as a result of intensive use of vancomycin to treat hospital-associatedinfections.

Enterococci are known to be intrinsically resistant to a number of antibiotics,including cephalosporins, β-lactams, sulfonamides, and low levels of clindamycinand aminoglycosides (Murray, 1990; Leclercq, 1997; Morrison et al., 1997).Acquired resistance, based on acquisition of plasmids and transposons, has beenobserved for chloramphenicol, erythromycin, high levels of clindamycin and amino-glycosides, tetracycline, β-lactams (by β-lactamase or penicillinase), fluoro-quinolones and glycopeptides (Murray, 1990; Moellering, 1991; Landman andQuale, 1997; Leclercq, 1997; Morrison et al., 1997). Antibiotic resistant enterococciare well known to occur in various beef, poultry or pork products and such strainsmay be resistant to chloramphenicol, tetracycline, erythromycin or high levels ofaminoglycoside (Klein et al., 1998; Quednau et al., 1998; Teuber et al., 1999; Paviaet al., 2000; Baumgartner et al., 2001). Enterococci with multiple resistances includ-ing resistance to cephalosporins, macrolide and tetracycline classes of antimicro-bials, as well as resistance to streptogramin quinopristin-dalfopristin were shownto be associated with the poultry environment (Joseph et al., 2001). Enterococci isolated from pig gastrointestinal sources in Spain, Denmark and Sweden alsoshowed resistances to erythromycin, chloramphenicol, tetracycline and amino-glycosides, with higher levels of resistance observed for isolates stemming fromSpain and Denmark when compared to those from Sweden (Aarestrup et al., 2002).Aarestrup et al. (2002) suggested that this effect was a reflection of the fact thathigher amounts of antibiotics were fed as growth promoters in Spain and Denmark,when compared to Sweden.

5.4. E. coli

In Germany, 42% of investigated E. coli strains were resistant and 36% were multi-ple resistant in isolations from cattle, pigs and poultry (BfR, 2003). Especially iso-lates from poultry and pigs showed a high prevalence of resistant E. coli (61 and59%, respectively). Of the poultry isolates, 33% were quinolone resistant with13.5% being fluoroquinolone resistant (BfR, 2003). In total over 300 isolates havebeen tested. However, representative studies sensu stricto have to include more isolates from different regions in Germany. E. coli has been tested very intensely indifferent countries and always isolates from animals as well from food were foundto be resistant to a variety of antibiotics in significant percentages (e.g., Lehn et al.,1996; Trolldenier, 1996; Altieri and Massa, 1999; Mathew et al., 1999). Theseresistances comprise tetracycline (up to 83%, Bensink et al., 1981), ciprofloxacin(0–13%, Sáenz et al., 2001) or ampicillin (0–47%, Sáenz et al., 2001).


A specific subgroup of the E. coli population, the potential enterohaemorrhagic E. coli (EHEC) bacteria with verocytotoxic E. coli-associated virulence factors(VTEC-AFV), has also been tested. In agreement with data from Germany for cat-tle, lamb and sheep as well as food (Klein and Bülte, 2003), EHEC and VTEC-AFVseem to be less resistant to antimicrobials and they have a high percentage of susceptible isolates compared to E. coli isolates in general (CDC, 2000). Data concerning this subpopulation are very rare and studies are needed to elucidate thepossible development of antibiotic resistance in these potentially pathogenic agentsentering the food chain. This is not necessary for effective antibiotic therapy inhuman medicine, because infections with EHEC bacteria are normally treated with-out antibiotics. However, the existence of such resistances may be an indicator andcan give an overview on the distribution of resistances in food of animal origin.

6. RESISTANCE TRANSFER IN THE GROWING ANIMAL

In principle the resistance transfer and the acquisition of antibiotic resistance is theresult of one of the following steps (Berger-Bächi, 2001):

acquisition of resistant non-pathogenic bacteria and subsequent transfer of theresistance to the autochthonous gut flora and/or to gut-associated pathogenicbacteria (fig. 1A),


Fig. 1. Transfer mechanisms and routes of antibiotic resistances from the farm animal via food to the humangastrointestinal tract. (A) Antibiotic resistance gene transfer from the non-pathogenic, physiological animaland/or food microflora to pathogenic microorganisms of the human gastrointestinal tract (example:vancomycin resistant enterococci (non-pathogenic); transfer mechanism: conjugation). *VRE: vancomycinresistant enterococci. (B) Transfer of antibiotic-resistant pathogenic microorganisms from animal and/or foodto the human gastrointestinal tract (example: fluoroquinolone resistant Campylobacter spp.; no transfer ofresistance possible, but dissemination of resistant strains/clones).

acquisition of resistant pathogenic bacteria from the environment or by directcontact (fig. 1B), or

spontaneous mutation under selective pressure or without selective pressure.The transfer of the resistance is dependent on the nature of the resistance

mechanism. Resistance resulting from mutation events will be transferred by the dissemination of the resistant strain itself, whereas transfer by conjugation results indissemination of the relevant resistance gene in different strains or even bacterialspecies. Therefore the acquisition of resistant non-pathogenic bacteria leads only to resistant pathogens if a transfer mechanism is available. An example for a non-transferable resistance caused by a single or double mutation step is the quinoloneresistance in Campylobacter or Salmonella (Bachoual et al., 2001). The inductionof the resistance by quinolone treatment is described for Campylobacter in broiler production (Jacobs-Reitsma et al., 1994). An example for a conjugational transfer ofresistance is the transfer of the vanA-related glycopeptide resistance in enterococci.This transfer mechanism, especially for the conjugative transfer of plasmids, is wide-spread amongst Gram-positive bacteria (Grohmann et al., 2003).

A simplified scheme of transfer routes and possible influence factors for thetransfer of antibiotic resistances from the farm animal to humans is given in fig. 2.

7. EFFECT OF ANTIMICROBIAL GROWTH PROMOTERS ONANTIBIOTIC RESISTANCE

Antimicrobial growth promoters (AGPs) are used to enhance the growth of younganimals in order to gain the slaughter weight at an early stage. The working princi-ple of AGPs is not fully understood. Some effects may be caused by the preventionof simple enteric infections and the reduction of the microbial population in the gutin general (Kamphues, 1999). Furthermore, such effects may include metaboliceffects, improvement of digestion or absorption of certain nutrients, nutrient sparingeffects in which antibiotics may reduce the animal’s dietary requirements andincreased feed and/or water intake (Gersema and Helling, 1986). Unwanted side-effects are the possible development of antibiotic resistances in enteric bacteria,


Fig. 2. Transfer of antibiotic resistances. Elements and transfer routes influencing the amount of antibioticresistance in farm animals, food and human medicine.

because some substances used as AGPs can induce cross-resistance to substancesalso used in human medicine. A well documented case is the application of AGPs inpoultry and pigs and the effects on the glycopeptide resistance of enterococci.

However, not only the much quoted VRE example shows that feeding of antibioticsat sub-inhibitory concentrations can cause problems for humans. For example, the useof the streptothricin antibiotic nuorseothricin as a porcine growth promoter in the former East Germany between 1983 and 1990 resulted in the development of anantibiotic resistance transposon (Khachatourians, 1998; Witte, 1998; Aarestrup, 1999).Two years after its inroduction, resistant E. coli isolates were found in pig guts, meatproducts, as well as the intestines of pig farmers and their families, patients with urinary tract infections and the general public (Witte, 1997; Khachatourians, 1998). Asthis antibiotic has not been applied in human medicine, these observations showed that growth promoting antibiotics do induce resistances in the animal populationwhich later disseminate via resistant enterobacteria into the human population(Teuber, 1999).

In the Netherlands, water medication with the fluoroquinolone antibioticenrofloxacin in poultry production was followed by the emergence of fluoro-quinolone resistant Campylobacter species among poultry and humans (Endtz et al.,1991; Aarestrup, 1999).


For the safety of food of animal origin and for the prevention of severe infectionswithout antibiotic treatment options, the reduction of antibiotic resistance in animalhusbandry to a minimum is essential. On the other hand, antibiotic therapy in vet-erinary medicine is indispensable for reasons of animal welfare, and the necessarytreatment of infections with zoonotic agents. A balance between both objectivesmust be established. Therefore, in the future, representative monitoring systems forthe evaluation of the prevalence of antibiotic resistance in farm animals as well asin food of animal origin are required. Target organisms should be representatives ofzoonotic bacteria, commensal bacteria and animal pathogens. These systems mustcover nationwide development and should be evaluated European-wide or on aninternational level. Representative samples are crucial for the value of these exami-nations. Another important point is that the methods for the determination of antibi-otic resistance (MIC values are recommended worldwide) and also the breakpointsshould be standardized to enable a comparison of the results. Also important forfuture investigation should be the molecular characterization of antibiotic resistance,so as to detect new variants of resistance mechanisms or emerging resistance pat-terns. The application of the DNA microarray technique may in future facilitate suchmolecular characterization both quickly and accurately. With the informationobtained from monitoring programmes the prudent use of veterinary therapeutics ispossible and the need for restrictions or the substitution of therapy schemes are more


easily recognized and performed. However, antibiotic usage in animal husbandry isnot the only source for resistant strains in human medicine and even in many casesis only a minor contribution. Therefore, this should not be the only step, but alsoclinical therapy in humans should be critically evaluated. In any case, a reduction of theamount of antibiotics used, whether in human or veterinary medicine, is a step forwardto reducing the level of antibiotic resistance and thus to a more reliable therapy.

Finally, it should be mentioned that reduction in antibiotic uses at subtherapeu-tic levels will come with a price. As an example, increased production cost as aresult of abstaining from the use of antibiotics as growth promoters was reported forpork production in Denmark. Apparently, 25% of the piglets suffered from diar-rhoea. Piglets also grew slower, so that in the worst case it took 20 days longer fora piglet to reach a weight of 30 kg. In 2000 it was calculated that abstaining fromsubtherapeutic use of antibiotics in a production unit with 200 sows and a yearlyproduction of 5000 piglets would result in a cost increase of 5700 to 6600 DM(about E 3000) (Verseput, 2000). Clearly, increases in production costs will increaseproduct price and consumers have to expect price increases for meat products as aresult of the discontinuation of the use of antibiotic growth promoters.

REFERENCES

Aarestrup, F.M., 1999. Association between the consumption of antimicrobial agents in animal husbandryand the occurrence of resistant bacteria among food animals. Int. J. Antimicrob. Agents 12, 279–285.

Aarestrup, F.M., Hasman, H., Jensen, L.B., Moreno, M., Herrero, I.A., Domíngez, L., Finn, M., Franklin, A., 2002. Antimicrobial resistance among enterococci from pigs in three European countries. Appl. Environ. Microbiol. 68, 4127–4129.

Altieri, C., Massa, S., 1999. Antibiotic resistant Escherichia coli from food of animal origin. Adv. FoodSci. 21, 166–169.

Angulo, F.J., 1997. Multidrug resistant Salmonella typhimurium definitive type 104. Emerg. Infect. Dis.3, 414.

Anonymous, 1998. A Review of Antimicrobial Resistance in the Food Chain. Ministry of Agriculture,Food and Fisheries, London.

Bachoual, R., Ouobdessalam, S., Mory, F., Lascols, C., Soussy, S.-J., Tankovic, J., 2001. Single or dou-ble mutational alterations of GyrA associated with fluoroquinolone resistance in Campylobacterjejuni and Campylobacter coli. Microb. Drug Resist. 7, 257–261.

Bates, J., Jordens, J.Z., Griffiths, D.T., 1994. Farm animals as a putative reservoir for vancomycin-resist-ant enterococcal infection in man. J. Antimicrob. Chemother. 34, 507–516.

Baumgartner, A., Kueffer, M., Rohner, P., 2001. Occurrence and antibiotic resistance of enterococci invarious ready-to-eat foods. Arch. Lebensmittelhyg. 52, 1–24.

Bensink, J.C., Frost, A.J., Mathers, W., Mutimer, M.D., Rankin, G., Woolcock, J.B., 1981. The isolationof antibiotic resistant coliforms from meat and sewage. Aust. Vet. J. 57, 12–16, 19.

Berger-Bächi, B., 2001. Mechanisms of antibiotic resistances in bacteria. Mitt. Lebensm. Hyg. 92, 3–9.BfR (Bundesinstitut für Risikobewertung), 2003. Zweiter Zwischenbericht zum Forschungsvorhaben:

Erfassung phänotypischer und genotypischer Resistenzeigenschaften bei Salmonella und E. coliIsolaten vom Tier, Lebensmitteln, Futtermitteln und der Umwelt. (Second Report of a ResearchProject covering phenotypic and genotypic resistance marker for salmonella and E. coli from animals,food, feed and the environment.) BfR, Berlin.

Boerlin, P., Wissing, A., Aarestrup, F.M., Frey, J., Nicolet, J., 2001. Antimicrobial growth promoter banand resistance to macrolides and vancomycin in enterococci from pigs. J. Clin. Microbiol. 39,4193–4195.


Bogaard, A.E. van den, Stobberingh, E.E., 2001. Antibiotic resistance epidemiology: links betweenantibiotic use in food animals and resistance in human bacteria. Mitt. Lebensm. Hyg. 92, 42–58.

Caprioli, A., Busani, L., Martel, J.L., Helmuth, R., 2000. Monitoring of antibiotic resistance in bacteria ofanimal origin: epidemiological and microbiological methods. Int. J. Antimicrob. Agents 14, 295–301.

CDC, 2000. National antimicrobial resistance monitoring system: enteric bacteria. In: 2000 AnnualReport. Centers for Disease Control, Atlanta, GA.

Clewell, D.B., 1990. Movable genetic elements and antibiotic resistance in enterococci. Eur. J. Clin.Microbiol. Infect. Dis. 9, 90–102.

Davies, J., 1996. Origins and evolution of antibiotic resistance. Microbiología SEM 12, 9–16.Endtz, H.P., Ruijs, G.J., van Klingeren, B., Jansen, W.H., van der Reyden, T., Mouton, R.P., 1991.

Quinolone resistance in campylobacter isolated from man and poultry following the introduction offluoroquinolones in veterinary medicine. J. Antimicrob. Chemother. 27, 199–208.

European Commission, 1999. Opinion of the scientific steering committee on antimicrobial resistance.European Commission, DG XXIV, Brussels.

Feinman, S.E., 1998. Antibiotics in animal feed: drug resistance revisited. Am. Soc. Microbiol. News 64,24–30.

Gersema, L.M., Helling, D.K., 1986. The use of subtherapeutic antibiotics in animal feed and its impli-cations on human health. Drug Intel. Clin. Pharm. 20, 214–218.

Grohmann, E., Muth, G., Espinosa, M., 2003. Conjugative plasmid transfer in Gram-positive bacteria.Microbiol. Mol. Rev. 67, 277–301.

Hoogkamp-Korstanje, J.A.A., 1999. Antimicrobiële grocibevorderaars. (Antimicrobial growth promoters.)Ned Tijdschr. Geneeskd. 143, 1293–1295.

Jacobs-Reitsma, W.F., Kan, C.A., Bolder, N.M., 1994. The induction of quinolone resistance inCampylobacter bacteria in broilers by quinolone treatment. Lett. Appl. Microbiol. 19, 228–231.

Joseph, S.W., Hayes, J.R., English, L.L., Carr, L.E., Wagner, D.D., 2001. Implications of multiple antimicrobial-resistant enterococci associated with the poultry environment. Food Add. Contam. 18,1118–1123.

Jukes, T., 1986. Effects of low levels of antibiotics in livestock feed. Effects Antibiotics Livestock Feeds10, 112–126.

Kamphues, J., 1999. Leistungsförderer mit antibiotischer Wirkung aus Sicht der Tierernährung. (Growthpromoters with antibiotic effect under the aspect of animal nutrition.) Berl. Münchn. Tierärztl. Wschr.112, 370–379.

Khachatourians, G.G., 1998. Agricultural use of antibiotics and the evolution and transfer of antibiotic-resistant bacteria. Can. Med. Ass. J. 159, 1129–1136.

Klare, I., Heier, H., Claus, H., Reissbrodt, R., Witte, W., 1995a. vanA-mediated high-level glycopeptideresistance in Enterococcus faecium from animal husbandry. FEMS Microbiol. Lett. 125, 165–172.

Klare, I., Heier, H., Claus, H., Böhme, G., Marin, S., Seltmann, G., Hakenbeck, R., Antanassova, V., Witte, W., 1995b. Enterococcus faecium strains with vanA-mediated high-level glycopeptide resist-ance isolated from animal foodstuffs and fecal samples in the community. Microb. Drug Resist. 1, 265–272.

Klare, I., Badstübner, D., Konstabel, C., Böhme, G., Claus, H., Witte, W., 1999. Decreased incidence ofVanA-type vancomycin-resistant enterococci isolated from poultry meat and from fecal samples ofhumans in the community after discontinuation of avoparcin usage in animal husbandry. Microb.Drug Res. 5, 45–52.

Klein, G., Bülte, 2003. Antibiotic susceptibility pattern of Escherichia coli with enterohemorrhagicEscherichia coli-associated virulence factors from food and animal feces. Food Microb. 20, 27–33.

Klein, G., Pack, A., Reuter, G., 1998. Antibiotic resistance patterns of enterococci and occurrence of van-comycin-resistant enterococci in raw minced beef and pork in Germany. Appl. Environ. Microbiol.64, 1825–1830.

Kluge, K., Ungemach, F., 2000. Alle nach EU Recht erlaubten Wirkstoffe und ihre Zulassung inDeutschland. (Complete list of substances allowed according to EU law and their approval inGermany.) Dtsch. Tierärztebl. 48, Suppl. 06/00.

Landman, D., Quale, J.M., 1997. Management of infections due to resistant enterococci: A review oftherapeutic options. J. Antimicrob. Chemother. 40, 161–170.

Leclercq, R., 1997. Enterococci acquire new kinds of resistance. Clin. Infect. Dis. 24, Suppl. 1, S80–S84.


Lee, L.A., Puhr, N.D., Malony, F.K., Bean, N.H., Taupe, R.V., 1994. Increase in antimicrobial-resistantSalmonella infections in the United States, 1989–1990. J. Infect. Dis. 170, 128–134.

Lehn, N., Stöwer-Hoffmann, J., Kott, T., Strassner, C., Wagner, H., Krönke, M., Schneider-Brachert, W.,1996. Characterization of clinical isolates of Escherichia coli showing high levels of fluoroquinoloneresistance. J. Clin. Microbiol. 34, 597–602.

Levy, S.B., 1998. The challenge of antibiotic resistance. Sci. Am. 278, 46–53.Luber, P., Bartelt, E., Genschow, E., Wagner, J., Hahn, H., 2003. Comparison of broth microdilution, E

Test, and agar dilution methods for antibiotic susceptibility testing of Campylobacter jejuni andCampylobacter coli. J. Clin. Microbiol. 41, 1062–1068.

Madigan, M.T., Martinko, J.M., Parker, J., 1997. Biology of Microorganisms, 8th edition. Prentice-Hall,Englewood Cliffs, NJ, p. 309.

Mathew, A.G., Saxton, A.M., Upchurch, W.G., Chattin, S.E., 1999. Multiple antibiotic resistance patternsof Escherichia coli isolates from swine farms. Appl. Environm. Microbiol. 65, 2770–2772.

McDonald, L.C., Kuehnert, M.J., Tenover, F.C., Jarvis, W.R., 1997. Vancomycin-resistant enterococcioutside the healthcare setting: prevalence, sources, and public health implications. Emerg. Infect. Dis.3, 311–317.

McEwen, S.A., Fedorka-Cray, P.J., 2002. Antimicrobial use and resistance in animals. Clin. Infect. Dis.34(Suppl. 3), S93–S106.

Moellering, R.C. Jr., 1991. The enterococcus: a classic example of the impact of antimicrobial resistanceon therapeutic options. J. Antimicrob. Chemother. 28, 1–12.

Morrison, D., Woodford, N., Cookson, B., 1997. Enterococci as emerging pathogens of humans. J. Appl.Microbiol. Symp. Suppl. 83, 89S–99S.

Murray, B.E., 1990. The life and times of the Enterococcus. Clin. Microbiol. Rev. 3, 46–65.OIE, 2001. Second International Conference on Antimicrobial Resistance. Office International des

Epizooties, Paris, France, 2−4 October 2001.Pavia, M., Nobile, C.G.A., Salpietro, L., Angelillio, I.F., 2000. Vancomycin resistance and antibiotic

susceptibility of enterococci in raw meat. J. Food Prot. 7, 912−915.Peters, J., 2003. Antibiotikaresistenz von Enterokokken aus landwirtschaftlichen Nutztieren und

Lebensmitteln tierischer Herkunft. (Antibiotic Resistance of Enterococci from Farm Animals andFood of Animal Origin.) Vet. Med. Diss., Frei Universität Berlin.

Peters, J., Mac, K., Wichmann-Schauer, H., Klein, G., Ellerbroek, L., 2003. Resistenzen vonEnterokokken aus Lebensmitteln tierischer Herkunft Deutschland. (Resistances of enterococci fromfood of animal origin in Germany). 43. Arbeitstagung des Arbeitsgebietes Lebensmittelhygiene inGarmisch-Partenkirchen, 24−27 September 2002. Gießen: Deutsche VeterinärmedizinischeGesellschaft, 2003, Proceedings, pp. 206−211.

Quednau, M., Ahrne, S., Petersson, A.C., Molin, G., 1998. Antibiotic-resistant strains of Enterococcusisolated from Swedish and Danish retailed chicken and pork. J. Appl. Microbiol. 84, 1163−1170.

Rosdahl, V.T., Pedersen, K.B. (Eds.), 1998. The Copenhagen Recommendations. Report from theInvitational EU Conference on the Microbial Threat. Ministry of Health, Ministry of Food,Agriculture and Fisheries, Copenhagen.

Sáenz, Y., Zarazaga, M., Lantero, M., Gastañares, M.J., Baquero, F., Torres, C., 2000. Antibiotic resist-ance in Campylobacter strains isolated from animals, foods, and humans in Spain in 1997–1998.Antimicrob. Agents Chemother. 44, 267–271.

Sáenz, Y., Zarazaga, M., Briñas, L., Lantero, M., Ruiz-Larrea, F., Torres, C., 2001. Antibiotic resistanceof Escherichia coli isolates obtained from animals, foods and humans in Spain. Int. J. Antimicrob.Agents 18, 353–358.

SCOPE, 2002. Opinion of the Scientific Committee on Animal Nutrition on the Criteria for Assessing theSafety of Micro-organisms Resistant to Antibiotics of Human Clinical and Veterinary Importance.European Commission Health and Consumer Protection Directorate-General, Directorate C –Scientific Opinions, pp. 1–20.

Skirrow, M.B., Blaser, M.J., 2000. Clinical aspects of Campylobacter infection. In: Nachamkin, I.,Blaser, M.J. (Eds.), Campylobacter, 2nd edition. ASM Press, Washington, DC, pp. 69–88.

Smith, K.E., Bender, J.B., Osterholm, M.T., 2000. Antimicrobial resistance in animals and relevance tohuman infection. In: Nachamkin, I., Blaser, M.J. (Eds.), Campylobacter, 2nd edition. ASM Press,Washington, DC, pp. 483–495.


SVARM, 2000. Swedish Veterinary Antibiotic Resistance Monitoring. National Veterinary Institute,Uppsala, Sweden.

Talsma, E., Goettsch, W.G., Nieste, H.L., Schrijnemakers, P.M., Sprenger, M.J., 1999. Resistance inCampylobacter species: increased resistance to fluoroquinolones and seasonal variation. Clin. Infect.Dis. 29, 845–848.

Teuber, 1999. Spread of antibiotic resistance with food-borne pathogens. Cell Mol. Life Sci. 56,755–763.

Teuber, M., 2001. Mitt Antibiotikaresistente Bakterien in Lebensmitteln. (Antibiotic resistant bacteria infood.) Mitt Lebensm. Hyg. 92, 10–27.

Teuber, M., Meile, L., Schwarz, F., 1999. Acquired antibiotic resistance in lactic acid bacteria from food.Ant. v. Leeuwenhoek 76, 115–137.

Trolldenier, H., 1996. Resistenzentwicklungen von Infektionserregern landwirtschaftlicher Nutztiere inDeutschland (1990–1994) – ein Überblick. (Trends of resistances of farm animal pathogens inGermany (1990–1994) – an overview.) Dt. Tierärztl. Wschr. 103, 256–260.

Verseput, W., 2000. Ban of antibiotics is expensive. Fleischwirtsch. 8, 15.Visek, W., 1978. The mode of growth promotion by antibiotics. J. Anim. Sci. 46,1447–1469.Ungemach, F.R., 1999. Antibiotics and the resistance problem. Deutsches Tierärzteblatt 3, 224–227.Walsh, C., 2003. Antibiotics: Actions, Origins, Resistance, 1st edition. ASM Press, Washington, DC.Witte, W., 1997. Impact of antibiotic use in animal feeding on resistance of bacterial pathogens in

humans. In: Chadwick, D.J., Goode, J. (Eds.), Antibiotic Resistance: Origins, Evolution, Selectionand Spread. Ciba Foundation. John Wiley & Sons, New York, pp. 61–75.

Witte, W., 1998. Medical consequences of antibiotic use in agriculture. Science 279, 996–997.


208

A diverse range of bacteria and viruses is associated with diseases of fish. The skin,lateral line, gills and gastrointestinal tract or a combination of these organs are sug-gested to be infection routes. The purpose of this review is to present current knowl-edge on adhesion, colonization and translocation of pathogenic agents in thegastrointestinal tract of growing fish.

1. INTRODUCTION

As fish live in an aqueous environment, their external surfaces will be regularlyexposed to potential pathogens, and water taken into the gastrointestinal (GI) tractduring feeding can deliver them to the mucosal surfaces of this tract. Even in thenon-feeding early stages of development of marine fish larvae, drinking of water isrequired for osmotic regulation (Tytler and Blaxter, 1988) and this provides an earlyentry into the GI tract for bacteria. As in all animals, the GI tract is the route ofnutrient uptake and any perturbation by microbial action can be harmful. This isparticularly so in the early stages of fish larval development. In contrast to mam-mals, where numerous bacterial and viral pathogens produce severe diarrhoealdisease there are no directly equivalent pathogens known for fish. However, anumber of bacteria cause pathology in the gut of fish and this can be a route of sys-temic infection in many instances, comparable to that of invasive enteropathogensof mammals.

10 Pathogenesis and the gastrointestinal tractof growing fish

T.H. Birkbecka and E. Ringøb

aDivision of Infection and Immunity, Institute of Biomedical andLife Sciences, University of Glasgow, Joseph Black Building,Glasgow G12 8QQ, UKbSection of Arctic Veterinary Medicine, Department of Food Safety andInfection Biology, The Norwegian School of Veterinary Science,NO-9292 Tromsø, Norway


The bacterial pathogens of major importance in aquaculture are, with few excep-tions, Gram-negative microorganisms. Aeromonas salmonicida, A. hydrophila,Vibrio anguillarum, V. salmonicida, V. viscosus (Moritella viscosa or M. marina)and V. ordalii belong to the Vibrionaceae; Yersinia ruckeri, Edwardsiella ictaluriand E. tarda are members of the Enterobacteriaceae, and Piscirickettsia salmonis isa member of Pisciriickettsiaceae. Of the Gram-positive bacterial pathogens,Renibacterium salmoninarum belongs to the Corynebacteriaceae, Carnobacteriumpiscicola to the lactic acid bacteria, and others, e.g. Streptococcus iniae, S. difficileand Lactococcus garviae, are Gram-positive cocci. Microbial pathogenicity hasbeen defined as the biochemical mechanisms whereby microorganisms cause disease (Smith, 1990). Not all pathogens have an equal probability of causing infec-tion and disease. In this review, the term infection will be used to describe success-ful persistence or multiplication of a pathogen on or within the host, while diseasewill be described as an infection which causes significant overt damage to the host.

Intensive fish production has increased the risk of infectious diseases all over theworld (Press and Lillehaug, 1995; Karunasagar and Karunasagar, 1999), but to pre-vent microbial entry fish have various protective mechanisms, such as production ofmucus by goblet cells, the apical acidic microenvironment of the intestinal epithe-lium, cell turnover, peristalsis, gastric acidity, lysozyme and antibacterial activity ofepidermal mucus. At the same time, pathogenic microorganisms have evolvedmechanisms to target the skin, gills or GI tract as points of entry. The three majorroutes of infection are through: a) skin (Kawai et al., 1981; Muroga and De La Cruz,1987; Kanno et al., 1990; Magarinos et al., 1995; Svendsen and Bøgvald, 1997;Spanggard et al., 2001), b) gills (Baudin Laurencin and Germon, 1987; Hjeltnes et al., 1987; Svendsen et al., 1999), and c) the GI tract (Sakai, 1979; Rose et al., 1989;Chair et al., 1994; Olsson, 1995; Grisez et al., 1996; Olsson et al., 1996; Romalde et al., 1996; Jöborn et al., 1997; Robertson et al., 2000; Lødemel et al., 2001).

Pathogenicity can be divided into four different phases: 1) the initial phase wherethe pathogen enters the host’s environment, including the GI tract, 2) the exponentialphase where the pathogen adheres to and colonizes mucosal surfaces, replicates to suf-ficient numbers and/or translocates into host enterocytes, 3) the stationary phase wherethe pathogen replicates within the host and circumvents the host defence system; in thisphase the host is moribund and this can quickly be followed by 4) the death phase.

In order to adhere successfully, colonize and produce disease, the pathogen mustovercome the host defence system. It is well known that stress from environmentalfactors, such as oxygen tension, water temperature and water salinity, are importantin increasing the susceptibility of fish to microbial pathogens. The water milieu canalso facilitate transmission of these pathogens.

The purpose of this review is to present information on 1) adhesion of bacteria tomucosal surfaces, 2) protection against bacterial adhesion, 3) bacterial translocation,4) invasion of host cells, 5) effect of diet in disease resistance and 6) data obtainedfrom endothermic animals which may have relevance to pathogenesis of fish.

Pathogenesis and GIT of growing fish 209

2. ADHESION OF BACTERIA TO MUCOSAL SURFACES

2.1. General factors

A number of environmental factors determine whether bacteria can adhere to and colo-nize the digestive tract of endothermic animals and these have been extensively reviewedby Savage (1983). Among these are: 1) gastric acidity (Gilliland, 1979); 2) bile salts(Floch et al., 1972); 3) peristalsis; 4) digestive enzymes (Marmur, 1961); 5) immuneresponse; and 6) indigenous microorganisms and the antibacterial compounds whichthey produce. In order to replicate to a sufficient number to allow transmission to a newsusceptible host, a microbial pathogen must enter a host, find a unique niche, circum-vent competing microbes and host defence barriers, and obtain nutrients from the host.

Adhesion of bacteria to surfaces such as epithelial cells involves four differenttypes of interaction, depending on the distance separating the bacteria from thesurface. Attraction is initially by van der Waal’s forces operating at distances greaterthan 50 nm, but at closer distances electrostatic interactions become more signifi-cant. As epithelial and bacterial cells are usually negatively charged, electrostaticrepulsion normally prevents closer association. In regions of lower ionic strengthcloser interaction may occur allowing hydrophobic interaction and specific receptor–ligand binding within circa 1 μm separation. This leads to strong binding betweenbacteria and host cell surfaces (Fletcher, 1996).

2.2. Adhesins

Several bacterial surface components can be involved in specific binding to epithe-lial cell ligands. The best characterized bacterial adhesins are the fimbriae (or pili)which are widely distributed on Gram-negative bacteria (Smyth et al., 1996), but arealso found on some Gram-positive bacteria (Klemm et al., 1998). Although fimbriaeare the most widely used adhesins in Gram-negative bacteria, flagella, capsules, pro-tein fibrils, outer membrane proteins (Gram-negative bacteria), surface proteins(Gram-positive bacteria) and crystalline protein surface arrays can all be used asadhesins (Henderson et al., 1999).

A range of fimbriae can be expressed by any one bacterial species; for example,14 different types of fimbriae are known in Escherichia coli, more than one of whichcan be expressed at the same time (Hacker, 1992; Klemm et al., 1998; Nataro andKaper, 1998). Type 1 fimbriae of E. coli are perhaps the best studied example and asingle cell may express over 500 fimbriae. Of approximately 7 nm in diameter and1 μm in length, type 1 fimbriae are composed of about 1000 copies of the majorstructural protein FimA, in a helical cylinder (Brinton, 1965) capped by the FimHprotein which recognizes mannose-containing receptors on the target eukaryoticcell. Other minor proteins, FimF and FimG are involved in binding FimH to theFimA helix and other genes in the Fim complex are required for assembly of thefimbriae and translocation through the bacterial membranes (Krogfeld et al., 1990;Klemm et al., 1998).

210 T.H. Birkbeck and E. Ringø

Despite their widespread occurrence in Gram-negative bacteria and their impor-tance in the pathogenesis of numerous infections, fimbriae have not yet been provedto be important in bacterial infections of fish. Saeed (1983) observed that E. ictaluriwas heavily piliated and suggested that this could be important in infection,although this awaits further examination.

The crystalline protein array (S layer or A layer) of A. salmonicida renders thesurface of this bacterium extremely hydrophobic to the extent that bacteria in brothcultures autoagglutinate and sediment rapidly when allowed to stand unshaken.Loss of the S layer can occur spontaneously, or can be induced by culture at elevatedtemperature or by Tn5 mutagenesis; in all cases the loss of the S layer is accompa-nied by loss of virulence (Ishiguro et al., 1981; Belland and Trust, 1985; Trust, 1986)and loss of adherence to macrophages (Trust et al., 1983).

Flagella are important adhesins for bacteria such as V. cholerae (Guentzel andBerry, 1975; Richardson, 1991) and Campylobacter jejuni (Wassenaar et al., 1991;Nachamkin et al., 1993). Although flagella have been shown to be important for thevirulence of V. anguillarum, this was not at the level of adhesion, as motility-deficient mutants which had much reduced virulence had similar adhesion levels tochinook salmon embryo (CHSE) cells as the wild-type organism (Ormonde et al.,2000). However, chemotactic motility and active motility are important for virulencein waterborne infections of fish (O’Toole et al., 1996, 1999; Ormonde et al., 2000).

Infectivity studies revealed that disruption of the flagellum and subsequent lossof motility correlated with an approximate 500-fold decrease in virulence when fishwere inoculated by immersion in bacteria-containing water. Once the pathogenshave reached the mucosal surface, several options exist: depending on their intrin-sic colonizing or invasive capacities, the nature of the toxin(s) they produce and theirability to resist host defences.

2.3. Electron microscopy studies of adhesion of fish-pathogenic bacteria totissues of the GI tract

In a recent study, Knudsen et al. (1999) tested pathogenic and non-pathogenic bacte-ria isolated from fish for their adhesion to cryosections from different mucosal sur-faces of Atlantic salmon by immunohistochemistry. The majority of the bacteriatested – V. anguillarum serotype O1, V. salmonicida, V. viscosus, Flexibacter maritimus,“gut vibrios” and intestinal isolates of V. salmonicida – all adhered to mucus from thepyloric caeca, foregut and hindgut. In contrast to these results, V. anguillarumserotype O2 (O2a and O2b), did not adhere to mucus.

The past decade has seen an explosion of information on our understanding ofbacterial adhesion at both the molecular and genetic level of endothermic animals,and electron microscopy has contributed significantly to this knowledge (Knutton,1995). Although several papers have described pathogenesis in fish, few investiga-tions have used transmission electron microscopy (TEM) and/or scanning electronmicroscopy (SEM) to evaluate the effect of bacterial infection on morphology in the


GI tract of fish. The advantage of using SEM is that large areas of the mucosal andcell surfaces can be examined rapidly for adherent bacteria. Adhesion can then beassessed qualitatively or quantitatively. For quantitative analysis a defined numberof fields is selected at random, photographed, and bacterial adherence assessed togive an adhesion index consisting of numbers of bacteria per unit area (Yamamotoet al., 1991) or percentage of area colonized by bacteria (Knutton et al., 1991). Theresolution of SEM is rarely sufficient to obtain detailed information about themechanisms of adhesion, although it has proved useful to determine bacterialadhesion/colonization of gut enterocytes of fish (Magarinos et al., 1996; Ringøet al., 2001, 2002). Magarinos et al. (1996) demonstrated that Photobacteriumdamselae (Pasteurella piscicida) strains adhered strongly to the intestines from seabream, sea bass and turbot in numbers ranging from 104 to 105 bacteria per gram ofintestine depending on the bacterial isolate and the fish species employed. Theseresults are clearly supported by scanning electron microscopy studies. Sometimes,bacteria colonizing the GI tract had their luminal ends protruding above the levelsof the microvilli (figs. 1 and 2). Micrographs displayed clear differences in levels ofbacterial association over a small area, as some enterocytes were heavily colonizedwhile others had no associated bacteria. Ringø et al. (2001) showed that some ente-rocytes were heavily colonized by bacteria when charr were fed dietary soybean oil,whereas a different situation was observed when fish were fed dietary linseed oil(Ringø et al., 2002). In the latter situation, most bacteria associated with enterocyteswere located at the apical brush border (fig. 3).


Fig. 1. Scanning electron micrograph of theapical aspects of enterocytes in the midgut ofArctic charr (Salvelinus alpinus L.) fed dietarysoybean oil. The borders between adjacent cellsare clearly visible, as microvilli which coverthe cell apex. The luminal ends of bacterialocated in the intestines between microvilli arealso visible (arrows). × 7500. After Ringø et al.(2001).


Fig. 2. Scanning electron micrograph show-ing cell apices in the hindgut of Arctic charr(Salvelinus alpinus L.) fed dietary soybean oil.Cell borders can be seen and all cells haveassociated bacteria (arrows), although numbersvary from cell to cell. Note the small spaces(arrowheads) between microvilli. These mayrepresent the transit paths of more deeplyembedded bacteria, or they may be created by bacterial loss. The latter may be an artefactof tissue preparation or a consequence of local bacterial cell division. × 5000. AfterRingø et al. (2001).

Fig. 3. Scanning electron micrograph show-ing bacteria associated with enterocytes in thehindgut of Arctic charr (Salvelinus alpinus L.)fed dietary linseed oil. Associated bacteria(arrows) are located at the apical brush border.× 7500. After Ringø et al. (2002).

Fish pathogenic bacteria, such as V. salmonicida and V. anguillarum, have beenshown in vivo to adhere to the intestinal epithelium of fish larvae and promote severedestruction of microvilli (Olafsen and Hansen, unpublished data). In contrast,Lødemel et al. (2001) did not show any destruction of microvilli in the pyloriccaeca, midgut or hindgut regions of adult Arctic charr (Salvelinus alpinus L.)infected by A. salmonicida subsp. salmonicida. SEM investigations of human intes-tinal mucosa infected with enteropathogenic E. coli (EPEC) showed that EPECadhere intimately in microcolonies and cause gross alterations of the brush bordersurface of infected enterocytes (Knutton et al., 1987, 1989; Knutton, 1995). Thesecharacteristic “attaching-and-effacing” lesions are formed on epithelial cells in athree-stage process. After initial adhesion, mediated by bundle-forming (type IV)fimbriae, a type III secretory system is activated in E. coli allowing secretion of areceptor (translocated intimin receptor) into the epithelial cell membrane which actsas a receptor for the E. coli outer membrane protein intimin. This leads to reorgan-ization of the cellular actin cytoskeleton and formation of the characteristic elevatedpedestal to which E. coli is bound.

2.4. Host cell ligands

A wide range of potential receptors is present on the eukaryotic cell membraneinvolved in the normal cellular functions of transport, signal transduction andcell–cell communication, and bacteria can bind to many of these molecules. In addi-tion, proteins of the extracellular matrix, such as fibronectin, fibrinogen and colla-gen, are receptor molecules for which specific adhesins have been characterized inbacteria such as Staphylococcus aureus (Smeltzer et al., 1997). In glycoproteins, thesugar residues commonly act as receptor ligands for fimbriae; binding of E. coli toeukaryotic cells via type 1 or type 5 fimbriae is inhibited by mannose leading to theconclusion that mannose-containing glycoproteins are cellular targets for binding bythis organism (Krogfeld et al., 1990). Other sugars, e.g. fucose and galactose havebeen similarly identified as receptor targets for other types of fimbriae (Ofek andDoyle, 1994). Similar work by Wang and Leung (2000) has shown that strains ofVibrio anguillarum differ in the types of receptors used. Two invasive strains of theorganism, G/Virus/5(3) and 811218-5W adhered strongly to three different fish tissue culture cell lines. Adherence of strain G/Virus/5(3), and of nine other vibrios,was inhibited by galactose-containing sugars, but adherence by strain 811218-5Wwas not affected by a range of sugars tested. As no fimbriae could be detected ineither strain it was concluded that non-fimbrial adhesins were involved in both cases(Wang and Leung, 2000).

The ability of Photo. damselae subsp. piscicida to adhere to fish tissue culturecell lines was inhibited by galactose and mannose but not fucose, indicating a pos-sible glycoprotein target for adhesins of this organism (Magarinos et al., 1996).However, prior treatment of bacteria with proteinase K did not affect their capacity


to bind to tissue culture cells and the Photo. damselae subsp. piscicida adhesinremains unidentified.

2.5. Consequences of bacteria/ligand interactions

As noted above, many cell surface molecules are receptors involved in transmem-brane signalling. For bacteria, interaction with eukaryotic cells can lead to alteredcell growth patterns, induction of adhesins, e.g. for enteropathogenic E. coli(Donnenberg and Kaper, 1992), or secretion of proteins required for invasion, e.g. for Yersinia (Cornelis and Wolf-Watz, 1997). For the eukaryotic cell, uptake of bacteria may result in cytokine release (Wilson et al., 1998), morphology alteration(enteropathogenic E. coli) (Nataro and Kaper, 1998) or intercellular adhesion molecule synthesis may be stimulated (enteroinvasive E. coli).

3. PROTECTION AGAINST BACTERIAL ADHESION

3.1. Mucus

The internal surface of the host is the first defence barrier to infection. Intestinalmucins secreted by specialized epithelial goblet cells located in the intestinal entero-cytes form a viscous, hydrated blanket on the surface of the intestinal mucosa thatprotects the delicate columnar epithelium. This is thought to be a vital componentof the intestinal mucosal barrier in prevention of colonization by pathogens in bothfish and endothermic animals (Florey, 1962; Forstner, 1978; Westerdahl et al., 1991;Maxson et al., 1994; Henderson et al., 1999; Mims et al., 2000). Gastrointestinalmucus is thought to have three major functions: 1) protection of the underlyingmucosa from chemical and physical damage, 2) lubrication of the mucosal surface,and 3) to provide a barrier against entero-adherence of pathogenic organisms to theunderlying mucosal epithelium. Intestinal mucus is composed almost entirely ofwater (90–95%) and the electrolyte composition is similar to plasma, accounting forabout 1% of the mucus weight. The remaining 4−10% is composed of high molec-ular weight glycoproteins (mucins), consisting of a protein core with numerous carbohydrate (fucose and galactose) side chains. Hydrolysis of intestinal mucusmaterial of rainbow trout liberated increased amounts of N-acetylgalactosamine andN-acetylglucosamine (O’Toole et al., 1999), indicating that these carbohydratesmay be present as mucin-bound moieties in fish intestinal mucus as is the case formucus from other animal species (Roussel et al., 1988). The majority of intestinalmucus-associated lipids in rainbow trout partitioned to the organic phase duringextraction with chloroform/methanol and this contained saturated and unsaturatedfree fatty acids, phospholipids, bile acid, cholesterol, and monoglycerides anddiglycerides (O’Toole et al., 1999).

The mucous blanket is constantly renewed by the secretion of high molecularweight glycoproteins from individual goblet cells throughout the epithelium.


Goblet cells differentiate in the lower portion of the crypts of both small and largeintestine and gradually migrate on to the villi or mucosal surface.

In an early study on histopathological changes caused by V. anguillarum,Ransom et al. (1984) found large amounts of goblet (mucus producing) cells in theanterior part of the GI tract of infected chum salmon. The first reaction of Arcticcharr (Salvelinus alpinus L.) infected by pathogenic bacteria (A. salmonicida) is toslough off the infected mucus by increasing goblet cell production (fig. 4A) com-pared to uninfected fish (fig. 4B). A similar reaction to that found in infected fish isalso observed in rabbits and rats infected by pathogenic bacteria (Enss et al., 1966;Mantle et al., 1989, 1991), and this may be considered a normal host response toparticular intestinal infections (Mims et al., 2000).

Gastrointestinal mucus is rich in nutrients that organisms, including pathogens,may utilize for growth (Blomberg et al., 1995; Wadolkowski et al., 1988). Manyendothermic studies have implicated growth in mucus as a critical factor for intes-tinal colonization by pathogens and several outer membrane proteins are necessaryfor establishment of an infection focus (Freter et al., 1983; Myhal et al., 1982;Krivan et al., 1992; Burghoff et al., 1993). Olsson et al. (1992) suggested that the GItract is a site of colonization of V. anguillarum as the pathogen could utilize dilutedturbot (Scophthalmus maximus L.) intestinal mucus as its sole nutrient source. Morerecently, Garcia et al. (1997) examined the ability of V. anguillarum to grow insalmon intestinal mucus, which they concluded is an excellent growth medium forthis species. This is an important aspect of the pathogenesis of this organism.

3.2. Ultrastructural changes in enterocytes caused by dietary manipulation

Recently, Olsen et al. (1999, 2000) showed that extensive accumulation of lipiddroplets occurred in Arctic charr enterocytes when the fish were fed a diet contain-ing linseed oil and this caused significant damage to the epithelium with focal loss


Fig. 4. Light microscopic view of villi in the midgut from Arctic charr (Salvelinus alpinus L.) fed soybeanoil (A) post and (B) prior to challenge with Aeromonas salmonicida subsp. salmonicida. Note the substantialmore conspicuous goblet cells (arrows) along the villi of infected fish. After Lødemel et al. (2001).

of enterocytes and consequent loss of the epithelial barrier. Such damage (fig. 5) islikely to be pathological and therefore detrimental to fish health. Rupture in themembranous system may also represent a major microbial infection route for poten-tially pathogenic bacteria provided they are present in sufficient numbers in the gut.

The prebiotic potential of dietary fibres is well known in endothermic animals(Gibson, 1998), and may also have interesting applications in aquaculture. However,a recent study clearly demonstrated that feeding Arctic charr a diet supplemented with15% inulin led to the occurrence of a large number of spherical lamellar bodies in theenterocytes of the pyloric caeca and the hindgut. These structures were not observedwhen fish were fed 15% dextrin (Olsen et al., 2001). Feeding inulin had a destructiveeffect on microvillus organization which may increase translocation of pathogenicbacteria if they are present in relatively high concentrations in the GI tract.

3.3. Autochthonous bacteria and antagonistic activity

Savage (1983) defined bacteria isolated from the digestive tract as being eitherindigenous (autochthonous) or transient (allochthonous) depending on whether ornot they are able to colonize epithelial surface of the digestive tract of the host animal. Recently, Ringø and Birkbeck (1999) presented a list of criteria for testingautochthony of bacteria from the GI tract of fish. These were that they should i) befound in healthy animals, ii) colonize early stages and persist throughout life, iii) befound in both free-living and hatchery-cultured fish, iv) grow anaerobically, and v) be found associated with epithelial mucosa in the digestive tract. The presence ofan autochthonous microflora fitting the above criteria was demonstrated recently byRingø et al. (2002) in that bacteria in the gut were found closely associated with theintestinal epithelium and between the microvilli. On the basis of this observation,one might hypothesize that the autochthonous microflora of fish which is associatedclosely with the intestinal epithelium forms a barrier serving as the first defence tolimit direct attachment or interaction of pathogenic bacteria with the mucosa asreported for endothermic animals (van der Waaij et al., 1972; Snoeyenbos, 1979;


Fig. 5. Linseed-oil-fed Arctic charr(Salvelinus alpinus L.). These enterocytes are obviously damaged by the fat vacuoles(droplets). Note especially cellular membraneruptures (arrows). Bar = 7 μm. After Olsen et al. (1999).

Tancrède, 1992). In fish, situations such as stress, antibiotic administration, or evensmall dietary changes affect the GI tract microflora. Stability of this microflora isimportant in natural resistance to infections produced by bacterial pathogens in theintestinal tract. The existence of antibacterial substances produced by bacteria isolated from the digestive tract of fish has been demonstrated in several studies(Schrøder et al., 1980; Dopazo et al., 1988; Strøm, 1988; Westerdahl et al., 1991;Olsson et al., 1992; Bergh, 1995; Sugita et al., 1996, 1997, 1998; Jöborn et al., 1997;Gram et al., 1999; Ringø, 1999; Ringø et al., 2000). However, a recent study by Gramet al. (2001) demonstrated that in vitro activity in well diffusion assays and broth cul-tures cannot be used to predict a possible in vivo effect even if a reduction of in vivomortality was observed in another system (Gram et al., 1999). These studies under-line the importance of developing and testing cultures for each specific combinationof different pathogens, different fish species and environment that might occur.

4. BACTERIAL INVASION AND TRANSLOCATION MECHANISMS

The indigenous intestinal flora is prevented from gaining access to other sites in thebody by a single epithelial cell layer on the mucosa. In endothermic animals the M cells of the intestinal epithelium are specialized structures that may allow naturalentry of bacterial pathogens (Jones et al., 1995; Neutra et al., 1996; Vazques-Torresand Fang, 2000). Information about the interactions between intracellular pathogenicbacteria and M cells in fish is not available, however, and is a topic of further studies.

The mechanisms by which bacteria can translocate from the gut to appear inother organs are an important phenomenon in the pathogenesis of “opportunistic”infections by indigenous intestinal bacteria (Finlay and Falkow, 1997). Once insidea host cell, pathogens have a limited number of ways to ensure their survivalwhether remaining within a host vacuole or escaping into the cytoplasm.

In endothermic animals the primary defence mechanisms preventing indigenousbacteria from translocating from the gastrointestinal tract are: a) a stable GI tractmicroflora preventing bacterial overgrowth of certain indigenous bacteria or colo-nization by more pathogenic exogenous bacteria, b) the host immune defences andc) an intact mucosal barrier. More than one of these defence mechanisms can beinvolved, depending upon the animal model or clinical situation. An example of thisis Lactobacillus casei, which can prevent E. coli infection in a neonatal rabbit modeland inhibits translocation of E. coli in an enterocyte cell culture model (Mattar et al.,2001). However, in fish these defence mechanisms are not well understood.

The pathogenesis of V. cholerae infections in mammals is primarily a non-invasive toxin-mediated gut infection but such infections have not been found infish. Translocation of intact Vibrio antigens and bacterial cells by endocytosis hasbeen reported in the gastrointestinal tract of fish larvae (Hansen and Olafsen, 1990,1999; Hansen et al., 1992; Olafsen and Hansen, 1992; Grisez et al., 1996). However,when discussing endocytosis, the development of the digestive tract is an important


factor to be considered. At the time of hatching, the digestive tract of fish is an undif-ferentiated straight tube which is morphologically and physiologically less elaboratethan that of the adult (Govoni et al., 1986). However, endocytosis was demonstratedin pyloric caeca (fig. 6), midgut (fig. 7) and hindgut (fig. 8) of adult Arctic charr(Ringø et al., 2002).


Fig. 6. Transmission electron micrograph ofthe apical regions of enterocytes in the pyloric caecum of adult Arctic charr (Salvelinusalpinus L.). Bacterial profiles are seen scatteredat different levels within the brush border fromthe tips to bases of microvilli. In addition, onebacterial profile (arrowhead) is seen to becontained in an internalized, membrane-boundendocytic vacuole. × 15000. After Ringø et al.(2001).

Fig. 7. High power transmission electronmicrograph of the midgut of adult Arctic charr(Salvelinus alpinus L.). The opposed surfacesof two enterocytes are shown. Both cells haveappreciable numbers of bacterial profilesbetween their microvilli. Note the internalizedbacterium in the subapical cytoplasm (arrow-head). × 15000. After Ringø et al. (2001).

5. INVASION OF HOST CELLS

Entry into host cells is a specialized strategy for survival and multiplication utilizedby a number of pathogens which can exploit existing eukaryotic internalizationpathways (Finlay and Falkow, 1989, 1997; Sansonetti, 1993). Three general mechanisms are recognized by which bacteria can invade epithelial cells. The mostcommon method, as employed by Yersinia, Shigella and Salmonella, is by inducingrearrangement of the actin cytoskeleton of the epithelial cell. EnteropathogenicYersinia spp. induce uptake into endocytic vacuoles of epithelial cells followingclose contact of the bacteria at many points to the cell surface (zippering). Thisinvolves three adhesins – the invasin, Ail and YadA proteins – and interactionbetween invasin and its cell-surface receptor, α5β1 integrin induces actin cytoskele-ton rearrangement via a protein tyrosine kinase signalling system (Cornelius andWolf-Watz, 1997; Lloyd et al., 2001). Invasion by Shigella is dependent upon possession of a 220 kb plasmid encoding 32 invasion-associated genes (Menard et al., 1996), including those for a type III secretion system which directly secretesShigella proteins into the cytoplasm of the epithelial cell; this induces actincytoskeleton rearrangement and pseudopodia formation to internalize the bacterialcell. Once internalized, lysis of the vesicle is mediated by a Shigella protein releas-ing the organism into the cytoplasm where it can multiply and spread through thecytoplasm propelled by an actin “tail” (Menard et al., 1996). Inhibitors of actinpolymerization, such as cytochalasin D, block entry of such pathogens into cells.However, invasion of epithelial cells by Campylobacter jejuni is unaffected bycytochalasin D but is sensitive to the microtubule depolymerizing drug colchicine,


Fig. 8. Transmission electron micrographshowing bacteria associated with the microvilliof enterocytes in the hindgut of adult Arcticcharr (Salvelinus alpinus L.). Enterocytes inthis region show endocytic activity and arecharacterized by large numbers of intracyto-plasmic vacuoles (V) with contents of varyingelectron density. An internalized bacterium isdiscernible (arrowhead). × 15000. After Ringøet al. (2002).

indicating an actin-independent, microtubule-dependent pathway of entry (Russelland Blake, 1994; Biswas et al., 2000). A direct invasive mode of entry is utilized byrickettsiae, which bind to the phospholipid cell membrane and gains entry viaexpression of phospholipase A (Silverman et al., 1992). For fish pathogens, exam-ples are known of invasion of epithelial cells. Although none has been characterizedto the extent of human pathogens, current knowledge is summarized below.

5.1. Aeromonas

Aeromonas salmonicida is the causative agent of furunculosis, a disease whichcaused very serious losses in European aquaculture in the early 1990s (Munro andHastings, 1993) and which had previously caused major epizootics in wild fish(Mackie et al., 1930). All salmonid species are affected, but Atlantic salmon andbrook trout appear to be more susceptible to infection than rainbow trout or Pacificsalmon (Cipriano, 1983). The intestine has long been considered a route of infectionfor A. salmonicida as Plehn (1911) found inflammation of the gut to be a commoncharacteristic of furunculosis. However, there is still debate about the route of entryof this pathogen (Bernoth et al., 1997). The presence of A. salmonicida in the intes-tine of Atlantic salmon has been demonstrated using an enzyme-linked immuno-sorbent assay by Hiney et al. (1994), who suggested that the intestine could be theprimary location of A. salmonicida in stress-inducible infections. O’Brien et al.(1994) also detected A. salmonicida in faeces using a species-specific DNA probe,in conjunction with a polymerase chain reaction (PCR) assay. Although the organ-ism can be detected in the intestinal tract in the above assays, McCarthy (1977)failed to infect brown trout (Salmo trutta) either by administering food pelletssoaked in a culture of A. salmonicida, or by direct intubation into the stomach. In the latter case, 105–106 A. salmonicida were recovered per ml homogenized stom-ach within 12 h of introduction, no organisms could be recovered by 48 h and nomortalities occurred, despite recovery of low numbers of organisms fromhomogenates of kidney within 5 h. Despite failing to cause disease by the intestinalroute the organism killed five of six fish exposed for 5 days to an aqueous suspen-sion of the bacteria (106 cells/ml).

In experimental infections of turbot (Scophthalmus maximus L.) and halibut(Hippoglossus hippoglossus L.) yolk sac larvae with A. salmonicida subsp. salmonicida, Bergh et al. (1997) failed to re-isolate the pathogen from halibut larvae,but using immunohistochemical techniques showed the bacteria to be present in theintestinal lumen of some turbot larvae, but not associated to mucus or gut microvilli.

A recent study by Lødemel et al. (2001) clearly demonstrated that A. salmonicidasubsp. salmonicida could be detected within enterocytes of the midgut of Arctic charr(Salvelinus alpinus L.).

In summary, although A. salmonicida can be detected in the intestine of infectedfish there is still doubt that this is the principal route by which systemic infection


occurs, although translocation of organisms from stomach to kidney has beendemonstrated (McCarthy, 1977).

A range of freshwater fish are susceptible to motile aeromonad septicaemiacaused by A. hydrophila (Thune et al., 1993) and this organism is capable of bindingto collagen and fibronectin (Ascencio et al., 1991), and invading EPC (epitheliosumpapillosum of carp) tissue culture cells (Tan et al., 1998). Studies with inhibitors oftyrosine kinase, protein kinase C and protein tyrosine phosphatase indicated that theorganism initiated a signalling cascade involving tyrosine kinase, leading to actinmicrofilament reorganization involving actin “clouds” (Tan et al., 1998).

5.2. Edwardsiella

Two species of this genus, E. ictaluri and E. tarda are serious pathogens of fish(Plumb, 1993), causing distinctly different diseases in a range of fish species.

Edwardsiella septicaemia, which affects warm water fish is widely distributed inthe environment and can cause severe losses in farmed catfish, Ictalurus punctatus(Plumb, 1993). Although Darwish et al. (2000) found no histological lesions in theintestine of catfish during experimentally induced infections, a different type ofstudy by Ling et al. (2000) employing green fluorescent protein (GFP)-labelled bac-teria showed that 3 days after intramuscular injection of 1.2 × 105 E. tarda into bluegorami approximately 106 bacteria were recovered from the intestine, although thehighest concentrations of bacteria were found in the muscle and liver. However, E. tarda is not considered a pathogen with significant involvement of the gut ininfection. Nevertheless, it has a pronounced capacity to invade both human and fishtissue culture cells (Janda et al., 1991; Ling et al., 2000). The invasion of both tissueculture cell types by E. tarda was sensitive to cytochalasin D (Janda et al., 1991;Ling et al., 2000), and in fish cells was also dependent on protein tyrosine kinaseactivity (Ling et al., 2000).

The second species, E. ictaluri, causes enteric septicaemia of catfish which canresult in high mortalities. Two disease conditions are known with infection of brainvia the olfactory organ or the intestine (Shotts et al., 1986; Francis-Floyd et al.,1987). Doses of 5 × 109 bacteria were intubated into the stomach of fingerling catfish and within 2 weeks fish developed enteritis and other chronic lesions.Horizontal transmission occurred to cohabiting fish, which also developed lesionsbeginning in the intestine (Shotts et al., 1986). As yet, the invasion pathway from thegut to other tissues and organs has not been established.

5.3. Photobacterium damselae subsp. piscicida

This organism was found to adhere to tissue sections of intestine from sea breamSparus aurata, sea bass Dicentrarchus labrax, and turbot Scophthalmus maximus atconcentrations of 104–105 per gram by Magarinos et al. (1996).


Evaluation of the invasive capacities of the Photo. damselae subsp. piscicida ondifferent poikilothermic cell lines indicated that according to the Janda index (Jandaet al., 1991), the strains studied were weakly or moderate invasive, with the numberof intracellular bacteria ranging from 101 to 103. Photo. damselae subsp. piscicidawas able to invade CHSE-214 tissue culture cells and to remain viable for at least 2 days inside the infected cells.

5.4. Piscirickettsia salmonis

Piscirickettsia salmonis is an obligate, intracellular, Gram-negative organism andsuch fastidious bacteria have been increasingly detected as emerging pathogens in arange of fish species in different geographic locations (Fryer and Mauel, 1997). In 1990 it was recognized that the causative agent responsible for the loss of 1.5 million coho salmon in the previous year (Cvitanich et al., 1990, 1991; Fryer et al.,1990; Branson and Diaz-Munoz, 1991; Garces et al., 1991) was a rickettsial agentof a new genus and species (Fryer et al., 1992). Whereas the Rickettsiaceae aremembers of the α-Proteobacteria, P. salmonis is assigned to the γ-Proteobacteria.The disease was termed salmonid rickettsial septicaemia because of the systemicnature of the disease (Cvitanich et al., 1991). Several organs were affected in dis-eased fish, including the intestine, which was severely damaged with necrosis andinflammation of the lamina propria and sloughing off of epithelial cells (Bransonand Diaz-Muoz, 1991). The route of infection was studied by Smith et al. (1999)who investigated various routes as possible portals of entry for the pathogen.Subcutaneous injection of P. salmonis (104 TCID50) resulted in 100% cumulativemortality of fish by day 33 post injection. Application to the skin or gills of patchessoaked in P. salmonis (104.2 TCID50 per patch) resulted in 52 and 24% mortalities,respectively, whereas 24 and 2% cumulative mortalities occurred following intestinalor gastric intubation (104 TCID50 administered in both cases). The authors concludedthat rickettsia could infect the fish directly through the skin or gills and that the intestinal route was not the normal route of infection.

5.5. Vibrio anguillarum

Vibrio anguillarum is an important pathogen of marine and estuarine fish speciesand is the causative agent of vibriosis. This disease is one of the major bacterial dis-eases affecting fish, as well as bivalves and crustaceans (Austin and Austin, 1999),and vibriosis can cause substantial losses to the aquaculture industry. Vibriosis ischaracterized by deep focal necrotizing myositis and subdermal haemorrhages, withthe intestine and rectum becoming swollen and filled with fluid (Horne et al., 1977;Munn, 1977). The GI tract of fish appears to be a site of colonization and amplifi-cation for pathogenic Vibrio species (Horne and Baxendale, 1983; Ransom et al.,1984; Olsson et al., 1996), and Olsson et al. (1998) recently demonstrated that orally


ingested V. anguillarum can survive passage through the stomach of feeding juve-nile turbot (Scophthalmus maximus L.). Vibrio anguillarum and V. ordalii have beenfound primarily in the pyloric caeca and the intestinal tracts of three species ofPacific salmon (chum salmon, Oncorhynchus keta; coho salmon, Oncorhynchuskisutch; and chinook salmon, Oncorhynchus tshawytscha) (Ransom et al., 1984).In addition, Olsson (1995) and Olsson et al. (1996) demonstrated that the GI tractcan serve as a portal of entry for V. anguillarum and it can utilize intestinal mucusas its sole nutrient source (Olsson et al., 1992; Garcia et al., 1997). Although some evidence indicates that V. anguillarum can invade fish either via the skin or the GItract, Grisez et al. (1996) showed that the organism is transported across the intes-tinal epithelium by endocytosis. Chemotactic motility mediated by a single polarsheathed flagellum is essential for virulence as bacteria deficient in this activitywere unable to infect fish when administered by immersion in bacteria-containingwater but were virulent when given by intraperitoneal injection (O’Toole et al.,1996). These findings imply that V. anguillarum responds chemotactically to certainfish-derived products in a manner that promotes the infection process prior to penetration of the fish epithelium.

Recently, it was shown that V. anguillarum exhibited a stronger chemotacticresponse towards intestinal mucus than towards skin (O’Toole et al., 1999). Of thefree amino acids identified in the intestinal mucus, glutamic acid, glutamine,glycine, histidine, isoleucine, leucine, serine and threonine, and carbohydrates suchas fucose, glucose, mannose and xylose behaved as chemoattractants, while the lipidcomponents identified, bile acid, taurocholic acid and taurochenodeoxycholic acidinduced only a weak chemotactic response. A combination of all individualchemoattractants identified from mucus reconstituted a high level of chemotacticactivity similar to that present in the intestinal mucus homogenate. On the basis ofthese results, the authors proposed that multiple chemoattractants in rainbow troutmucus indicated a strong relationship between chemotaxis and bacterial virulence.

The invasion mechanism of vibrios for fish cells has been investigated recently byWang et al. (1998) in a comparison of 24 isolates of seven different species. Thirteenisolates were invasive for gruntfin (GF) and EPC tissue culture cell lines including allfive V. vulnificus and both V. harveyii isolates. Of the 11 V. anguillarum isolatestested, three were invasive, two of which adhered strongly to EPC cells. Cytochalasin Dinhibited invasion by both strains although one was also sensitive to inhibition by vin-cristin, a microtubule depolymerizing agent, indicating different routes of invasion forthe two strains. This difference was confirmed by the difference in response of thestrains to inhibitors of the signalling molecules protein kinase C and tyrosine kinase.

5.6. Streptococcosis

Streptococcosis is a septicaemic disease that affects freshwater and marine fish inboth farmed and wild populations. Among commercially important fish species, this


disease has been reported worldwide in yellowtail (Seriola spp.), eels (Anguillajaponica), menhaden (Brevoortia patronus), striped mullet (Mugil cephalus),striped bass (Morone saxatilis) and turbot. Romalde et al. (1996) demonstrated thecapacity of Enterococcus sp. to overcome adverse conditions in the stomach whenassociated with food or faecal materials, since the pathogen was able to establish aninfective state and to produce mortalities after 16 to 20 days post ingestion.

6. VIRUSES

With the availability of effective vaccines against many major bacterial fish pathogens(Gudding et al., 1997) agents such as infectious pancreatic necrosis (IPN) virus, infec-tious salmon anaemia (ISA) virus, viral haemorrhagic septicaemia (VHS) virus,infectious haematopoietic necrosis (IHN) virus and nodavirus have emerged as moreprominent threats to aquaculture. In mammals, the enteroviruses, rotaviruses, corona-viruses and Norwalk virus group are important causes of diarrhoeal disease trans-mitted by the faecal–oral route (Mims et al., 2000). For poliovirus the initialreplication in the GI tract can be followed by invasion of the bloodstream and pene-tration of the blood–brain barrier to cause paralytic poliomyelitis (Mims et al., 2000).

In salmonids, IPN virus is a serious pathogen causing major losses in Atlanticsalmon aquaculture in Norway, Scotland and Chile (Smail and Munro, 2001). As itsname implies this virus causes significant necrosis of the pancreas in salmonids butother organs, including the intestinal tract, may also be affected (Wolf, 1988).Pathological changes in the intestinal tract have also been shown in larval sea bass(Bonami et al., 1983) and larval halibut (Biering et al., 1994). In the latter study,focal necrosis was observed in the intestinal tract with sloughing off of epithelialcells, and the GI tract was considered the most likely route of entry and replicationfor the virus (Bergh et al., 2002). However, there was no evidence of damage to thepancreas in larval halibut.

Viral encephalopathy and retinopathy (VER), caused by nodaviruses, is arecently recognized serious disease of Atlantic halibut which poses a serious threatto larval culture of this fish (Grotmol et al., 1995, 1997; Munday and Nakai, 1997).Although pathology is largely restricted to lesions in the brain, spinal chord andretina (Grotmol et al., 1995), experimental infection models indicate that the intes-tinal epithelium is the probable route of entry for this virus into the larval fish(Grotmol et al., 1999). However, as with IPN virus, little is known of the pathogenicmechanisms involved in invasion from the intestinal tract to the sites where signifi-cant pathological damage is caused, and this awaits further investigation.

7. THE EFFECT OF DIET ON DISEASE RESISTANCE

Intensive fish production has increased the risk of infectious diseases. Therefore,there is a growing need to find alternatives to antibiotic treatments for disease con-trol, as indiscriminate use of antibiotics in many parts of the aquaculture industry


has led to the development of antibiotic resistance in bacteria. Nutritional status isconsidered an important factor in determining disease resistance. The complex rela-tionship between nutritional status, immune function and disease resistance hasbeen documented for higher vertebrates in several comprehensive reviews andbooks (Gershwin et al., 1985; Chandra, 1988; Bendich and Chandra, 1990). Theinfluence of dietary factors on disease resistance in fish has been extensivelyreviewed (Lall, 1988; Landolt, 1989; Blazer, 1992; Lall and Olivier, 1993; Waagbø,1994; Olivier, 1997), and micronutrients such as vitamins have received particularattention. Studies on the essential fatty acid, vitamin and trace element requirementsof several warm and cold water fish have demonstrated their integral role in themaintenance of epithelial barriers of skin and the gastrointestinal tract. Althoughthere is some information on the relationship between disease resistance and dietarylipid (Salte et al., 1988; Erdal et al., 1991; Obach et al., 1993; Waagbø et al., 1993;Li et al., 1994; Bell et al., 1996; Thompson et al., 1996), there is a lack of informa-tion about the functional role of dietary lipid on intestinal microbiota, their antago-nism and disease resistance. However, a recent study showed clear differences in thegut microbiota of fish fed different oils (post and prior to challenge) and the abilityof the gut microbiota to inhibit growth of three fish pathogens (A. salmonicidasubsp. salmonicida, V. anguillarum and V. salmonicida) (Ringø et al., 2002). Also,Lødemel et al. (2001) clearly demonstrated that survival of Arctic charr after challenge with A. salmonicida subsp. salmonicida was improved by dietary soybeanoil. These results are in agreement with those reported by Hardy (1997) thatreplacement of dietary fish oil with plant- or animal-derived fats increases resistanceof catfish (Ictalurus punctatus) to disease caused by experimental challenge with E. ictaluri.


Bacterial and viral diarrhoeal diseases are major causes of mortality and morbidityin mammals but no equivalent diseases are recognized in fish, presumably becausedramatic fluid loss does not occur so readily in an aquatic environment. However,the GI tract still presents a route of infection, especially for opportunistic bacteriapresent on ingested food particles, and there is clear evidence for this as a route forinvasion to affect other organs and tissues.

The main reasons why studies on fish pathogenic bacteria have lagged behindthose of mammalian pathogens is because intensive aquaculture has developed quiterecently as a significant industry, several of the pathogens are novel, and there hasbeen a relatively small research effort in this field, in comparison with human andveterinary medicine. The past decade has seen major developments in methodologyfor studying microbial pathogenicity, and the techniques applied to humanpathogens are only now being applied to fish pathogens (O’Toole et al., 1996, 1999;Tan et al., 1998; Ling et al., 2000; Mathew et al., 2001). Undoubtedly, the most


significant development in microbiology for 50 years has been the genome sequencedetermination for many prokaryotes. Since the first complete sequence was pub-lished in 1995, a total of 56 genome sequences have been completed to date and afurther 210 are in progress (see www.tigr.org and www.integratedgenomics.com).Those in progress include genomes for the fish pathogens A. salmonicida and P. salmonis and this will provide unique insights into the potential pathogenic mech-anisms of these bacteria, including evolutionary distances between P. salmonis andRickettsia prowazekii and whether pseudogenes are also prevalent in P. salmonis(Andersson et al., 1998; Andersson and Andersson, 1999).

Other methods, including the use of expressed markers such as green fluorescentprotein and laser confocal microscopy (e.g. Ling et al., 2000), will provide moredefinitive analysis of pathways of invasion by pathogens taken up via the GI tract.One area in which there is particular deficiency at present is in the nature of fish-cell-surface colonization by bacteria and any downstream signalling which occurs.This would be of considerable practical value in designing pathogen preventionstrategies using probiotic bacteria to prevent colonization by pathogens.

REFERENCES

Andersson, J.O., Andersson, S.G., 1999. Insights into the evolutionary process of genome degradation.Curr. Opin. Genet. Dev. 9, 664–671.

Andersson, S.G., Zomorodipour, A., Andersson, J.O., Sicheritz-Ponten, T., Alsmark, U.C.M., Podowski,R.M., Naslund, A.K., Eriksson, A.-S., Winkler, H.H., Kurland, C.G., 1998. The genome sequence ofRickettsia prowazekii and the origin of mitochondria. Nature 396, 133−140.

Ascencio, F., Ljungh, Å., Wadström, T., 1991. Comparative study of extracellular matrix protein bindingto Aeromonas hydrophila. Microbios 65, 135–146.

Austin, B., Austin, D.A., 1999. Bacterial Fish Pathogens: Diseases in Farmed and Wild Fish, 3rd edition.Ellis Horwood Ltd., Chichester.

Baudin Laurencin, F., Germon, E., 1987. Experimental infections of rainbow trout, Salmo gairdneri R.,by dipping in suspensions of Vibrio anguillarum: ways of bacterial penetration: influence of temper-ature and salinity. Aquaculture 67, 203−205.

Bell, J.G., Ashton, I., Secombes, C.J., Weitzel, B.R., Dick, J.R., Sargent, J.R., 1996. Dietary lipid affectsphospholipid fatty acid composition, eicosanoid production and immune function in Atlantic salmon(Salmo salar). Prostagl. Leuk. Essent. Fatty Acids 54, 173−182.

Belland, R.J., Trust, T.J., 1985. Synthesis, export, and assembly of Aeromonas salmonicida A-layer ana-lyzed by transposon mutagenesis. J. Bacteriol. 163, 877−881.

Bendich, A., Chandra, R.K., 1990. Micronutrients and Immune Functions. New York Academy ofSciences, New York.

Bergh, Ø., 1995. Bacteria associated with early life stages of halibut, Hippoglossus hippoglossus L.,inhibit growth of a pathogenic Vibrio sp. J. Fish Dis. 18, 31−40.

Bergh, Ø., Hjeltnes, B., Skiftesvik, A.B., 1997. Experimental infections of turbot Scophthalmus maximusand halibut Hippoglossus hippoglossus yolk sac larvae with Aeromonas salmonicida subsp. salmonicida.Disease Aquat. Org. 29, 13−20.

Bergh, Ø., Nilsen, F., Samuelsen, O.B., 2002. Diseases, prophylaxis and treatment of the Atlantic halibutHippoglossus hippoglossus: a review. Disease Aquat. Org. 41, 57−74.

Bernoth, E.-M., Ellis, A.E., Midtlyng, P.J., Olivier, G., Smith, P., 1997. Furunculosis. Academic Press,San Diego.

Biering, E., Nilsen, F., Rodseth, O.M., Glette, J., 1994. Susceptibility of Atlantic halibut Hippoglossushippoglossus to infectious pancreatic necrosis virus. Dis. Aquat. Org. 20, 183–190.


Biswas, D., Itoh, K., Sasakawa, C., 2000. Uptake pathways of clinical and healthy animal isolates ofCampylobacter jejuni into INT-407 cells. FEMS Immunol. Med. Microbiol. 29, 203−211.

Blazer, V.S., 1992. Nutrition and disease resistance. Ann. Rev. Fish Dis. 2, 309−323.Blomberg, L., Gustafsson, L., Cohen, P.S., Conway, P.L., Blomberg, A., 1995. Growth of Escherichia

coli K88 in piglet ileal mucus: protein expression as an indicator of type of metabolism. J. Bacteriol.177, 6695−6703.

Bonami, J.R., Cousserans, F., Weppe, M., Hill, B.J., 1983. Mortalities in hatchery-reared sea bass fryassociated with a birnavirus. Bull. Eur. Assoc. Fish Pathol. 3, 41.

Branson, E.J., Diaz-Munoz, D.N., 1991. Description of a new disease condition occurring in farmed cohosalmon, Oncorhynchus kisutch (Walbaum), in South America. J. Fish Dis. 14, 147−156.

Brinton, C.C., 1965. The structure, function, synthesis and genetic control of bacterial pili and a molecularmodel for DNA and RNA transport in Gram negative bacteria. Trans. N.Y. Acad. Sci. 27, 1003−1054.

Burghoff, R.L., Pallesen, L., Krogfelt, K.A., Newman, J.V., Richardson, M., Bliss, J.L., Laux, D.C.,Cohen, P.S., 1993. Utilization of the mouse large intestine to select a Escherichia coli F-18 DNAsequence that enhances colonizing ability and stimulates synthesis of type I fimbriae. Infect. Immun.61, 1293−1300.

Chair, M., Dehasque, M., van Poucke, S., Nelis, H., Sorgeloos, P., de Leener, A.P., 1994. An oral chal-lenge for turbot larvae with Vibrio anguillarum. Aquacult. Int. 2, 270−272.

Chandra, R.K., 1988. Nutrition and Immunology. Alan R. Liss, New York.Cipriano, R.C., 1983. Resistance of salmonids to Aeromonas salmonicida: relation between agglutinins

and neutralizing activities. Trans. Amer. Fish. Soc. 112, 95−99.Cornelis, G.R., Wolf-Watz, H., 1997. The Yersinia Yop virulon: a bacterial system for subverting eukary-

otic cells. Molec. Microbiol. 23, 861−867.Cvitanich, J.D., Garate, O., Smith, C.E., 1990. Etiological agent in a Chilean coho disease isolated and

confirmed by Koch’s postulates. FHS/AFS Newsletter 18, 1−2.Cvitanich, J.D., Garate, O., Smith, C.E., 1991. The isolation of a rickettsia-like organism causing disease

and mortality in Chilean salmonids and its confirmation by Koch’s postulate. J. Fish Dis. 14, 121−145.Darwish, A., Plumb, J.A., Newton, J.C., 2000. Histopathology and pathogenesis of experimental infec-

tion with Edwardsiella tarda in channel catfish. J. Aquat. Anim. Health 12, 255–266.Donnenberg, M.S., Kaper, J.B., 1992. Enteropathogenic Escherichia coli. Infect. Immun. 60, 3953−3961.Dopazo, C.P., Lemos, M.L., Lodeiros, C., Bolinches, J., Barja, J.L., Toranzo, A.E., 1988. Inhibitory activ-

ity of antibiotic producing marine bacteria against fish pathogens. J. Appl. Bacteriol. 65, 97−101.Enss, M.L., Müller, H., Schmidt-Witting, U., Kownatzki, R., Coenen, M., Hedrich, H.J., 1966. Effects of

perorally applied endotoxin on colonic mucins of germfree rats. Scand. J. Gastroenterol. 31, 868−874.Erdal, J.I., Evensen, Ø., Kaurstad, O.K., Lillehaug, A., Solbakken, R., Thorud, K., 1991. Relationship

between diet and immune response in Atlantic salmon (Salmo salar L.) after feeding various levelsof ascorbic acid and omega-3 fatty acids. Aquaculture 98, 363−379.

Finlay, B.B., Falkow, S., 1989. Common themes in microbial pathogenicity. Microbiol. Rev. 53, 210−230.Finlay, B.B., Falkow, S., 1997. Common themes in microbial pathogenicity revisited. Microbiol. Molec.

Biol. Rev. 61, 136−169.Fletcher, M., 1996. Bacterial attachment in aquatic environments: a diversity of surfaces and adhesion

strategies. In: Fletcher, M. (Ed.), Bacterial Adhesion: Molecular and Ecological Diversity. Wiley-Liss,New York, pp. 1−24.

Floch, M.H., Binder, H.J., Filburn, B., Gesshengoren, W., 1972. The effect of bile acids on intestinalmicroflora. Amer. J. Clin. Nutr. 25, 1418−1426.

Florey, H.W., 1962. The secretion and function of intestinal mucus. Gastroenterology 43, 326−329.Forstner, J.F., 1978. Intestinal mucins in health and disease. Digestion 17, 234−263.Francis-Floyd, R., Beleau, M.H., Waterstrat, P.R., Bowser, P.R., 1987. Effect of water temperature on the

clinical outcome of infection with Edwardsiella ictaluri in channel catfish. J. Amer. Vet. Med. Assoc.191, 1413−1416.

Freter, R., Brickner, H., Botney, M., Cleven, D., Aranki, A., 1983. Mechanisms that control bacterial popu-lations in continuous-flow culture models of mouse large intestinal flora. Infect. Immun. 39, 676−685.

Fryer, J.L., Mauel, M., 1997. The Rickettsia: an emerging group of pathogens in fish. Emerg. Infect. Dis.3, 137−144.


Fryer, J.L., Lannan, C.N., Garces, L.H., Larenas, J.J., Smith, P.A., 1990. Isolation of a rickettsiales-likeorganism from diseased coho salmon Oncorhynchus kisutch in Chile. Fish Pathol. 25, 107−114.

Fryer, J.L., Lannan, C.N., Giovannoni, S.J., Wood, N.D., 1992. Piscirickettsia salmonis gen. nov., thecausative agent of an epizootic disease in salmonid fishes. Int. J. System. Bacteriol. 42, 120−126.

Garces, L.H., Larenas, J.J., Smith, P.A., Sandino, S., Lannan, C.N., Fryer, J.L., 1991. Infectivity of a rickettsia isolated from coho salmon (Oncorhynchus kisutch). Disease Aquat. Org. 11, 93−97.

Garcia, T., Otto, K., Kjelleberg, S., Nelson, D.R., 1997. Growth of Vibrio anguillarum in salmon intes-tinal mucus. Appl. Environ. Microbiol. 63, 1034−1039.

Gershwin, M.E., Leach, R., Hurley, L.S. (Eds.), 1985. Nutrition and Immunity. Academic Press, New York.Gibson, G.R., 1998. Dietary modulation of the human gut microflora using prebiotics. Brit. J. Nutr. 80,

S209−S212.Gilliland, S.E., 1979. Beneficial interrelationships between certain microorganisms and humans: candi-

date organisms for use as dietary adjuncts. J. Food Protect. 42, 164−167.Govoni, J.J., Boehlert, G.W., Watanabe, Y., 1986. The physiology of digestion in fish larvae. Environ.

Biol. Fish 16, 59−77.Gram, L., Melchiorsen, J., Spanggard, B., Huber, I., Nielsen, T.F., 1999. Inhibition of Vibrio anguillarum

by Pseudomonas fluorescens strain AH2, a possible probiotic treatment of fish. Appl. Environ.Microbiol. 65, 969−973.

Gram, L., Løvold, T., Nielsen, J., Melchiorsen, J., Spanggaard, B., 2001. In vitro antagonism of the probiont Pseudomonas fluorescens strain AH2 against Aeromonas salmonicida does not confer protection of salmon against furunculosis. Aquaculture 199, 1−11.

Grisez, L., Chair, M., Sorgeloos, P., Ollevier, F., 1996. Mode of infection and spread of Vibrio anguil-larum in turbot Scophthalmus maximus larvae after oral challenge through live feed. Disease Aquat.Org. 26, 181−187.

Grotmol, S., Totland, G.K., Kvellestad, A., Fjell, K., Olsen, A.B., 1995. Mass mortality of larval and juvenile hatchery reared halibut Hippoglossus hippoglossus L. associated with the presence of virus-like particles in vacuolating lesions in the central nervous system and retina. Bull. Eur. Assoc. Fish Pathol. 15, 176−180.

Grotmol, S., Totland, G.K., Thorud, K., Hjeltnes, B.K., 1997. Vacuolating encephalopathy and retino-pathy associated with a nodavirus-like agent: a probable cause of mass mortality of cultured larvaland juvenile Atlantic halibut Hippoglossus hippoglossus. Disease Aquat. Org. 29, 85−97.

Grotmol, S., Bergh Ø., Totland, G.K., 1999. Transmission of viral encephalopathy and retinopathy (VER)to yolk sac larvae of the Atlantic halibut Hippoglossus hippoglossus: occurrence of nodavirus in various organs and a possible route of infection. Disease Aquat. Org. 36, 95−106.

Gudding, R., Lillehaug, A., Midtylyng, P., Brown, F., 1997. Fish Vaccinology. Dev. Biol. Standard.Vol. 90. Karger, Basel.

Guentzel, M.N., Berry, L.J., 1975. Motility as a virulence factor for Vibrio cholerae. Infect. Immun. 11,890−897.

Hacker, J., 1992. Role of fimbrial adhesins in the pathogenesis of Escherichia coli infections. Can. J.Microbiol. 38, 720−727.

Hansen, G.H., Olafsen, J.A., 1990. Endocytosis of bacteria in yolksac larvae of cod (Gadus morhua L.).In: Lesel, R. (Ed.), Microbiology in Poecilotherms. Elsevier, Amsterdam, pp. 187−191.

Hansen, G.H., Olafsen, J.A., 1999. Bacterial interactions in early life stages of marine cold water fish.Microbial Ecol. 38, 1−26.

Hansen, G.H., Strøm, E., Olafsen, J.A., 1992. Effect of different holding regimes on the intestinalmicroflora of herring (Clupea harengus) larvae. Appl. Environ. Microbiol. 58, 461−470.

Hardy, R.W., 1997. 25th Fish feed and nutrition workshop. Aquaculture Magazine 23, 69−72.Henderson, B., Wilson, M., McNab, R., Lax, A.J., 1999. Cellular Microbiology. John Wiley and Sons,

Chichester.Hiney, M.P., Kilmartin, J.J., Smith, P.R., 1994. Detection of Aeromonas salmonicida in Atlantic salmon

with asymptomatic furunculosis infections. Disease Aquat. Org. 19, 161−167.Hjeltnes, B., Andersen, K., Ellingsen, H.M., Egedius, E., 1987. Experimental studies on pathogenicity of

a Vibrio sp. isolated from Atlantic salmon, Salmo salar L., suffering from Hitra disease. J. Fish Dis.10, 21−27.


Horne, M.T., Baxendale, A., 1983. The adhesion of Vibrio anguillarum to host tissues and its role inpathogenesis. J. Fish Dis. 6, 461−471.

Horne, M.T., Richards, R.H., Roberts, R.J., Smith, P.C., 1977. Peracute vibriosis in juvenile turbotScophthalmus maximus. J. Fish Pathol. 11, 355−361.

Ishiguro, E.E., Kay, W.W., Ainsworth, T., Chamberlain, J.B., Austen, R.A., Buckley, J., Trust, T.J., 1981.Loss of virulence during culture of Aeromonas salmonicida at high temperature. J. Bacteriol. 148,333−340.

Janda, J.M., Abbot, S.L., Oshiro, L.S., 1991. Penetration and replication of Edwardsiella spp. in Hep-2cells. Infect. Immun. 59, 154−161.

Jöborn, A., Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1997. Colonization in the fishintestinal tract and production of inhibitory substances in intestinal mucus and faecal extracts byCarnobacterium sp. strain K1. J. Fish Dis. 20, 83−92.

Jones, B., Pascopella, L., Falkow, S., 1995. Entry of microbes into the host: using M cells to break themucosal barrier. Curr. Op. Immunol. 7, 474−478.

Kanno, T., Nakai, T., Muroga, K., 1990. Mode of transmission of vibriosis among ayu Plecoglossusaltivelis. J. Aquat. Anim. Health 1, 2−6.

Karunasagar, I., Karunasagar, I., 1999. Diagnosis, treatment and prevention of microbial diseases of fishand shellfish. Curr. Sci. 76, 387−399.

Kawai, K., Kusuda, R., Itami, T., 1981. Mechanisms of protection in ayu orally vaccinated for vibrios.Fish Pathol. 15, 257−262.

Klemm, P., Schembri, M., Stentebjerg-Olesen, B., Hasman, H., Hasty, D.L., 1998. Fimbriae: detection,purification, and characterization. In: Williams, P., Ketley, J., Salmond, G. (Eds.), BacterialPathogenesis, Methods in Microbiology, 27. Academic Press, London, pp. 239−248.

Knudsen, G., Sørum, H., Press, C.M., Olafsen, J.A., 1999. In situ adherence of Vibrio sp. to cryosectionsof Atlantic salmon, Salmo salar L., tissue. J. Fish Dis. 22, 409−418.

Knutton, S., 1995. Electron microscopical methods in adhesion. Methods Microbiol. 253, 145−158.Knutton, S., Lloyd, D.R., McNeish, A.S., 1987. Adhesion of enteropathogenic Escherichia coli to human

intestinal enterocytes and cultured human intestinal mucosa. Infect. Immun. 56, 69−77.Knutton, S., Baldwin, T., Williams, P.H., McNeish, A.S., 1989. Actin accumulation at sites of adhesion

to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagicEscherichia coli. Infect. Immun. 57, 1290−1298.

Knutton, S., Phillips, A.D., Smith, H.R., Gross, R.J., Shaw, R., Watson, P., Price, E., 1991. Screening forenteropathogenic Escherichia coli in infants with diarrhea by the fluorescent actin staining test.Infect. Immun. 41, 365−371.

Krivan, H.C., Franklin, D.P., Wang, W., Laux, D.C., Cohen, P.S., 1992. Phosphatidylserine found in intes-tinal mucus serves as a sole source of carbon and nitrogen for salmonellae and Escherichia coli.Infect. Immun. 60, 3943−3946.

Krogfeld, K.A., Bergmans, H., Klemm, P., 1990. Direct evidence that the FimH protein is the mannosespecific adhesin of Escherichia coli type 1 fimbriae. Infect. Immun. 58, 1995−1998.

Lall, S.P., 1988. Disease control through nutrition. In: Proceedings of the Aquaculture InternationalCongress and Exposition, B.C. Pavilion Corporation, Vancouver, B.C., pp. 607−612.

Lall, S.P., Olivier, G., 1993. Role of micronutrients in immune response and disease resistance in fish.In: Kaushik, S.J., Luquet, P. (Eds.), Fish Nutrition in Practice. INRA Editions, Paris, pp. 101−118.

Landolt, M.L., 1989. The relationship between diet and the immune response of fish. Aquaculture 79,193−206.

Li, M.H., Wise, D.J., Johnson, M.R., Robinson, E.H., 1994. Dietary menhaden oil reduced resistance ofchannel catfish (Ictalurus punctatus) to Edwardsiella ictaluri. Aquaculture 128, 335−344.

Ling, S.H.M., Wang, X.H., Xie, L., Lim, T.M., Leung, K.Y., 2000. Use of green fluorescent protein(GFP) to study the invasion pathways of Edwardsiella tarda in in vivo and in vitro fish models.Microbiology 146, 7−19.

Lloyd, S.A., Forsberg, A., Wolf-Watz, H., Francis, M.S., 2001. Targeting exported substrates to theYersinia TTSS: different functions for different signals? Trend. Microbiol. 9, 367−371.

Lødemel, J.B., Mayhew, T.M., Myklebust, R., Olsen, R.E., Espelid, S., Ringø, E., 2001. Effect of threedietary oils on disease susceptibility in Arctic charr (Salvelinus alpinus L.) during cohabitant chal-lenge with Aeromonas salmonicida ssp. salmonicida. Aquacult. Res. 32, 935–945.


Mackie, T.J., Arkwright, J.A., Pryce-Tannatt, T.E., Mottram, J.C., Johnston, W.D., Menzies, W.J., 1930.Interim Report of the Furunculosis Committee. HMSO, Edinburgh.

Magarinos, B., Pazos, F., Santos, Y., Romalde, J.L., Toranzo, A.E., 1995. Response of Pasteurella piscicida and Flexibacter maritimus to skin mucus of marine fish. Disease Aquat. Org. 21, 103−108.

Magarinos, B., Romalde, J.L., Noya, M., Barja, J.L., Toranzo, A.E., 1996. Adherence and invasive capac-ities of the fish pathogen Pasteurella piscicida. FEMS Microbiol. Letts. 138, 29−34.

Mantle, M., Thakore, E., Hardin, J., Gall, D.G., 1989. Effect of Yersinia enterocolitica on intestinalmucin secretion. Amer. J. Physiol. 256, G319−G327.

Mantle, M., Atkins, E., Kelly, J., Thakore, E., Buret, A., Gall, D.G., 1991. Effects of Yersinia entero-colitica infection on rabbit intestinal and colonic goblet cells and mucin: morphometrics, histochem-istry and biochemistry. Gut 32, 1131−1138.

Marmur, J., 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol.Biol. 3, 208−218.

Mathew, J.A., Tan, Y.P., Rao, P.S.S., Lim, T.M., Leung, K.Y., 2001. Edwardsiella tarda mutants defec-tive in siderophore production, motility, serum resistance and catalase activity. Microbiology 147,449–457.

Mattar, A.F., Drongowski, R.A., Coran, A.G., Harmon, C.M., 2001. Effect of probiotics on enterocytebacterial translocation in vitro. Pediat. Surg. Int. 17, 265−268.

Maxson, R.T., Dunlap, J.P., Tryka, F., Jackson, R.J., Smith, S.D., 1994. The role of the mucus gel layerin intestinal bacterial translocation. J. Surg. Res. 57, 682−686.

McCarthy, D.H., 1977. Some ecological aspects of the bacterial fish pathogen Aeromonas salmonicida.In: Skinner, F.A., Shewan, J.M. (Eds.), Aquatic Microbiology. Society for Applied BacteriologySymposium Series No. 6. London, Academic Press, pp. 299−324.

Menard, R., Prevost, M.C., Gounon, P., Sansonetti, P., Dehio, C., 1996. The secreted Ipa complex ofShigella flexneri promotes entry into mammalian cells. Proc. Natl. Acad. Sci. USA 93, 1254−1258.

Mims, C.A., Dimmock, N.J., Nash, A., Stephen, J., 2000. Mims Pathogenesis of Infectious Disease,5th edition. Academic Press, London, pp. 474.

Munday, B.L., Nakai, T., 1997. Special topic review: nodaviruses as pathogens in larval and juvenilemarine finfish. World J. Microbiol. Biotechnol. 13, 375−381.

Munn, C.B., 1977. Vibriosis in fish and its control. Fish Manage. 8, 11−15.Munro, A.L.S., Hastings, T.S., 1993. Furunculosis. In: Inglis, V., Roberts, R.J., Bromage, N.R. (Eds.),

Bacterial Diseases of Fish. Blackwell Scientific Publications, Oxford, pp. 122−142.Muroga, K., De La Cruz, M.C., 1987. Fate and location of Vibrio anguillarum in tissues of artificially

infected ayu, Plecoglossus altivelis. Fish Pathol. 22, 99−103.Myhal, M.L., Laux, D.C., Cohen, P.S., 1982. Relative colonizing abilities of human fecal and K12 strains of

Escherichia coli in the large intestines of streptomycin treated mice. Eur. J. Clin. Microbiol. 1, 186−192.Nachamkin, I., Yang, X.H., Stern, N.J., 1993. Role of Campylobacter jejuni flagella as colonization fac-

tors for 3-day-old chicks: analysis with flagellar mutants. Appl. Environ. Microbiol. 59, 1269−1273.Nataro, J.P., Kaper, J.B., 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11, 142−201.Neutra, M.R., Pringault, E., Kraehenbuhl, J.-P., 1996. Antigen sampling across epithelial barriers and

induction of immune responses. Ann. Rev. Immunol. 14, 275−300.Obach, A., Quentel, C., Baudin Laurencin, F., 1993. Effects of alpha-tocopherol and dietary oxidized fish

oil on immune response of sea bass Dicentrarchus labrax. Disease Aquat. Org. 15, 175−185.O’Brien, D., Mooney, J., Ryan, D., Powell, E., Hiney, M., Smith, P.R., Powell, R., 1994. Detection of

Aeromonas salmonicida, causal agent of furunculosis in salmonid fish, from the tank effluent ofhatchery-reared Atlantic salmon smolts. Appl. Environ. Microbiol. 60, 3874−3877.

Ofek, I., Doyle, R.J., 1994. Bacterial Adherence to Cells and Tissues. Chapman and Hall, London.Olafsen, J.A., Hansen, G.H., 1992. Intact antigen uptake in intestinal epithelial cells of marine fish larvae.

J. Fish Biol. 40, 141−156.Olivier, G., 1997. Effect of nutrition on furunculosis. In: Bernoth, E.-M., Ellis, A.E., Midtlyng, P.J.,

Olivier, G., Smith, P. (Eds.), Furunculosis: Multidisciplinary Fish Disease Research. Academic Press,San Diego, pp. 327−344.

Olsen, R.E., Myklebust, R., Kaino, T., Ringø, E., 1999. Lipid digestibility and ultrastructural changes inthe enterocytes of Arctic charr (Salvelinus alpinus L.) fed linseed oil and soybean lecithin. FishPhysiol. Biochem. 21, 35−44.


Olsen, R.E., Myklebust, R., Ringø, E., Mayhew, T.M., 2000. The influence of dietary linseed oil and saturated fatty acids on caecal enterocytes of Arctic charr (Salvelinus alpinus L.): a quantitative ultrastructural study. Fish Physiol. Biochem. 22, 207−216.

Olsen, R.E., Myklebust, R., Kryvi, H., Mayhew, T.M., Ringø, E., 2001. Damaging effect of dietary inulinto intestinal enterocytes in Arctic charr (Salvelinus alpinus L.). Aquacult. Res. 32, 931−932.

Olsson, C., 1995. Bacteria with inhibitory activity and Vibrio anguillarum in fish intestinal tract. Fil. Dr.Thesis. Gothenburg University, Gothenburg.

Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1992. Intestinal colonization potential of turbot (Scophthalmus maximus)- and dab (Limanda limanda)-associated bacteria with inhibitoryeffects against Vibrio anguillarum. Appl. Environ. Microbiol. 58, 551−556.

Olsson, J.C., Jöborn, A., Westerdahl, A., Blomberg, L., Kjelleberg, S., Conway, P.L., 1996. Is the turbot,Scophthalmus maximus L., intestine a port of entry for the fish pathogen Vibrio anguillarum? J. FishDis. 19, 225−234.

Olsson, J.C., Jöborn, A., Westerdahl, A., Blomberg, L., Kjelleberg, S., Conway, P.L., 1998. Survival, persistence and proliferation of Vibrio anguillarum in juvenile turbot, Scophthalmus maximus (L.),intestine and faeces. J. Fish Dis. 21, 1−10.

Ormonde, P., Horstedt, P., O’Toole, R., Milton, D.L., 2000. Role of motility in adherence to and invasionof a fish cell line by Vibrio anguillarum. J. Bacteriol. 182, 2326−2328.

O’Toole, R., Milton, D.M., Wolf-Watz, H., 1996. Chemotactic motility is required for invasion of the hostby the fish pathogen Vibrio anguillarum. Mol. Microbiol. 19, 625−637.

O’Toole, R., Lundberg, S., Fredriksson, S.A., Jansson, A., Nilsson, B., Wolf-Watz, H., 1999. The chemo-tactic response of Vibrio anguillarum to fish intestinal mucus is mediated by a combination of multi-ple mucus components. J. Bacteriol. 181, 4308−4317.

Plehn, M., 1911. Die furunkulose der salmoniden. Central. Bakteriol. Parasit. Infect. Hyg. Abt. 1. Orig.60, 609−624.

Plumb, J.A., 1993. Edwardsiella septicaemia. In: Inglis, V., Roberts, R.J., Bromage, N.R. (Eds.),Bacterial Diseases of Fish. Blackwell Scientific Publications, Oxford, pp. 61–79.

Press, C.M., Lillehaug, A., 1995. Vaccination in European salmonid aquaculture: a review of practicesand prospects. Brit. Vet. J. 151, 45−69.

Ransom, D.P., Lannan, C.N., Rohovec, J.S., Fryer, J.L., 1984. Comparison of histopathology caused byVibrio anguillarum and Vibrio ordalii in three species of pacific salmon. J. Fish Dis. 7, 107−115.

Richardson, K., 1991. Roles of motility and flagellar structure in pathogenicity of Vibrio cholerae: analy-sis of motility mutants on three animal models. Infect. Immun. 59, 2727−2736.

Ringø, E., 1999. Lactic acid bacteria in fish: antibacterial effect against fish pathogens. In: Krogdahl, Å.,Mathiesen, S.D., Pryme, I. (Eds.), Effects of Antinutrients on the Nutritional Value of Legume Diets.Vol. 8. COST 98. EEC Publication, Luxembourg, pp. 70−75.

Ringø, E., Birkbeck, T.H., 1999. Intestinal microflora of fish larvae and fry. Aquacult. Res. 30, 73−93.Ringø, E., Bendiksen, H.R., Wesmajervi, M.S., Olsen, R.E., Jansen, P.A., Mikkelsen, H., 2000. Lactic

acid bacteria associated with the digestive tract of Atlantic salmon (Salmo salar L.). J. Appl.Microbiol. 89, 317−322.

Ringø, E., Lødemel, J.B., Myklebust, R., Kaino, T., Mayhew, T.M., Olsen, R.E., 2001. Epithelium-associated bacteria in the gastrointestinal tract of Arctic charr (Salvelinus alpinus L.). An electronmicroscopical study. J. Appl. Microbiol. 90, 294−300.

Ringø, E., Lødemel, J.B., Myklebust, R., Jensen, L., Lund, V., Mayhew, T.M., Olsen, R.E., 2002. Aerobicgut microbiota of Arctic charr (Salvelinus alpinus L.). Effect of soybean, linseed and marine oils on priorto and post challenge with Aeromonas salmonicida subsp. salmonicida. Aquacult. Res. 33, 591–606.

Robertson, P.A.W., O’Dowd, C., Burrells, C., Williams, P., Austin, B., 2000. Use of Carnobacterium sp.as a probiont for Atlantic salmon (Salmo salar L.) and rainbow trout (Oncorhynchus mykiss,Walbaum). Aquaculture 185, 235−243.

Romalde, J.L., Magarinos, B., Nunez, S., Barja, J.L., Toranzo, A.E., 1996. Host range susceptibility ofEnterococcus sp. strains isolated from diseased turbot: Possible routes of infection. Appl. Environ.Microbiol. 62, 607−611.

Rose, A.S., Ellis, A.E., Munro, A.L.S., 1989. The infectivity by different routes of exposure and sheddingrates of Aeromonas salmonicida subsp. salmonicida in Atlantic salmon, Salmo salar L., held in seawater. J. Fish Dis. 12, 573−578.


Roussel, P., Lamblin, G., Lhermitte, M., Houdret, N., Lafitte, J.-J., Perini, J.-M., Klein, A., Scharfman, A.,1988. The complexity of mucins. Biochimie 70, 1471−1482.

Russell, R.G., Blake, D.C., 1994. Cell association and invasion of Caco-2 cells by Campylobacter jejuni.Infect. Immun. 62, 3773−3779.

Saeed, M., 1983. Chemical characterization of the lipopolysaccharides of Edwardsiella ictaluri and theimmune response of channel catfish to this function and to whole cell antigen with histopathologicalcomparisons. Ph.D. Thesis. Auburn University, USA.

Sakai, D.K., 1979. Invasive routes of Aeromonas salmonicida subsp. salmonicida. Sci. Rep. HokkaidoFish Hatch. 34, 61−89.

Salte, R., Åsgård, T., Liestøl, K., 1988. Vitamin E and selenium prophylaxis against Hitra disease infarmed Atlantic salmon. A survival study. Aquaculture 75, 45−55.

Sansonetti, P.J., 1993. Bacterial pathogens, from adherence to invasion: comparative strategies. Med.Microbiol. Immunol. 182, 223−232.

Savage, D.C., 1983. Mechanisms by which indigenous microorganisms colonize gastrointestinal epithe-lial surface. Progr. Food Nutr. Sci. 7, 65−74.

Schrøder, K., Clausen, E., Sandberg, A.M., Raa, J., 1980. Psychrotropic Lactobacillus plantarum fromfish and its ability to produce antibiotic substances. In: Connell, J. (Ed.), Advances in Fish Scienceand Technology. Fishing News Books, Surrey, pp. 480−483.

Shotts, E.B., Blazer, V.S., Waltman, W.D., 1986. Pathogenesis of experimental Edwardsiella ictaluriinfections in channel catfish (Ictalururs punctatus). Can. J. Fish. Aquat. Sci. 43, 36−42.

Silverman, D.J., Santucci, L.A., Meyers, N., Sekeyova, Z., 1992. Penetration of host-cells byRickettsia rickettsii appears to be mediated by a phospholipase of rickettsial origin. Infect. Immun. 60,2733−2740.

Smail, D.A., Munro, A.L.S., 2001. The virology of teleosts. In: Roberts, R.J. (Ed.), Fish Pathology, 3rdedition. W.B. Saunders, London.

Smeltzer, M.S., Gillaspy, A.F., Pratt, F.L., Thames, M.D., Iandolo, J.J., 1997. Prevalence and chromo-somal map location of Staphylococcus aureus adhesin genes. Gene 196, 249−259.

Smith, H., 1990. Pathogenicity and the microbe in vivo. J. Gen. Microbiol. 136, 377−383.Smith, P.A., Pizarro, P., Ojeda, P., Contreras, J., Oyanedel, S., Larenas, J., 1999. Routes of entry of

Piscirickettsia salmonis in rainbow trout Oncorhynchus mykiss. Disease Aquat. Org. 37, 165−172.Smyth, C.J., Marron, M.B., Twohig, J.M.G.J., Smith, S.G.J., 1996. Fimbrial adhesins: similarities and

variations in structure and biogenesis. FEMS Immunol. Med. Microbiol. 16, 127−139.Snoeyenbos, G.H., 1979. Role of native intestinal microflora in protection against pathogens. Proc. Ann.

Meet. U.S. Anim. Health Ass. 83, 388−393.Spanggard, B., Huber, I., Nielsen, J., Nielsen, T., Gram, L., 2001. Proliferation and location of Vibrio anguil-

larum during infection of rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Dis. 23, 423−427.Strøm, E., 1988. Melkesyrebakterier i fisketarm. Isolasjon, karakterisering og egenskaper (in Norwegian).

Msci. Thesis. The Norwegian College of Fishery Science, University of Tromsø, Norway, pp. 88.Sugita, H., Shibuya, K., Shimooka, H., Deguchi, Y., 1996. Antibacterial abilities of intestinal bacteria in

freshwater cultured fish. Aquaculture 145, 195−203.Sugita, H., Shibuya, K., Hanada, H., Deguchi, Y., 1997. Antibacterial abilities of intestinal microflora of

the river fish. Fish. Sci. 63, 378−383.Sugita, H., Hirose, Y., Matsuo, N., Deguchi, Y., 1998. Production of the antibacterial substance by Bacillus

sp. strain NM 12, and intestinal bacterium of Japanese coastal fish. Aquaculture 165, 269−280.Svendsen, Y.S., Bøgvald, J., 1997. Influence of artificial wound and non-intact mucus layer on mortality

of Atlantic salmon (Salmo salar L.) following a bath challenge with Vibrio anguillarum andAeromonas salmonicida. Fish Shellfish Immunol. 7, 317−325.

Svendsen, Y.S., Dalmo, R.A., Bøgvald, J., 1999. Tissue localization of Aeromonas salmonicida inAtlantic salmon, Salmo salar L., following experimental challenge. J. Fish Dis. 22, 125−131.

Tan, E., Low, K.W., Wong, W.S.F., Leung, K.Y., 1998. Internalisation of Aeromonas hydrophila by fishepithelial cells can be inhibited with a tyrosine kinase inhibitor. Microbiology 144, 299−307.

Tancrède, C., 1992. Role of human microflora in health and disease. Eur. J. Clin. Microbiol. Infect. Dis.11, 1012−1015.

Thompson, K.D., Tatner, M.F., Henderson, R.J., 1996. Effects of dietary (n-3) and (n-6) polyunsaturatedfatty acid ratio on the immune response of Atlantic salmon, Salmo salar L. Aquacult. Nutr. 2, 21−31.


Thune, R.L., Stanley, L.A., Cooper, K., 1993. Pathogenesis of Gram-negative bacterial infections inwarm water fish. Ann. Rev. Fish Dis. 3, 37−68.

Trust, T.J., 1986. Pathogenesis of infectious diseases of fish. Ann. Rev. Microbiol. 40, 479−502.Trust, T.J., Kay, W.W., Ishiguro, E.E., 1983. Cell surface hydrophobicity and macrophage association of

Aeromonas salmonicida. Curr. Microbiol. 9, 315−318.Tytler, P., Blaxter, J.H.S., 1988. Drinking in yolk-sac stage larvae of the halibut, Hippoglossus

hippoglossus (L.). J. Fish Biol. 32, 493−494.van der Waaij, D., Berghuid-de Vries, J.M., Lekkerkerk-van der Wees, J.E.C., 1972. Colonization resist-

ance of the digestive tracts of mice during systemic antibiotic treatment. J. Hyg. 70, 405−411.Vazquez-Torres, A., Fang, F.C., 2000. Cellular routes of invasion by enteropathogens. Curr. Opin.

Microbiol. 3, 54−59.Waagbø, R., 1994. The impact of nutritional factors on the immune system in Atlantic salmon, Salmo

salar L.: a review. Aquacult. Fish. Manage. 25, 175−197.Waagbø, R., Sandnes, K., Jørgensen, J., Engstad, R., Glette, J., Lie, Ø., 1993. Health aspects of dietary lipid

sources and vitamin E in Atlantic salmon (Salmo salar L.). II. Spleen and erythrocyte phospholipid fattyacid composition, nonspecific immunity and disease resistance. Fisk. Dir. Skr. Ser. Ern. 6, 63−80.

Wadolkowski, E.A., Laux, D.C., Cohen, P.S., 1988. Colonization of the streptomycin treated mouse largeintestine by a human fecal Escherichia coli strain. Role of growth in mucus. Infect. Immun. 56,1030−1035.

Wang, X.H., Leung, K.Y., 2000. Biochemical characterization of different types of adherence of Vibriospecies to fish epithelial cells. Microbiology 146, 989−998.

Wang, X.H., Oon, H.L., Ho, G.W.P., Wong, W.S.F., Lim, T.M., Leung, K.Y., 1998. Internalization andcytotoxicity are important virulence mechanisms in vibrio-fish epithelial cell interactions.Microbiology 144, 2987−3002.

Wassenaar, T.M., Bleuminkpluym, N.M.C., Vanderzeijst, B.A.M., 1991. Inactivation of Campylobacterjejuni flagellin genes by homologous recombination demonstrates that flaA but not flaB is requiredfor invasion. EMBO J. 10, 2055−2061.

Westerdahl, A., Olsson, J.C., Kjelleberg, S., Conway, P.L., 1991. Isolation and characterization of turbot(Scophthalmus maximus)-associated bacteria with inhibitory effects against Vibrio anguillarum.Appl. Environ. Microbiol. 57, 2223−2228.

Wilson, M., Seymour, R., Henderson, B., 1998. Bacterial perturbation of cytokine networks. Infect.Immun. 66, 2401−2409.

Wolf, K., 1988. Fish Viruses and Fish Viral Diseases. Cornell University Press, Ithaca.Yamamoto, T., Endo, S., Yokota, T., Escheverria, P., 1991. Characteristics of adherence of entero-

aggregative Escherichia coli to human and animal mucosa. Infect. Immun. 59, 3722−3739.


Salmonella continues to pose questions in terms of its pathogenicity and host speci-ficity and also remains a key organism in the study of general infection mechanisms.It elicits a disease that can vary from localized gut disorder to severe systemic bacteremia, depending on the strain or serovar of the pathogen. In humans, theprevalent form is a self-limiting infection confined primarily to the gastrointestinaltract. The severity of the illness is, however, greatly influenced by factors such asage, dietary history and health/immune status of the individual.

A plethora of animal models have been adopted to study salmonellosis. Fewappear to effectively model the overall infection in humans. In particular, there isgreat variation in the levels of colonization, invasion and systemic spread that areobserved, as well as in the incidence of bacteremia and the effects of the pathogenon long-term health. However, their use has allowed very detailed study of specificaspects of pathogenesis.

Ideally the animal model chosen would be that which best exhibits the facet of salmonellosis to be studied. However, other factors such as susceptibility to thedisease, the infective dose required, the ease of non-invasive monitoring or readyavailability of species-specific reagents often influence our choice. The mouse, ratand pig are widely used. In this chapter, their strengths and weaknesses as modelsof salmonellosis will be evaluated.

1. INTRODUCTION

Infections caused by Salmonella species continue to be a worldwide health problem.They usually occur as a result of consumption of contaminated food or water and

11 Modelling of salmonellosis1

P.J. Naughtona and G. Grantb

aNorthern Ireland Centre for Food and Health, School of Biomedical Sciences,University of Ulster, Cromore Road, Coleraine, Co. Londonderry BT52 1SA, UKbRowett Research Institute, Greenburn Road, Bucksburn, Aberdeen AB21 9SB, UK

235

1This work was supported in part by the Scottish Executive Environment and Rural Affairs Department (SEERAD).



can manifest in a variety of disease states. These range from asymptomatic carrierstatus through to localized gastroenteritis or to severe systemic infections that leadto death (Buchwald and Blaser, 1984; Kotova et al., 1988; Bean and Griffin, 1990;Darwin and Miller, 1999; Tsolis et al., 1999b; Kingsley and Baumler, 2000; Ohl andMiller, 2001; Santos et al., 2001a). The nature and severity of the infection variesaccording to bacterial strain and is also greatly influenced by host factors, such asage and health status. The young, the elderly and immunocompromised individualsare particularly susceptible.

For epidemiological purposes, the Salmonella can be placed into three groups: 1) those that infect humans only e.g., S. typhi, 2) the host-adapted serovars,some of which are human pathogens and may be contracted from foods, including S. gallinarum (poultry), S. dublin (cattle), S. abortis-equi (horses), S. abortus-ovis(sheep) and S. cholerasuis (swine) and 3) non-adapted serovars, such as S. enteritidisand S. typhimurium, that are pathogenic for humans and other animals and encompassmost food-borne serovars.

In Europe and North America, the majority of salmonellosis cases are non-typhoidal, mostly of the self-limiting gastroenteritis type and in the main caused byS. typhimurium and S. enteritidis (Rampling, 1993; Mandal, 1994; Tauxe and Pavia,1998; Mead et al., 1999). In contrast, typhoid-type diseases remain the predominantforms in Asia, Africa and South America. These are due to S. typhi and S. paratyphiA, B and C and continue to cause a high incidence of mortality in these regions(Candy and Stephen, 1989; Mandal, 1994; Merican, 1997). The occurrence oftyphoid-like diseases is very low in Europe and North America but there is a tendencyfor the levels to increase with time, possibly as a result of increased foreign travel(Ryan et al., 1989; Mead et al., 1999).

2. SALMONELLA TYPHI

S. typhi colonizes the human gastrointestinal tract and adheres to and invades theepithelium. It passes through into the subepithelium, where it is phagocytosed intomacrophages. It survives in these cells, rapidly spreads via the reticuloendothelialsystem to internal tissues, such as liver and spleen and triggers chronic systemicresponses including onset of fever (Hornick et al., 1970; Mandal, 1994; Weinsteinet al., 1998; Santos et al., 2001a). Invasion appears to be primarily via the ileum,where infiltration of mononuclear cells and thickening of the mucosa is evident soonafter infection. Furthermore, haemorrhage, ulceration and perforation occur in thisregion of the gut in the longer term. The mesenteric lymph nodes, liver and spleenenlarge and cellular dysfunction and lesioning develops in these tissues. Chronicdamage to the intestine appears, however, to be a primary cause of death (Hornicket al., 1970; Weinstein et al., 1998; Santos et al., 2001a).

S. typhi is host-specific and elicits typhoid-like disorders only in humans andchimpanzees (Pascopella et al., 1995; Weinstein et al., 1998). S. typhi does colonize

236 P.J. Naughton and G. Grant

the gut of ex-germfree mice and enters Peyer’s patches (Collins and Carter, 1978).However, it does not spread to other tissues or cause chronic infection (Collins andCarter, 1978). S. typhi may be unable to proliferate, persist or cause damage in cellsof the murine Peyer’s patch (O’Brien, 1982; Kohbata et al., 1986; Pascopella et al.,1995). The absence of a suitable animal model has severely hampered the study ofS. typhi infection.

3. SALMONELLA TYPHIMURIUM AND SALMONELLA ENTERITIDIS

S. typhimurium and S. enteritidis colonize the human gut, attach to and invade theepithelium and pass into the subepithelium. They may translocate to the mesentericlymph nodes but, unlike S. typhi, they do not generally reach the liver and spleen.Rapid clearance of the pathogens by macrophages is thought to limit the spreadbeyond the level of the lymph nodes. Extensive infiltration of inflammatory cellsinto the intestine is triggered by infection. There is also severe disruption of gutstructure and integrity, loss of fluid and electrolytes and onset of acute diarrhoea.The disease is, however, self-limiting, usually clearing within 7 days (Old, 1990;Tsolis et al., 1999a). Salmonella may nonetheless continue to be shed in faeces fora significant period after the disappearance of clinical signs. Severe systemic infec-tion is rare (2–5% of cases) in otherwise healthy individuals (Blaser and Feldman,1981). It is, however, a significant risk for immunocompromised patients, the veryyoung and the elderly (Old, 1990; Tsolis et al., 1999a)

S. typhimurium and S. enteritidis are not host-specific and infect a wide range ofdomesticated or wild animals (Old, 1990; Lax et al., 1995). In general, they cause aself-limiting gastroenteritis or settle as a carrier population with no overt detrimen-tal effects on the host. Chronic systemic infection is infrequent (Old, 1990; Lax et al., 1995). One exception is the mouse, in which these serovars cause severe systemic infection and death (Tsolis et al., 1999a).

4. GENERAL ASPECTS OF SALMONELLOSIS IN HUMANS

Salmonella spp. can thus cause two types of disease in humans: an acute gastro-enteritis or systemic typhoid disease. While some non-specific symptoms inSalmonella disease (e.g. nausea and abdominal pain) are shared, the clinical featuresof Salmonella-induced diarrhoea and systemic typhoid infections are quite different.For example, gastroenteritis may occur within 8−36 h after ingestion of contami-nated food, whereas typhoid may follow after a period of 10−20 days. Diarrhoea(which is usually watery, may be severe, and sometimes bloody) is the predominat-ing feature of gastroenteritis, whereas in the case of adults constipation can occur inthe early clinical stages of typhoid.

Current perceptions are that gastroenteritis results from the interactions ofS. typhimurium or S. enteritidis and/or their products with the gut mucosa. Diarrhoea is

Modelling of salmonellosis 237

thought to be triggered by host-derived inflammatory products liberated as a result ofinvasion of the intestine by the pathogen (Wallis et al., 1986; Worton et al., 1989;Lodge et al., 1995) or by an enterotoxin or cell-free bioactive components in theabsence of invasion (Sandefur and Peterson, 1977; Finkelstein et al., 1983).

Certain aspects of pathogenesis are considered to be dose-dependent (Mintz et al.,1994). High infective doses appear to provoke a more intense gastrointestinal response;the onset of diarrhoea occurs earlier, vomiting is more common and stool frequency isincreased. Repeat exposure to Salmonella spp. also resulted in higher rates of vomitingand hospitalization. However, clear relationships between inoculum, incubation periodsand other symptoms were not apparent (McCullough and Eisele, 1951).

Antibiotic treatment of gastroenteritis caused by Salmonella species has had lim-ited success. It does not appear to reduce the duration or severity of illness and mayin fact prolong asymptomatic carriage (Jewes, 1987). Combined with the growingresistance of Salmonella species to clinically important antibiotics (Lee et al., 1994),this has meant that alternative therapeutic strategies to prevent or ameliorate salmo-nellosis are becoming increasingly important.

Typhoid fever appears to be triggered by S. typhi organisms which translocate themucosa, survive within macrophages, multiply rapidly in systemic tissues and releaseendotoxin which triggers the highly complex endotoxin-cascade (Aleekseev et al.,1960; Santos et al., 2001a). Nonetheless, gut damage may be a primary cause ofdeath (Hornick et al., 1970; Weinstein et al., 1998; Santos et al., 2001a). Higher inci-dences of infection and shorter incubation periods were reported for volunteers givenincreasingly larger doses of S. typhi (Hornick et al., 1970). The clinical course, onceillness occurred, did not, however, appear to vary with infectious dose. Inoculum sizewas therefore not a strong predictor of intensity, nor of duration of fever.

The dose of Salmonella required to initiate infection is not well defined and canbe variable. Bryan (1977) showed it to be approximately 104 colony forming units(CFU) for S. typhi and more for other serotypes. However, lower doses (<103 CFUsalmonellae per gram of food) may also cause disease and outbreaks of entercolitisin receptive individuals (Silliker, 1980). In addition, there are compelling data thatSalmonella infection can follow the ingestion of even very small numbers of bacteria(<10 CFU) (Gill et al., 1983; Mintz et al., 1994).

The nature and severity of the infection caused by Salmonella spp. in humansthus varies according to serovar, dose, virulence factors and host factors including ageand prior health or immune status. It follows, therefore, that the choice of organism/host combination for experimental study of salmonellosis and precise definition ofthe question(s) to be addressed, are of crucial importance.

5. SALMONELLOSIS IN THE RAT

Salmonella infection was first described in rats by Orskøv et al. (1928) and thepathogenesis of the disease has been described by several investigators over the past


decades (Bakken and Vogelsang, 1950; Habermann and Williams, 1958; Böhme et al., 1959; Maenza et al., 1970; Naughton et al., 1996, Havelaar et al., 2001). S. enteritidis and S. typhimurium colonize the whole gastrointestinal tract of the rat(fig. 1) and cause disruption of the small intestine epithelium and possibly the caecal epithelium, induce epithelial cell hyperplasia and trigger infiltration byinflammatory cells into the tissue. Invasion appears to be mainly via the ileum, highnumbers of Salmonella reach the mesenteric lymph nodes and moderate numbersspread to and persist in the liver and spleen (fig. 1). In general, experimental ratscontinue to grow, albeit slowly. Despite historical evidence suggesting thatSalmonella can be an effective rodenticide (Leslie, 1941; Threlfall et al., 1996),recent studies indicate that death of experimental Hooded Lister rats owing toSalmonella infection is rare, unless infective doses are very high or health orimmune status or gut integrity or function has been compromised by other factors(Naughton et al., 1996, 2000; Islam et al., 2000; Havelaar et al., 2001; Takumi et al.,2002). This ability to survive despite carriage of a significant Salmonella infectionwould be consistent with the suspected role of rats in the spread of Salmonellabetween poultry units (Davies and Wray, 1995).

Early work with rats concentrated on pathogenicity of Salmonella, the determi-nation of infective dosage in specific pathogen free (SPF) albino Wistar rats


Fig. 1. Tissue distribution of S. enteritidis LA51 and 6 days post-infection (single oral dose of≈108 CFU) of rats. (A) St, stomach; J, jejunum;I, ileum; Ca, caecum; Co, colon. (B) MLN,mesenteric lymph nodes; L, liver; Sp, spleen.

(Kampelmacher et al., 1968), the study of diarrhoea in Walter Reed strain albino rats(Maenza et al., 1970), the comparison of Salmonella strains (Radoucheva et al.,1994; Naughton et al., 1996) and evaluating the effects of nutrition on pathogenesisin Sprague-Dawley rats (Omoike et al., 1990) and in Hooded Lister rats(Guggenheim and Buechler, 1947; Naughton et al., 2000). Suprisingly, little workhas been done on the effects of Salmonella on rat ileal morphology (Naughton et al.,1995; Havelaar et al., 2001) and in understanding the early events following entryof Salmonella into the host. Morphologically, the lesions in the rat have been shownto exhibit features evident in both severe human gastroenteritis with S. typhimurium(Aleekseev et al., 1960) and human typhoid fever (Gay, 1918). Nonetheless, thetime scales are different. The timecourse of Salmonella entercolitis in the rat resem-bles human gastroenteritis far more closely than it does typhoid fever.

Habermann and Williams (1958) stated that a Salmonella infection in rats couldtake an acute or chronic form, or any degree of severity between the two extremes,depending on the level of exposure (Bloomfield and Lew, 1942; Maenza et al.,1970), virulence of the organism, host age and the dietary treatment (Guggenheimand Buechler, 1947). Fasting overnight or prolonged protein or calorie intake defi-ciency significantly increased susceptibility to infection (Guggenheim andBuechler, 1947; Maenza et al., 1970; Omoike et al., 1990). Kampelmacher et al.(1968) found that infection could occur at very low doses (e.g., 102 CFU) but withconsiderable inter-individual variation. However, others found no persistent infec-tion following a single dose of 103 CFU/ml of S. enteritidis (Naughton et al., 1996).Colonization, invasion and persistence was, however, achieved at higher (up to 108 CFU) infective doses (Kampelmacher et al., 1968; Naughton et al., 1996). Age had marked effects on the susceptibility of rats to oral Salmonella infection(Kampelmacher et al., 1968). A persistent infection was evident in 4-day-old ratsgiven an infectious dose of 106 CFU/ml, but not in 20-day-old rats. However, aninfectious dose of 108 CFU/ml did cause a persistent infection in 100-day-old rats.

Kampelmacher and co-workers gave S. typhimurium, S. dublin and S. oranienburgat varying doses to rats aged 4, 20 and 100 days in order to study persistence of theorganism in the faeces and the duration of the infection in the organs. It wasassumed the three variables – age, dosage and type of Salmonella – would influencepersistence. However, it was found to correlate directly only with age and dosage.Repeated administration of Salmonella did not appear to lead to an increase in persistence, although there was considerable inter-animal and inter-group variabilityin the data. S. typhimurium, S. dublin and S. oranienburg showed unmistakable differences in their ability to survive but different strains of the same type gavealmost identical results.

In experiments where the fate of intraperitoneally administered S. dublin wastested in rats and mice, rats expressed greater cellular responses against thepathogen and quickly suppressed or limited its growth. In contrast, the cellularresponse of mice to the pathogen was poor. Later work by Veljanov et al. (1984)


showed that S. dublin caused systemic infection in mice but was avirulent in rats.The avirulence of the bacteria for rats was related to the high phagocytic activity ofleucocytes against S. dublin (Veljanov et al., 1984). Bacterial growth in rat peri-toneal fluid was attenuated and the microorganisms were found only at the site ofinoculation (Veljanov et al., 1984). This suggested that the systemic immune systemof the rat was more effective at suppression or clearance of salmonella than that ofthe mouse (Havelaar et al., 2001; Takumi et al., 2002).

Only a limited number of studies have been done on the involvement of virulencefactors in salmonellosis of the Hooded Lister rat. Type 1 (SEF21) and other (SEF14,SEF17, pef, lpf) fimbriae have transient roles in the early interactions of thepathogen with the gut (Robertson, 2000; Naughton et al., 2001a,b; Robertson et al.,2003). However, their presence does not appear to be a prerequisite for the occur-rence of infection. Inability to produce functional flagella or flagellin significantlyreduces pathogenicity of Salmonella in the rat (Robertson, 2000; Robertson et al.,2003). Despite this, aflagellate strains still cause a severe and persistent infectionsuggesting that virulence factors in addition to flagella and fimbriae are involved inthe early interactions of Salmonella with the rat gut.

S. typhimurium and S. enteritidis in general cause a self-limiting gastroenteritis-like disorder in rats. The infection is located primarily in the gut and associated tissues and its severity is dependent on the infective dose, age and health and dietary status. Salmonellosis in the rat thus has many features that are similar to thegastroenteritis-type infection commonly elicited by these bacteria in humans anddomesticated animals. It is thus a useful model for the study of this disorder and theeffects of diet or environmental factors on pathogenicity.

6. SALMONELLOSIS IN THE MOUSE

Mouse strains vary greatly in their ability to deal with infection by S. enterica (table 1).Balb/c or C57Bl/6 mice are very susceptible to S. typhimurium and S. enteritidis. As little as 10 CFU of either of these serovars administered systemically or around 105 CFU given orally can cause death. In contrast, CBA and A/J mice are resistantto S. typhimurium and S. enteritidis. With these mouse strains, ≥ 104 CFU given sys-temically or ≥108 CFU orally are needed to cause a lethal infection. All mice appear,however, to be resistant to S. typhi, even at inputs above 109 CFU orally (Collins andCarter, 1978; O’Brien, 1982).

The resistance of mice to S. typhimurium has been linked to at least five differ-ent gene loci. Most of these, including ity (Nramp-1) and lps, are important duringthe early phases of infection whilst the others appear to be associated with late-phase responses to infection (Mattsby-Baltzer et al., 1996; Pie et al., 1996;Cannone-Hergaux et al., 1999; Forbes and Gros, 2001). Disablement of these genesgreatly increases susceptibility to infection. For example, the intraperitoneal LD50

for the resistant C3H/HeN mouse is around 103 CFU but is less than 10 CFU for the


C3H/HeJ mouse, a related strain that is LPS and TLR4 defective (table 1). Mousestrains that are most susceptible to Salmonella, such as the Balb/c or C57Bl/6, aredefective in ity (Nramp-1) (Mattsby-Baltzer et al., 1996; Pie et al., 1996).

S. typhimurium colonizes the whole gastrointestinal tract of susceptible (itys) mice,attaches to enterocytes and invades primarily through M cells of ileal Peyer’s patches(Tannock et al., 1975; Jones et al., 1994; Darwin and Miller, 1999; Kingsley andBaumler, 2000). There is proliferation of the pathogen within the Peyer’s patches,destruction of M cells, infiltration of inflammatory cells into the gut and considerabledisruption to the structure and integrity of the intestine (Tannock et al., 1975; Jones et al., 1994; Santos et al., 2001a). A recent study indicates that colonization of thePeyer’s patches may be caspase-1 dependent since it is attenuated in casp-1−/− (Monacket al., 2000). In the subepithelium, macrophages or dendritic cells captureS. typhimurium. The pathogen can, however, survive within these cells and is carried tothe mesenteric lymph nodes, liver and spleen (Carter and Collins, 1974; Vazquez-Torres et al., 1999; Isberg and Barnes, 2000; Resigno and Barrow, 2001). In susceptiblemice, very rapid proliferation of the pathogen occurs in the liver and spleen. Indeed,bacterial numbers can increase more than ten-fold per day (Tsolis et al., 1999a). Thistriggers an influx of polymorphonuclear cells into the tissues and leads to formation ofgranulomata and development of hepato- and speno-megaly. Subsequently, when bac-terial levels in the liver and spleen reach over 108 CFU/tissue, there is a breakout of thepathogen and development of bacteraemia. Liver and spleen damage and dysfunctionappear to be the main cause of death in susceptible mice (Tsolis et al., 1999a).


Table 1. Comparison of LD50 values (CFU) of Salmonella enterica serovars Typhimurium andEnterica reported for various strains of mice

S. typhimurium S. enteritidis

Route SC IV IP Oral Oral

Balb/c <10 20 <10 2 × 104−9 × 105 20−2 × 105

C57Bl/6 20 <10 <10 3 × 104−2 × 105

B10 <10 20B10.D2 <10−50 <10DBA/2 2 × 105 40C3H/HeN 1 × 106 1 × 103−4 × 103 >1 × 107

C3H/HeJ 1 × 104

CD-1 2 × 106

CBA 1 × 107 1 × 103−1 × 104 >1 × 108 >1 × 108

A/J 2 × 106 2 × 104 1 × 104 −1 × 105

SC, subcutaneous injection; IV, intravenous injection; IP, intraperitoneal injection; Oral, administered by gavage.

Based on data from: Carter and Collins, 1974; Plant and Glynn, 1974; Hormaeche, 1979; Eisenstein et al., 1982; Lockman and Curtiss, 1990; Morrissey and Charrier, 1991; Baumler et al., 1996;van der Velden et al., 1998; Lu et al., 1999; Mastroeni et al., 1999; Shea et al., 1999; Edwards et al.,2000; Monack et al., 2000; Ikeda et al., 2001; Nicholson and Baumler, 2001; Schmitt et al., 2001; van der Straaten et al., 2001.

S. typhimurium numbers in the liver and spleen of resistant (ityr) mice do notincrease in this uncontrolled manner. This is probably as a result of effective clear-ance from the tissues, suppression of proliferation or limitation of systemic spreadas a result of ity (Nramp-1) expression in macrophages (Hormaeche, 1980; Lang et al., 1997; Mastroeni et al., 1999; Cuellar-Mata et al., 2002).

Many virulence factors and their modes of action have been identified throughstudy of S. typhimurium infection in mice (Darwin and Miller, 1999; Galan, 2001;Ohl and Miller, 2001). Fimbrial adhesins have been shown to be important in facil-itating initial attachment of the pathogen to Peyer’s patch cells (Baumler et al.,1996; van der Velden et al., 1998; Humphries et al., 2001). The type III secretionsystem encoded on Salmonella pathogenicity island 1 (SPI1) and the array of bio-active components it releases are essential for invasion of the epithelial layer and forproliferation of the pathogen in the subepithelium (Galan, 2001; Ohl and Miller,2001). Salmonella pathogenicity island 2 (SPI2), in particular its type III secretionsystem and effector proteins, Salmonella pathogenicity island 3 (SPI3) andSalmonella plasmid virulence (spv) genes are required for survival and proliferationof the pathogen at systemic sites (Galan, 2001; Ohl and Miller, 2001).

Flagellin, the core protein of the flagella, has also been identified as a potent trigger of cytokine release and inflammation (Eaves-Pyles et al., 2001; Gewirtz et al.,2001; Ikeda et al., 2001). However, its role in S. typhimurium infection of miceremains equivocal. Bacteria unable to express flagellin (FliC blocked or deleted) werefound to be less able to cause infection than wild-type counterparts (Carsiotis et al.,1984; Schmitt et al., 2001). However, in other cases, the mutant strains appeared asinfective as wild-type strains (Lockman and Curtiss, 1990; Ikeda et al., 2001).

Colonization, invasion, systemic spread, persistence and proliferation by S. typhimurium in the mouse occurs progressively, as a result of the integrated actionof the various virulence factors expressed by the bacterium. However, no one viru-lence factor appears to be an absolute requirement for these processes to occur.Whilst interference with SPI1 or SPI2 or deletion of multiple fimbriae greatlyreduces pathogenicity of S. typhimurium, the mutated strains can still cause a lethalinfection, albeit that the bacterial numbers required are significantly increased. Thiswould indicate the bacterium has the inherent capacity to compensate, at least inpart, for functional impairment or loss of individual virulence factors.

Few studies appear to have dealt with the effects of S. enteritidis in mice.However, the susceptibility of various mouse strains to S. enteritidis (table 1) and thenature of the infection it causes appear, at least in general, to be similar to those foundfor S. typhimurium (Carter and Collins, 1974; Humphrey et al., 1996; Lu et al., 1999).

7. OTHER INFLUENCES ON PATHOGENICITY IN THE MOUSE

The severity of the infection caused by Salmonella can vary significantly when thepathogen is administered orally to mice. This is, at least in part, due to actions of the


resident gut flora, which can significantly limit the ability of S. typhimurium to colonize the gut and/or translocate and spread systemically (Miller and Bohnhoff,1963; Que and Hentges, 1985). If the flora is disrupted or removed completely, this pro-tection is lost. Thus, as little as 10 CFU S. typhimurium can be lethal to susceptible/antibiotic-treated mice whereas at least 105 CFU is needed in conventional counter-parts (Miller and Bohnhoff, 1963; Que and Hentges, 1985; Murray and Lee, 2000).The level and diversity of the commensal flora can differ significantly betweenbatches of mice and even between individuals, depending on conditions underwhich they have been bred and reared. There is thus likely to be an inherent vari-ability in the susceptibility of mice to oral pathogen challenge. However, this prob-lem applies equally to other animal models.

Re-infection through coprophagy and cross-contamination can also be a factor inmouse studies. For ease of management and on animal welfare grounds, mice aregenerally housed as groups in solid floored caging. As a result, infected mice are exposed to and consume significant amounts of faeces and can therefore be re-infected with the pathogen. This in itself may not be a problem. However, theexpression of key virulence factors by Salmonella and pathogenicity can be changeddramatically due to passage through the gut and the overall virulence of a strain maybe greatly increased (Kampelmachar et al., 1968; Hormaeche, 1979; Old, 1990). Theresponses observed in infected animals, in particular the intestine–pathogen inter-action may therefore be greatly affected by these re-ingested bacteria. Group-housedanimals appear to succumb to Salmonella far more quickly than singly housed ones(Kampelmachar et al., 1968).

Food intake and food quality significantly influence the susceptibility of animalsto Salmonella (Omoike et al., 1990; Peck et al., 1992). Mice are usually given freeaccess to food and it is difficult to control or monitor intake owing to group housing.However, mice develop a strong hierarchy within groups, particularly strains such asthe Balb/c or C57BL. Even though food is freely available, dominated animals tendto eat less. This in turn is likely to affect their susceptibility to infection.

Many studies on the roles of virulence compounds in Salmonella infection usethe LD50 dose as an index of pathogenicity. However, this may be a poor measure.Salmonella produces a plethora of virulence factors and appears able to compensate,at least partially, for loss of individual virulence factors. Equally, a dose that causesdeath of the mouse is likely to trigger multiple tissue and systems dysfunction. Anyeffects due to loss of a virulence compound may thus be swamped by general meta-bolic responses to infection. An example of this may be with flagellin. In its puri-fied form, flagellin has pronounced effects on the metabolism of mice (Eaves-Pyleset al., 2001; Gewirtz et al., 2001). However, inability to express flagellin has limitedor no effect on the pathogenicity of S. typhimurium (Carsiotis et al., 1984; Lockmanand Curtiss, 1990; Ikeda et al., 2001; Schmitt et al., 2001). Studies that utilize lowerinfective doses in combination with monitoring of tissue distribution and keycellular responses in the gut and systemic tissues may be more revealing as to theroles and impact of flagella or other virulence factors in the infection process.


8. MOUSE MODEL OF TYPHOID FEVER

S. typhimurium usually causes a self-limiting gastroenteritis-type infection inhumans and most domesticated animals. The lethal infection caused by the serovarin susceptible (itys) mice is thus atypical. However, it has strong similarities to thedisease caused by S. typhi in humans and chimpanzees. In the absence of a direct in vivo model for S. typhi infection, the S. typhimurium-infected mouse is thereforewidely accepted and used as a model to study typhoid-like disease in vivo (Tsoliset al., 1999a; Ohl and Miller, 2001; Santos et al., 2001a). Use of the mouse modelof typhoid has enabled identification of the routes and mechanisms by whichS. typhi may colonize, invade and cause gut and systemic tissue dysfunction (Tsoliset al., 1999a; Santos et al., 2001a; Ohl and Miller, 2001). In addition, it has high-lighted a number of key bacterial genes that may be targeted in order to reduce vir-ulence. This in turn has facilitated development of attenuated S. typhi strains forpossible use as live oral vaccines. The mouse model of typhoid has thus been essen-tial for development of treatments of this disease and can continue to make a con-tribution in the future.

The present mouse model of typhoid has, however, serious limitations. The incu-bation period for typhoid-like disease in mice is much shorter than that for S. typhi-infection in humans. In addition, the mode of lethality is different (Tsolis et al.,1999a). For S. typhi infection, death appears to be linked primarily to intestinal dam-age whereas liver and spleen dysfunction are the main causes in the mouse model.In addition, S. typhi lacks a number of the factors known to be of particular impor-tance for the infectivity of S. typhimurium (Woodward et al., 1989). As a result, mostof the potential live oral vaccines developed are Typhi strains with defects in specific functional or regulatory genes rather than in those for key virulence factors.Furthermore, although data from the mouse model suggest that live oral vaccinesshould be most effective against S. typhi, killed vaccines actually appear to be moreprotective in human subjects (Eisenstein, 1998; Pang, 1998). S. typhi infection clearlyinvolves additional factors or pathways that do not manifest in the mouse model oftyphoid. A major challenge for the future will be to identify these mechanisms. Thisis unlikely to be achievable with the present model or possibly even in the absence ofS. typhi itself as the infective agent. At present no suitable alternatives exist. However,development of transgenic or knockout mouse strains may in the future lead to amodel that exhibits more of the characteristics of typhoid-like infection.

9. INTERACTIONS IN THE MOUSE GUT

Although the infection caused by S. typhimurium in susceptible (itys) mice is atypi-cal, it has generally been assumed that the nature of early interactions of the pathogenwith the gut would be fairly similar to those in other animal species. Bacterial fimbriae are of particular importance for the initial interactions of S. typhimuriumwith the gut epithelium, in particular with the M cells and for overall pathogenicityin susceptible (itys) mice (Baumler et al., 1996; van der Velden et al., 1998;


Humphries et al., 2001; Santos et al., 2001a). However, in other animal models (rator chick), virulent S. enteritidis strains unable to express five fimbriae (SEF14,SEF17, SEF21, pef and lpf), although possibly transiently disadvantaged, were ableto colonize, invade and persist in the long term as effectively as wild-type counter-parts (Allen-Vercoe et al., 1999; Robertson, 2000; Robertson et al., 2003). Thesecontrasting findings might have been ascribed to differences betweenS. typhimurium and S. enteritidis in their modes of action. However, the nature andseverity of the infection caused by both serovars appear very similar in mice (table 1;Humphrey et al., 1996; Lu et al., 1999; Santos et al., 2001a,b). This is also the casein rats (Naughton et al., 1996). The major difference is in the type of disease elicitedin the respective models. Whilst S. typhimurium and S. enteritidis cause a lethalinfection in mice, they elicit a self-limiting gastroenteritis-like infection in rats(Naughton et al., 1996, 2000; Havelaar et al., 2001; Takumi et al., 2002). This maysuggest that the requirement and potential roles for particular surface appendagesvary according to the animal model or severity of disease. Since the infectioninduced by S. typhimurium in susceptible (itys) mice is atypical, caution may there-fore need to be applied in extrapolating from the S. typhimurium-mouse model tolikely effects of S. typhimurium and S. enteritidis in other species.

10. SALMONELLA IN PIGS

Salmonella infections in domestic pigs Sus scrofa are associated with significant animal suffering and economic losses and pigs may indeed be a reservoir of infectionfor humans (Berends et al., 1997). Pigs acquire a severe systemic disease at leastsuperficially similar to typhoid fever as a result of infection with S. cholerasuis (Grayet al., 1996; Baumler et al., 1998, Anderson et al., 1998). In contrast, they do not contract systemic disease when infected with S. typhi (Metcalf et al., 2000) or S. typhimurium (Wilcock and Schwartz, 1992). Carriage of S. typhi in the porcine tonsilhas, however, been observed (Metcalf et al., 2000). This organ was previously notedto be important in S. cholerasuis (Gray et al., 1996) and S. typhimurium infection ofpigs (Wood et al., 1989, 1991; Wood and Rose, 1991; Fedorka-Cray et al., 1995).

Natural infection of pigs with Salmonella typhimurium is associated with anenteric disease (Wilcock and Schwartz, 1992), that is a major problem affecting grow-ing pigs in several parts of the world (Reed et al., 1986; Grøndahl et al., 1998). Themechanisms by which Salmonella typhimurium cause this, particularly scouring, arepoorly understood and have largely been extrapolated from work done in rodents.Bacterial invasion is not correlated with the incidence or severity of diarrhoea (Walliset al., 1986, 1989). An inflammatory response with an influx of polymorphonuclearleucocytes is needed to trigger scouring, since S. typhimurium fails to induce fluidsecretion in the absence of leucocyte influx (Giannella et al., 1975). Another virulencefactor seems to be an enterotoxin with a partial resemblance to cholera toxin (CT) andEscherichia coli heat labile enterotoxin (LT) (Wallis et al., 1986; Chopra et al., 1991).


However, despite the efforts of many groups of investigators, a role for an enterotoxinin S. typhimurium-induced fluid secretion has yet to be defined. Unlike CT and LT, theS. typhimurium enterotoxin is not secreted (Wallis et al., 1986).

The digestive tract of the domestic pig has distinct structural and functional similarities to that of humans (Kidder et al., 1961; Hill, 1970). In addition, experi-mental pigs infected orally with S. cholerasuis or S. typhimurium exhibit diseasesthat are almost identical to those reported in the field (Reed et al., 1986). Pigs there-fore provide a good model for study of the potential effects of gut–microbe–foodinteractions on colonization, invasion and persistence in humans by Salmonella spp.

11. SALMONELLA AND CALVES

Salmonellosis is a significant cause of ill health, poor growth or weight loss andoccasionally death in cattle. Many serotypes can cause infection but S. dublin and S. typhimurium are the serovars most often associated with disease (Wray, 1991). S. dublin is predominant in older animals, causing both enteric and severe systemicinfections (Wray, 1991; Wallis et al., 1995; Watson et al., 1995; Libby et al., 1997).However, it can also be present in a chronic carrier state or may trigger abortion incows and heifers without otherwise eliciting other overt clinical signs of infection(Sojka et al., 1974; Hall et al., 1979). Scouring is generally the major clinical signof S. dublin infection of calves. Nonetheless, in the long term, bacteraemia, feverand severe systemic dysfunction can also develop, often in the absence of ongoing gastroenteritis (Wray and Sojka, 1981; Wray, 1991).

S. typhimurium is the main cause of salmonellosis in calves. It causes acute gas-troenteritis, with severe scouring developing around 12−48 h post-infection (Wray,1991; Wallis et al., 1995; Santos et al., 2001a,b). At moderate infective doses(approximately 106 CFU), the infection is self-limiting and usually clears by around8 days. The disorder can be lethal at higher (up to 1010 CFU) infective doses, duemainly to extensive intestinal lesions and dehydration due to extreme fluid loss(Tsolis et al., 1999a; Santos et al., 2001a,b).

S. typhimurium colonizes the calf gastrointestinal tract, attaches to enterocytesand invades the tissue (Santos et al., 2001a,b). Invasion may occur initially viaM cells of ileal Peyer’s patches but can be observed in all regions of the ileal epithe-lium soon after infection. The pathogen translocates rapidly to the mesenteric lymphnodes but, in most cases, only limited spread to systemic organs is evident, evenwith very high infective doses (Kingsley and Baumler, 2000; Santos et al., 2001a,b).Infection causes severe disruption of the epithelium and leads to loss of gut integrity.There is infiltration of polymorphonuclear cells, potent inflammatory responses inthe tissue and extensive loss of fluid and electrolytes.

A number of virulence factors have been identified as potentially important in S. typhimurium infection of calves (Wallis and Galyov, 2000; Wallis et al., 1999).In particular, bioactive components associated with Salmonella pathogenicity


island 1 (SPI1) are crucial for triggering release of inflammatory cytokines, recruit-ment of neutrophils, development of intestinal lesions and onset of scouring. Theflagellar secretory apparatus also appears to be required for maximal influx of neu-trophils and fluid secretion in a bovine ligated intestinal loop model (Ikeda et al.,2001; Schmitt et al., 2001). However, the exact mechanisms responsible for theonset and severity of scouring and gut damage at high and moderate Salmonelladoses remain unclear (Tsolis et al., 1999a,b; Santos et al., 2001a,b).

Bioactive factors linked to other Salmonella pathogenicity islands appear less cru-cial for S. typhimurium-linked gastrointestinal disease in calves (Tsolis et al., 1999a,b).They do, however, have key roles in S. dublin infection, in particular systemic spread,survival and proliferation of the pathogen (Wallis et al., 1999; Wood et al., 1998). Aswith S. typhimurium, SPI1 appears to be important in the early responses of the gut toS. dublin, but other SPIs, such as SPI5, may also have roles (Wood et al., 1998).

S. typhimurium infection of calves in vivo is used to model human gastroenteritisin order to study pathogen–gut interactions during development of this disorder. Theintestinal loop in situ model has been particularly useful in short-term (up to 12 h)study of inflammatory responses, neutrophil recruitment and fluid secretion provokedby Salmonella and the likely roles of specific virulence components (Wallis et al.,1995; Tsolis et al., 1999a; Santos et al., 2001a,b). In some circumstances, the modelsin situ and in vivo may, however, give contrasting results, because they are dealingwith short- or long-term responses respectively. For example, neutrophil recruitmentand fluid secretion were attenuated when sopB-blocked mutants of S. typhimurium orS. dublin were tested in ligated loops (Wood et al., 1998; Santos et al., 2001a,b).Nonetheless, a sopB-blocked S. typhimurium mutant was as pathogenic as a wild-typestrain in vivo when administered at a dose of 1010 CFU (Tsolis et al., 1999a,b). Thismay have arisen because the Salmonella could compensate for the loss of sopB andwas thus only transiently disadvantaged (Tsolis et al., 1999a; Santos et al., 2001a,b).

S. typhimurium causes a lethal infection if given in very high numbers to calvesbut a self-limiting gastroenteritis if administered in moderate amounts (Tsolis et al.,1999a; Santos et al., 2001a,b). In both cases, the infection is located primarily inthe intestine and mesenteric lymph nodes. However, the physiological responses andgut disruption triggered by the lethal infective dose are far more severe and likely toinvolve additional virulence components. Any changes in pathogenicity owing toblocking of an individual virulence factor, such as sopB, may therefore be maskedwhen high infective doses of Salmonella are used. Study of the roles of individualvirulence factors may be better achieved with lower infective doses, where thedisease caused by the wild-type strain is of a self-limiting type.

12. HOST SPECIFICITY

The factors influencing Salmonella serotype host-specificity remain poorly defined,although there is some evidence that bacterial survival within macrophages is


important (Kok et al., 2001; Santos et al., 2001a). S. typhimurium can persist in significantly higher numbers than S. typhi in primary murine macrophages in vitro(Pascopella et al., 1995; Schawn and Kopecko, 1997), which correlates with the virulence of these serotypes for mice. Similarly S. typhi persists better in humanmacrophages compared to murine macrophages in vitro, which again correlates withvirulence. However, the relative persistence of S. typhi and S. typhimurium in humanmacrophages varies considerably between different studies (Vladoianu et al., 1990;Alpuche-Aranda et al., 1995; Ishibashi and Arai, 1996). This variation may beexplained by non-quantifiable differences in the magnitude of macrophage lysis,which can have a major effect on the interpretation of results from bacterial persist-ence studies (Guilloteau et al., 1996; Sizemore et al., 1997).

13. DIET AND SALMONELLA INFECTION

Diarrhoea is highly prevalent among malnourished children (Scrimshaw, 1970;Bovee-Oudenhoven et al., 1997). Furthermore, epidemiological studies have shownthat the duration and severity of diarrhoea increase with malnutrition (Chen et al.,1981; Black et al., 1984). However, the relationship between infection and malnu-trition is complex and its determinants remain poorly understood (Guggenheim andBuechler, 1947; Maenza et al., 1970; Peck et al., 1992). Omoike et al. (1990)showed that adherence of Salmonella to the intact mucosa in situ was consistentlyhigher in well-fed rats rather than malnourished Sprague-Dawley rats. Nonetheless,the infection caused very marked cell destruction and lysis in tissues of malnour-ished rats, whereas only limited histological changes were detected in mucosa fromthe well-fed animals. Adherence of pathogen to the gut is clearly not the only factordictating the severity of the infection. It is likely that malnutrition severely impairsintestinal cell metabolism and reduces the efficacy of the local and systemic immunesystems. This may greatly increase the permeability of the gut to the pathogen or itsproducts and predispose the malnourished animal to salmonellosis.

Rats on a low calcium milk diet had a significantly impaired colonization resist-ance to S. enteritidis (Bovee-Oudenhoven et al., 1996). The barrier potential of thegut was reduced and the pathogen was able to translocate in significant numbersacross the intestine to the systemic circulation. Biliary factors may have a significantrole in infection by S. enteritidis in the rat since diversion of bile from the intestinelumen greatly reduced the ability of the pathogen to invade and spread systemically(Islam et al., 2000). Newberne et al. (1968) observed a much higher mortality rate inS. typhimurium-infected rats if they were fed diets deficient in copper or in protein.The severity of infection tended to be greater in Charles River and Holtzman rats feda 12% casein diet than in those fed an 18% casein diet (Newberne et al., 1968).Similarly, nitrogen losses by Salmonella-infected rats were higher if they were fedlow-protein diets instead of diets containing adequate amounts of protein (McGuireet al., 1968). A net protein-catabolic response to an infection stress has also been


shown to occur in humans (Beisel, 1966). Muscle polysome profiles are also alteredwhen Salmonella-infected rats are given a low-protein diet (Young et al., 1968).Faecal nitrogen outputs by rats were found to be much higher with S. typhimuriuminfection than for S. enteritidis infection (Naughton et al., 1996). It has also beenshown that dietary plant lectins, such as GNA (from Galanthus nivalis, the snow-drop) and Concanavlin A (ConA, from the Jack bean), can alter the course ofSalmonella infection (Naughton et al., 2000). The lectin GNA was shown todecrease the numbers of Salmonella and ConA was shown to increase the numbersof Salmonella in the small intestine for rats. Particular plant lectins may, therefore,interfere with the association of type 1 fimbriae with the rat gastrointestinal tract(Naughton et al., 2001a,b). The rat may have significant advantages as a model forthe study of dietary influences on susceptibility to and severity of salmonellosis.


The growth in the elderly human population in much of the developed world, thehigh incidence of immune deficiency diseases worldwide and rising degrees ofantibiotic resistance amongst pathogens mean that novel strategies to prevent orlimit pathogenic infections are urgently required. A significant amount of researchwill be required into diseases affecting the gastrointestinal tract. Ways will have tobe found to ameliorate the effects of old age or immunodeficiency on susceptibilityto infection. There will be increasing efforts to modify and supplement the diet in amanner that will prevent or limit pathogenic infections, support gut function andpromote gut integrity and health. With the plethora of animal models available to us,it belies researchers to choose wisely and address the correct questions.

REFERENCES

Aleekseev, P.A., Berman, M.I., Korneeva, E.P., 1960. Clinical and histological picture of Salmonellatyphimurium infection in children. J. Microbiol. Epid. Immunobiol. 31, 133–139.

Allen-Vercoe, E., Sayers, A.R., Woodward, M.J., 1999. Virulence of Salmonella enterica serotype Enteritidisaflagellate and afimbriate mutants in a day-old chick model. Epidemiol. Infect. 122, 395–402.

Alpuche-Aranda, C.M., Berthiaume, E.P., Mock, B., Swanson, J.A., Miller, S.I., 1995. Spacious phago-some formation within mouse macrophages correlates with Salmonella serotype pathogenicity andhost susceptibility. Infect. Immun. 63, 4456–4462.

Anderson, R.C., Nisbet, D.J., Buckley, S.A., 1998. Experimental and natural infection of early weanedpigs with Salmonella cholerasuis. Res. Vet. Sci. 64, 261–262.

Bakken, K., Vogelsang, T.M., 1950. The pathogenesis of Salmonella in mice. Acta Path. Microbiol.Scand. 27, 41–50.

Baumler, A.J., Tsolis, R.M., Heffron, F., 1996. The lpf fimbrial operon mediates adhesion of Salmonellatyphimurium to murine Peyer’s patches. Proc. Natl. Acad. Sci. USA 93, 279–283.

Baumler, A.J., Tsolis, R.M., Ficht, T.A., Adams, L.G., 1998. Evolution of host adaption in Salmonellaenterica. Infect. Immun. 66, 4579–4587.

Bean, N.H., Griffin, P.M., 1990. Foodborne disease outbreaks in the United States 1973−1987,pathogens, vehicles and trends. J. Food Prot. 53, 804–817.

Beisel, W.R., 1966. Effect of infection on human protein metabolism. Fed. Proc. 25, 1682–1687.


Berends, B.R., van Kappen, F., Snijers, J.M., Mossel, D.A., 1997. Identification and quantification of riskfactors regarding Salmonella species on pork carcasses. Int. J. Food Microbiol. 36, 199.

Black, R.E., Brown, K.H., Becker, S., 1984. Malnutrition is a determining factor in diarrheal duration,but not incidence among young children in a longitudinal study in rural Bangladesh. Amer. J. Clin.Nutr. 39, 87–94.

Blaser, M.J., Feldman, R.A., 1981. From the centers for disease control. Salmonella bacteremia, reportsto the Centers for Disease Control, 1968−1979. J. Infect. Dis. 143, 743–746.

Bloomfield, A.L., Lew, A., 1942. Significance of Salmonella in ulcerative cecitis of rats. Proc. Soc. Exp.Biol. Med. 51, 179–182.

Böhme, D.H., Schneider, H.A., Lee, J.M., 1959. Some physiopathological parameters of natural resist-ance to infection in murine salmonellosis. J. Exp. Med. 110, 9–25.

Bovee-Oudenhoven, I.M., Termont, D.S., Dekker, R., van der Meer, R., 1996. Calcium in milk and fer-mentation by yoghurt bacteria increase the resistance of rats to Salmonella infection. Gut 38, 59–65.

Bovee-Oudenhoven, I.M., Termont, D.S., Weerkamp, A.H., Faassen-Peters, M.A., Van der Meer, R.,1997. Dietary calcium inhibits the intestinal colonization and translocation of Salmonella in rats.Gastroenterology 113, 550–557.

Bryan, F.I., 1977. Diseases transmitted by foods contaminated by wastewater. J. Food Protect. 40, 45–56.Buchwald, D.S., Blaser, M.J., 1984. A review of human salmonellosis II. Duration of excretion follow-

ing infection with nontyphi Salmonella. Rev. Infect. Dis. 6, 345–356.Candy, D.C.A., Stephen, J., 1989. Salmonella. In: Farthing, M.J.G., Keusch, G.T. (Eds.), Enteric Infection:

Mechanisms, Manifestations and Management. Chapman and Hall Medical, London, p. 289.Canonne-Hergaux, F., Gruenheid, S., Govoni, G., Gros, P., 1999. The Nramp1 protein and its role in

resistance to infection and macrophage function. Proc. Assoc. Amer. Phys. 111, 283–289.Carsiotis, M., Weinstein, D.L., Karch, H., Holder, I.A., O’Brien, A.D., 1984. Flagella of Salmonella

typhimurium are a virulence factor in infected C57BL/6J mice. Infect. Immun. 46, 814–818.Carter, P.B., Collins, F.M., 1974. The route of enteric infection in normal mice. J. Exp. Med. 139,

1189–1203.Chen, L.C., Huq, E., Huffman, S.L., 1981. A prospective study of the risk of diarrheal diseases accord-

ing to the nutritional status of children. Amer. J. Epidemiol. 114, 284–292.Chopra, A.K., Peterson, J.W., Houston, C.W., Pericas, R., Prasad, R., 1991. Enterotoxin associated DNA

sequence homology between Salmonella species and Escherichia coli. FEMS Microbiol. Lett. 77,133–138.

Collins, F.M., Carter, P.B., 1978. Growth of salmonellae in orally infected germfree mice. Infect. Immun.21, 41–47.

Cuellar-Mata, P., Jabado, N., Liu, J., Furuya, W., Finlay, B.B., Gros, P., Grinstein, S., 2002. Nramp1 mod-ifies the fusion of Salmonella typhimurium-containing vacuoles with cellular endomembranes inmacrophages. J. Biol. Chem. 277, 2258–2265.

Darwin, K.H., Miller, V.L., 1999. Molecular basis of the interaction of Salmonella with the intestinalmucosa. Clin. Microbiol. Rev. 12, 405–428.

Davies, R.H., Wray, C., 1995. Mice as carriers of Salmonella enteritidis on persistently infected poultryunits. Vet. Rec. 137, 337–341.

Eaves-Pyles, T., Murthy, K., Liaudet, L., Virag, L., Ross, G., Soriano, F.G., Szabo, C., Salzman, A.L.,2001. Flagellin, a novel mediator of Salmonella-induced epithelial activation and systemic inflam-mation: I kappa B alpha degradation, induction of nitric oxide synthase, induction of proinflamma-tory mediators, and cardiovascular dysfunction. J. Immunol. 166, 1248–1260.

Edwards, R.A., Schifferli, D.M., Maloy, S.R., 2000. A role for Salmonella fimbriae in intraperitonealinfections. Proc. Natl. Acad. Sci. USA 97, 1258–1262.

Eisenstein, T.K., 1998. Intracellular pathogens, the role of antibody-mediated protection in Salmonellainfection. Trends Microbiol. 6, 135–136.

Eisenstein, T.K., Deakins, L.W., Killar, L., Saluk, P.H., Sultzer, B.M., 1982. Dissociation of innate sus-ceptibility to Salmonella infection and endotoxin responsiveness in C3HeB/FeJ mice and other strainsin the C3H lineage. Infect. Immun. 36, 696–703.

Fedorka-Cray, P.J., Kelley, L.C., Stabel, T.J., Gray, J.T., Laufer, J.A., 1995. Alternative routes of invasionmay affect pathogenesis of Salmonella typhimurium in swine. Infect. Immun. 63, 2658–2664.


Finkelstein, R.A., Marchiewicz, B.A., McDonald, R.J., Boesman-Finkelstein, M., 1983. Isolation andcharacterisation of a cholera related enterotoxin from Salmonella typhimurium. FEMS Microbiol.Lett. 17, 239.

Forbes, J.R., Gros, P., 2001. Divalent-metal transport by NRAMP proteins at the interface of host-pathogen interactions. Trends Microbiol. 9, 397–403.

Galan, J.E., 2001. Salmonella interactions with host cells, Type III secretion at work. Ann. Rev. Cell. Dev.Biol. 17, 53–86.

Gay, F.P., 1918. Typhoid Fever Considered as a Problem of Scientific Medicine. Macmillan, New York,p. 286.

Gewirtz, A.T., Simon, P.O. Jr, Schmitt, C.K., Taylor, L.J., Hagedorn, C.H., O’Brien, A.D., Neish, A.S.,Madara, J.L., 2001. Salmonella typhimurium translocates flagellin across intestinal epithelia,inducing a proinflammatory response. J. Clin. Invest. 107, 99–109.

Giannella, M.D., Gots, R.E., Cahaney, A.N., Greenough, W.B., Formal, S.B., 1975. Pathogenesis ofSalmonella-mediated intestinal fluid secretion. Activation of adenylate cyclase and inhibition byindomethacin. Gastroenterology 69, 1238–1245.

Gill, O.N., Sockett, P.N., Barlett, C.I.R., 1983. Outbreak of Salmonella napoli infection caused bycontaminated chocolate bars. Lancet 1, 574–577.

Gray, J.T., Fedorka-Cray, P.J., Stabel, T.J., Kramer, T.T., 1996. Natural transmission of Salmonellacholerasuis in swine. Appl. Environ. Microbiol. 62, 141–146.

Grøndahl, M.L., Jensen, G.M., Nielsen, C.G., Skadhauge, E., Olsen, J.E., Hansen, M.B., 1998. Secretorypathways in Salmonella typhimurium-induced fluid accumulation in the porcine small intestine.J. Med. Microbiol. 47, 151–157.

Guggenheim, K., Buechler, E., 1947. Nutrition deficiency and resistance to infection. The effect ofcaloric and protein deficiency on the susceptibility of rats and mice to infection with Salmonellatyphimurium. J. Hyg. 45, 103–109.

Guilloteau, L.A., Wallis, T.S., Gautier, A.V., MacIntyre, S., Platt, D.J., Lax, A.J., 1996. The Salmonellavirulence plasmid enhances Salmonella-induced lysis of macrophages and influences inflammatoryresponses. Infect. Immun. 64, 3385–3393.

Habermann, R.T., Williams, Jr. F.P., 1958. Salmonellosis in laboratory animals. J. Nat. Cancer Inst. 20,933–942.

Hall, G.A., Jones, P.W., Parsons, K.R., Chanter, N., Aitken, M.M., 1979. Studies of the virulence ofSalmonella dublin in experimental infections of cattle and rats. Brit. Vet. J. 135, 243–248.

Havelaar, A.H., Garssen, J., Takumi, K., Koedam, M.A., Dufrenne, J.B., van Leusden, F.M., de LaFonteyne, L., Bousema, J.T., Vos, J.G., 2001. A rat model for dose-response relationships ofSalmonella enteritidis infection. J. Appl. Microbiol. 91, 442–452.

Hill, K.J., 1970. Development and comparative aspects of digesta. In: Swenson, M.J. (Ed.), DukesPhysiology of Domestic Animals. Cornell University Press, Ithaca, NY, pp. 409–423.

Hormaeche, C.E., 1979. Natural resistance to Salmonella typhimurium in different inbred mouse strains.Immunology 37, 311–318.

Hormaeche, C.E., 1980. The in vivo division and death rates of Salmonella typhimurium in the spleensof naturally resistant and susceptible mice measured by the superinfecting phage technique ofMeynell. Immunology 41, 973–979.

Hornick, R.B., Greisman, S.E., Woodward, T.E., DuPont, H.L., Dawkins, A.T., Snyder, M.J., 1970.Typhoid fever: pathogenesis and immunologic control. New Engl. J. Med. 283, 686–691.

Humphrey, T.J., Williams, A., McAlpine, K., Lever, M.S., Guard-Petter, J., Cox, J.M., 1996. Isolates ofSalmonella enterica Enteritidis PT4 with enhanced heat and acid tolerance are more virulent in miceand more invasive in chickens. Epidemiol. Infect. 117, 79–88.

Humphries, A.D., Townsend, S.M., Kingsley, R.A., Nicholson, T.L., Tsolis, R.M., Baumler, A.J., 2001.Role of fimbriae as antigens and intestinal colonization factors of Salmonella serovars. FEMSMicrobiol. Lett. 201, 121–125.

Ikeda, J.S., Schmitt, C.K., Darnell, S.C., Watson, P.R., Bispham, J., Wallis, T.S., Weinstein, D.L.,Metcalf, E.S., Adams, P., O’Connor, C.D., O’Brien, A.D., 2001. Flagellar phase variation ofSalmonella enterica serovar Typhimurium contributes to virulence in the murine typhoid infectionmodel but does not influence Salmonella-induced enteropathogenesis. Infect. Immun. 69, 3021–3030.


Isberg, R.R., Barnes, P., 2000. Border patrols and secret passageways across the intestinal epithelium.Trends Microbiol. 8, 291–293.

Ishibashi, Y., Arai, T., 1996. A possible mechanism for host specific pathogenesis of Salmonella serovars.Microb. Pathog. 21, 435–446.

Islam, A.F.M.W., Moss, N.D., Dai, Y., Smith, M.S.R., Collins, A.M., Jackson, G.D.F., 2000.Lipopolysaccharide-induced biliary factors enhance invasion of Salmonella enteritidis in a rat model.Infect. Immun. 68, 1–5.

Jewes, L.A., 1987. Antimicrobial therapy of non-typhi Salmonella and Shigella infection. J. Antimicrob.Chemother. 19, 557–560.

Jones, B.D., Ghori, N., Falkow, S., 1994. Salmonella typhimurium initiates murine infection by penetrating and destroying the specialized epithelial M cells of the Peyer’s patches. J. Exp. Med. 180, 15–23.

Kampelmacher, E.H., Guinée, P.A.M., Lucretia, M., van Noorle, J.L.M., 1968. Artificial Salmonellainfection in rats. Zentralbl. Veterinarmed. 16b, 173–182.

Kidder, D.E., Manners, M.J., McCrea, M.R., 1961. The passage of food through the alimentary tract ofthe piglet. Res. Vet. Sci. 2, 227–231.

Kingsley, R.A., Baumler, A.J., 2000. Salmonella interactions with professional phagocytes. In:Oelschlaeger, T.A., Hacker, J. (Eds.), Bacterial Invasion into Eukaryotic Cells. KluwerAcademic/Plenum Publishers, New York, pp. 321–342.

Kohbata, S., Yokoyama, H., Yabuuchi, E., 1986. Cytopathogenic effect of Salmonella typhi GIFU 10007on M cells of murine ileal Peyer’s patches in ligated ileal loops, an ultrastructural study. Microbiol.Immunol. 30, 1225–1237.

Kok, M., Buhlmann, E., Pechere, J., 2001. Salmonella typhimurium thyA mutants fail to grow intracel-lularly in vitro and are attenuated in mice. Microbiology 147, 727–733.

Kotova, A.L., Konsratskaya, S.A., Yasutis, L.M., 1988. Salmonella carrier state and biological character-istics of the infectious agent. J. Hyg. Epidemiol. Microbiol. Immunol. 32, 71–78.

Lang, T., Prina, E., Sibthorpe, D., Blackwell, J.M., 1997. Nramp1 transfection transfers Ity/Lsh/Bcg-related pleiotropic effects on macrophage activation, influence on antigen processing and presenta-tion. Infect. Immun. 65, 380–386.

Lax, A.J., Barrow, P.A., Jones, P.W., Wallis, T.S., 1995. Current perspectives in salmonellosis. Brit. Vet.J. 151, 351–377.

Lee, K.R., Leggiadro, R.J., Burch, K.J., 1994. Drug use evaluation of antibiotics in a pediatric teachinghospital. Infect. Control Hosp. Epidemiol. 15, 710–712.

Leslie, P.H., 1941. The bacteriological classification of the principal cultures used in rat and mouse con-trol in Great Britain. J. Hyg. 42, 552–562.

Libby, S.J., Adams, L.G., Ficht, T.A., Allen, C., Whitford, H.A., Buchmeier, N.A., Bossie, S., Guiney,D.G., 1997. The spv genes on the Salmonella dublin virulence plasmid are required for severe enteri-tis and systemic infection in the natural host. Infect. Immun. 65, 1786–1792.

Lockman, H.A., Curtiss, R., 1990. Salmonella typhimurium mutants lacking flagella or motility remainvirulent in BALB/c mice. Infect. Immun. 58, 137–143.

Lodge, J., Douce, G.R., Amin, I.I., Bolton, A.J., Martin, G.D., Chatfield, S., Dougan, G., Brown, N.L.,Stephen, J., 1995. Biological and genetic characterization of TnphoA mutants of Salmonellatyphimurium TML in the context of gastroenteritis. Infect. Immun. 63, 762–769.

Lu, S., Manges, A.R., Xu, Y., Fang, F.C., Riley, L.W., 1999. Analysis of virulence of clinical isolates ofSalmonella enteritidis in vivo and in vitro. Infect. Immun. 67, 5651–5657.

Maenza, R.M., Powell, D.W., Plotkin, G.R., Formal, S.B., Jervis, H.R., Sprinz, H., 1970. Experimentaldiarrhoea, salmonella enterocolitis in the rat. J. Infect. Dis. 121, 475–485.

Mandal, B.K., 1994. Salmonella typhi and other salmonellas. Gut 35, 726–728.Mastroeni, P., Clare, S., Khan, S., Harrison, J.A., Hormaeche, C.E., Okamura, H., Kurimoto, M., Dougan,

G., 1999. Interleukin 18 contributes to host resistance and gamma interferon production in miceinfected with virulent Salmonella typhimurium. Infect. Immun. 67, 478–483.

Mattsby-Baltzer, I., Ahlstrom, B., Edebo, L., de Man, P., 1996. Susceptibility of lipopolysaccharide-responsive and -hyporesponsive ItyS mice to infection with rough mutants of Salmonellatyphimurium. Infect. Immun. 64, 1321–1327.


McCullough, N.B., Eisele, C.W., 1951. Experimental human Salmonellosis III. Pathogenicity of strainsof Salmonella newport, Salmonella derby, and Salmonella bareilly obtained from spray-dried wholeegg. J. Infect. Dis. 113, 59–66.

McGuire, E.A., Young, V.R., Newberne, P.M., Payne, B.J., 1968. Effects of Salmonella typhimuriuminfection. Arch. Path. 86, 60–68.

Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C., Griffin, P.M., Tauxe, R.V.,1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5, 607–625.

Merican, I., 1997. Typhoid fever, present and future. Med. J. Malaysia 52, 299–308.Metcalf, E.S., Almond, G.W., Pouth, P.A., Horton, J.R., Dillman, R.C., Orndorff, P.E., 2000.

Experimental Salmonella typhi infection in the domestic pig, Sus scrofa domestica. Microb. Pathog.29, 121–126.

Miller, C.P., Bohnhoff, M., 1963. Changes in the mouse’s enteric microflora associated with enhancedsusceptibility to Salmonella infection following streptomycin treatment. J. Infect. Dis. 113, 59–66.

Mintz, E.D., Carter, M.L., Halder, J.L., Wassell, J.T., Zingeser, J.A., Tauxe, R.V., 1994. Dose-responseeffects in an outbreak of Salmonella enteritidis. Epidemiol. Infect. 112, 13–23.

Monack, D.M., Hersh, D., Ghori, N., Bouley, D., Zychlinsky, A., Falkow, S., 2000. Salmonella exploitscaspase-1 to colonize Peyer’s patches in a murine typhoid model. J. Exp. Med. 192, 249–258.

Morrissey, P.J., Charrier, K., 1991. Interleukin-1 administration to C3H/HeJ mice after but not prior toinfection increases resistance to Salmonella typhimurium. Infect. Immun. 59, 4729–4731.

Murray, R.A., Lee, C.A., 2000. Invasion genes are not required for Salmonella enterica serovartyphimurium to breach the intestinal epithelium, evidence that salmonella pathogenicity island 1 hasalternative functions during infection. Infect. Immun. 68, 5050–5055.

Naughton, P.J., Grant, G., Ewen, S.W., Spencer, R.J., Brown, D.S., Pusztai, A., Bardocz, S., 1995.Salmonella typhimurium and Salmonella enteritidis induce gut growth and increase the polyaminecontent of the rat small intestine in vivo. FEMS Immunol. Med. Microbiol. 12, 251–258.

Naughton, P.J., Grant, G., Spencer, R.J., Bardocz, S., Pusztai, A., 1996. A rat model of infection bySalmonella typhimurium or Salm. enteritidis. J. Appl. Bacteriol. 81, 651–656.

Naughton, P.J., Grant, G., Bardocz, S., Pusztai, A., 2000. Modulation of Salmonella infection by thelectins of Canavalia ensiformis Con A. and Galanthus nivalis GNA in a rat model in vivo. J. Appl.Microbiol. 88, 720–727.

Naughton, P.J., Grant, G., Bardocz, S., Allen-Vercoe, E., Woodward, M.J., Pusztai, A., 2001a. Expressionof type 1 fimbriae SEF 21 of Salmonella enterica serotype enteritidis in the early colonisation of therat intestine. J. Med. Microbiol. 50, 191–197.

Naughton, P.J., Grant, G., Sojka, M., Bardocz, S., Thorns, C.J., Pusztai, A., 2001b. Survival and distri-bution of cell-free SEF 21 of Salmonella enterica serovar Enteritidis in the stomach and various com-partments of the rat gastrointestinal tract in vivo. J. Med. Microbiol. 50, 1049–1054.

Newberne, P.M., Hunt, C.E., Young, V.R., 1968. The role of diet and the reticuloendothelial system in theresponse of rats to Salmonella typhimurium infection. J. Nutr. 49, 448–457.

Nicholson, T.L., Baumler, A.J., 2001. Salmonella enterica serotype typhimurium elicits cross-immunityagainst a Salmonella enterica serotype enteritidis strain expressing LP fimbriae from the lac promoter.Infect. Immun. 69, 204–212.

O’Brien, A.D., 1982. Innate resistance of mice to Salmonella typhi infection. Infect. Immun. 38,948–952.

Ohl, M.E., Miller, S.I., 2001. Salmonella, a model for bacterial pathogenesis. Ann. Rev. Med. 52,259–274.

Old, D.C., 1990. Salmonella. In: Parker, N.T., Collier, L.E. (Eds.), Topley & Wilson’s Principles ofBacteriology, Virology and Immunity. Edward Arnold, London, pp. 469–493.

Omoike, I., Lindquist, B., Abud, R., Merrick, J., Lebenthal, E., 1990. The effect of protein-energy mal-nutrition and refeeding on the adherence of Salmonella typhimurium to small intestinal mucosa andisolated enterocytes in rats. J. Nutr. 120, 404–411.

Orskøv, J., Jensen, K.A., Kobayashi, K., 1928. Studien über Bresinulnfektion der Mäuse, speziell mitRücksicht auf die Bedeutung des Retikuloendothelialgewebes (Studies on the Bresinu-infection ofmice, with special reference to the significance of the reticulo-endothelial tissue). Z. Immun.Allergieforsch 55, 34–68.


Pang, T., 1998. Vaccination against intracellular bacterial pathogens. Trends Microbiol. 6, 433.Pascopella, L., Raupach, B., Ghori, N., Monack, D., Falkow, S., Small, P.L., 1995. Host restriction

phenotypes of Salmonella typhi and Salmonella gallinarum. Infect. Immun. 63, 4329–4335.Peck, M.D., Babcock, G.F., Alexander, J.W., 1992. The role of protein and calorie restriction in outcome

from Salmonella infection in mice. J. Parenter. Enteral. Nutr. Nov–Dec, 16 (6), 561–565.Pie, S., Matsiota-Bernard, P., Truffa-Bachi, P., Nauciel, C., 1996. Gamma interferon and interleukin-10

gene expression in innately susceptible and resistant mice during the early phase of Salmonellatyphimurium infection. Infect. Immun. 64, 849–854.

Plant, J., Glynn, A.A., 1974. Natural resistance to salmonella infection, delayed hypersensitivity andIr genes in different strains of mice. Nature 248, 245–247.

Que, J.U., Hentges, D.J., 1985. Effect of streptomycin administration on colonization resistance toSalmonella typhimurium in mice. Infect. Immun. 48, 169–174.

Radoucheva, T., Kurteva, J., Markova, N., Veljanov, D., Najdenski, H., 1994. Behaviour of Salmonelladublin in mice and rats upon intraperitoneal infection. Zentralbl. Bakteriol. 2804, 520–525.

Rampling, A., 1993. Salmonella enteritidis five years on. Lancet 342, 317–318.Reed, W.M., Olander, H.J., Thacker, H.I., 1986. Studies on the pathogenesis of Salmonella typhimurium

and Salmonella cholerasuis var kunzendrof infection in weaning pigs. Amer. J. Vet. Res. 47, 75–83.Resigno, M., Barrow, P., 2001. The host-pathogen interaction, new themes from dendritic cell biology.

Cell 106, 267–270.Robertson, J.M.R., 2000. Roles of Fimbriae and Flagella during Salmonella enterica Serovar Enteritidis

Infection in the Rat. PhD Thesis, University of Aberdeen.Robertson, J.M.R., McKenzie, N.H., Duncan, M., Allen-Vercoe, E., Woodward, M.J., Flint, H.J., Grant,

G., 2003. Lack of flagella disadvantages Salmonella enterica serovar Enteritidis during the earlystages of infection in the rat. J. Med. Microbiol. 52, 91–99.

Ryan, C.A., Hargrett-Bean, N.T., Blake, P.A., 1989. Salmonella typhi infections in the United States,1975−1984, increasing role of foreign travel. Rev. Infect. Dis. 11, 1–8.

Sandefur, P.D., Peterson, J.W., 1977. Notes: neutralization of Salmonella toxin-induced elongation ofchinese hamster ovary cells by cholera antitoxin. Infect. Immun. 15, 988–992.

Santos, R.L., Zhang, S., Tsolis, R.M., Kingsley, R.A., Garry, A.L., Baumler, A.J., 2001a. Animal modelsof Salmonella infections, enteritis versus typhoid fever. Microbes Infect. 3, 1335–1344.

Santos, R.L., Tsolis, R.M., Zhang, S., Ficht, T.A., Baumler, A.J., Adams, L.G., 2001b. Salmonella-induced cell death is not required for enteritis in calves. Infect. Immun. 69, 4610–4617.

Schawn, W.R., Kopecko, D.J., 1997. Serovar specific differences in Salmonella survival withinmacrophage cells. Adv. Exp. Med. Biol. 412, 277–278.

Schmitt, C.K., Ikeda, J.S., Darnell, S.C., Watson, P.R., Bispham, J., Wallis, T.S., Weinstein, D.L., Metcalf,E.S., O’Brien, A.D., 2001. Absence of all components of the flagellar export and synthesis machinery dif-ferentially alters virulence of Salmonella enterica serovar Typhimurium in models of typhoid fever, sur-vival in macrophages, tissue culture invasiveness, and calf enterocolitis. Infect. Immun. 69, 5619–5625.

Scrimshaw, N.S., 1970. Synergism of malnutrition and infection. Evidence from field studies inGuatemala. JAMA 212, 1685–1692.

Shea, J.E., Beuzon, C.R., Gleeson, C., Mundy, R., Holden, D.W., 1999. Influence of the Salmonellatyphimurium pathogenicity island 2 type III secretion system on bacterial growth in the mouse. Infect.Immun. 67, 213–219.

Silliker, J.H., 1980. Status of Salmonella, ten years later. J. Food Protect. 43, 307–313.Sizemore, D.R., Elinghorst, E.A., Eck, L.C., Branstrom, A.A., Hoover, D.L., Warren, R.L., Rubin, F.A.,

1997. Interaction of Salmonella typhi strains with cultured human monocyte-derived macrophages.Infect. Immun. 65, 309–312.

Sojka, W.J., Thomson, P.D., Hudson, E.B., 1974. Excretion of Salmonella dublin by adult bovine carriers.Brit. Vet. J. 130, 482–488.

Takumi, K., Garssen, J., Havelaar, A., 2002. A quantitative model for neutrophil response and delayed-type hypersensitivity reaction in rats orally inoculated with various doses of Salmonella enteritidis.Int. Immunol. 14, 111–119.

Tannock, G.W., Blumershine, R.V., Savage, D.C., 1975. Association of Salmonella typhimurium with,and its invasion of, the ileal mucosa in mice. Infect. Immun. 11, 365–370.


Tauxe, R.V., Pavia, A.T., 1998. Salmonellosis, nontyphoidal. In: Evans, A.S., Brachman, P.S. (Eds.),Bacterial Infections of Humans, Epidemiology and Control. Plenum Medical Book Co., New York,pp. 613–630.

Threlfall, E.J., Ridley, A.M., Ward, L.R., Rowe, B., 1996. Assessment of health risk from Salmonella-based rodenticides. Lancet 348, 616–617.

Tsolis, R.M., Adams, L.G., Ficht, T.A., Baumler, A.J., 1999a. Contribution of Salmonella typhimuriumvirulence factors to diarrheal disease in calves. Infect. Immun. 67, 4879–4885.

Tsolis, R.M., Kingsley, R.A., Townsend, S.M., Ficht, T.A., Adams, L.G., Baumler, A.J., 1999b. Of mice,calves, and men. Comparison of the mouse typhoid model with other Salmonella infections. Adv.Exp. Med. Biol. 473, 261–274.

van der Straaten, T., van Diepen, A., Kwappenberg, K., et al., 2001. Novel Salmonella enterica serovartyphimurium protein that is indispensable for virulence and intracellular replication. Infect. Immun.69, 7413–7418.

van der Velden, A.M.W., Baumler, A.J., Tsolis, R.M., Heffron, F., 1998. Multiple fimbrial adhesins arerequired for full virulence of Salmonella typhimurium in mice. Infect. Immun. 66, 2803–2808.

Vazquez-Torres, A., Jones-Carson, J., Baumler, A.J., Falkow, S., Valdivia, R., Brown, W., Le M.,Berggren, R., Parks, W.T., Fang, F.C., 1999. Extraintestinal dissemination of Salmonella by CD18-expressing phagocytes. Nature 401, 804–808.

Veljanov, D.K., Todorov, I.G., Harbov, D.D., 1984. Virulence for different hosts of strains S. Dublin iso-lated during epizootic and sporadic salmonellosis. Compt. Rend. Acad. Bulg. Sci. 37, 69–71.

Vladoianu, I., Chang, H.R., Pechere, J.C., 1990. Expression of host resistance to Salmonella typhi andSalmonella typhimurium, bacterial survival within macrophages of murine and human origin.Microb. Pathog. 8, 83–90.

Wallis, T.S., Galyov, E.E., 2000. The secreted effector protein of Salmonella dublin, SopA, is translo-cated into eukaryotic cells and influences the induction of enteritis. Cell Microbiol. 2, 293–303.

Wallis, T.S., Starkey, W.G., Stephen, J., Haddon, S.J., Osborne, M.P., Candy, D.C., 1986. Enterotoxinproduction by Salmonella typhimurium strains of different virulence. J. Med. Microbiol. 21, 19–23.

Wallis, T.S., Hawker, R.J., Candy, D.C., Qi, G.M., Clarke, G.J., Worton, K.J., Osborne, M.P., Stephen, J.,1989. Quantification of the leucocyte influx into rabbit ileal loops induced by strains of Salmonellatyphimurium of different virulence. J. Med. Microbiol. 30, 149–156.

Wallis, T.S., Paulin, S.M., Plested, J.S., Watson, P.R., Jones, P.W., 1995. The Salmonella dublin virulenceplasmid mediates systemic but not enteric phases of salmonellosis in cattle. Infect. Immun. 63,2755–2761.

Wallis, T.S., Wood, M., Watson, P., Paulin, S., Jones, M., Galyov, E., 1999. Sips, Sops, and SPIs but notstn influence Salmonella enteropathogenesis. Adv. Exp. Med. Biol. 473, 275–280.

Watson, P.R., Paulin, S.M., Bland, A.P., Jones, P.W., Wallis, T.S., 1995. Characterization of intestinalinvasion by Salmonella typhimurium and Salmonella dublin and effect of a mutation in the invH gene.Infect. Immun. 63, 2743–2754.

Weinstein, D.L., O’Neill, B.L., Hone, D.M., Metcalf, E.S., 1998. Differential early interactions betweenSalmonella enterica serovar Typhi and two other pathogenic Salmonella serovars with intestinalepithelial cells. Infect. Immun. 66, 2310–2318.

Wilcock, B.P., Schwartz, K.L., 1992. Salmonellosis. In: Leman, A.D., Ames, I.A. (Eds.), Diseases ofSwine, 7th edition. Iowa University Press, Ames, Iowa, pp. 570–583.

Wood, M.W., Jones, M.A., Watson, P.R., Siber, A.M., McCormick, B.A., Hedges, S., Rosqvist, R.,Woodward, M.J., McLaren, I., Wray, C., 1989. Distribution of virulence plasmids withinSalmonellae. J. Gen. Microbiol. 135, 503–511.

Wood, M.W., Jones, M.A., Watson, P.R., Hedges, S., Wallis, T.S., Galyov, E.E., 1998. Identification of apathogenicity island required for Salmonella enteropathogenicity. Mol. Microbiol. 29, 883–891.

Wood, R.L., Rose, R., 1991. Population of Salmonella typhimurium in internal organs of experimentallyinfected carrier swine. Amer. J. Vet. Res. 53, 653–658.

Wood, R.L., Pospischil, A., Rose, R., 1989. Distribution of persistent Salmonella typhimurium infectionin internal organs of swine. Amer. J. Vet. Res. 50, 1015–1021.

Wood, R.L., Rose, R., Coe, N.E., Ferris, K.E., 1991. Experimental establishment of persistent infectionin swine with a zoonotic strain of Salmonella newport. Amer. J. Vet. Res. 52, 813–819.


Woodward, M.J., McLaren, I., Wray, C., 1989. Distribution of virulence plasmids within Salmonellae.J. Gen. Microbiol. 135 (Pt 3), 503–511.

Worton, K.J., Candy, D.C.A., Wallis, T.S., Clarke, C.J., Osborne, M.P., Haddon, S.J., Stephen, J., 1989.Studies on early association of Salmonella typhimurium with intestinal mucosa in vivo and in vitro,relationship to virulence. J. Med. Microbiol. 29, 283.

Wray, C., 1991. Salmonellosis in calves. In Pract. 13, 13–15.Wray, C., Sojka, W.J., 1981. Salmonella dublin infection of calves, use of small doses to simulate natu-

ral infection on the farm. J. Hyg. 87, 501–509.Young, V.R., Chen, S.C., Newberne, P.M., 1968. Effect of infection on skeletal muscle ribosomes in rats

fed adequate or low protein. J. Nutr. 94 (3), 361–368.


258

The skin and particularly the mucous membranes of humans and animals are heavilycolonized by indigenous bacteria and, in poultry, bacterial colonization of themucosal surfaces occurs within the first few hours of life (Gibbons, 1977). Neonatalbirds inhale or ingest the majority of the microorganisms that go on to colonize and develop into the normal flora, a flora that contains many diverse populations ofbacteria. In healthy birds, the bacterial composition of the mucosal surfaces remainsrelatively stable and many of these bacteria play an essential role in the developmentand well being of the host. However, if the stability of the native flora is broken,opportunistically pathogenic organisms are able to colonize which may lead topotentially fatal septicaemia. Infectious disease, on the other hand, requires as a firststep, colonization of the host. Colonization for many pathogenic bacteria occurs ini-tially on a mucosal or skin surface (Smith, 1972) and in competition with the nativeflora that is considered to have a protective effect (Nurmi and Rantala, 1973).

Colonization of host tissues by commensal and pathogenic bacteria commonlyinvolves adhesin–receptor interaction to overcome host defence mechanisms such asurination, desquamation and peristaltic propulsion through the intestinal tract,although other factors also play a role in the establishment of infection. Aspects ofcolonization are discussed along with the particular mechanisms employed by spe-cific bacterial species. Additionally, the role of intervention strategies for the modi-fication of the commensal microflora to control bacterial pathogens is discussed.

1. INTRODUCTION

Colonization of host tissues by commensal and pathogenic bacteria commonlyinvolves adhesin–receptor interaction to overcome host defence mechanisms such as

12 Bacterial colonization of avian mucosalsurfaces

R.M. La Ragione, D.G. Newell and M.J. Woodward

Department of Bacterial Diseases, Veterinary Laboratories Agency (Weybridge),Woodham Lane, Addlestone, New Haw, Surrey KT15 3NB, UK


urination, desquamation and peristaltic propulsion through the intestinal tract,although other factors also play a role in colonization. Motility may aid in penetra-tion of mucus (Smyth, 1988). Lipopolysaccharides and capsule may contribute to resistance to bile salts and protect against host iron-binding proteins and non-specific serum factors such as complement (Mims, 1987; Woolcock, 1988).

Gastrointestinal surfaces are dynamic three-dimensional compartments. Thesecompartments are composed of mucilaginous glycoproteins, flowing on glycocalyxoverlaying the epithelial membrane that is convoluted into microvilli (Dyce et al.,1987). The membranes themselves are presumably in a dynamic fluid state withcontinual movement and transposition of their macromolecular constituents(Savage, 1983). One of the many functions of the intestine is to prevent lumen bac-teria and endotoxins from reaching systemic organs and tissues. Failure of the intes-tinal barrier function results in the systemic spread of bacteria from the gut tosystemic organs, a process that is known as translocation (Allori et al., 2000). Berg(1992, 1995, 1999) has defined several routes for bacterial translocation that may beinduced by disruption of the ecological equilibrium in the intestinal tract whichenables bacterial overgrowth, increased permeability of the mucosal barrier, includ-ing disruption, and deficiencies in the host immune system. Interestingly, virulentstrains of certain enteric pathogens were also observed to associate with the intes-tinal mucosa, whereas avirulent mutants did not (Drucker et al., 1967; Smith andHalls, 1968; Bertschinger et al., 1972).

The intestinal mucosa undergoes a continuous renewal process whereby prolif-erating cells differentiate predominantly to enterocytes that migrate up the villus andare sloughed into the lumen from the villus tip (Wright and Alison, 1994). Duringthis migration process, the enterocytes acquire differentiated functions in terms ofdigestion, absorption and mucin secretion (Imondi and Bird, 1966; Weiser, 1973;Meddings et al., 1990; Traber et al., 1991; Ferraris et al., 1992; Thomson et al.,1994). The normal flora of the gastrointestinal (GI) tract contains many diversepopulations of bacteria that play a role in the development and well being of the host(Berg, 1990, 1996). This is an area of novel research and, by way of example,Freitas et al. (2001) have shown that heat labile soluble factors of Bacteroidesthetaiotaomicron, a common gastrointestinal tract inhabitant, increase galactosyla-tion of target cells. In poultry, bacterial colonization of the mucosal surfaces occursin the first few hours of life. The majority of the microorganisms are inhaled oringested by the neonatal birds. However, some bacterial pathogens including E. coliand Salmonella may be acquired by transovarial infection and subsequently the neonatal bird may hatch already infected. In healthy, adult animals, the bacterialcomposition of the mucosal surfaces remains relatively stable. However, that stabilitymay be broken. Internal factors such as hormonal shifts (Havenaar and Huis in’tVeld, 1992) and external factors such as feed deprivation, antibiotic administration,obstruction and radiation (Tannock and Savage, 1974; Berg, 1981; Lenener et al.,1984; Lizko et al., 1984; Deitch et al., 1990; Havenaar and Huis in’t Veld, 1992)

Colonization of avian mucosal surfaces 259

may induce sudden flora changes. Although many of these quoted studies refer tothe mouse model, it is highly likely that the same events are also probable in avianspecies. Under such circumstances, pathogenic and opportunistic pathogenic organ-isms may be able to colonize, leading to localized lesions, bacteraemia and or evensepticaemia (Berg, 1999).

The bacterial composition of the gut microflora varies considerably from oneregion of the GI tract to another. For example, in poultry the crop is lined withLactobacilli, which form a close association with the epithelium and are involved infermentation processes (Barnes et al., 1972, 1980; Fuller and Brooker, 1974).Additionally, Lactobacilli are the predominant culturable flora in the crop and smallintestine, whereas obligate anaerobes dominate the caeca with Coliforms such as E. coli found mainly in the large intestine. The natural microflora also varies withage with a greater diversity of organisms found in older birds. However, it has beenshown that as soon as chicks hatch, even before feeding, several species of bacteriamay be found in the caeca (Barnes, 1982). Barnes et al. (1980) examined manygroups of commercial and experimental chickens and found that the microflora ofday-old chicks consisted of combinations of Streptococci, Coliforms or Clostridia.Interestingly, Lactobacilli were never found in the caeca of day-old chicks whereasvarious other organisms were isolated, some of which are not normally found inadult birds. These included Pseudomonas aeruginosa and Proteus spp. From 3 daysof age, the caecal flora was shown to be remarkably consistent, and Lactobacilliwere present in high numbers together with Coliforms, Streptococci and Clostridia.However, it was reported to be difficult to isolate any non-sporing anaerobes thatonly established after the facultative aerobes such as E. coli had multiplied. It is pos-tulated that the facultative anaerobes contribute to environmental changes in the GItract so that by 2 weeks of age the non-spore-forming anaerobes colonize to becomethe predominant flora (Barnes and Impey, 1970).

2. AVIAN BACTERIAL PATHOGENS

2.1. Escherichia coli

Escherichia coli is the aetiological agent of numerous disease syndromes in poultryincluding colisepticaemia, egg peritonitis, yolk sac infection and coligranuloma(Jordan and Pattison, 1996). Interestingly, numerous serotypes of E. coli includingthose implicated in disease are found as commensals on the mucosal surfacesof clinically healthy birds. Colonization of the alimentary and respiratory tractswith these serotypes occurs in the first few hours of life (fig. 1). The source of theseprimary colonizers is commonly faecal contamination of housing and occasionallytransovarian infection. It has been demonstrated in experimental infection that E. coli isolates of non-avian origin, such as human O157:H7 isolates, are also able tocolonize the avian host (Schoeni and Doyle, 1994; Best, personal communication).Siccardi (1966) identified 74 of 154 (48%) of the known serotypes present in

260 R.M. La Ragione, D.G. Newell and M.J. Woodward

commercial poultry and amongst these were serotypes O1:K1, O2:K1, O3, O6, O8,O15, O18, O35, O71, O74, O78:K80, O87, O95, O103 and O109 which are recog-nized avian pathogenic E. coli types. Of these O1:K1, O2:K1, O8, O35 andO78:K80 are very common (Sojka and Carnaghan, 1961). However, clinical diseaseonly occurs when there are predisposing factors.

It has been shown that several bacterial factors influence colonization andinclude, amongst others, surface antigens that may be elaborated under favourableconditions (La Ragione et al., 2000a). It has also been shown that commensal isolates not associated with disease are often less able to colonize to the same degreeas isolates directly associated with clinical disease. The vast majority of E. coli ofavian origin are capable of elaborating flagella and possibly three distinct morpho-logical types of fimbriae, type 1, curli and P, respectively.

Type 1 fimbriae are the most commonly cited fimbriae on E. coli isolated from poul-try irrespective of source (fig. 2). Additionally type 1 fimbriae have been shown to beexpressed in vivo in the chicken (Pourbakhsh et al., 1997). It has been hypothesizedthat type 1 fimbriae recognize receptors at the epithelial surface and are involvedin intimate attachment to epithelial cells, a prerequisite for subsequent invasion.


Fig. 1. Scanning electron micrograph(magnification × 3500) of chicken trachealtissue infected with E. coli O78:K80 (previouslypublished in La Ragione et al., 2000a).

Fig. 2. Transmission electron micrograph(magnification × 525 000) of E. coli O78:K80showing peritrichous type 1 fimbriae.

Type 1 fimbriae have also been shown to adhere to mucus thus aiding colonization(Klemm, 1985; Mooi and De Graaf, 1985). It has been demonstrated experimentallythat type 1 fimbriae are important in adherence to avian gut and tracheal explanttissue. Additionally, isogenic type 1 fimbriae mutants constructed in an avian E. coliO78:K80, were reduced in their ability to adhere and invade cultured epithelial cellsand explant tissues (La Ragione et al., 2000a). Also, knockout mutants lacking type 1fimbriae have been shown to be attenuated in terms of colonization, invasion andpersistence in the specific pathogen free (SPF) chick compared to the wild-typeparent strain (La Ragione et al., 2000b).

Curli fimbriae are thin, coiled filamentous structures found on the bacterial sur-face of E. coli and Salmonella spp. (Olsen et al., 1989; Collinson et al., 1993) andare commonly associated with avian E. coli clinical disease isolates. Curli mediatebacterial binding to extracellular matrix and serum proteins such as fibronectin,laminin, plasminogen and plasminogen activator protein (Olsen et al., 1989, 1993;Arnqvist et al., 1992; Sjobring et al., 1994; Hammar et al., 1995). Also, curli havebeen cited as important in attachment to avian gut tissue and in pathogenesis inexperimentally infected SPF chicks (La Ragione et al., 2000a,b). However, theirexact role in pathogenesis is unclear but it is possible that curli fimbriae mediateaggregation of curliated bacterial cells and so enhance the number of bacteria colo-nizing epithelial surfaces (Collinson et al., 1993).

P fimbriae have been cited as important in the pathogenesis of urinary tract infec-tions in the mammalian host (Hagberg et al., 1983; Lindberg et al., 1984; Roberts et al., 1989) but their role in colonization in the avian host is unclear. P fimbriaedo not appear to be involved in bacterial adherence to tracheal or pharyngeal cells in vitro (Van den Bosch et al., 1993; Vidotto et al., 1997). However, by immuno-fluorescence P fimbriae have been shown to be expressed in the lungs, air sacsand internal organs of experimentally infected chickens (Pourbakhsh et al., 1997)and this may indicate a role in the later stages of infection as well as in initialcolonization.

Many non-fimbrial adhesins (NFAs) have been described on E. coli that are elab-orated by many diverse serotypes, some of which are found colonizing avian species(Schmidt, 1994). These NFAs are defined as not showing any detectable orderedstructure on the bacterial surface as visualized by electron microscopy. However,they may appear as amorphous masses and appear similar to capsule. The distribu-tion of fimbrial operons among E. coli is assumed to be widespread and, therefore,it is assumed that these are common mechanisms whereby both pathogenic and non-pathogenic E. coli may colonize.

Flagella have been cited as important in colonization for many pathogens in anumber of hosts (Yancey et al., 1978; Montie et al., 1982; Attridge and Rowley,1983; Smyth, 1988; Allen-Vercoe and Woodward, 1999a,b; Dibb-Fuller et al., 1999;Lee et al., 1999). It is thought that flagella enable bacteria to swim through themucus sheet covering the mucosal surfaces and reach the underlining epithelium,


where flagella may act as an anchor, subsequently allowing fimbriae to attachintimately (Smyth, 1988; La Ragione et al., 2000a). E. coli are regarded to bemonophasic, possessing a single flagella antigen type that may be phase on or phaseoff. The regulatory mechanisms controlling expression are complex and linked withenvironmental factors. Interestingly, only one serotype of E. coli, O148 which isnot regarded as an avian pathogenic serotype, has been described as biphasic,namely possessing two antigenic types of flagella and therefore is similar to manySalmonella serovars (Ratiner, 1999). Additionally, motility conferred by the flagella,aids bacteria to move towards chemical stimuli such as food sources and awayfrom adverse environmental conditions (Macnab, 1987a,b). However, aflagellatebacterial avian pathogens, such as S. gallinarum and S. pullorum (see below), havebeen described which lends further support to the concept that colonization ismultifactorial.

2.2. Salmonella

S. enterica serotype Gallinarum biotypes Gallinarum and Pullorum cause disease inmany avian species. Both biotypes are antigenically very similar but lack flagella,however, and cause two distinct systemic diseases – fowl typhoid and pullorum dis-ease (Shivaprasad et al., 1997). Transmission is environmental, largely faecal–oral,and transovarial with colonization starting in the distal ileum and then the caecumalthough colonization of the gastrointestinal tract is considered to be poor by comparison with other Salmonella serotypes (Barrow et al., 1987, 1994). Clinicalsigns are inconsistent but rapid death in the very young and copious diarrhoea isoften noted. Mortality rates may be high and systemic spread and lesions in manyorgans may be noted. Only Gallinarum infection may result in mortality in birdsaged 3 weeks or older. Prince and Garren (1966) described mortality and resistanceto infections by S. gallinarum in certain inbred lines of poultry, a finding that hasbeen confirmed in more recent studies (Bumsted and Barrow, 1993). Egg transmis-sion and the cycle of S. pullorum infection has been well documented by Rettger andPlastridge (1932) who indicated that those birds surviving infection often becomecarriers that consistently contaminate the environment.

Disease caused by non-host-specific Salmonella infections is uncommon and isusually seen in chicks, poults or ducklings under 2 weeks of age and rarely in birdsover 4 weeks of age. Clinical signs are not pathognomic and diagnosis relies on bac-terial isolation. The serotypes of current concern for human health are Salmonellaenterica serotypes Typhimurium and Enteritidis, although others such as Hadar,Heidelberg and Kentucky, amongst others, are associated with human illness mediatedby poultry as the vector. The main site for colonization by Salmonella in the intestinaltract of chickens is the caecum, however, other sites may be colonized including thecrop (Fanelli et al., 1971; Snoeyenbos et al., 1978). Barrow et al. (1988) reported thatthe majority of Salmonella organisms in the caecum were isolated from the lumen


rather than epithelial homogenates. Evidence has been gained to support the hypoth-esis that the advantage of caecal colonization, an organ with a low flow rate of contents, could be to provide an inoculum for fresh chyle when the caecum isrefilled after emptying (Hill and Smith, 1969; Barrow et al., 1988; Craven andWilliams, 1997). However, the number of Salmonellae associated with the epithe-lium in the crop frequently exceeded the numbers in the lumenal contents (Cravenand Williams, 1997; Barrow et al., 1988).

Salmonellae produce a variety of putative virulence determinants encoded by upto seven “pathogenicity islands” that are associated with all aspects of systemicdisease. For the avian species-specific serovar S. gallinarum, Jones et al. (2001)showed that virulence in the chicken requires the Salmonella pathogenicity island(PAI) 2 type III secretion system but not PAI 1. Other virulence determinantsinclude haemagglutinins, adhesins, invasins, fimbriae, exotoxins, endotoxins, typeIII secretion systems and so forth (Duiguid et al., 1966; Formal, 1983; Freter, 1981;Koo et al., 1984; Halula and Stocker; 1987; Galan and Curtiss, 1989; Baumler et al.,2000). Apart from commonly shared fimbrial operons (see below), the virulenceplasmids of S. gallinarum, S. pullorum and S. dublin shared genes that with homol-ogy K88 fimbrial genes and, for S. gallinarum, mutants in these genes resulted inattenuation for all aspects of chicken pathogenesis (Rychlik et al., 1998). In themouse model, S. Typhimurium virulence is dependent on its ability to attach, invadeand multiply in the Peyer’s patches (Carter and Collins, 1974). Interestingly,S. Typhimurium also attaches to cells of the intestinal epithelium (Finlay et al.,1988; Galan and Curtiss, 1989). The role of specific fimbriae (LPF and PEF; seebelow) is important for initial attachment to Peyer’s patches (Baumler et al., 1997,2000). However, loss of flagella and loss of either mannose-sensitive or mannose-resistant haemagglutination structures reduces the organism’s ability to adhere to orinvade epithelial cells in vitro (Curtiss et al., 1993). Interestingly, non-motile or non-haemagglutinating mutants are fully virulent, following oral mouse challengewhereas mutants deficient in both are attenuated (Carsiotis et al., 1984; Lockmanand Curtiss, 1990, 1992; Curtiss et al., 1993). S. pullorum and S. gallinarum are fully virulent although lacking flagella. However, Holt and Chaubal (1997) have demonstrated that these serovars are capable of expression of these organellesin vivo in the chicken, which suggests tight regulation of this function and a prob-able role in colonization.

Salmonella enterica serotype Enteritidis is able to elaborate monophasic flagellaand at least five morphological distinct fimbriae have been observed, although thegenome sequence indicates a genetic potential to express up to 12 discrete fimbriae(Baumler, personal communication). The well characterized S. Enteritidis fimbriaeinclude SEF14 fimbriae that are serovar restricted to most group D Salmonellasonly, the curli orthologue SEF17, the type 1 orthologue SEF21, the long polar fim-briae (LPF) and plasmid encoded fimbriae (PEF). In gut explant adhesion assays itwas observed that flagella were important in adhesion to avian proximal gut tissue,


but fimbriae were not (Allen-Vercoe and Woodward, 1999a; Dibb-Fuller et al.,1999). In these studies it was observed that motility conferred by flagella and thepresence of the flagellum itself, were important in adhesion to avian gut tissue. It may be hypothesized that the flagella either increase bacterial surface area for initial contact or may help in overcoming repulsive forces. Interestingly, Jones et al.(1992) reported that the direction of flagellar rotation affected the ability of S. Typhimurium to invade cultured epithelial cells and, in addition flagella havebeen implicated in assisting survival of S. Typhimurium within murine macrophages(Weinstein et al., 1984). It has been observed that type 1 fimbriae of S. Typhimuriumare important in adhesion to gut epithelial cells (Takeuchi, 1967; Lockman andCurtiss, 1992; Vidotto et al., 1997) and invasion (Ernst et al., 1990). In studies conducted in mice and in 3-week-old chicks, Turner et al. (1998) used a signaturetagged mutagenesis approach to identify factors essential for persistence. LPSO antigen amongst many other factors was found to be an important factor whereasfimbriae were not identified by this approach. Several studies have demonstratedthat lipopolysaccharide (LPS) of S. Typhimurium plays a role in the colonization of mice and chickens (Craven, 1994; Licht et al., 1996; Turner et al., 1998).Interestingly, LPS endotoxin promotes translocation of bacteria from the gastro-intestinal tract, at least in mouse studies (Deitch et al., 1987).

In vivo studies with S. Enteritidis showed that aflagellate mutants were attenu-ated but the role of fimbriae was less clear although they did contribute to persist-ence (Allen-Vercoe et al., 1999; Allen-Vercoe and Woodward, 1999b; Dibb-Fullerand Woodward, 2000). In 5-day-old SPF chicks that have a gastrointestinalmicroflora not dissimilar to that of an adult bird (Coloe et al., 1984), similar resultswere observed. However, it had been reported that a natural microflora significantlyreduces the susceptibility of these animals to colonization by Salmonella spp.(Barrow and Tucker, 1986; Berchieri and Barrow, 1990; Duchet-Suchaux et al.,1995). Collectively, these studies suggest that motile flagella may contribute to gutcolonization whereas the contribution, if any, from fimbriae was too subtle toobserve in the models used. It must be borne in mind that although significant dif-ferences were not seen between different ages of birds, only caecal contents wereexamined and differences may have been observed if epithelial surfaces and wholetissue were tested separately.

2.3. Campylobacter

Particular concern about poultry and poultry products as a source of human food-borne illness has arisen in recent years with the S. Enteritidis epidemic of thelate 1980s and early 1990s, and the increasing awareness of Campylobacter jejuni.Of the food-borne pathogens carried by poultry, Campylobacter, and, in particularC. jejuni, is now recognized as the most significant in terms of human illness andpotential economic impact.


C. jejuni and C. coli colonize the gastrointestinal tract of most animals asympto-matically (Shane, 1991; Hu and Kopecko, 2000) and are primarily associated withdisease in susceptible humans in the industrialized world (Van Vlient and Ketley,2001). All members of the genus Campylobacter are motile, Gram-negativemicroaerophilic, spiral, uniflagellate organisms (Skirrow and Benjamin, 1980) butC. jejuni and its close relative C. coli are thermophilic, growing optimally at 42°C(Genigeorgis, 1986) and appear to have evolved to colonize the avian GI tract.Surveys show that 15−95% of broiler flocks are naturally infected (Blaser, 1982;Newell and Wagenaar, 2000) and subsequent faecal contamination of carcassesresults in up to 75% of retail poultry product contamination. Campylobacter is rec-ognized as the most significant in terms of human illness and potential economicimpact. It is estimated that there are 500 000 cases of Campylobacteriosis perannum in the United Kingdom alone.

Chickens appear to acquire Campylobacters horizontally from their environ-ment. Although oviducts can be colonized (Camarda et al., 2000), hatched chicksare Campylobacter-negative. Flocks do not usually become positive until 2−3 weeksof age (Newell and Wagenaar, 2000). Interestingly, experimental infection can beestablished in chicks by oral dosing from day of hatch onwards (Ruiz-Palacios et al.,1981). Once colonization occurs in one animal, it spreads rapidly throughout theflock, reaching levels of up to 106−107 CFU/g in the caecal contents (Corry andAtabay, 2001). This colonization appears to have no adverse effects on the flockhealth. Thus, in poultry, Campylobacters act as a highly successful GI tract com-mensal. However, in ostrich chicks, C. jejuni is associated with enteritis and evendeath (Newell, 2001). A severe disease “Vibrionic hepatitis” was prevalent in the1950s and 1960s in chickens in North America and Europe, and may have beencaused by C. jejuni (Peckham, 1984).

The reason for the lag phase in colonization is intriguing. It occurs even in organ-ically reared chicks surrounded by an environment heavily contaminated withcampylobacters suggesting that lack of exposure is not the explanation and indicat-ing that the immature avian GI tract is refractory to infection. Newly hatched chickshave circulating and mucosal antibodies directed against Campylobacter surfaceantigens (Cawthraw et al., 1994) but experimentally infected chicks are susceptibleto infection (Cawthraw et al., 1996) suggesting that these antibodies are not protec-tive. Alternative explanations include antimicrobial feed additives and competitivemembers of the native flora. An understanding of the mechanism of the lag phasemay be an important factor in the development of targeted strategies for the controlof this organism.

Little is known about the nature and mechanism of avian GI tract colonizationby Campylobacters. Experimentally, the colonizing organisms are confined prima-rily to the intestinal mucus layer in the crypts of the intestine and caecum (Beeryet al., 1988). There appears little close association of Campylobacters with intestinalepithelial cells but Campylobacters have been recovered from extraintestinal sites


including the liver and spleen. This suggests that adherence (Ziprin et al., 1999) is arare or possibly intermittent event that may lead to invasion across the mucosa.Motility is an essential factor for Campylobacter colonization of the intestinalmucous layer (Newell et al., 1985; Lee et al., 1986; Szymanski et al., 1995). Thismotility is conferred by polar flagella, and combined with their “cork-screw” formallows them to efficiently penetrate the mucous layer. The possession of active flagella (Newell et al., 1985; Guerry et al., 1991) and the ability to undergo chemo-tactic motility (Yao et al., 1994; Penn, 2001) are well established prerequisites forvirulence in mammalian systems. However, in experimental chick models, motilitymay not be so important (Wassenar et al., 1993) presumably because the primarycolonization site, the caecum, has restricted mucus flow and, once organisms havereached this site, colonization can be persistent despite lack of motility.

Flagella also appear to play a role in invasion into host cells (Wassenar et al.,1991; Penn, 2001), possibly reflecting some adhesive capacity to the host intestinalcells (Newell et al., 1985). There are two tandem flagellin genes, flaA and flaB, andmutants unable to express the products of these genes exhibit differences in motil-ity and colonization, as well as invasive potential (Wassenar et al., 1991, 1994a;Nuitjen et al., 1992; Penn, 2001). Several other putative adhesins have been described.Early observations of fimbrial-like structures on C. jejuni (Doig et al., 1996) appear to be artefactual (Gaynor et al., 2001). Experimental studies suggest that the outer-membrane proteins (OMPs) and lipopolysaccharides (LPS) may play roles in colonization (McSweegan and Walker, 1986; McSweegan et al., 1987; Fauchere et al.,1989; Ziprin et al., 1999). There are a number of surface exposed antigens (PEBs)which have been investigated as adhesins. Pei et al. (1998) showed that peb1A mutantshad reduced adherence in in vitro models and significantly reduced colonization in themouse but still colonize chicks. Interestingly, cadF (fibronectin-binding protein)mutants were unable to colonize newly hatched chicks (Ziprin et al., 1999).

2.4. Spirochaetes

Avian intestinal spirochaetosis (AIS) is a poorly understood disease. Although itwas first described by Fantham in 1910, only in the past two decades have detailedpathogenesis studies been undertaken (Dwars et al., 1989; Swayne et al., 1995). AISis defined as a sub-acute to chronic, non-septicaemic intestinal disorder character-ized by variable clinical illness, morbidity and mortality (Swayne and McLaren,1997). AIS has been reported in broilers, laying hens, pheasants, geese and ducks(Swayne and McLaren, 1997; Webb et al., 1997). The spirochaetes that colonize theintestinal tract of birds are a heterogeneous group of bacteria, with few character-ized sufficiently for taxonomic identification. These pathogenic spirochaetes may bean integral part of the autochthonous intestinal flora, or may be opportunistic colo-nizers. Of the better characterized Spirochaetes, Brachyspira (Serpulina) pilosicoli(Trott et al., 1996) colonizes the crypts of the lower intestinal tract of the pig, but


may extend to the small intestinal tract. More needs to be done in avian speciesregarding colonization by these highly motile organisms, but all show polar attach-ment to the mucosa. Trampel et al. (1994) studied natural AIS cases and showed bydark-field and light microscopy that spirochaetes were located in the caeca.Typically, the apical surfaces of caecal enterocytes were covered by a dense layer ofspirochaetes aligned parallel to each other and perpendicular to the mucosal surface.Sueyoshi et al. (1986) orally inoculated Treponema hyodysenteriae into youngbroiler chicks and the organisms were present both on the mucosal surface and inthe deep crypts of the caecum 7 and 14 days later. The numbers colonizing weredose dependent but were readily observed by light microscopy. The caeca of chicksinfected with treponemes were atrophied and the lumen was filled with a whitewatery fluid instead of digested feed. The lesions were primarily confined to thecaecum and comprised desquamation of epithelial cells, oedema, leucocytic infil-tration and haemorrhage of the mucosa. This model was developed further bySueyoshi and Adachi (1990) as a surrogate model for swine dysentery. Trott andHampson (1998) dosed specific pathogen free chicks aged 1 day with strains of fivedifferent species of intestinal spirochaete originally isolated from pigs or humans.The findings in this model were similar to those described above with, by way of anexample, a virulent strain of Serpulina hyodysenteriae (WA 15) colonizing thechicks well to cause retarded growth rate and histological changes, including caecalatrophy, epithelial and goblet cell hyperplasia, and crypt elongation. Sagartz et al.(1992) described necrotizing typhlocolitis in 13 juvenile common rheas (Rheaamericana) from three separate, geographically isolated Ohio flocks, with mortalityranging from 25 to 80%. At post mortem examination, a diphtheritic membrane cov-ered ulcerated caecal mucosa was observed. Caecal sections showed necrosis andgranulomatous-to-suppurative inflammation that extended into the submucosa andoften surrounded large eosinophilic colonies of bacteria. Warthin-Starry stainingshowed these colonies to be composed of entangled spirochaetes that invaded thesubmucosa and frequently were present transmurally. Muniappa and Duhamel(1997) have identified a partially heat stable, subtilisin-like, serine protease in theouter membrane of all spirochaetes. They suggest this factor may be essential for survival in the intestinal environment and may indirectly contribute to intestinaldamage caused by pathogenic spirochaetes during association with the mucosal surface of the host.

2.5. Clostridia

A number of species of Clostridium are normal inhabitants of the intestinal tract of healthy chickens and can be recovered from the intestines of chicks from approximately 1 week of age (Ficken and Wages, 1997). However, many, includingC. botulinum, C. colinum, C. septicum and C. perfringens cause disease in poultry.The main mechanism of virulence is the production of toxins. Yet, other putative


virulence determinants such as the mechanisms of colonization have not been studied in detail.

C. perfringens (mainly type A or type C toxin) is the aetiological agent of aviannecrotic enteritis (NE), which is a sporadic disease of 4-week or older chickens andturkeys. NE was first described in 1930 by Bennetts and later by Parish (1961) and iseconomically very important. The disease is initiated by a predisposing factor such aspoor husbandry practices, intestinal mucosal damage caused by high-fibre litter orconcurrent infection with Coccidia (Al-Sheikhly and Al-Saieg, 1980). Kimura et al.(1976) investigated the bacteriological and histopathological changes in the caeca ofyoung chickens after infection with sporulated oocysts of Eimeria tenella. Lactobacilliand Bifidobacteria showed a remarkable decrease in number on the fifth day afterinfection when destruction of mucosa along with severe haemorrhaging was noticed.Clostridium perfringens proliferated 5 days after infection following the decline ofLactobacilli and Bifidobacteria. Proliferation of Clostridia was so intense that thenumber was almost a million times greater than that of the uninfected chicken. Otherpredominant bacteria, such as Bacteroidaceae, Catenabacteria and Peptostreptococcishowed only moderate and temporal decrease in number during the infection.

Colonization of the mucosa by Clostridia has not been described in detail largelybecause the focus of study has been towards the toxins that mediate the necrosis. Forexample, Al-Sheikhly and Truscott (1977) reproduced necrotic enteritis by intraduode-nal infusion of chicks with Clostridium perfringens type A. Typical lesions of necroticenteritis, characterized by oedema in the lamina propria and desquamation of epithelialcells, were seen as early as 5 h after infusion. Large numbers of Clostridia were seenamong these sloughed cells and clumps of clostridia were obvious among the necrotictissue. Thus it is recognized that C. perfringens can proliferate in the large intestine andcaeca with subsequent migration to the small intestine, or sites of mucosal damagecaused by coccidial infection that leads to further proliferation and production of toxinthat results in tissue damage and necrosis. In severe manifestations necrosis can extendthe entire width of the intestinal mucosa (Murakami et al., 1989) although acute catarrhwithout necrosis is more common in the lower intestine. Histological examination oftissues reveals Gram-positive bacilli that are often detected in the lamina propria andfrequently attached to cellular debris (Songer, 1997).

In a hamster model, Borriello and Barclay (1985) demonstrated that a non-toxigenicstrain of C. difficile adhered to gut mucosa and afforded some protection against toxi-genic trains. It is proposed that the protection afforded by the non-toxigenic strains maybe due to competition for ecological niches. This approach could be investigated for NEin chickens.

2.6. Mycoplasma

Mycoplasmas are highly fastidious microorganisms that differ from other eubacte-ria in that they are significantly smaller and lack a cell wall. This characteristic


accounts for their “fried egg” colonial morphology and total insensitivity to antibi-otics that have cell wall biosynthesis as a target (Kleven, 1998). Mycoplasmas arenormal inhabitants of the mucosal surfaces of poultry and gain entry into the hostthrough the respiratory tract or via the infected embryo. A number of serotypes ofMycoplasma have been isolated from healthy poultry. However, M. gallisepticumand M. synoviae cause disease in chickens, M. iowae in chickens and turkeys andM. meleagridis solely in turkeys. Mycoplasmas cause a variety of clinical syndromesin poultry including keratoconjunctivitis, chronic respiratory disease, embryonicmortality, skeletal abnormalities, exudative synovitis, tendovaginitis, bursitis, salp-ingitis and airsacculitis (Yoder, 1991). The severity and sometimes the appearanceof disease may be influenced by concomitant infection with other pathogens ordebilitating factors (Cullen et al., 1977).

As with many bacterial pathogens, a characteristic that permits Mycoplasmato effectively establish infection and proliferate within the host is associated withits capability of attachment to tissues, mainly epithelial linings of the oviductsand respiratory tract (Fiorentin et al., 1998). Adherence to the surface receptorsof host cells by Mycoplasmas is generally mediated by adhesins expressed onthe cell surface (Kahane et al., 1985; Razin and Jacobs, 1992). At least three cytad-herence proteins are involved in M. gallisepticum adherence and colonization(Fiorentin et al., 1998). Keeler et al. (1996) cloned and sequenced the 3366-nucleotide open reading frame, mgc1, that encodes a 150-kDa cytadhesin-like pro-tein. The 1122-amino-acid protein, MGC1, has characteristics of a class I membraneprotein and has homology with the MgPa cytadhesin of M. genitalium and the P1cytadhesin of M. pneumoniae. Hnatow et al. (1998) identified a second cytadhesin-like protein, MGC2, and Boguslavsky et al. (2000) identified a third, PVP-A. Theencoding genes have been sequenced and all share homology one with anotherand other related Mycoplasma adhesins. Variable expression, possibly phase-onphase-off, for the cytadhesins is observed as is variability in their respective sizes.Functional studies with attachment inhibition assays have confirmed their role(Hnatow et al., 1998; Boguslavsky et al., 2000). These data suggest that there is afamily of cytadhesin genes conserved among pathogenic Mycoplasmas infectingwidely divergent hosts. Additionally, some Mycoplasmas possess surface organellesthat have been implicated as potential colonization factors. The first to be describedwas the “bleb” of M. gallisepticum, a small organelle at one or both poles of the cell(Maniloff et al., 1965).

2.7. Other pathogens

Avian tuberculosis caused by Mycobacterium avium was a common disease ofdomestic fowl that is largely controlled by improved husbandry practices. Birddensity, poor hygiene and deep litter are regarded as influencing factors. The hostrange is limited to domestic fowl in the main although some caged birds may be


affected also. Pigs and humans may also be affected by this pathogen. Initial infec-tion is considered to be by entry of the organisms into the premises by unsanitarypractices. Many studies have focused upon Johne’s disease in ruminants and littlehas been done on poultry. In the mouse model, M. avium was found to invade theintestinal mucosa by interacting primarily with enterocytes and not with M cells(Sangari et al., 2001). Thus, it may be assumed that colonization of the gastroin-testinal tract is largely an invasive process. Once within host cells, tubercles developin the intestine and large numbers of bacteria may be excreted in the droppings tospread contamination rapidly. The bacterium is hardy and may survive in the envi-ronment for prolonged periods. Tubercles may develop in the reproductive tract butegg transmission is not considered important (Thoen and Karlson, 1991).

Yersinia pseudotuberculosis is found commonly in domestic poultry, other avianspecies and rodents. Again, poor hygiene and overcrowding are associated with thespread of this agent that colonizes via the intestine or through breaks in the skin(Rhoades and Rimler, 1991a). Many studies on other Yersinia spp. have identifiedvirulence factors including adhesins and secretory proteins essential for invasion.Mecsas et al. (2001) used signature-tagged mutagenesis to isolate mutants of Y. pseudotuberculosis that failed to survive in the caecum of mice after orogastricinoculation; Yersinia pseudotuberculosis localizes to the distal ileum, caecum, andproximal colon in this model. Several mutations were in operons encoding compo-nents of the type III secretion system, including Yop proteins and O-antigen biosyn-thesis; these mutants were also unable to invade epithelial cells as efficiently aswild-type Y. pseudotuberculosis. This indicates that one or more Yops and LPS maybe necessary for colonization of the gastrointestinal tract.

Avian pasteurellosis is caused by a number of different bacteria that include Y. pseudotuberculosis, Moraxella spp. and Pasteurella multocida. P. multocida isthe highly infectious cause of fowl cholera of which several pathogenic and highlyinvasive serotypes have been defined. Access is gained to the host via broken skinand nasal, conjunctival and intestinal routes. The organisms of this group are all verysusceptible to air drying and airborne routes are not considered important in theirspread (Rhoades and Rimler, 1991b).

Of the Gram-positive cocci, Staphylococcus aureus readily colonizes poultryskin and the nasal passages and is considered ubiquitous. It is well established thatS. aureus possesses various bacterial adhesins that bind skin and various cell typesand that adhesion is mediated by fibronectin-binding protein (Fnbp), fibrinogen-binding protein (Clf) and collagen-binding protein (Cna). Recent studies (Cho et al.,2001; Massey et al., 2001) have proven that fibronectin and fibrinogen, but not col-lagen, play major roles in the enhanced adherence of S. aureus to skin and cells. It must be assumed they play similar roles in colonization of avian species. However,the co-aggulase positive strains are opportunistic pathogens that are associated withmany clinical outcomes after translocation (e.g. having gained access to the bird viabroken skin). Arthritis, synovitis, spondylitis, other general abscesses, dermatitis,


septicaemia and yolk infection have been recorded frequently. Enterococcus faecalis is a common inhabitant of the intestine, and may invade causing septi-caemia and endocarditis, as are S. aureus, Streptococcus zooepidemicus andErysipelothrix insidosa (Skeeles, 1991; Wages, 1991). With regard to Enterococci,adherence to intestinal, urinary tract epithelial cells and heart cells is by means ofadhesins expressed on the bacterial surface (Johnson, 1994). The expression of theseis growth phase dependent and is enhanced if the organisms produce a proteina-ceous surface material that aggregates bacteria and that also facilitates plasmidtransfer (Johnson, 1994). Enterococci are capable of inducing platelet aggregationand tissue factor-dependent fibrin production, which may be relevant to the patho-genesis of enterococcal endocarditis (Johnson, 1994).

Erysipelothrix rhusiopathiae is a common pathogen of turkeys, but may alsoaffect other avian species. Faecal shedding is noted, as is frequent carrier status. Thesource of infection is largely environmental and entry to the bird is via damagedmucosal layers. The early work of Sadler and Corstvet (1965) demonstrated therequirement for direct access to the host because oral, nasal and conjunctival inoc-ulation with known pathogenic strains did not cause morbidity or mortality. Theclosely related bacterial species Listeria monocytogenes causes sporadic outbreakswith up to 40% mortality and which was first reported as a serious problem in thepoultry industry in the mid 1930s. The source was believed to be silage, especiallyfrom maize, rye and grass, and it also affected sheep and cattle, both considered riskfactors in transmission. The infection is septicaemic and the bacterium is shed innasal exudate and faeces. With regard to mucosal colonization, little is known, espe-cially in avian species. Bubert et al. (1992) showed that the p60 extracellular proteinof Listeria monocytogenes is required for adherence to and invasion of 3T6 mousefibroblasts but not for adherence to human epithelial Caco-2 cells. However, Cowartet al. (1990) proposed that attachment of Listeria to eukaryotic cells occurs by inter-action of the microbial alpha-D-galactose with that of a eukaryotic galactose receptor.The role of these factors in the avian species is untested. Of concern is contaminationby this organism of the skin of processed carcasses (Gitter, 1976), especially as theorganism may multiply at typical domestic storage temperatures (Mead et al., 1990).

3. THE NATIVE COMMENSAL FLORA

The native commensal flora of avian species is highly complex with at least 200 dis-tinct species either identified after isolation and classic bacteriology or partiallyidentified by genetic means. Direct analysis of the microbiota of GI tract contentsby 16S rRNA analysis (Suau et al., 1999) as well as by DNA analysis techniques(Apajalahti et al., 1998) has been used to determine the species and relative num-bers of types of bacteria in the GI tract. The early work of Barnes and Impey (1970)demonstrated various distinct predominant classes of bacteria in the caeca of chick-ens and turkeys. Animals of 5 weeks of age with what may be regarded as a mature


flora possessed Gram-negative non-spore-forming anaerobes, largely Bacteroidesspp., and Gram-positive non-spore-forming anaerobes with Bifidobacteria compris-ing approximately 80% of the cultivable bacterial population.

Many of the species identified in the intestine also inhabit the oral cavity in mam-mals and the adherence of these has been studied intensively. Leung et al. (1989)demonstrated that Bacteroides intermedius possessed four distinct fimbriae andwhereas Bacteroides forsythus possesses an adherence factor, the BSP-A surfaceantigen, that binds oral tissues and triggers host immune responses (Sharma et al.,1998). Similar surface structures play a role in intestinal colonization. Brook andMyhal (1991) demonstrated that piliated and capsulated strains of Bacteroidesfragilis adhered at least five times greater than the adherence of their non-piliatedand non-capsulated or capsulated only counterparts to the gastrointestinal mucosa.However, the efficiency of adherence to oral epithelial cells and INT-407 cells variedbetween 10 and 40 bacteria per host depending upon the source and strain of theBacteroides (Guzman et al., 1997). Also, Baba et al. (1992) showed that the balanceof the normal flora is readily perturbed by infections. In conventional chickens,fewer Bacteroides vulgatus and Bifidobacterium thermophilum adhered to the E. tenella-infected caeca than to the uninfected caeca.

The remaining flora comprise Peptostreptococcus, Prevotella, Ruminococcus,Propionibacterium, Fusobacterium, Lactobacillus and Eubacterium, amongst manyother genera. Particular interest has focused on certain species because of their asso-ciation with reducing colonization by avian pathogens and these will be discussedin detail below.

3.1. Lactic acid bacteria

Large populations of Lactobacilli inhabit the digestive tracts of fowl. Interestingly,some gastrointestinal strains of Lactobacilli have the ability to adhere to and colo-nize the surface of stratified squamous epithelium in the oesophagus, crop and stom-ach whereas other Lactobacillus strains appear to be inhabitants of thegastrointestinal lumen only (Fuller and Turvey, 1971; Fuller and Brooker, 1974).Lactobacilli, irrespective of origin and whether shed from epithelial surfaces or mul-tiplying in ingested food, permeate all regions of the digestive tract in a number ofanimals (Fuller et al., 1978; Tannock, 1987; Gusils et al., 1999a). Although it hasbeen suggested that several factors are associated with host specificity, one signifi-cant speculation is that Lactobacilli adhere to epithelial surfaces by interactionsoccurring between specific molecules on bacterial cells and on the gastrointestinalsurface of the host. Fuller (1975) reported the possible involvement of a carbohy-drate in the adhesion process. Henriksson et al. (1991) reported that Lactobacillusstrains adhere to porcine stomach epithelium through proteinaceous componentslocated on the bacterial surface. Additionally, Savage (1969) demonstrated thatLactobacilli colonized the surface of non-secreting keratinized epithelial cells in the


stomach of rats and mice, but not the surface of the secreting stomach epithelium.Interestingly, Tannock (1990) mentioned that carbohydrate-specific molecules(lectins) contribute to the adherence of Lactobacilli to epithelial surfaces and Gusils et al. (1999b) found lectin-like structures in the external layers of L. animalis,L. fermentum and L. cellobiosus. These molecules, present on the cell surfaces,would favour adhesion to epithelial cells. Also, immuno-stimulating activity of cellwalls from L. casei CRL 431 can be induced by an adhesion phenomenon in whichlectin-like structures are involved (Morata de Ambrosini et al., 1988a,b). Feedingdiets containing Lactobacillus spp. has been shown to enhance the production ofanti-salmonella IgM antibodies and the function of T cells in newly hatched chicksand poults (Dunham et al., 1993).

3.2. Bifidobacteria

Bifidobacteria are Gram-positive, anaerobic, non-motile, non-spore-forming, fermentative rods which are normal inhabitants of the mammalian and avian gastrointestinal tract (Van der Werf and Venema, 2001). It has been reported thatBifidobacteria have a positive effect on the hosts’ health status and that humans withhigher numbers of Bifidobacteria are less at risk from diarrhoea (Albert et al., 1978;Saavedra et al., 1994; Heinig and Dewey, 1996). A total of 32 species ofBifidobacteria have been identified in the genus Bifidobacterium in the present clas-sification. Bifidobacteria are host specific and age specific. For example, in humans,B. infantis and B. breve are predominant in infants whereas B. adolescentis andB. longum are predominant in adults (Mitsuoka, 1982). It has been reported thatBifidobacteria bind to host mucus glycoproteins and these vary from species tospecies (Von Nicolai et al., 1981; He et al., 2001). Bifidobacteria adhere to mucusand, in human studies, the number of Bifidobacteria in faeces and intestinal contentsreduces with increasing age of the host. Some strains also showed decreased adhe-sion to mucus isolated from infants compared with that from adults (Ouwehand et al., 1999). It has been reported that Bifidobacteria competitively exclude otherbacteria such as Proteus vulgarius and Klebsiella pneumoniae (Timoshko et al.,1981). However, many factors have been shown to reduce the number ofBifidobacteria in the GI tract of chickens including thermal kestoses (Patterson et al., 1997). The mechanisms by which Bifidobacteria competitively exclude otherbacteria are beneficial to health and are considered to be: 1) inhibiting the growth ofpathogenic bacteria by producing bacteriocins; 2) lowering the gut pH by acetateand lactate production; 3) colonization of the gastrointestinal tract; 4) production ofvitamins; and 5) stimulating the immune system (Deguchi et al., 1985; Fuller, 1989;Mitsuoka, 1990; Ballongue, 1993). Lievin et al. (2000) described two Bifidobacteriastrains that expressed antagonistic activity against S. Typhimurium SL1344 in vitro.The Bifidobacteria inhibited Salmonella cell entry and killed intracellularSalmonella in Caco-2 cells and the effector was identified as a low molecular weight


lipophilic molecule. Additionally, in the axenic C3/He/Oujco mouse model, thesetwo Bifidobacteria strains colonized the intestinal tract and protected mice against alethal S. typhimurium oral dose.

In poultry, Bifidobacteria have been shown to be obligate inhabitants of the cropand intestine of adult birds. In addition Bifidobacteria are found in high numbers inthe intestines. Interestingly, it has been shown that those inhabiting the intestine andcrop are specific to those tissues, respectively (Petr and Rada, 2001).

4. COMPETITIVE EXCLUSION

As early as the late nineteenth century the antagonistic interaction between somebacterial strains was observed (Pasteur and Joubert, 1877; Metchnikoff, 1907).However, the therapeutic use of bacterial cultures in human and veterinary medicineis more recent for two main reasons. First, control of zoonotic pathogens by anti-biotics is associated with increased antimicrobial resistance amongst many commonbacterial pathogens. Secondly, an increased awareness that antibiotic treatment perturbs a native and therefore “protective” flora that may predispose to later infec-tions. Thus, the use of bacteria for prophylaxis has become more commonplace(Bengmark, 2000).

It has been recognized that the normal intestinal microflora of animals can bedivided into three categories. 1) The autochthonous microbiota – microbes that arepresent at high population levels, usually throughout the life of an animal and foundin all members of a host population (e.g., Lactobacilli). 2) The normal microbiota –populations of microbes frequently present in the digestive tract, but of variablepopulation size and absent from some populations (e.g., Escherichia coli). 3) Truepathogens (e.g., Salmonella enterica serotypes) – accidentally acquired and capableof persisting in tissues and causing disease (Dubos et al., 1965).

The phenomenon of controlling or modifying the resident microflora is knownas “competitive exclusion” or the Nurmi concept (Pivnick and Nurmi, 1982).Competitive exclusion is used either to protect young chicks by early establishmentof adult microflora or to compensate for the side-effects of antibiotic treatment inolder birds. However, little is known about the bacteria responsible for the barriereffect or the mechanisms involved (Impey et al., 1982; Nuotio et al., 1992; Nuotioand Mead, 1993).

Protection against colonization with bacterial pathogens depends upon oraladministration of viable non-pathogenic bacteria, especially anaerobes (Schneitzand Mead, 2000). Indeed, Impey et al. (1982) described a 48-strain mixture ofspecies that comprised part of the mature flora of chickens that competitivelyexcluded Salmonella. Notable in that mixed population were 11 Lactobacillus spp.,10 Clostridium spp., eight commensal E. coli, three Bacteroides spp., threeStreptococcus spp., two Eubacteria spp., two Bifidobacteria spp., various undefinedanaerobes and Bacillus coagulans.


With regard to the colonization of the chick gut by members of theEnterobacteriaceae, undefined complex cultures were more effective than definedtreatments (Stavric et al., 1991). Also, the most effective exclusion agents werederived directly from healthy adult chicken intestines (Cameron and Carter, 1992;Spencer et al., 1998).

Competitive exclusion (CE) cultures may offer alternatives to antimicrobialagents for disease prophylaxis in poultry (McReynolds et al., 2000). Following thestudies of Nurmi and Rantala (1973), it is widely recognized that the autochthonousgut microbiota provides an effective barrier in poultry against colonization by humanfood-borne pathogens such as Salmonella enterica serotypes Enteritidis andTyphimurium.

It has been reported that chicks are immunologically naive (Jeurissen et al.,1989) and prone to rapid and persistent colonization by both commensal and path-ogenic bacteria in the first 3 to 4 weeks of life (Barrow et al., 1988). Recent studiesby Holt et al. (1999) indicated that the avian immune system may be compromisedby vaccination at 1 day old and that an optimum time for delivery of vaccines wasat about 4 weeks of age. Up to that age, birds remain vulnerable to infectionalthough maternally derived immunity may reduce colonization by pathogens, butthis is unlikely to be fully protective. Therefore, if vaccines alone cannot militateagainst bacterial pathogens, the development of defined, reliable competitive exclu-sion agents is essential. However, it is interesting to note that Weinack et al. (1984)showed that the immune system played no role in the control of Salmonella in CEtreated birds and van der Wielen et al. (2000) showed that volatile fatty acid concentrations were important in controlling microbial populations in the gut.Additionally, it has been suggested that Bifidobacteria may increase antibody syn-thesis through its mitogenic influence on B cells. Bifidobacteria induce spleen B cellsto be reactive to transforming growth factor beta 1 and interleukin-5, so resulting inincreased surface IgA expression (Ko et al., 1999). Bifidobacteria have also beenshown to have a protective effect against clostridia in quails (Butel et al., 1998).

Many of the commercially available agents are ill-defined mixtures of adult birdgut content. However, recent studies have shown that defined single organism agentscan offer good protection to day-old chicks subsequently challenged with patho-genic organisms. In one study, birds were dosed at 1 day old with Bacillus subtilisspores and subsequently challenged with avian pathogenic E. coli O78:K80 24 hlater. In these studies, a significant reduction in E. coli O78:K80 colonization andinvasion was seen. Additionally, compared to a control group, the level of faecalshedding of the pathogenic E. coli was reduced over a 35-day period (Stavric, 1992;La Ragione et al., 2001).

The criteria for competitive exclusion agents include bile and acid stability, adhe-sion to the intestinal mucosa, temporary colonization of the gastrointestinal tract,production of antimicrobial components and safety in medical and veterinary use(Wesney and Tannock, 1979; Saxellin et al., 1995). The most often used genera are


Lactobacilli and Bifidobacteria (Lee et al., 1999). It has also been suggested that probiotic supplementation is of greatest benefit when birds are exposed to stressfulconditions (Jin et al., 1997). However, for organisms such as C. jejuni, occupation ofthe same apparently unique niche by the competitive agent appears to be necessarywhich is probably why competitive exclusion is not consistently successful. It hasbeen suggested that exclusion may only be achieved using other Campylobacterstrains (Wassenar et al., 1994b). If reduction of pathogenic Campylobacters in thefood chain is the aim, then non-pathogenic, genetically stable strains will be required.

5. REMOVAL OF ANTIBIOTICS FROM FOOD ANIMALS

Antibiotic growth promoters were introduced because of their association with thebenefits of decreased mortality, increased feed conversion efficiency and improvedwelfare through improvements in litter quality. The risks associated with their useappeared low as all antibiotics used as growth promoters are absorbed minimallyfrom the gut and do not have systemic action and, therefore, were considered topresent little or no risk of residues in meat. Recently, EU Agricultural Ministersvoted in favour of a ban on the use of four antibiotic growth promoters (zinc baci-tracin, spiramycin, tylosin phosphate and virginiamycin) and these products werebanned in July 1999 (EC Council Regulation, 1999). As a consequence, only twoantibiotic growth promoters are currently commercially available: avilamycin andbambermycin. However, they are closely related to antibiotics for human use andtheir use is very limited. Additionally, studies by Elwinger et al. (1992) showed thatcommercially available competitive exclusion agents are more effective than anti-microbial growth promoters in controlling necrotic enteritis, at least.

Antibacterial substances have been used widely as growth promoters in animalhusbandry (Witte et al., 1999). In recent years, there has been growing concern overthe use of growth promoters in animal feed and the possible acquisition of resistanceby normal resident flora which may have a zoonotic potential. Different means ofinteraction between microecological systems in different animal hosts and the envi-ronment may occur during the transfer of resistant bacteria and their resistancegenes. Spread of resistance takes place in different ways with respect to clonalspread of resistance strains by the spread of wide host range plasmids and trans-locatable elements. Commensals in ecosystems have a special significance anda pronounced capacity for acquisition and transfer of resistance genes as withEnterococcus faecium and E. coli in the normal gut flora (Witte, 2000).

Certain antibiotics are used at subtherapeutic doses in poultry feed for growthpromotion and also at higher doses to treat clinical disease (Gustafson and Bowen,1997). The removal of antibiotics from animal feed has led to a sharp increasein infections caused by opportunistic pathogens. For example, the incidenceof necrotic enteritis which is caused by C. perfringens has been reported to haveincreased sharply since the withdrawal of growth promoters. Of particular concern


is the increase in disease in poultry by potentially zoonotic pathogens such asSalmonella enterica serotypes, E. coli and C. perfringens. Many of these diseases areon the increase as primary predisposing infections are not susceptible to treatment.For example, the incidence of avian colibacillosis is on the increase as primaryMycoplasma infections leave birds predisposed to E. coli infections.


There is much information covering bacterial colonization in poultry although,unlike in other species, such as pigs and sheep where specific host receptors for bacterial colonization have been identified, no poultry-specific receptors have beenidentified. Recently, some of the mechanisms of colonization such as flagella, fimbriae and LPS of well studied bacterial pathogens (e.g., S. Enteritidis and E. coli)have been elucidated. For other organisms such as Lactobacilli, Bifidobacteria,Clostridium perfringens, Campylobacter and Spirochaetes, only limited informationexists and for these organisms the exact mechanisms of colonization and persistenceneed to be elucidated. This is especially important as there is a desire to enhance thestability of the native “protective” flora and eliminate zoonotic pathogens such asCampylobacter jejuni. To this end, a greater understanding of the host–pathogenrelationship needs to be gained possibly through insight into the relationshipbetween the in vivo environment and in vivo bacterial responses. Apart from thehost–bacterium interactions, there is a wealth of data on the efficacy of competitiveexclusion agents reducing colonization by bacterial pathogens. However, there isscant information on how these agents mediate their effect. Within this complex pic-ture is bacterium–bacterium interactions about which little is known. It may beassumed that bacteria do “talk” to each other as evidenced by the phenomenon ofquorum sensing. Is it fanciful to consider positive cross talk that enhances the “pro-tective” flora and negative cross talk that eliminates the pathogens? Identifyingthese putative messages, which may be small metabolites such as the homoserinelactones, will be important in modulating the gut flora. Other mechanisms for mod-ifying flora that could be developed include dietary management, addition of spe-cific dietary supplements such as oligomannans, enhancing the secretory immuneresponses by mucosal vaccines, passive immunization possibly by plant-derivedantibodies in feed, use of bacteriophage for selective lysis, antibacterial toxins andnovel antibiotics. The list of possibilities is long but the knowledge base limited.There is much to do!

REFERENCES

Al-Sheikhly, F., Truscott, R.B., 1977. The pathology of necrotic enteritis of chickens following infusionof broth cultures of Clostridium perfringens into the duodenum. Avian Dis. 21, 230–240.

Al-Sheikhly, F., Al-Saieg, A., 1980. Role of Coccidia in the occurrence of necrotic enteritis of chickens.Avian Dis. 24, 324–333.


Albert, M.J., Bhat, P., Rajan, D., Maiya, P.P., Pereira, S.M., Baker, S.J., 1978. Faecal flora of South Indianinfants and young children in healthy and with acute gastroenteritis. J. Med. Microbiol. 11, 137–143.

Allen-Vercoe, E., Woodward, M.J., 1999a. The role of flagella, but not fimbriae, in the adherence ofSalmonella enterica serotype Enteritidis to chick gut explant. J. Med. Microbiol. 48, 771–780.

Allen-Vercoe, E., Woodward, M.J., 1999b. Colonisation of the chicken caecum by afimbriate and afla-gellate derivatives of Salmonella enterica serotype Enteritidis. Vet. Microbiol. 69, 265–275.

Allen-Vercoe, E., Sayer A.R., Woodward, M.J., 1999. The virulence of Salmonella enterica serotypeEnteritidis afimbriate and aflagellate mutants in a day old chick model. Epidem. Infect. 122, 395–402.

Allori, C., Aguero, G., De Ruiz Holgado, A.P., De Nader, O.M., Perdigon, G., 2000. Gut mucosa mor-phology and microflora changes in malnourished mice after re-nutrition with milk and administrationof Lactobacillus casei. J. Food Protect. 63, 83–90.

Apajalahti, J.H., Sarkilahti, L.K., Maki, B.R., Heikkinen, J.P., Nurminen, P.I., Holben, W.E., 1998.Effective recovery of bacterial DNA and percent-guanine-plus cytosine-based analysis of communitystructure in the gastrointestinal tract of broiler chickens. Appl. Environ. Microbiol. 64, 4084–4088.

Arnqvist, A., Olsen, A., Pfeifer, J., Russel, D.G., Normark, S., 1992. The Crl protein activates crypticgenes for curli formation and fibronectin binding E. coli HB101. Mol. Microbiol. 6, 2443–2452.

Attridge, S.R., Rowley, D., 1983. The role of the flagellum in the adherence of Vibrio cholerae. J. Infect.Dis. 147, 864–872.

Baba, E., Wakeshima, H., Fukui, K., Fukata, T., Arakawa, A., 1992. Adhesion of bacteria to the cecalmucosal surface of conventional and germ-free chickens infected with Eimeria tenella. Amer. J. Vet.Res. 53, 194–197.

Ballongue, J., 1993. Bifidobacteria and probiotic action. In: Salminen, S., Von Wright, A. (Eds.), LacticAcid Bacteria. Dekker, New York, pp. 357–428.

Barnes, E.M., 1982. Factors influencing the avian caecal flora. Eur. J. Chemother. Antibiot. 2, 57–62.Barnes, E.M., Impey, C.S., 1970. The isolation and properties of the predominant anaerobic bacteria in

the caeca of chickens and turkeys. Brit. Poultry Sci. 11 (4), 467–481.Barnes, E.M., Mead, G.C., Barnum, D.A., Harry, E.G., 1972. The intestinal flora of the chicken in the period

2-6 weeks of age with particular reference to the anaerobic bacteria. Brit. Poultry Sci. 13, 311–326.Barnes, E.M., Impey, C.S., Cooper, D.M., 1980. Manipulation of the crop and intestinal flora in the newly

hatched chick. Amer. J. Clin. Nutr. 33, 2426–2433.Barrow, P.A., Tucker, J.F., 1986. Inhibition of colonisation of the chicken caecum with Salmonella

typhimurium by pre-treatment with strains of Escherichia coli. J. Hyg. 96, 161–169.Barrow, P.A., Simpson, J.M., Lovell, M.A., Binns, M.M., 1987. Contribution of Salmonella gallinarum

large plasmid towards virulence in fowl typhoid. Infect. Immun. 55, 388–392.Barrow, P.A., Simpson, J.M., Lovell, M.A., 1988. Intestinal colonisation in the chicken by food poison-

ing Salmonella serotypes; microbial characteristics associated with faecal excretion. Avian Pathol. 17,571–588.

Barrow, P.A., Higgins, M.B., Lovell, M.A., 1994. Host specificity of Salmonella infection in chickensand mice is expressed in vivo primarily at the level of the reticuloendothelial system. Infect. Immun.62, 4602–4610.

Baumler, A.J., Tsolis, R.M., Heffron, F., 1997. Fimbrial adhesins of Salmonella typhimurium. Role inbacterial interactions with epithelial cells. Adv. Exp. Med. Biol. 412, 149–158.

Baumler, A.J., Tsolis, R.M., Heffron, F., 2000. Virulence mechanisms of Salmonella and their geneticbasis. In: Wray, C., Wray, A. (Eds.), Salmonella in Domestic Animals. CABI, Wallingford, Oxon,Chapter 4, pp. 57–87.

Beery, J.T., Hugdahl, M.B., Doyle, M.P., 1988. Colonisation of gastrointestinal tracts of chicks byCampylobacter jejuni. Appl. Environ. Microbiol. 54, 2365–2370.

Bengmark, S., 2000. Probiotics and phage. In: Andrew, P.W., Oyston, P., Smith, G.L., Stewart-Tull, E.E.(Eds.), Fighting Infection in the 21st Century. Blackwell Science, Oxford.

Bennetts, H.W., 1930. Bacillus welchii and bowel lesions. Austral. Vet. J. 4, 153–154.Berchieri, A., Barrow, P.A., 1990. Further studies on the inhibition of colonisation of the chicken ali-

mentary tract with Salmonella typhimurium by pre-colonisation with an avirulent mutant. Epidem.Infect. 104, 427–441.

Berg, R.D., 1981. Promotion of the translocation of enteric bacteria from the gastrointestinal tract of miceby oral treatment with penicillin, clindamycin, or metronidazole. Infect. Immun. 33, 854–861.


Berg, R.D., 1990. Bacterial translocation from the gastrointestinal tract. Comp. Ther. 3, 149–154.Berg, R.D., 1992. Translocation of enteric bacteria in health and disease. Curr. Stud. Hematol. Blood

Transfusion 59, 44–65.Berg, RD., 1995. Inhibition of bacterial translocation from the gastrointestinal tract to the mesenteric

lymph nodes in specific pathogen-free mice but not gnotobiotic mice by non-specific macrophageactivation. Adv. Exp. Med. Biol. 371A, 447–452.

Berg, R.D., 1996. The indigenous gastrointestinal microflora. Trends Microbiol. 4, 430–435.Berg, R.D., 1999. Bacterial translocation from the gastrointestinal tract. Adv. Exp. Med. Biol. 473,

11–30.Bertschinger, H.V., Moon, H.W., Whipp, S.C., 1972. Association of Escherichia coli with the small intes-

tinal epithelium. Comparison of enteropathogenic and non-enteropathogenic porcine strains in pigs.Infect. Immun. 5, 595–605.

Blaser, M.J., 1982. Campylobacter jejuni and food. Food Technol. 36, 89–92.Boguslavsky, S., Menaker, D., Lysnyansky, I., Liu, T., Levisohn, S., Rosengarten, R., Garcia, M., Yogev, D.,

2000. Molecular characterization of the Mycoplasma gallisepticum pvpA gene which encodes aputative variable cytadhesin protein. Infect. Immun. 68, 3956–3964.

Borriello, S.P., Barclay, F.E., 1985. Protection of hamsters against Clostridium difficile ileocaecitis byprior colonization with non-pathogenic strains. J. Med. Microbiol. 19, 339–350.

Brook, I., Myhal, M.L., 1991. Adherence of Bacteroides fragilis group species. Infect. Immun. 59,742–744.

Bubert, A., Kuhn, M., Goebel, W., Kohler, S., 1992. Structural and functional properties of the p60 pro-teins from different Listeria species. J. Bacteriol. 174, 8166–8171.

Bumsted, N., Barrow, P., 1993. Resistance to Salmonella gallinarum, S. pullorum, and S. enteritidis ininbred lines of chickens. Avian Dis. 37, 189–193.

Butel, M.J., Roland, N., Hibert, A., Popot, F., Favre, A., Tessedre, A.C., Bensaada, M., Rimbault, A.,Szylit, O., 1998. Clostridial pathogenicity in experimental necrotising enterocolitis in gnotobioticquails and protective role of Bifidobacteria. J. Med. Microbiol. 47, 391–399.

Camarda, A., Newell, D.G., Nasti, R., Di Modugnoa, G., 2000. Genotyping Campylobacter jejuni strainsisolated from the gut and oviduct of laying hens. Avian Dis. 44, 907–912.

Cameron, D.M., Carter, J.N., 1992. Evaluation of the efficacy of BROILACT in preventing infection ofbroiler chickens with S. enteritidis PT4. Int. J. Food Microbiol. 15, 319–326.

Carsiotis, M., Weinstein, D.L., Karch, H., Holder, I.A., O’Brien, A.D., 1984. Flagella of Salmonellatyphimurium are a virulence factor in infected C57BL/6J mice. Infect Immun. 46, 814–818.

Carter, P.B., Collins, F.M., 1974. The route of enteric infection in normal mice. J. Exp. Med. 139,1189–1203.

Cawthraw, S., Ayling, R., Newell, D.G., 1994. The isotype, specificity and kinetics of systemic andmucosal antibodies to Campylobacter jejuni during experimental oral infections of chickens. AvianDis. 38, 341–349.

Cawthraw, S.A., Wassenaar, T.M., Ayling, R., Newell, D.G., 1996. Increased colonisation potential ofCampylobacter jejuni strain 81116 after passage through chickens and its implication on the rate oftransmission within flocks. Epidemiol. Infect. 117, 213–215.

Cho, S.H., Strickland, I., Boguniewicz, M., Leung, D.Y., 2001. Fibronectin and fibrinogen contribute tothe enhanced binding of Staphylococcus aureus to atopic skin. J. Allergy Clin. Immunol. 108,269–274.

Collinson, S.K., Doig, P.C., Doran, J.L., Clouthier, S., Trust, T.J., Kay, W.W., 1993. Thin aggregative fimbriae mediate binding of Salmonella enteritidis to fibronectin. J. Bacteriol. 175, 12–18.

Coloe, P.J., Bagust, T.J., Ireland, L., 1984. Development of the normal gastrointestinal microflora of specific pathogen free chickens. J. Hyg. 92, 79–87.

Corry, J.E.L., Atabay, H.I., 2001. Poultry as a source of Campylobacter and related organisms. J. Appl.Microbiol. 90, 96S–114S.

Cowart, R.E., Lashmet, J., McIntosh, M.E., Adams, T.J., 1990. Adherence of a virulent strain of Listeriamonocytogenes to the surface of a hepatocarcinoma cell line via lectin-substrate interaction. Arch.Microbiol. 153, 282–286.

Craven, S.E., 1994. Altered colonizing ability for the ceca of broiler chicks by lipopolysaccharide-deficient mutants of Salmonella typhimurium. Avian Dis. 38, 3, 401–408.


Craven, S.E., Williams, D.D., 1997. Inhibition of Salmonella typhimurium attachment to chicken cecalmucus by intestinal isolates of Enterobacteriaceae and Lactobacilli. Avian Dis. 41, 548–558.

Cullen, G.A., Gorden, R.F., Harry, E.G., Jordan, F.T.W., 1977. Bacterial diseases. In: Gordon, R.F. (Ed.),Poultry Diseases. Bailliere Tindall, London, pp. 10–65.

Curtiss, R. 3rd, Kelly, S.M., Hassan, J.O., 1993. Live oral avirulent Salmonella vaccines. Vet. Microbiol.37, 397–405.

Deguchi, Y., Morishita, T., Mutai, M., 1985. Comparative studies on synthesis of water soluble vitaminsamong species of Bifidobacteria. Agr. Biol. Chem. 49, 13–19.

Deitch, E.A., Berg, R.D., Specian, R.D., 1987. Endotoxin promotes translocation of bacteria from the gut.Arch. Surg. 122, 185–190.

Deitch, E.A., Bridges, W.M., Ma, J.W., Berg, R.D., Specian, R.D., 1990. Obstructed intestine as areservoir for systemic infection. Amer. J. Surg. 159, 394–401.

Dibb-Fuller, M.P., Woodward, M.J., 2000. Contribution of fimbriae and flagella of Salmonella enteritidisto colonization and invasion of chicks. Avian Pathol. 29, 295–304.

Dibb-Fuller, M.P., Allen-Vercoe, E., Thorns, C.J., Woodward, M.J., 1999. Fimbriae- and flagella-mediatedassociation with and invasion of cultured epithelial cells by Salmonella enteritidis. Microbiology 145,1023–1031.

Doig, P., Yao, R.J., Burr, D.H., Guerry, P., Trust, T.J., 1996. An environmentally regulated pilus-likeappendage involved in Campylobacter pathogenesis. Mol. Microbiol. 20, 885–894.

Drucker, M.M., Yeivin, R., Sacks, T.G., 1967. Pathogenesis of Escherichia coli enteritis in the ligatedrabbit gut. Isr. J. Med. Sci. 3, 445–452.

Dubos, R., Schaedler, R.W., Costello, R., Hoet, P., 1965. Indigenous normal and autochthonous flora ofthe gastrointestinal tract. J. Exp. Med. 122, 67–76.

Duchet-Suchaux, M., Lechopier, P., Marly, J., Bernardet, P., Delauney, R., Pardon, P., 1995.Quantification of experimental Salmonella typhimurium. Avian Dis. 38, 401–408.

Duguid, J.P., Anderson, E.S., Campbell, I., 1966. Fimbriae and adhesive properties in salmonellae.J. Pathol. Bacteriol. 92, 107–138.

Dunham, H.J., Williams, C., Edens, F.W., Casas, L.A., Dobrogosz, W.J., 1993. Lactobacillus reuteri immuno-modulation of stress associated diseases in newly hatched chickens and turkeys. Poultry Sci. 72, 103.

Dwars, R.M., Smit, H.F., Davelaar, F.G., Van T. Veer, W., 1989. Incidence of spirochaetal infections incases of intestinal disorders in chickens. Avian Pathol. 18, 591–595.

Dyce, K.M., Sack, W.O., Wensing, C.J.G., 1987. Textbook of Veterinary Anatomy. The DigestiveApparatus. Saunders, London, Chapter 3, pp. 96–142.

EC Council Regulation 2821, 1999. Regulation Antibiotic Resistance in the European Union Associatedwith Therapeutic Use of Veterinary Medicines. Report and Qualitative Risk Assessment by theCommittee for Veterinary Medicinal Products. 14th July 1999. The European Agency for theEvaluation of Medicinal Products. EMEA/CVMP/342/99-Final.

Elwinger, K., Schneitz, C., Berndtson, E., Fossum, O., Teglof, B., Engstrom, B., 1992. Factors affectingthe incidence of necrotic enteritis, caecal carriage of Clostridium perfringens and bird performance inbroiler chicks. Acta Vet. Scand. 33, 369–378.

Ernst, R.K., Dombroski, D.M., Merrick, J.M., 1990. Anaerobiosis, type 1 fimbriae and growth phaseare factors that affect invasion of HEp-2 cells by Salmonella typhimurium. Infect. Immun. 58,2014–2016.

Fanelli, M.J., Sadler, W.W., Franti, C.E., Brownell, J.R., 1971. Localisation of Salmonellae within theintestinal tract of chickens. Avian Dis. 15, 366–375.

Fantham, H.B., 1910. Observations on the parasitic protozoa of red grouse (Lagopus scoticus), with anote on the grouse fly. Proc. Zoo. Soc. Lond. 692–708.

Fauchere, J.L., Kervella, M., Rosenau, A., Mohanna, K., Veron, M., 1989. Adhesion to Hela cells ofCampylobacter jejuni and C. coli outer membrane components. Res. Microbiol. 140, 379–392.

Ferraris, R.P., Villenas, S.A., Diamond, J., 1992. Regulation of brush-border activities and enterocytemigration rates in mouse small intestine. Amer. J. Physiol. 262, G1047–1059.

Ficken, M.D., Wages, D.P., 1997. Necrotic enteritis. In: Calnek, B.W. (Ed.), Diseases of Poultry,10th edition. Mosby-Wolfe: Ames, IA, pp. 261–264.

Finlay, B.B., Gumbiner, B., Falkow, S., 1988. Penetration of Salmonella through a polarized Madin-Darby canine kidney epithelial cell monolayer. J. Cell Biol. 107, 221–230.


Fiorentin, L., Panangala, S., Zhang, Y., Toivio-Kinnucan, M., 1998. Adhesion inhibition of Mycoplasmaiowae to chicken lymphoma DT40 cells by monoclonal antibodies reacting with a 65-KD polypep-tide. Avian Dis. 42, 721–731.

Formal, S.B., Hale, T.L., Sansonetti, P.J., 1983. Invasive enteric pathogens. Rev. Infect. Dis. 5, 5702–5707.Freitas, M., Cayuela, C., Antoine, J.-M., Piller, F., Sapin, C., Trugan, G., 2001. A heat labile soluble

factor from Bacteroides thetaiotaomicron VPI-5482 specifically increases the galatosylation patternof HT29-MTX cells. Cellular Microbiol. 3, 289–300.

Freter, R., 1981. Mechanisms of association of bacteria with mucosal cells. CIBA Found. Symp. 80,36–55.

Fuller, R., 1975. Nature of the determinant responsible for the adhesion of Lactobacilli to chicken cropepithelial cells. J. Gen. Microbiol. 87, 245–250.

Fuller, R., 1989. Probiotics in man and animals. J. Appl. Bacteriol. 66, 365–378.Fuller, R., Brooker, B.E., 1974. Lactobacilli which attach to the crop epithelium of the fowl. Amer. J.

Clin. Nutr. 27, 1305–1312.Fuller, R., Turvey, A., 1971. Bacteria associated with the intestinal wall of the domestic fowl (Gallus

domesticus). J. Appl. Bacteriol. 34, 617–622.Fuller, R., Barrow, P.A., Brooker, B.E., 1978. Bacteria associated with the gastric epithelium of neonatal

pigs. Appl. Environ. Microbiol. 35, 582–591.Galan, J.E., Curtiss, R. III., 1989. Cloning and molecular characterisation of genes whose products allow

Salmonella typhimurium to penetrate tissue culture cells. Proc. Natl. Acad. Sci. USA 86, 6383–6387.Gaynor, E.C., Ghori, N., Falkow, S., 2001. Bile-induced “pili” in Campylobacter jejuni are bacteria-

independent artefacts of the culture medium. Mol. Microbiol. 39, 1546–1549.Genigeorgis, C., 1986. Significance of Campylobacter in poultry. Proceedings of the 35th West Poultry

Disease Conference, pp. 54–59.Gibbons, R.J., 1977. Adherence of bacteria to host tissue. In: Schlessinger D. (Ed.), Microbiology.

American Society for Microbiology, Washington, DC, pp. 395–406.Gitter, M., 1976. Listeria monocytogenes in “oven-ready” poultry. Vet. Rec. 99, 336.Guerry, P., Alm, R.A., Power, M.E., Logan, S.M., Trust, T.J., 1991. Role of two flagellin genes in

Campylobacter motility. J. Bacteriol. 173, 4757–4764.Gusils, C., Gonzalez, S.N., Oliver, G., 1999a. Some probiotic properties of chicken Lactobacilli. Can. J.

Microbiol. 45, 981–987.Gusils, C., Palacios, J., Gonzalez, S., Oliver, G., 1999b. Lectin-like protein in lactic acid bacteria isolated

from chickens. Biol. Pharm. Bull. 22, 11–15.Gustafson, R.H., Bowen, R.E., 1997. Antibiotic use in animal agriculture. J. Appl. Microbiol. 83, 531–541.Guzman, C.A., Biavasco, F., Pruzzo, C., 1997. Adhesiveness of Bacteroides fragilis strains isolated from

faeces of healthy donors, abscesses, and blood. Curr. Microbiol. 34, 332–334.Hagberg, L., Hull, R., Hull, S., Falkow, S., Freter, R., Svanborg-Eden, C., 1983. Contribution of adhe-

sion and bacterial persistence in the mouse urinary tract. Infect. Immun. 40, 265–272.Halula, M.C., Stocker, B.A.D., 1987. Distribution and properties of the mannose-resistant haemagglu-

tinin produced by Salmonella species. Micro. Path. 3, 455–459.Hammar, M., Arnqvist, A., Bian, Z., Olsen, A., Normark, S., 1995. Expression of two csg operons is

required for the production of fibronectin and Congo red binding curli polymers in E. coli K-12. Mol.Biol. 18, 661–670.

Havenaar, R., Huis in’t, Veld, J.H.J., 1992. Probiotics: a general view. The lactic acid bacteria. In:Wood, B.J.B. (Ed.), The Lactic Acid Bacteria in Health and Disease. Elsevier Applied Science. Essex,UK, pp. 151–170.

He, F., Ouweha, A.C., Hashimoto, H., Isolauri, E., Benno, Y., Salmien, S., 2001. Adhesion ofBifidobacterium spp. to human intestinal mucus. Micro. Immun. 45, 259–262.

Heinig, M.J., Dewey, K.G., 1996. Health advantages of breast feeding for infants: a critical review. Nutr.Res. Rev. 9, 89–110.

Henriksson, A., Szewzyk, R., Conway, P.L., 1991. Characteristics of the adherence determinants ofLactobacillus fermentum. Appl. Environ. Microb. 57, 499–502.

Hill, R., Smith, I.M., 1969. The influence of diets containing low or high levels of fish meal, casein plusgelatin or meat meal on the survival of chicks infected with Salmonella gallinarum. Proc. Nutr. Soc.28, 5A–6A.


Hnatow, L.L., Keeler, C.L. Jr., Tessmer, L.L., Czymmek, K., Dohms, J.E., 1998. Characterization ofMGC2, a Mycoplasma gallisepticum cytadhesin with homology to the Mycoplasma pneumoniae30-kilodalton protein P30 and Mycoplasma genitalium P32. Infect. Immun. 66, 3436–3442.

Holt, P.S., Chaubal, L.H., 1997. Detection of motility and putative synthesis of flagellar proteins inSalmonella pullorum cultures. J. Clin. Microbiol. 35, 1016–1020.

Holt, P.S., Gast, R.K., Porter, R.E., Stone, H.D., 1999. Hyporesponsiveness of the systemic and mucosalhumoral immune systems in chickens infected with Salmonella enterica serovar Enteritidis at one dayof age. Poultry Sci. 78, 1510–1517.

Hu, L., Kopecko, D.J., 2000. Interactions of Campylobacter with eukaryotic cells: Gut luminal colo-nization and mucosal invasion mechanism. In: Nachamkin, I., Blaser, M.J. (Eds.), Campylobacter,2nd edition, pp. 191–215.

Imondi, A.R., Bird, F.H., 1966. The turnover of intestinal epithelium in the chick. Poultry Sci. 45, 142–147.Impey, C.S., Mead, G.C., George, S.M., 1982. Competitive exclusion of salmonellas from the chick cae-

cum using a defined mixture of bacterial isolates from the caecal microflora of an adult bird. J. Hyg.Lond. 89, 479–490.

Jeurissen, S.H.M., Janse, E.M., Koch, G., DeBoer, G.F., 1989. Postnatal development of mucosa-associated lymphoid tissue in chickens. Cell Tissue Res. 258, 119–124.

Jin, L.Z., Ho, Y.W., Abdullah, N., Jalaludin, S., 1997. Probiotics in poultry: modes of action. World’sPoultry Sci. J. 53, 351–368.

Johnson, A.P., 1994. The pathogenicity of Enterococci. J. Antimicrob. Chemother. 33, 1083–1089.Jones, B.D., Lee, C.A., Falkow, S., 1992. Invasion by Salmonella typhimurium is affected by the direc-

tion of flagellar rotation. Infect. Immun. 60, 2475–2480.Jones, M.A., Wigley, P., Page, K.L., Hulme, S.D., Barrow, P.A., 2001. Salmonella enterica serovar gallinarum

requires the Salmonella pathogenicity island 2 type III secretion system but not the Salmonella patho-genicity island 1 type III secretion system for virulence in the chickens. Infect. Immun. 69, 5471–5476.

Jordan, F.T.W., Pattison, M., 1996. Poultry Diseases, 4th edition. W.B. Saunders Company Ltd.,Cambridge University Press, Cambridge.

Kahane, I., Tucker, S., Leith, D.K., Morrison-Plummer, J., Baseman, J.B., 1985. Detection of the majoradhesin P1 in triton shells of virulent Mycoplasma pneumoniae. Infect. Immun. 50, 944–946.

Keeler, C.L. Jr., Hnatow, L.L., Whetzel, P.L., Dohms, J.E., 1996. Cloning and characterisation of a puta-tive cytadhesin gene (mgc1) from Mycoplasma gallisepticum. Infect. Immun. 64, 1541–1547.

Kimura, N., Mimura, F., Nishida, S., Kobayashi, A., 1976. Studies on the relationship between intestinalflora and cecal coccidiosis in chicken. Poultry Sci. 55, 1375–1383.

Klemm, P., 1985. Fimbrial adhesins of Escherichia coli. Rev. Infect. Dis. 7, 321–340.Kleven, S.H., 1998. Mycoplasmas in the etiology of multifactorial respiratory disease. Poultry Sci. 77,

1146–1149.Ko, E.J., Goh, J.S., Lee, B.J., Choi, S.H., Kim, P.H., 1999. Bifidobacterium bifidum exhibits a

lipopolysaccharide-like mitogenic activity for murine B lymphocytes. J. Dairy Sci. 82, 1869–1876.Koo, F.C.W., Peterson, J.W., Houston, C.W., Molina, N.C., 1984. Pathogenesis of experimental

Salmonellosis: inhibition of protein synthesis by cytotoxin. Infect. Immun. 43, 93–100.La Ragione, R.M., Cooley, W.A., Woodward, M.J., 2000a. The role of fimbriae and flagella in the adher-

ence of avian strains of Escherichia coli O78:K80 to tissue culture cells and tracheal and gut explants.J. Med. Microbiol. 48, 327–338.

La Ragione, R.M., Sayers, A.R., Woodward, M.J., 2000b. The role of fimbriae and flagella in the colo-nization, invasion and persistence of Escherichia coli O78:K80 in the day-old-chick model. Epidem.Infect. 124, 351–363.

La Ragione, R.M., Casula, G., Cutting, S., Woodward, M.J., 2001. Bacillus subtilis spores competitivelyexclude Escherichia coli O78:K80. Vet. Microbiol. 79, 133–142.

Lee, A., O’Rourke, J.L., Barrington, P.J., Trust, T.J., 1986. Mucus colonisation as a determinant of patho-genicity in intestinal infection by Campylobacter jejuni: a mouse cecal model. Infect. Immun. 51,536–546.

Lee, Y., Nomoto, K., Salminen, S., Gorbach, S.L., 1999. Handbook of Probiotics. Selection andMaintenance of Probiotic Strains. Wiley-Interscience. New York, pp. 23–35.

Lenener, A.A., Lenener, C.H.P., Mikelsaar, M.E., 1984. The quantitative composition of the gut lactoflorabefore and after cosmic flights of different duration. Die Nahrung 28, 607–713.


Leung, K.P., Fukushima, H., Sagawa, H., Walker, C.B., Clark, W.B., 1989. Surface appendages, hemag-glutination, and adherence to human epithelial cells of Bacteroides intermedius. Oral Microbiol.Immunol. 4, 204–210.

Licht, T.R., Krogfelt, K.A., Cohen, P.S., Poulsen, L.K., Urbance, J., Molin, S., 1996. Role of lipopolysac-charide in colonization of the mouse intestine by Salmonella typhimurium studied by in situ hybridi-sation. Infect. Immun. 64, 3811–3817.

Lievin, V., Peiffer, I., Hudault, S., Rochat, F., Brassart, D., Neeser, J.R., Servin, A.L., 2000.Bifidobacterium strains from resident infant human gastrointestinal microflora exert antimicrobialactivity. Gut 47, 646–652.

Lindberg, F.P., Lund, B., Normark, S., 1984. Genes of pyelonephritogenic E. coli required fordigalactosidase-specific agglutination of human cells. EMBO J. 3, 1167–1173.

Lizko, N.N., Silov, V.M., Syrych, G.D., 1984. Particularities in the formation of an intestinal dysbacte-riosis in man under extreme conditions. Die Nahrung 28, 599–605.

Lockman, H.H., Curtiss, R., 1990. Salmonella typhimurium mutants lacking flagella or motility remainvirulent to BALB/C mice. Infect. Immun. 58, 137–143.

Lockman, H.H., Curtiss, R., 1992. Isolation and characterization of conditional adherent and non-type 1fimbriated Salmonella typhimurium mutants. Mol. Microbiol. 6, 933–945.

Macnab, R.M., 1987a. Flagella. In: Neidhardt, F.C., Ingraham, J.L., Low, K.B., Magasanik, B.,Schaechter, M., Umbarger, H.E. (Eds.), Escherichia coli and Salmonella typhimurium: Cellular andMolecular Biology. American Society for Microbiology, Washington, DC, pp. 123–145.

Macnab, R.M., 1987b. Motility and chemotaxis. In: Neidhardt, F.C., Ingraham, J.L., Low, K.B.,Magasanik, B., Schaechter, M., Umbarger, H.E. (Eds.), Escherichia coli and Salmonella typhimurium:Cellular and Molecular Biology. American Society for Microbiology, Washington, DC, pp. 70–83.

Maniloff, J., Morowitz, H.J., Barrnett, R.J., 1965. Ultrastructure and ribosomes of Mycoplasmagallisepticum. J. Bacteriol. 90, 193–204.

Massey, R.C., Kantzanou, M.N., Fowler, T., Day, N.P., Schofield, K., Wann, E.R., Berendt, A.R., Hook, M.,Peacock, S.J., 2001. Fibronectin-binding protein A of Staphylococcus aureus has multiple, substituting, binding regions that mediate adherence to fibronectin and invasion of endothelial cells.Cell Microbiol. 3, 839–851.

McReynolds, J.L., Caldwell, D.Y., McElroy, A.P., Hargis, B.M., Caldwell, D.J., 2000. Antimicrobialresidue detection in chicken yolk samples following administration of egg-producing chickens andeffect of residue detection on competitive exclusion culture (PREEMPT) establishment. J. Agr. FoodChem. 25, 6435–6438.

McSweegan, E., Walker, R.I., 1986. Identification and characterization of two Campylobacter jejuniadhesins for cellular and mucous substrates. Infect. Immun. 53, 141–148.

McSweegan, E., Burr, D.H., Walker, R.I., 1987. Intestinal mucus gel and secretory antibody are barriersto Campylobacter jejuni adherence to INT 407 cells. Infect. Immun. 55, 1431–1435.

Mead, G.C., Hudson, W.R., Ariffin, R., 1990. Survival and growth of Listeria monocytogenes on irradi-ated poultry carcasses. Lancet 335, 1036.

Mecsas, J., Bilis, I., Falkow, S., 2001. Identification of attenuated Yersinia pseudotuberculosis strains andcharacterization of an orogastric infection in BALB/c mice on day 5 postinfection by signature-tagged mutagenesis. Infect. Immun. 69, 2779–2787.

Meddings, J.B., Desouza, D., Goel, M., Thiesen, S., 1990. Glucose transport and microvillus membranephysical properties along the crypt-villus axis of the rabbit. J. Clin. Inv. 5, 1099–1107.

Metchnikoff, E., 1907. The Prolongation of Life, Optimistic Studies. William Heinemann, London.Mims, C.A., 1987. The Pathogenesis of Infectious Disease, 3rd edition. Academic Press, London.Mitsuoka, T., 1982. Recent trends in research on intestinal flora. Bifidobact. Microflora 1, 3–24.Mitsuoka, T., 1990. Bifidobacteria and their role in human health. J. Ind. Microbiol. 6, 263–286.Montie, T.C., Doyle-Huntzinger, D., Craven, R.C., Holder, I.A., 1982. Loss of virulence associated with

absence of flagellum in an isogenic mutant of Pseudomonas aeruginosa in the burned-mouse model.Infect. Immun. 38, 1296–1298.

Mooi, F.R., De Graaf, F.K., 1985. Molecular biology of fimbriae of enterotoxigenic Escherichia coli.Curr. Top. Micro. Immun. 118, 119–138.

Morata de Ambrosini, V., Gonzalez, S.P., de Ruiz Holgado, A., Oliver, G., 1988a. Study of the morphologyof the cell walls of some strains of lactic acid bacteria and related species. J. Food Prot. 61, 557–562.


Morata de Ambrosini, V., Gonzalez, S.P., de Ruiz Holgado, A., Oliver, G., 1988b. Immuno-stimulatingactivity of cell walls from lactic acid bacteria and related species. Food Agr. Immun. 10, 183–191.

Muniappa, N., Duhamel, G.E., 1997. Outer membrane-associated serine protease of intestinal spiro-chetes. FEMS Microbiol. Lett. 154, 159–164.

Murakami, S., Okazaki, Y., Kazama, T., Suzuki, T., Iwabuchi, I., Kairioka, K., 1989. A dual infection ofClostridium perfringens and Escherichia coli in broiler chicks. J. Japan. Vet. Med. Assoc. 42, 405–409.

Newell, D.G., 2001. Animal models of Campylobacter jejuni colonisation and disease and the lessons tobe learned from similar Helicobacter pylori models. Symp. Ser. Soc. Appl. Microbiol. 30, 57S-67S.

Newell, D.G., Wagenaar, J.A., 2000. Poultry infections and their control at the farm level. In: Nachamkin, I.,Blaser, M.J. (Eds.), Campylobacter, 2nd edition. American Society for Microbiology, Washington, DC,pp. 497–509.

Newell, D.G., McBride, H., Dolby, J.M., 1985. Investigation on the role of flagella in the colonisation ofinfant mice with Campylobacter jejuni and attachment of Campylobacter jejuni to human epithelialcell lines. J. Hyg. Lond. 95, 217–227.

Nuitjen, P.J.M., Wassenar, T.M., Newell, D.G., Van der Zeijst, B.A.M., 1992. Molecular characterisationand analysis of Campylobacter jejuni flagellin genes and proteins. In: Nachamkin, I., Blaser M.J.,Tomkins L.S. (Eds.), Campylobacter jejuni: Current Status. ASM Press, Washington, DC, pp. 282–296.

Nuotio, L., Mead, G.C., 1993. An in vitro model for studies on bacterial interactions in the avian caecum.Lett. Appl. Microbiol. 17, 65–67.

Nuotio, L., Schneitz, C., Halonen, U., Nurmi, E., 1992. Use of competitive exclusion to protect newly-hatched chicks against intestinal colonisation and invasion by Salmonella enteritidis PT4. Brit.Poultry Sci. 32, 775–779.

Nurmi, E., Rantala, M., 1973. New aspects of Salmonella infection in broiler production. Nature 241,210–211.

Olsen, A., Jonsson, A., Normark, S., 1989. Fibronectin binding mediated by a novel class of surfaceorganelles on Escherichia coli. Nature 338, 652–655.

Olsen, A., Arnqvist, A., Hammar, M., Sukupolvi, S., Normark, S., 1993. The RpoS sigma factor relievesH-NS-mediated transcriptional repression of csgA, the sub-unit gene of fibronectin-binding curli inE. coli. Mol. Microbiol. 7, 523–536.

Ouwehand, A.C., Isolauri, E., Kirjavainen, P.V., Salminen, S.J., 1999. Adhesion of four Bifidobacteriumstrains to human intestinal mucus from subjects in different age groups. FEMS Microbiol. Lett. 172,61–64.

Parish, W.E., 1961. Necrotic enteritis in the fowl (Gallus gallus domesticus). I. Histopathology of the dis-ease and isolation of a strain of Clostridium welchii. Comp. Path. 71, 377–333.

Pasteur, L., Joubert, J.F., 1877. Charbon et septicemie. C.R. Soc. Biol. Paris 85, 101–115.Patterson, J.A., Orben, J.I., Sutton, A.L., Richards, G.N., 1997. Selective enrichment of Bifidobacteria in

the intestinal tract of broilers by thermally produced kestoses and effect on broiler performance.Poultry Sci. 76, 497–500.

Peckham, M.C., 1984. Avian vibrionic hepatitis. In: Hofstad, M.S. (Ed.), Diseases of Poultry, 8th edition.Iowa State University Press, Ames, Iowa, pp. 221–229.

Pei, Z., Burucoa, C., Grignon, B., Baqar, S., Huang, X.Z., Kopecko, D.J., Bourgeosis, A.L., Fauchere, J.L.,Blaser, M.J., 1998. Mutation in the peb1A locus of Campylobacter jejuni reduces interactions withepithelial cells and intestinal colonisation of mice. Infect. Immun. 66, 938–943.

Penn, C.W., 2001. Surface components of Campylobacter and Helicobacter. J. Appl. Microbiol. 90, 25S–35S.Petr, J., Rada, V., 2001. Bifidobacteria are obligate inhabitants of the crop of adult laying hens. J. Vet.

Med. B. Infect. Dis. Vet. Pub. Health 48, 227–233.Pivnick, H., Nurmi, E., 1982. The Nurmi concept and its role in the control of Salmonellae in poultry. In:

Davies, R. (Ed.), Developments in Food Microbiology – 1. Applied Science Pub, London, pp. 41–70.Pourbakhsh, S.A., Dho-Moulin, M., Bree, A., Desautels, C., Martineau-Doize, B., Fairbrother, J.M.,

1997. Localisation of the in vivo expression of P and F1 fimbriae in chickens experimentally inocu-lated with pathogenic Escherichia coli. Microbiol. Path. 22, 331–342.

Prince, W.R., Garren, H.W., 1966. An investigation of the resistance of White Leghorn chicks toSalmonella gallinarum. Poultry Sci. 45, 1149–1153.

Ratiner, Y.A., 1999. Temperature-dependent flagellar antigen phase variation in Escherichia coli. Res.Microbiol. 150, 457–463.


Razin, S., Jacobs, E., 1992. Mycoplasma adhesion. J. Gen. Microbiol. 138, 407–422.Rettger, L.F., Plastridge, W.N., 1932. Pullorum disease in domestic fowl. Storrs Agr. Exp. Stat. Bull. 178,

107–192.Rhoades, K.R., Rimler, R.B., 1991a. Pseudotuberculosis. In: Calnek, B.W. (Ed.), Diseases of Poultry,

9th edition. Wolfe Publishing Ltd, USA, pp. 141–166.Rhoades, K.R., Rimler, R.B., 1991b. Pasteurellosis. In: Calnek, B.W. (Ed.), Diseases of Poultry,

9th edition. Wolfe Publishing Ltd, USA, pp. 165–177.Roberts, J.A., Kaack, M.B., Fuussell, E.N., 1989. Bacterial adherence in urinary tract infections: prelim-

inary studies in a primate model. Infection 17, 401–404.Ruiz-Palacios, G.M., Escamilla, E., Torres, N., 1981. Experimental Campylobacter diarrhoea in chickens.

Infect. Immun. 34, 250–255.Rychlik, I., Lovell, M.A., Barrow, P.A., 1998. The presence of genes homologous to the K88 genes faeH

and faeI on the virulence plasmid of Salmonella gallinarum. FEMS Microbiol. Lett. 15, 255–260.Saavedra, J.M., Bauman, N.A., Oung, I., Perman, J.A., Yolken, R.H., 1994. Feeding of Bifidobacterium

bifidum and Streptococcus thermophilus to infants in hospital for prevention of diarrhoea and shed-ding of rotavirus. Lancet 344, 1046–1049.

Sadler, W.W., Corstvet, R.E., 1965. The effect of Erysipelothrix insidiosa infection on wholesome ofmarket turkeys. Amer. J. Vet. Res. 26, 1429–1436.

Sagartz, J.E., Swayne, D.E., Eaton, K.A., Hayes, J.R., Amass, K.D., Wack, R., Kramer, L., 1992. Necrotizingtyphlocolitis associated with a spirochete in rheas (Rhea americana). Avian Dis. 36, 282–289.

Sangari, F.J., Goodman, J., Petrofsky, M., Kolonoski, P., Bermudez, L.E., 2001. Mycobacterium aviuminvades the intestinal mucosa primarily by interacting with enterocytes. Infect. Immun. 69, 1515–1520.

Savage, D.C., 1969. Microbial interference between indigenous yeast and Lactobacilli in the rodentstomach. J. Bact. 98, 1278–1283.

Savage, D.C., 1983. Mechanisms by which indigenous micro-organisms colonize gastrointestinal epithe-lial surfaces. Prog. Food. Nutr. Sci. 7, 65–74.

Saxelin, M., Pessi, T., Salminen, S., 1995. Fecal recovery following oral administration of LactobacillusGG (ATCC 53103) in gelatin capsules to healthy volunteers. Int. J. Food Microbiol. 25, 199–203.

Schmidt, M.A., 1994. Nonfimbrial adhesins of Escherichia coli. In: Klemm, P. (Ed.), Fimbriae:Adhesion, Genetics, Biogenesis and Vaccines. CRC Press, London, Chapter 5, pp. 85–96.

Schneitz, C., Mead, G.M., 2000. Competitive exclusion. In: Wray, C., Wray, A. (Eds.), Salmonella inDomestic Animals. Wallingford, CABI Publishing, pp. 301–322.

Schoeni, J.L., Doyle, M.P., 1994. Variable colonisation of chickens perorally inoculated with Escherichiacoli O157:H7 and subsequent contamination of eggs. Appl. Environ. Microbiol. 60, 2958–2962.

Shane, S.M., 1991. Campylobacteriosis. In: Calnek, B.W. (Ed.), Diseases of Poultry, 9th edition. WolfePublishing, USA, pp. 236–246.

Sharma, A., Sojar, H.T., Glurich, I., Honma, K., Kuramitsu, H.K., Genco, R.J., 1998. Cloning, expres-sion, and sequencing of a cell surface antigen containing a leucine-rich repeat motif from Bacteroidesforsythus ATCC 43037. Infect. Immun. 66, 5703–5710.

Shivaprasad, H.L., Nagaraja, K.V., Pomeroy, B.S., Williams, J.E., 1997. Salmonella infections: arizonosis.In: Diseases of Poultry, 10th edition. Iowa State University Press, Ames, Iowa, Chapter 3, pp. 122–129.

Siccardi, F.J., 1966. Identification and disease producing ability of Escherichia coli associated withE. coli infection of chickens and turkeys. MSc thesis, University of Minnesota, USA.

Sjobring, U., Pohl, G., Olsen, A., 1994. Plasminogen, absorbed Escherichia coli expressing curli or bySalmonella enteritidis expressing thin, aggregative fimbriae, can be activated by simultaneouslycaptured tissue-type plasminogen activator (t0PA). Mol. Microbiol. 14, 443–452.

Skeeles, J.K., 1991. Staphylococcosis. In: Calnek, B.W. (Ed.), Diseases of Poultry, 9th edition. WolfePublishing Ltd, USA, pp. 293–299.

Skirrow, M.B., Benjamin, J., 1980. ‘1001’ Campylobacters: Cultural characteristics of intestinalCampylobacters from man and animals. J. Hyg. Lond. 85, 427–442.

Smith, H., 1972. The little-known determinants of microbial pathogenicity. Soc. Gen. Microbiol. 22, 1–24.Smith, H.W., Halls, S., 1968. The production of oedema disease and diarrhoea in weaned pigs by the oral

administration of Escherichia coli: factors that influence the course of the experimental disease.J. Med. Microbiol. 1, 45–59.


Smyth, C.J., 1988. Flagella: their role in virulence. In: Owen, P., Foster, T.J. (Eds.), Immunochemical andMolecular Genetic Analysis of Bacterial Pathogens. Elsevier, Amsterdam, pp. 3–11.

Snoeyenbos, G.H., Weinack, O.M., Smyser, F., 1978. Protecting chicks and poults from Salmonellae byoral administration of “normal” gut microflora. Avian Dis. 22, 273–287.

Sojka, W.J., Carnaghan, R.B.A., 1961. Escherichia coli infection in poultry. Res. Vet. Sci. 2, 340–352.Songer, J.G., 1997. Clostridial diseases of animals. In: Rood, J.I., McClaine, B.A., Songer, J.G.,

Titball, R.W. (Eds.), The Clostridia Molecular Biology and Pathogenesis. Academic Press, London,pp. 153–182.

Spencer, J.L., Chambers, J.R., Modler, H.W., 1998. Competitive exclusion of Salmonella typhimurium inbroilers fed with Vermipost and complex carbohydrates. Avian Path. 27, 244–249.

Stavric, S., 1992. Defined cultures and prospects for probiotics. Int. J. Food. Microbiol. 15, 173–180.Stavric, S., Gleeson, T.M., Blanchfield, B., 1991. Effect of avian intestinal microflora possessing adhering

and hydrophobic properties on competitive exclusion of Salmonella typhimurium from chicks.J. Appl. Bacteriol. 70, 414–421.

Suau, A., Bonnet, R., Sutren, M., Godon, J.J., Gibson, G.R., Collins, M.D., Dore, J., 1999. Direct analysisof genes encoding 16S rRNA from complex communities reveals many novel species within thehuman gut. Appl. Environ. Microbiol. 65, 4799–4807.

Sueyoshi, M., Adachi, Y., 1990. Diarrhea induced by Treponema hyodysenteriae: a young chick cecalmodel for swine dysentery. Infect. Immun. 58, 3348–3362.

Sueyoshi, M., Adachi, Y., Shoya, S., Miyagawa, E., Minato, H., 1986. Investigations into the location ofTreponema hyodysenteriae in the cecum of experimentally infected young broiler chicks by light- andelectron microscopy. Zentralbl. Bacteriol. Mikrobiol. Hyg. (A) 261, 447–453.

Swayne, D.E., McLaren, A.J., 1997. Avian intestinal spirochaetes and avian spirochaetosis. In: Hamson, D.J.,Stanton, T.B. (Eds.), Intestinal Spirochaetes in Domestic Animals and Humans. CAB International,Wallingford, pp. 267–300.

Swayne, D.E., Eaton, K.A., Stoutenberg, J., Trott, D.J., Hampson, D.J., Jenson, N.S., 1995. Identificationof a new intestinal spirochaete with pathogenicity for chickens. Infect. Immun. 63, 430–436.

Szymanski, C.M., King, M., Haardt, M., Armstrong, G.D., 1995. Campylobacter jejuni motility andinvasion of Caco-2 cells. Infect. Immun. 63, 4295–4300.

Takeuchi, A., 1967. Electron microscopy studies of experimental Salmonella infections. Amer. J. Pathol.50, 109–136.

Tannock, G.W., 1987. Demonstration of mucosa-associated microbial populations in the colons of mice.Appl. Environ. Microbiol. 53, 1965–1968.

Tannock, G.W., 1990. In: Marshal, K.C. (Ed.), Advances in Microbial Ecology, Vol. 11. Plenum Press.New York, pp. 147–171.

Tannock, G.W., Savage, D.C., 1974. Influences of dietary and environmental stress on microbial popula-tions in the murine gastrointestinal tract. Infect. Immun. 9, 591–598.

Thoen, C.O., Karlson, A.G., 1991. Tuberculosis. In: Calnek, B.W. (Ed.), Diseases of Poultry, 9th edition.Wolfe Publishing Ltd, USA, pp. 172–185.

Thomson, A.B.R., Cheesman, C.L., Keelan, M., Fedorak, R., Clandinin, M.T., 1994. Crypt cell produc-tion rate, enterocyte turnover time and appearance of transport along the jejunal villus of the rat.Biochim. Biophys. Acta 863, 197–204.

Timoshko, M.A., Vil’Shanskaia, F.L., Pospelova, W., Rakhimova, N.G., 1981. Interactions betweenBifidobacteria bifidum, Proteus vulgarius and Klebsiella pneumonia 204 in the gastrointestinal tractof gnotobiotic chicks. Zh. Mikrobiol. Epidemiol. Immunobiol. 3, 58–61.

Traber, P.G., Gumucio, D.L., Wang, W., 1991. Isolation of intestinal epithelial cells for the study ofdifferential gene expression along the crypt-villus axis. Amer. J. Physiol. 260, G895–903.

Trampel, D.W., Jensen, N.S., Hoffman, L.J., 1994. Cecal spirochetosis in commercial laying hens. AvianDis. 38, 895–898.

Trott, D.J., Hampson, D.J., 1998. Evaluation of day-old specific pathogen-free chicks as an experi-mental model for pathogenicity testing of intestinal spirochaete species. J. Comp. Pathol. 118,365–381.

Trott, D.J., Stanton, T.B., Jenson, N.S., Duhamel, G.E., Johnson, J.L., Hampson, D.J., 1996. Serpulinapilosicoli spp. The agent of porcine intestinal Spirochaetosis. Int. J. Syst. Bact. 46, 206–215.


Turner, A.K., Lovell, M.A., Hulme, S.D., Zhang-Barber, L., Barrow, P.A., 1998. Identification ofSalmonella typhimurium genes required for colonization of the chicken alimentary tract and forvirulence in newly hatched chicks. Infect. Immun. 66, 2099–2106.

Van den Bosch, J.F., Hendriks, J.H., Gladigau, I., Willems, H.M., Storm, P.K., De Graaf, F. K., 1993.Identification of F11 fimbriae on chicken Escherichia coli strains. Infect. Immun. 61, 800–806.

Van der Werf, M.J., Venema, K., 2001. Bifidobacteria: Genetic modification and the study of their rolein the colon. J. Agr. Food Chem. 49, 378–383.

Van der Wielen, P.W., Biesterveld, S., Notermans, S., Hofstra, H., Urlings, B.A., Van Knapen, F., 2000.Role of volatile fatty acids in development of cecal microflora of broiler chickens during growth.Appl. Environ. Microbiol. 66, 2536–2540.

Van Vlient, A.H.M., Ketley, J.M., 2001. Pathogenesis of enteric Campylobacter infection. J. Appl.Microbiol. 90, 45S–56S.

Vidotto, M.C., Navarro, H.R., Gaziri, L.C., 1997. Adherence pili of pathogenic strains of avianEscherichia coli. Vet. Microbiol. 5, 79–87.

Von Nicolai, H., Esser, P., Lauer, E., 1981. Partial purification and properties of neuraminidase fromBifidobacterium lactentis. Hoppe Seylers Z. Physiol. Chem. 362, 153–162.

Wages, D.P., 1991. Streptococcosis. In: Calnek, B.W. (Ed.), Diseases of Poultry, 9th edition. WolfePublishing Ltd, USA, pp. 299–303.

Wassenar, T.M., van der Zeijst, B.A.M., Newell, D.G., 1993. Colonisation of chicks by motility mutantsof Campylobacter jejuni demonstrates the importance of flagellin A expression. J. Gen. Microbiol.139, 1171–1175.

Wassenaar, T.M., Bleumink-Pluym, N.M., van der Zeijst, B.A., 1991. Inactivation of Campylobacterjejuni flagellin genes by homologous recombination demonstrates that flaA but not flaB is requiredfor invasion. EMB J. 8, 2055–2061.

Wassenar, T.M., Bleumink-Pluym, N.M.C., Newell, D.G., Nuitjen, P.J.M., van der Zeijst, B.A.M., 1994a.Differential flagellin expression in a flaA-flaB mutant of Campylobacter jejuni. Infect. Immun. 62,3901–3906.

Wassenar, T.M., Newell, D.G., van der Zeijst, B.A.M., 1994b. Towards the use of attenuatedCampylobacter mutants as a probioticum in poultry. In: Jensen, J.F., Hinton, M.M., Mulder, R.W.A.W.(Eds.), FLAIR No. 6, Prevention and Control of Potentially Pathogenic Microorganisms in Poultryand Poultry Meat Processing. Probiotics and Pathogenicity. COVP-DLO Het Spelderholt, TheNetherlands, Vol. 12, pp. 85–90.

Webb, D.M., Duhamel, G.E., Mathiesen, M.R., Muniappa, N., White, A.K., 1997. Cecal Spirochaetosisassociated with Serpulina pilosicoli in captive juvenile ring-necked pheasants. Avian Dis. 41, 997–1002.

Weinack, O.M., Snoeyenbos, G.H., Smyser, C.F., Soerjadi-Liem, A.S. 1984. Influence of Mycoplasmagallisepticum, infectious bronchitis, cyclophosphamide on chickens protected by native intestinalmicroflora against Salmonella typhimurium or Escherichia coli. Avian Diseases 28, 416–425.

Weinstein, D.L., Carsiotis, M., Lessner, C.R., O’Brien, A.D., 1984. Flagella help Salmonellatyphimurium survive within murine macrophages. Infect. Immun. 46, 814–818.

Weiser, M.M., 1973. Intestinal cell surface membrane glycoprotein synthesis. I. An indicator of cellulardifferentiation. J. Biol. Chem. 248, 2536–2541.

Wesney, E., Tannock, G.W., 1979. Association of rat, pig and fowl biotypes of lactobacilli with the stom-ach of gnotobiotic mice. Micro. Ecol. 5, 35–42.

Witte, W., Klare, I., Werner, G., 1999. Selective pressure by antibiotics as feed additives. Infection 27,35–38.

Witte, W., 2000. Ecological impact of antibiotic use in animals on different complex microflora: envi-ronment. Int. J. Antimicrob. Agents 14, 321–325.

Woolcock, J.B., 1988. Bacterial resistance to humoral defence mechanisms: an overview. In: Roth, J.A.(Eds.), Virulence Mechanisms of Bacterial Pathogens. American Society for Microbiology,Washington, DC, pp. 73–93.

Wright, M.A., Alison, M., 1994. The Biology of Epithelial Cell Populations. Oxford, Oxford UniversityPress.


Yancey, R.J., Willis, D.L., Berry, L.J., 1978. Role of motility in experimental cholera in adult rabbits.Infect. Immun. 22, 387–392.

Yao, R., Burr, D.H., Doig, P., Trust, T.J., Niu, H., Guerry, P., 1994. Isolation of motile and non-motileinsertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukary-otic cells. Mol. Microbiol. 14, 883–893.

Yoder, H.W. Jr., 1991. Mycoplasmosis. In: Calnek, B.W. (Ed.), Diseases of Poultry. 9th edition.Wolfe Publishing Ltd, USA, pp. 196–235.

Ziprin, R.L., Young, C.R., Stanker, L.H., Hume, M.E., Konkel, M.E., 1999. The absence of cecal colo-nization of chicks by a mutant of Campylobacter jejuni not expressing bacterial fibronectin-bindingprotein. Avian Dis. 43, 586–589.


293

Immunology has all the beauty and fascination but also all the complexitiesand uncertainties of research in biology (Zinkernagel, 2000).

Animals have the ability to develop an active response in their contact with thevariety of pathogens from their environment owing to their evolutionary adaptedimmune system. The mammalian immune system consists of innate (non-specific)and acquired (specific) humoral and cellular mechanisms of self defence whichwere developed by the co-evolution between the organism and the environment. A special role is played by the mucosal immune system (MIS) which, together with the skin immune system (SIS), is critical for efficient protection against infec-tious agents and also in developing a tolerance towards food antigens. Mucosal-associated lymphoid tissues (MALT) form a network of functional integrity, whichis achieved by communication of cells via secreted cytokines/chemokines and bytrafficking of immune cells between different mucosae. Development of this part of the immunity is also attributed to microflora colonizing mucosal surfaces. In this review we present and discuss the mechanisms of immune protection in the

13 Mucosal immunity and the bovine entero-mammary link: evolutionaryestablished dialogue between antigen and arms of immune system1

M. Niemialtowskia, A. Schollenbergera and W. Klucinskib

aImmunology Laboratory, Division of Virology, Mycology and Immunology,Department of Preclinical Sciences, Warsaw Agricultural University, Grochowska272, PL-03-849 Warsaw, PolandbDepartment of Clinical Sciences, Faculty of Veterinary Medicine,Warsaw Agricultural University (SGGW), Ciszewskiego 8, 02-786 Warsaw,Poland

1Supported by the State Committee for Scientific Research (KBN, Grant No. 6 P06K 020 21) and by the Foundation forPolish Sciences: (i) Professors of 2000 grants (No. 10/2000 for M. Niemialtowski), and (ii) IMMUNO Program (1999) –both institutions in Warsaw, Poland.



294 M. Niemialtowski, A. Schollenberger and W. Klucinski

gastrointestinal tract and in the bovine mammary gland as an entero-mammary link,which is extremely important for the newborn.

1. INTRODUCTION

Immunology, as an experimental and clinical science, developed at the end of the eighteenth century and was directly associated with the work of Edward Jenner(1749–1823), an English physician who, knowing nothing about the true nature ofsmallpox, used the material from cowpox lesions as a “vaccine” to protect humansagainst smallpox variolation. From the work of Louis Pasteur (1822–1895), followedby others, the exogenous infectious nature of many human and animal diseasesbecame well established. Nonetheless, we still know little about the functioning of theimmune system from birth to maturity – from immune innocence to immune memory.

The new threats of the new transmissible agents prions have been described, andevidence is growing that they may cross the interspecies barrier (reviewed in detail byPrusiner, 1996; Johnson, 1998). Consumers fear that bovine spongiform encephalo-pathy (BSE) agents may be present in beef and in bovine milk so that the food maytransmit the non-curable, deadly disease. There are also fears of “old” animal infectiousdiseases such as foot-and-mouth disease (FMD) or the “new” diseases caused by HIVin humans or SIV (simian immunodeficiency virus) in primates (Luciw, 1996).

Our understanding is based mostly on studies performed on laboratory animals,which differ significantly from farm animals and also from humans. There is, how-ever, a universal model (pattern) of cell-to-cell communication that can be achievedby a variety of receptor–ligand interactions. These molecules, either expressed onthe cell surface or secreted from the cell, establish a functional network for homeo-stasis within which integrity is achieved. The mucosal immune system, togetherwith the skin’s immune system, directly communicate with the outside environmentwith its enormous number of foreign antigens, many of them being a pathologicalthreat for the host (VanCott et al., 2000). Local mechanisms of defence in the gastrointestinal, respiratory and genitourinary tracts and also within the mammarygland(s) are an integral part of the whole immune system. The immune system islocalized in every type of mucosa, and each is well adjusted to the function that it plays(Savage, 1977; Kaganoff, 1996; Shanahan, 1997). In order to maintain homeostasis, abalance must be kept between signals that activate necessary immune responses andsignals that may induce improper immune reactions leading to pathology, i.e.autoimmune diseases (Abbas et al., 1994; Bell, 1998; Zinkernagel, 2000).

This short introduction is an effort to provoke questions about immune tolerance/immune response towards commensal microflora. Ruminants, unlike othermammals, depend entirely on microorganisms that enable them to utilize cellulose,hemicellulose, lignin, gums, pectin and cellobiose from plants. Digestion of dietaryfibres takes place in the forestomach, mostly in the rumen, which is a huge fermen-tative vat. A newborn does not have bacteria in its immature gastrointestinal (GI)

Mucosal immunity and entero-mammary link 295

tract and during the first few days of life when subsequent bacterial colonizationtakes place, the GI tract also matures (Tizard, 2000). Thus, not only enzymatic/digestive activity but also immune mechanisms need some time to be developed. In horses and in pigs, degradation of plant fibres takes place in the colon where com-mensal bacteria secrete cellulolytic enzymes and also synthesize vitamins. Intestinalautochthonic microflora are very competitive so, in cooperation with local hostdefences, create a hostile environment for many pathogenic microorganisms.Therefore, favouring the growth of commensals by feeding animals with probioti-cally formulated food is advantageous for the host.

A very special example of local immunity operates within the bovine mammarygland (Salmon, 1999). Milk is a stable constituent of the human diet in most partsof the world and the increasing rate of subclinical mastitis in highly geneticallyselected dairy cattle presents a serious problem for consumer health. Inflammatoryresponses within the mammary gland are also the cause of huge economical losses.The immune response within the bovine mammary gland is subjected to enormousphysiological pressure if highly producing dairy cows have to give daily 20 litres ofmilk. On the other hand the non-lactating period is characterized by massive apop-tosis of secreting parenchymal cells. Immune cells located within mammary glandare challenged by various modulatory factors in colostrums/milk during lactationand by apoptotic signals during the dry period.

2. IMMUNOLOGICAL AND NON-IMMUNOLOGICAL PROTECTIONOF THE GASTROINTESTINAL TRACT

In mammals the surface area of all mucosal tissues exceeds the skin surface area bymore than 200 times. Mucosal-associated lymphoid tissues (MALT) are composedof gut-associated lymphoid tissues (GALT), bronchus-associated lymphoid tissues(BALT), nasal (nasopharynx)-associated lymphoid tissues (NALT) and lymphoidtissues in the reproductive tract, urinary tract, mammary gland(s), lacrimal glandsand salivary glands. Immune cells are able to migrate within the tightly regulatednetwork of mucosal tissues, therefore local immune responses in one system canresult in stimulation of local immunity within the whole MALT (Stokes and Bourne,1989; Herbert et al., 1995; Green et al., 1997; Holmgren and Rudin, 1999; Lee andMekalanos, 1999; McGhee et al., 1999).

The alimentary tract is the most important route for foreign antigen entry. It hasa unique property of simultaneous immune exclusion, immune elimination, devel-opment of immune tolerance and certainty of mounting an efficient immuneresponse by associated lymphoid tissue. It requires perfect cooperation betweenantigen presenting cells (APCs) and immune cells, and precise regulation throughthe locally secreted cytokines/chemokines/hormones. Conditions within the alimen-tary tract are challenging, yet healthy individuals survive, owing to the subtle inter-play between various bioactive molecules and immune cells within the mucosa.


Non-specific, i.e. innate, defence mechanisms in the GI tract, include an acidousgastric environment, mucosal protection, an epithelial barrier and peristalsis. Animportant factor of host defence is autochthonic microflora in the forestomach ofruminants and in the large intestine of monogastric animals. Intestinal peristalsismixes the food and promotes the broad contact of ingested nutrients with themucosa. In the colon, peristaltic movement prolongs such contact. Intestinal smoothmuscle contractions are under intrinsic (Meissner and Auerbach plexuses) andextrinsic neural control. Primary and accessory cells involved in the immune andnervous systems create a network through surface receptors of neurons and immunecells, i.e. lymphocytes (Trautmann and Vivier, 2001). Khan et al. (2001) showed thatagrin (glycoprotein in neuromuscular junctions/synapses between neurons and muscle cells) is also present in the immune system network and may participate in rosette formation by APCs and T cells. Thus, agrin may be also critical for presentation of foreign antigens to APCs which are located within the GI tract andin regulation of the mucosal immune system. The GI tract mucosa in the small intestine provides an absorptive function, extending into the lumen on the finger-likeprojections of the innumerable villi. Their surface contains three major types ofepithelial cells: columnar absorptive cells that may also serve as APCs, mucin-producing goblet cells and several endocrine cells. Within the crypts are undifferen-tiated crypt cells, goblet cells and Paneth cells (Voynow and Rose, 1994; Ayabe et al., 2000; Ganz, 2000; Hermann, 2000; Nordman, 2000).

In the colon, the mucosa does not absorb digested nutrients but retains water andelectrolytes from liquid content; it does not have villi and is almost flat, but itsepithelium does have columnar absorptive cells and goblet mucous cells. Within theGI mucosa, different immune cells are distributed and communication viacytokines, chemokines, chemotactic peptides, neuropeptides, leukotrienes, vasoac-tive peptides and many more, allows not only the coordination of the absorptive anddigestive function but also the immune surveillance and protection of the host(Voynow and Rose, 1994; Bagglioni and Loetscher, 2000). In reference to the GItract, antigen-specific immune cells (cytotoxic T lymphocytes, CTLs) have alsobeen identified between epithelial cells and lamina propria lymphocytes. These areeffector cells not only for their killing capacity but also for secreting many bioregu-latory proteins/peptides (Garcia, 1999; Kerksiek and Pamer, 1999; Rothkötter et al.,1999). It is worth stressing the extraordinary regenerative capacities of intestinalepithelium. Cells proliferate within the crypt, then migrate out and upwards to thetip of a villus, where they are shed into the lumen. It takes usually 96 to 144 h, sorenewal of the epithelium in the small intestine is possible within 4–6 days (Stokesand Bourne, 1989).

The epithelial surface is covered with a layer of mucus that creates a barrier formicroorganisms and also facilitates the digestion and absorption of different nutrients(Stokes and Bourne, 1989). The entrapment of some bacteria is due to their binding tomucosal analogues of epithelial receptors. In this way bacteria may be easily expelled

from the GI tract. However, an important factor in their virulence is the production ofmucolytic enzymes that enable bacterial colonization of the intestine epithelium.Mucus is also the best milieu for secretory IgA (sIgA), lysozyme, interferon (IFN) andother humoral factors active in hosts’ protection (Shanahan, 1997).

Another important mechanism of intestinal defence, active mostly within theduodenum and small intestine, is the bile acids (Stokes and Bourne, 1989). Theyinactivate viruses having a lipid-containing envelope and also some enteropatho-genic bacteria; however, enterococci and Bacteroides spp. can degrade bile salts.Moreover, in some species (rats, rabbits and chicken) bile is very rich in IgA andover 75% of IgA reaches the intestine by this route, whereas in ruminants, swine anddogs less than 5% of IgA enters the bile.

Briefly, the GI tract immune system is a part of MALT, which consists of lym-phoid nodules, mucosal lymphocytes, isolated lymphoid follicles and mesentericlymph nodes. The gut-associated lymphoid tissues (GALT) are the largest lymphoidorgan in the body and contain different effector mechanisms to fight potentialpathogens entering the host via food. Nodules of lymphoid tissue may lie within themucosa or span the mucosa and partially the submucosa. Peyer’s patches (PP) areorganized masses of lymphocytes and are macroscopically visible follicles in theileum. GALT may be divided into lymphoid compartments which communicatewithin a functional network (reviewed by Cebra et al., 1999). In ruminants and inpigs there are two types of PP in the small intestine. Jejunal PP are a secondary lym-phoid tissue playing an important role in intestinal defence throughout life, whereasileocaecal PP are primary lymphoid tissues responsible for the development ofcompetent immune cells and they regress in animals over 6 months of age (Stokesand Bourne, 1989; Tizard, 2000). The epithelium covering lymphoid follicles con-tains columnar absorptive cells and epithelial microfold (M) cells which transcytosemacromolecules from the lumen to underlying lymphocytes, and thus are probablythe portal of entry for viruses and bacteria. In the fundus of mucosal crypts in theintestine are Paneth cells which secrete lysozyme and peptides called cryptdins thatare related to α-defencins and are also toxic for microorganisms − lipopolysaccha-rides (LPS), lipid A, lipoteichoic acid, Gram-positive and Gram-negative bacteriamay significantly elicit cryptdin production and secretion (Eisenhauer et al., 1992;Pang et al., 1994; Ganz, 1999; Ayabe et al., 2000).

Throughout the intestines T cells (usually of CD8+ and largely γ /δ T cell recep-tor (TCR) phenotype, however, TCR α /β T cells may be found in the lamina propria)as intraepithelial lymphocytes (IELs) are spread within epithelium – these cells mayconstitute up to 27% of the epithelial cell population (Emoto et al., 1996; Wyatt et al., 1996, 1999; Marrack et al., 2000; Tizard, 2000). In ruminants as many as 90%of IELs have γ /δ TCR and they are specialized for epithelial surveillance in the GItract (McGhee et al., 1999; Tizard, 2000). Their unique property is antigen recogni-tion without previous processing. They also communicate with epithelial cells,which do express major histocompatibility complex (MHC) class II molecules,


through E-cadherin and the novel integrin αEβ7 (CD103). However, IELs do notrespond by proliferation of conventional antigens; they have the ability to producecytokines and also to kill target cells just as natural killer (NK) cells do (Lanier,2001). They are involved in regulation of IgA production by B cells; some are suppressive while others express contrasuppressive activity.

Bovine lymphocytes are composed of three subsets: i) B lymphocytes bearing IgMsurface receptors and several non-immunoglobulin (non-Ig) molecules, ii) T lympho-cytes with α/β TCR (TCR2), co-expressing either CD4 or CD8 molecules, thus repre-senting bovine CD4 (BoCD4) (T helpers), and bovine CD8 (BoCD8) (T cytotoxic/suppressors) subpopulations and iii) γ /δ TCR (TCR1) T cells (Tizard, 2000). The lastsubset dominates in young animals; in calves 40–80% of peripheral blood lympho-cytes and 35% of spleen lymphocytes represent the TCR1 phenotype. ProfessionalAPCs, dendritic cells (DC), are also present within the intestinal epithelium. Epithelialcells may express MHC class II molecules and may produce many cytokines that regulate intestine inflammatory response and also interleukin-7 (IL-7), which is agrowth factor for γ /δ T cells.

B cells and plasma cells secrete sIgA which evolved to protect mucosal surfaces(Shanahan, 1997). These cells are located in the submucosa especially in the cryptregion. Secretory IgA antibodies are responsible for immune exclusion, i.e. theyprevent adherence of microorganisms to the epithelium; however, they are not bac-tericidal and activate complement only by an alternative pathway. They neutralizeviruses and some bacterial enzymes/toxins and also opsonize particles and mayoperate in antibody-dependent cell-mediated cytotoxicity (ADCC). Recently,Macpherson et al. (2000) showed that in C57 BL/6 (H–2b) mice maintained underspecific pathogen free (SPF) conditions, sIgA against commensals were produced ina T cell independent manner. This took place within lymphoid follicles where theB cell subpopulation, derived from B1 peritoneal cells, was induced to synthesizeIgA independently from helper T (Th) cells, thus lymphoid follicular organizationrepresents a premature (primitive, less sophisticated) form of specific humoralresponse.

Another important component of local cellular defence are mast cells. Believedto be the effector cells of anti-parasitic immunity and to play a key role duringimmediate hypersensitivity reactions (Herbert et al., 1995; Miller, 1996; Smith andWeis, 1996), these cells were also found to be of significance for local immuneresponse on mucosal surfaces. There are two subpopulations of mast cells:

mucosal mast cells (MMC), of a lymphoid appearance with sparse granules,located within mucosal surface; their proliferation and differentiation is T celldependent and they are involved in anti-gastrointestinal nematode immuneresponses as effector cells,

connective mast cells (CMC) with profuse cytoplasmic granules; T cellindependent, they are apparently associated with hypersensitivity reactionsand were found within fibrotic lesions within the gastrointestinal tract.


Although MMC need interleukin 3 as a growth factor, Miller (1996) describedthe regulatory activity of these cells towards T lymphocytes, which makes MMC animportant link of a local immune network in the GI tract. During nematode invasionMMC produce, as one of their effector molecules, specific serine protease and theactivity of these cells may be quantified by the level of chymases (chymotrypsin-like proteases, MMCP-I and RMCP-II) released from their granules.

GALT should not be mobilized against food antigens mostly because immuneresponse to food may produce severe intestinal pathology to say nothing about seri-ous reduction in food uptake, absorption and metabolic utilization. Therefore theintestinal immune system has to discriminate between harmless food and potentiallyharmful pathogens. It is well accepted now that every intake of food providesenough intact proteins to easily stimulate immune responses if given by other routes.Another challenge comes from the normal microflora in the intestines and/or inforestomachs (rumen) in ruminants (Szynkiewicz, 1975; Lee, 1985; Rojas Vasquez,1996; Cebra et al., 1999; Roos, 1999).

Improper immune responses to food antigens and the harmful effect on homeo-stasis exerted by commensal GI bacteria may lead to serious pathological conditionssuch as coeliac disease and inflammatory bowel disease in humans (Abbas et al.,1994). To avoid unnecessary and dangerous specific immune response to food-derived antigens, many inhibitory/suppressory mechanisms were developed thatlead to immune tolerance. Ingested proteins during short-term contact with GALTare locally processed by antigen presenting cells (APCs) and presented in the con-text of MHC II molecules (Picker, 1992). Depending on the nature of the antigen,its quantity, the frequency of administration and the response of the host, T cellenergy, T cell deletion or antigen dose-related suppression of immune response isdeveloped. It is accompanied by production of inhibitory cytokines such as TGF-βand by inhibiting co-stimulatory signals for CD4+ T lymphocytes. Intraepitheliallymphocytes (IELs) of TcR γδ phenotype play a special role in establishing toler-ance to food antigens. On the other hand it is obvious that the active immuneresponse may also develop, including local sIgA production and generation of spe-cific and memory cellular defences. Moreover, owing to the trafficking of antigen-primed immune cells, mechanisms operating in the GI tract may lead to generationof a local immune response on other mucosal surfaces (i.e., in the respiratory tract)as well as systemic immunity. Therefore this method of stimulation was chosen forsuccessful oral vaccination.

If dangerous pathogens target the epithelial cells they secrete chemotactic/acti-vating molecules – chemokines/cytokines – that alert the surrounding immune andinflammatory cells. Chemokines are small molecules produced almost instantlyunder proinflammatory conditions and GI epithelial cells may secrete either type Ias IL-8, gro-α or ENA78 or type II chemokines as RANTES, MCP-1 and MIP-1α(Agace et al., 1993). Certainly, immune cells are major producers of cytokines thatmodulate and direct immune reactions within the GI tract. Major cytokines active



during inflammatory response are: TGF-β with IL-4 and IL-5, switching B cells inlamina propria to IgA synthesis and IL-6, inducing IgA+B cells to differentiate intosIgA-producing cells. One recently described reaction is the modulation (i.e. devel-opment and activation) of IELs with IL-7 (Fujihashi et al., 1996). It is thereforelogical that epithelial cells and IELs of the mucosal surface regulate each other tomaintain an immunological homeostasis in the GI tract. Cytokines exert theirmodulatory activity through autocrine, paracrine or endocrine pathways (Abbaset al., 1994). They are responsible for the dialogue between immune cells. It mayresult in establishing a humoral response, i.e. sIgA production or a cellular response,i.e. generation of antigen-specific CTLs (Emoto et al., 1996). A well known mecha-nism of cytokine cascade may be responsible for the clinical signs of endotoxaemia asa result of LPS challenge (Abbas et al., 1994). Activation of macrophage MΦ takesplace in the presence of IL-1, tumour necrosis factor (TNF) and IFN-γ, whereas poly-morphonuclear neutrophils (PMNs) are attracted by IL-8 (Pang et al., 1994).Cytokines also can down-regulate cellular activities, such as TGF-β, which reducesthe lymphocyte proliferation rate. Within the Th cell population a cytokine networkfunctions, i.e. IL-4 and IL-10 inhibit the growth of Th1 cells and IFN-γ decreases theproliferation of Th2 cells (Abbas et al., 1994; Tizard, 2000). Moreover, IL-1 has a directinfluence on intestinal glutamine utilization – the primary source of energy for epithelialcells and also a substrate for DNA synthesis (Austgen et al., 1992). Changed gluta-mine metabolism may result in altered permeability of the mucosa and inductionof bacterial translocation. Another cytokine, IL-6, together with IL-1 and IFN-γ, isassociated with inflammatory bowel disease in humans and it is suggested that itsrole is to induce production of autoantibodies to epithelial antigens (Powrie, 1995;Groux and Powrie, 1999).

Intestinal epithelial cells express very low levels of MHC class II proteins ontheir surface and usually do not act as APCs. However, in contact with bacterialendotoxins expression of these molecules is augmented which allows a response tobe mounted against epithelial cells, resulting in their damage. Enteropathogenicbacteria have many virulence factors, such as adhesins, toxins/enzymes and trans-missible resistance to chemotherapeutics. Amongst the exotoxins produced are: i) enterotoxins, ii) cytotoxins, iii) toxins damaging the cytoskeleton, and iv) neuro-toxins (Sears and Kaper, 1996). For example, the cytotoxic effect (CTE) of entero-pathogenic E. coli in chinese hamster ovary (CHO) cells is shown in fig. 1(Niemialtowski and Toka, 1996). These are bacterial exotoxins that act differentlyin various systems so they are classified within more than one group. Enterotoxinsmay stimulate secretory activity of intestinal epithelial cells, with virtually no dam-age to cells, whereas cytotoxins may suppress FcR-dependent phagocytosis andintracellular killing by phagocytic cells leading to their death (Niemialtowski et al.,1993; Rappuoli et al., 1999).

Changes in organization of F-actin lead to cytoskeleton damage, but this does notkill the cell. Cytotoxins, however, may inhibit migration of phagocytic cells and


diminish FcR-dependent phagocytic and bactericidal activity – they usually maycause the death of cells (Niemialtowski et al., 1993; Rappuoli et al., 1999). The neu-rotoxins are an important group produced by intestinal pathogens, which release neurotransmitters within the intestinal wall and can modulate smooth muscle activity(Sears and Kaper, 1996).

During each individual development, the pattern of immune response character-izing mature T cells depends on interactions between genetic constitution and envi-ronmental factors influencing maturation. Cytokine response of mononuclear bloodcells from newborns and the foetus indicates that differences occur when compar-ing foetal cells with adult cells. Such immature responses are characterized by a pre-ponderance of cytokines of Th2 type (Martinez et al., 1995). In mice, as in mostmammalian species there are three subsets of Th cells: Th1, which produce IFN-γ(type 1 cytokines), Th2, which synthesize IL-4 (type 2 cytokines) and Th0 of indef-inite cytokine pattern (Abbas et al., 1994; Tizard, 2000). One may suggest that astrong Th1 response in utero and immediately after birth may result in serious harmto the offspring or that trophoblasts are a source of a high level of Th1 inhibitorssuch as IL-4, IL-10, progesteron, and prostaglandins. Functional immaturity of theAPC system may also be significant in favouring Th2 early postnatal responses.Another factor, IL-12 pathway deficiency, can contribute in directing the cytokinesof newborns towards Th2 (Ridge et al., 1996). Animal experiments suggest stronglythat microorganisms are the primer for Th1 maturation and that removing suchstimuli (by creating a bacteria-free environment) locks the immune system into a

Fig. 1. Cytotoxic effect of enteropathogenicE. coli in chinese hamster ovary (CHO) cells(B) and control untreated CHO cells (A).Original magnification = 400 × (OlympusBX60 microscope; own experiments).

Th2 bias (Holt and Macaubas, 1997). The most important sources for these matura-tion signals are commensals within the GI microflora (Sudo et al., 1997). Alsoimportant seems to be the role of bacterial LPS in the maturation of APC precursorsand/or the upregulation of either bound or soluble CD14 by mononuclear cells,at least in humans. However, many pathogens modulate the expression, secretionand biological activity of host regulatory proteins, i.e. cytokines, MHC moleculesand adhesins, thus providing evidence for their delicate adaptive mechanisms ofescaping the immune recognition and response (Banks and Rouse, 1992;Niemialtowski et al., 1997; Alcami and Koszinowski, 2000; Tortorella et al., 2000).In bovines, the upregulation of IL-2Rα on B cells and MHC II expression on T andB cells by bovine leukaemia virus (BLV) infection is well documented (Torre et al.,1992; Stone et al., 1995).

Microbiota that colonize the intestinal ecosystem have evolved with their hostsfor over a 100 million years. More than 400 different species of bacteria, dividedinto indigenous, transient and permanent inhabitants, and shed from other niches,colonize the GI tract of the average animal species. This stable climax population isassociated with the intestinal epithelium and is required for IEL activation and NKcell stimulation (Berg, 1996). The majority of them are anaerobic bacteria tightlybound to the epithelial layer (reviewed in Cebra et al., 1999).

The GI tract of the newborn calf is colonized during the first hours of life byLactobacillus spp., E. coli and Enterococcus spp. and later by anaerobic bacteria ofrumen microflora from the forestomachs. As long as the calf is milk-fed, lactoferrinand κ-casein exert their bactericidal activity against coliform bacteria and favour thegrowth of Bifidobacterium spp. (reviewed by Schanbacher et al., 1997). Bovine milkis also rich in proteins that may act as probiotics helping the GI microflora togrow (Lee and Salminen, 1995; Dugas et al., 1999). Anaerobic conditions and verylow Eh (i.e. –280 mV or lower) support a friendly environment for Bacteroides spp.,Ruminococcus spp., Selenomonas spp., Lampropedia spp., Oscillospira spp.,Clostridium spp. and many more bacteria (over 30 species) that occur exclusivelywithin the rumen (Szynkiewicz, 1975; Holt et al., 1994). Maturation of the GI tracttogether with enriching microflora with cellulolytic microorganisms enable growinganimals to utilize fibre from plants and become independent of their mothers’ milk.

The passive mucosal protection of the newborn animal until weaning is depend-ent on the continuous supply of maternal antibodies and other milk proteins/peptides (reviewed by Salmon, 1999). The lactogenic humoral immunity is linkedto the GI tract (mostly to the gut) and is called the entero-mammary link, becauseof the translocation of IgA (in monogastric animals) and IgG1 (in polygastric ani-mals) or the migration of primed lymphocytes to the mammary gland. Studies onlymphocyte subsets within the mammary gland have sustained the view of a truelocal immune response, depending on the udder stage development. Accordingly,the increase of the lactogenic, i.e. maternal immunity focuses on: i) antigen priming(inducing) sites; possibly the intestines and also the mammary gland, ii) increase of


cell trafficking from the gut into the mammary gland and iii) enhancement of Ig pro-duction and secretion within the udder. A very special environment, rich incytokines and hormones is responsible for the IgM/IgA switch, the induction ofmucosal homing receptors on to lymphocytes/lymphoblasts, and the recruitment andalso the induction of mucosal vascular addressins (Brandtzaeg et al., 1999). Younganimals after weaning are able to mount an intestinal immune response and thedecreasing of the suckling period does not seem to be detrimental for its onset(Dreau et al., 1994, 1995; Salmon, 1995; Butler, 1998).

Milk bioactive proteins such as insulin-like growth factors I and II not only areimportant for GI physiology but also may modulate non-immune protection(Grosvenor et al., 1992; Zinn, 1997). Another milk protein α-lactalbumin (α-LA)(but not β-lactoglobulin, β-LG), that escapes degradation by digestive enzymeswithin the GI tract, may alter the proliferation and/or maturation of intestinal epithe-lial cells (as was shown in vitro with Caco-2 and HT-29 cell lines) (Hendriks, 1996;Alston-Mills et al., 1997). Other factors are reviewed in detail by Schanbacher et al.(1997) and by Weaver (1997).

3. IMMUNOBIOLOGY OF THE BOVINE MAMMARY GLAND IN THECONTEXT OF THE ENTERO-MAMMARY LINK

The mammary gland has evolved in all mammalian species to nourish neonatal off-spring. Lactation is the final phase of the mammalian reproductive cycle and mater-nal milk, especially in ruminants, is essential for the survival of newborns during theneonatal period by providing not only nutrients but moreover colostral maternalantibodies that protect the offspring from infections (passive immune protection).Milk of various mammalian species contains a “cocktail” of bioactive and immuno-regulatory substances, i.e. cytokines, digestive enzymes, hormones and hormonallyactive peptides (Koldovsky and Goldman, 1999), as well as anti-infectious cells(leukocytes/PMNs, MΦ, lymphocytes B and T), molecules (i.e. Ig, lysozyme, lacto-ferrin, opsonins, glycoconjugates, oligosaccharides, lipids) and pathogenic contami-nants (Weaver, 1997; Asai et al., 1998, 2000; Butler, 1999; Goldman and Ogra,1999; van Kampen et al., 1999). Dairy cows were domesticated a long time ago andthrough genetic selection their mammary gland yields far more milk than is neededto nourish their offspring. Advanced milking technologies and other factors associ-ated with intense dairy cow management have increased significantly the exposureto many potential pathogens and also profoundly affect mammary gland immunitywhich results in dramatically reduced resistance to mastitis (Hillerton et al., 1995).

The bovine mammary gland is most susceptible to invasion by differentpathogens, usually bacteria, during the period of transition from involution tocolostrogenesis and from lactation to the involution phase (Stokes and Bourne,1989; Annemüller et al., 1999). Thus, during the dry period, mammary gland phys-iology remains the fundamental factor for the production of high-quality milk


throughout the subsequent lactation. The early dry period and periparturient periodhave been considered as extremely important for the control of bovine mastitis(Hillerton, 1999). Since we cannot eliminate the selective pressure on dairy cattle toincrease production we should be aware that the only ways to combat mastitis aregenetic selection of cattle for mastitis resistance and rigorous hygiene procedures.

3.1. Development of mastitis

Mastitis is a very old problem and one of the most expensive diseases in farm ani-mals (Schalm and Lasmanis, 1968; McDonald and Anderson, 1981). The infectiousagents gain entrance to the gland via the teat canal (McDonald, 1984). Bacteria thatcause mastitis are able to overcome the bactericidal barrier of the teat canal keratinand also the physical barrier of the teat sphincter (Grindal et al., 1991). Moreoverthey can escape bacteriostatic and bactericidal factors present in the milk and are notsensitive to resident leukocytes. Many pathogens are able to adhere to epithelialcells, some may have capsule and also produce exotoxins. Enterotoxins produced by staphylococci, acting as superantigens, may modulate cytokine response, other factors may protect microorganisms within phagocytic cells and thus establish intra-cellular invasion (Almeida et al., 1999; Lee and Mekalanos, 1999).

As inflammation proceeds, with increased vascular permeability and subsequentPMN infiltration of mammary parenchyma, the secretory activity decreases. Theduration and severity of the inflammatory response within the udder have a majorimpact on the quality and quantity of produced milk. The most common pathogens thatare the cause of mastitis are staphylococci (S. aureus), streptococci (S. dysgalactiae,S. agalactiae, S. uberis), E. coli, mycoplasmas (M. agalactiae) and, less frequently,Corynebacterium bovis, Pseudomonas aeruginosa, Enterobacter aerogenes,Klebsiella pneumoniae and Actinobacillus pyogenes (Holt et al., 1994; Quinn et al.,1994; Cifrian et al., 1996; Calvinho and Oliver, 1998; Douglas et al., 2000).Furthermore, any inflammatory process that is raised within the secretory parenchymacauses damage to the parenchymal cells. Interestingly, Milner et al. (1995, 1996)showed that changes in electrical conductivity of foremilk can be used for the detec-tion of clinical mastitis caused by S. aureus and S. uberis before visible changes inmilk would occur. Routine application of this method requires automated sensorsand accurate computer software.

Bovine coliform mastitis is an example of compromised mammary gland defence,thus an insufficient local immune response is a major factor in pathogenesis (Quinnet al., 1994). The disease develops around the time of calving and is regarded as aserious problem because of its severity and difficulty to control with antibiotics for-mulated for intramammary administration. The reason why E. coli colonization is soeffective is the reduced antibacterial activity of PMNs migrating to the site of inflam-mation. They indiscriminately ingest milk constituents, release their granule content andlose the energy sources and the phagocytic activity for low opsonin concentration.



This is accompanied by elevated levels of glucocorticoids in the postpartum periodand lower chemotactic activity owing to copious milk production, so that intracellu-lar killing of phagocytosed bacteria is severely reduced. Once the infection is estab-lished, E. coli uses its own immunoregulatory strategy through endotoxin (endotoxicshock) and exotoxin (enterotoxins and other factors) production.

In staphyloccocal mastitis, however, the pathogen exhibits different mechanismsof infection (Quinn et al., 1994). S. aureus escapes mammary gland defence byenterotoxins, proteases and selective adhesin production that usually leads to severeimmunosuppression. Properly activated PMNs can assist the host in clearing theinfection but, when dysfunctional, they become a reservoir of viable bacteria thusprotecting them from antibiotics. Day et al. (2001) showed by multilocus sequencetyping (MLST), a typing system based on DNA sequences of ~450 base pairs (bp)from seven loci scattered throughout S. aureus chromosome, that a link existsbetween virulence and ecological abundance of these strains. It was found thatpathogenic strains of S. aureus represent a restricted subset within the species andare important for studying the mechanisms of mammary gland colonization, i.e. forpathogenesis of staphylococcal mastitis.

It is well documented that PMNs within the mammary gland have much lowerphagocytic activity than PMNs from peripheral blood (Targowski and Klucinski,1985; Targowski and Niemialtowski, 1986a,b, 1988). Since these cells migrateto the udder in great numbers during the first days of lactation, this diminishedantibacterial activity may be identified as an immune paradox (Targowski andKlucinski, 1985; Targowski and Niemialtowski, 1986a). It can also be accompaniedby a high level of immunoglobulins (IgG1 predominates in ruminants, whereas IgAprevails in monogastric animals) that form immune complexes (Ic) and also by anincreased number of MΦ (Butler, 1998, 1999; Tizard, 2000).

3.2. Immune paradox

Targowski and Klucinski (1985) and Targowski and Niemialtowski (1986a) demon-strated the presence of a certain factor/activity in the bovine colostrum, which waslethal for phagocytic cells – their vitality was severely reduced when compared withthe bovine milk fraction(s) (P < 0.01). In a period from 3 days to 24 h before partu-rition, PMN vitality dropped from 70 to 10%, respectively. Then, from parturitionup to day 21 of lactation, a systematic increase of viable cells up to 70% wasobserved. When comparing colostral fractions, the cytotoxic factor was found in thefat fraction. It was active against PMNs and lymphocytes isolated from bovineperipheral blood. Simultaneously conducted studies showed that the reduced phago-cytic activity of mammary gland PMNs and MΦ was also associated with saturationof their FcR with Ic that together with non-limited ingestion of fat from the mam-mary secretion, significantly diminished the ability to engulf and kill other antigens,i.e. bacteria (P < 0.01). What is important is that α-LA and β-LG may also form Ic,


thus additionally abusing phagocytic cells (Kilshaw and Slade, 1981). In normalmilk the level of Ic is significantly lower.

As was shown by the rosette test with Ig coated sensitized sheep red blood cells(SRBC), most of the colostral leukocytes have FcRs on their surface (Targowski andKlucinski, 1985; Targowski and Niemialtowski, 1986a,b; Klucinski et al., 1990;Worku et al., 1994). When isolated 3 days before parturition and consecutivelythrough the periparturient period up to day 21, the percentage of rosette formationincreased from 15 to 25% (fig. 2). Moreover, it was found that cytotoxic factor(s) is(are) produced locally and amounts may differ as the viability of phagocytic cellschanges between udder quarters. All these findings may suggest that cytotoxic activ-ity: i) is dependent on the local microenvironment, ii) is associated with the fattyacids fraction, and iii) is yet to have its source and control of synthesis described.Human female milk contains a cytostatic factor able to inhibit the proliferation ofmitogen-stimulated peripheral blood lymphocytes and that also has cytotoxic factoractivity. Colostrum-associated cytotoxicity was also found in women (Drew et al.,1984). It seems that the role of these factors is to prevent excessive maternal lymphocyte activity in the GI tract of newborns.

In human females and in cows, when colostral cytotoxic activity and phagocyticcells are overloaded with fat droplets this significantly diminishes the bactericidalactivity of colostrum and phagocytic cells, and this should be considered the mainreason for an overwhelmed local immunity and recurrent bacterial infections (fig. 3).Every immune response is orchestrated through a variety of cytokines (Butler, 1999;Koldovsky and Goldman, 1999). This prompted the idea of prevention and treatmentof bovine mastitis with exogenous administration of colony stimulating factors(CSFs) (G- and GM-CSF), interleukins (IL-1, IL-2), and IFN-γ. Their common feature is to stimulate directly phagocytic cells and/or to inhibit production of othercytokines such as TNF-α which may contribute in PMN and MΦ suppression, thusincreasing the severity of coliform mastitis (Sordillo et al., 1995).

Recently it was found that soluble Fas is present in human milk and it is assumedthat it is bound to FasL (also called CD95L or Apo-1L) on mammary cells. Thus,Fas may modulate the apoptotic machinery (programmed cell death, PCD); an

Fig. 2. Hallmark of rosette formation bybovine milk leukocyte and sensitized sheep red blood cells (SRBC). Original magnification =400 × (Olympus BX60 microscope; ownexperiments).


essential and complex regulated process for balancing cell numbers during animaldevelopment and adult homeostasis (Khler et al., 1999; Porter, 1999; Srivastavaand Srivastava, 1999; Hunt and Evan, 2001). During the dry period almost no secre-tion is produced and bovine mammary gland parenchyma is subdued in the invo-lution process which is associated with apoptosis (Wilde et al., 1997). Thepro-apoptotic microenvironment can thus favour conditions facilitating the bacterialinvasion through the teat canal. Håkansson (1999) and Håkansson et al. (1995,1999) described that the casein fraction of human milk contains multimeric α-LAwhich induces apoptosis. In contrast, the monomeric form of α-LA, which ispresent in the milk whey, does not induce apoptosis. These examples suggest thatelimination of secretory cells during mammary tissue involution is physiologicallyregulated and is due by PCD.

Thus, mammary gland and GI tract immune mechanisms play an important rolein the entero-mammary link, a bridge between GI destruction (food and microbialpathogens) and production of milk for the offspring.

4. CONCLUSIONS AND FUTURE PERSPECTIVES

This brief presentation of the MIS in the context of the entero-mammary link mayprovide the reader with a basis for understanding more complex issues involvingimmune defence against foreign antigens and the role of mucosal immunity. In thepast few years the danger of rapidly spreading animal infectious diseases such as

Fig. 3. Phagocytosis by PMNs isolated frombovine milk: (A) fat droplets (marked byarrows), and (B) S. aureus (marked by arrows).Original magnification = 400 × (OlympusBX60 microscope; own experiments).


foot-and-mouth disease, and also the threat of new transmissible diseases (like spongi-form encephalopathies) in humans and in animals, has been shown. There is also anincreasing risk of infection with antibiotic-resistant strains of bacteria (i.e. nosocomialinfections), which presents a real danger especially in the immunocompromised host.Hazardous decisions on feeding of animals with prion-containing feed undoubtedlycaused BSE, and also set up the prospect of transmission of these agents to the humanpopulation (infectious Creutzfeldt–Jakob disease/CJD/variant).

Viruses and bacteria evolve faster than the mammalian immune system may adaptand moreover they develop very effective strategies of immune evasion. We are nowfacing the real problem of an increasing number of antibiotic-resistant bacteria owingto the many years of selective pressure applied by using antibiotics in farm animalsand overprescribing in human medicine. On the other hand many new perspectivesmay arise from projects on safe, edible mucosal vaccines. Local immunity presents afield of a great interest for immunologists (Sheoran et al., 1997). Studies on mecha-nisms of apoptosis within the mammary gland and regulation of immune and neuralsystems are in progress. Recently, there have been some interesting findings on therole of agrin in establishing immunological synapses – the question on their role inmucosal immunity will hopefuly be answered in the near future.

One has to remember that we are “in a race with the replicative capacity of micro-organisms, immune responses must not be ‘too cold’ … neither ‘too hot’, … but ‘justright’ – rapid, vigorous, properly modulated, and of the correct quality” (Germain,2001). The many unresolved questions and doubts regarding the entero-mammary linkleave the door open to develop new theories and hypotheses. A better understandingof the multifactorial regulatory network of the local immune system may help uscompete more successfully with faster contestants.

ACKNOWLEDGEMENT

We express our sincere appreciation to Professor Stefan G. Pierzynowski (Institute of Cell and OrganismBiology, Lund University, and Gramineer Int. AB, Lund, Sweden) for helpful scientific discussions andfruitful long-term collaboration.

REFERENCES

Abbas, A.K., Lichtman, A.H., Pober, J.S., 1994. Cellular and Molecular Immunology. W.B. Saunders,Philadelphia.

Agace, W.W., Hedges, S.R., Ceska, M., Svanborg, C., 1993. Interleukin-8 and the neutrophil response tomucosal gram-negative infection. J. Clin. Invest. 92, 780–785.

Alcami, A., Koszinowski, U.H., 2000. Viral mechanisms of immune evasion. Immunol. Today 21, 447–455.Almeida, R.A., Calvinho, L.F., Oliver, S.P., 1999. Influence of protein kinase inhibitors on Streptococcus

uberis internalization into bovine mammary epithelial cells. Microb. Path. 28, 9–16.Alston-Mills, B., Allen, J.C., Sternhagen, L., Hepler, C.D., 1997. Effects of whey-milk proteins on Caco-2

and HT-29 intestinal cell lines. Livest. Prod. Sci. 50, 147–148.Annemüller, C., Lämmler, Ch., Zschöck, M., 1999. Genotyping of Staphylococcus aureus isolated from

bovine mastitis. Vet. Microbiol. 69, 217–224.


Asai, K., Kai, K., Rikiishi, H., Sugawara, S., Maruyama, Y., Yamaguchi, T., Ohta, M., Kumagai, K., 1998.Variation in CD4+ T and CD8+ T lymphocyte subpopulations in bovine mammary gland secretionsduring lactating and non-lactating periods. Vet. Immunol. Immunopathol. 65, 51–61.

Asai, K., Komine, Y., Kozutsumi, T., Yamaguchi, T., Komine, K., Kumagai, K., 2000. Predominant subpopulations of T lymphocytes in the mammary gland secretions during lactation and intraepithe-lial T lymphocytes in the intestine of dairy cows. Vet. Immunol. Immunopathol. 73, 233–240.

Austgen, T.R., Chen, M.K., Dudrick, P.S., Copeland, E.M., Souba, W.W., 1992. Cytokine regulation ofintestinal glutamine utilization. Amer. J. Surg. 163, 174–179.

Ayabe, T., Satchell, D.P., Wilson, C.L., Parks, W.C., Selsted, M.E., Ouellette, A.J., 2000. Secretion of microbicidal α-defensins by intestinal Paneth cells in response to bacteria. Nature Immunol. 1, 113–118.

Bagglioni, M., Loetscher, P., 2000. Chemokines in inflammation and immunity. Immunol. Today 21,418–420.

Banks, T.A., Rouse, B.T., 1992. Herpesviruses – immune escape artists? Clin. Inf. Dis. 14, 933–941.Bell, J.I., 1998. Molecular anatomy of the immune system. Immunol. Rev. 163, 5–10.Berg, R.D., 1996. The indigenous gastrointestinal microflora. Trends Microbiol. 4, 430–435.Brandtzaeg, P., Farstad, I.N., Haraldsen, G., 1999. Regional specialization in the mucosal immune

system: primed cells do not always home along the same track. Immunol. Today 20, 267–276.Butler, J.E., 1998. Immunoglobulin diversity, B-cell and antibody repertoire development in large farm

animals. Rev. Sci. Tech. 17, 43–70.Butler, J.E., 1999. Immunoglobulins and immunocytes in animal milks. In: Ogra, P.L., Lamm, M.E.,

Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology, 2nd edition.Academic Press, San Diego, pp. 1531–1554.

Calvinho, L.F., Oliver, S.P., 1998. Characterization of mechanisms involved in uptake of Streptococcusdysgalactiae by bovine mammary epithelial cells. Vet. Microbiol. 63, 261–274.

Cebra, J.J., Jiang, H.-Q., Sterzl, J., Tlaskalová-Hogenová, H., 1999. The role of mucosal microbiota inthe development and maintenance of the mucosal immune system. In: Ogra, P.L., Lamm, M.E.,Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology, 2nd edition.Academic Press, San Diego, pp. 267–280.

Cifrian, E., Guidry, A.J., Bramley, A.J., Norcross, N.L., Bastida-Corcuera, F.D., Marquardt, W.W., 1996.Effect of staphylococcal β toxin on the cytotoxicity, proliferation and adherence of Staphylococcusaureus to bovine mammary epithelial cells. Vet. Microbiol. 48, 187–198.

Day, N.P.J., Moore, C.E., Enright, M.C., Berendt, A.R., Smith, J.M., Murphy, M.F., Peacock, S.J.,Spratt, B.G., Feil, E.J., 2001. A link between virulence and ecological abundance in natural popula-tions of Staphylococcus aureus. Science 292, 114–116.

Dreau, D., Lalles, J.P., Philouzerome, V., Toullec, R., Salmon, H., 1994. Local and systemic immune-responses to soyabean protein ingestion in early-weaned pigs. J. Anim. Sci. 72, 2090–2098.

Dreau, D., Lalles, J.P., Lejan, C., Toullec, R., Salmon, H., 1995. Hypersensitivity to soyabean proteins inearly weaned piglets: humoral and cellular components. Adv. Exp. Med. Biol. 371B, 865−869.

Drew, P.A., Petrucco, O.M., Sheraman, D.J., 1984. A factor present in human milk, but not colostrum,which is cytotoxic for human lymphocytes. Clin. Exp. Immunol. 55, 437–443.

Douglas, V.L., Fenwick, S.G., Pfeiffer, D.U., Williamson, N.B., Holmes, C.W., 2000. Genomic typing ofStreptococcus uberis isolates from cases of mastitis, in New Zealand dairy cows, using pulsed-fieldgel electrophoresis. Vet. Microbiol. 75, 27–41.

Dugas, B., Mercenier, A., Lenoir-Wijnkoop, I., Arnaud, C., Dugas, N., Postaire, E., 1999. Immunity andprobiotics. Immunol. Today 20, 387–390.

Eisenhauer, P.B., Hartwig, S.S.L., Lehrer, R.I., 1992. Cryptdins: antimicrobial defensins of the murinesmall intestine. Infect. Immun. 60, 3556–3565.

Emoto, M., Neuhaus, O., Emoto, Y., Kaufmann, S.H.E., 1996. Influence of β2-microglobulin expressionon gamma-interferon secretion and target cell lysis by intraepithelial lymphocytes during intestinalListeria monocytogenes infection. Infect. Immun. 64, 569–575.

Fujihashi, K., Yamamoto, M., McGhee, J.R., Nishikawa, S.I., Kijono, H., 1996. Interleukin 2 (IL-2) andinterleukin 7 (IL-7) and IL-2 receptors on T cell receptor positive intraepithelial lymphocytes. Proc.Natl. Acad. Sci. USA 93, 3613–3618.

Ganz, T., 1999. Defensins and host defense. Science 286, 420–421.


Ganz, T., 2000. Paneth cells – guardians of the gut cell hatchery. Nature Immunol. 1, 99–100.Garcia, K.C., 1999. Structural basis of T cell recognition. Annu. Rev. Immunol. 17 369–397.Germain, R.N., 2001. The art of the probable: system control in the adaptive immune system. Science

293, 240–245.Goldman, A.S., Ogra, P.L., 1999. Anti-infectious and infectious agents in human milk. In: Ogra, P.L.,

Lamm, M.E., Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology,2nd edition. Academic Press, San Diego, pp. 1511–1521.

Green, W.B., Eaton, K., Krakowka, S., 1997. Porcine gastric mucosa associated lymphoid tissue(MALT): stimulation by colonization with the gastric bacterial pathogen, Helicobacter pylori. Vet. Immunol. Immunopathol. 56, 119–131.

Grindal, R.J., Walton, A.W., Hillerton, J.E., 1991. Influence of milk flow rate and streak canal length onnew intramammary infection in dairy cows. J. Dairy Res. 58, 383–388.

Grosvenor, C.E., Picciano, M.F., Baumrucker, C.R., 1992. Hormones and growth factors in milk.Endocrine Rev. 14, 710–728.

Groux, H., Powrie, F., 1999. Regulatory T cells and inflammatory bowel disease. Immunol Today 20,442–445.

Håkansson, A., 1999. Apoptosis induced by a human milk protein complex. PhD Thesis. LundUniversity, Sweden.

Håkansson, A., Zhivotovsky, B., Orrenius, S., Sabharwal, H., Svanborg, C., 1995. Apoptosis induced bya human milk protein. Proc. Natl. Acad. Sci. USA 92, 8064–8068.

Håkansson, A., Andréasson, J., Zhivotovsky, B., Karpman, D., Orrenius, S., Svanborg, C., 1999.Multimeric α-lactalbumin from human milk induces apoptosis through a direct effect on cell nuclei.Exp. Cell Res. 246, 451–460.

Hendriks, H.G., 1996. Effects of legume lectins on structure and function of Caco-2 cells. PhD Thesis.University of Utrecht Press, pp. 1–111.

Herbert, W.J., Wilkinson, P.C., Stott, D.I., 1995. The Dictionary of Immunology, 4th edition. AcademicPress, London.

Hermann, A., 2000. Intestinal mucins “soluble and insoluble problems”. PhD Thesis. Lund University,Sweden.

Hillerton, J.E., 1999. Balancing mastitis and quality. Proc. British Mastitis Conf. Institute for AnimalHealth, Compton, UK, pp. 31–36.

Hillerton, J.E., Bramley, A.J., Staker, R.T., McKinnon, C.H., 1995. Patterns of intramammary infectionand clinical mastitis over a 5 year period in a closely monitored herd applying mastitis control meas-ures. J. Dairy Res. 62, 39–50.

Holmgren, J., Rudin, A., 1999. Mucosal immunity and bacteria. In: Ogra, P.L., Lamm, M.E.,Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology, 2nd edition.Academic Press, San Diego, pp. 685–693.

Holt, P.G., Macaubas, C., 1997. Development of long-term tolerance versus sensitization to environ-mental allergens during the perinatal period. Curr. Opin. Immunol. 9, 782–787.

Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Bergey’s DeterminativeBacteriology, 9th edition. Williams & Wilkins, Baltimore.

Hunt, A., Evan, G., 2001. Till death us do part. Science 293, 1784–1785.Johnson, R.T., 1998. Viral Infections of the Nervous System, 2nd edition. Lippincott-Raven, Philadelphia.Kaganoff, M.F., 1996. Mucosal immunology: new frontiers. Immunol. Today 17, 57–59.Kerksiek, K.M., Pamer, E.G., 1999. T cell responses to bacterial infection. Curr. Opin. Immunol. 11,

400–405.Khan, A.A., Rose, C., Yam, L.S., Soloski, M.J., Rupp, F., 2001. Physiological regulation of the immuno-

logical synapse by agrin. Science 292, 1681–1686.Khler, C., Håkansson, A., Svanborg, C., Orrenius, S., Zhivotovsky, B., 1999. Protease activation in apop-

tosis induced by MAL. Exp. Cell Res. 249, 260–268.Kilshaw, P.J., Slade, H., 1981. Milk protein immune complexes in the cow and calf. J. Reprod. Immunol.

3, 227–236.Klucinski, W., Degórski, A., Miernik-Degórska, E., Targowski, S., Winnicka, A., 1988. Effect of ketone

bodies on the phagocytic activity of bovine milk macrophages and polymorphonuclear leukocytes.J. Vet. Med. A. 35, 626.


Klucinski, W., Niemialtowski, M., Winnicka, A., Degórski, A., Gonzales-Gonzales, M., 1990. Thephagocytic activity of neutrophil granulocytes isolated from blood, mammary gland and uterus ofcows. Pol. Arch. Wet. 30, 89–98.

Koldovsky, O., Goldman, A.S., 1999. Growth factors and cytokines in milk. In: Ogra, P.L., Lamm, M.E.,Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology, 2nd edition.Academic Press, San Diego, pp. 1523–1530.

Lanier, L.L., 2001. On guard – activating NK cell receptors. Nature Immunol. 2, 23–27.Lee, A., 1985. Neglected niches: the microbial ecology of the gastrointestinal tract. In: Marshall, K.C.

(Ed.), Advances in Microbial Ecology. Plenum Press, London, pp. 115–162.Lee, C., Mekalanos, J., 1999. Bacterial interactions with intestinal epithelial cells. In: Ogra, P.L.,

Lamm, M.E., Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology,2nd edition. Academic Press, San Diego, pp. 657–669.

Lee, Y.-K., Salminen, S., 1995. The coming age of probiotics. Trends Food Sci. Technol. 6, 241–245.Luciw, P.A., 1996. Human immunodeficiency viruses and their replication. In: Fields, B.N., Knipe, D.M.,

Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., Straus, S.E. (Eds.),Fundamental Virology. Lippincott-Raven, Philadelphia and New York, pp. 845–916.

Macpherson, A.J., Gatto, D., Sainsbury, E., Harriman, G.R., Hengartner, H., Zinkernagel, R.M., 2000.A primitive T cell-independent mechanism of intestinal mucosal IgA responses to commensal bacteria.Science 288, 2222–2226.

Marrack, P., Bender, J., Hildeman, D., Jordan, M., Mitchell, T., Murakami, M., Sakamoto, A., Schaefer, B.C.,Swanson, B., Kappler, J., 2000. Homeostasis of αβTCR+ T cells. Nature Immunol. 1, 107–111.

Martinez F.D., Stern, D.A., Wright, A.L., Holberg, C.J., Taussig, L.M., Halonen, M., 1995. Association ofinterleukin-2 and interferon-gamma production by blood mononuclear cells in infancy with parentalallergy skin tests and with subsequent development of atopy. J. Allergy Clin. Immunol. 96, 652–660.

McDonald, J.S., 1984. Streptococcal and staphylococcal mastitis. Vet. Clin. North America: LargeAnimal Practice 6, 305–321.

McDonald, J.S., Anderson, A.J., 1981. Total and differential somatic cell counts in secretions from non-infected bovine mammary glands: the early nonlactating period. Amer. J. Vet. Res. 42, 1360–1365.

McGhee, J.R., Lamm, M.E., Strober, W., 1999. Mucosal immune responses. An overview. In: Ogra, P.L.,Lamm, M.E., Bienenstock, J., Mestecky, J., Strober, W., McGhee, J.R. (Eds.), Mucosal Immunology,2nd edition. Academic Press, San Diego, pp. 485–506.

Miller, H.R.P., 1996. Mucosal mast cells and the allergic response against nematode parasites. Vet.Immunol. Immunopathol. 54, 331–336.

Milner, P., Page, K.L., Walton, A.W., Hillerton, J.E., 1995. Detection of clinical mastitis by changes inelectrical conductivity of foremilk before visible changes in milk. J. Dairy Sci. 79, 83–86.

Milner, P., Page, K.L., Hillerton, J.E., 1996. The effects of early antibiotic treatment following diagnosisof mastitis by a change in the electrical conductivity of milk. J. Dairy Sci. 80, 859–863.

Niemialtowski, M.G., Klucinski, W., Malicki, K., Spohr de Faundez, I., 1993. Cholera toxin (cholera-gen)–polymorphonuclear leukocyte interactions: effect on migration in vitro and FcR-dependentphagocytic and bactericidal activity. Microbiol. Immunol. 37, 55–62.

Niemialtowski, M., Toka, F.N., 1996. Enteric bacteria and their toxins – from the microscope to molec-ular biology. In: Niemialtowski, M., Wedrychowicz, H. (Eds.), Proceedings of the Conference onInfectious Diseases of the Alimentary Tract of Human and Animals. Kazimierz Dolny n. Wisla, 24−26 October. Warsaw Argicultural University Press, pp. 82–95.

Niemialtowski, M.G., Toka, F.N., Malicka, E., Spohr de Faundez, I., Gierynska, M., Schollenberger, A.,1997. Immune escape of orthopoxviruses. Rev. Med. Virol. 7, 35–47.

Nordman, H., 2000. Gastric mucins – molecular characterisation and interactions with Helicobacterpylori. PhD Thesis, Lund University, Sweden.

Pang, G., Ciuch, L., Batey, R., Clancy, R., Cripps, A., 1994. GM-csf, IL-1α, IL-1β, IL-6, IL-8, IL-10,ICAM-1 and VCAM-1 gene expression and cytokine production in human duodenal fibroblasts stim-ulated by lipopolisaccharide, IL-1α and TNF-β. Clin. Exp. Immunol. 96, 437–443.

Picker, L.J., 1992. Mechanisms of lymphocyte homing. Curr. Opin. Immunol. 4, 277–286.Porter, A.G., 1999. Protein translocation in apoptosis. Cell Biol. 9, 394–401.Powrie, F., 1995. T cells in inflammatory bowel disease: protective and pathogenic roles. Immunity 3,

171–174.


Prusiner, S., 1996. Prions. In: Fields, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L.,Monath, T.P., Roizman, B., Straus, S.E. (Eds.), Fundamental Virology. Lippincott-Raven,Philadelphia and New York, pp. 1245–1294.

Quinn, P.J., Carter, M.E., Markey, B., Carter, G.R., 1994. Clinical Veterinary Microbiology. Wolfe Publ.Mosby-Year Book Europe Limited London.

Rappuoli, R., Pizza, M., Douce, G., Dougan, G., 1999. Structure and mucosal adjuvanticity of choleraand Escherichia coli heat-labile enterotoxins. Immunol. Today 20, 493–500.

Ridge, J.P., Fuchs, E.J., Matzinger, P., 1996. Neonatal tolerance revisited: turning on newborn T cellswith dendritic cells. Science 271, 1723–1726.

Rojas Vasquez, M., 1996. Studies on an adhesion promoting protein from Lactobacillus and its role incolonization of the gastrointestinal tract. Doctoral thesis. Göteborg University, Sweden.

Roos, S., 1999. Adhesion and autoaggregation of Lactobacillus reuteri and description of a newLactobacillus species with mucus binding properties. Doctoral thesis. Universitatis AgriculturaeSueciae, Agraria 148, Swedish University of Agricultural Sciences, Uppsala.

Rothkötter, H.J., Pabst, R., Bailey, M., 1999. Lymphocyte migration in the intestinal mucosa: entry, transitand emigration of lymphoid cells and the influence of antigen. Vet. Immunol. Immunopathol. 72,157–165.

Salmon, H., 1995. Lactogenic immunity and vaccinal protection in swine. Vet. Res. 26, 232–237.Salmon, H., 1999. The mammary gland and neonate mucosal immunity. Vet. Immunol. Immunopathol.

72, 143–155.Savage, D.C., 1977. Microbial ecology of the gastrointestinal tract. Annu. Rev. Microbiol. 31, 107–133.Schalm, O.W., Lasmanis, J., 1968. The leukocytes: origin and function in mastitis. J. Amer. Vet. Med.

Assoc. 153, 1688–1694.Schanbacher, F.L., Talhouk, R.S., Murray, F.A., 1997. Biology and origin of bioactive peptides in milk.

Livest. Prod. Sci. 50, 105–123.Sears, C.L., Kaper, J.B., 1996. Enteric bacterial toxins: mechanisms of action and linkage to intestinal

secretion. Microbiol. Rev. 60, 167–215.Shanahan, F., 1997. The gastrointestinal immune system. In: Sipes, I.G., McQueen, C.A., Gandolfi, A.J.

(Editors-in-Chief), Comprehensive Toxicology. Vol. 9. McCuskey, R.S., Earnest, D.L. (Eds.), Hepaticand Gastrointestinal Toxicology. Pergamon, Cambridge, pp. 539–548.

Sheoran, A.S., Sponseller, B.T., Holmes, M.A., Timoney, J.F., 1997. Serum and mucosal isotyperesponses to M-like protein (SeM) of Streptococcus equi in convalescent and vaccinated horses. Vet.Immunol. Immunopathol. 59, 239–251.

Smith, T.J., Weis, J.H., 1996. Mucosal T cells and mast cells share common adhesion receptors.Immunol. Today 17, 60–63.

Sordillo, L.M., Pighetti, G.M., Davis, M.R., 1995. Enhanced production of bovine tumor necrosis factor-αduring the periparturient period. Vet. Immunol. Immunopathol. 49, 263–270.

Srivastava, M.D., Srivastava, B.I., 1999. Soluble Fas and soluble Fas ligand proteins in human milk:possible significance in the development of immunological tolerance. Scand. J. Immunol. 49, 51–54.

Stokes, C., Bourne, J.F., 1989. Mucosal immunity. In: Halliwell, R.E.W., Gorman, N.T. (Eds.), VeterinaryClinical Immunology. W.B. Saunders, Philadelphia, pp. 164–231.

Stone, D.M., Hof, A.J., Davis, W.C., 1995. Up-regulation of IL-2 receptor α and MHC class II expres-sion on lymphocyte subpopulations from bovine leukemia virus infected lymphocytotic cows. Vet.Immunol. Immunopathol. 48, 65–76.

Sudo, N., Sawamura, S., Tanaka, K., Aiba, Y., Kubo, C., Koga, Y., 1997. The requirement of intestinalbacterial flora for the development of an IgE production system fully susceptible to oral toleranceinduction. J. Immunol. 159, 1739–1745.

Szynkiewicz, Z.M., 1975. Autochthonic microflora of animals (in Polish). In: Szynkiewicz, Z.M. (Ed.),Microbiology. PWN, Warszawa, pp. 488–521.

Targowski, S.P., Klucinski, W., 1985. Effect of immune complexes from mastitic milk on blocking of Fcreceptors and phagocytosis. Infect. Immun. 47, 484–488.

Targowski, S.P., Niemialtowski, M., 1986a. Inhibition of lacteal leukocyte phagocytosis by colostrum,non-lactating secretion, and mastitic milk. Amer. J. Vet. Res. 47, 1940–1945.


Targowski, S.P., Niemialtowski, M., 1986b. Appearance of Fc receptors on poly-morphonuclear leuko-cytes after migration and their role in phagocytosis. Infect. Immun. 52, 798–802.

Targowski, S.P., Niemialtowski, M., 1988. Cytotoxic and blocking effect of bovine colostrum. J. Vet.Med. B, 35, 96–104.

Tizard, I.R., 2000. Veterinary Immunology. An Introduction, 6th edition. W.B. Saunders, Philadelphia.Torre, P.M., Konur, P.K., Oliver, S.P., 1992. Proliferative response of mammary gland mononuclear cells

to recombinant bovine interleukin-2. Vet. Immunol. Immunopathol. 32, 351–358.Tortorella, D., Gewurz, B.E., Furman, M.H., Schust, D.J., Ploegh, H.L., 2000. Viral subversion of the

immune system. Annu. Rev. Immunol. 18, 861–926.Trautmann, A., Vivier, E., 2001. Agrin – a bridge between the nervous and immune systems. Science 292,

1667–1668.Van Kampen, C., Mallard, B.A., Wilkie, B.N., 1999. Adhesion molecules and lymphocyte subsets in milk

and blood of periparturient Holstein cows. Vet. Immunol. Immunopathol. 69, 23–32.VanCott, J.L., Yamamoto, S., McGhee, J.R., 2000. Mucosal immunity, In: Cunningham, M.W., Fujinami, R.S.

(Eds.), Effects of Microbes on the Immune System. Lippincott Williams & Wilkins, Philadelphia,pp. 233–250.

Voynow, J.A., Rose, M.C., 1994. Quantitation of mucin mRNA in respiratory and intestinal epithelialcells. Amer. J. Respir. Cell. Mol. Biol. 11, 742–750.

Weaver, L.T., 1997. Significance of bioactive substances in milk to the human neonate. Livest. Prod. Sci.50, 139–146.

Wilde, C.J., Quarrie, L.H., Tonner, E., Flint, D.J., Peaker, M., 1997. Mammary apoptosis. Livest. Prod.Sci. 50, 29–37.

Worku, M., Paape, M.J., Filep, R., Miller, R.H., 1994. Effect of in vitro and in vivo migration of bovineneutrophils on binding and expression of Fc receptors for IgG2 and IgM. Amer. J. Vet. Res. 55,221–226.

Wyatt, C.R., Brackett, E.J., Perryman, L.E., Davis, W.C., 1996. Identification of γδ lymphocyte subsetsthat populate calf ileal after birth. Vet. Immunol. Immunopathol. 52, 91–103.

Wyatt, C.R., Barrett, W.J., Brackett, E.J., Davis, W.C., Besser, T.E., 1999. Phenotypic comparison of ilealintraepithelial lymphocyte populations of suckling and weaned calves. Vet. Immunol. Immunopathol.67, 213–222.

Zinkernagel, R.M., 2000. What is missing in immunology to understand immunity? Nature Immunol. 1,181–185.

Zinn, S.A., 1997. Bioactive components in milk: introduction. Livest. Prod. Sci. 50, 101–103.

314

Newly hatched fish do not have a functionally developed specific immune system andprotection against bacterial infection is provided by non-specific mechanisms and fac-tors, some of which may be derived from maternal origins. There may be scope foroptimizing the transfer of these factors from mother to offspring or for improving theability of larvae to synthesize non-specific defence factors by manipulation of dietaryor environment parameters of the mother prior to ovulation. In the larvae, lympho-cytes first appear in the thymus and T cells differentiate before B cells, which developfirst in the kidney. The gut becomes populated with T and B cells quite early in devel-opment, with T cells being present in the mucosal epithelium and B cells in the lam-ina propria. The ability to produce specific immune responses begins at the time offirst feeding in fish such as salmonids with large eggs and larvae, but in many marinefish species, with small eggs and larvae, there is some time between the onset of feed-ing and functional maturation of the specific immune response. This appears to be atime when the fry are very susceptible to bacterial infection. Furthermore, there is evi-dence that if exposure to some types of antigen occurs during this period, a state ofimmunological suppression can be induced. Thus, there are important consequencesfor the earliest time to vaccinate fish. The onset of maturation of the specific immuneresponse relates to the rate of general differentiation of the tissues and this is temper-ature related in fish. For a given species there is a particular degree/days relationshipgoverning this differentiation. Thereafter, the magnitude of antibody responsesincreases with increasing age of the fish until maximum levels are achieved. This maytake several months but is different for different species.

14 Development of the immune response in relation to bacterial disease in thegrowing fish

A.E. Ellis

Marine Laboratory, Victoria Road, Aberdeen AB11 9DB, UK


Immune response in growing fish 315

1. INTRODUCTION

Different fish species hatch at different stages of development but in most (if not allspecies studied) the newly hatched fry do not have a functionally developed specificimmune system. Protection against diseases must, therefore, be based on non-specific mechanisms, some of which may initially be derived from the mother.Transfer of specific antibody to the egg from the mother does occur but in mostspecies this does not appear to be in sufficient amounts for protection. The lymphoidsystem usually develops some time after hatching and takes further time to becomefunctionally mature. There is some evidence that antigenic exposure of very youngfry may result in immunological tolerance. Thus, knowledge of the functional devel-opment of the immune system of fish is important for the strategic use of vaccines.This chapter will review this knowledge, especially in relation to defences againstbacterial diseases.

2. MATERNAL INFLUENCES

Little is known about the influence of maternal investment in protection of eggs andlarvae of fish, especially with respect to the physiological status of the mother at thetime of egg development in the ovary. However, it is known that the ova of manyspecies of fish contain substances that are considered to play important roles indefence against bacterial infections. These include C-reactive protein (CRP), lectinsand lysozyme (Ellis, 1999) which are present in fairly high concentration and pre-sumably are selectively incorporated into the yolk. Lysozyme levels have been meas-ured in the embryos and larvae of sea bass and these were elevated in embryos (48and 72 h after fertilization) and larvae (24 h after hatching) originating from mothersfed a diet enriched in vitamin C for 1 month prior to spawning (Cecchini et al., 2000).

Immunoglobulin (Ig) has been detected in the ova of fish but at very low con-centration (3–12 μg/g egg weight, compared to about 12 mg/ml serum of adult fish;Scapigliati et al., 1999) indicating that active secretion of Ig into fish ova is absent.Attempts to protect larval Atlantic salmon against Enteric Redmouth by immuniz-ing mothers prior to spawning were unsuccessful although some specific antibodywas detected in the eggs (Lillehaug et al., 1996).

It is interesting that newly hatched fry appear to be much more resistant to chal-lenge by bacteria than older fry. Rainbow trout fry were completely resistant toimmersion challenge by Vibrio anguillarum at 2 and 4 weeks post-hatch while 20%mortality was obtained in fry of 6 weeks post-hatch (0.2−0.3 g) and over (Tatner andHorne, 1983). Similarly, when Atlantic salmon fry were challenged with Yersiniaruckeri 2 weeks after hatching, 8% mortality occurred, rising to 60% mortality whenchallenged 6 weeks post-hatch onwards (Lillehaug et al., 1996). This occurred inoffspring of both immunized and non-immunized mothers, suggesting that non-specific factors of maternal origin were responsible and were depleted with time.

316 A.E. Ellis

The situation in mouth-brooding fish appears to be different. In these species,relatively small numbers of eggs are produced and the larvae are carefully attendedby the parents. In the tilapia Oreochromis aureus, mothers that were vaccinatedagainst Ichthyophthirius multifiliis (a protozoan parasite) 4 weeks before spawning,produced offspring that were highly protected against challenge following hatching(10 days after spawning). It was further shown that this protection was not only pas-sively transferred from the mother via antibodies in the eggs but also enhanced viathe mucus secretions in the mouth during mouth-brooding (Sin et al., 1994). In thesespecies it may be possible to effectively protect the fry against diseases before theirspecific immune system is fully developed by vaccination of the mothers.

3. NON-SPECIFIC DEFENCE FACTORS OF ENDOGENOUS ORIGIN IN FISH FRY

Little is known about the expression of non-specific defence factors in fish ontogenybut recently the initiation of expression of pleurocidin (an antibacterial peptide) inthe fry of winter flounder has been demonstrated 13 days post-hatch (dph) (Douglaset al., 2001). In the pre-larvae of tilapia, lysozyme has been detected immunohisto-chemically in the epithelial cells of the primordial swimbladder and the hepatocytessuggesting these organs produced lysozyme before the haemopoietic tissue devel-oped in the kidney (Takemura, 1996).

4. DIFFERENTIATION OF THE PHAGOCYTIC SYSTEM

The development of this system in rainbow trout has been studied followingintraperitoneal (ip) injection of carbon particles (Tatner and Manning, 1985). Themajor tissue sites containing phagocytic cells in trout of different ages are shown intable 1. In the fry, phagocytic cells are mainly present in the integument, particularlythe gills. As the fish ages, the kidney and spleen become the major sites of phago-cytic cells coinciding with the progressively developing lymphoid tissue in theseorgans. In the carp, the use of monoclonal antibodies specific for monocytes/macrophages, indicates that these cells differentiate very early, possibly in the yolksac, before other leucocytes and before the lymphoid organs, being detectable by 1 week post-fertilization (Romano et al., 1997).

Table 1. Development of the phagocytic system in tissues of the rainbow trout: days after hatching

Age Gill Skin Gut Kidney Spleen

4 days ++ + + − −18 days ++ − − ++ +8 months − − − ++ ++

Data from Tatner and Manning, 1985.


5. DIFFERENTIATION OF LYMPHOID ORGANS

The larvae of different fish species hatch at different stages of development of theirvarious organ systems and lymphocytes differentiate within the lymphoid organs atdifferent times relative to hatching. For example, in Atlantic salmon, the thymus andkidney are fully lymphoid at the time of hatching, while lymphocytes do not appearin lymphoid organs of some other species, e.g. sea bream and plaice, until 6 weeksafter hatching. Nevertheless, the order in which lymphocytes first appear in the lym-phoid organs is the same for all species studied. Lymphocytes first appear in the thy-mus followed by the blood and pronephros and lastly, after some considerable delay,in the spleen (table 2).

The origin of the thymic stem cells is not known, but most workers have identi-fied haemopoietic blast cells in the kidney prior to differentiation of lymphocytes inthe thymus (Ellis, 1977; Razquin et al., 1990; Josefsson and Tatner, 1993) and it isbelieved that the thymic rudiment is initially populated by stem cells from the yolksac or pronephros. Using in situ hybridization, genes associated with haemopoiesis(e.g., GATA-1, c-myb) are first expressed in zebra fish in the yolk sac, 12 h post-fertilization (hpf), and definitive haemopoiesis is established in the kidney by day 5when Scl/tal-1 is first detected. Lymphocyte specific gene markers (e.g., Rag-1,Rag-2) and the T cell specific lck gene have been found to be expressed in the zebrafish thymus as early as 68 hpf (Trede and Zon, 1998).

Following proliferation of lymphocytes in the thymus, many lymphocytes can be observed in the connective tissue separating the thymus from the pronephros,

Table 2. Appearance of lymphocytes in the developing lymphoid organs of fish: days prior to (−)or post (+) hatching

Size when lymphocytes first appear in

Thymus Blood Pronephros Spleen thymus (mm) References

Plaice +28 NK +49 NK 8 Lele, 1933Salmon −22 −14 −14 +42 NK Ellis, 1977Rainbow +1–3 +5 +5 +14–25 17 Grace and

trout Manning, 1980;Razquin et al., 1990

Carp +5 +7–8 +7–8 +8–9 4 Botham andManning, 1981

Sebasticus +21 NK +30 +44 NK Nakanishi, 1991marmoratus*

Sea bream +47 +54 +54 >+77 6 Josefsson andTatner, 1993

Sea bass +21–27 NK +30 +44 NK Abelli et al., 1996;Breuil et al., 1997

*Days post-birth.NK, not known.

318 A.E. Ellis

leading to the suggestion that lymphocytes migrate from the thymus to populate thepronephros (Tatner, 1985; Josefsson and Tatner, 1993; Abelli et al., 1996).

The thymus in fish first develops as separate buds situated dorsal to several gillarches. These later coalesce into a single organ in each branchial cavity. The thymicbuds develop as collections of lymphocytes between the pharyngeal epithelial cells andthe epidermal basement membrane (Ellis, 1977). In young fish the fully differentiatedthymus is separated from the external environment only by a single layer of simpleepithelial cells. However, this still provides an effective barrier to entry of antigens intothe thymic tissue from the environment (Castillo et al., 1998). In older fish the epithe-lium thickens and the organ progressively becomes encapsulated in connective tissue.

During the first few months there is intense mitotic activity of lymphocytes in thethymus and then a decrease. Apoptosis of thymocytes has been reported, with anonset in carp at about 4 weeks of age (Romano et al., 1999) and sea bass at 5 weeks(Abelli et al., 1998). It is believed that many thymocytes migrate to peripheralorgans during the first 2–3 months post-hatching and signs of involution of the thymus appear with the onset of sexual maturation. Histologically, lymphocytesappear in the kidney and blood a few days after their appearance in the thymus whilesplenic lymphocytes appear a considerable time later (table 2).

6. DEVELOPMENT OF T LYMPHOCYTES

At present monoclonal antibodies (Mabs) specific for mature T lymphocytes haveonly been developed for sea bass (Scapigliati et al., 1995, 2000). Mabs have alsobeen produced which are specific for early thymocytes in the carp (Rombout et al.,1997) or a carp mucosal T cell subpopulation (Rombout et al., 1998). These Mabsprovide evidence for initial differentiation of T lymphocytes within the thymus fol-lowed by their migration to the gut mucosa, kidney and spleen. In sea bass, T cellsfirst appeared in the thymus at 30 dph, 3 days after the first appearance of lymphoidcells, shortly after in the epithelium of the gut mucosa, and then in the kidney (35 dph) and spleen (44 dph). Throughout development, T cells were very numer-ous in the thymus and increased markedly in the intestinal mucosa from 44 dphonward while they remained infrequent in the developing kidney and spleen (Abelliet al., 1996; Picchietti et al., 1997). Using flow cytometry of larval homogenates,cells with T cell determinants have been detected in sea bass as early as 12 dph,which suggests they may originate in an extrathymic compartment as the thymus isnot lymphoid at this stage (dos Santos et al., 2000). Adult levels of T cells in seabass are attained by about 140 dph (dos Santos et al., 2000).

7. DEVELOPMENT OF B LYMPHOCYTES

A polyclonal antiserum to Atlantic salmon serum Ig, which stained all blood lymphocytes of adult salmon was used to study the differentiation of surface


Ig positive (sIg+) cells in salmon fry (Ellis, 1977). Although small lymphocytes werepresent in the thymus and kidney prior to hatching, sIg+ lymphocytes were notdetected in whole fry homogenates until 45 dph, coinciding with the time of firstfeeding. By 48 dph, the majority of lymphocytes stained with this antiserum. It isthus apparent, that while lymphocytes are present in the lymphoid organs of fishfrom an early age they do not initially express the surface markers characteristic ofmature lymphocytes and a further period of time is necessary for functional differ-entiation to mature.

Mabs that react with the sIg determinants expressed on B cells are available fora number of species of fish (Scapigliati et al., 1999). Studies using these Mabs todetermine the appearance of B cells in the lymphoid tissues clearly demonstrate theimportance of the kidney in their development. In rainbow trout, B cells first appearin the kidney at about 8 dph and in the spleen at about 30 dph (Razquin et al., 1990).Using homogenates of embryos, B cells could first be detected as early as 8 daysprior to hatching in rainbow trout (Castillo et al., 1993). In sea bass, B cells developafter T cells, first appearing in the kidney at about 38 dph (Breuil et al., 1997) andin the spleen at about 50 dph (Picchietti et al., 1997). The proportion of lymphocytesthat are B cells in the different organs of adult sea bass is reached by about 140 dph(dos Santos et al., 2000).

The proportion of sIg+ cells within the lymphoid organs in the developing carp has shown them to first appear in the kidney about 2 weeks post-hatch and their numbers increase with age to the adult levels (about 20%) at 30 weeks(Koumans-van Diepen et al., 1994; Romano et al., 1997) (table 3). Plasma cellswere first detected in the kidney about 1 month after hatching. B cells first appearedin the spleen at about 4 weeks, in the blood at 6 weeks and in the gut at about 11 weeks (Romano et al., 1997). The percentage of B cells in all these organs continued to increase and did not reach adult levels until after 30 weeks of age. Thethymus never contained many sIg+ cells. These data indicate that the humoralimmune system in the carp begins to mature at about 2–4 weeks of age.

Table 3. First appearance (days after hatching) of lymphocytes expressing thymocyte (T)-specific(T cells) or immunoglobulin (Ig)-specific (B cells) markers in lymphoid tissues of fish detected byimmunohistochemistry

Thymus Kidney Spleen

Mab reactivity T+ Ig+ T+ Ig+ Ig+ References

Carp 5 Absent 10 14 28 Secombes et al., 1983; Romano et al., 1997

Sea bass 30 Absent 35 38 50 Abelli et al., 1996; Picchietti et al., 1997; Breuil et al., 1997

Rainbow trout 8 30 Razquin et al., 1990

320 A.E. Ellis

8. MUCOSA-ASSOCIATED LYMPHOID TISSUE (MALT)

The entire integument of fish is a mucous membrane that is in intimate contact witha pathogen-rich aquatic environment. It is to be expected therefore that well devel-oped immune responses will occur in the mucosal compartment but there is compa-ratively little known. There is no well organized lymphoid tissue in the mucosae butleucocytes are diffusely scattered throughout. These include lymphocytes, antibody-secreting cells (ASCs), macrophages, granulocytes and eosinophilic granular cells(EGCs, considered to be mast cell equivalents; Reite, 1998). EGCs form a denselayer, the stratum granulosum, in the intestine of many fish species. In rainbowtrout, EGCs first appear about the time of first feeding (Bergeron and Woodward,1982). Lymphocytes and ASCs are present in skin (Davidson et al., 1993a), gill(Davidson et al., 1997; Lin et al., 1999) and gut (Davidson et al., 1993b; Lin et al.,2000). In the intestine, ASCs and sIg+ (B) cells are present mainly in the lamina pro-pria (Rombout et al., 1993; McMillan and Secombes, 1997). Most lymphocytes inthe gut mucosal epithelium of fish are sIg− (McMillan and Secombes, 1997). SuchsIg− lymphocytes are also present in the epithelium of skin and gills of carp and areregarded as a distinct population of T lymphocytes in the mucosae, with some sim-ilarities to CD8+ T cells of mammals (Rombout et al., 1998). Immuno-purificationof sIg− lymphocytes from the gut epithelium and blood of sea bass using a mono-clonal antibody has been performed and these cells have been confirmed to be T cells by their expression of the T cell receptor, TCR (Scapigliati et al., 2000).While antibody-producing cells have been reported in the gill and gut, and antibodyis present in gill and skin mucus (Lumsden et al., 1995), Ig appears to be absentfrom gut mucus, at least in salmonids (Hatten et al., 2001).

9. ONTOGENY OF IMMUNE RESPONSIVENESS

9.1. Antibody responses

Evidence from assays of Ig concentration in ova and larvae of fish indicates thatsmall amounts are present at the time of spawning and this decreases post-hatchreaching a nadir at about 2 months post-hatch in rainbow trout (Castillo et al., 1993),12 dph in tilapia (Takemura, 1993) and 5–15 dph in sea bass (Breuil et al., 1997).Thereafter, Ig concentrations begin to increase, suggesting that the onset of endoge-nous production occurs at the time of depletion of maternally derived Ig and yolkand the start of feeding.

The timing of maturation of the immune system in fish is of importance in defin-ing the earliest time at which different fish species can be vaccinated. As fish growmore slowly at lower temperatures, development correlates better with size ratherthan age. Accordingly the onset of immunomaturation has been claimed to correlatebetter with the weight of the fish rather than time after hatch (Manning and Mughal,1985). However, individual fish of the same hatch can grow at different rates but this does not appear to affect the rate of development of T and B lymphocytes


(dos Santos et al., 2000). Thus, a developmental age expressed as degree/days maybe the best way of standardizing data for comparison purposes. As these parameters(age in days, weight and rearing temperature) are infrequently reported in the liter-ature, it is difficult to directly compare results from different studies. This is furthercomplicated because there is evidence that exposure to antigen at a very early ageof development may result in tolerance (see below).

In general the specific antibody titres induced by immunization increase with theage of the fish. Sea bass fry are capable of specific antibody responses to immersiondelivery of Photobacterium damselae spp. piscicida vaccine when only 0.1 g buthigher numbers of antibody-producing cells were induced in 2 g fish and the responsewas faster. These responses occurred mainly in the gill tissue (dos Santos et al.,2001). The serum antibody titres in carp intramuscularly immunized with Vibrioanguillarum when 85, 99 and 128 days old significantly increased with age (Joostenet al., 1995). These results are in agreement with the finding of Koumans-van Diepenet al. (1994) who suggested that the immune system of carp was completed between3 and 8 months of age, based on the percentages of B cells and plasma cells found inthe lymphoid tissues (see table 4). In addition, van Loon et al. (1981) stated that theadult level of serum Ig was attained when carp were 5–8 months of age.

Oral exposure of carp and gilthead sea bream (Sparus aurata) to V. anguillarumantigen at an earlier age (58 dph, just prior to weaning from Artemia nauplii to commercial diet; weight not reported) resulted in memory formation, as the anti-body titres induced by an injection of the antigen 10 weeks after oral administrationwere significantly higher than in the control group which had not received the oralpriming. However, when carp of only 15 and 29 dph were similarly exposed toorally administered Vibrio antigens, the antibody titres induced by an injection route10 weeks later were significantly lower than in the control group (Joosten et al.,1995). This indicates that very young carp can develop a degree of immunologicaltolerance to intestinal exposure to antigens which lasts for at least 10 weeks; animportant feature in devising protocols for oral priming of juvenile fish.

9.2. Ontogeny of memory versus tolerance

The age of onset of specific antibody production appears to depend upon the natureof the antigen, possibly pertaining to whether it is T independent or T dependent, aswell as the route of exposure. As shown in table 4, rainbow trout fry, injected withAeromonas salmonicida antigens at 7 and 14 dph, were unresponsive; they showedno antibody production, or tolerance on secondary immunization at a later stage(Manning et al., 1982). On the other hand, injection of this antigen into rainbowtrout and carp at 3–4 weeks post-hatch, induced an antibody response and develop-ment of memory (Manning et al., 1982; Manning and Mughal, 1985). Similarly,with bath at this age, although primary antibody levels were low, the bath primingimproved the secondary response to subsequent injection immunization (Manningand Mughal, 1985). In contrast, injection of human gamma globulin (HGG) or

Tabl

e 4.

Dev

elop

men

t of

imm

une

resp

onsi

vene

ss in

fis

h

Rai

nbow

trou

tC

arp

Ant

ibod

y re

spon

se to

Ant

ibod

y re

spon

se to

Age

(w

eeks

) w

hen

firs

t im

mun

ized

,Vi

brio

(fo

llow

ing

Salm

ongr

afte

d or

ass

ayed

A.s

HG

GG

raft

rej

ectio

nA

.sH

GG

Srbc

oral

pri

min

g)4

Gra

ft r

ejec

tion

ML

R6

1−5

−(T

)5In

com

plet

e7−

2−5

−(T

)5+

7S

+(M

)8,9

3+

5−

(T)5

+(M

)9−

4+

(M)1

+(M

)1−T

1−T

2S

−6

+8

+5

+(M

)1+

3P

−, N

o re

spon

se; +

, pos

itive

res

pons

e; M

, mem

ory

indu

ced;

T, t

oler

ance

indu

ced;

S, P

, sup

pres

sed

or e

nhan

ced

resp

onse

to s

ubse

quen

t inj

ectio

n re

spec

tivel

y;

ML

R, m

ixed

leuc

ocyt

e re

actio

n; H

GG

, hum

an g

amm

a gl

obul

in; S

rbc,

she

ep e

ryth

rocy

tes;

A.s

,A.s

alm

onic

ida.

Dat

a fr

om: 1

Man

ning

and

Mug

hal (

1985

); 2

Van

Loo

n et

al.

(198

1); 3

Van

Mui

swin

kel e

t al.

(198

5); 4

Joos

ten

et a

l. (1

995)

; 5M

anni

ng e

t al.

(198

2); 6

Elli

s (1

977)

; 7 T

atne

r an

d M

anni

ng (

1983

); 8

Bot

ham

and

Man

ning

(19

81);

9B

otha

m e

t al.

(198

0).


sheep red blood cells (Srbc) into trout and carp up to 4 weeks post-hatch resulted inthe induction of tolerance which persisted for up to 23 weeks (Van Loon et al., 1981;Manning et al., 1982; Manning and Mughal, 1985; Van Muiswinkel et al., 1985). A positive response and development of memory was not detected to these antigensuntil the fish were 8 weeks of age.

This lack of tolerance induction to bacterial antigens by injection or immersionexposure of young fish contrasts with the effect of oral exposure. As mentionedabove, oral priming of carp at 2 and 4 weeks post-hatch with V. anguillarum anti-gens suppressed antibody production in response to injection of the antigen 10 weeks later in comparison to non-orally primed controls. However, oral primingat 8 weeks of age resulted in enhanced responses to injected antigen 10 weeks latercompared with controls (Joosten et al., 1995).

Taken together, these data suggest that B cells and T suppressor functions in carpmature at about 4 weeks of age but T helper functions and memory cells mature laterat about 8 weeks.

9.3. Ontogeny of immune protection

The earliest time to vaccinate fish is an important issue in aquaculture. Some studieswith salmonid species indicate that immersion vaccination with Vibrio anguillarumvaccines is ineffective in fish below 1 g (Johnson et al., 1982a). Above this size,duration of protection lengthened with age. Maximum duration of protection wasachieved in rainbow trout when they were immunized at about 4 g (Johnson et al., 1982b). Other work indicates that rainbow trout fry could be effectively pro-tected against vibriosis when immersion vaccinated at 0.5 g body weight (10 weekspost-hatch) and challenged by ip injection 4 weeks later (Tatner and Horne, 1983).These fish began feeding at 4 weeks post-hatch.

However, the situation in channel catfish may be quite different. Eyed ova wereimmersed in a live attenuated Edwardsiella ictaluri vaccine 4 days prior to hatching.When challenged by immersion 60 days post-vaccination significant protection wasobserved compared to non-vaccinated controls (Shoemaker et al., 2002). However,it is considered that channel catfish do not attain immuno-competence (antibodyproduction) until about 4 weeks post-hatch (Petrie-Hanson and Ainsworth, 1999).The mechanism of the protection by in ovo vaccination is not understood and maybe to do with the vaccine being live and persisting until the immune system matures,or stimulating the system in a different way to that of inactivated antigens. It may berelevant that the mechanism of protection against enteric septicaemia in catfish hasbeen shown to be cell mediated rather than antibody mediated (Shoemaker andKlesius, 1997; Shoemaker et al., 1997) and cell-mediated immunity (CMI) developsslightly earlier than the antibody response (see below). Furthermore, it has beendemonstrated that live attenuated bacterial vaccines, such as the AroA Aeromonassalmonicida, preferentially induce T cell responses relative to B cell responses inrainbow trout (Marsden et al., 1996).

324 A.E. Ellis

9.4. Ontogeny of cell-mediated immunity

Development of the CMI response in young fish has been investigated using twotests: the mixed leucocyte response (MLR) and the allograft rejection response(table 4). In the Atlantic salmon the MLR develops at 45 dph, coincidentally withfirst feeding (Ellis, 1977). Skin rejection responses in carp and rainbow trout fry aredeveloped by 16 dph (Botham and Manning, 1981) and 14 dph (Tatner andManning, 1983), respectively. In the trout, grafting at 5 dph resulted in incompleterejection by day 30 post-grafting but the experiment was not continued to establishif allograft application at this age eventually results in rejection or tolerance.

The ontogeny of cell-mediated immune memory has been investigated followingthe application of first set skin grafts to 26-day-old rainbow trout and 16-day-old carp(Botham et al., 1980). In both cases memory developed. It is thus apparent that in troutand carp the CMI system has fully matured by 2–4 weeks post-hatch and suggests thatthis system, with the production of cytotoxic T-like cells, matures a little earlier thanthe humoral immune response particularly to the T-dependent antigen, HGG.


Fish larvae appear to hatch when their specific immune responsiveness is still non-functional and defence against bacterial infection is presumably provided by non-specific factors derived from the yolk or endogenously produced. Little is known aboutthese factors or how they are affected by the health status of the mother. However, theywould be expected to play an important role in affecting “egg quality” and the chancesof survival of the larvae. Optimizing these factors with respect to resistance to impor-tant diseases in aquaculture will be a useful challenge for future research.

As the yolk is utilized by the sac-fry, the lymphoid organs develop functionalmaturity, which in some species, e.g. salmonids, coincides with the virtual depletionof the yolk sac and the onset of first feeding. However, in some species, e.g. flatfishand sea bass, where the eggs and larvae are very small, feeding on zooplanktonappears to precede the development of the lymphoid organs. For instance, Breuil et al. (1997) reported that the onset of feeding on Artemia nauplii was at 7–10 dphin their sea bass larvae, yet the thymus did not become lymphoid until 21 dph andsIg+ cells began to increase after 30 dph, at about the time the fish were weaned onto a commercial sea bass diet. These authors concluded that the fry are competentfor antibody responses when the weaning period is achieved at about 40 dph and the fish were about 50 mg. Thus, in these species there appears to be a rather pro-longed period from first feeding until the capacity for specific antibody responses.During this time, defence against bacterial infection would appear to be entirelydependent on non-specific mechanisms but it is also a period when sea bass fry are highly sensitive to bacterial infection (Breuil and Haffner, 1989). Sea bass arecertainly capable of antibody responses to immersion immunization at 100 mg,especially in the gill tissue (dos Santos et al., 2001), but the effect of earlier exposure


to antigen has not yet been reported. In larval carp, oral delivery of Vibrio antigensbio-encapsulated in Artemia nauplii, before 58 dph resulted in immunosuppression(Joosten et al., 1995). It will therefore be of great importance to define the timing ofvaccination of larvae of different fish species in order to achieve protection ratherthan risk rendering the fish more susceptible to disease.

REFERENCES

Abelli, L., Picchietti, S., Romano, N., Mastrolia, L., Scapigliati, G., 1996. Immunocytochemical detec-tion of a thymocyte antigenic determinant in developing lymphoid organs of sea bass, Dicentrarchuslabrax (L). Fish Shellfish Immunol. 6, 493–505.

Abelli, L., Baldassini, M.R., Meschini, R., Mastrolia, L., 1998. Apoptosis of thymocytes in developingsea-bass Dicentrarchus labrax (L.). Fish Shellfish Immunol. 8, 13–24.

Bergeron, T., Woodward, B., 1982. The development of the stratum granulosum of the small intestine ofthe rainbow trout Salmo gairdneri. Can. J. Zool. 60, 1513–1516.

Botham, J.W., Manning, M.J., 1981. Histogenesis of the lymphoid organs in the carp, Cyprinus carpio L. and the ontogenic development of allograft reactivity. J. Fish Biol. 49, 403–414.

Botham, J.W., Grace, M.F., Manning, M.J., 1980. Ontogeny of first set and second set of immune reac-tivity in fishes. In: Manning, M.J. (Ed.), Phylogeny of Immunological Memory. Elsevier/NorthHolland, Biomedical Press, Amsterdam, pp. 83–92.

Breuil, G., Haffner, P., 1989. A field report on Vibrio disease of sea bass, Dicentrarchus labrax, in thesouth of France. Adv. Tropic. Aquacult. 9, 161−169.

Breuil, G., Vassiloglou, B., Pepin, J.F., Romestand, B., 1997. Ontogeny of IgM-bearing cells and changesin the immunoglobulin-like protein level (IgM) during larval stages in sea bass (Dicentrarchuslabrax). Fish Shellfish Immunol. 7, 29–43.

Castillo, A., Sanchez, C., Dominguez, J., Kaattari, S.L., Villena, A., 1993. Ontogeny of IgM and IgM-bearing cells in rainbow trout. Develop. Comp. Immunol. 17, 419–424.

Castillo, A., Razquin, B., Villena, A.J., Zapata, A.G., Lopez-Fierro, P., 1998. Thymic barriers to antigenentry during the post-hatching development of the thymus of rainbow trout, Oncorhynchus mykiss.Fish Shellfish Immunol. 8, 157–170.

Cecchini, S., Terova, G., Caricato, G., Saroglia, M., 2000. Lysozyme activity in embryos and larvae ofsea bass (Dicentrarchus labrax L.), spawned by broodstocks fed with vitamin C enriched diets. Bull.Eur. Ass. Fish Pathol., 20, 120–124.

Davidson, G.A., Ellis, A.E., Secombes, C.J., 1993a. Novel cell types isolated from the skin of rainbowtrout, Oncorhynchus mykiss. J. Fish Biol. 42, 301–306.

Davidson, G.A., Ellis, A.E., Secombes, C.J., 1993b. Route of immunization influences the generation ofantibody secreting cells in the gut of rainbow trout. Develop. Comp. Immunol. 17, 373–376.

Davidson, G.A., Lin, S.-H., Secombes, C.J., Ellis, A.E., 1997. Detection of specific and “constitutive”antibody secreting cells in the gills, head kidney and peripheral blood leucocytes of dab (Limandalimanda). Vet. Immunol. Immunopathol. 58, 363–374.

dos Santos, N.M.S., Romano, N., de Sousa, M., Ellis, A.E., Rombout, J.H.W.M., 2000. Ontogeny of Band T cells in sea bass (Dicentrarchus labrax, L.). Fish Shellfish Immunol. 10, 583–596.

dos Santos, N.M.S., Taverne-Thiele, J.J., Barnes, A.C., van Muiswinkel, W.B., Ellis, A.E., Rombout, J.H.W.M., 2001. The gill is a major organ for antibody secreting cell production followingdirect immersion of sea bass (Dicentrarchus labrax L.) in a Photobacterium damselae spp. piscicidabacterin: an ontogenetic study. Fish Shellfish Immunol. 11, 65–74.

Douglas, S.E., Gallant, J.W., Gong, Z., Hew, C., 2001. Cloning and developmental expression of a familyof pleurocidin-like antimicrobial peptides from winter flounder, Pleuronectes americanus(Walbaum). Develop. Comp. Immunol. 25, 137−147.

Ellis, A.E., 1977. Ontogeny of the immune response in Salmo salar. Histogenesis of the lymphoid organsand appearance of membrane immunoglobulin and mixed leucocyte reactivity. In: Solomon, J.B.,Horton, J.D. (Eds.), Developmental Immunobiology. Elsevier/North Holland, Biomedical Press,Amsterdam, pp. 225–231.

326 A.E. Ellis

Ellis, A.E., 1999. Immunity to bacteria in fish. Fish Shellfish Immunol. 9, 291–308.Grace, M.F., Manning, M.J., 1980. Histogenesis of the lymphoid organs in rainbow trout, Salmo gairdneri.

Develop. Comp. Immunol. 4, 255–264.Hatten, F., Fredriksen, A., Hordvik, I., Endresen, C., 2001. Presence of IgM in cutaneous mucus, but not

in gut mucus of Atlantic salmon, Salmo salar. Serum IgM is rapidly degraded when added to gutmucus. Fish Shellfish Immunol. 11, 257–268.

Johnson, K.A., Flynn, J.K., Amend, D.F., 1982a. Onset of immunity in salmonid fry vaccinated by directimmersion in Vibrio anguillarum and Yersinia ruckeri bacterins. J. Fish Dis. 5, 197–205.

Johnson, K.A., Flynn, J.K., Amend, D.F., 1982b. Duration of immunity in salmonids vaccinated by directimmersion with Yersinia ruckeri and Vibrio anguillarum bacterins. J. Fish Dis. 5, 207–213.

Joosten, D.H.M., Aviles-Trigueros, M., Sorgeloos, P., Rombout, J.H.W.M., 1995. Oral vaccination ofjuvenile carp (Cyprinus carpio) and gilthead seabream (Sparus aurata) with bioencapsulated Vibrioanguillarum bacterin. Fish Shellfish Immunol. 5, 289–299.

Josefsson, S., Tatner, M.F., 1993. Histogenesis of the lymphoid organs in sea bream (Sparus aurata L.).Fish Shellfish Immunol. 3, 35–49.

Koumans-van Diepen, J.C.E., Taverne-Thiele, J.J., van Rens, B.T.T.M., Rombout, J.H.W.M., 1994.Immunocytochemical and flow cytometric analysis of B cells and plasma cells in carp (Cyprinuscarpio L.); an ontogenetic study. Fish Shellfish Immunol. 4, 19–28.

Lele, S.H., 1933. On the phasical history of the thymus gland in plaice of various ages with note on theinvolution of the organ, including also notes on the other ductless glands in this species. J. Univ.Bombay 1, 37–53.

Lillehaug, A., Sevatdal, S., Endal, T., 1996. Passive transfer of specific maternal immunity does not pro-tect Atlantic salmon (Salmo salar L.) fry against yersiniosis. Fish Shellfish Immunol. 6, 521–535.

Lin, S.-H., Ellis, A.E., Davidson, G.A., Secombes, C.J., 1999. Migration, respiratory burst and mitogenicresponses of leucocytes from the gills of rainbow trout (Oncorhynchus mykiss). Fish ShellfishImmunol. 9, 211–226.

Lin, S.-H., Davidson, G.A., Secombes, C.J., Ellis, A.E., 2000. Use of a lipid-emulsion carrier for immu-nisation of dab (Limanda limanda) by bath and oral routes: an assessment of systemic and mucosalantibody responses. Aquaculture 181, 11–24.

Lumsden, J.S., Ostland, V.E., MacPhee, D.D., Ferguson, H.W., 1995. Production of gill-associated andserum antibody by rainbow trout (Oncorhynchus mykiss) following immersion immunisation withacetone-killed Flavobacterium branchiophilum and the relationship to protection from experimentalchallenges. Fish Shellfish Immunol. 5, 151–165.

Manning, M.J., Mughal, M.S., 1985. Factors affecting the immune responses of immature fish. In: Ellis,A.E. (Ed.), Fish and Shellfish Pathology. Academic Press, London, pp. 27–40.

Manning, M.J., Grace, M.F., Secombes, C.J., 1982. Ontogenetic aspects of tolerance and immunityin carp and rainbow trout: Studies on the role of the thymus. Develop. Comp. Immunol. Suppl. 2,75–82.

Marsden, M.J., Vaughan, L.M., Foster, T.J., Secombes, C.J., 1996. A live (ΔaroA) Aeromonas salmonicidavaccine for furunculosis preferentially stimulates enhanced T cell responses relative to B cell responsesin rainbow trout (Oncorhynchus mykiss). Inf. Imm. 5, 199–210.

McMillan, D.N., Secombes, C.J., 1997. Isolation of rainbow trout (Oncorhychus mykiss) intestinalintraepithelial lymphocytes (IEL) and measurement of their cytotoxic activity. Fish ShellfishImmunol. 7, 527–541.

Nakanishi, T., 1991. Ontogeny of the immune system in Sebastiscus marmoratus: histogenesis of thelymphoid organs and effects of thymectomy. Exp. Biol. Fishes 30, 135–145.

Petrie-Hanson, L., Ainsworth, A.J., 1999. Humoral immune response of channel catfish (Ictaluruspunctatus) fry and fingerlings exposed to Ewardsiella ictaluri. Fish Shellfish Immunol. 9, 579–589.

Picchietti, S., Terribili, F.R., Mastrolia, L., Scapigliati, G., Abelli, L., 1997. Expression of lymphocyteantigenic determinants in developing gut-associated lymphoid tissue of the sea bass Dicentrarchuslabrax (L.). Anat. Embryol. 196, 457–463.

Razquin, B.E., Castillo, A., Lopez-Fierro, P., Alvarez, F., Zapata, A., Villena, A.J., 1990. Ontogeny ofIgM-producing cells in the lymphoid organs of rainbow trout, Salmo gairdneri Richardson: animmuno- and enzyme histochemical study. J. Fish Biol. 36, 159–173.

Reite, O.B., 1998. Mast cells/eosinophilic granular cells of teleostean fish: A review focusing on stain-ing properties and functional responses. Fish Shellfish Immunol. 8, 489–513.


Romano, N., Taverne-Thiele, J.J., van Maanen, J.C., Rombout, J.H.W.M., 1997. Leucocyte subpopula-tions in the developing carp (Cyprinus carpio L.): immunocytochemical studies. Fish ShellfishImmunol. 7, 439–453.

Romano, N., Taverne-Thiele, A.J., Fanelli, M., Baldassini, M.R., Abelli, L., Mastrolia, L., Van Muiswinkel, W.B., Rombout, J.H.M.W., 1999. Ontogeny of the thymus in a teleost fish, Cyprinuscarpio L.: developing thymocytes in the epithelial microenvironment. Develop. Comp. Immunol. 23, 123–137.

Rombout, J.H.W.M., Taverne-Thiele, J.J., Villena, M.I., 1993. The gut associated lymphoid tissue (GALT)of carp (Cyprinus carpio): an immunocytochemical analysis. Develop. Comp. Immunol. 15, 55–66.

Rombout, J.H.W.M., Van de Wal, J.W., Companjen, A., Taverne-Thiele, J.J., 1997. Characterisation of aT cell lineage marker in carp Cyprinus carpio L. Develop. Comp. Immunol. 21, 35–46.

Rombout, J.H.W.M., Joosten, P.H.M., Engelsma, A.P.V., Taverne, N., Taverne-Thiele, J.J., 1998.Indications for a distinct putative T cell population in mucosal tissue of carp (Cyprinus carpio L.).Develop. Comp. Immunol. 22, 63–77.

Scapigliati, G., Mazzini, M., Mastrolia, L., Romano, N., Abelli, L., 1995. Production and characterisa-tion of a monoclonal antibody against the thymocytes of the sea bass Dicentrarchus labrax (L.)(Teleostea, Percicthydae). Fish Shellfish Immunol. 5, 393–405.

Scapigliati, G., Romano, N., Abelli, L., 1999. Monoclonal antibodies in fish immunology: identification,ontogeny and activity of T- and B-lymphocytes. Aquaculture 172, 3–28.

Scapigliati, G., Romano, N., Abelli, L., Meloni, S., Ficca, A.G., Buonocore, F., Bird, S., Secombes, C.J.,2000. Immunopurification of T-cells from sea bass Dicentrarchus labrax (L.). Fish ShellfishImmunol. 10, 329–341.

Secombes, C.J., Van Groningen, J.J.M., Van Muiswinkel, W.B., Egberts, E., 1983. Ontogeny of the immunesystem in carp (Cyprinus carpio L.). The appearance of antigenic determinants on lymphoid cellsdetected by mouse anti-carp thymocyte monoclonal antibodies. Develop. Comp. Immunol. 7, 455–464.

Shoemaker, C.A., Klesius, P.H., 1997. Protective immunity against enteric septicaemia in channel cat-fish, Ictalurus punctatus (Rafinesque), following controlled exposure to Edwardsiella ictaluri. J. FishDis. 20, 101–108.

Shoemaker, C.A., Klesius, P.H., Plumb, J.A., 1997. Killing of Edwardsiella ictaluri by macrophagesfrom channel catfish immune and susceptible to enteric septicaemia of catfish. Vet. Immunol.Immunopathol. 58, 181–190.

Shoemaker, C.A., Klesius, P.H., Evans, J.J., 2002. In ovo methods for utilizing the modified liveEdwardsiella ictuluri vaccine against enteric septicemia in Channel catfish. Aquaculture 203, 221–227.

Sin, Y.M., Ling, K.H., Lam, T.J., 1994. Passive transfer of protective immunity against ichthyophthiriasisfrom vaccinated mother to fry in tilapias, Oreochromis aureus. Aquaculture 120, 229–237.

Takemura, A., 1993. Changes in an immunoglobulin (IgM)-like protein during larval stages in tilapia,Oreochromis mossambicus. Aquaculture 115, 233–241.

Takemura, A., 1996. Immunohistochemical localisation of lysozyme in the prelarvae of tilapia,Oreochromis mossambicus. Fish Shellfish Immunol. 6, 75–77.

Tatner, M.F., 1985. The migration of labelled thymocytes to the peripheral lymphoid organs in rainbowtrout, Salmo gairdneri Richardson. Develop. Comp. Immunol. 9, 85–91.

Tatner, M.F., Horne, M.T., 1983. Susceptibility and immunity to Vibrio anguillarum in post-hatchingrainbow trout fry, Salmo gairdneri Richardson 1836. Develop. Comp. Immunol. 7, 465–472.

Tatner, M.F., Manning, M.J., 1983. The ontogeny of cellular immunity in the rainbow trout, Salmogairdneri Richardson, in relation to the stage of development of the lymphoid organs. Develop.Comp. Immunol. 7, 69–75.

Tatner, M.F., Manning, M.J., 1985. The ontogenic development of the reticuloendothelial system in therainbow trout, Salmo gairdneri Richardson. J. Fish Dis. 8, 35–41.

Trede, N.S., Zon, L.I., 1998. Development of T-cells during fish embryogenesis. Develop. Comp.Immunol. 22, 253–263.

Van Muiswinkel, W.B., Anderson, D.P., Lamars, C.H.J., Egberts, E., Van Loon, J.J.A., Ijssel, J.P., 1985.Fish immunology and fish health. In: Manning, M.J., Tatner, M.F. (Eds.), Fish Immunology.Academic Press, London, pp. 1–8.

Van Loon, J.J.A., van Oosterom, R., van Muiswinkel, W.B., 1981. Development of the immune systemin carp. In: Solomon, J.B. (Ed.), Aspects of Developmental and Comparative Immunology, Vol. 1.Pergamon Press, Oxford, pp. 469–470.

328

Lactobacilli have a number of properties that render them highly suited as vehiclesfor delivery to the mucosa of compounds that are of pharmaceutical interest. Oneattractive property of many lactobacilli is that they survive passage through thehighly acidic stomach. As such, the desired compounds can be delivered in situwithout the risk of being degraded prior to reaching the actual target site. Indeed,many strains of the genus Lactobacillus are capable of colonizing specific regionsof the body, e.g. the oral cavity and the gastrointestinal and urogenital tracts, wherethey play an important role in maintaining a balanced ecosystem.

Recent years have seen an impressive growth in our understanding of the molecular genetic properties of lactobacilli and how to exploit this knowledge for the expression of foreign proteins. The immunomodulating capacity of lactobacillitogether with the possibility to target antigens to specific sites of the bacterium, offerattractive opportunities for the treatment of infectious diseases through vaccination,and of autoimmune diseases or other immune disorders by modulating the immuneresponse in a directed and predetermined way.

In this overview the present state of the art regarding Lactobacillus systems forthe high-level expression of foreign antigens and their delivery will be presented.Some immunological properties of lactobacilli will be discussed and their potentialuse as delivery vehicles for oral immunization purposes will be highlighted.

1. INTRODUCTION

Why mucosal immunization? First of all, delivery of vaccines to the mucosa (orally,intranasally, vaginally) is simple and relatively inexpensive and does not require the

15 Development of lactobacilli for mucosalimmunization

J.F.M.L. Seegers, C.E.G. Havenith, S.H.A. Kremer and P.H. Pouwels

Toegepast Natuurkundig Onderzoek (TNO) Prevention and Health, Department of Infection and Immunology, Special Programme InfectiousDiseases, Post Box 2215, NL-2301 CE Leiden, The Netherlands


Lactobacilli for mucosal immunization 329

use of needles with all the risks of contamination involved. This is especially impor-tant for large vaccination programmes in which large numbers of subjects areinvolved. Moreover, for most pathogens mucosal sites are the porte d’entrée andtherefore the most likely sites for a vaccine to elicit the desired response. Parenteraladministration of antigens stimulates induction of systemic immune responses, but is generally ineffective for induction of secretory immunoglobulin A (sIgA).Vaccines administered via mucosal surfaces can elicit biologically active serum IgGantibodies and cell mediated immune responses. Furthermore, they can elicitmucosal sIgA and cell mediated immunity at mucosal surfaces to prevent pathogeninfiltration and inflammation (Levine and Dougan, 1998; van Ginkel et al., 2000).

Why lactobacilli? Several live microbial vaccine-vectors that are active at targetsites of mucosal immunization have been shown to be efficient delivery systems infacilitating immune responses at mucosal and systemic sites concurrently (Walker,1994). These vectors are usually derived from attenuated pathogenic microorgan-isms such as Salmonella typhimurium, Listeria monocytogenes and Mycobacteriumbovis BCG. Several excellent recent reviews describe the use of these pathogens invaccination and their future development (Thole et al., 2000; Medina and Guzman,2001; Mollenkopf et al., 2001). However, safety considerations, particularly theimmune status of the vaccine recipients in developing countries, make it doubtfulwhether these vector candidates can be used on a large scale. Furthermore, theseorganisms are highly immunogenic themselves, drawing unnecessary attention ofthe immune system, and this might even hamper repetitive use of the carrier withother antigens (Hone et al., 1991). Moreover, this system could be extremely usefulfor the development of safe oral vaccines for infants of whom the immune systemhas not been fully developed and for whom vaccines based on attenuated pathogenscould pose a serious health risk. Therefore, non-pathogenic, food grade or com-mensal bacterial vectors have started to receive attention for their vaccine potential(Pouwels et al., 1996; Slos et al., 1998; Shaw et al., 2000; Havenith et al., 2002).These bacteria are in general amenable to genetic transformation, and the fact thatthey can survive the passage through the stomach and can persist in the gastroin-testinal tract makes them useful delivery vehicles of antigens to be presented to themucosal immune system.

A potential limitation of oral immunization is that the immunogenicity of anti-gens administered in a soluble form is low. In fact, this administration route mayeven induce tolerance. To circumvent this, antigens should be presented togetherwith an adjuvant, either chemically coupled to a carrier or in the form of a geneti-cally engineered bacterium or virus. Proteins delivered in this way can induce a pro-tective immune response in the host without the need for actual contact with thepathogen itself.

The past decade has seen intensive research, focused on the use of lactic acid bac-teria (LAB) as vaccine delivery vehicles. Overviews of earlier research in this fieldhave appeared in a number of articles (Wells et al., 1996; Pouwels et al., 1998;

330 J.F.M.L. Seegers et al.

Mercenier, 1999). In this chapter, an overview is given of the current state of the artwith respect to the development of lactic acid bacteria in general and Lactobacillusspp. in particular as immune modulating entities with an emphasis on developmentsover the past five years.

2. LAB AS LIVE MICROBIAL CARRIERS

Many LAB species are being used for the manufacturing of fermented food prod-ucts, varying from wine and yoghurt to cheese and dried sausages. LAB species canalso be found as members of the natural microflora of living organisms and, as aresult, have long been the subject of scientific research. Over the past two decadesLAB have received enormous attention in international research, also as a result oflarge financial funding from the EU. This has led to an explosive growth in ourknowledge on many aspects of these bacteria, such as carbohydrate metabolism,genetics, stress regulation, and phage resistance. By now, the genomes of severalrepresentatives of LAB have been determined, which will give a new stimulus toresearch in this field.

The possibility to use LAB for vaccination purposes was seriously considered forthe first time about 10 years ago. Gerritse (1990) showed that killed lactobacilli thatcarried the hapten trinitrophenyl chemically coupled to the cell surface, could beused for the immunization of mice. Other groups exploited the use of Streptococcusgordonii (Pozzi et al., 1992) or Lactococcus lactis (Wells et al., 1993) as expressionand delivery vehicles. A first joint effort was initiated through the EU Biotech 1programme. This programme was organized as a single integrated project aimed at the development of LAB for industrial application and combined the expertise of36 participating laboratories. The project was divided into subgroups to facilitate atargeted focus on key objectives. The key objective of one subgroup was LAB invaccines. This subgroup continued in the fourth framework programme under theacronym LABVAC, comprising nine groups from all over Europe. Their workresulted in the development of sophisticated cloning systems that allow expressionat the different cellular compartments and led to several interesting findings. It wasshown that the level of expression of the antigen is a crucial factor. This in itself isnot a surprise, but poses some limits to what can be administered. It was also shownthat the immune response could be enhanced when mice were immunizedintranasally with different expression strains of L. lactis which secreted IL-2 or IL-6 in addition to the production of tetanus toxin fragment C (TTFC). In this case, the anti-TTFC antibody titres increased more rapidly and were substantiallyhigher (Steidler et al., 1998). This group also showed that lactococci, expressing and secreting IL-10, can be used for the treatment of colitis in mice, and this modelis currently being tested as a possible treatment for Crohn’s disease in humans(Steidler et al., 2000). An overview of viral and bacterial antigens that have beencloned in non-pathogenic Gram-positive bacteria is given in table 1.


Experiments performed by Drouault et al. (1999) with Lactococcus lactis express-ing green fluorescent protein suggest that the antigens are being released as a resultof cell lysis either before or after uptake by macrophages and are thus exposed tothe immune system. This difference in immunogenicity could reflect a difference in expression levels as a result of promoter activity or as a result of extracellularproteolytic activity.

An ongoing topic of research is strain containment. It is desirable that recombinantstrains that have been used for immunization experiments are not able to surviveoutside the treated subject in order to prevent the unwanted spread of these micro-organisms in the environment.

3. SELECTION OF STRAINS

Several criteria can be defined for the choice of the proper strain. These might forexample be related to the capacity to stimulate or suppress an immune response, to the efficacy with which such strains can express and secrete a foreign antigen, or to theirresistance to the hostile environment that is present in the gastrointestinal tract. Thereis an increasing accumulation of evidence showing that LAB, notably lactobacilli, arecapable of influencing the expression of a number of cytokines, both in vivo and in vitro. It has also been shown that different strains induce distinct mucosal cytokineprofiles and possess differential intrinsic adjuvanticity (Maassen et al., 2000), a pro-perty that enables the investigator to aim at Th1- or Th2-dominated immune responses.

Table 1. Antigens that have been cloned in non-pathogenic bacteria for vaccine development

Carrier strain Pathogen Antigen Reference

L. lactis Brucella abortus L7/L12 Ribeiro et al., 2002L. lactis Helicobacter pylori Ure B Lee et al., 2001L. lactis Rotavirus NSP4 Enouf et al., 2001L. lactis Clostridium tetani TTFC Wells et al., 1993Lb. fermentum HIV gp41 Turner and Giffard, 1999Lb. fermentum Chlamydia psittaci OmpA Turner and Giffard, 1999Lb. casei Bacillus anthracis PA Zegers et al., 1999Lb. plantarum HIV Epitope gp41 Hols et al., 1997Lb. casei Influenza virus Hackett epitope Pouwels et al., 1996Lb. casei FMD virus fgmt VP1 Pouwels et al., 1996S. gordonii Porphyromonas gingivalis FimA Sharma et al., 1999S. gordonii Measles virus HA Maggi et al., 2000S. gordonii Measles virus F Maggi et al., 2000S. gordonii Escherichia coli LTB Ricci et al., 2000S. gordonii HIV Epitope gp120 Pozzi et al., 1994S. gordonii Bordetella pertussis Toxin S1 subunit Lee et al., 1999S. gordonii Human papillomavirus E7 Pozzi et al., 1992

L., Lactococcus; Lb., Lactobacillus; S., Streptococcus.


A strong Th2-dependent adjuvanticity of a Lactobacillus strain might be consideredas an advantageous property, if the strain is to be used for active vaccination.However, for the deliberate induction of mucosal tolerance, e.g. in the case of expres-sion of autoimmune antigens, strains with immunosuppressing properties might bemore suitable.

A number of Lactobacillus strains, in contrast to lactococci, can persist in thegastrointestinal tract for an extended period of time and are found as part of thenormal microflora. Both in vitro and animal models have been used to select foradherent Lactobacillus strains. In a study performed by the groups of Collins andMarteau, the survival of four non-recombinant lactic acid bacterial strains that weregiven orally to human volunteers has been analysed. The two major conclusions fromthese studies are: i) there are considerable differences in the survival of differentstrains, and ii) no strain colonized the gastrointestinal tract permanently (Dunne et al., 1999; Vesa et al., 2000). Depending on the application, persistence of the straincould be a major selection criterion.

Other selection criteria are the transformability of a selected strain and the levelat which it can express the antigen. Although a variety of protocols have been devel-oped for the transformation of many Lactobacillus strains, some strains can still notbe transformed. In addition, the synthesis of antigens is not equally effective in allLactobacillus strains. An important observation has been that the efficiency withwhich a promoter is recognized can differ greatly between strains. In studies per-formed by McCracken and Timms (1999), mutations were introduced in a singlepromoter, and the transcription levels were compared between two Lactobacillusstrains. In another study (McCracken et al., 2000), several promoters were isolatedand their activity was tested in three different Lactobacillus strains. These studieshave identified some attributes, such as a UP element and the TG DNA sequencemotif, that can play a role in the control of transcription in lactobaccilli. A selectioncriterion therefore could be the level of expression from a certain promoter in aspecific strain. Another difference that can be observed between Lactobacillusstrains is codon usage (Pouwels and Leunissen, 1994). These differences are beingcorroborated by the increasing availability of sequence data. The implications of thishave still to be assessed.

Some studies have been performed aimed at determining the fate of LAB afteroral or intranasal administration. The green fluorescent protein (GFP) and luciferasegenes have been used to determine survival and metabolic activity of L. lactis in the gastrointestinal tract (Droualt et al., 1999). It was shown that the way in whichbacteria are administered had a dramatic impact. Lactococci that transit with the dietappeared quite resistant to gastric acidity (90 to 98% survival), whereas only 10 to30% of the bacteria survived in the duodenum. Viable cells were shown to be meta-bolically active in each compartment of the digestive tract, while most dead cellswere subjected to rapid lysis. Furthermore, it was shown by in vitro studies thattrypsin strongly affected survival while pepsin, a stomach enzyme, had far less


influence on survival rates. These findings were corroborated by in vivo experimentswhich indicated that viability of L. lactis was greatly reduced in the duodenumwhere trypsin, secreted by the pancreas, is present. In another study with GFP-expressing Lb. plantarum, ingestion of cells by bronchoalveolar macrophagesfollowing intranasal administration could be shown (Geoffroy et al., 2000). As esti-mated by flow cytometry, the proportion of macrophages having phagocytosedlactobacilli could reach 10% of the total bronchoalveolar macrophage population.

4. LACTOBACILLUS AS DELIVERY VEHICLE FOR ORAL IMMUNIZATION

As outlined above, lactobacilli are capable of persisting in the gastrointestinal tractfor an extended period of time and show a significant immunomodulatory capability.One could argue that these criteria would favour lactobacilli over lactococci for the development of live oral vaccines. As an added benefit many lactobacilli haveprobiotic properties and can prevent adherence, establishment, and replication ofseveral enteric pathogens through different antimicrobial mechanisms such ascompetition for nutrients, occupation of mucosal sites, and the production of anti-microbial peptides. Apart from these intrinsic properties of lactobacilli, several othercriteria are important for the induction of an adequate immune response. Theseinvolve the regime of immunization and the level of expression of the antigen aswell as the location of expression.

In the following section, probiotic properties, some immunogenetic traits oflactobacilli, and the development of genetic tools will be discussed.

4.1. Probiotic properties

In recent studies sponsored by the EU, a consortium of laboratories have definedand applied a number of criteria for the selection of probiotic strains (Dunne et al.,1999). Some of these criteria that also apply to the selection of vaccine-carrierstrains, are acid resistance, bile tolerance and adherence to host epithelial tissues.Many lactobacilli are acid resistant and bile salt resistant, which enables them tosurvive the passage through the stomach and ileum, and to be maintained in largenumbers in the small intestine and upper part of the large intestine, despite the pres-ence of considerable amounts of bile salts. The property of survival in the presenceof bile salts originates from the capacity of numerous lactobacilli to hydrolyse con-jugated bile salts. Deconjugation not only renders bile salts less toxic for intestinalbacteria, but also diminishes their concentration in the lumen, since deconjugatedbile salts can be resorbed in the duodenum into the blood, in contrast to conjugatedbile salts (Leer et al., 1993; de Roos and Katan, 2000).

Very little is known about the mechanisms that are used by lactobacilli to adhereto the mucosa of the intestinal tract. This contrasts with the detailed knowledge that


has become available on the structure and properties of adhesins and the mode ofregulation of adhesin biogenesis of pathogenic bacteria such as Salmonella, Yersiniaand Listeria (Finlay and Cossart, 1997; Finlay and Falkow, 1997).

Persistence of lactobacilli in the gastrointestinal tract is a host-specific andpossibly also tissue-specific process that is mediated by a number of cell wall asso-ciated components such as lipoteichoic acids (LTA), polysaccharides and proteins(Tannock, 1999). A surface-associated protein has been identified that enhances theadhesion of L. reuteri 104R to squamous epithelial cells and to mucus (Henrikssonand Conway, 1992). It was shown that mutants lacking this MapA protein had asignificantly reduced persistence in mice (Satoh, Leer, Conway and Pouwels,unpublished observations). The fact that the strain still persisted for a period of morethan 3 to 4 days, suggests that other surface-associated bacterial factors are involvedin its adhesion to mucosal surfaces. Lactobacilli can also associate with gastroin-testinal and urogenital cells and co-aggregate with microorganisms owing to thepresence of so-called surfactants at the bacterial surface (Reid et al., 2000). Thesemolecules, which are characterized by the ability to reduce liquid surface tension,can inhibit the adhesion of enterococci and other uropathogens to solid surfaces(Velraeds et al., 1996).

Yoghurt consumption has some well acclaimed benefits such as prevention,reduction of duration or severity of infant diarrhoea and has even been linked to anti-tumour effects. These benefits are thought to be largely a result of metabolic fermen-tation products of the commonly used lactic acid bacterial strains Lb. bulgaricus andStreptococcus thermophilus (Djouzi et al., 1997). Table 2 shows some health bene-fits that have been attributed to a number of other Lactobacillus strains.

4.2. Intrinsic immunogenicity of lactobacilli

Immune responses against bacteria are usually and understandably directed againstcell surface associated moieties such as peptides, lipopolysaccharides (LPSs) andcarbohydrates. Since the composition of these elements differs strongly betweenbacteria, they influence the response of the immune system in different ways. LPS,for instance, is a strong immunogen, but is completely absent in Gram-positivebacteria. It has been reported that lipoteichoic acid (LTA), a major constituent of thebacterial cell wall of Gram-positive bacteria, is able to weakly stimulate cytokinesynthesis, and that this effect depends on the presence of D-alanine substituents(Bhakdi et al., 1991; Tsutsui et al., 1991).

On the basis of specific antigenic determinants (mostly cell wall located sugarmoieties), lactobacilli have been classified into seven serological groups (Sharpe,1981). More recent studies have shown that lactobacilli can stimulate or inhibit the expression of several cytokines and can thus influence the immune response(Yasui et al., 1999; Maassen et al., 2000). The intrinsic properties of lactobacilli tomodulate the immune system render them attractive for health applications, and in


particular for the in vivo production and delivery of biologically active molecules(Geoffroy et al., 2000). Administration of lactobacilli can lead to activation of innateimmune effector functions, can affect cytokine expression in a specific or non-specificmanner, and can influence humoral responses.

With regard to the innate immune function, many studies have shown that differ-ent strains of lactobacilli are able to activate macrophages and induce production ofTNF-α, IL-1, IL-6, IL-12, IL-18 and/or IFN-γ (reviewed by Maassen, 1999). Recentstudies indicate that in general Gram-positive bacteria induced more Th1-likecytokines, e.g. IL-12, IL-18, and IFN-γ, than Th2-like IL-10 in human monocytes,while Gram-negative bacteria induced more IL-10 than IL-12 (Mietinnen et al.,1998; Haller et al., 2000; Hessle et al., 2000). The differential effects of variousbacteria, and especially of Lactobacillus strains, on the immune system are at leastpartially due to differences in cell wall composition. Peptidoglycan, a major com-ponent of the cell wall of Gram-positive bacteria, has been shown to induce theproduction of IL-1, IL-6, and TNF-α by human blood cells (Schrijver et al., 1999).However, it is still uncertain whether peptidoglycan is responsible for macrophageactivation (de Ambrosini et al., 1996). In addition, the cell wall components pepti-doglycan and LTA, could not completely mimic the impressive IL-12 inducingeffect of intact Gram-positive bacteria (Hessle et al., 2000). It was suggested that yetundefined component(s) of Gram-positive bacteria is (are) responsible for their

Table 2. Reported activities of probiotic bacteria

Activity Mode of action* Species**

Intestinal health Balancing microflora Lb. acidophilus,Lb. bulgaricus

Anti-cancer Enhancing the immune response, Lb. acidophilus, Lb. caseibinding/removal of carcinogens

Immune system modulation Production of immunopeptides Lb. reuteri, Lb. plantarumMilk tolerance (lactose Lactose metabolism Lb. bulgaricus

intolerance/milk allergy)Vaginal/urinary tract health Low pH, bacteriocin production, Lb. rhamnosus,

hydrogen peroxide production Lb. acidophilus,Lb. fermentum

Stomach health Anti-Helicobacter pylori Lb. acidophilusHypertension Bioactive peptides resulting Lb. Helveticus, Lb. casei

from proteolysis

Cholesterol lowering ??? Lb. acidophilus

*Since in many cases the exact modes of action are unknown, only some likely possibilities are given. All these activities and their possible modes of action are extensively reviewed in a recent issue of the Journal of Nutrition that was in part dedicated to a symposium on probiotic bacteria, held in Washington DC in April 1999 (J. Nutr. 130, 2000, 382S−416S).

**Most of the reported activities are not exclusive to the species that are mentioned. Other species with these activities have been reported, but only the most important ones are listed.


strong IL-12 inducing capacity. These data suggest that Gram-positive bacteriacould be especially suited as inducers of Th1-type responses. Recently, it has beendemonstrated that different types of Toll-like receptor (TLR) recognize different cellwall components. This new class of receptors is indicated as a key initiator of innateimmunity by activating the NF-κB and AP1 pathway. The triggering of these recep-tors leads to intracellular signalling and activation of a variety of inflammatorymediators and cytokines. Different members of the TLR class are triggered by dif-ferent bacterial components (Aderem and Ulevitch, 2000). TLRs have also beenidentified on intestinal epithelial cells, and it has been speculated that they mightplay a frontline role in monitoring lumenal bacteria (Cario et al., 2000). NF-κB hasbeen indicated as a central regulator of intestinal epithelial cell innate immuneresponses induced by infection with enteroinvasive bacteria (Elewaut et al., 1999).Recently it was shown that a non-pathogenic strain of Salmonella (commensalbacterium) is able to abrogate the synthesis of inflammatory cytokines by gutepithelial cells. The bacteria accomplish this by blocking degradation of IκB, aninhibitor of NF-κB. It was speculated that this might represent one of the probioticworking mechanisms of commensal bacteria in general (Neish et al., 2000).

4.2.1. Natural killer cells

Natural killer (NK) cells play an important role in innate immune resistance, partic-ularly through the synthesis of the proinflammatory cytokine IFNγ. Besidesmacrophages (vide infra), NK cells constitute the primary targets for bacterial stim-ulation. NK cells do upregulate the IL-2Rα chain (CD25) and undergo proliferationwhen stimulated by L. johnsonii. In the presence of bacterially primed macrophages,expression of CD25 and/or secretion of IFNγ from purified NK cells was signi-ficantly increased, indicating that full activation required both bacterial and cellcontact-based signals derived from accessory cells (Haller et al., 2000).

4.2.2. Immunoglobulins

The capacity of certain lactobacilli to enhance systemic and mucosal immunity ingeneral, has been shown both in animal model systems and in humans. There havebeen several reports describing the effects of lactobacilli on upregulation of IgA pro-duction, locally as well as systemically, and enhanced sIgA production andincreased numbers of IgA-producing plasma cells was observed (Maassen, 1999;Perdigón et al., 1999; Erickson and Hubbard, 2000). For instance, oral administra-tion of Lactobacillus GG (LGG) during acute rotavirus diarrhoea promotes recoveryvia augmentation of the local immune defences. LGG enhanced non-specific humoralresponses during the acute phase of the infection, reflected in the IgG, IgA, and IgMsecreting cell numbers. Furthermore, a specific IgA response to rotavirus is endorsed,which is possibly relevant in protection against reinfection (Kaila et al., 1992;


Isolauri et al., 1995). In addition, LGG has an immunostimulating effect on oralrotavirus vaccination (DxRRV reassortant rotavirus vaccine), with higher serocon-version (IgA and IgM).

A role for lactobacilli in treatment of allergic disorders was indicated by theobservation that oral feeding of killed Lb. casei (strain Shirota) was able to stimulatethe production of Th1 cytokines, resulting in repressed production of IgE antibodiesin an animal model. A potential role in attenuating autoimmune diseases came fromthe observation that cofeeding of antigen with probiotic bacteria can suppress both antibody and cellular immune responses. The tumour-suppressive activity oflactobacilli seems to depend on the activation of macrophages producing IL-1 andTNF-α (Maassen, 1999; Matsuzaki and Chin, 2000).

4.3. Molecular biological considerations

Over the past decade, considerable progress has been made in developing techniquesto genetically modify lactobacilli. As mentioned before, one of the great obstacles forthe genetic manipulation of lactobacilli is their often poor transformability. However,for most strains, adequate transformation protocols have been established. To dateessentially all tools necessary for investigation of the properties of lactobacilli at themolecular level and for their modification in a directed and predictable manner havebecome available. Plasmid vectors have been constructed with which differentspecies of Lactobacillus can be genetically modified as well as vectors for general orsite-specific integration of vector material into the Lactobacillus chromosome.Expression signals have been analysed and are currently being used in expressionvectors enabling the efficient and targeted expression of antigens or enzymes.

While site-specific chromosomal integration is possible for lactobacilli, its mainapplication has been the adaptation of an inducible system based on the autoregula-tory properties of the bacteriocin nisin (Pavan et al., 2000). This system was derivedfrom L. lactis and extensive studies of its regulatory mechanisms have resulted inthe development of the nisin-controlled expression (NICE) system (Kleerebezem et al., 1997). The original two-plasmid NICE system turned out to be poorly suitedto Lb. plantarum. In order to obtain a stable and reproducible nisin dose-dependentsynthesis of a reporter protein or a model antigen, the lactococcal nisRK regulatorygenes were integrated into the chromosome of Lb. plantarum. This gave satisfactoryresults with regard to stability and regulation. This system turned out to be wellsuited to determine a dose–response relationship between the amount of antigenproduced by the bacteria and the corresponding antibody titre. For vaccine develop-ment using non-pathogenic Gram-positive microorganisms chromosomal integra-tion is mainly being exploited for Streptococcus gordonii. This research has led tosome impressive results in eliciting immune responses to different antigens andcombating pathogens at vaginal and oral sites (Oggioni et al., 1999; Loeffler et al.,2001; table 2).


Indeed, integration into the chromosome usually results in genetically fully stabletransformants. This method, however, results in multiple copies of the gene andtherefore generally gives rise to higher levels of expression, which has been shownto be an important criterion in eliciting an appropriate immune response. At TNO (the Dutch Organization for Applied Research), a series of expression vectors wasdesigned that allows direct cloning of antigens and expression of these antigens indifferent cellular compartments. These vectors are organized in easily replaceablecassettes, and their main differences are i) the promoter that is employed for theexpression of the antigen, ii) the presence or absence of a secretion signal, and iii) thepresence or absence of a cell wall anchor that, when present, results in surface expo-sure of the gene product. The general structure of these vectors is depicted in fig. 1.

4.3.1. Promoter sequences

As for most prokaryotic promoters, the majority of Lactobacillus promoters showtypical and conserved –35 and –10 sequences. In some cases an additional TG motifis found at position –16. The appearance of a TG motif often coincides with thepresence of a less conserved or even total absence of a –35 region. The TG motif,which is not essential for promoter activity, might enhance a rate limiting reactionsuch as docking of RNA polymerase near the –35 region (Pouwels and Chaillou,2003). For the current series of vectors, two promoters were chosen. One promoter(Pamy), driving the expression of the L. paracasei α-amylase gene, carries a creelement to which the global repressor CcpA can bind in the presence of a rapidlymetabolizable energy source. Expression from Pamy can be repressed, e.g. by thepresence of glucose in the growth medium, and can be derepressed by replacingglucose by other sugars such as mannitol or galactose. In other vectors, the promoterof the Lb. casei lactose dehydrogenase gene (Pldh) is used. A new series of vectors

Fig. 1. General structure of the Lactobacillusexpression vectors that can be used for theexpression of antigens in a variety ofLactobacillus species. mcs = multiple cloningsite, ss = secretion signal. All elements such aspromoter, selection marker, anchor sequences,etc. are present in the vectors as cassettes thatcan be easily exchanged between differentvectors. Upstream and downstream of the openreading frame (orf) region a number of uniquerestriction enzyme sites are present to allowinsertion or replacement of antigen encodingsequences. For more details aboutthe structure and properties of these vectors,see Pouwels, Leer and Boersma (1996) orPouwels et al. (2001).


is being developed for which a third promoter, the slpA promoter (PslpA) that drivesexpression of the surface layer protein of Lb. acidophilus, is being employed (Bootet al., 1996). Similar to the ldh promoter, expression from PslpA is constitutive, butit has been shown to give higher levels of expression than the other two promoters.

Although protocols have been developed for the transformation of manyLactobacillus spp., the transformation frequencies are still considerably lower thanthose obtained for E. coli. Consequently, cloning of antigens is mostly carried out inE. coli. As cloning of foreign genes in E. coli frequently results in structural insta-bility with concomitant loss of parts of the plasmid vector (Pouwels and Leer, 1993),a phenomenon which was also observed in the initial stages of the construction of these vectors, a strategy was designed to circumvent instability in E. coli. Theobserved instability seemed to be correlated with the activity on cloned fragmentsof promoter sequences. For this reason, an 80-basepair cassette containing the rho-independent terminator of the ldh gene, flanked by NotI sites, was inserted in allvectors downstream from the promoter and the translation initiation region. Thisresulted in complete structural stability of the hybrid Lactobacillus-E. coli vectorsin E. coli. Before the vectors are transferred to Lactobacillus, this terminator isremoved by NotI digestion and religation.

4.3.2. Secretion signals

In order to elicit an immune response, antigens have to be presented to the immunesystem. Three possibilities exist for bacteria: intracellular, extracellular and surface-bound expression. Extracellular and surface-bound expression require the presence ofan efficient secretion signal. The secretion signal that was used for the current seriesof vectors normally drives the secretion of the α-amylase gene of L. amylovorus.Another secretion signal that is currently under investigation, is that of the highlyexpressed surface layer protein slpA (Boot et al., 1993).

4.3.3. Anchor sequences

Surface exposure of the antigen is obtained by adding a peptide anchor sequence to the protein in addition to the secretion signal. Five different anchors can bedistinguished, tentatively named A1 to A5 (for reviews, see Leenhouts et al., 1999;Navarre and Schneewind, 1999). The A1 anchor comprises a transmembrane span-ning domain (TMD) that links the antigen to the cytoplasmic membrane. Thismethod, however, requires the addition of at least 100 amino acids between theTMD and the protein of interest in order to span the cell wall so that the antigenbecomes exposed at the cell surface. Its efficacy has not yet been tested in intactcells. The A2 anchor binds the antigen covalently to the lipid bilayer through N-acyldiglyceride modification of a cysteine residue, located immediately C-terminal tothe signal sequence cleavage site (Pugsley, 1993). The A3 anchor is responsible for


covalent binding to the cell wall through a sorting reaction that is characterized bya LPXTG box, a hydrophobic region and a charged tail (Navarre and Schneewind,1999). This region is preceded by a 50 to 125 amino acid residues long region tospan the cell wall, and has a high percentage of proline/glycine and/or threonine/serine residues. Binding of the immunoglobulin binding protein A and of numerousother Gram-positive bacterial surface proteins, occurs through a type A3 anchor. A fourth type of anchor (A4) has a modular design and is applied by nearly all bac-terial cell wall hydrolases. The cell wall binding domain often comprises repeatedamino acid sequences and the nature of binding to the cell wall is as yet unknown,but is most likely of a non-covalent nature. Surface layer proteins (S-proteins) ofbacilli and lactobacilli are anchored with the outer part of the cell wall by hydrogenbonds and/or ionic interactions through a region (A5) that is highly positivelycharged and contains a large number of tyrosine residues. The tyrosine residues maybe involved in binding of the S-proteins to saccharide moieties in the cell wall. S-proteins of bacteria belonging to the acidophilus group A, such as Lb. acidophilus,Lb. crispatus and Lb. helveticus contain a highly conserved C-terminal region(one-third of the protein) that interacts with the cell wall, while the anchoring region in bacilli and L. brevis is found near the N-terminal end (Jarosch et al., 2000;Smit et al., 2001).

The anchor sequence that is used in the vectors that are presented in fig. 1 isderived from the L. paracasei protease PrtP and belongs to the A3 family of anchors.It has successfully been employed for surface exposure of β-glucuronidase of E. coli, rotavirus structural proteins VP7 and VP8 and Urease A and B of H. pylori(Pouwels et al., 1998), and tetanus toxin fragment C (TTFC) (Maassen, 1999).Other anchor types that are reported to have been successfully employed in lacto-bacilli are an A4 type anchor from the AcmA protein of L. lactis for surface exposureof β-lactamase, and an A5 type anchor of the surface layer protein SlpA of L. brevis(Leenhouts et al., 1999).

4.4. Future development of vectors

Results from several groups indicate a strong correlation between the level of expres-sion and the immune response that is elicited. Therefore, high-level expression of theantigens seems to be a prerequisite. This can be obtained by choosing strong pro-moters. It has been shown, however, that a strong promoter in one Lactobacillusstrain is not necessarily a strong promoter in another strain (McCracken and Timms,1999). This necessitates the evaluation of the used promoter in the chosen strain and,when necessary, either optimizing the promoter or selection of another promoter thatdoes display strong expression levels for the desired strain.

Another variable is the immunogenicity of the chosen antigen itself. Someproteins hardly elicit an immune response while other proteins, such as TTFC, are potent stimulants of the immune system. In this respect, translational coupling


of a low immunogenic substrate such as glutathione S-transferase with TTFC has beensuccessfully employed to elicit an immune response against this substrate (Khan et al.,1994b). The immune response could further be enhanced by coupling TTFC with mul-tiple tandem copies of a short peptide of this protein (Khan et al., 1994a).

A necessary further development is the construction of food grade vectors. Thecurrent vectors carry antibiotic resistance genes as selection markers and these willhave to be replaced by adequate food grade selection markers to allow their use inhumans without risking the inadvertent release of the antibiotic resistance trait intothe environment. Several metabolic genes such as the lacF gene, which is essentialfor lactose utilization, have been employed as food grade selection markers(MacCormick et al., 1995; Platteeuw et al., 1996). This requires inactivation of the concomitant gene on the chromosome and limits its use to that specific strain. A wider application can be obtained with bacteriocin resistance (immunity) markers.An example of this is the L. johnsonii LafI, which confers immunity to lactacin F(Allison and Klaenhammer, 1996). Since lactacin F does not inhibit growth of alllactobacilli, additional resistance markers against different bacteriocins are required.

5. LACTOBACILLI AS VACCINE CARRIERS: A CASE STUDY

In the previous subsection, we mentioned that one of the factors that can influence theimmune response is the immunization regime. In our group, oral and nasal immu-nizations of mice have been performed with vaccines constructed through an originalsystem for heterologous gene expression in Lactobacillus in which the 50 kDa TTFCis expressed as either an intracellular or a surface-exposed protein (Shaw et al., 2000).

Following nasal delivery of recombinant Lb. plantarum expressing TTFC intra-cellularly, antigen-specific antibody secreting cells (ASC; IgA as well as IgG) couldbe detected in the cervical lymph nodes, indicating that local mucosal activation hasoccurred. These observations are underlined by the results from an analysis of frag-ment cultures of nasal-associated lymphoid tissues (NALT), in which TT-specificIgA could be detected. In addition, activation of the local gut mucosa occurredfollowing oral delivery of recombinant Lactobacillus, as indicated by the presenceof TT-specific ASC in mesenteric lymph nodes (MLN) and Peyer’s patches (PP),and by the presence of TT-specific IgA in fragment cultures of the duodenum.Substantial levels of TTFC-specific immunoglobulin G (IgG) in serum were inducedafter nasal delivery of live recombinant lactobacilli, confirming the efficiency of thenasal route for immunization. Although less efficient, oral application also inducedsignificant levels of TTFC-specific IgG in serum (fig. 2).

These results led us to the hypothesis that nasal delivery could facilitate theinduction of gastrointestinal tract responses by subsequent oral deliveries. Indeed, it was found that induction of systemic, and more importantly of gastrointestinaltract responses are augmented in mice that received recombinant lactobacilli via theintranasal route before intragastric deliveries were applied (fig. 3). Moreover, the


results gave further proof of the compartmentalization of the immune system, andindicated that responses found at the gastrointestinal level were not due to accidentalintranasal cross-contamination.

The goal for successful mucosal vaccination will be to optimize priming of bothsystemic and mucosal immune compartments. Our data indicate that systemic andlocal activation of the mucosal sites occurs, in which both sites, nasal and intestinal,can be targeted. Intranasal application leads to better induction at the nasal mucosalsite, whereas oral application leads to better induction of local responses at the intes-tinal site. Intranasal priming, in addition, can enhance the local responses at theintestinal sites following oral booster applications. This indicates that recombinantlactobacilli interact with the mucosa and opens up considerable potential for thegeneration of more potent mucosally administered vaccines.

Fig. 2. TTFC-specific IgG following immunization of groups of mice with live recombinant lactobacilli,measured by ELISA. (A)16 C57BL/6 mice were immunized with three doses of 5×109 Lb. plantarumpLP503-TTFC intranasally in 20 μl of phosphate buffered saline (PBS) (•) or orally in 200 μl of NaHCO3 (▲)on days 1–3. Identical booster immunizations were administered on days 28–30. (B) Three C57BL/6 micewere immunized with 5×109 Lb. plantarum pLP503-TTFC intranasally in 20 μl of PBS on day 1 and day 28(■) or three doses of Lb. plantarum pLP401-TTFC intranasally in 20 μl of PBS on days 1–3 followed by abooster with either 5×109 Lb. plantarum pLP503-TTFC on days 28–30 (▲), or Lb. plantarum pLP401-TTFCon days 28–30 and 49–51 (◆), intranasally in 20 μl of PBS. This figure is adapted from Shaw et al. (2000).


In a comparative study, Lb. plantarum 256 turned out to be more effective ineliciting an immune response than Lb. casei 393 (Shaw et al., 2000). These strains wereinitially selected for study because of their different periods of persistence in the GItract. It is interesting to speculate that differences in persistence could explain thedifferences found in immunogenicity. Moreover, the levels of TTFC expressed intra-cellularly were higher in Lb. casei than in Lb. plantarum. However, sustainable levelsof non-degraded TTFC, interaction of the bacteria with M cells, dendritic cells andmacrophages, and intrinsic adjuvanticity will each have its influence. These character-istics may help to increase the immunogenicity of the antigen delivery vehicle. In addi-tion, the influence of immunization schemes, with variations in time between primingand boost, and combinations of application routes, seem to have their influence on theefficiency of the induction of responses. Furthermore, delivery of TTFC expressed asan intracellular antigen turned out to be more effective than cell-surface expressionunder the conditions tested (fig. 2). All these factors together have to be taken into con-sideration in order to optimize strain and expression system combinations for a rationalaugmentation of the efficiency of lactobacilli as vaccine delivery vehicles.

6. PASSIVE VACCINATION

Following infection with a pathogen, the immune system will need several days tomount a full blown attack against this pathogen. It was discovered over 100 years

Fig. 3. Measurements of TT-specific IgA antibodies in fragment cultures of the NALT (I), and duodenumand ileum (II) of animals immunized intranasally (in) and/or intragastrically (o) with live recombinant Lb. plantarum expressing TTFC. Each animal was given one prime and two boosters each 2 weeks apart. Group A was primed intranasally followed by two intranasal boosters, group B primed intranasally followed by two intragastric boosters and group C primed intragastrically, followed by two intragastric boosters.The absorbance (OD405nm) values of each individual animal sample are given.


ago by Adolph von Behring that rabbits and guinea pigs could be protected fromdiphtheria or tetanus toxin by administering serum of previously immunizedanimals (Behring and Kitasato, 1890). For this work he was awarded the first NobelPrize for medicine in 1901. It was discovered later that this passive immunizationdepends on the presence of antibodies in the serum, specific for these toxins. The introduction of monoclonal antibody (MAb) technology in the 1970s has led toa renewed interest in this immunization strategy. Its potential is now well recognizedand 80 monoclonal antibodies are in clinical development, not just for infectiousdiseases but also for therapy in chronic immune-mediated diseases. A major prob-lem in the oral application of antibodies is their instability in the digestive tract. Yet, most pathogens enter the host exactly via this route which therefore remains thepreferred anatomical site for delivery. The use of lactobacilli for delivery of anti-bodies appears an excellent solution to this dilemma. While bacteria cannot producethe full structure of the different disulphide-linked immunoglobulin chains thattogether make up a normal antibody, there is a way around this. By combining theantigen-specific portions VH and VL from an original MAb, an artificial single-chainprotein is constructed that retains the original antigen specificity, but is sufficientlysimple to be produced by bacteria. Several groups have demonstrated the feasibilityof this approach by producing such so-called single-chain Fv’s (scFv) in E. coli.ScFv’s are now widely studied for therapeutic use, for example in cancer treatment.An additional advantage of scFv’s is the relative ease with which the genes encoding

Fig. 4. Expression of anti-Shiga toxin scFv, coupled to an E-tag for detection purposes, in Lb. casei.(A) Western analysis. Lane 1: total cell free extract of Lb. casei, expressing cell wall anchored scFv; lane 2: total cell free extract of Lb. casei, expressing cell wall anchored GusA; lane 3: concentrated extracellular proteins of Lb. casei, secreting GusA; lane 4: total cell free extract of Lb. casei, secreting scFv;lane 5: concentrated extracellular proteins of Lb. casei, secreting scFv. (B) Fluorescence activated cell sorter (FACS) analysis showing Lb. casei with cell wall bound scFv. Bound anti-E-tag antibody was detected with optimally diluted fluorescein isothiocyanate (FITC)-conjugated anti-mouse antibody. 10 000bacteria were analysed for each experiment.


them can be manipulated, for example to achieve higher specificity, or to fuse themwith other bioactive molecules. In addition, by directing scFv’s to a bacterial sur-face, a bacterium can be made to adhere to predetermined structures.

To test the feasibility of using lactobacilli for the production of scFv’s to combatinfections we have cloned a scFv that was derived from a hybridoma cell line thatexpresses the anti-Shiga toxin monoclonal antibody 13C4 in E. coli and subse-quently in Lactobacillus expression vectors. Both secreted and cell wall anchoredscFv’s could be detected using either western analysis or a fluorescence activatedcell sorter (fig. 4). Multiple bands in the sample of cell wall anchored fraction are aresult from cell wall fragments that remain attached to the protein, that is composedof the scFv and a cell wall anchoring domain. The cell free extract of the cells thatare secreting the scFv shows two bands. The upper band represents precursor scFv,while the lower band represents processed scFv from which the signal sequence hasalready been cleaved off. In ELISA assays Shiga toxin binding activity could beshown for this scFv. Tests are currently being performed to determine the in vivoneutralizing capacity of the Lactobacillus generated anti-Shiga toxin scFv.


The past decade has witnessed a considerable increase in our knowledge of the molecular genetics of lactobacilli, a biotechnologically very important group of microorganisms. The structure of regulatory elements such as promoters has beenelucidated and the mode of regulation of gene expression has been investigated indetail. Tools have been developed for the directed integration of foreign DNA intothe chromosome of various Lactobacillus species and plasmid-derived vectors havebeen constructed with which foreign DNA sequences can be efficiently expressed,and their products be targeted to the cytoplasm, the cell surface or the medium. Yet, much needs to be learnt on how cellular processes interact, how signals aretransmitted and how the cell responds to such signals.

Lactobacilli are natural commensals of the gastrointestinal and urogenital tracts.We have, however, little understanding of the mechanisms through which lacto-bacilli colonize the mucosa. To be “on speaking terms” with its host, e.g. to avoidan immune response against the bacterium, lactobacilli have to closely interact withthe mucosal cells and emit signals to the host. How they do that is as yet unknown.To get a better understanding of these mechanisms, two adherence factors have beenstudied in some detail. The main conclusions from this research are that coloniza-tion most likely is a multifactorial mechanism in which several adhesion factorsparticipate. The structure and mode of interaction with the target receptors on themucosa of these adherence factors can greatly differ, making a multidisciplinaryapproach necessary if we are to fully understand how the interaction of a bacteriumand a host cell takes place. Recent developments in genomics, proteomics and DNA chip technology are expected to result in dramatic changes in knowledge


acquisition in these areas in the near future. The nucleotide sequences of the genomeof L. acidophilus, Lb. plantarum and L. sakei have recently been determined. Oncethe complete genome sequences are available, we will be in a position to determinethe function of all individual genes and to exploit that information for a detailedanalysis of the regulatory mechanisms. These technological developments, whichmark a revolution in biology, will have a great impact on industrial applications oflactobacilli, not only for their traditional market, the food industry, but also forapplications in public and animal health.

The developments in genetics and the demonstration of immunogenicity follow-ing delivery of recombinant lactobacilli by the oral route in animal models willendorse the development of safe Lactobacillus-based oral vaccines. Accelerating thekinetics of the response, and increasing the specific immune responses at systemicand mucosal levels, by defining optimal host–vector combinations, will provide asound basis to analyse protective efficacy. Application of this technology to otherpathogens, which manifest their pathology through the mucosal surfaces, has to beassessed. Further improvements of the mucosal immuno-adjuvanticity of lactobacilliand understanding of the basic mechanisms behind this, will also be a prerequisite toobtaining optimal protective efficacy. Recent discoveries in the involvement of regu-latory mechanisms through NF-κB in distinguishing between self and non-self in themicroflora could provide clues for a further development of mucosal vaccines, usingnon-pathogenic microorganisms.

Application of the Lactobacillus delivery systems and combination of these sys-tems with existing or newly developed mucosal vaccine systems should lead to moreknowledge of the mucosal immune system and finally to safe vaccines for both thehuman as well as animal population.

REFERENCES

Aderem, A., Ulevitch, R.J., 2000. Toll-like receptors in the induction of the innate immune response.Nature 406, 782–787.

Allison, G.E., Klaenhammer, T.R., 1996. Functional analysis of the gene encoding immunity to lactacin F,lafI, and its use as a Lactobacillus-specific, food-grade genetic marker. Appl. Environ. Microbiol. 62,4450–4460.

Behring, E.A., Kitasato, S., 1890. Ueber das Zustandekommen der Diphtherie-Immunität und derTetanus-Immunität bei Thieren. Deuts. Med. Wochen. 16, 1113–1114.

Bhakdi, S., Klonisch, P., Nuber, P., Fischer, W., 1991. Stimulation of monokine production by lipotei-choic acids. Infect. Immun. 59, 4614–4620.

Boot, H.J., Kolen, C.P., van Noort, J.M., Pouwels, P.H., 1993. S-layer protein of Lactobacillusacidophilus ATCC 4356: purification, expression in Escherichia coli, and nucleotide sequence of thecorresponding gene. J. Bacteriol. 175, 6089–6096.

Boot, H.J., Kolen, C.P.A.M., Andreadaki, F.J., Leer, R.J., Pouwels, P.H., 1996. The Lactobacillusacidophilus S-layer protein gene expression site comprises two consensus promoter sequences, oneof which directs transcription of stable mRNA. J. Bacteriol. 178, 5388–5394.

Cario, E., Rosenberg, I.M., Brandwein, S.L., Beck, P.L., Reinecker, H.C., Podolsky, D.K., 2000.Lipopolysaccharide activates distinct signaling pathways in intestinal epithelial cell lines expressingtoll-like receptors. J. Immunol. 164, 966–972.


De Ambrosini, V.M., Gonzalez, S., Perdigon, G., de Ruiz, H., Oliver, G., 1996. Chemical composition ofthe cell wall of lactic acid bacteria and related species. Chem. Pharm. Bull. 44, 2263–2267.

De Roos, N.M., Katan, M.B., 2000. Effects of probiotic bacteria on diarrhea, lipid metabolism, andcarcinogenesis: a review of papers published between 1988 and 1998. Amer. J. Clin. Nutr. 71, 405–411.

Djouzi, Z., Andrieux, C., Degrivy, M.-C., Bouley, C., 1997. The association of yogurt starters withLactobacillus casei DN 114.001 in fermented milk alters the composition and metabolism of intes-tinal microflora in germ-free rats and in human flora-associated rats. J. Nutr. 127, 2260–2266.

Drouault, S., Corthier, G., Ehrlich, S.D., Renault, P., 1999. Survival, physiology, and lysis of Lactococcuslactis in the digestive tract. Appl. Environ. Microbiol. 65, 4881–4886.

Dunne, C., Murphy, L., Flynn, S., O’Mahony, L., O’Halloran, S., Feeney, M., Morrissey, D., Thornton, G.,Fitzgerald, G., Daly, C., Kiely, B., Quigley, E.M.M., O’Sullivan, G.C.C., Shanahan, F., Collins, J.K.,1999. Probiotics: from myth to reality. Demonstration of functionality in animal models of diseaseand in human clinical trails. Antonie van Leeuwenhoek 76, 279–292.

Elewaut, D., DiDonato, J.A., Kim, J.M., Truong, F., Eckmann, L., Kagnoff, M.F., 1999. NF-κB is acentral regulator of the intestinal epithelial cell innate immune response induced by infection withenteroinvasive bacteria. J. Immunol. 163, 1457–1466.

Enouf, V., Langella, P., Commissaire, J., Cohen, J., Corthier, G., 2001. Bovine rotavirus nonstructuralprotein 4 produced by Lactococcus lactis is antigenic and immunogenic. Appl. Environ. Microbiol.67, 1423–1428.

Erickson, K.L., Hubbard, N.E., 2000. Probiotic immunomodulation in health and disease. J. Nutr. 130,403S–409S.

Finlay, B.B., Cossart, P., 1997. Exploitation of mammalian host cell functions by bacterial pathogens.Science 276, 718–725.

Finlay, B.B., Falkow, S., 1997. Common themes in microbial pathogenicity revisited. Microbiol. Mol.Biol. Rev. 61, 136–169.

Geoffroy, M.C., Guyard, C., Quatannens, B., Pavan, S., Lange, M., Mercenier, A., 2000. Use of greenfluorescent protein to tag lactic acid bacterium strains under development as live vaccine vectors.Appl. Environ. Microbiol. 66, 383–391.

Gerritse, K., Posno, M., Schellekens, M.M., Boersma, W.J., Claassen, E., 1990. Oral administration ofTNP-Lactobacillus conjugates in mice: a model for evaluation of mucosal and systemic immuneresponses and memory formation elicited by transformed lactobacilli. Res. Microbiol. 141, 955–962.

Haller, D., Blum, S., Bode, C., Hammes, W.P., Schiffrin, E.J., 2000. Activation of human peripheral bloodmononuclear cells by nonpathogenic bacteria in vitro: evidence of NK cells as primary targets. Infect.Immun. 68, 752–759.

Havenith, C.E.G., Seegers, J.F.M.L., Pouwels, P.H., 2002. Gut-associated lactobacilli for oral immunisa-tion. Food Res. Int. 35, 151–163.

Henriksson, A., Conway, P.L., 1992. Adhesion to porcine squamous epithelium of saccharide and proteinmoieties of Lactobacillus fermentum strain 104-S. J. Gen. Microbiol. 138, 2657–2661.

Hessle, C., Andersson, B., Wold, A.E., 2000. Gram-positive bacteria are potent inducers of monocyticinterleukin-12 (IL-12) while gram-negative bacteria preferentially stimulate IL-10 production. Infect.Immun. 68, 3581–3586.

Hols, P., Slos, P., Dutot, P., Reymund, J., Chabot, P., Delplace, B., Delcour, J., Mercenier, A., 1997.Efficient secretion of the model antigen M6-gp41E in Lactobacillus plantarum NCIMB 8826.Microbiology 143, 2733–2741.

Hone, D.M., Harris, A.M., Chatfield, S., Dougan, G., Levine, M.M., 1991. Construction of geneticallydefined double aro mutants of Salmonella typhi. Vaccine 9, 810–816.

Isolauri, E., Joensuu, J., Suomalainen, H., Luomala, M., Vesikari, T., 1995. Improved immunogenicity oforal D x RRV reassortant rotavirus vaccine by Lactobacillus casei GG. Vaccine 13, 310–312.

Jarosch, M., Egelseer, E.M., Mattanovich, D., Sleytr, U.B., Sara, M., 2000. S-layer gene sbsC of Bacillusstearothermophilus ATCC 12980: molecular characterisation and heterologous expression inEscherichia coli. Microbiology 146, 273–281.

Kaila, M., Isolauri, E., Soppi, E., Virtanen, E., Laine, S., Arvilommi, H., 1992. Enhancement of the cir-culating antibody secreting cell response in human diarrhea by a human Lactobacillus strain. Pediat.Res. 32, 141–144.


Khan, C.M., Villarreal-Ramos, B., Pierce, R.J., Demarco de Hormaeche, R., McNeill, H., Ali, T.,Chatfield, S., Capron, A., Dougan, G., Hormaeche, C.E., 1994a. Construction, expression, andimmunogenicity of multiple tandem copies of the Schistosoma mansoni peptide 115-131 of the P28glutathione S-transferase expressed as C-terminal fusions to tetanus toxin fragment C in a live aroattenuated vaccine strain of Salmonella. J. Immunol. 153, 5634–5642.

Khan, C.M., Villarreal-Ramos, B., Pierce, R.J., Riveau, G., Demarco de Hormaeche, R., McNeill, H., Ali, T., Fairweather, N., Chatfield, S., Capron, A., Dougan, G., Hormaeche, C.E., 1994b. Construction,expression, and immunogenicity of the Schistosoma mansoni P28 glutathione S-transferase as agenetic fusion to tetanus toxin fragment C in a live Aro attenuated vaccine strain of Salmonella. Proc.Natl. Acad. Sci. USA 91, 11 261–11 265.

Kleerebezem, M., Beerthuyzen, M.M., Vaughan, E.E., de Vos, W.M., Kuipers, O.P., 1997. Controlledgene expression systems for lactic acid bacteria: transferable nisin-inducible expression cassettes forLactococcus, Leuconostoc, and Lactobacillus spp. Appl. Environ. Microbiol. 63, 4581–4584.

Lee, M.H., Roussel, Y., Wilks, M., Tabaqchali, S., 2001. Expression of Helicobacter pylori urease subunit Bgene in Lactococcus lactis MG1363 and its use as a vaccine delivery system against H. pylori infec-tion in mice. Vaccine 19, 3927–3935.

Lee, S.F., March, R.J., Halperin, S.A., Faulkner, G., Gao, L., 1999. Surface expression of a protectiverecombinant pertussis toxin S1 subunit fragment in Streptococcus gordonii. Infect. Immun. 67,1511–1516.

Leenhouts, K., Buist, G., Kok, J., 1999. Anchoring of proteins to lactic acid bacteria. Antonie vanLeeuwenhoek 76, 367–376.

Leer, R.J., Christiaens, H., Peters, L., Posno, M., Pouwels, P.H., 1993. Gene-disruption in Lactobacillusplantarum strain 80 by site-specific recombination: isolation of a mutant strain deficient in conjugatedbile salt hydrolase activity. Mol. Gen. Genet. 239, 269–272.

Levine, M.M., Dougan, G., 1998. Optimism over vaccines administered via mucosal surfaces. Lancet351, 1375–1376.

Loeffler, J.M., Nelson, D., Fischetti, V.A., 2001. Rapid killing of Streptococcus pneumoniae with abacteriophage cell wall hydrolase. Science 294, 2170–2172.

Maassen, C.B.M., van Holten-Neelen, J.C.P.A., Balk, F., Heijne den Bak-Glashouwer, M.J., Leer, R.J.,Laman, J.D., Boersma, W.J.A., Claassen, E., 2000. Strain-dependent induction of cytokine profiles inthe gut by orally administered Lactobacillus strains. Vaccine 18, 2613–2623.

Maassen, K., 1999. Lactobacilli as antigen delivery system for mucosal tolerance in autoimmune disease.PhD thesis. Erasmus University of Rotterdam, The Netherlands.

MacCormick, C.A., Griffi, H.G., Gasson, M.J., 1995. Construction of a food-grade host/vector systemfor Lactococcus lactis based on the lactose operon. FEMS Microbiol. Lett. 127, 105–109.

Maggi, T., Oggioni, M.R., Medaglini, D., Bianchi Bandinelli, M.L., Soldateschi, D., Wiesmuller, K.H.,Muller, C.P., Valensin, P.E., Pozzi, G., 2000. Expression of measles virus antigens in Streptococcusgordonii. New. Microbiol. 23, 119–128.

Matsuzaki, T., Chin, J., 2000. Modulating immune responses with probiotic bacteria. Immunol. Cell.Biol. 78, 67–73.

McCracken, A., Timms, P., 1999. Efficiency of transcription from promoter sequence variants inLactobacillus is both strain and context dependent. J. Bacteriol. 181, 6569–6672.

McCracken, A., Turner, M.S., Giffard, P., Hafner, L.M., Timms, P., 2000. Analysis of promoter sequencesfrom Lactobacillus and Lactococcus and their activity in several Lactobacillus species. Arch.Microbiol. 173, 383–389.

Medina, E., Guzman, C.A., 2001. Use of live bacterial vaccine vectors for antigen delivery: potential andlimitations. Vaccine 19, 1573–1580.

Mercenier, A., 1999. Lactic acid bacteria as live vaccines. In: Tannock, G.W. (Ed.), Probiotics: A CriticalReview. Horizon Scientific Press, Wymondham, UK, pp. 113–127.

Miettinen, M., Matikainen, S., Vuopio-Varkila, J., Pirhonen, J., Varkila, K., Kurimoto, M., Julkunen, I.,1998. Lactobacilli and streptococci induce interleukin-12 (IL-12), IL-18, and gamma interferonproduction in human peripheral blood mononuclear cells. Infect. Immun. 66, 6058–6062.

Mollenkopf, H., Dietrich, G., Kaufmann, S.H., 2001. Intracellular bacteria as targets and carriers forvaccination. Biol. Chem. 382, 521–523.


Navarre, W.W., Schneewind, O., 1999. Surface proteins of gram-positive bacteria and mechanisms oftheir targeting to the cell wall envelope. Microbiol. Mol. Biol. Rev. 63, 174–229.

Neish, A.S., Gewirtz, A.T., Zeng, H., Young, A.N., Hobert, M.E., Karmali, V., Rao, A.S., Madara, J.L.,2000. Prokaryotic regulation of epithelial responses by inhibition of IκB-alpha ubiquitination.Science 289, 1560–1563.

Oggioni, M.R., Medaglini, D., Maggi, T., Pozzi, G., 1999. Engineering the gram-positive cell surface forconstruction of bacterial vaccine vectors. Methods 19, 163–173.

Pavan, S., Hols, P., Delcour, J., Geoffroy, M.C., Grangette, C., Kleerebezem, M., Mercenier, A., 2000.Adaptation of the nisin-controlled expression system in Lactobacillus plantarum: a tool to study invivo biological effects. Appl. Environ. Microbiol. 66, 4427–4432.

Perdigón, G., Vintini, E., Alvarez, S., Medina, M., Medici, M., 1999. Study of the possible mechanismsinvolved in the mucosal immune system activation by lactic acid bacteria. J. Dairy Sci. 82, 1108–1114.

Platteeuw, C., van Alen-Boerrigter, I., van Schalkwijk, S., de Vos, W.M., 1996. Food-grade cloning andexpression system for Lactococcus lactis. Appl. Environ. Microbiol. 62, 1008–1013.

Pouwels, P.H., Chaillou, S., 2003. Gene expression in lactobacilli. In: Wood, B.J., Warner, Ph. (Eds.), TheLactic Acid Bacteria, The Genetics of The Lactic Acid Bacteria. Plenum, New York.

Pouwels, P.H., Leer, R.J., 1993. Genetics of lactobacilli: plasmids and gene expression. Antonie vanLeeuwenhoek 64, 85–107.

Pouwels, P.H., Leunissen, J.A., 1994. Divergence in codon usage of Lactobacillus species. NucleicAcids. Res. 1994, 22, 929–936.

Pouwels, P.H., Leer, R.J., Boersma, W.J.A., 1996. The potential of Lactobacillus as a carrier for oralimmunisation: Development and preliminary characterisation of vector systems for targeted deliveryof antigens. J. Biotechnol. 44, 183–192.

Pouwels, P.H., Leer, R.J., Shaw, M., Heijne den Bak-Glashouwer, M.J., Tielen, F.D., Smit, E., Martinez, B.,Jore, J., Conway, P.L., 1998. Lactic acid bacteria as antigen delivery vehicles for oral immunisationpurposes. Int. J. Food Microbiol. 41, 155–167.

Pouwels, P.H., Vriesema, A., Martinez, B., Tielen, F.J., Seegers, J.F.M.L., Leer, R.J., Jore, J., Smit, E.,2001. Lactobacilli as vehicles for targeting antigens to mucosal tissues by surface exposition of for-eign antigens. Methods Enzymol. 336, 369–389.

Pozzi, G., Contorni, M., Oggioni, M.R., Manganelli, R., Tommasino, M., Cavalieri, F., Fischetti, V.A.,1992. Delivery and expression of a heterologous antigen on the surface of streptococci. Infect.Immun. 60, 1902–1907.

Pozzi, G., Oggioni, M.R., Manganelli, R., Medaglini, D., Fischetti, V.A., Fenoglio, D., Valle, M.T.,Kunkl, A., Manca, F., 1994. Human T-helper cell recognition of an immunodominant epitope of HIV-1 gp120 expressed on the surface of Streptococcus gordonii. Vaccine 12, 1071–1077.

Pugsley, A.P., 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol. Rev.57, 50–108.

Reid, G., Heinemann, C., Velraeds, M., van der Mei, H.C., Busscher, H.J., 2000. Biosurfactants producedby Lactobacillus. Meth. Enzymol. 310, 426–433.

Ribeiro, L.A., Azevedo, V., Le Loir, Y., Oliveira, S.C., Dieye, Y., Piard, J.C., Gruss, A., Langella, P., 2002.Production and targeting of the Brucella abortus antigen L7/L12 in Lactococcus lactis: a first steptowards food-grade live vaccines against Brucellosis. Appl. Environ. Microbiol. 68, 910–916.

Ricci, S., Medaglini, D., Rush, C.M., Marcello, A., Peppoloni, S., Manganelli, R., Palu, G., Pozzi, G.,2000. Immunogenicity of the B monomer of Escherichia coli heat-labile toxin expressed on thesurface of Streptococcus gordonii. Infect. Immun. 68, 760–766.

Schrijver, I.A., Melief, M.J., Eulderink, F., Hazenberg, M.P., Laman, J.D., 1999. Bacterial peptidoglycanpolysaccharides in sterile human spleen induce proinflammatory cytokine production by humanblood cells. J. Infect. Dis. 179, 1459–1468.

Sharma, A., Honma, K., Sojar, H.T., Hruby, D.E., Kuramitsu, H.K., Genco, R.J., 1999. Expression ofsaliva-binding epitopes of the Porphyromonas gingivalis FimA protein on the surface ofStreptococcus gordonii. Biochem. Biophys. Res. Commun. 258, 222–226.

Sharpe, M.E., 1981. The genus Lactobacillus. In: Starr, M.P., Stolp, H., Trueper, H.G., Balows, A.,Schlegel, H.G. (Eds.), The Prokaryotes. A Handbook on Habitats, Isolation and Identification ofBacteria. Springer, New York, pp. 1653–1679.


Shaw, D.M., Gaerthé, B., Leer, R.J., van der Stap, J.G.M.M., Smittenaar, C., Heijne den Bak-Glashouwer, M.-J., Thole, J.E.R., Tielen, F.J., Pouwels, P.H., Havenith, C.E.G., 2000. Engineering themicroflora to vaccinate the mucosa: serum immunoglobulin G responses and activated drainingcervical lymph nodes following mucosal application of tetanus toxin fragment C-expressing lacto-bacilli. Immunology 100, 510–518.

Slos, P., Dutot, P., Reymund, J., Kleinpeter, P., Prozzi, D., Kieny, M.P., Delcour, J., Mercenier, A., Hols, P.,1998. Production of cholera toxin B subunit in Lactobacillus. FEMS Microbiol. Lett. 169, 29–36.

Smit, E., Oling, F., Demel, R., Martinez, B., Pouwels, P.H., 2001. The S-layer protein of Lactobacillusacidophilus ATCC 4356: identification and characterisation of domains responsible for S-proteinassembly and cell wall binding. J. Mol. Biol. 305, 245–257.

Steidler, L., Robinson, K., Chamberlain, L., Schofield, K.M., Remaut, E., Le Page, R.W., Wells, J.M.,1998. Mucosal delivery of murine interleukin-2 (IL-2) and IL-6 by recombinant strains ofLactococcus lactis coexpressing antigen and cytokine. Infect. Immun. 66, 3183–3189.

Steidler, L., Hans, W., Schotte, L., Neirynck, S., Obermeier, F., Falk, W., Fiers, W., Remaut, E., 2000.Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 289, 1352–1355.

Tannock, G.W., 1999. Analysis of the intestinal microflora: a renaissance. Antonie van Leeuwenhoek 76,265–278.

Thole, J.E.R., van Dalen, P.J., Havenith, C.E.G., Pouwels, P.H., Seegers, J.F.M.L., Tielen, F.D., van derZee, M.D., Zegers, N.D., Shaw, M., 2000. Live bacterial delivery systems for development of mucosalvaccines. Curr. Opin. Mol. Ther. 2, 94–99.

Tsutsui, O., Kikeguchi, S., Matsumara, K., Kto, K., 1991. Relationship of the chemical structure andimmunobiological activities of lipoteichoic acids of Streptococcus faecalis (Enterococcus hirae)ATCC 9790. FEMS Microbiol. Immunol. 76, 211–218.

Turner, M.S., Giffard, P.M., 1999. Expression of Chlamydia psittaci- and human immunodeficiencyvirus-derived antigens on the cell surface of Lactobacillus fermentum BR11 as fusions to bspA.Infect. Immun. 67, 5486–5489.

Van Ginkel, F.W., Nguyen, H.H., McGhee, J.R., 2000. Vaccines for mucosal immunity to combat emerg-ing infectious diseases. Emer. Infect. Dis. 6, 123–132.

Velraeds, M.M., Heinemann, C., Velraeds, M., van der Mei, H.C., Busscher, H.J., 1996. Inhibition of ini-tial adhesion of uropathogenic Enterococcus faecalis by biosurfactants from Lactobacillus isolates.Appl. Environ. Microbiol. 62, 1958–1963.

Vesa, T., Pochart, P., Marteau, P., 2000. Pharmacokinetics of Lactobacillus plantarum NCIMB 8826,Lactobacillus fermentum KLD, and Lactococcus lactis MG 1363 in the human gastrointestinal tract.Aliment. Pharmacol. Ther. 14, 823–828.

Walker, R.I., 1994. New strategies for using mucosal vaccination to achieve more effective immuniza-tion. Vaccine 12, 387–400.

Wells, J.M., Wilson, P.W., Norton, P.M., Gasson, M.J., Le Page, R.W., 1993. Lactococcus lactis: high-level expression of tetanus toxin fragment C and protection against lethal challenge. Mol. Microbiol.8, 1155–1162.

Wells, J.M., Robinson, K., Chamberlain, L.M., Schofield, K.M., Le Page, R.W.F., 1996. Lactic acidbacteria as vaccine delivery vehicles. Antonie van Leeuwenhoek 70, 317–330.

Yasui, H., Shida, K., Matsuzaki, T., Yokokura, T., 1999. Immunomodulatory function of lactic acidbacteria. Antonie Van Leeuwenhoek 76, 383–389.

Zegers, N.D., Kluter, E., van der Stap, S.H., van Dura, E., van Dalen, P., Shaw, M., Baillie, L., 1999.Expression of the protective antigen of Bacillus anthracis by Lactobacillus casei: towards the devel-opment of an oral vaccine against anthrax. J. Appl. Microbiol. 87, 309–314.

351

The intestinal flora exerts a strong effect on gut-associated lymphoid tissue (GALT)activation and development of the regulatory mechanisms involved in the mainte-nance of the healthy state at the intestinal level. GALT is under constant exposureto environmental antigens, and the digestive flora is the main antigenic stimulus.Bacterial numbers and composition vary considerably along the gastrointestinaltract, constituting a complex ecosystem which depends on the physiology of thehost and on the interaction between bacteria and eukaryotic cells. The gut microflorais an important protective barrier against intestinal infections by bacteria, fungi andviruses. This knowledge is the basis for the use of live microbial food supplements(probiotics) as a method for repairing microbial deficiencies and enhancing resist-ance to disease. Lactic acid bacteria (LAB) have been used for many centuries in thepreparation and processing of food. These microorganisms are the main candidateto be used as probiotics. Currently, lactobacilli are also known for their health-stimulating activities by inducing an effect on the immune system. The knowledge ofthe role played on the immune system and on the functioning of the systemic immuneresponse as a whole, is essential for the successful development of probiotics for usein animals and humans. The natural route for the majority of pathogens to enter the body is via the mucosa of the gastrointestinal–respiratory and urogenital tracts.

16 The influence of the lactic acid bacteriaand other resident microflora on theimmune system of the growing animal1

G. Perdigóna,b, R. Fullerc and M. Medinaa

aResearch Centre for Lactobacilli (CERELA), Chacabuco 145, 4000 Tucumán,ArgentinabImmunology Department Faculty of Biochemistry, National Tucumán University,Argentinac59 Ryeish Green, Three Mile Cross, Reading RG7 1ES, UK

1This research was supported by Grants from CIUNT 26D/127 and PICT/97 No. 05-02312.



352 G. Perdigón, R. Fuller and M. Medina

Most of the information about the mechanisms on the pathway of internalizationrefers to the interaction between host and pathogen resulting in disease; however,these interactions do not always result in disease, depending on many factors suchas the immune state of the host, and the virulence and pathogenicity of the microbes.There are few reports about the interaction of non-pathogen bacteria such as LABwith the intestine and the mechanisms involved in the adjuvant capacity on the gutmucosal immune activation. There is no doubt of the immunomodulatory effectplayed by some probiotics administered by the oral route. However, how thesemicroorganisms interact with the intestinal epithelium and with the immune cellsassociated to modulation of the immune response is still poorly understood. Here,we review the role of the normal microflora in the maintenance of adequate immunestimulation and we also analyse the different immunomodulating activities of LABwhich would be related to the mechanisms of interaction or to the pathway of inter-nalization by these microorganisms with the intestine. The intensity of such inter-action induces the intercellular signals between epithelial and immune cellsassociated with the gut to stimulate the immune system. Both parameters are essen-tial for the induction and/or downregulation of the immune response.

1. ACQUISITION AND DEVELOPMENT OF GUT FLORA

The origin and development of the microflora colonizing the intestinal tract variesbetween animal species, but for all animals there is a very large number of differentmicroorganisms resident in the lower gut; this may comprise over 400 differenttypes of microorganisms, including numerous variants of each microbial speciespresent. This complex population is acquired shortly after birth and forms a symbi-otic association with the host contributing to its nutrition and immune states.

The newborn animal (or newly hatched bird) is sterile and rapidly acquires itsflora from the mother, other adults and the general environment. From the vast arrayof different organisms which the young animal will ingest, it must select those thatare required to form its characteristic microflora. Various host and flora-derived factors allow the “right” organisms to colonize the gut. By far the most importantsource of gut microorganisms is the mother, because a) she has intimate contact withthe infant and b) she provides microorganisms that are components of the flora characteristic for that species of animal.

In the case of the newborn human infant, the high level of hygiene practised inhospitals and homes of developed countries aims to prevent faecal contamination ofthe baby. However, in spite of the precautions taken, transfer of microorganismsfrom the mother to the infant does occur (Tannock et al., 1990) but, no doubt, therewill be a delay in the development of the normal flora resulting from this restrictedaccess to the mother’s microorganisms. In a sense we are being too hygienic, but wehave to strike a balance between preventing contamination by pathogens and allow-ing the transfer of desirable bacteria which constitute the indigenous microflora.

Microflora and the immune system 353

The development of probiotics containing defined microbial species has been influ-enced by this consideration.

So there is a progression from a sterile gut to one in which a large number of dif-ferent microbial species become established and influence the host animal. The speedand composition of this change will depend on the species of animal and the nature ofthe environment. Animals will vary in the structure of the gut and the food which theyeat. The ruminant animal is a unique example which has a large sac (the rumen) at theanterior end of the gastrointestinal tract. This contains a very complex population ofmicroorganisms with the strictly anaerobic species being dominant. This arrangementallows for the digestion of fibre which forms a large part of the ruminant’s diet.

The more normal organization of the gut in mammals is to have the major micro-bial presence in the lower gut (caecum and colon). For monogastric herbivores thedigestion of fibre occurs in the lower gut and supplies significant amounts of nutri-ents. For omnivores, such as the pig and humans, the digestion of fibre is less impor-tant and the volatile fatty acids (VFAs) produced in the lower gut form only a smallpart (about 7%) of the energy requirements of the host animal. Graminivorous birds,such as the chicken have two caeca attached to the rectum. These organs contain alarge number of microorganisms but no fibre digestion occurs. Even grazing birds,such as the goose, are unable to metabolize cellulose and they depend on the cellcontents of the grass for nutrients.

Humans have a low stomach pH (like carnivores) which tends to limit the survival and growth of bacteria in the stomach and upper small intestine. Humansare also unique in that a large proportion of their food is heated or processed in someway that reduces the number of viable bacteria ingested. Despite the inimical conditions in the stomach, some organisms, with special attributes, can exist there.For example, Helicobacter pylori can survive in the mucous layer of the stomachantrum. Although this organism can be the cause of gastric or duodenal ulcers, it isoften present in the gut of healthy individuals (Lee et al., 1993). This is an excellentillustration of the delicate balance that exists between the host animal and its gutmicroflora, where, according to the prevailing conditions, an organism can be aharmless commensal or a dangerous pathogen.

The total number of microorganisms in the healthy stomach rarely exceeds about104 CFU/g. But, in achlorhydric patients this will increase, showing the importance ofpH in flora control in this region of the gut. In the duodenum the pH rises and becomesslightly alkaline, but the rapid transit time of food in the small intestine prevents anylarge increases in bacterial numbers. It is not until the food enters the ileum (lower thirdof the small intestine) that numbers begin to increase. This may be due to an increaseoccurring in this region or to backflow from the colon. Lactobacilli and enterococciwhich formed the dominant flora in the stomach and small intestine, are joined in thelower gut by enterobacteriaceae and obligate anaerobes. The flora in this regionexploits the availability of the time allowed for growth by the stasis which occurs in thispart of the gut. The breakdown of polysaccharides by one section of the flora releases


simple carbohydrates which can be used by other members of the colonic flora. Thereis a nutritional interdependence of the various components of the gut flora, creating animportant balance which must be maintained to ensure the optimal health of the host.

The bacteria in the gut are not uniformly distributed through it. They colonize thewhole of the lumen by occupying numerous habitats. These include:1. Bacteria living in the crypts at the base of the villi.2. Bacteria attached to the cell membrane of the epithelial cells. A good example

of this is the Lactobacillus flora of the chicken crop which adheres to thesquamous epithelial cells (Barrow, 1992).

3. Bacteria entrapped in the mucous of the gut.4. Bacteria floating in the lumen contents.5. Bacteria attached to food particles. This is an important habitat for rumen

organisms that digest cellulose (Latham et al., 1978). The intimate associationof the bacteria and the food particle ensures a high concentration of the enzymeon the substrate surface.Association of bacteria with the epithelial surface is an important ecological

niche. By immobilizing itself on the gut wall, the microorganism can resist beingremoved by the peristaltic flow of the intestinal contents. These attached bacteria aresome of the most important organisms inhabiting the intestinal tract; they are in aposition where they can benefit from nutrients and oxygen concentrations not avail-able to those organisms that colonize sites remote from the gut wall. They are alsoable to exert their influence on the host animal either adversely by the production oftoxins or positively by modulation of the immune response (see later).

Attachment is also a very effective colonization factor. The epithelial cells withattached bacteria are continuously being sloughed off and will inoculate the incom-ing food ensuring that the attaching bacteria remain as major components of the gutmicroflora. The attachment of non-pathogenic indigenous bacteria, such as lacto-bacilli, may protect the animal host against infection by not allowing the pathogensto attach and colonize the gut. This is a feature which has been made use of in theselection of organisms for inclusion in probiotic products. There is in vitro evidencethat lactobacilli can compete with enteric pathogens for adhesion receptors on the surface of gut epithelial cells (Bernet et al., 1994). This ability to compete foradhesion receptors and the potential for immunomodulation are the most importantfeatures for determining the benefits derived from probiotic microorganisms.

1.1. Factors affecting the gut flora

The gut flora that develops in the adult animal is a complex mixture of organismswhich are interacting with each other and which are themselves influencing the hostanimal in adverse or beneficial ways. If the animal is to remain healthy, it must keepthis interaction under control; the balance of the gut flora must be maintained.However, there are several factors which can affect this balance.


1.1.1. Change in the diet

The effect of diet on the microflora is best assessed by studying the changes in meta-bolic activity (Mallett and Rowland, 1988). Changes in enzyme activity and endproduct concentration have been related to differences in polysaccharide, carbo-hydrate and protein content of the diet.

The change from milk to solid foods that occurs when mammals are weaned canhave marked effects on flora especially when, as is the case with most farm animals andmany human infants, the weaning is done at an earlier age than would occur naturally.

1.1.2. Stress

The word stress can cover a multitude of conditions such as overcrowded transportto the slaughter house or other locations or withdrawal of food, and these have allbeen related to changes in the gut microflora (Tannock, 1983). These claims shouldbe regarded with caution because it is often not possible to separate stress from otherenvironmental factors such as diet. Often the stress causes a reduction in thenumbers of lactobacilli with a corresponding increase in the numbers of entero-bacteriaceae. It is tempting to conclude that the one is responsible for the other. Thereasons for the effect of stress on gut flora are not well defined, but it is suggestedthat it may be related to the autonomic nervous control of the gut musculaturemodifying transit times, or the change in hormone balance which could affect thegut mucous living.

1.1.3. Medication

Oral administration of antibodies can often result in a condition known aspseudomembranous colitis caused by Clostridium difficile (normally present in thegut in small numbers). Although Clostridium difficile is pathogenic for human adultsit can occur in large numbers in the gut of infants without causing adverse effects(Corthier, 1997). The gut of human infants may also contain large numbers ofStaphylococcus aureus without showing symptoms of disease (Kay et al., 1990).Even though these microorganisms are producing the relevant toxins, they are without effect because the toxin receptors are absent from the gut wall. This againillustrates the complicated relationship between the host and its gut flora. The mere presence of an organism in the gut does not induce a predictable response inthe host.

1.1.4. Environment

As mentioned earlier, the use of antiseptics and a high level of hygiene can delay thetransfer of microorganisms from the mother to the infant. A difference in the speciesof Bifidobacterium found in baby’s faeces can be detected between hospitals and


even between wards of the same hospital (Lundequist et al., 1985). The age of theinfant at weaning will also be important. Premature removal of pigs and calves fromthe dam is a common practice on farms which will seriously affect the developmentof the gut flora. The effect of environment on gut flora is seen in an extreme formin the case of the incubator reared chicken. There the chick hatches into a clean envi-ronment totally divorced from any contact with the mother hen. On hatching, thechick acquires organisms from the incubator and is deficient in those bacteria thatgive protection against intestinal infections. It was found that when newly hatchedchicks were dosed with a faecal suspension from a healthy adult chicken, the resist-ance to infection found in farmyard reared chickens was restored (Nurmi andRantala, 1973). Although effective, the administration of faeces to day old chicks isundesirable and much effort has gone into identifying the organisms responsible forthis so called “competitive exclusion” phenomenon (Barrow, 1992). Other prepara-tions based mainly on lactic acid bacteria (LAB) (probiotics) have also been used torepair the deficiencies in the flora created by incubator rearing techniques.

1.2. The probiotic concept

The very large and complex population of microorganisms that develops in the gastrointestinal tract is an important part of the overall metabolic activity of the hostanimal. In addition to the nutritional aspects of the association, the gut microflora isan important protective barrier against intestinal infection by bacteria, fungi andviruses. But, as mentioned earlier, the gut flora is subject to several factors that tendto change it adversely and it is important, if optimal health is to be enjoyed, thatthese changes be reversed.

This forms the basis of the move towards the use of live microbial food supple-ments as a method for repairing the microbial deficiencies and enhancing resistanceto disease: the so-called probiotic effect. The precise mechanisms of this importantnew approach are still not clearly understood, but the gut flora presents a formidableantigenic challenge to the host animal’s immune system and it seems likely thatimmunomodulation will be a significant aspect of the way in which probiotics manifest their beneficial effects.

The remainder of this chapter is devoted to a critical review of the publishedwork on the immune potential of LAB which are the most common microorganismsused in probiotic preparations.

2. HOST RESPONSE TO THE GUT MICROFLORA. DEVELOPMENTAND MAINTENANCE OF THE MUCOSAL IMMUNE SYSTEM

Discussions of the host–microbe interaction usually concern the behaviour ofpathogens or descriptions of the effects of particular organisms rather than attempt-ing to define a more general host–microbe relationship. Commensal microbes also


interact with the host but without harmful effect. Certain members of the intestinalmicroflora are always present in high numbers throughout life and may confer benefits on the host (autochthonous); others are incidental and generally innocuousinhabitants (normal flora) that can colonize the gut from the environment, and others are transient. The presence of pathogens in the gut does not always result indisease. The harmful effect will depend on whether the pathogen is present in a highnumber, its virulence and the immune state of the host (Casadevall and Pirofski,2000). Beneficial associations between bacteria and eukaryotic cells have been studied by many researchers in recent years, and they have discovered a whole spectrum of interactions, ranging from obligatory symbiosis to loose associations.The benefits derived from these interactions include nutritional exchange and protection from infection. This has been described as important adaptations to environmental gradients (Polz et al., 1994), as well as less integrated associationsfor bacteria involved in cellulose degradation in the bovine rumen.

The understanding of the diversity of ecological niches and impact of bacteria–eukaryote cell interaction is in the early stages. It has been known for decades thatgut commensal microbes or indigenous bacteria colonizing the neonatal mammalcan influence intestinal immune function. The effect is related to the activation anddevelopment of the systemic immune system, especially to the increase of circulat-ing specific and “natural” antimicrobial antibodies (Tlaskavlová-Hogenová et al.,1983, 1994; Shroff et al., 1995). It was also demonstrated that microbes or micro-bial products participate in the granulocyte lineage formation in the bone marrow(Tada et al., 1996). Immune development does not occur in the intestine of germ-free animals. Stimulation of the intestinal immune system development by the intes-tinal microflora and host tolerance of the microflora indicate a complex symbioticmechanism, which is only beginning to be explored at the mechanistic level.

The following generalizations are the result of numerous studies comparinggerm-free and conventional animals, and observations with germ-free animals colonized with specific microbes. It has been established that: a) conventional animals develop “natural” antibody-secreting cells in the spleen and the peripherallymph nodes (PLN) against autoantigens and bacterial antigens (especiallylipopolysaccharides, LPS), whereas in germ-free neonates the development ofantibody-secreting cells is greatly delayed (Tlaskalová-Hogenová and Stepánková,1980; Tlaskalová-Hogenová, 1997); b) there is a profound hypotrophy of lymphoidtissues and lymphoid cells reactive to mitogenic stimuli (Moreau et al., 1978).Bacterial antigens such as LPS may regulate B cell development in neonates(Monroe et al., 1993). There is a delay in the competence of neonatal B cells, butneonatal and adult B cells are equally responsive to mitogenic stimulation with LPS.

Studies with germ-free animals have demonstrated alterations of immune parame-ters which include under-developed Peyer’s patches and mesenteric lymphoid nodes(MLN) and smaller but more numerous goblet cells; macrophages are present, but theiractivities (intracellular killing of bacteria) are diminished (Wells and Balish, 1980;


Rodming et al., 1982; Sotoh, 1988). Intestinal lymphocyte population and antibodyprofiles are altered in germ-free animals. The intraepithelial lymphocytes (IEL) sub-population is also altered; in conventional animals the number of antigen T cell recep-tor (TCR) αβ+ is approximately the same as or greater than the number of γδ+,depending on the mouse strain, whereas in germ-free mice the population of γδ+ IEL ispresent in higher numbers than that of αβ+ IEL (Kaeaguchi et al., 1993).

When germ-free animals are associated with bacteria, the gut morphology andthe immune system develop quickly. Peyer’s patches develop antibodies, particu-larly those specific for intestinal bacteria. After conventionalization, the IEL popu-lation is predominantly αβ+ IEL, and the γδ T population acquires the ability toregulate oral tolerance (Carter and Pollard, 1971; Kaeaguchi et al., 1993; Ke et al.,1997). The role of individual bacteria or groups of bacteria in stimulating immunity hasbeen studied using germ-free rodents; it was demonstrated that some bacterial speciesinduce a host immune response, whereas others are poorly or non-immunogenic (Bergand Savage, 1975; Shroff et al., 1995; Umesaki et al., 1996). However, it is difficultto predict which indigenous bacterial species will be immunogenic. Mice and othermammals normally harbour an extensive bacterial flora, not only in the large intes-tine, but also in the small intestine, and it is very difficult to determine the exact roleof specific populations in a gut microflora that includes facultative anaerobes andobligate anaerobes such as lactobacilli, enterococci and enterobacteria. The smalland large gut are first colonized by lactobacilli, then enterococci followed by bac-teroides in the large intestine. The enterobacteria are a minor component of the com-plete gut flora. Certain populations of enterobacteria and enterococci decrease afterhaving reached a maximum level. The presently available immunological evidenceshows that the gut mucosal immune response to some luminal microbes does notprevent their continued successful colonization (Van der Waaij et al., 1994; Frimanet al., 1996). A high proportion of the normal bacteria of the gut are found to becoated with “natural” IgA secreted into the gut lumen. It is not easy to determine theconditions for immunogenicity with allochthonous or autochthonous strains. It wasalso suggested that an autochthonous, non-immunogenic organism can induce anti-body responses if encountered outside its normal habitat. The question of why somenon-pathogenic bacteria induce immunological development, while others are appar-ently ignored by the host, remains unanswered. Although an individual autochtho-nous species may not induce a response from the host, the same strain in germ-freeanimals can induce an immune response. Certain bacterial species corresponding tothe normal microflora, individually stimulate rapid development of the intestinalimmune response. Understanding the difference between homologous bacteria whichare able or not able to influence the immune system in the same way as heterologousbacteria, will be crucial for the characterization of immunogenic strains in probioticpreparations.

The indigenous microflora is occasionally implicated in disease, for exampleulcerative colitis and Crohn’s disease (Sartor, 1995), although the exact mechanisms


responsible for disease induction remain unknown. It is also uncertain if bacteria initiate the disease process, or if bacterial-induced inflammation is a secondary con-sequence of intestinal immune dysfunction. These observations should be consideredbefore recommendations are made for ingestion of high concentrations of live organ-isms. Some of the microbial components that are highly effective inducers ofinflammatory reaction, can take part in the induction of autoimmune mechanisms(Kotzin et al., 1993). These microorganisms are capable of stimulating T or B cellsand inducing pathological symptoms, among them bacterial translocation by epithe-lial modification of tight junctions. They exhibit high antibody titres for many speciesof intestinal bacteria.

In conventional animals, the incomplete knowledge of the complex microflora ofthe intestinal tract makes it difficult to relate immune response to specific micro-organisms. Novel advances in molecular technologies are increasingly being usedfor the identification and analysis of the intestinal ecosystem and yield a betterunderstanding of the interaction between host and microbes in the intestinal tract(Vaughan et al., 1999, 2000). Populations of intestinal bacteria have been mappedby molecular techniques and shown to be constant over time (Tannock, 1999). Thereis a continuous natural process of selection exerted perhaps by the immune systemand matched by selective adaptation of specific strains of microorganisms. Recently,the influence of the major histocompatibility complex (MHC) on the composition ofthe gastrointestinal flora was also described, indicating that monozygotic twins thathave identical MHC also have identical faecal floras (Toivanen et al., 2001). Thesestudies highlight the intimate relationship between microflora and immune system.

3. BACTERIAL ADHESION TO GUT MUCOSAL SURFACE

The mucosal colonization by bacteria is preceded by attachment to epithelial cellsor to the mucin coating these mucosal cells. Cellular and molecular studies haverevealed that many bacteria express adhesins on their surfaces, which bind to recep-tors on the surface of mucosal cells. This specific binding allows the bacteria toattach firmly to particular sites on the mucosal surfaces and resist dislocation byforces that act on these surfaces. Adhesion of a bacterium leading to mucosal colo-nization determines its site and density. Other post-adhesion events enable the bac-teria to establish themselves successfully on the mucosal surfaces; even pathogenicbacteria to initiate infection need to upregulate virulence factors and to induce phys-iological changes on the mucosal surface, such as proliferation of epithelial cells,increased mucus secretion and induction of pro- and anti-inflammatory mediators,by mucosal and submucosal cells (Ofek and Beachey, 1980; Sharon, 1996). The fac-tors that determine the stability of the normal microflora are numerous.Microorganisms on mucosal surfaces such as the gut are separated from cells of themucosal immune system by epithelial barriers. To enter the external environment itis necessary for the epithelial tissue to transport the antigen across these barriers


without compromising the integrity and protective functions of the epithelium. Mostof the knowledge on bacterial adhesion to mucosal cells is based on studies of adhe-sion of pathogens, which includes interaction of adhesin–receptor (lectin, carbo-hydrate recognition) protein–protein recognition, and interaction between hydrophobicmoieties of protein and lipids, on the host cells with the bacterial surface (Silvesteret al., 1996; Sharon and Lis, 1997). On the other hand, mucosal cells are coveredby a layer of mucus composed of a heterogeneous mixture of proteins with variousdegrees of glycosylation. This layer includes several different glycoproteins includ-ing mucins, collagens, elastin, fibronectin, laminin and proteoglycans; fibronectin isan important receptor for bacterial pathogens (Hasty et al., 1994). Many mucosalcolonizers express adhesins that specifically recognize one or more of those con-stituents. These findings have stimulated the development of anti-adhesin drugs forthe prevention and therapy of microbial infections in humans.

The mechanism of adhesion of non-pathogenic bacteria to mucosal surfaces isnot well known. The polysaccharides or the lipopolysaccharides of Gram-negativebacteria can interact with lectins on macrophages and initiate an immune response(Jacques, 1996) when the antigen crosses the epithelial barrier. Bacteria possess arepertoire of strategies for triggering the immune response; however, often, mucosalresponses are marked by the development of a tolerogenic response, for example byresponses elicited by a common mucosal antigen associated with the resident micro-bial flora or with food proteins. The oral tolerance phenomenon has been mostextensively studied for soluble antigens in various animal models. The ability tomount both an immunologic and a mucosal tolerogenic response posed the questionof how the mucosal immune system knows which type of response is most appro-priate in any given circumstance. To answer this question, it was suggested that thetolerogenic response is a failure of cellular signalling elicited by the antigen admin-istered. Conversely, the immunogenic response is the result of cellular signallingelicited if the antigen is accompanied by a mucosal adjuvant or is by itself a mucosaladjuvant. Both immunogenic and tolerogenic mucosal responses are characterizedby the generation of immune effector cells. Thus, the immunity is always generatedbut, the magnitude of the response is reduced. Bacteria and all antigens derived fromthem have the ability to induce an immune response owing to the intimate interac-tion occurring at the mucosal surfaces. Microbial signals are seen by the gutimmune system as being immunogenic, but we cannot ignore the tolerance to com-mensal bacteria, which may maintain a level of suppressor mechanisms. It wasdemonstrated that oral tolerance against enteric microorganisms is broken in theevent of an inflammatory mucosal response, where both humoral and cellular immu-nity can be generated against intestinal bacterial populations (Duchman et al.,1995). Many reports have shown the complexity of the immune mechanisms thatoperate to maintain oral tolerance (Friedman and Weiner, 1994; MacDonald, 1995;Gaborious-Routhian and Moreau, 1996). This becomes even more complex if weconsider the presence of the normal microflora and the live microorganisms that


enter the gut with the consumption of food (Chin et al., 2000). This implies that sev-eral mechanisms may be operative, and the possibility of inducing an immunologicalresponse to foreign bacteria is strong. It will depend on the strength of interactionwith the epithelial cells of the gut and also on the interaction with immune cellsassociated to the lamina propria of the intestine. This aspect will be discussed later.

3.1. Bacterial interaction with gut epithelium and associated immune cells

Foreign antigens and bacteria (pathogens or non-pathogens) on the gastrointestinaland all mucosal surfaces are separated from cells of the mucosal immune system byan epithelial barrier. Thus, in order for the antigens from the external environmentto induce an effect on the mucosal immune system, it is necessary for the epithelialcells to transport the antigen across this barrier, without compromising the integrityand protective functions of the epithelium. However, this pathway of internalizationof antigens carries the risk that pathogens may exploit the mechanisms to crossepithelial barriers and invade the body. Antigen sampling strategies at diversemucosal sites differ dramatically, because they are adapted to the cellular organiza-tion of the local epithelial barrier (stratified or simple).

The gastrointestinal tract is lined with a single layer of epithelial cells; in thissimple epithelium, the intercellular spaces are sealed by tight junctions and special-ized epithelial M cells which deliver samples of foreign material from the lumen toorganized lymphoid tissues within the mucosa. Tight junctions are generally effec-tive in excluding peptides and macromolecules with antigenic potential (Madaraet al., 1990; Matter and Mellman, 1994).

The uptake of particulate antigens or bacteria across the intestinal epithelium canoccur only by active transepithelial vesicular transport, and this is restricted by mul-tiple mechanisms including local secretions containing mucins and secretory IgAantibodies. The epithelial cells have rigid microvilli and a glycocalyx (thick layer ofglycoproteins). The glycocalyx of enterocytes is an effective diffusion barrier thatcontains negatively charged mucin-like molecules (Maury et al., 1995; Neutra et al.,1996b). It prevents the uptake of bacteria (pathogens or not) by enterocytes, and alsois impermeable to most macromolecular aggregates, particles and viruses(Amerongen et al., 1994).

There is considerable evidence that the pathways of antigen internalization canbe: a) through the enterocytes that transport small amounts of intact proteins andpeptides across the epithelium (Neutra and Kraehenbuhl, 1993), although the induc-tion of immune response or immune tolerance is still controversial, and b) throughthe Peyer’s patches that are the inductor sites of mucosal immunity. In these mucosallymphoid follicles the epithelium is characterized by special epithelial cells calledfollicle-associated epithelium (FAE) cells and M cells. Both FAE and M cells allowaccess of macromolecules, particles and bacteria from the surface, and promote theiruptake by transepithelial transport (Neutra et al., 1996a). The M cell apical surface


differs from the intestinal absorptive cells in that they lack the highly organizedbrush border with microvilli typical of enterocytes, and also the uniform thick glycocalyx seen on enterocytes. Thus, the hydrolytic enzymes present in the glyco-calyx are often reduced or absent on M cells. M cells are the major pathway forendocytosed material and for induction of a mucosal immune response. From thecharacteristic of M cells, the microorganisms with a hydrophobic surface can inter-act selectively with them. The M cells are the gateways to immune inductive siteswhere the bacteria are destroyed as well as processing and presenting as antigen,which will induce immunoglobulin A (IgA) specific antibodies. Thus IgA plays animportant role as a second line of defence, eliminating invasive bacteria and so pre-venting disease (Kerr, 2000). Summarizing, the antigen uptake can be through theepithelial cell and more specifically by M or FAE cells from Peyer’s patches.However, even when epithelial M cells of the intestine are continuously exposed tothe lumen of the gut and are relatively accessible to attachment and invasion ofpathogens, the occurrence of mucosal disease may be reduced by the close interac-tions of the FAE cells with professional antigen processing and presenting cells ofmucosal lymphoid tissues immediately under the epithelium. Nevertheless, thetransport of pathogens by M cells may result in initiation of mucosal and/or sys-temic infections. A wide range of pathogenic Gram-negative bacteria, such as Vibriocholerae, Salmonella typhi, Salmonella typhimurium, Shigella and Yersinia canselectively bind M cells and initiate a variety of molecular mechanisms includingrecognition (lectin–carbohydrate interaction) followed by more intimate associa-tions that conclude with the recruitment of bacteria to the interaction site. They canalso react with the immune cells associated with Peyer’s patches, and initiate animmune response or systemic invasion, depending on the ability of M cells andassociated phagocytic cells to digest and inactivate microorganisms. When patho-genic microorganisms bind to the host epithelial cells, they activate signal transduc-tion with membrane activation; such signalling molecules may directly or indirectlyupregulate different adhesion molecules (addressins) on mucosal small venules, tofacilitate polynuclear and mononuclear cell extravasation. Those immune cells enablesome enteropathogens such as Salmonella typhimurium to disseminate into deeptissue by apoptosis induction of infected macrophages (Van der Velden et al., 2000).It is well known that the ability of a pathogen to initiate an infection is dependent onfactors such as dose and virulence. Some pathogens can be efficiently transported byM cells, but they are not equipped to survive in the mucosa or spread systematically,as is the case for V. cholerae or S. flexneri. S. flexneri is unable to invade the apical surfaces but it infects by inducing release of chemotactic signals which attract inflam-matory cells causing breakdown of normal epithelial function (Perdomo et al., 1995).

The commensal microflora plays a crucial role in the development of organizedmucosal lymphoid tissues. This microflora is able to influence the differentiation programme of mucosal epithelial cells, thereby creating favourable niches whileshaping the host’s adaptative mucosal system (Bry et al., 1996). The microorganisms


also modulate the migration of professional antigen presenting cells and lymphocytes into simple or stratified epithelia (Eckmann et al., 1995).

Most of the information about the mechanisms or pathways of internalizationrefers to pathogens and in particular the Gram-negative bacteria. There are few reportson Gram-positive pathogens and even fewer for Gram-positive non-pathogenic bacte-ria, e.g. lactic acid bacteria. Enterococcus faecalis, a Gram-positive, facultative anaer-obic bacterium that belongs to the normal flora of the intestinal tract, has increasinglygained attention as a pathogen, especially those strains resistant to all antimicrobialagents commonly used for therapy. The mechanisms involved in extraintestinal dissemination of enterococci are related to an aggregation substance protein involvedin virulence. It is expressed on the surface of E. faecalis and facilitates bacterial adher-ence and internalization by epithelial cells from the colon and duodenum, but not bycells derived from the ileum (Sartingen et al., 2000; Wells et al., 2000). Enterococcaltranslocation through the epithelial cells was not observed, but these studies indicatethat for some Gram-positive bacteria the route of internalization is not via M cells, butvia epithelial cells allowing the establishment of infection by these microorganisms.

The normal microflora by itself exerts a barrier effect against pathogens. It wasdemonstrated that some microorganisms from the flora, e.g. Bifidobacterium canhelp in the maintenance of barrier function by developing antimicrobial activityagainst pathogens (Lievin et al., 2000). The improvement of the intestinal microfloraand the immune response increase the host’s resistance to intestinal infections.It was shown that non-pathogenic, segmented filamentous bacteria (anaerobic Gram-positive microorganisms) colonize the ileum of many young animals by attachingto intestinal epithelial cells. These bacteria strongly stimulate the mucosal immunesystem and induce intestinal epithelial cells to express major histocompatibilitycomplex class II molecules. These segmented filamentous bacteria can be phago-cytosed into the ileal epithelial cells and intracellularly processed by lysosomalheterophagy, and this phagocytosis could be the first triggering step for the immuno-logical response induced (Yamauchi and Snel, 2000).

The knowledge that microorganisms from the normal microflora can stimulate theimmune system and preserve the barrier functions, led to the use of non-pathogenicmicroorganisms for the prevention and treatment of intestinal infections. In thissense, Saccharomyces boulardii (non-pathogenic yeast) was used in the treatment ofinfectious diarrhoea, because it preserves an effective barrier and can modulate thesignal transduction pathway induced by enteropathogenic E. coli (Czerucka et al.,2000; Qamar et al., 2001). S. boulardii exerts a protective effect on epithelial cellsafter adhesion of E. coli by modulating the signalling induced by bacterial infectionand provides protection against intestinal lesions caused by the pathogen.

Lactic acid bacteria are non-pathogenic, Gram-positive bacteria frequently used asprobiotics. They are associated with fermented products and, if consumed in the dailydiet, LAB can improve the intestinal microflora and the immune state. There is nodoubt about the beneficial effect of LAB on the mucosal immune system (see later).


However, the way in which these bacteria influence the immune system, how theyinteract with the gut and how they are internalized to make contact with the immunecells associated with the mucosa, are information that is essential for understandingthe behaviour of the different LAB on the mucosal immunostimulation.

In our laboratory, we have studied the pathway of internalization of different LAB,e.g. Lactobacillus casei CRL 431, L. acidophilus CRL 724, Lactobacillus plantarumCRL 936, Lactobacillus delbrueckii spp. bulgaricus CRL 423 and Streptococcusthermophilus CRL 412. These LAB were selected on the basis of previous studies ortheir effect on the mucosal immune system, after oral administration, where wedemonstrated that the LAB induced different immune responses, e.g. non-specific(inflammatory), specific or both (Vintiñi et al., 2000). These results led us to analysethese different responses, by studying the mechanisms of gut interaction, internaliza-tion and contact with the immune cells associated with the mucosa. Our aim was tounderstand how the LAB induce immunostimulation and therefore to know 1) whysome of them enhance the inflammatory immune response and others the specificresponse and 2) why not all LAB, as Gram-positive microorganisms carrying the basicstructure in their cell wall (muramyldipeptide), induce the same immunostimulation.

We performed animal studies of immunofluorescence and transmission electronmicroscopy (TEM). For the first assay, the LAB were labelled with fluorescein iso-thiocyanate (FITC) (Perdigón et al., 2000). We looked for fluorescent bacteria in histo-logical slices from the small intestine, Peyer’s patches and the large intestine. Wedemonstrated that all LAB were present in the Peyer’s patches, all except L. delbrueckiispp. bulgaricus and L. acidophilus were present in the villi of the small intestine, and onlyL. acidophilus and L. plantarum were present in the large intestine (figs 1 a,b and 2 a,b).These observations led us to outline what was the pathway of internalization to Peyer’spatches, M cells or FAE cells. To investigate this, we performed TEM studies.Unlabelled bacteria were administered by intubation and we analysed by histologicalslices of Peyer’s patches and epithelial cells from the small and large intestine, the

Fig. 1. (a) Histological slice of Peyer’s patch tissue from control animal that received unlabelled lactic acidbacteria. Magnification × 40. (b) Peyer’s patch from mice that received labelled L. plantarum (108 cells) byoral intubation. Numerous FITC-labelled bacteria are observed in Peyer’s patch. Magnification × 40.


lysosomal activation induced by antigen interaction as an indirect measure of cell acti-vation by LAB. We found that L. casei and L. plantarum interact with M cells and FAEcells. L. acidophilus, L. delbrueckii spp. bulgaricus and S. thermophilus were internal-ized to Peyer’s patches only by FAE cells and no activity in M cells was observed. L. acidophilus and L. delbrueckii spp. bulgaricus also induced a marked lysosomal acti-vation in the epithelial cells of the small intestine and L. acidophilus and L. plantaruminteracted with the epithelial cells of the large intestine (figs 3 a,b and 4 a,b).

Fig. 2. (a) Histological slice of large intestine tissue from control animals treated with unlabelled bacteria.Magnification × 40. (b) Large intestine from mice that were given labelled L. plantarum. Fluorescent bacteria can be seen. Magnification × 40.

Fig. 3. (a) Transmission electron micrograph of murine Peyer’s patch containing M and FAE cells from thecontrol animal. Magnification × 7800. (b) Electron micrograph of lymphoid cell associated with M cell frommice following oral inoculation of L. plantarum. Intense lysosomal activity in lymphoid cell and FAE cell areobserved. Magnification × 7800.


Taking into account the above results, we proposed for the LAB studied a modelof interaction with the gut as shown in fig. 5. These findings confirm that Gram-positive non-pathogenic bacteria such as LAB can stimulate the immune system bydifferent pathways of antigen uptake with M or FAE cells from Peyer’s patches orepithelial cells. These observations are important because they explain the differentimmunostimulations observed. The pathway of antigen internalization in the mucosal

Fig. 4. (a) Micrograph of control mouse epithelial cell from large intestine. Magnification × 7800. (b) Micrograph of epithelial cell from large intestine after oral administration of L. plantarum. Lysosomal andreticule ergoplasmic activity is observed. Magnification × 7800.

Fig. 5. Scheme of pathways of internalization of different lactic acid bacteria. Oral administration of LABcould modulate different immune responses depending on the route of entry.


immune system is important in the generation of a variety of distinct responses in thehost, so, if the route of entry is by M cells, a specific immune response with antibodyproduction and increase in the number of IgA secreting cells will be induced. If theinternalization is by FAE or epithelial cells, an inflammatory response would beelicited due to the transported antigen interacting with dendritic cells or macrophagesassociated with the lamina propria of the intestine with the consequent cellularmigration, release of cytokines and T cell stimulation (Sozzani et al., 1995).

There is information that the intestinal flora is the driving force of an excessive T cell response (Th1) inducing a slight inflammatory state. If stimulation is byexogenous antigens, this Th1 response can lead to undesirable effects such as colonicinflammation by the cytokines that the Th1 cells release (Strober et al., 2000).

Although we demonstrated the different ways of gut internalization by LABwhich influence the mucosal immune response, other questions should be asked. Forexample, if the LAB are internalized giving a short residence in the gut, how can weexplain the effect of probiotic microorganisms as a repair of a deficient microflora,or in their metabolic activity by beneficial enzyme production? Does it justify thelong period of administration that is required to produce the desired effect? If this isthe case, it is important to adjust the dose to avoid oral tolerance. Similarly, the useof a heterologous strain as a probiotic should be restricted to those that give theeffect that we want to obtain. Finally, how important is the use of homologousstrains? We think that this last aspect must be considered carefully for animal probiotics because of the diversity of the intestinal ecosystem and microflora. Forhumans, the situation is different because not all the strains used in commercialproducts are isolated from humans. Thus, the importance of host specificity in thechoice of probiotic strains (human and animal) should be reassessed.

4. INFLUENCE OF LAB ON THE INTESTINAL IMMUNE RESPONSE

In recent years, the food industry has indicated the importance of the consumption offunctional food as dietary supplements to promote health. Such dietary supplementsare called probiotics, and the development of probiotics is based on the knowledgethat the gut microflora is involved in resistance to disease. A probiotic is defined asa “live microbial feed supplement which beneficially affects the host animal byimproving its intestinal microbial balance” (Fuller, 1989). Probiotic administrationcauses changes in the composition and/or activity of the gut microflora and, althoughthe word probiotics originally referred to feed for farm animals, the concept has morerecently been applied to humans. The major consumption by humans is through dairyfermented products containing species of lactobacilli and bifidobacteria. Severalstudies have demonstrated that some lactic acid bacteria that can inhabit the lumen of gastrointestinal tract are beneficial to animal and human health (Holzapfel et al., 1998). It was demonstrated that fermented milk can improve digestion andassimilation of lactose in patients suffering from lactose maldigestion. It also


prevents diarrhoea and constipation and decreases the serum cholesterol level(Gilliland and Walker, 1990; Sanders, 1993; Saavedra et al., 1994).

Most of the protection attributed to probiotics may result from colonizationresistance, but need not necessarily be related to the establishment of the microbialspecies administered in the gut. Inhibition of colonization by other strains may beby competition for nutrients or attachment sites, change in pH, and production ofbacteriocins or other antimicrobial substances (Rolfe, 1996). In addition to prevent-ing colonization resistance by pathogens, the probiotics contribute to the enhance-ment of the immune response (systemic and mucosal) and can reinforce theepithelial barrier and prevent bacterial translocation. Numerous studies were con-ducted to demonstrate that probiotic strains can prevent or decrease the number ofpathogens (bacteria or virus) occurring in human or animal intestinal infections(Muralidhara et al., 1977; Kaila et al., 1995; Kimura et al., 1997).

Probiotics have also been reported to stimulate non-specific immunity includingincreased macrophage activation, natural killer (NK) cell numbers and inflamma-tory release of cytokines (Sato et al., 1988; Schiffrin et al., 1995; Haller et al., 2000;Maasen et al., 2000). Acquired or specific immune response is also enhancedincluding activation of lymphocytes with antibody production and anti-tumouractivity (De Simone et al., 1993; Link-Amster et al., 1994; Kato, 2000). A primarymechanism by which probiotics mediate immune improvement is through an adju-vant effect. This adjuvant effect is induced by whole bacteria as well as by productsof their cell wall. Probiotics have been used in immunosuppressed hosts (Perdigónand Oliver, 2000). Most information is derived from experiments with animals andit may not always be possible to extrapolate to humans.

With regard to the immunomodulatory effect of LAB on the mucosal immunesystem, since they are usually ingested as part of the normal daily diet, it is impor-tant to know their influence on the immune cells associated with the gut. In thisrespect there are many reports analysing their behaviour. Although most studies wereperformed using experimental models (Yasui et al., 1992; Famularo et al., 1997) theyare a useful basis for the design of human experiments (Perdigón et al., 2001).

In our laboratory we evaluated the mucosal adjuvant capacity of different LABorally administered with particular reference to L. casei CRL 431 (Perdigón et al.,1991, 1993, 1995; Perdigón and Pesce de Ruiz Holgado, 2000; Alvarez et al., 1998).We determined that our viable strain of L. casei was able to 1) protect against aSal. typhimurium experimental infection in the mouse, 2) enhance the secretory IgA(S-IgA) specific for the pathogen Sal. typhimurium, 3) increase the number of IgA+ cellsassociated with the lamina propria of the small intestine, 4) protect against Salmonellaup to 5 days post-treatment and 5) have a boosting effect on day 15 post-priming andto increase the protective effect against a new challenge with the pathogen (table 1).

We also determined the effect of the oral administration of L. casei, L. acidophilus,L. rhamnosus, L. delbrueckii spp. bulgaricus, L. plantarum, S. thermophilus andLactococcus lactis on bronchus immunity, by analysing the IgA secreting cells


associated with bronchus. We showed that, with the exception of L. acidophilus, theLAB were able to increase the IgA-producing cells at bronchus level (Perdigón et al.,1999a). These results were very interesting because the demonstration that oral admin-istration of LAB induced immunity at bronchus level means that those bacteriaincreased the IgA cycle by cellular migration of IgA B cells. This led us to study thenumber of CD4+ T cells on the lamina propria of the small intestine after LAB feeding.We saw that only L. casei and L. plantarum induced CD4+ T cell migration fromPeyer’s patches and they increased the number of these T cells on the laminapropria of the small intestine (Vintiñi et al., 2000).

The interaction with M cells of Peyer’s patches can induce cellular migration.Considering that the LAB produced cellular migration and a strong immuno-stimulation, we analysed whether or not the LAB are presenting and processing asantigen, by determining antibodies specific against their epitopes (a-LAB). Wefound antibodies against L. casei, L. plantarum, S. thermophilus and L. rhamnosus(Perdigón et al., 1999b) (table 2).

We demonstrated that not all the LAB are able to stimulate the mucosal immunesystem in the same way, and that the immunostimulatory capacity was not related tothe genus or species, but it was strain specific. The different immunomodulatingactivities of LAB would be related to the mechanisms of interaction of thesemicroorganisms with the intestine, and with the intensity of such interaction that ini-tiates the intercellular signals of transductions to stimulate the immune system. Ourresearch has yielded a sounder knowledge of interactions of the existing probioticLAB, however, it may also be necessary to search for new probiotic organisms withbetter probiotic activity and improved immunomodulating powers.

Table 1. Effect of Lactobacillus casei protection against Salmonella typhimurium andgut immunity

Test group Control group(treated with L. casei) (untreated with L. casei)

Assays 2 × 109 cells/day/mouse treated with NFM 10%

Protection against Sal. typhimurium. + + −Optimal dose of 2 days

Levels of S-IgA anti-Salmonella. OD = 493 nm 2.5 ± 0.03 1 ± 0.05

IgA+ secreting cells/10 villi 100 ± 10 85 ± 5Duration of preventive effect after 5 days −

priming optimal doseEffect of boosting on day 15 + + −

post-priming in prevention of Sal. typhimurium infection

Lactobacillus casei was administered at an optimal dose of 2 days, previously determined. Values are the mean of n = 5 ± SD. Results expressed as + or − represent positive or negative protection against Sal. typhimurium challenge.

NFM, non-fat milk; OD, optical density.


5. CONCLUSIONS

Progress in the knowledge of the intestinal microflora and their interaction with hostcells is essential for the use of viable microorganisms to improve animal and humanhealth. Since the human digestive tract is different in anatomy and physiology tothose of animals, the research in that field will require improved understanding ofhost intestinal physiology and its relationship with intestinal microbes. The identi-fication in the host cell surface of receptors able to bind microbes (indigenous orexogenous non-pathogenic) will allow selection of desirable microorganisms forprobiotic use. A better understanding in the conditions necessary for adhesion ofbacteria to mucin or the host cell may help to establish a reliable test for the selec-tion of strains. The value of single- or multi-strain probiotic preparations should alsobe assessed. The correct characterization and identification by molecular biology ofthe strains isolated from intestinal microflora from each ecological niche is neces-sary for the complete understanding of the significance of the gut flora. The studieson immune effects of non-pathogenic microbial strains should be performed inin vivo animal models with a complete gut flora (conventional). Although the useof mono-associated animals has provided a reasonable understanding of the role ofmicroflora on the immune system, direct transpose of these findings to explain whatis happening in the conventional animal, is not correct. Similarly, while some in vitro tests could be useful for the understanding of the probiotic effect, theyshould not be extrapolated directly to the in vivo situation.

It is also necessary to determine the importance of the viable probiotic strains,because we cannot ignore several studies that suggest that the use of cell wall or lysatepreparations is also active. The modifications induced in the intestinal micro-flora after long-term microbial administration of probiotics should be evaluated.

Table 2. Gut immunostimulation by lactic acid bacteria

Increase of Increase of Increase of IgA antibodyMicroorganisms IgA B cells in IgA+ cells in CD4+ T cells a-LABassayed LP intestine bronchus in LP intestine response

L. casei + + + + + +L. acidophilus + − − −L. rhamnosus + + − +L. plantarum + + + +L. delbr. spp. bulgaricus + + + + − −S. thermophilus + + + − +Lac. lactis + + − −

Lactic acid bacteria were orally administered to mice. Animals were sacrificed and small intestines were removed for histological techniques. B and T cells were determined by fluorescence test. IgA antibodies anti-LAB were measured in the intestinal fluid by ELISA test. Results expressed as + or – represent enhancement of cells or antibody detection.

LP, lamina propria.


The mechanisms of action and potential side-effects of each selected strain must becarefully studied to avoid compromising the important role of the intestinalmicroflora as a barrier to intestinal infections.

ACKNOWLEDGEMENTS

The authors wish to thank Dr Marta Medici for typing the manuscript, and also the co-worker from theImmunology Laboratory at CERELA and Tucumán University.

REFERENCES

Alvarez, S., Gobbato, N., Bru, E., de Ruiz Holgado P.A., Perdigón, G., 1998. Specific immunity induc-tion at the mucosal level by viable Lactobacillus casei. A perspective for oral vaccine development.Food Agr. Immunol. 10, 79–87.

Amerongen, H., Wilson, G., Fields, B., Neutra, M., 1994. Proteolytic processing of reovirus is requiredfor adherence to intestinal M cells. J. Virol. 68, 8428–8432.

Barrow, P.A., 1992. Probiotics of chickens. In: Fuller, R. (Ed.), Probiotics. The Scientific Basis. Chapmanand Hall, London, pp. 223–257.

Berg, R., Savage, D., 1975. Immune response of specific pathogen-free and gnotobiotic mice to antigensof indigenous and non-indigenous microorganisms. Infect. Immun. 11, 320–329.

Bernet, M.F., Brassart, D., Neeser, J.R., Servin, A.L., 1994. Lactobacillus acidophilus LA 1 binds to cul-tured human intestinal cells and inhibits cell attachment and cell invasion by entero-virulent bacteria.Gut 35, 483–489.

Bry, L., Falk, P., Midtvedt, T., Gordon, J., 1996. A model of host microbial interactions in an open mam-malian ecosystem. Science 273, 1380–1383.

Carter, P., Pollard, M., 1971. Host response to normal microbial flora in germ-free mice. Res. J.Reticuloendothel. Soc. 9, 580–587.

Casadevall, A., Pirofski, L., 2000. Host-pathogen interactions: Basic concepts of microbial commen-salisms, colonization, infection and disease. Infect. Immun. 68, 6511–6518.

Chin, J., Turner, B., Barchia, I., Mullbacher, A., 2000. Immune response to orally consumed antigens andprobiotic bacteria. Immunol. Cell. Biol. 78, 55–66.

Corthier, G., 1997. Antibiotic-associated diarrhoea: treatments by living organisms given by the oralroute (probiotics). In: Fuller, R. (Ed.), Probiotics 2. Application and Practical Aspects. Chapman andHall, London, pp. 40–64.

Czerucka, D., Dahan, S., Mograbi, B., Rossi, B., Rampal, P., 2000. Saccharomyces boulardii preserve thebarrier function and modulate the signal transduction pathway induced in enteropathogenicEscherichia coli infected T84 cells. Infect. Immun. 68, 5998–6004.

De Simone, C., Vesely, R., Bianchi Salvadori, B., Jirillo, E., 1993. The role of probiotics in modulationof the immune system in man and in animals. Int. J. Immunother. 9, 23–28.

Duchman, R., Kaiser, I., Hermann, E., Mayer, W., Ewe, K., Buschenfelde, K., 1995. Tolerance existstowards resident intestinal flora but is broken in active inflammatory bowel disease (IBD). Clin. Exp.Immunol. 102, 448–455.

Eckmann, L., Kagnoff, M., Fierer, J., 1995. Intestinal epithelial cells as watch dogs for the naturalimmune system. Trends Microbiol. 8, 118–120.

Famularo, G., Moretti, S., Marcellini, S., De Simone, C., 1997. Stimulation of immunity by probiotics.In: Fuller, R. (Ed.), Probiotics 2. Applications and Practical Aspects. Chapman and Hall, London,pp. 133–161.

Friedman, A., Weiner, H., 1994. Induction of anergy or active suppression following oral tolerance isdetermined by antigen dosage. Proc. Natl. Acad. Sci. USA 91, 6688–6692.

Friman, V., Adlerberth, I., Connell, H., Svanborg, C., Hanson, L., Wold, A., 1996. Decreased expressionof mannose-specific adhesions by Escherichia coli in the colonic microflora of immunoglobulin A-deficient individuals. Infect. Immun. 64, 2794–2798.


Fuller, R., 1989. Probiotics in man and animals. J. Appl. Bacteriol. 66, 365–378.Gaborious-Routhian, V., Moreau, M., 1996. Gut flora allows recovery of oral tolerance to ovoalbumin in

mice after transient breakdown mediated by cholera toxin or Escherichia coli heat-labile enterotoxin.Pediat. Res. 39, 625–629.

Gilliland, S., Walker, D., 1990. Factors to consider when selecting a culture of Lactobacillus acidophilusas a dietary adjunct to produce a hypocholesterolemic effect in humans. J. Dairy Sci. 73, 905–911.

Haller, D., Bode, C., Hammes, W., Pfeifer, A., Schiffrin, E., Blum, S., 2000. Non-pathogenic bacteria elicita differential cytokine response by intestinal epithelial cell/leukocytes co-cultures. Gut 47, 79–87.

Hasty, D., Courtney, H., Sokwrenko, E., Ofek, I., 1994. Bacterial extracellular matrix interactions. In: Kem, P. (Ed.), Fimbriae, Adhesion, Genetics, Biogenesis and Vaccines. CRC Press, Boca Raton,pp. 197–221.

Holzapfel, W., Haberer, P., Snel, J., Schillinger, U., Huis, J., 1998. Overview of gut flora and probiotics.Int. J. Food Microbiol. 41, 85–101.

Jacques, M., 1996. Role of lipo-oligosaccharides and lipopolysaccharides in bacterial adherence. TrendsMicrobiol. 4, 408–410.

Kaeaguchi, M., Nanno, M., Umesaki, Y., Matsumoto, Y., Okada, Z., Cai, Z., Shimamura, T., Matsuoka, Y.,Ohwaki, M., Ishikawa, H., 1993. Cytolytic activity of intestinal intraepithelial lymphocytes in germ-free mice is strain dependent and determined by T cells expressing δι T-cell receptors. Proc. Nat.Acad. Sci. USA 90, 8591–8594.

Kaila, M., Isolauri, E., Saxelin, M., Arvilommi, A., Vesikari, T., 1995. Viable versus inactivatedLactobacillus strain GG in acute rotavirus diarrhoea. Arch. Dis. Child. 72, 51–53.

Kato, I., 2000. Antitumour activity of LAB. In: Fuller, R., Perdigón, G. (Eds.), Probiotics 3.Immunomodulation by the Gut Microflora and Probiotics. Kluwer, Dordrecht, pp. 115–134.

Kay, B., Fuller, R., Wilkinson, A., Hall, M., McMichael, J.E., Cole, C.B., 1990. High levels ofStaphylococci in the faeces of breast-fed babies. Microbiol. Ecol. Health Dis. 3, 277–281.

Ke, Y., Pearce, J., Lake, J., Ziegler, H., Kapp, J., 1997. δι T lymphocytes regulate the induction and main-tenance of oral tolerance. J. Immunol. 158, 3610–3618.

Kerr, M., 2000. Function of immunoglobulin A in immunity. Gut 47, 751–752.Kimura, K., McCartney, A., McConnell, A., Tannock, G., 1997. Analysis of fecal populations of bifi-

dobacteria and lactobacilli and investigation of the immunological responses of their human host tothe predominant strains. Appl. Environ. Microbiol. 63, 3394–3398.

Kotzin, B., Leung, D., Kappler, J., Marrack, P., 1993. Superantigens and their potential role in human dis-eases. Adv. Immunol. 54, 99–121.

Latham, M.J., Brooker, B.E., Pettipher, G.L., Harris, P.J., 1978. Adhesion of Bacteroides succinogenes inpure culture and in presence of Ruminococcus flavetaciens to cell walls in leaves of perennial rye-grass (Lolium perenne). Appl. Environ. Microbiol. 35, 1163–1173.

Lee, A., Fox, J., Hazell., S., 1993. Pathogenicity of Helicobacter pylori: a perspective. Infect. Immun. 61,1601–1610.

Liévin, V., Peiffer, I., Hudault, S., Rochat, F., Brassart, D., Neeser, J., Servin, A., 2000. Bifidobacterium strainsfrom resident infant human gastrointestinal microflora exert antimicrobial activity. Gut 47, 646–652.

Link-Amster, H., Rochat, F., Saudan, K., Mignot, O., Aeschlimann, J., 1994. Modulation of specifichumoral immune response and changes in intestinal flora mediated through fermented milk. FEMSImmunol. Med. Microbiol. 10, 55–63.

Lundequist, B., Nord, C.E., Winberg, J., 1985. The composition of the faecal microflora in breast-fed andbottle fed infants from birth to eight weeks. Acta Ped. Scand. 74, 45–51.

Maasen, C., Laman, J., Boersma, W., Claassen, E., 2000. Modulation of cytokine expression by lactobacilliand its possible therapeutic use. In: Fuller, R., Perdigón, G. (Eds.), Probiotics 3. Immunomodulationby the Gut Microflora and Probiotics. Kluwer, Dordrecht, pp. 176–192.

MacDonald, T., 1995. Breakdown of tolerance to the intestinal bacterial flora in inflammatory boweldisease (IBD). Clin. Exp. Immunol. 102, 445–447.

Madara, J., Nash, S., Moore, R., Atisook, K., 1990. Structure and function of the intestinal epithelialbarrier in health and disease. Monogr. Pathol. 31, 306–324.

Mallet, A.K., Rowland, I.R., 1988. Factors affecting the gut microflora. In: Rowland, I.R. (Ed.), Role ofthe Gut Flora in Toxicity and Cancer. Academic Press, London, pp. 347–382.


Matter, K., Mellman, I., 1994. Mechanisms of cell polarity: sorting and transport in epithelial cells. Curr.Opin. Cell. Biol. 6, 545–554.

Maury, J., Nicoletti, C., Guzzo-Chambrand, L., Maroux, S., 1995. The filamentous brush border glyco-calyx, a mucin-like marker of enterocyte hyper-polarization. Eur. J. Biochem. 228, 323–331.

Monroe, J., Yellen-Shaw, A., Seyfert, V., 1993. Molecular basis for unresponsiveness and tolerance induc-tion in immature stage B lymphocytes. Adv. Mol. Cell Immunol. 1B, 1–32.

Moreau, M., Ducluzeau, R., Guy-Grand, D., Muller, M., 1978. Increase in the population of duodenalimmunoglobulin A plasmocytes in axenic mice associated with different living or dead bacterialstrains of intestinal origin. Infec. Immun. 21, 532–539.

Muralidhara, K., Sheggeby, G., Elliker, P., England, D., Sandine, W., 1977. Effect of feeding lactobacillion the coliform and Lactobacillus flora of the intestinal tissue and feces from pigs. J. Food Prot. 40,288–295.

Neutra, M., Kraehenbuhl, J., 1993. Transepithelial transport of protein by intestinal epithelial cells. In:Audus, K., Raub, T. (Eds.), Biological Barriers to Protein Delivery. Plenum, New York, pp. 107–129.

Neutra, M., Frey, A., Kraehenbuhl, J., 1996a. Epithelial M cells: gateways for mucosal infection andimmunization. Cell 86, 345–348.

Neutra, M., Pringault, E., Kraehenbuhl, J., 1996b. Antigen sampling across epithelial barriers and induc-tion of mucosal immune response. Ann. Rev. Immunol. 14, 275–300.

Nurmi, E., Rantala, M., 1973. New aspects of Salmonella infection in broiler production. Nature 241,210–211.

Ofek, I., Beachey, E., 1980. General concepts and principles of bacterial adherence in animals and man.In: Beachey, E. (Ed.), Bacterial Adherence Receptors and Recognition series B. Chapman and Hall,London, pp. 1–27.

Perdigón, G., Oliver, G., 2000. Modulation of the immune response of the immunosuppressed host byprobiotics. In: Fuller, R., Perdigón, G. (Eds.), Probiotics 3. Immunomodulation by the Gut Microfloraand Probiotics. Kluwer, Dordrecht, pp. 148–171.

Perdigón, G., Pesce de Ruiz Holgado, A., 2000. Mechanisms involved in the immunostimulation bylactic acid bacteria. In: Fuller, R., Perdigón, G. (Eds.), Probiotics 3. Immunomodulation by theGut Microflora and Probiotics. Kluwer, Dordrecht, pp. 213–233.

Perdigón, G., Alvarez, S., Pesce de Ruiz Holgado, A., 1991. Immunoadjuvant activity of oral Lactobacilluscasei: Influence of dose on the secretory immune response and protective capacity in intestinal infections.J. Dairy Res. 58, 485–496.

Perdigón, G., Alvarez, S., Medici, M., Pesce de Ruiz Holgado, A., 1993. Influence of the use ofLactobacillus casei as an oral adjuvant on the levels of secretory immunoglobulin A during an infec-tion with Salmonella typhimurium. Food Agr. Immunol. 5, 27–37.

Perdigón, G., Alvarez, S., Gobbato, N., Valverde de Budeguer, M., Pesce de Ruiz Holgado, A., 1995.Comparative effect of the adjuvant capacity of Lactobacillus casei and lipopolysaccharide on theintestinal secretory antibody response and resistance to Salmonella infection in mice. Food Agr.Immmunol. 7, 283–294.

Perdigón, G., Alvarez, S., Medina, M., Vintiñi, E., Roux, E., 1999a. Influence of the oral administrationof lactic acid bacteria on IgA producing cells associated to bronchus. Int. J. Immunopathol. Pharmacol.12, 97–102.

Perdigón, G., Vintiñi, E., Alvarez, S., Medina, M., Medici, M., 1999b. Study of the possible mechanismsinvolved in the mucosal immune system activation by lactic acid bacteria. J. Dairy Sci. 82,1108–1114.

Perdigón, G., Medina, M., Vintiñi, E., Valdez, J., 2000. Intestinal pathway of internalization of lactic acidbacteria and gut mucosal immunostimulation. Int. J. Immunopathol. Pharmacol. 13, 141–150.

Perdigón, G., Fuller, R., Raya, R., 2001. Lactic acid bacteria and their effect on the immune system. Curr.Issues Intest. Microbiol. 2, 27–42.

Perdomo, J., Gaunon, P., Sansonetti, P., 1995. Polymorphonuclear leukocyte transmigration promotesinvasion of colonic epithelial monolayer by Shigella flexneri. J. Exp. Med. 180, 1307–1319.

Polz, M.F., Distel, D., Zarda, B., Amann, R., Felbeck, H., Ott, A., Cavanaugh, M., 1994. Phylogeneticanalysis of a highly specific association between ectosymbiotic, sulfur-oxidizing bacteria and amarine nematode. Appl. Environ. Microbiol. 60, 4461–4467.

Qamar, A., Aboudola, S., Warny, M., Michetti, P., Pothoulakis, C., Lamont, T., Kelly, C., 2001.Saccharomyces boulardii stimulates intestinal immunoglobulin A immune response to Clostridiumdifficile toxin A in mice. Infect. Immun. 69, 2762–2765.

Rodming, C., Erlandsen, S., Wilson, D., Carpenter, A., 1982. Light microscopic morphometric analysisof rat ileal mucosa. Component quantification of Paneth cells. Anat. Res. 204, 33–38.

Rolfe, R., 1996. Colonisation resistance. In: Mackie, R., White, B., Isaacson, R. (Eds.), GastrointestinalMicrobiology. Gastroinestinal Microbes and Host Interactions. Chapman and Hall, New York,pp. 501–536.

Saavedra, J., Bauman, N., Oung, I., Perman, J., Yolken, R., 1994. Feeding of Bifidobacterium bifidum andStreptococcus thermophilus to infants in hospital for prevention of diarrhoea and shedding ofrotavirus. Lancet 344, 1046–1049.

Sanders, M., 1993. Effect of consumption of lactic cultures on human health. Adv. Food Nutr. Res. 37,67–130.

Sartingen, S., Rozdzinski, E., Muscholl-Silberhorn, A., Marre, R., 2000. Aggregation substances increaseadherence and internalization, but not translocation of Enterococcus faecalis through different intes-tinal epithelial cells in vitro. Infect. Immun. 68, 6044–6047.

Sartor, R., 1995. Insights into the pathogenesis of inflammatory bowel disease provided by new modelsof spontaneous colitis. Inflamm. Bowel Dis. 1, 64–75.

Sato, K., Saito, H., Tonioka, H., 1988. Enhancement of host resistance against Listeria infection byLactobacillus casei: activation of liver macrophages and peritonea macrophages by Lactobacilluscasei. Microbiol. Immunol. 32, 689–698.

Schiffrin, E., Rochat, F., Link-Amster, H., Aeschlimann, J., Donnet-Hughes, A., 1995. Immunomodulationof human blood cells following the ingestion of lactic acid bacteria. J. Dairy Sci. 78, 491–497.

Sharon, N., 1996. Carbohydrate-lectin interaction in infectious disease. Adv. Exp. Med. Biol. 408, 2–7.Sharon, N., Lis, H., 1997. Microbial lectins and their glycoprotein receptors. In: Montreuil J.,

Vliegenthart, J., Schachter, H. (Eds.), Glycoprotein II. Elsevier, Amsterdam, pp. 475–506.Shroff, K., Meslin, K., Cebra, J., 1995. Commensal enteric bacteria engender a self-limiting humoral

mucosal immune response while permanently colonizing the gut. Infect. Immun. 63, 3904–3913.Silvester, F., Philpott, D., Gold, B., Lastovica, A., Forstner, J., 1996. Adherence to lipids and intestinal mucin

by a recently recognized human pathogen, Campylobacter upsaliensis. Infect. Immun. 64, 4060–4066.Sotoh, Y., 1988. Effect of live and heat-killed bacteria on the secretory activity of Paneth cells in germ-

free mouse. Cell Tissue Res. 251, 87–93.Sozzani, S., Sallusto, F., Luini, W., Zhou, D., Piemonti, L., Aliavena, P., Veandemme, J., Valitutti, S.,

Lanzavecchia, A., Mantovani, A., 1995. Migration of dendritic cells in response to formyl peptides,C5a and a distinct set of chemokines. J. Immunol. 155, 3292–3295.

Strober, W., Fuss, I., Kitani, A., 2000. Counter-regulation in the mucosal immune system and the mech-anism of mucosal inflammation. Mucosal Immunol. Update 4, 3–9.

Tada, T., Yamamura, S., Kuwano, J., Abo, L., 1996. Level of myelopoiesis in the bone marrow is influ-enced by intestinal flora. Cell. Immunol. 173, 155–161.

Tannock, G., 1983. Effect of dietary and environmental stress on the gastrointestinal microbiota. In:Hentges, D.J. (Ed.), Human Intestinal Microflora in Health and Disease. Academic Press, New York,pp. 517–539.

Tannock, G., 1999. Analysis of the intestinal microflora: a renaissance. Antonie van Leeuwenhoek 76,265–278.

Tannock, G.W., Fuller, R., Smith, S.L., Hall, M.A., 1990. Plasmid profiling of members of the familyEnterobacteriaceae, Lactobacilli and Bifidobacteria to study the transmission of bacteria from motherto infant. J. Clin. Microbiol. 28, 1225–1228.

Tlaskalová-Hogenová, H., 1997. Gnotobiology as a tool. In: Lefkovits, I. (Ed.), Manual ofImmunological Methods. Academic Press, London, pp. 1524–1527.

Tlaskalová-Hogenová, H., Stepánková, R., 1980. Development of antibody formation in germ-free andconventionally reared rabbits: the role of intestinal lymphoid tissue in antibody formation to E. coliantigens. Folia Biol. (Praha) 26, 81–93.

Tlaskalová-Hogenová, H., Sterzl, J., Stepánková, R., Dlabac, V., Vetvicka, V., Rossmann, P., Mandel, L.,Rejnek, J., 1983. Development of immunological capacity under germfree and conventional condi-tions. Ann. N.Y. Acad. Sci. 409, 96–113.



Tlaskalová-Hogenová, H., Mandel, L., Trbichavsky, I., Kovaru, F., Barot, R., Sterzl, J., 1994.Development of immune responses in early pig ontogeny. Vet. Immunol. Immunopath. 43, 135–142.

Toivanen, P., Vaahtorrio, J., Eerola, E., 2001. Influence of the major histocompatibility complex on bac-terial composition of fecal flora. Infect. Immun. 69, 2372–2377.

Umesaki, Y., Okada, Y., Matsumoto, S., Imaoka, A., Setoyama, H., 1996. Segmented filamentousbacteria are indigenous intestinal bacteria that activate intraepithelial lymphocytes and induce MHCClass II molecules and fucosyl asielo GM1 glicolipids on the small intestinal epithelial cells in theex-germ-free mouse. Microbiol. Immunol. 39, 555–562.

Van der Velden, A., Lindgren, S., Worley, M., Heffron, F., 2000. Salmonella pathogenicity island1-independent induction of apoptosis in infected macrophages by Salmonella enteric serotypetyphimurium. Infect. Immun. 68, 5702–5709.

Van der Waaij, L., Mesander, G., Limburg, P., van der Waaij, D., 1994. Direct flow cytometry of anaer-obic bacteria in human feces. Cytometry 16, 270–279.

Vaughan, E., Mollet, B., de Vos, W., 1999. Functionality of probiotics and intestinal lactobacilli: light inthe intestinal tract tunnel. Curr. Opin. Biotech. 10, 505–510.

Vaughan, E., Schut, F., Heilig, H., Zoetendal, E., de Vos, W., Akkermans, A., 2000. A molecular view ofthe intestinal ecosystem. Curr. Issues Intest. Microbiol. 1, 1–12.

Vintiñi, E., Alvarez, S., Medina, M., Medici, M., Budeguer, M.V., Perdigón, G., 2000. Gut mucosalimmunostimulation by lactic acid bacteria. Biocell. 24, 223–232.

Wells, C., Balish, E., 1980. Modulation of the rat’s immune status by monoassociation with anaerobicbacteria. Can. J. Microbiol. 26, 1192–1198.

Wells, C., Moore, E., Hoag, J., Hirt, H., Dunny, G., Erlandsen, S., 2000. Inducible expression ofEnterococcus faecalis aggregation substance surface protein facilitates bacterial internalization bycultured enterocytes. Infect. Immun. 68, 7190–7194.

Yamauchi, K., Snel, J., 2000. Transmission electron microscopic demonstration of phagocytosis andintracellular processing of segmented filamentous bacteria by intestinal epithelial cells of the chickileum. Infect. Immun. 68, 6496–6504.

Yasui, H., Nagaoka, A., Mike, A., Hayakawa, K., Ohwaki, M., 1992. Detection of Bifidobacterium strainsthat induce large quantities of IgA. Microbiol. Ecol. Health Dis. 5, 155–162.

379

The chapter provides a “state of the art” for the use of probiotics (live microbial cul-tures that improve the health of the host) in aquaculture. In fish farming, probioticcultures are not exclusively aimed at the gastrointestinal tract but may just as wellact via skin or gills. Some cultures can therefore be added directly to tank or pondwater. Current strategies for selecting potential fish or crustacean probiotic culturesare outlined, including in vitro antagonism, survival, adhesion to mucus and aggre-gating abilities. However, there is a lack of scientific evidence comparing in vitroactivities with in vivo effects. A multitude of studies have, based on in vitro tests,reported on the isolation of what are believed to be potential fish or crustaceanprobiotic cultures. The chapter emphasizes the need for in vivo testing in challengemodel or field trials as a prerequisite before the term probiotic can be used about aculture. Many in vivo trials are hampered by too few replicates and/or poor statisti-cal analyses and the need for improvement in this area is discussed. An importantarea of future research will be to gain insight into mechanisms, stability, resistanceand specificity of fish probiotics, and knowledge of bacterial physiology and molecu-lar approaches is needed. Despite the shortcomings outlined, several trials do docu-ment that, in some systems, live microbial cultures can indeed be used to reducedisease and/or decrease mortality. Probiotics should especially be investigated in therearing of fin fish larvae or shellfish where vaccination is difficult or not possible.

1. INTRODUCTION

World fish production, including wild catches, has increased steadily during the pastseveral decades (FAO, 1998), however, catches from the wild populations have

17 Prospects of fish probiotics

L. Grama and E. Ringøb

aDanish Institute for Fisheries Research, Department of Seafood Research,Søltofts Plads, c/o Technical University of Denmark bldg. 221, DK-2800 Kgs.Lyngby, DenmarkbSection of Arctic Veterinary Medicine, Department of Food Safety and InfectionBiology, The Norwegian School of Veterinary Science, NO-9292 Tromsø, Norway



380 L. Gram and E. Ringø

remained static at approximately 90 million metric tonnes (mt) per year since 1990.This is currently believed to be the maximum sustainable yield from this resource.In contrast, the contribution from reared fish is increasing, and currently approxi-mately 30 million mt per year are produced by mari- and aquaculture (fig. 1). In1996, aquaculture provided 20% of global fisheries production (and 29% of foodfish) (FAO, 1998). This makes aquaculture the fastest growing protein producingsector in the world. This development is desired as one source of food to the grow-ing world population. The aquaculture industry has also become a major source ofincome for several countries, particularly in Southeast Asia, such as Thailand, wherethe culture of crustaceans is an expanding industry (fig. 2).

Fig. 1. World global fishproduction – catch andaquaculture (FAO, 1998).

Fig. 2. Cultured and wild-capturedshrimp harvests in Thailand(Dierberg and Kiattisimkul, 1996;cited from FAO/NACA, 1995).

Disease is one of the major constraints on the aquaculture sector, as fish in inten-sive rearing experience stress and rapid spread of infections. Thus, the sustainablecontrol of fish diseases has become one of the most important prerequisites for thefuture development of aquaculture. In the early days of commercial large-scaleaquaculture, treatment with antibiotics was the method of choice for controlling fishdisease. The widespread use of high amounts of antibiotics has caused muchconcern because of the risk of developing antibiotic resistance (Sørum, 1999).Antibiotic resistance may be detrimental to commercial fish production as the fishpathogen is no longer sensitive (Alderman and Hastings, 1998). Thus, prophylacticuse of antibiotics in a shrimp (Penaeus monodon) hatchery led to antibiotic resistantVibrio harvyei that was responsible for mass mortalities and could not be controlled by addition even of very high levels of antibiotics (e.g., 1000 μg/ml ofchloramphenicol, erythromycin and nifurpirinol) to which the organism is normallysensitive (Karunasagar et al., 1994).

Although the resistance of fish pathogens has immediate serious effects for theaquaculture sector even worse scenarios can be envisaged if such resistance spreadsto human pathogens (Levy, 1998). Several studies have shown that the microflora inaquaculture environments is more resistant to antibiotics than microorganisms inenvironments not exposed to antibiotics (Austin and Al-Zahrani, 1988; Spanggaardet al., 1993; DePaola et al., 1995; Schmidt et al., 2000). Thus, during treatment withoxytetracycline, approximately 80% of the intestinal Gram-negative bacteria fromchannel catfish (Ictalurus punctatus) were resistant to the compound (DePaola et al.,1995). Despite reports on occurrence of antibiotic resistance in fish farms, there iscurrently no information available about the potential transfer to human pathogenicbacteria (MAFF, 1998) and the risk may be smaller than assumed. More recently, theUS Food and Drug Administration (FDA) conducted an investigation of the preva-lence of Salmonella in imported foods and the level of antibiotic resistance (Zhao,2001). Of 187 Salmonella strains investigated, 15 were resistant to one or severalantibiotics, and of these strains, 10 were isolated from imported seafoods, particu-larly from Southeast Asian countries. Whilst this could be caused by an uncontrolleduse of antibiotics in clinics or agriculture eventually leaking to the aquatic environ-ment, it could also be linked to improper use of antibiotics in aquaculture.

No statistics are available on the worldwide use of antibiotics in aquaculture. In a study of 11 000 aquaculture farms in Southeast Asia in 1995, it was found thatonly 5% of inland carp farms used antibiotics (FAO/NACA/WHO, 1999) and, similarly, few of the coastal shrimp farmers used antibiotics. In contrast, in intensiveshrimp farming, there was a higher frequency of antimicrobial use with oxytetra-cycline and oxolinic acid being the main compounds.

During the past 10 years, significant advances have been made in several moreenvironmentally compatible disease control measures. Of major importance is thedevelopment of fish vaccines that now enable the control of a range of bacterial fishpathogens in several fish species. Thus, the use of antibiotics in Norway has

Prospects of fish probiotics 381

decreased from 48 570 to 591 kg active substance per year from 1987 to 1999 –whilst the production of Atlantic salmon has increased from 60 000 to 464 000 mtper year during the same period (Grave et al., 1990, 1996; Alderman and Hastings,1998; Norm-Vet, 1999). This significant reduction in the use of antibiotics is primarily due to the introduction and use of efficient vaccines. However, vaccinationof fish smaller than 35 g is not recommended. Also, during hatching and in the larval stage, vaccination may not be possible due to the slow development of the fishimmune system and the small size of the fish. Owing to the lack of acquired immu-nity in crustaceans and molluscs, conventional vaccination cannot be used as adisease control strategy for such organisms.

One alternative means of disease prevention that has attracted attention duringrecent years is the addition or inclusion of assumed beneficial bacteria to rearingwater or feed, the so-called probiotics (Gatesoupe, 1999; Gomez-Gil et al., 2000;Verschuere et al., 2000a). This chapter reviews some of the findings in terms of (in vitro) strategies for selecting probiotics and their use in vivo. Whilst the overallconclusion of the chapter is that there is indeed reason to follow this path, the readershould be alerted that one common feature of probiotic research whether in warm-and cold-blooded animals is the huge variability of experimental results, particularlyin in vivo studies (Berg, 1998). As will become evident, no studies so far have shownwhich strategy is optimal for selection of probiotic cultures, and a better under-standing of the fish indigenous flora and of the basic ecological principles control-ling its development must be built. Further, results from in vitro antagonism testsmust be compared with in vivo experiments. To understand the areas in which theprobiotic concept is viable, we must understand the mechanisms by which suchtreatments work (Atlas, 1999). This will require detailed knowledge of:

the pathogen, its virulence, its proliferation and invasion sites;the host, its immune defence, and its natural microflora;the surrounding environment, including nutrients, microorganisms, etc.;the probiont, its functional features, its mechanisms of action, and its effect

on the general microflora, etc.A range of microorganisms has either been suggested or evaluated as fish

probiotics. These include lactic acid bacteria (LAB) (Gatesoupe, 1994; Gildberg et al., 1995, 1997; Gildberg and Mikkelsen, 1998; Robertson et al., 2000), Bacillusspecies (Moriarty, 1998), Pseudomonas species (Bly et al., 1997; Gram et al., 1999;Spanggaard et al., 2001), Vibrio species (Austin et al., 1995) and other Gram-negative bacteria (Nogami et al., 1997).

As well as adding a beneficial microorganism to the system, attempts have alsobeen made to increase the disease resistance in fish by other means than probiotics.Several compounds, such as glucans and vitamins, can have an immunomodulatingeffect (Raa, 1996; Sakai, 1999), and it has been suggested that certain polysaccha-ride compounds can be used as prebiotics, i.e. host-indigestible compounds that canbe metabolized by “beneficial” microorganisms colonizing the digestive tract


(Roberfroid, 1993, 1998, 2001). However, to date no information is available onprebiotics and their effect on intestinal fish microflora and fish welfare. Other disease control attempts include the use of “natural compounds” as agents againstpathogens: e.g. the reduction of Vibrio alginolyticus infection in juvenile rockfish(Sebastes schlegeli) fed an aloe-containing diet (Kim et al., 1999).

2. PROBIOTICS: DEFINITION OF TERMINOLOGY

When initially introduced in the 1970s, the term “probiotic” described microbialfeed supplements (Berg, 1998) and stems from combining the Latin word pro (for)with the Greek word bios (for life) (Zivkovic, 1999). Fuller (1989) further detailedthis as “live microbial feed supplements which beneficially affect the host animal byimproving its intestinal microbial balance”. Since then, many variations to the definition(s) have been proposed (table 1). As discussed below, the term “microbialbalance” is ill-defined and “The Lactic Acid Bacteria Industrial Platform” at theirworkshop in 1995 provided a more concise definition of the term as “oral probioticsare live microorganisms which upon ingestion in certain numbers, exert health benefits beyond inherent basic nutrition” (Guarner and Schaafsma, 1998). In thischapter, we define a probiotic as a live microbial preparation that when added to thefish, crustaceans or molluscs (larvae, fry, young or adult animals) has a beneficialeffect on the health of the host.

Some points should be made with respect to the definitions proposed. The term probiotic originates from warm-blooded animals where the importance

of the gastrointestinal tract for onset of diseases and of the mucosal immune systemis clearly recognized. In contrast, in fish (and other cold-blooded animals), areas suchas skin and gills are probably as important sites for proliferation of pathogens and initiation sites for the infection. Therefore, in the context of fish and aquatic animals,a probiotic is a live microbial supplement added to the host environment or feed. By expanding the area of application, the term probiotic parallels the “biocontrol”term used for microbial cultures applied to prevent plant (rhizosphere) disease.

Several definitions imply that the probiotic “beneficially affects the microbialbalance”. The term “microbial balance” is not well defined and whilst it must beassumed that this refers to a change in the host microbial community to diminishnumbers and/or metabolism of pathogenic organisms, it may be difficult and timeconsuming to measure. There are examples in the literature where addition of a probiotic culture (Bacillus) to pond water decreased the numbers of potentially pathogenic bacteria (luminescent vibrios) in penaeid culture and subsequentlyincreased survival of the animals (Moriarty, 1998). Furthermore, Skjermo et al.(1997) showed that so-called “microbial maturation” of water may increase the survival of Atlantic halibut (Hippoglossus hippoglossus). This process involveschanging the microflora of pond water from opportunistic, substrate-(nutrient)-requiring microorganisms (r-strategists) to non-opportunistic so-called K-strategists


Tabl

e 1.

Def

initi

ons

of p

robi

otic

s by

var

ious

aut

hors

Def

initi

onR

efer

ence

Com

men

ts

Mic

robi

al c

ompo

unds

whi

ch p

rom

ote

body

fun

ctio

ns a

nd b

enef

icia

l mic

roor

gani

sms

Ver

gin,

195

4Pr

oduc

ts –

not

live

cul

ture

?M

icro

bial

ly p

rodu

ced

“fac

tors

” w

hich

pro

mot

e gr

owth

of

othe

r or

gani

sms

Lill

y an

d St

illw

ell,

1965

Gro

wth

pro

mot

ion

Ani

mal

fee

d su

pple

men

ts –

org

anis

ms

and

subs

tanc

es th

at h

ave

a be

nefi

cial

eff

ect

Park

er, 1

974

Onl

y fe

ed w

hat i

s “m

icro

bial

on th

e ho

st a

nim

al b

y co

ntri

butin

g to

its

inte

stin

al m

icro

bial

bal

ance

bala

nce”

Liv

e m

icro

bial

fee

d su

pple

men

ts w

hich

ben

efic

ially

aff

ect t

he h

ost a

nim

al b

yFu

ller,

1989

Onl

y fe

ed w

hat i

s “m

icro

bial

impr

ovin

g its

inte

stin

al m

icro

bial

bal

ance

bala

nce”

A m

icro

bial

die

tary

adj

uvan

t tha

t ben

efic

ially

aff

ects

the

host

phy

siol

ogy

byN

aidu

et a

l., 1

999

Onl

y fe

ed w

hat i

s “m

icro

bial

mod

ulat

ing

muc

osal

and

sys

tem

ic im

mun

ity, a

s w

ell a

s im

prov

ing

nutr

ition

alba

lanc

e”an

d m

icro

bial

bal

ance

in th

e in

test

inal

trac

tL

ive

mic

roor

gani

sms

supp

lem

ente

d in

foo

d or

fee

d w

hich

giv

e be

nefi

cial

eff

ects

Gild

berg

et a

l., 1

997

Onl

y fe

ed w

hat i

s “m

icro

bial

on th

e in

test

inal

mic

robi

al b

alan

ceba

lanc

e”A

live

mic

robi

al s

uppl

emen

t whi

ch b

enef

icia

lly a

ffec

ts th

e ho

st a

nim

al b

yG

ram

et a

l., 1

999

Wha

t is

“mic

robi

al b

alan

ce”?

impr

ovin

g its

mic

robi

al b

alan

ceIs

bas

ed o

n el

imin

atio

n of

har

mfu

l mic

rofl

ora

from

the

anim

al’s

dig

estiv

e tr

act

Bog

ut e

t al.,

199

8E

limin

atio

n? O

nly

diet

ary

trac

tM

ono

or m

ixed

cul

ture

s of

live

mic

roor

gani

sms

whi

ch w

hen

appl

ied

to a

nim

alH

aven

aar

et a

l., 1

992

Wha

t are

“pr

oper

ties

of

or m

an, b

enef

icia

lly a

ffec

t the

hos

t by

impr

ovin

g th

e pr

oper

ties

of th

e C

onw

ay, 1

996

indi

geno

us m

icro

flor

a”?

indi

geno

us m

icro

flor

aH

olza

pfel

et a

l., 1

998

Via

ble

mic

roor

gani

sms

(bac

teri

a or

yea

sts)

that

exh

ibit

a be

nefi

cial

eff

ect o

n th

eSa

lmin

en e

t al.,

199

8he

alth

of

the

host

whe

n th

ey a

re in

gest

edB

enef

icia

l bac

teri

a w

hich

may

ove

rrid

e pa

thog

ens

by p

rodu

cing

inhi

bito

ryR

ique

lme

et a

l., 2

000

Wha

t is

“ove

rrid

e”?

subs

tanc

es, o

r by

pre

vent

ing

path

ogen

ic c

olon

izat

ion

of th

e ho

stB

enef

icia

l bac

teri

a th

at d

ispl

ace

path

ogen

s by

com

petit

ive

proc

esse

s or

by

rele

ase

Mor

iart

y, 1

997

of g

row

th in

hibi

tors

Liv

e in

test

inal

bac

teri

a th

at a

re a

dded

to p

rom

ote

the

viab

ility

of

the

host

, but

the

Skje

rmo

and

Vad

stei

n, 1

999

How

is “

colo

niza

tion”

reg

ulat

ed?

term

is a

lso

prop

er f

or b

acte

ria

able

to r

egul

ate

colo

niza

tion

of th

e ou

ter

surf

aces

Liv

e m

icro

bial

adj

unct

whi

ch h

as a

ben

efic

ial e

ffec

t on

the

host

by

mod

ifyi

ng th

eV

ersc

huer

e et

al.,

200

0aho

st-a

ssoc

iate

d or

am

bien

t mic

robi

al c

omm

unity

, by

ensu

ring

impr

oved

use

of

the

feed

or

enha

ncin

g its

nut

ritio

nal v

alue

, by

enha

ncin

g th

e ho

st r

espo

nse

tow

ards

dise

ase,

or

by im

prov

ing

the

qual

ity o

f its

am

bien

t env

iron

men

tO

ral p

robi

otic

s ar

e liv

ing

mic

roor

gani

sms,

whi

ch u

pon

inge

stio

n in

cer

tain

Gua

rner

and

Sch

aafs

ma,

nu

mbe

rs, e

xert

hea

lth b

enef

its b

eyon

d in

here

nt b

asic

nut

ritio

n19

98

Liv

e m

icro

bial

cul

ture

s ad

ded

to f

eed

or e

nvir

onm

ent (

wat

er)

to in

crea

se v

iabi

lity

Thi

s ch

apte

r(s

urvi

val)

of

the

host


with high substrate (nutrient) affinity (Salvesen et al., 1999). Despite these studiesthat show that alterations in microbial composition may affect fish survival, this maynot be a prerequisite for a successful probiotic culture. Therefore, we propose thatthe effect of a probiotic be measured by its ability to decrease frequency of diseaseand/or increase survival from lethal diseases.

Live microbial cultures may also be used in aquaculture for other purposes suchas feed or water quality improvement that indirectly will benefit the health/survivalof the aquatic organism (Moriarty, 1997). If a clear effect on, e.g. health condition(disease survival) can be documented, we suggest that such preparations be denotedprobiotics. Otherwise, they should be classified as “feed”, “water treatment” etc.(Gatesoupe, 1999). Examples of this include the nitrification (e.g. the conversion oftoxic NH3 and NO2

− to NO3−) of water by various bacterial species such as

Nitrosomonas and Nitrobacter (Hagopian and Riley, 1998), and the enhancement offish growth by microorganisms that has been observed in several studies (Byun et al., 1997; Bogut et al., 1998; McCausland et al., 1999; Thompson et al., 1999).

In an attempt to clarify some of the inconsistencies, we define a probiotic as a live microbial preparation that when added to the fish, crustacean or mollusc (larvae, fry, young or adult animals) has a beneficial effect on the health of the host.Most commonly a beneficial effect (which it can be argued is somewhat imprecise)refers to reduced mortality (or increased survival). We do not use “probiotic” todescribe microorganisms that improve water quality, e.g. by denitrifying, ormicroorganisms that serve as food supply. Whilst most studies have used bacterialcultures as potential probiotics, yeast (Scholz et al., 1999) has also been suggestedas a disease control measure. The addition of microalgae to the rearing waters oflarvae presents an intermediate. Whilst this use of “green water” improves thegrowth rate of several larvae, and thereby classifies the algae as feed, it alsoimproves survival and thereby qualifies as a probiotic (Alves et al., 1999; Planas andCunha, 1999). It has also been suggested that the addition of viruses specific for thefish/larval pathogen could be used to prevent disease (Park et al., 2000).

3. MICROFLORA OF FISH

Whilst no clear strategy exists for the isolation and selection of fish probiotic cul-tures, most studies have based their work on isolates from the natural microflora offish or fish larvae. This is due to the assumption that pathogen-antagonizing bacte-ria from the fish or fish larvae are better adapted for this niche than cultures derivedfrom terrestrial or other sources. Fish and other aquatic organisms harbour a micro-bial flora on the surfaces of skin, gills and in the intestinal tract, either in the lumenor associated to mucus or epithelial cells. Fish live in extreme environments, fromthe cold Arctic and Antarctic oceans to tropical freshwater lakes. Despite these dif-ferences in environmental conditions, certain general patterns emerge in terms ofcomposition of the culturable microflora (Cahill, 1990; Hansen and Olafsen, 1999;Ringø and Birkbeck, 1999; Gram and Huss, 2000; table 2).

Tabl

e 2.

Bac

teri

al g

ener

a as

soci

ated

with

var

ious

raw

fin

fish

and

cru

stac

eans

, and

thei

r pe

rcen

tage

of

the

tota

l mic

rofl

ora

(IC

MSF

, 200

4)

% c

ompo

sitio

n

Fish

type

Ref

eren

ce

Mar

ine

fish,

tem

pera

teN

orth

Sea

fis

h (1

932)

51

5611

5—

23—

—Sh

ewan

, 197

1N

orth

Sea

fis

h (1

960)

162

2327

1018

4—

—Sh

ewan

, 197

1N

orth

Sea

fis

h (1

970)

221

4110

21—

1—

—Sh

ewan

, 197

1H

addo

ck (

Nor

th A

tlant

ic)

262

4515

—4

42

—L

ayco

ck a

nd R

egie

r, 19

70Fl

atfi

sh (

Japa

n)21

2922

13—

132

——

Sim

idu

et a

l., 1

969

“Pes

cada

” (B

razi

l)32

—35

518

44

—3

Wat

anab

e, 1

965

Shri

mp

(Nor

th P

acif

ic)

10—

4722

—3

74

8H

arri

son

and

Lee

, 196

9Sc

ampi

(U

K)

3—

112

—81

——

3W

alke

r et

al.,

197

0M

arin

e fis

h, t

ropi

cal

Mul

let (

Aus

tral

ia)

18—

98

—12

512

—G

illes

pie

and

Mac

rae,

197

5Pr

awn

(Ind

ia)

1110

236

—13

6—

—Su

rend

ran

et a

l., 1

985

Sard

ine

(Ind

ia)

2028

304

——

7—

11Su

rend

ran

et a

l., 1

989

Shri

mp

(Tex

as G

ulf)

222

149

140

11—

2V

ande

rzan

t et a

l., 1

970

Shri

mp

(Tex

as, p

ond)

2—

1525

1243

30.

51

Van

derz

ant e

t al.,

197

1F

resh

wat

er fi

sh, t

empe

rate

Pike

(Sp

ain)

10—

15—

5510

—5

5G

onzá

lez

et a

l., 1

999

Bro

wn

trou

t (Sp

ain)

—6

——

40—

15—

34G

onzá

lez

et a

l., 1

999

Tro

ut (

Spai

n, r

eare

d)11

—7

—26

545

—6

Gon

zále

z et

al.,

199

9T

rout

(D

enm

ark,

rea

red)

193

50—

186

4—

—Sp

angg

aard

et a

l., 2

001

Fre

shw

ater

fish

, tro

pica

lN

ile p

erch

(K

enya

)6

243

—9

530

5—

Gra

m e

t al.,

199

0C

atfi

sh (

Indi

a)—

—10

——

—50

40—

Ven

kata

rana

n an

d Sr

eeni

vasa

n, 1

953

Car

p (I

ndia

)—

—20

1130

—39

——

Ven

kata

rana

n an

d Sr

eeni

vasa

n, 1

953

Pseudomonas

Vibrionaceae

Acinetobacter–Moraxella

Flavobacterium–Cytophaga

OtherGram-negative

Coryneforms

Gram-positivecocci

Bacillusspp.

Otherornot

It should be borne in mind that whilst the majority of the bacteria in the gastrointestinal tract and on fish surfaces are probably culturable (Spanggaard et al.,2000a), some studies using molecular techniques clearly demonstrate large propor-tions of unculturable bacteria associated with the gastrointestinal tract and fishsurfaces (Bernadsky and Rosenberg, 1992; Huber et al., 2004). Thus, this area ofmicrobial ecology would benefit from the use of molecular techniques such as dena-turing gradient gel electrophoresis and in situ rRNA hybridization methods for theevaluation of microflora composition.

3.1. Fish eggs and larvae

The interior of the fish egg is assumed to be sterile, but the outer surface of Atlantichalibut, Atlantic cod, turbot and Atlantic salmon eggs is typically occupied by non-fermentative Gram-negative bacteria such as Cytophaga, Flavobacterium andPseudomonas (Bell, 1966; Cahill, 1990; Bergh, 1995; Hansen and Olafsen, 1999).Upon hatching, the yolk-sac larvae start to drink (Magnor-Jensen and Adoff, 1987;Reitan et al., 1998), and, as a natural consequence, the intestinal tract of non-feeding larvae becomes colonized by the same genera present in the rearing water(Strøm and Ringø, 1993; Bergh et al., 1994). Furthermore, some studies (Ringø et al., 1996; Makridis et al., 2000b) have shown that the gut microflora of earlydeveloping turbot larvae fed rotifers reflect the water microflora by adding Vibriopelagius (Ringø et al., 1996) or unidentified Gram-negative bacteria (Makridis et al.,2000b) to the water. Upon feeding, a dramatic shift occurs in the intestinalmicroflora which becomes dominated by fermentative Gram-negative rods such asVibrio and Aeromonas (Munro et al., 1993, 1995; Bergh et al., 1994; Bergh, 1995;Blanch et al., 1997). In agreement with these results, Gatesoupe (1990) reported thatthe microflora on both rotifers and turbot larvae feeding on the rotifers were domi-nated by Vibrio spp. The occurrence of Vibrio species as the dominant gut micro-organisms has also been reported in white shrimp (Litopenaeus vannamei) larvae ina recent study by Vandenberghe et al. (1999). The exact role of different bacterialgroups in larval disease or disease prevention is not known, however, it has beensuggested that some members of the natural flora are part of the non-specific dis-ease defence of the larvae (Bergh et al., 1994; Makridis et al., 2000b). Accordingly,some studies have shown that bacteria isolated from larvae may be inhibitory – inlaboratory based in vitro assays – against potential fish pathogenic organisms(Bergh, 1995). However, in vitro studies may not reflect the actual role of eachorganism in its ecological niche as will become evident from studies that have compared in vitro and in vivo antagonism. Even though few studies have identifiedspecific microbial larval pathogens, there seems no doubt that they are the cause ofthe significant mortalities often experienced in larval rearing. Thus, ultraviolet (UV)treatment of rotifers used for feeding turbot larvae, reduced bacterial numbers by more than 90% and caused a significant increase in survival of the larvae as



compared to larvae fed non-UV-treated rotifers (Munro et al., 1999), and Munro et al. (1995) obtained survival of up to 100% of larvae in bacteria-free larval rearing.Also, surface disinfection of eggs results in a much higher survival of the hatched larvae than when non-disinfected eggs are used (Vadstein et al., 1993).

Readers with special interest in bacterial species colonizing the gastrointestinaltract of larvae and fry are referred to the recent reviews of Hansen and Olafsen(1999) and Ringø and Birkbeck (1999).

3.2. Skin flora

Despite large variations in the environment of fish (cold marine fjords, tropicalfreshwater ponds) and the variation this imposes on the microbiology of cold-blooded animals, several groups of microorganisms are typical of fish skin. In general, the microflora of fish skin has been found to reflect the microflora in thesurrounding water (Horsley, 1973; Cahill, 1990). Thus, fish caught in temperatewaters harbour microflora dominated by Gram-negative bacteria belonging toPseudomonas, Acinetobacter, Vibrionaceae (including Photobacterium spp.),Flavobacterium and Moraxella, with Gram-positive bacteria represented by speciessuch as Kurthia, Streptococcus, Micrococcus and the coryneforms. In warmerwaters, the percentage of Gram-positive bacteria tends to be higher, and Bacillusand Micrococcus are more often isolated. Also, Enterobacteriaceae, which areisolated from temperate water fish, typically occur in higher numbers in tropical fish species (Gram and Huss, 2000).

3.3. Gill flora

In his early study, Horsley (1973) demonstrated that the major microbial groups of thegill microflora of Atlantic salmon were similar to those present in the water, whichsupports the hypothesis that the external fish flora is a reflection of the environment.In contrast, Mudarris and Austin (1988) demonstrated in a study with turbot that bac-teria isolated from the surface of gills were quite distinct from the species isolatedfrom the surrounding water. Also, Austin (1982, 1983) demonstrated that the surfacemicroflora isolated from turbot did not closely reflect the type of bacteria found ineither the seawater that supplied the tanks, or the water in which the fish lived. Byusing scanning electron microscopy of gill preparations, Mudarris and Austin (1988)found that only protected areas on the gill, such as clefts between adjacent secondarylamellae were colonized by bacteria-like objects. The gill flora is somewhat similar tothe skin microflora and is typically dominated by non-fermentative Gram-negativerods (Gennari and Tomaselli, 1988; Mudarris and Austin, 1988; Cahill, 1990) andVibrionaceae (Mudarris and Austin, 1988; Cahill, 1990). Thus, Spanggaard et al.(2001) found that 43% of the culturable flora were non-fermentative Gram-negativerods whilst 35% belonged to Vibrionaceae and Enterobacteriaceae (table 3).

Tabl

e 3.

Com

posi

tion

of th

e cu

ltura

ble

mic

robi

al f

lora

(10

18 s

trai

ns)

of r

ainb

ow tr

out (

isol

ated

fro

m 4

9 fi

sh)

and

the

num

ber

of s

trai

ns c

apab

le o

f in

hibi

ting

Vibr

io a

ngui

llar

umin

an

agar

-wel

l dif

fusi

on a

ssay

(m

odif

ied

from

Spa

ngga

ard

et a

l., 2

001;

Hub

er e

t al.,

200

4)

Skin

flo

raG

ill f

lora

Gut

flo

ra

No.

of

No.

of

No.

of

No.

of

No.

of

No.

of

Gro

up/s

peci

esis

olat

es%

anta

goni

sts

isol

ates

%an

tago

nist

sis

olat

es%

anta

goni

sts

Pse

udom

onas

2719

199

2424

235

3A

cine

toba

cter

/Mor

axel

la73

501

133

324

327

0V

ibri

onac

eae

43

060

154

100

220

Ent

erob

acte

riac

eae

00

046

111

167

363

Oth

er G

ram

-neg

ativ

es1

2618

050

120

7817

0G

ram

-pos

itive

s215

101

236

038

83

Yea

sts

00

01

<1

023

50

Tota

l14

510

03

412

100

3346

110

09

1 In

clud

e Sh

ewan

ella

, Fla

voba

cter

ium

, Ple

siom

onas

, Xan

thom

onas

.2

Incl

ude

Stre

ptoc

occu

s, S

taph

yloc

occu

s, C

arno

bact

eriu

m.

3.4. Gut flora

The fish gut microflora is markedly different from the environment and is typicallydominated by a few genera. However, there is a large variation in numbers of micro-organisms associated with epithelial mucosa and in the lumen of individual fish(Ringø and Olsen, 1999; Huber et al., 2004). Faecal bacterial flora (Sugita et al.,1988, 1990) and the gut microflora of one fish species may vary between differentgeographical regions. Furthermore, it is well known that dietary components affectthe gut microflora (Sugita et al., 1988; Strøm and Olafsen, 1990; Ringø, 1993;Ringø and Strøm, 1994; Ringø and Olsen, 1999) and there may also be large day-to-day fluctuations in both numbers and composition (Sugita et al., 1987, 1990). Thechange in microflora from the outer surfaces is probably due to both a selective pres-sure from the gut acid and bile, and just as much to the nutrient-rich, low-oxygenenvironment selecting for fermentative Gram-negative rods. Thus, Vibrionaceae andEnterobacteriaceae are typical of the fish gut where they may occur in high num-bers (106–108 CFU/g) (Austin and Al-Zahrani, 1988; Cahill, 1990; Spanggaard et al., 2001; Huber et al., 2004) although the overall number of bacteria in fish gutmay vary from 103 to 108 CFU/g (Huber et al., 2004). Most of these bacteria seemto reside in the lumen with only a minor proportion of the bacterial cells presentalong the gut wall (Austin and Al-Zahrani, 1988; Huber et al., 2004). In Arctic charr,scanning and transmission electron microscopy has shown that bacterial cells areclosely associated with epithelial cells of pyloric caeca, midgut and hindgut(Lødemel et al., 2001; Ringø et al., 2001).

In most studies on intestinal microflora of fish, only aerobic or facultative anaerobic bacteria have been isolated (Hansen and Olafsen, 1999; Ringø andBirkbeck, 1999). However, some investigations revealed the occurrence of signifi-cant numbers of strict anaerobic bacteria in fish gut (Trust and Sparrow, 1974; Trust,1975; Trust et al., 1979; Sakata et al., 1981; Huber et al., 2004). Huber et al. (2004)reported that an unculturable bacterium with a gene-sequence clustering amongstanaerobic bacteria accounted for 99% of the gut microflora of rainbow trout, butthen only in a few fish. Gram-positive bacteria such as LAB are also commonly isolated from fish gut (Ringø and Gatesoupe, 1998), and Ringø et al. (2000) reportedthat for Atlantic salmon sampled in Northern Norway approximately 30% of a gut flora of 103 CFU/g were LAB, in particular carnobacteria (Carnobacterium piscicola).

4. SELECTION STRATEGIES FOR PROBIOTIC CULTURES

Studies on the use of probiotics in warm-blooded animals and, as outlined above, infish have often been non-conclusive or even contradictory (Conway, 1996). A variety of traits have been considered important, including the production of



anti-pathogen substances and the ability to survive and colonize the host. Amongstthe parameters listed for selecting a probiotic strain are (Conway, 1996):

specified targethost originsurvival in vivo/in situcolonization potentialnon-pathogenic/safebiological activity against targetdemonstrable efficacystability/robustness.At present little is known about the in vivo mechanisms that are operational when

probiotic culture is used with success, and a scientific rationale for the selection of the best species or strains for use as probiotics is not possible without more information on the mechanisms by which probiotics exert their beneficial effect in vivo (Berg, 1998). For instance, Moriarty (1998) found that a preparation ofBacillus spp. was effective in reducing shrimp disease but also that a constant supply was required, indicating that colonization did not take place and was not aprerequisite. Ringø and co-authors (Ringø, personal communication, 2001) recentlydemonstrated by transmission electron microscopy (TEM) that a pathogenic Vibrio(V. anguillarum) attached to the gut epithelial cells of the midgut and hindgut of common wolffish (Anarhichas lupus L.) fry, whereas a potential probiotic LAB strain, a Carnobacterium divergens originally isolated from wolffish fry, colonized enterocytes of the pyloric caeca and foregut. Such studies by using TEM and scanning electron microscopy (SEM) emphasize the need to understandthe pathogen proliferation and infection sites, so that the probiotic culture can be targeted to these sites.

4.1. In vitro antagonism

Most – if not all – studies initiate the search for probiotic microorganisms by evalu-ating the in vitro inhibitory activity of a collection of microorganisms against thetarget pathogen, although it has not been shown that inhibitory substances producedin vitro are actually effective in vivo (Atlas, 1999).

Agar-based diffusion assays exist in several forms, where the target organism isincorporated into an agar layer and exposed to the test organism as either a live cul-ture (Gram, 1993; Bly et al., 1997) or the inactivated test organism (Lemos et al.,1985; Dopazo et al., 1988; Westerdahl et al., 1991; Bergh, 1995; Sugita et al.,1996a,b, 1997a; Jöborn et al., 1997). Also, the two organisms may be cross-streakedagainst each other (Smith and Davey, 1993).

The test organism may be cultured in liquid media and the filter-sterilized culturesupernatant (or extracts thereof) tested for its inhibitory activity against the pathogen.


This is typically done following the growth of the pathogen by absorbance meas-urement (Smith and Davey, 1993; Jöborn et al., 1997; Gram et al., 1999; Ringøet al., 2000). Some studies have also used co-culture in broth systems to evaluate theinhibitory properties of an organism (Gram et al., 1999, 2001; Spanggaard et al.,2001), however, such studies require that a specific, quantitative method exists forenumeration of the pathogenic target organism.

Using such approaches, a range of bacteria with inhibitory properties against fishpathogenic bacteria have been identified. Typically, only a few per cent of the cul-turable flora from fish can inhibit fish pathogenic bacteria in vitro (Sugita et al.,1996a,b; Riquelme et al., 1997; Spanggaard et al., 2001) but some studies havereported that almost 1/3 of the microflora were inhibitory to target organisms(Westerdahl et al., 1991; Burgess et al., 1999). A high proportion (20%) of bacteriafrom intertidal seaweeds was also inhibitory to other bacteria (Lemos et al., 1985).

Media composition strongly influences the results of such screenings, as theinhibitory activity of the test strains is medium dependent. One classical example isthe antimicrobial properties of fluorescent pseudomonads in which iron limitationcauses production of iron chelating siderophores which deprive the competingpathogen of iron. Under iron-surplus conditions, no inhibition is detected, whereasunder iron-limited conditions, a pronounced inhibition is seen (Gram, 1993; Smithand Davey, 1993). Bearing the influence of media and culture conditions in mind, it is perhaps not surprising that no clear pattern emerges in terms of which bacterialspecies harbour anti-pathogen strains. Some studies have identified Vibrio (andAeromonas spp.) as the most inhibitory (measured in in vitro assays) as comparedto other culturable bacteria isolated (Westerdahl et al., 1991; Bergh et al., 1995),whereas others have identified Pseudomonas and Alteromonas as potent inhibitors(Lemos et al., 1985; Tanasomwang et al., 1998; Spanggaard et al., 2001). Also, a large proportion of LAB is inhibitory to fish pathogens (Ringø and Gatesoupe,1998; Ringø et al., 2000). In the studies by Sugita et al. (1996a,b, 1997b) almost2000 strains were isolated from the intestinal tract of various fish species. Nospecific bacterial group emerged as more inhibitory than others.

4.2. Adherence to mucus and survival of intestinal conditions

Probiotic bacteria must persist – either transiently or permanently – in the environ-ment in which they are to act. Thus, they must be able to adhere to the fish surfaces.Several authors pursuing fish probiotics have used the ability of an organism toadhere to intestine or skin mucus as a selection criterion (Olsson et al., 1992; Jöbornet al., 1997; Nikoskelainen et al., 2001a). Whilst the organisms do adhere, it shouldbe noted that fish mucus does not, in general, cause an increase in numbers ofadhered bacteria, as adhesion to neutral surfaces, e.g. polystyrene, may be as highor better (Nikoskelainen et al., 2001a).

Also, if a probiotic microorganism is to act in the intestinal system, it must beable to resist compounds that may act as antimicrobials. Thus, Nikoskelainen et al.


(2001a) found that a range of LAB that were tested for potential as probioticsresisted 10% fish bile for 11/2 hours.

4.3. Colonization and persistence

“Colonization potential” has been suggested as one of several selection criteria forprobionts (Conway, 1996). In this chapter, colonization is defined not only as theability of an organism to proliferate for some time in a niche to which it is added (in high numbers), but also as the ability to establish itself more permanently andremain after exogenous supply has stopped. Innumerable studies of the mammaliantract have found that unless the existing microflora is eliminated, e.g. by antibiotictreatment, it is virtually impossible for an exogenous microorganism to becomeestablished. This ability of the existing microbiota in a niche to resist invaders hasbeen called the colonization resistance (van der Waaj et al., 1971). As describedbelow, this phenomenon appears important in aquatic organisms as well.

Robertson et al. (2000) found 106–107 CFU of Carnobacterium spp. per gram inthe gut, when trout were fed a diet containing approximately 107 per gram of theorganism, indicating that no significant growth took place. However, when the fishreverted to the control diet without Carnobacterium, the level dropped to less than 1 per gram in 6 days indicating that no colonization had taken place at all and thatthe supplement was unable to compete with the natural flora. Similarly, Strøm andRingø (1993) added 105 CFU/ml of an Atlantic cod-derived Lactobacillus plantarumto cod fry. This resulted in an immediate dominance of this organism, however, after9 days the flora of the Lactobacillus treated fry was similar to that of non-treated fry.

Colonization by added microflora may be possible if very young, and pre-sumably not completely colonized animals are exposed to an exogenous sourceof microorganisms. Electron microscopy revealed extensive colonization of the cae-cal mucosal epithelium of 3-day-old chicks that were fed a mixture of 29 differentmicroorganisms on the day of hatch (Droleskey et al., 1995). This so-called com-petitive exclusion treatment increased resistance to colonization with Salmonellaand reduced Salmonella contamination of floor pen litter (Corrier et al., 1998).

4.4. Aggregation with pathogens

Some potential probiotic bacteria may have the ability to aggregate the pathogenicorganisms, and it has been suggested that this may prevent the adhesion of thepathogen to the mucosa. This has been studied in bacteria relevant to mammaliansystems where Spencer and Chesson (1994) isolated five strains of lactobacilli ableto aggregate Escherichia coli 0129-K88+ within a sample of 43 lactobacilli strains.Later, Kmet and Lucchini (1999) isolated 20 strains of lactobacilli and demonstratedaggregation activity between six homofermentative autoaggregative lactobacilli andthree strains of pathogenic E. coli with F4, F5 and F6 fimbriae. As no informationis available about aggregation of probiotics with pathogens in fish, this topic might


be of some interest in future studies on fish. One could argue that if such aggrega-tion occurs, it may also prevent the probiont from reaching attachment sites andthereby reduce the potential beneficial effect.

4.5. Site specificity in vivo

A probiotic microorganism may exert an “overall” effect on the environment, e.g. producing a diffusible antibacterial compound that affects the pathogenic organ-ism. However, it is possible that the effect of a probiotic microorganism has to beexerted against the pathogen in very specific spatial areas. Therefore, the probioticmicroorganism must be targeted to the niches where the pathogenic organismresides, proliferates or invades. This will require better knowledge of fish pathogensthan currently available. As mentioned, Ringø and co-workers (unpublished data)observed using TEM and SEM that a pathogenic Vibrio adhered to another part ofthe intestinal tract than a potential probiotic Carnobacterium. Such observationscould point to selection of probiotic strains from the same species as the pathogen,assuming that they will compete more efficiently for the same nutrients and spaceas the pathogen.

5. USE OF MICROBIAL CULTURES FOR PREVENTION OF DISEASEIN AQUACULTURE

The number of studies testing probiotics in aquaculture is rapidly increasing. Whilstinitial studies were carried out with grown fish, the focus has shifted to fish larvae,crustaceans and molluscs where vaccines are unlikely to be available within the nearfuture. This section gives an overview of studies that have involved in vivo evaluationof probiotics as disease preventive measure, and major results are summarized intables 4, 5 and 6. In vivo trials have typically evaluated a probiotic effect by compar-ing the accumulated mortality of control (infected) tanks or ponds to that of treated(infected) tanks or ponds. Unfortunately, the statistical treatment of suchsurvival/mortality data is often lacking or is based on simple comparisons of meanvalues, e.g. using analysis of variance (Gram et al., 1999) or student’s t-test (Queirozand Boyd, 1998). It would be statistically more appropriate to use other means ofanalysis commonly used to analyse survival data. The standard analysis for analysingsurvival data, when the exact survival times are observed, is the proportional hazardmodel (Cox, 1972; Cox and Oakes, 1984). When the survival times are only observedat certain points in time, this model reduces to a so-called generalized linear modelwith a clog-log link (Fahrmeir and Tutz, 1994; Spanggaard et al., 2001).

As mentioned, in vivo infection trials may be difficult to standardize and, evenwith tight control of fish, water temperature, oxygen concentration, etc., large vari-ations, e.g. in accumulated mortality may be seen between tanks. Thus, in a set oftrials testing the probiotic bacterium P. fluorescens strain AH2 (Gram et al., 1999),

Tabl

e 4.

Eff

ect o

f ad

ditio

n of

pro

biot

ic m

icro

orga

nism

s on

fis

h an

d cr

usta

cean

larv

al s

urvi

val

Pres

umed

pro

bion

tPa

thog

enH

ost o

rgan

ism

Eff

ect o

n su

rviv

alR

efer

ence

Car

noba

cter

ium

div

erge

nsVi

brio

ang

uill

arum

cod

fry

no e

ffec

tG

ildbe

rg a

ndM

ikke

lsen

, 199

8

Car

noba

cter

ium

div

erge

nsVi

brio

ang

uill

arum

cod

fry

incr

ease

acc

umul

ated

sur

viva

l fro

m 4

0 to

60%

G

ildbe

rg e

t al.,

199

7af

ter

3 w

eeks

Car

noba

cter

ium

div

erge

nsA

erom

onas

A

tlant

ic s

alm

on f

ryde

crea

se a

ccum

ulat

ed s

urvi

val f

rom

50

to

Gild

berg

et a

l., 1

995

salm

onic

ida

30%

aft

er 4

wee

ksL

actic

aci

d ba

cter

ium

Vib

rio

Ptu

rbot

larv

aein

crea

se a

ccum

ulat

ed s

urvi

val f

rom

19

to 2

8%

Gat

esou

pe, 1

994

(exp

t 1)

afte

r 48

hou

rs o

r fr

om 1

9 to

23%

(e

xpt 2

) an

d fr

om 8

to 5

0%

(exp

t 3)

afte

r 72

hou

rsVi

brio

pel

agiu

sno

t kno

wn

turb

ot la

rvae

incr

ease

acc

umul

ated

sur

viva

l fro

m 6

to 9

% o

nR

ingø

and

Vad

stei

n, 1

998

day

12 a

nd f

rom

0 to

3%

on

day

16 a

fter

ha

tchi

ngVi

brio

med

iter

rane

iQ

40no

t kno

wn

turb

ot la

rvae

incr

ease

acc

umul

ated

sur

viva

l (5

days

pos

t H

uys

et a

l., 2

001

hatc

hing

) in

fiv

e se

para

te e

xper

imen

ts; e

.g. 1

4

to 5

5% in

tria

l 1 o

r 75

to 8

1% in

tria

l 4A

erom

onas

med

iaVi

brio

tub

iash

iioy

ster

larv

aein

crea

se s

urvi

val a

fter

6 d

ays

from

4 to

100

%G

ibso

n et

al.,

199

8P

seud

omon

asan

d un

know

nVi

brio

ang

uill

arum

sc

allo

p la

rvae

incr

ease

sur

viva

l fro

m 5

to 6

0% a

fter

14

days

Riq

uelm

e et

al.,

199

7st

rain

like

Pse

udom

onas

and

Vibr

iofi

eld

tria

l? n

ot k

now

n sc

allo

p la

rvae

sam

e su

rviv

al a

fter

48

h as

ant

ibio

tic tr

eate

d R

ique

lme

et a

l., 2

001

tank

sT

hala

ssob

acte

r ut

ilis

fiel

d tr

ial n

ot k

now

n cr

ab la

rvae

incr

ease

sur

viva

l fro

m 1

6 to

26%

(se

e ta

ble

7)N

ogam

i and

Mae

da, 1

992;

(Vib

rio

spp.

)N

ogam

i et a

l., 1

997

Tabl

e 5.

Eff

ect o

f ad

ditio

n of

pro

biot

ic m

icro

orga

nism

s on

fis

h su

rviv

al

Pres

umed

pro

bion

tPa

thog

enH

ost o

rgan

ism

Eff

ect o

n su

rviv

alR

efer

ence

Bac

teri

opha

ge1

Pse

udom

onas

ay

uin

crea

se s

urvi

val f

rom

35

to 7

5%Pa

rk e

t al.,

200

0pl

ecog

loss

icid

aTe

tras

elm

is s

ueci

case

vera

l Gra

m-n

egat

ives

salm

onin

crea

se s

urvi

val f

rom

0–1

5 to

20–

100%

Aus

tin e

t al.,

199

2

Bac

illus

spp

.fi

eld

tria

lch

anne

l cat

fish

incr

ease

sur

viva

l fro

m 5

6 to

80%

Que

iroz

and

Boy

d, 1

998

not k

now

nC

arno

bact

eriu

m s

pp.

Vibr

io a

ngui

llar

umsa

lmon

no e

ffec

t on

surv

ival

Rob

erts

on e

t al.,

200

0Vi

brio

ord

alii

incr

ease

sur

viva

l fro

m 2

3 to

74%

Yers

inia

ruc

keri

incr

ease

sur

viva

l fro

m 4

2 to

71%

Aer

omon

as s

alm

onic

ida

incr

ease

sur

viva

l fro

m 0

to 2

0%A

erom

onas

sal

mon

icid

atr

out

incr

ease

sur

viva

l fro

m 3

2 to

74%

Vibr

io a

lgin

olyt

icus

Aer

omon

as s

alm

onic

ida

salm

onin

crea

se s

urvi

val f

rom

0 to

82%

Aus

tin e

t al.,

199

5Vi

brio

ang

uill

arum

incr

ease

sur

viva

l fro

m 1

0 to

26%

Vibr

io o

rdal

iiin

crea

se s

urvi

val f

rom

0 to

26%

Yers

inia

ruc

keri

ino

eff

ect o

n su

rviv

alP

seud

omon

as f

luor

esce

nsA

erom

onas

sal

mon

icid

asa

lmon

incr

ease

hea

lthy

fish

fro

m 2

5–70

to 9

0–10

0%Sm

ith a

nd D

avey

, 199

3P

seud

omon

as f

luor

esce

ns

Vibr

io a

ngui

llar

umtr

out

incr

ease

sur

viva

l fro

m 5

0 to

70%

Gra

m e

t al.,

199

9P

seud

omon

assp

p.Vi

brio

ang

uill

arum

trou

tin

crea

se s

urvi

val (

som

e st

rain

s)Sp

angg

aard

et a

l., 2

001

no e

ffec

t (so

me

stra

ins)

Pse

udom

onas

flu

ores

cens

A

erom

onas

sal

mon

icid

asa

lmon

no e

ffec

t on

surv

ival

Gra

m e

t al.,

200

1

1 It i

s de

bata

ble

whe

ther

or

not a

bac

teri

opha

ge q

ualif

ies

as a

pro

bion

t, “a

live

mic

robi

al...

”, b

ecau

se o

f th

e in

ert n

atur

e of

vir

uses

.


rainbow trout were immersion infected with V. anguillarum and the fish divided into16 tanks each with 30 fish (Slierendrecht and Gram, 2001, unpublished data). Eighttanks were treated with AH2 by adding the probiont to the water on a daily basis.The accumulated mortality in control tanks varied from 64 to 89% whereas theaccumulated mortality in treated tanks varied from 28 to 63% with an outlier of 84%(fig. 3). The importance of knowing the inherent variation in a particular infectiontrial system and of using an appropriate number of replicates cannot be under-estimated. With a large variation, a too low number of replicates may mask

Table 6. Effect of addition of probiotic microorganisms on crustacean (shrimp) survival

Presumed Host probiont Pathogen organism Effect on survival Reference

Saccharomyces not known shrimp increase survival Scholz et al., 1999cerevisiae from 60 to 80%

Phaffia not known shrimp increase survival Scholz et al., 1999rhodozyma from 60 to 80%

Bacillus spp. field trial shrimp increase survival Moriarty, 1998luminescent

vibriosBacillus S11 Vibrio harveyi tiger shrimp increase survival Rengpipat et al., 2000

from 36 to 54%Bacillus S11 field trial tiger shrimp increase survival Rengpipat et al., 1998

Vibrio harveyi from 16 to 33% increase survival

from 30 to 100%

Fig. 3. Accumulated mortality of rainbow trout following immersion infection with Vibrio anguillarum.Approximately 480 fish were infected; half of which were also submerged in Pseudomonas fluorescens strainAH2 and subsequently treated with AH2 on a daily basis. Fish were divided into 16 tanks: eight served ascontrols and eight were treated with AH2 (Slierendrecht and Gram, 2001; unpublished data).


a significant effect of a treatment. Conversely, single determinations may, by chance,be different and appear to show an effect, yet may be only reflecting the normalwindow of variation in the experiment.

By contrast, if the variation is smaller, a lower number of replicates is required.This is depicted in fig. 4 where the accumulated mortality of Atlantic salmon co-habitant infected with furunculosis, varies from 66 to 76% in the control andfrom 66 to 78% in the tanks treated with AH2 (Gram et al., 2001).

5.1. Disease prevention in Artemia

Despite many attempts, it has not been possible to develop commercial, non-live feedfor fish larvae. Thus in a very sensitive stage of the life cycle, fish larvae depend onlive feed, typically small shrimp, Artemia, and rotifers. As these cultures also expe-rience microbial disease, probiotic disease prevention has been studied for the livefeed. Rico-Mora et al. (1998) showed that the number of the pathogenic Vibrio alginolyticus in a diatom culture could be reduced by the addition of an Aeromonasspecies. Later, Verschuere et al. (2000b) tested nine bacterial strains isolated fromwell-performing Artemia cultures for their in vivo effect on Artemia survival follow-ing Vibrio proteolyticus infection. All Artemia in infected cultures to which no otherbacterial strains were added, died within 2 days. Five of the nine bacterial strains hada probiotic effect as their addition resulted in survival (90%) comparable to the non-infected Artemia. These data were derived 48 h following infection. Two strains werealso tested over a 4 day period, and one strain (belonging to the Vibrionaceae) alsoincreased survival over this longer period. This strain had the highest colonizationpotential (measured as CFU/Artemia) and caused a slower growth of the pathogen

Fig. 4. Accumulated mortality of Atlantic salmon following co-habitant infection with Aeromonassalmonicida. Six tanks were infected with furunculosis and three tanks were treated with Pseudomonasfluorescens AH2 three times per day (see also Gram et al., 2001).


V. proteolyticus, resulting in a lower cell density of V. proteolyticus in the culturewater. Interestingly, this Vibrio strain was also tested as a food source for rotiferswhere it caused a significant reduction in growth rate of the rotifer (Rombaut et al.,1999). Addition of filter-sterilized supernatant from the probiotic Vibrio strain did notincrease survival of Artemia and none of the nine strains showed in vitro activityagainst the pathogen Vibrio alginolyticus when tested by Dopazo’s double layer tech-nique. The addition of potential probiotic microbial cultures to Artemia or rotifer cultures has also been studied as a way of introducing the probionts to the larval rearing systems (Gatesoupe, 1994). Introducing a LAB culture to rotifers (addingoutgrown culture to the rotifer medium) increased the counts of LAB in the larvae a100-fold from 102 to 104 per larvae (Gatesoupe, 1994).

5.2. Fish and crustacean larvae

For a range of fish species, in particular marine species, the main obstacle for cultiva-tion is a very high mortality in the larval stage. In these marine larvae, the pathogenicorganisms have not been unequivocally identified but trials have shown that culturingturbot larvae under bacteria free conditions or at very low bacterial densities enhancessurvival dramatically (Munro et al., 1993, 1995), thus indicating the importance ofbacteria for disease development. Also, exposure to V. anguillarum increases mortal-ity of turbot larvae (Munro et al., 1995). The live food (rotifers, Artemia) can be asource of pathogenic bacteria (Makridis et al., 2000a). These bacteria may be removedby disinfection regimes, however, re-colonization with opportunistic, potentially patho-genic bacteria may occur rapidly (Skjermo and Vadstein, 1999). Therefore, it has beensuggested that the live food, following disinfection, may be exposed to bacteria bene-ficial to larvae which are then allowed to colonize the rotifers or Artemia and subsequently transferred to the fish larvae (Makridis et al., 2000a).

Gatesoupe (1994) added LAB originally isolated from the non-dominant flora ofrotifers (Brachionus plicatilis) to rotifers which were used as feed for turbot larvaethat subsequently were infected with a Vibrio in three separate experiments. In oneof the experiments, the author reported a marginal effect on survival, whereas in thetwo other trials addition of LAB increased larval survival from 13 to 28% and from8 to 50%, respectively. Unfortunately, the study does not give mortality kinetics butonly survival percentages at 48 and 72 h after infection. LAB addition did not alterthe number of vibrios in the tank water. In a recent study, Huys et al. (2001)searched for beneficial bacterial strains for turbot larviculture and found increasedsurvival of larvae by addition of a Vibrio mediterranei Q40 originally isolated fromsea bream larvae and an unknown organism isolated from turbot larvae at a concen-tration of 105 bacteria per ml water.

Several studies from Tromsø, North Norway, have focused on the possible use ofLAB, in particular carnobacteria, as oral probiotics for marine fish larvae and fry.The accumulated mortality of Atlantic salmon fry fed a diet with C. divergens and


challenged with Aeromonas salmonicida reached 65% in 3 weeks whereas the con-trol group was at approximately 40% after the same time period (Gildberg et al.,1995). The LAB used in this study showed no in vitro antagonism, neither as sterilefiltered supernatant nor in co-inoculation trials where equivalent numbers were usedas inoculum. In two studies, the feed for Atlantic cod larvae was supplementedeither with freeze dried C. divergens-like (Gildberg et al., 1997) or with a live C. divergens-like culture (Gildberg and Mikkelsen, 1998). Upon challenge with V. anguillarum, the mortality was lower in the group fed LAB-containing feed thanin the group fed the same diet with no supplementation of LAB (Gildberg et al.,1997). However, a control group fed a third (non-LAB containing) diet, had the lowest accumulated mortality of all. Following infection with V. anguillarum, all fish reached 80% accumulated mortality after 3 weeks. In in vitro co-culture, V. anguillarum was not inhibited by C. divergens. However, the cultures were inocu-lated at similar cell densities (106 CFU/ml), so no inhibition is to be expected.

Bacillus strains have been used successfully as probiotics in shrimp culturing(Moriarty, 1998) and some data exist on their potential use in larval rearing. Addition ofa Bacillus spp. combined with a slight decrease in salinity, resulted in excellent survival(80 –100%) of larvae of common snook in commercial systems (Kennedy et al., 1998).

In two studies, Nogami and Maeda (1992) and Nogami et al. (1997) added 105–106 CFU/ml of bacterial culture isolated from shrimp pond to seawater used for crab(Portunus trituberculatus) culture. The strain was a Gram-negative, non-fermentative,motile rod identified as Thalassobacter utilis (Maeda and Liao, 1992; Nogami et al.,1997). The culture, which was added once every 7 days, was selected based on its ability to improve survival in in vivo infection trials. The organism also inhibitedgrowth of V. anguillarum, in vitro. By adding the culture, a decline in concentration ofVibrio spp. in the seawater occurred and survival was significantly improved (table 7).It should be emphasized that the addition of five other microbial cultures (e.g. a Bacillus subtilis) accelerated mortality of the larvae (Nogami and Maeda, 1992).

Gatesoupe (1997) tested a siderophore-producing vibrio as probiotic by feedinginfected turbot larvae with rotifers enriched by the vibrio. This improved survivalwhen measured 48 h after infection. However, after 10 days, no difference was seenin survival. In contrast, Ringø and Vadstein (1998) found that addition of Vibriopelagius to early developing turbot (Scophthalmus maximus) larvae had no short-term effect but caused a slight increase in survival after 12–16 days. V. pelagius wastaken up by the larvae if added to the tank water at the day of hatching (Ringø et al.,1996). It is, however, not known whether the bacterium had colonized the gut andwould persist if the larvae were removed from the Vibrio-containing tank water.

5.3. Scallop and oyster larvae

The bivalve molluscs (e.g. oysters) are of increasing interest in the aquaculturesector. These invertebrate organisms do not possess acquired immunity (Roch, 1999)

and disease prevention by vaccination is therefore difficult. The animals do possessan innate immunity and are capable of specific humoral and cellular defence reac-tions (Roch, 1999).

Riquelme et al. (1997) investigated 506 bacterial strains for their potential probiotic effect in Chilean scallop (Argopecten purpuratus Lamarck 1819) larvalculture. Initially, both a Pseudomonas isolate showing in vitro activity against a V. anguillarum-related strain and an unidentified isolate with no in vitro inhibitoryactivity were found to improve survival from 5% in the non-probiotic treated to 60% in the probiotic treated over a 14 day period. Screening the 506 bacterial isolates, the authors found that only 2.2% of them were able to inhibit growth of aV. anguillarum related bacterium associated with mortality of scallop larvae.However, several strains showing in vitro activity increased mortality of scallop larvae. Thus, this study demonstrates the importance of in vivo testing, as strainswith in vitro effect may be dangerous to the animals – and strains with no in vitroeffect may have probiotic effects in vivo.

In a recent study, Riquelme et al. (2001) reported that growth and survival in fieldtrials with scallop larvae treated with pathogen-antagonizing bacteria at 103 CFU/mlwere the same as when the larvae were treated with antibiotics. The antagonizingbacteria were added to the water at the initiation of the experiment and again after48 h. Controls with no treatment were not included as the commercial producer


Table 7. Survival and production of a swimming crab (Protunus trituberculatus) in fiveconsecutive years when comparing untreated systems to systems treated with the bacteriumThalassobacter utilis. The bacterium was added at a density of 105–10 6 per ml once every6 to 8 days (modified from Nogami et al., 1997)

Avg. survival Final No. of Addition of No. of Larval no. rate (%) production production

Year biocontrol experiments at start to 1st crab stage (ind/m3) failures

1989 − 10 46 960 000 22.0 5158 0+ 4 20 300 000 30.4 7703 0

1990 − 9 42 930 000 6.8 1617 4+ 7 30 570 000 26.7 5838 0

1991 − 7 34 150 000 10.4 2543 4+ 7 34 790 000 27.9 6938 0

1992 − 8 35 710 000 17.8 3963 3+ 9 37 410 000 28.8 5994 1

1993 − 8 33 200 000 21.6 4474 1+ 6 26 610 000 27.7 6150 0

Total − 42 192 950 000 15.7 3605 12+ 33 149 680 000 28.2 6397 1

experienced rapid mortality when no treatment was used. Earlier, Riquelme et al.(2000) studied the uptake of pathogen-inhibiting bacterial cultures in Chilean scalloplarvae, and found that an Arthrobacter was ingested in significant numbers. This can bea way of continuously adding the probiotic culture to the scallop larvae. TheArthrobacter strain was not tested in in vivo infection trials.

Bacterial probiotics have also been tested in the culturing of Pacific oyster(Crassostrea gigas) larvae. Gibson et al. (1998) added an Aeromonas media strain, iso-lated from Koi carp (Cyprinus carpio) from the Hawksbury River (Gibson, personalcommunication, 2001), to waters of oyster larvae which had been challenged withV. tubiashii. The challenge caused an increase in numbers of the pathogen and a com-plete kill of the oyster population in 5 days but addition of the A. media strain togetherwith the V. tubiashii, reduced numbers of the pathogen and resulted in complete sur-vival of the population. These results are in accordance with earlier results demonstrat-ing that additions of both algae (McCausland et al., 1999) and bacteria (Douillet andLangdon, 1994) have been found to improve growth of Pacific oyster larvae.

5.4. Grown fish

Prevention of bacterial disease in grown fish is somewhat easier than in larvae andjuveniles. For a range of bacterial pathogen–host combinations, good vaccines havebeen developed, and their optimization and use will be facilitated as further under-standing of the pathogen virulence factors and of the host immune system emerges.Some of the studies described below should therefore rather be regarded as trials ofthe probiotic concept than as suggestions for actual use of probiotics. Several Gram-positive and Gram-negative bacteria have been tested and experiments cover bothadditions to the rearing water and feed supplementation.

Austin et al. (1992) found that feeding Atlantic salmon with a diet containing 1% of the alga Tetraselmis suecica resulted in significant improvement of survival ratewhen the fish were subsequently challenged with a range of Gram-negative pathogenssuch as A. salmonicida, A. hydrophila, V. anguillarum, V. ordalii, Y. ruckerii and S. liquefaciens. The algae produced an antibacterial compound as tested by agar-diffusion assays and the protective effect displayed a dose–response relationship withimprovement in survivals seen with 0.25 to 1% in the feed.

Incorporation of a killed (by exposure to air and lyophilization) preparation ofClostridium butyricum into feed of rainbow trout reduced susceptibility to Vibrioinfections (Sakai et al., 1995). Whilst this does not concur with the definition in thischapter of probiotics being a “live microbial supplement” it contributes to the under-standing of the potential applications of probiotics.

In order to improve water quality of channel catfish ponds, a commercial bacte-rial inoculum (BioStart®) based on Bacillus spp. was added (at 103–104 CFU/ml) topond water three times per week for 5 months (Queiroz and Boyd, 1998). This didnot change water quality parameters but resulted in significant improvement of


survival from 56 to 80%. Although the fish were smaller in the treated ponds, over-all production was 30% higher here.

A Carnobacterium inhibens strain, originally isolated from Atlantic salmon(Jöborn et al., 1999) which has a strong in vitro antagonism against two fish pathogens(V. anguillarum and A. salmonicida) (Jöborn et al., 1997), was incorporated into fishfeed at levels of 1 × 106–5 × 107 CFU/g and fed to Atlantic salmon and trout (15–20 gsizes) for 7–28 days. Subsequent co-habitant challenge with several Gram-negativebacteria, showed higher percentage survival in groups fed 14 days or more with theprobiont (Robertson et al., 2000). The study – like most studies – unfortunately hasnot recorded the kinetics and replicates. Also, variances are not presented. It is not pos-sible, as is the case with most probiotic studies, to evaluate if differences are statisti-cally significant. The effect of a Lactobacillus rhamnosus strain on mortality offurunculosis co-habitant infected rainbow trout was studied (Nikoskelainen et al.,2001b). The lactobacilli (LAB) were incorporated in the fish feed and a level of 109 LAB per gram feed reduced accumulated mortality from 50 to 20%. However,incorporation of 1012 LAB resulted in an accumulated mortality of 45%. The infectiontrial was only done with one tank per treatment and thus, the variation of survival/mortality curves is not known. Single determinations may be far from sufficient,bearing in mind the variation in mortality that sometimes is seen (fig. 3).

Bathing of salmon for 10 min in suspensions containing Vibrio alginolyticusoriginally isolated from a shrimp hatchery in Ecuador, resulted in establishment ofthe culture as it was isolated 21 days after the immersion (Austin et al., 1995).Improved survival of the fish following infection with A. salmonicida was noted.Some effect on survival from Vibrio infections was also seen, whereas the probiotictreatment had no effect on infection with Yersinia ruckerii.

Smith and Davey (1993) reported that bathing of Atlantic salmon pre smolts in5 × 105 CFU/ml fluorescent pseudomonad suspension, reduced subsequent diseasefrom stress-inducible furunculosis. Thus, between 0 and 10% of pseudomonadtreated fish became infected whereas 30–72% of the non-treated fish were infected.In contrast to this, Gram et al. (2001) did not find improved survival in Atlanticsalmon co-habitants infected with furunculosis and treated with Pseudomonasfluorescens strain AH2. Levels of AH2 varied from 10 to 105 CFU/ml. In otherexperiments, this strain caused a reduction in accumulated mortality of rainbowtrout bath infected with V. anguillarum (Gram et al., 1999; Spanggaard et al., 2001).When rainbow trout were infected with V. anguillarum, the organism proliferated tohigh numbers on the skin surface before significant growth was detected on the gillsand in the intestinal tract. The beneficial action of pseudomonads added to the rear-ing water could be explained by their direct action on the outer skin surface of thefish where V. anguillarum proliferated (Spanggaard et al., 2000b). Studies have notbeen performed determining the proliferation and infection sites of A. salmonicidain co-habitant infected salmon, but if this occurs in areas (e.g. the gut) where thepseudomonads are not favoured, this could explain the different outcomes.



Whilst most studies have aimed the search for probiotic candidates at the bacterialpopulation, it has been demonstrated that pathogen-specific bacteriophages can inhibitthe pathogen in vitro and reduce subsequent fish mortality in vivo (Park et al., 2000).Phage treatment of ayu (Plecoglossus altivelis) infected with Pseudomonaspleccoglossicida resulted in reduction of accumulated mortality from 60–70% to20–25% in 10 g fish and from 73–80% to 0–13% in 2.4 g fish (Park et al., 2000).

5.5. Shrimps

Shrimp and other crustaceans are important aquaculture animals. The immunedefence system of shrimp is poorly developed and, in spite of a primitive immunesystem relying on, e.g. phagocytosis and lysis activity of the haemocytes (cited fromScholz et al., 1999), vaccines are not likely to be successful. Therefore, several stud-ies have been dedicated to alternative disease control measures in crustacean culture.

Scholz et al. (1999) reported that addition of yeasts (Saccharomyces cerevisiaand Phaffia rhodozyme) to shrimp feed improved survival from 60 to 80% over a 7 week period. Interestingly, phenoloxydase activity, which is an indicator of bacte-rial clearance ability, was lowest in the animals treated with the Phaffia. In contrast,phenoloxydase and other immunity indicators in tiger shrimp (Penaeus monodon)showed an increase due to inclusion of a Bacillus strain S11 in the feed. This addi-tion also caused an increase in survival as compared to a control feed when shrimpwere infected with Vibrio haryvei (Rengpipat et al., 1998, 2000). The supplement ofBacillus S11 into the feed also increased survival in unchallenged shrimp(Rengpipat et al., 1998) where average survival after 100 days was 33% in theBacillus treated tanks as opposed to 16% in the control systems (fig. 5).

Fig. 5. Survival of shrimp (Penaus monodon) fed a diet with (�) and without (■) a Bacillus S11 diet. Feed contained 1010 CFU per gram of S11 and the shrimp were fed three times per day (modified from Rengpipat et al., 1998).

Bacillus S11 was also used to suppress levels of luminescent vibrios in shrimpponds. Vibrio levels reached almost 107 CFU/ml in untreated ponds but remainedat 10 2–104 CFU/ml in ponds treated with Bacillus S11 (Rengpipat et al., 1998).Moriarty (1998) similarly found that a Bacillus-based probiotic addition lowered the level of luminescent vibrios. In untreated ponds, the level varied from 44 to 281 CFU/ml, whereas the level in treated ponds varied from 0 to 75 CFU/ml.

6. USE OF MICROBIAL CULTURES AS FEED

Microorganisms are an important part of the diet of some fish food organisms andfish (Moriarty, 1997; Thompson et al., 1999). Therefore, trials have been carried outto evaluate the growth rate of both fish larvae and rotifers following addition ofmicroorganisms or microbial products to rearing water. Results from such studiesvary and are typically based on “trial and error”, and with few attempts to under-stand the microbial ecology and mechanisms of the system.

6.1. Microorganisms as feed for larval feed

Douillet (2000a,b) found that addition of bacteria significantly enhanced the growthrate of rotifers (Brachionus plicatilis) grown in synxenic culture. A range of Gram-positive and Gram-negative bacterial cultures were tested, both laboratory isolatesand commercial preparations. None of the commercial preparations showed consis-tent positive effects, but particularly a laboratory isolate of Alteromonas spp. repeat-edly increased growth rate. Rombaut et al. (1999) and Gatesoupe (1991) also foundthat a range of bacterial isolates had positive effects on growth rates of rotifers.

Makridis et al. (2000a) did not evaluate growth rates, but showed that bothrotifers (B. plicatilis) and Artemia franciscana grazed on bacteria added to culturewater, and that a significant proportion of their bacterial flora consisted of the intro-duced bacteria. This “bioencapsulation” may be a way to introduce probiotic bacte-rial cultures to fish larvae.

6.2. Microorganisms as larval feed

Addition of high numbers (107 CFU/ml) of a bacterial culture (CA2) to Pacificoyster larvae (Crassostrea gigas Thunberg) caused a depression of both survival andgrowth, whereas addition of lower bacterial numbers (104–106 CFU/ml) increasedgrowth rate of the larvae (Douillet and Langdon, 1994). Also, the addition of variousalgal preparations enhanced growth of oyster larvae (McCausland et al., 1999).McIntosh et al. (2000) added BioStart HB-1 or HB-2 with different Bacillus species totank water of juvenile saltwater shrimp (Litopenaeus vannamei), and saw no effect onweight gain, survival or water quality parameters. Unfortunately, no bacteriologicaldata are included and the concentration and persistence of the added bacteria cannot


be evaluated. Similarly, no weight gain was found for sea bass when placed in recircu-lating water with a bacterial cell density of 104 to 105 CFU/ml (Leonard et al., 2000).

6.3. Bacteria as fish feed

Several reports indicate that addition of bacteria or microalgae to fish feed can causean increase in weight gain (McCausland et al., 1999). Ramesh et al. (1999) addedplant carbohydrates to fingerlings of carp (Cyprinus carpio) and rohu (Labeo rohita).This caused a doubling of bacterial numbers in the water and significant biofilmgrowth on the carbohydrate particles. The increased weight gain of the carp being20–50% higher than the control, was attributed to this increase in microbial biomass.Also, the addition of Enterococcus faecium (strain M74) to fish feed at a level of5 × 105 CFU/g caused an increase in weight gain of carp (Cyprinus carpio) fry(Bogut et al., 1998). Incorporation of a Lactobacillus spp. in flounder (Paralichthysolivaceus) feed at a level of 5 × 103 CFU/g caused an increase in the intestinal countsof Lactobacillus spp. and an increase in weight gain of the fishes (Byun et al., 1997).McCausland et al. (1999) found that the growth of juvenile Pacific oysters(Crassostrea gigas) increased significantly if fed a diet containing live microalgaebut the addition had no effect on water quality parameters and disease survival.

7. USE OF MICROBIAL CULTURES AS WATER TREATMENT

A range of commercial bacterial products are used to control water quality in aqua-culture (Moriarty, 1997). This is due to the ability of the bacteria to participate in theturnover of organic nutrients in the ponds. However, there are few scientifically docu-mented cases in which bacteria have assisted in bioaugmentation, with the notableexception of manipulating the NH3/NO2/NO3 balance (Verschuere et al., 2000a) inwhich nitrifying bacteria are used to remove toxic NH3 (and NO2).

Fish expel nitrogen waste as NH3 or NH4+ resulting in rapid build up of ammonia

compounds which are highly toxic to fish (Hagopian and Riley, 1998). Nitrate, in con-trast, is significantly less toxic being tolerated in concentrations of several thousand mgper litre. Several bacteria, e.g. Nitrosomonas, convert ammonia to nitrite and other bac-teria, e.g. Nitrobacter, further mineralize nitrite to nitrate. Nitrifying bacteria excretepolymers (Hagopian and Riley, 1998), allowing them to associate with surfaces andform biofilms. Recirculating systems must employ biofilters to remove ammonia, andSkjølstrup et al. (1998) demonstrated a 50% reduction in both ammonia and nitrite inan experimental fluidized biofilter in a rainbow trout recirculating unit.

8. PREBIOTICS

The addition of high doses of probiotic strains to established microbial communi-ties of fish can provoke a temporary change in the composition of the microbial


community as described above. However, within a few days after administration hadstopped, the added strains disappeared from the system (Jöborn et al., 1997; Ringøand Gatesoupe, 1998). Prebiotics are a way to increase the population level ofalready colonized beneficial bacteria with antagonistic ability (Roberfroid, 1993,1998, 2001). Prebiotics have been defined with reference to the intestinal tract. Theyare “a non-digestible food ingredient beneficially affecting the host by selectivelystimulating the growth and/or activity of one or a limited number of bacteria in thecolon”. In mammalian systems focus has been on the non-digestible oligosaccha-rides, in particular inulin-type fructans. Inulins occur naturally in several vegetablesand greens, such as garlic, chicory, asparagus and grass (Roberfroid, 1993; Van Loo et al., 1995). Several LAB are able to ferment inulin and other fructans – thusLactobacillus plantarum strains isolated from Thai fermented fish products fermentinulin (Paludan-Müller et al., 1999). Some of the Carnobacterium strains isolatedfrom fish intestinal tract are also capable of fermenting inulin (Ringø et al., 1998;Ringø and Olsen, 1999). It is also known that dietary inulin resulted in damage tointestinal enterocytes of the salmonid fish Arctic charr (Salvelinus alpinus L.)(Olsen et al., 2001), and that dietary inulin alters the adherent gut microbiota ofArctic charr (Ringø, Myklebust, Mayhew and Olsen, unpublished results). However,the effects of dietary inulin on fish welfare are not yet known.

Another aspect related to fish health is modulation of immune response andbarrier function by dietary means (Sakai, 1999). It is well known from endothermicanimals that any method leading to an improved feed intake in the first days afterweaning, could reduce inflammatory symptoms, and improve both intestinal mor-phology and resistance against pathogens (Bosi, 2000). However, knowledge aboutdietary tools to modulate the immune response and barrier function in the fish isincompletely known, and this is a topic for further study. Even at a very young stagesome immunostimulation may occur, as Vadstein et al. (1993) showed that feedinghalibut larvae with alginate rich in mannuronic acid polymer enhanced viabilityduring 4 weeks from 10 to 55%. Such effects may be explained by increase in, e.g.,complement activation as was found for sea bass (Dicentrarchus labrax) which werefed a glucan and ascorbic acid containing diet (Bagni et al., 2000).


Even though several studies have shown that the probiotic concept has potential inthe aquaculture sector, much work is still needed. Some of the most promising datastem from field trials where addition of probiotics to the water on a routine basisincreased survival of fish or crustaceans (Nogami et al., 1997; Moriarty, 1998;Queiroz and Boyd, 1998; Rengpipat et al., 1998). Several aspects, however, need tobe dealt with, in order to develop this concept further. Owing to the relatively highnumbers of bacteria that must be added, fish probiotics currently should be developedfor hatcheries, larval rearing units and recirculating systems.


It is crucial that the mechanisms involved in the in vivo probiotic effect be deter-mined (Berg, 1998; Atlas, 1999). Some go as far as stating that “without specificcause and effect relationships that can be substantiated scientifically, the use of probiotics remains controversial and should not be endorsed by the scientific com-munity” (Atlas, 1999). Even with a slightly less rigorous attitude (e.g. acknowledg-ing the data from several field trials), understanding mechanisms is a requirementfor any long-term commercial use as this is needed to determine any possible sideeffects on the environment, e.g. will the addition of probiotics alter the microbialcommunity on a permanent scale and will this subsequently affect turnover oforganic and inorganic compounds in the particular environment? Thus, the anti-microbial effect of some Bacillus and Pseudomonas species is caused by productionof antibiotics (Marahiel et al., 1993; Duffy and Défago, 1999; Msadek, 1999) andthis is obviously not a viable path in an attempt to find non-antibiotic substitutes fordisease control. An understanding of the in vivo mechanism(s) would also allow fora much more efficient and intelligent selection of potential probionts. At present, nota single study has seriously compared in vitro and in vivo antagonism. Therefore,it is not known if the screening of thousands of isolates for antagonistic activity inin vitro assay has any importance for their in vivo effect. Determining mechanismsof activity is not an easy task, however, some options exist. Comparing phenotypiccharacteristics and disease suppressing abilities (against phytopathogenic fungi) offluorescent pseudomonads has shown that for some strains, production of cyanide isimportant (Ellis et al., 2000). Mutant strains, e.g. constructed by random transposonmutagenesis, could allow for identification of clones with no disease preventiveeffect. Subsequent cloning and sequencing of the genes affected by the knockoutcould help clarify mechanisms. It has been hypothesized that iron chelation isimportant for the antagonism of pseudomonads in the rhizosphere and this hypoth-esis has been tested by comparing the in vivo disease suppressing effects of a wildtype strain and siderophore negative mutants (Buysens et al., 1996).

A particular aspect concerns the testing of probiotic cultures. The use of field trials under real conditions is obviously the ultimate test. However, an intermediatestep in terms of infection model systems using live hosts, is often needed. Owing tothe very high inherent (biological) variation in such systems, the model infectionstudies should be carried out with a sufficient number of replicates to allow forproper statistical treatment. As discussed in Section 5, analyses normally used todescribe and compare survival data must be used. Even with more appropriate statistical analysis, the development of the probiotic principle would benefit greatlyfrom more stable infection models.

It must also be recognized that a particular probiont which may work in one sys-tem (Gram et al., 1999; Spanggaard et al., 2001) may be completely ineffective inanother host–pathogen system (Gram et al., 2001). Therefore, more detailed knowl-edge of the pathogenic agents, their virulence factors and their interaction with thehost would be of great importance.



Different approaches have been used for introducing the probiont to the system.The organism may be live or in a freeze dried state. It can be added directly to thewater or incorporated in the feed; either pelleted or live feed. Nothing is knownabout how each of these treatments affects the viability of the organisms or the probiotic effect. Knowledge of proliferation and invasion sites of the pathogenwould assist in determining whether a water borne or food borne vehicle is the mostappropriate. Such understanding is required for further technological developments.Several studies have found that a single treatment with probiotic culture is notenough and that the organism(s) must be added on a more continuous basis (Nogamiet al., 1997; Moriarty, 1998; Gram et al., 1999), however, the robustness of the systems, e.g. required concentration of probiont, required frequency of addition,effects of changing temperature, has not been documented.

Finally, legal matters must be resolved. Is probiotic treatment classified as amedical issue (treating animals) or an environmental issue (treating water) and ineither case, who is responsible for control? Also, no cost–benefit analysis has yetbeen carried out. Whilst the application of probiotic technology is likely to increasecosts per se, it must be emphasized that if used succesfully (e.g. table 7), there maybe tremendous benefits due to a more stable and therefore higher production. Also,as some uses of antibiotics may be prohibited, use of probiotics may gain widerinterest.

In conclusion, there are several promising developments for fish probiotics,however, this will certainly not become the cure for all maladies.

ACKNOWLEDGEMENTS

Critical reviewing and valuable comments from Dr Harry Birkbeck, University of Glasgow, Dr JorunnSkjermo, SINTEF, and Mette Hjelm, Danish Institute for Fisheries Research are appreciated.

REFERENCES

Alderman, D.J., Hastings, T.S., 1998. Antibiotic use in aquaculture: development of antibiotic resistance –potential for consumer health risks. Int. J. Food Sci. Technol. 33, 139–155.

Alves, D., Specker, J.L., Bengtson, D.A., 1999. Investigations into the causes of early larval mortality incultured summer flounder (Paralichthys dentatus L.). Aquaculture 176, 155–172.

Atlas, R.M., 1999. Probiotics – snake oil for the new millennium. Environ. Microbiol. 1, 377–382.Austin, B., 1982. Taxonomy of bacteria isolated from coastal marine fish-rearing units. J. Appl. Bacteriol.

52, 253–268.Austin, B., 1983. Bacterial microflora associated with a coastal marine fish-rearing unit. J. Mar. Biol.

Ass. UK 63, 585–592.Austin, B., Al-Zahrani, A.M.J., 1988. The effect of antimicrobial compounds on the gastrointestinal

microflora of rainbow trout, Salmo gairdneri Richardson. J. Fish Biol. 33, 1–14.Austin, B., Baudet, E., Stobie, M., 1992. Inhibition of bacterial fish pathogens by Tetraselmis suecica.

J. Fish Dis. 15, 55–61.Austin, B., Stuckey, L.F., Robertson, P.A.W., Effendi, I., Griffith, D.R.W., 1995. A probiotic strain of

Vibrio alginolyticus effect in reducing diseases caused by Aeromonas salmonicida, Vibrio anguillarumand Vibrio ordalii. J. Fish Dis. 18, 93–96.


Bagni, M., Archetti, L., Amadori, M., Marino, G., 2000. Effect of long-term administration of an immuno-stimulant diet on innate immunity in sea bass (Dicentrarchus labrax). J. Vet. Med. B 47, 745–751.

Bell, G.R., 1966. On the microbial flora of strea-incubated eggs of the Pacific salmon (Oncorhynchus).Bull. Off. Int. Epiz. 65, 769–776.

Berg, R.D., 1998. Probiotics, prebiotics or ‘conbiotics’? Trends Microbiol. 6, 89–92.Bergh, Ø., 1995. Bacteria associated with early life stages of halibut, Hippoglossus hippoglossus L.,

inhibit growth of a pathogenic Vibrio spp. J. Fish Dis. 18, 31–40.Bergh, Ø., Naas, K.E., Harboe, T., 1994. Shifts in the intestinal microflora of Atlantic halibut

(Hippoglossus hippoglossus) larvae during first feeding. Can. J. Fish. Aquat. Sci. 51, 1899–1903.Bernadsky, G., Rosenberg, E., 1992. Drag-reducing properties of bacteria from the skin mucus of the

cornetfish (Fistularia commensonii). Microb. Ecol. 24, 35–40.Blanch, A.R., Alsina, M., Simon, M., Jofre, J., 1997. Determination of bacteria associated with reared

turbot (Scophthalmus maximus) larvae. J. Appl. Bacteriol. 82, 729–734.Bly, J.E., Quiniou, S.M.-A., Lawson, L.A., Clem, L.W., 1997. Inhibition of Saprolegnia pathogenic for

fish by Pseudomonas fluorescens. J. Fish Dis. 20, 35–40.Bogut, I., Milakovic, Z., Bukvic, Z., Brkic, S., Zimmer, R., 1998. Influence of probiotic (Streptococcus

faecium M74) on growth and content of intestinal microflora in carp (Cyprinus carpio). Czech.J. Anim. Sci. 43, 231–235.

Bosi, P., 2000. Modulation of immune response and barrier function in the piglet gut by dietary means.Asian-Austr. J. Anim. Sci. 13, 278–293.

Burgess, J.G., Jordan, E.M., Bregu, M., Mearns-Spragg, A., Boyd, K.G., 1999. Microbial antagonism:a neglected avenue of natural product research. J. Biotechnol. 70, 27–32.

Buysens, S., Heungens, K., Poppe, J., Hofte, M., 1996. Involvement of pyochelin and pyoverdin in sup-pression of pythium-induced damping-off of tomato by Pseudomonas aeruginosa 7NSK2 Appl.Environ. Microbiol. 62, 865–871.

Byun, J.-W., Park, S.-C., Benno, Y., Oh, T.-K., 1997. Probiotic effect of Lactobacillus sp. DS-12 in floun-der (Paralichthys olivaceus). J. Gen. Appl. Microbiol. 43, 305–308.

Cahill, M.M., 1990. Bacterial flora of fishes: a review. Microb. Ecol. 19, 21–41.Conway, P.L., 1996. Selection criteria for probiotic microorganisms. Asia Pacific J. Clin. Nutr. 5, 10–14.Corrier, D.E., Nisbet, D.J., Byrde, J.A., Hargis, B.M., Keith, N.K., Peterson, M., Deloach, J.R., 1998.

Dosage titration of characterized competitive exclusion culture to inhibit Salmonella colonization inbroiler chickens during growout. J. Food Prot. 61, 796–801.

Cox, D.R., 1972. Regression models and life tables. J. Roy. Stat. Soc. B 34, 187–220.Cox, D.R., Oakes, 1984. Analysis of Survival Data. Monographs on Statistics and Applied Probability.

Chapman and Hall, London.DePaola, A., Peeler, J.T., Rodrick, G.E., 1995. Effect of oxytetracycline-medicated feed on antibiotic

resistance of Gram-negative bacteria in catfish ponds. Appl. Environ. Microbiol. 61, 2335–2340.Dierberg, F.E., Kiattisimkul, W., 1996. Issues, impacts and implications of shrimp aquaculture in

Thailand. Environ. Man. 20, 649–666.Dopazo, C.P., Lemos, M.L., Lodeiros, C., Bolinches, J., Barja, J.L., Toranzo, A.E., 1988. Inhibitory activity

of antibiotic-producing marine bacteria against fish pathogens. J. Appl. Bacteriol. 65, 97–101.Douillet, P.A., 2000a. Bacterial additives that consistently enhance rotifer growth under synxenic culture

conditions. 1. Evaluation of commercial products and pure isolates. Aquaculture 182, 249–260.Douillet, P.A., 2000b. Bacterial additives that consistently enhance rotifer growth under synxenic culture

conditions. 2. Use of single and multiple bacterial probiotics. Aquaculture 182, 241–248.Douillet, P.A., Langdon, C.J., 1994. Use of a probiotic for the culture of larvae of the Pacific oyster

(Crassostrea gigas Thunberg). Aquaculture 119, 25–40.Droleskey, R.E., Corrier, D.E., Nisbet, D.J., Deloach, J.R., 1995. Colonization of cecal mucosal epithe-

lium in chicks treated with a continuous-flow culture of 29 characterized bacteria – confirmation byscanning electron-microscopy. J. Food Prot. 58, 837–842.

Duffy, B.K., Défago, G., 1999. Environmental factors modulating antibiotic and siderophore biosynthe-sis by Pseudomonas fluorescens biocontrol strains. Appl. Environ. Microbiol. 65, 2429–2438.

Ellis, R., Timms-Wilson, T.M., Bailey, M.J., 2000. Identification of conserved traits in fluorescentpseudomonads with antifungal activity. Environ. Microbiol. 2, 274–284.

FAO, 1998. The State of World Fisheries and Aquaculture. FAO, Rome, Italy.

FAO/NACA, 1995. Regional Study and Workshop on the Environmental Assessment and Managementof Aquaculture Development (TCP/RAS/2253). NACA Environment and Aquaculture DevelopmentSeries No. 1. Network of Aquaculture Centres in Asia-Pacific, Bangkok, Thailand.

FAO/NACA/WHO, 1999. Food Safety Issues Associated with Products from Aquaculture. WHOTechnical Report Series No. 883. WHO, Geneva.

Fahrmeir, L., Tutz, G., 1994. Multivariate Statistical Modelling Based on Generalized Linear Models.Springer Series in Statistics. Springer Verlag, Berlin.

Fuller, R., 1989. Probiotics in man and animal. J. Appl. Bacteriol. 66, 365–378.Gatesoupe, F.J., 1990. The continuous feeding of turbot larvae Scophthalmus maximus, and control of the

bacterial environment of rotifers. Aquaculture 89, 139–148.Gatesoupe, F.J., 1991. The effect of three strains of lactic acid bacteria on the production rate of rotifers,

Brachionus plicatilis, and their dietary value for larval turbot, Scophthalmus maximus. Aquaculture96, 335–342.

Gatesoupe, F.J., 1994. Lactic acid bacteria increase the resistance of turbot larvae, Scophthalmusmaximus, against pathogenic vibrio. Aquat. Living Resour. 7, 277–282.

Gatesoupe, F.J., 1997. Siderophore production and probiotic effect of Vibrio spp. associated with turbotlarvae, Scophthalmus maximus. Aquat. Living Resour. 10, 239–246.

Gatesoupe, F.J., 1999. The use of probiotics in aquaculture. Aquaculture 180, 147–165.Gennari, M., Tomaselli, S., 1988. Changes in aerobic microflora of skin and gills of Mediterranean

sardines (Sardina pilchardus) during storage in ice. Int. J. Food Microbiol. 6, 341–347.Gibson, L.F., Woodworth, J., George, A.M., 1998. Probiotic activity of Aeromonas media on the pacific

oyster, Crassostrea gigas, when challenged with Vibrio tubiashii. Aquaculture 169, 111–120.Gildberg, A., Mikkelsen, H., 1998. Effects of supplementing the feed to Atlantic cod (Gadus morhua)

fry with lactic acid bacteria and immuno-stimulating peptides during a challenge trial with Vibrioanguillarum. Aquaculture 167, 103–113.

Gildberg, A., Mikkelsen, H., Sandaker, E., Ringø, E., 1997. Probiotic effect of lactic acid bacteria in thefeed on growth and survival of fry of Atlantic cod (Gadus morhua). Hydrobiologia 352, 279–285.

Gildberg, A., Johansen, A., Bøgwald, J., 1995. Growth and survival of Atlantic salmon (Salmo salar) frygiven diets supplemented with fish protein hydrolysate and lactic acid bacteria during a challenge trialwith Aeromonas salmonicida. Aquaculture 138, 23–34.

Gillespie, N.C., Macrae, I.C., 1975. The bacterial flora of some Queensland fish and its ability to causespoilage. J. Appl. Bacteriol. 39, 91–100.

Gomez-Gil, B., Roque, A., Turnbull, J.F., 2000. The use and selection of probiotic bacteria for use in theculture of larval aquatic organisms. Aquaculture 191, 259–270.

Gonzalez, C.J., Lopez-Diaz, T.M., Garcia-Lopez, M.L., Prieto, M., Otero, A., 1999. Bacterial microfloraof wild brown trout (Salmo trutta), wild pike (Esox lucius), and aquacultured rainbow trout(Oncorhynchus mykiss). J. Food Prot. 62, 1270–1277.

Gram, L., 1993. Inhibitory effect against pathogenic and spoilage bacteria of Pseudomonas strains isolatedfrom spoiled and fresh fish. Appl. Environ. Microbiol. 59, 2197–2203.

Gram, L., Huss, H.H., 2000. Fresh and processed fish and shellfish. In: Lund, B.M., Baird-Parker, T.C.,Gould, G.W. (Eds.), The Microbiological Safety and Quality of Foods. Chapman and Hall, London,pp. 472–506.

Gram, L., Wedell-Neergaard, C., Huss, H.H., 1990. The bacteriology of spoiling Lake Victorian Nileperch (Lates niloticus). Int. J. Food Microbiol. 10, 303–316.

Gram, L., Melchiorsen, J., Spanggaard, B., Huber, I., Nielsen, T.F., 1999. Inhibition of Vibrioanguillarum by Pseudomonas fluorescens AH2, a possible probiotic treatment of fish. Appl. Environ.Microbiol. 65, 969–973.

Gram, L., Løvold, T., Melchiorsen, J., Nielsen, J., Spanggaard, B., 2001. In vitro antagonism of theprobiont Pseudomonas fluorescens strain AH2 against Aeromonas salmonicida does not conferprotection of salmon against furunculosis. Aquaculture 199, 1–11.

Grave, K., Engelstad, M., Søli, N.E., Håstein, T., 1990. Utilization of antibacterial drugs in salmonidfarming in Norway during 1980–1988. Aquaculture 86, 347–358.

Grave, K., Markestad, A., Bangen, M., 1996. Comparison in prescribing patterns of antibacterial drugsin salmonid farming in Norway during the periods 1980–1988 and 1989–1994. J. Vet. Pharmacol.Therap. 19, 184–191.


Guarner, F., Schaafsma, G.J., 1998. Probiotics. Int. J. Food Microbiol. 39, 237–238.Hagopian, D.S., Riley, J.G., 1998. A closer look at the bacteriology of nitrification. Aquacult. Eng.

18, 223–244.Hansen, G.H., Olafsen, J.A., 1999. Bacterial interactions in early life stages of marine cold water fish.

Microb. Ecol. 38, 1–26.Harrison, J.M., Lee, J.S., 1969. Microbial evaluation of Pacific shrimp processing. Appl. Microbiol.

18, 188–192.Havenaar, R., Ten Brink, B., Huis in’t Veld, J.H.J., 1992. Selection of strains for probiotic use. In: Fuller, R.

(Ed.), Probiotics. The Scientific Basis. Chapman and Hall, London, pp. 209–224.Holzapfel, W.H., Haberer, P., Snel, J., Schillinger, U., Huis in’t Veld, J.H.J., 1998. Overview of gut flora

and probiotics. Int. J. Food Microbiol. 41, 85–101.Horsley, R.W., 1973. The bacterial flora of the Atlantic salmon (Salmo salar L.) in relation to its

environment. J. Appl. Bacteriol. 36, 377–386.Huber, I., Spanggaard, B., Appel, K.F., Rossen, L., Nielsen, T., Gram, L., 2004. Phylogenetic analysis

and in situ identification of the intestinal microbial community of rainbow trout. J. Appl. Microbiol.96, 117–132.

Huys, L., Dhert, P., Robles, R., Ollevier, F., Sorgloos, P., Swings, J., 2001. Search for beneficial bacte-rial strains for turbot (Scophthalmus maximus L.) larviculture. Aquaculture 193, 25–37.

ICMSF (International Commission for the Microbiological Specifications for Foods), 2004. TheMicrobial Ecology of Foods. Book 6. 3rd edition. Aspen, New York (in press).

Jöborn, A., Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1997. Colonization in the fishintestinal tract and production of inhibitory substances in intestinal mucus and faecal extracts byCarnobacterium sp. strain K1. J. Fish Dis. 20, 383–392.

Jöborn, A., Dorsch, M., Olsson, J.C., Westerdahl, A., Kjelleberg, S., 1999. Carnobacterium inhibens sp. nov.,isolated from the intestine of Atlantic salmon (Salmo salar). Int. J. System. Bacteriol. 49, 1891–1898.

Karunasagar, I., Pai, R., Malathi, G.R., Karunasagar, I., 1994. Mass mortality of Penaeus monodon larvaedue to antibiotic-resistant Vibrio harvyei infection. Aquaculture 128, 203–209.

Kennedy S.B., Tucker Jr, J.W., Neidig, C.L., Vermeer, G.K., Cooper, V.R., Jarrell, J.L, Sennett, D.G.,1998. Bacterial management strategies for stock enhancement of warmwater marine fish: a case studywith common snook (Centropomus undecimalis). Bull. Mar. Sci. 62, 573–588.

Kim, K.H., Hwang, Y.J., Bai, S.C., 1999. Resistance to Vibrio alginolyticus in juvenile rockfish (Sebastesschlegeli) fed diets containing different doses of aloe. Aquaculture 180, 13–21.

Kmet, V., Lucchini, F., 1999. Aggregation of sow lactobacilli with diarrhoeagenic Escherichia coli.Zentralbl. Veterinarmed. (B) 46, 683–687.

Laycock, R.A., Regier, L.W., 1970. Pseudomonads and achromobacters in the spoilage of irradiatedhaddock of different pre-irradiation quality. Appl. Microbiol. 20, 333–341.

Lemos, M.L., Toranzo, A.E., Barja, J.L., 1985. Antibiotic activity of epiphytic bacteria isolated fromintertidal seaweeds. Microb. Ecol. 11, 149–163.

Leonard, N., Blancheton, J.P., Guiraud, J.P., 2000. Populations of heterotrophic bacteria in an experi-mental recirculating aquaculture system. Aquacult. Eng. 22, 109–120.

Levy, S.B., 1998. The challenge of antibiotic resistance. Sci. Amer. March, 32–39.Lilly, D.M., Stillwell, R.H., 1965. Probiotics: growth-promoting factors produced by microorganisms.

Science 147, 747–748.Lødemel, J.B., Mayhew, T.M., Myklebust, R., Olsen, R.E., Espelid, S., Ringø, E., 2001. Effect of three

dietary oils on disease susceptibility in Arctic charr (Salvelinus alpinus L.) during cohabitant chal-lenge with Aeromonas salmonicida subsp. salmonicida. Aquacult. Res. 32, 935–945.

Maeda, M., Liao, I.C., 1992. Effect of bacterial population on the growth of prawn larva, Penaeus monodon.Bull. Natl. Res. Inst. Aquacult. 21, 25–29.

MAFF, 1998. A review of antimicrobial resistance in the food chain. A Technical Report for MAFF(Ministry of Agriculture, Fisheries, and Food, UK).

Magnor-Jensen, A., Adoff, G.R., 1987. Drinking activity of the newly hatched larvae of cod (Gadusmorhua). Fish Physiol. Biochem. 3, 99–103.

Makridis, P., Fjellheim, A.J., Skjermo, J., Vadstein, O., 2000a. Control of the bacterial flora ofBrachionus plicatilis and Artemia franciscana by incubation in bacterial suspensions. Aquaculture185, 207–218.


Makridis, P., Fjellheim, A.J., Skjermo, J., Vadstein, O., 2000b. Colonization of the gut in first feedingturbot by bacterial strains added to the water or bioencapsulated in rotifers. Aquacult. Intern. 8, 367–380.

Marahiel, M.A., Nakano, M.M., Zuber, P., 1993. Regulation of peptide antibiotic production in Bacillus.Mol. Microbiol. 7, 631–636.

McCausland, M.A., Malcolm, R.B., Barrett, S.M., Diemar, J.A., Heasman, M.P., 1999. Evaluation of livemicroalgae and microalgal pastes as supplementary food for juvenile Pacific oysters (Crassostreagigas). Aquaculture 174, 323–342.

McIntosh, D., Samocha, T.M., Jones, E.R., Lawrence, A.L., McKee, D.A., Horowitz, S., Horowitz, A., 2000.The effect of a commercial bacterial supplement on the high-density culturing of Litopenaeus vannameiwith low-protein diet in an outdoor tank system and no water exchange. Aquacult. Eng. 21, 215–227.

Moriarty, D.J.W., 1997. The role of microorganisms in aquaculture ponds. Aquaculture 151, 333–349.Moriarty, D.J.W., 1998. Control of luminous Vibrio species in penaeid aquaculture ponds. Aquaculture

164, 351–358.Msadek, T., 1999. When the going gets tough: survival strategies and environmental signaling networks

in Bacillus subtilis. Trends Microbiol. 7, 201–207.Mudarris, M., Austin, B., 1988. Quantitative and qualitative studies of the bacterial microflora of turbot,

Scophthalmus-maximum L., gills. J. Fish Biol. 32, 223–229.Munro, P.D., Birkbeck, T.H., Barbour, A., 1993. Influence of rate of bacterial colonisation of the gut of

turbot larvae on larval survival. In: Reinertsen, H., Dahle, L.A., Jørgensen, L., Tvinnereim, K. (Eds.),Fish Farming Technology. Balkema, Rotterdam, pp. 85–92.

Munro, P.D., Barbour, A., Birkbeck, T.H., 1995. Comparison of the growth and survival of larval turbotin the absence of culturable bacteria with those in the presence of Vibrio anguillarum, Vibrioalginolyticus, or a marine Aeromonas sp. Appl. Environ. Microbiol. 61, 4425–4428.

Munro, P.D., Henderson, R.J., Barbour, A., Birkbeck, T.H., 1999. Partial decontamination of rotifers withultraviolet radiation: the effect of changes in the bacterial load and flora of rotifers on mortalities instart-feeding larval turbot. Aquaculture 170, 229–244.

Naidu, A.S., Bidlack, W.R., Clemens, R.A., 1999. Probiotic spectra of lactic acid bacteria (LAB). Crit.Rev. Food Sci. Nutr. 38, 13–126.

Nikoskelainen, S., Salminen, S., Bylund, G., Ouwehand, A.C., 2001a. Characterization of the propertiesof human- and dairy-derived probiotics for prevention of infectious diseases in fish. Appl. Environ.Microbiol. 67, 2430–2435.

Nikoskelainen, S., Ouwehand, A.C., Salminen, S., Bylund, G., 2001b. Protection of rainbow trout(Oncorhynchus mykiss) from furunculosis by Lactobacillus rhamnosus. Aquaculture 198, 229–236.

Nogami, K., Maeda, M., 1992. Bacteria as biocontrol agents for rearing larvae of the crab Protunustrituberculatus. Can. J. Fish. Aquat. Sci. 49, 2373–2376.

Nogami, K., Hamasaki, K., Maeda, M., Hirayama, K., 1997. Biocontrol method in aquaculture for rear-ing the swimming crab larvae Protunus trituberculatus. Hydrobiologia 358, 291–295.

Norm-Vet, 1999. Usage of antimicrobial agents in animals and occurrence of antimicrobial resistance inbacteria from animals, feed and food in Norway 1999, Kruse, H. (Ed.). The Norwegian ZoonosisCentre, Oslo, Norway (http://www.vetinst.no/arkiv/Zoonosesenteret/Norm-Vet_99.pdf).

Olsen, R.E., Myklebust, R., Kryvi, H., Mayhew, T.M., Ringø, E., 2001. Damaging effect of dietary inulinto intestinal enterocytes in Arctic charr (Salvelinus alpinus L.). Aquacult. Res. 32, 931–934.

Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1992. Intestinal colonization potential of tur-bot (Scophthalmus maximus)- and dab (Limanda limanda)-associated bacteria with inhibitory effectsagainst Vibrio anguillarum. Appl. Environ. Microbiol. 58, 551–556.

Paludan-Müller, C., Huss, H.H., Gram, L., 1999. Characterization of lactic acid bacteria isolated from aThai low-salt fermented fish product and the role of garlic as substrate for fermentation. Int. J. FoodMicrobiol. 46, 219–230.

Park, S.C., Shimamura, I., Fukunaga, M., Mori, K.-I., Nakai, T., 2000. Isolation of bacteriophages spe-cific to a fish pathogen, Pseudomonas plecoglossicida, as a candidate for disease control. Appl.Environ. Microbiol. 66, 1416–1422.

Parker, R.B., 1974. Probiotics, the other half of the antibiotic story. Anim. Nutr. Health 29, 4–8.Planas, M., Cunha, I., 1999. Larviculture of marine fish: problems and perspectives. Aquaculture 177,

171–190.


Queiroz, J.F., Boyd, C.E., 1998. Effects of a bacterial inoculum in channel catfish ponds. J. WorldAquacult. Soc. 29, 67–73.

Raa, J., 1996. The use of immunostimulatory substances in fish and shellfish farming. Rev. Fish. Sci.4, 229–288.

Ramesh, M.R., Shankar, K.M., Mohan, C.V., Varghese, T.J., 1999. Comparison of three plant substratesfor enhancing carp growth through bacterial biofilm. Aquacult. Eng. 19, 119–131.

Reitan, K.I., Natvik, C.M., Vadstein, O., 1998. Drinking rate, uptake and microalgae in turbot larvae.J. Fish Biol. 53, 1145–1154.

Rengpipat, S., Phianphak, W., Piyatiratitivorakul, S., Menasveta, P., 1998. Effects of a probiotic bac-terium on black tiger shrimp Penaeus monodon survival and growth. Aquaculture 167, 301–313.

Rengpipat, S., Rukpratanporn, S., Piyatiratitivorakul, S., Menasveta, P., 2000. Immunity enhancement onblack tiger shrimp (Penaeus monodon) by a probiont bacterium (Bacillus S11). Aquaculture 191,271–288.

Rico-Mora, R., Voltolina, D., Villaescusa-Celaya, J.A., 1998. Biological control of Vibrio alginolyticusin Skeletonema costatum (Bacillariophyceae) cultures. Aquacult. Eng. 19, 1–6.

Ringø, E., 1993. The effect of chromic oxide (Cr2O3) on aerobic bacterial populations associated withepithelial mucosa of Arctic charr, Salvelinus alpinus (L.). Can. J. Microbiol. 39, 1169–1173.

Ringø, E., Birkbeck, T.H., 1999. Intestinal microflora of fish larvae and fry. Aquacult. Res. 30, 73–93.Ringø, E., Gatesoupe, F.-J., 1998. Lactic acid bacteria in fish: a review. Aquaculture 160, 177–203.Ringø, E., Olsen, R.E., 1999. The effect of diet on aerobic bacterial flora associated with intestine of

Arctic charr (Salvelinus alpinus L.). J. Appl. Microbiol. 86, 22–28.Ringø, E., Strøm, E., 1994. Microflora of Arctic charr, Salvelinus alpinus (L.); gastrointestinal microflora

of free-living fish, and effect of diet and salinity on the intestinal microflora. Aquacult. Fish. Manage.25, 623–629.

Ringø, E., Vadstein, O., 1998. Colonization of Vibrio pelagius and Aeromonas caviae in early developingturbot (Scophthalmus maximus L.) larvae. J. Appl. Microbiol. 84, 227–233.

Ringø, E., Birkbeck, T.H., Munro, P.D., Vadstein, O., Hjelmeland, K., 1996. The effect of early exposureto Vibrio pleagius on the aerobic flora of turbot, Schophthalmus maximus (L.). J. Appl. Bacteriol.81, 207–211.

Ringø, E., Bendiksen, H.R., Gausen, S.J., Sundsfjord, A., Olsen, R.E., 1998. The effect of dietary fattyacids on lactic acid bacteria associated with the epithelial mucosa and from faecalia of Arctic charr,Salvelinus alpinus (L.). J. Appl. Microbiol. 85, 855–864.

Ringø, E., Bendiksen, H.R., Wesmajervi, M.S., Olsen, R.E., Jansen, P.A., Mikkelsen, H., 2000. Lacticacid bacteria associated with the digestive tract of Atlantic salmon (Salmo salar L.). J. Appl.Microbiol. 89, 317–322.

Ringø, E., Lødemel, J.B., Myklebust, R., Kaino, T., Mayhew, T.M., Olsen, R.E., 2001. Epithelium-associated bacteria in the gastrointestinal tract of Arctic charr (Salvelinus alpinus L.). An electronmicroscopical study. J. Appl. Microbiol. 90, 294–300.

Riquelme, C., Araya, R., Vergara, N., Rojas, A., Guaita, M., Candia, M., 1997. Potential probiotic strainsin the culture of the Chilean scallop Argopecten purpuratus (Lamarck, 1819). Aquaculture 154, 17–26.

Riquelme, C., Araya, R., Escribano, R., 2000. Selective incorporation of bacteria by Argopecten purpuratuslarvae: implications for the use of probiotics in culturing systems of the Chilean scallop. Aquaculture181, 25–36.

Riquelme, C., Jorquera, M.A., Rojas, A.I., Avendano, R.E., Reyes, N., 2001. Addition of inhibitor-producing bacteria to mass cultures of Argopecten purpuratus larvae (Lamarck, 1819). Aquaculture192, 111–119.

Roberfroid, M.B., 1993. Dietary fiber, inulin, and oligofructose: a review comparing their physiologicaleffects. C. R. Food Sci. Nutr. 33, 103–148.

Roberfroid, M.R., 1998. Prebiotics and synbiotics: concepts and nutritional properties. Brit. J. Nutr.80, S197–S202.

Roberfroid, M.B., 2001. Prebiotics: preferential substrates for specific germs? Amer. J. Clin. Nutr.73, Suppl., 406S–409S.

Robertson, P.A.W., O’Dowd, C., Burrells, C., Williams, P., Austin, B., 2000. Use of Carnobacterium sp.as a probiotic for Atlantic salmon (Salmo salar L.) and rainbow trout (Oncorhynchus mykiss,Walbaum). Aquaculture 185, 235–243.


Roch, P., 1999. Defense mechanisms and disease prevention in farmed marine invertebrates. Aquaculture172, 125–145.

Rombaut, G., Dhert, Ph., Vandenberghe, J., Vershuere, L., Sorgeloos, P., Verstraete, W., 1999. Selectionof bacteria enhancing the growth rate of axenically hatched rotifers (Brachionus plicatilis).Aquaculture 176, 195–207.

Sakai, M., 1999. Current research status of fish immunostimulants. Aquaculture 172, 63–92.Sakai, M., Yoshida, T., Atsuta, S., Kobayashi, M., 1995. Enhancement of resistance to vibriosis in rainbow

trout Oncorhynchus mykiss (Walbaum), by oral administration of Clostridium butyricum bacterin.J. Fish Dis. 18, 187–190.

Sakata, T., Suit, H., Mitsuoka, T., Katimoto, D., Kadota, H., 1981. Characteristics of obligate anaerobicbacteria in the intestine of freshwater fish. Bull. Jap. Soc. Sci. Fish. 47, 421–427.

Salminen, S., von Wright, A., Morelli, L., Marteau, P., Brassart, D., de Vos, W.M., Fonden, R., Saxelin, M.,Collins, K., Mogensen, G., Birkeland, S.E., Mattila-Sandholm, T., 1998. Demonstration of safety ofprobiotics – a review. Int. J. Food Microbiol. 44, 93–106.

Salvesen, I., Skjermo, J., Vadstein, O., 1999. Growth of turbot (Scophthalmus maximus L.) during firstfeeding in relation to the proportion of r/K-strategists in the bacterial community of the rearing water.Aquaculture 175, 337–350.

Schmidt, A.S., Bruun, M.S., Dalsgaard, I., Pedersen, K., Larsen, J.L., 2000. Occurrence of antimicrobialresistance in fish-pathogenic and environmental bacteria associated with four Danish rainbow troutfarms. Appl. Environ. Microbiol. 66, 4908–4915.

Scholz, U., Garcia Diaz, G., Ricque, D., Cruz Suarez, L.E., Vargas Albores, F., Latchford, J., 1999.Enhancement of vibriosis resistance in juvenile Penaeus vannamei by supplementation of diets withdifferent yeast products. Aquaculture 176, 271–283.

Shewan, J.M., 1971. The microbiology of fish and fishery products – a progress report. J. Appl. Bacteriol.34, 299–315.

Simidu, W., Kaneko, E., Aiso, K., 1969. Microflora of fresh and stored flatfish Kareius bicoloratus.Nippon Suisan Gakkaishi 35, 77–82.

Skjermo, J., Vadstein, O., 1999. Techniques for microbial control in the intensive rearing of marine larvae.Aquaculture 177, 333–343.

Skjermo, J., Salvesen, I., Oie, G., Olse, Y., Vadstein, O., 1997. Microbially matured water: a techniquefor selection of a non-opportunistic bacterial flora in water that may improve performance of marinelarvae. Aquacult. Int. 5, 13–28.

Skjølstrup, J., Nielsen, P.H., Frier, J.-O., McLean, E., 1998. Performance characteristics of fluidisedbed biofilters in a novel laboratory-scale recirculation system for rainbow trout: nitrification rates,oxygen consumption and sludge collection. Aquacult. Eng. 18, 265–276.

Smith, P., Davey, S., 1993. Evidence for the competitive exclusion of Aeromonas salmonicida from fishwith stress-inducible furunculosis by a fluorescent pseudomonad. J. Fish Dis. 16, 521–524.

Sørum, H., 1999. Antibiotic resistance in Aquaculture Acta. Vet. Scand. Suppl. 92, 29–36.Spanggaard, B., Jørgensen, F., Gram, L., Huss, H.H., 1993. Antibiotic resistance in bacteria isolated from

three freshwater fish farms and an unpolluted stream in Denmark. Aquaculture 115, 195–207.Spanggaard, B., Huber, I., Nielsen, J., Appel, K.F., Nielsen, T.F., Gram, L., 2000a. The microflora of

rainbow trout intestine: A comparison of traditional and molecular identification. Aquaculture 182,1–16.

Spanggaard, B., Huber, I., Nielsen, J., Nielsen, T.F., Gram, L., 2000b. Proliferation and location of Vibrioanguillarum during infection of rainbow trout, Oncorhynchus mykiss (Walbaum). J. Fish Dis. 23,423–427.

Spanggaard, B., Huber, I., Nielsen, J., Sick, E.B., Slierendrecht, W.J., Gram, L., 2001. Antagonism of theindigenous fish microflora against Vibrio anguillarum and subsequent use of non-pathogenic culturesin protection against vibriosis. Environ. Microbiol. 3, 755–765.

Spencer, R.J., Chesson, A., 1994. The effect of Lactobacillus spp. on the attachment of enterotoxigenicEscherichia coli to isolated porcine enterocytes. J. Appl. Bacteriol. 77, 215–220.

Strøm, E., Olafsen, J.A., 1990. The indigenous microflora of wild-captured juvenile cod in net-pen rearing.In: Lesel, R. (Ed.), Microbiology in Poecilotherms. Elsevier, Amsterdam, pp. 181–185.

Strøm, E., Ringø, E., 1993. Changes in the bacterial composition of early developing cod, Gadus morhua (L.)larvae following inoculation of Lactobacillus plantarum into the water. In: Walther, B.T., Fyhn, H.J. (Eds.),


Physiological and Biochemical Aspects of Fish Development. University of Bergen, Norway,pp. 226–228.

Sugita, H., Fukumoto, M., Tsunohara, M., Deguchi, Y., 1987. The fluctuation of the fecal flora of gold-fish Carassius auratus. Nippon Suisan Gakkaishi 53, 1443–1447.

Sugita, H., Fukumoto, M., Deguchi, Y., 1988. Changes in the fecal microflora of goldfish Carassius auratus,associated with diets. Nippon Suisan Gakkaishi 54, 1641–1645.

Sugita, H., Miyajima, C., Kobiki, Y., Deguchi, Y., 1990. The daily fluctuation and inter-individual varia-tion of the faecal flora of carp, Cyprinus carpio. J. Fish Biol. 36, 103–105.

Sugita, H., Matsuo, N., Shibuya, K., Deguchi, Y., 1996a. Production of antibacterial substances by intes-tinal bacteria isolated from coastal crab and fish species. J. Mar. Biotechnol. 4, 220–223.

Sugita, H., Shibuya, K., Shimooka, H., Deguchi, Y., 1996b. Antibacterial abilities of intestinal bacteria infreshwater cultured fish. Aquaculture 145, 195–205.

Sugita, H., Matsuo, N., Hirose, Y., Iwato, M., and Deguchi, Y., 1997a. Vibrio spp. strain NM 10, isolatedfrom the intestine of a Japanese coastal fish, has an inhibitory effect against Pasteurella piscicida.Appl. Environ. Microbiol. 63, 4986–4989.

Sugita, H., Shibuya, K., Hanada, H., Deguchi, Y., 1997b. Antibacterial abilities of intestinal microflora ofthe river fish. Fish. Sci. 63, 778–783.

Surendran, P.K., Mahadeva, I.K., Gopakumar, K., 1985. Succession of bacterial genera during icedstorage of three species of tropical prawns, Penaeus indicus, Matapenaeus dobsoni and M. affinis.Fish. Technol. 22, 117–120.

Surendran, P.K., Joseph, J., Shenoy, A.V., Perigreen, P.A., Mahadeva, I., Gopakumar, K., 1989. Studieson spoilage of commercially important tropical fishes under iced storage. Fish. Res. 7, 1–9.

Tanasomwang, V., Nakai, T., Nishimura, Y., Muroga, K., 1998. Vibrio-inhibiting marine bacteria isolatedfrom black tiger shrimp hatchery. Fish Pathol. 33, 459–466.

Thompson, F.L., Abreu, P.C., Cavalli, R., 1999. The use of microorganisms as food source for Penaeuspaulensis larvae. Aquaculture 174, 139–153.

Trust, T.J., 1975. Facultative anaerobic bacteria in the digestive tract of chum salmon (Oncorhynchusketa) maintained in fresh water under defined culture conditions. App. Microbiol. 29, 663–668.

Trust, T.J., Sparrow, R.A.H., 1974. The bacterial flora in the alimentary tract of freshwater salmonidfishes. Can. J. Microbiol. 20, 1219–1228.

Trust, T.J., Bull, L.M., Currie, B.R., Buckley, J.T., 1979. Obligate anaerobic bacteria in the gastrointestinalmicroflora of the grass carp (Ctenopharyngodon idella), goldfish (Carassius auratus), and rainbowtrout (Salmo gairdneri). J. Fish. Res. Bd. Can. 36, 1174–1179.

Vadstein, O., Øie, G., Olsen, Y., Salvesen, I., Skjermo, J., Skjåk-Bræk, G., 1993. A strategy to obtainmicrobial control during larval development of marine fish. In: Reinertsen, H., Dahle, L.A.,Jørgensen, L., Tvinnereim, K. (Eds.), Fish Farming Technology. Balkema, Rotterdam, pp. 69–75.

van der Waaj, D., Berghuis-de Vries, J.M., Lekkerkerk-van der Wees, J.E.C., 1971. Colonization resistanceof the digestive tract in conventional and antibiotic-treated mice. J. Hyg. Camb. 69, 405–411.

Van Loo, J., Coussement, P., De Leenheer, L., Hoebregs, H., Smits, G., 1995. On the presence of inulinand oligofructose as natural ingredients in the western diet. Crit. Rev. Food Sci. Nutr. 35, 525–552.

Vandenberghe, J., Verdonck, L., Robles-Arozarena, R., Rivera, G., Bolland, A., Balladares, M., Gomez-Gil, B., Calderon, J., Sorgeloos, P., Swings, J., 1999. Vibrios associated with Litopenaeus vannameilarvae, postlarvae, broodstock, and hatchery probionts. Appl. Environ. Microbiol. 65, 2592–2597.

Vanderzant, C., Mroz, E., Nickelson, R., 1970. Microbial flora of Gulf of Mexico and pond shrimp.J. Milk Food Technol. 33, 346–350.

Vanderzant, C., Nickelson, R., Judkins, P.W., 1971. Microbial flora of pond reared brown shrimp(Peneaus aztecus). Appl. Microbiol. 21, 916–921.

Venkataranan, R., Sreenivasan, A., 1953. The bacteriology of freshwater fish. India J. Med. Res. 41,385–392.

Vergin, F. von, 1954. Anti- und Probiotika. Hippokrates, Heft 4, 116–119.Verschuere, L., Rombaut, G. Sorgeloos, P. Verstraete, W., 2000a. Probiotic bacteria as biological control

agents in aquaculture. Microbiol. Mol. Biol. Rev. 64, 655–671.Verschuere, L., Heang, H., Criel, G., Sorgeloss, P., Verstraete, W., 2000b. Selected bacterial strains pro-

tect Artemia spp. from the pathogenic effects of Vibrio proteolyticus CW8T2. Appl. Environ.Microbiol. 66, 1139–1146.



Walker, P., Cann, D., Shewan, J.M., 1970. The bacteriology of “scampi” (Nephrops norvegicus).I. Preliminary bacteriological, chemical, and sensory studies. J. Food Technol. 5, 375–385.

Watanabe, K., 1965. Technological problems of handling and distribution of fresh fish in southern Brazil.In: Kreuzer, R. (Ed.), The Technology of Fish Utilization. Fishing News, London, pp. 44–46.

Westerdahl, A., Olsson, J.C., Kjelleberg, S., Conway, P.L., 1991. Isolation and characterization of turbot(Scophthalmus maximus)-associated bacteria with inhibitory effects against Vibrio anguillarum.Appl. Environ. Microbiol. 57, 2223–2228.

Zhao, S., 2001. Antimicrobial resistance of Salmonella isolated from imported food. Paper presented atthe 101st annual meeting of the American Society for Microbiology, Orlando, Florida, 19–23 May2001.

Zivkovic, R., 1999. Probiotics or microbes against microbes. Acta Med. Croatica 53, 23–28.

418

This review presents updated knowledge on the antagonistic ability of lactic acidbacteria (LAB) isolated from aquatic animals. Numerous strains of LAB includingthe genera Lactobacillus, Carnobacterium, Aerococcus-like, Lactococcus andStreptococcus, are capable of producing antibacterial substances against differentpotential fish-pathogenic bacteria such as Aeromonas salmonicida subsp. salmonicida,Aeromonas hydrophila, Aeromonas caviae, Flavobacterium psychrophilium,Photobacterium damselae subsp. piscicida, Vibrio salmonicida, Vibrio anguillarum,Vibrio splendidus and other Vibrio species, Carnobacterium piscicola, Streptococcusmilleri, several strains of Listeria including Listeria monocytogenes, several strainsof Enterobacteriaceae and Clostridium tyrobytyricum. In addition, many LAB havethe ability to inhibit growth of closely related bacteria including strains of carnobac-teria, lactobacilli, lactococci, leuconostoc and pediococci. This review will focus onthe antagonistic ability of LAB, the effect of LAB administration on intestinal micro-biota, in vivo challenge tests and the use of commercial preparations of live LAB inaquaculture.

1. INTRODUCTION

The aquaculture industry is plagued by many disease problems. Presently, the pre-vention and control of these diseases have concentrated on good husbandry practices

18 Antimicrobial activity of lactic acid bacteria isolated from aquatic animals and the use of lactic acid bacteria in aquaculture1

E. Ringøa, U. Schillingerb and W. Holzapfelb

aSection of Arctic Veterinary Medicine, Department of Food Safety and InfectionBiology, The Norwegian School of Veterinary Science, NO-9292 Tromsø, NorwaybInstitute of Hygiene and Toxicology BFE, D−76131 Karlsruhe, Germany

1Financial support from the Norwegian Research Council (Grant No. 122851/122) is gratefully acknowledged.


Lactic acid bacteria in aquaculture 419

and the use of vaccines and antibiotics. However, treatment or feeding with anti-biotics may cause the development of resistant bacteria (Aoki et al., 1985; Amàbile-Cuevas et al., 1995; Towner, 1995). An alternative method of prevention is the useof probiotics (non-pathogenic bacteria) able to inhibit colonization of and to exertinhibitory effects against undesired microorganisms, and that support the naturalhost microbial defence mechanisms. Six recent reviews have hinted at the impor-tance of the inhibition of pathogens by antibacterial substances produced by the fishmucosal microbiota (Ringø and Gatesoupe, 1998; Gatesoupe, 1999; Hansen andOlafsen, 1999; Ringø and Birkbeck, 1999; Gomez-Gill et al., 2000; Verschuereet al., 2000; Gram and Ringø, 2005, Chapter 17 in this book). The presence of a non-pathogenic microbiota with antibacterial ability on mucosal surface is of importancefor protection as skin, lateral line, gills and the gastrointestinal tract (GIT) or a com-bination of these organs, are suggested to be infection routes of pathogenic bacteria.From their studies, Tatner et al. (1984), Muroga and De La Cruz (1987), Kannoet al. (1989, 1990), Magarinos et al. (1995), and Svendsen and Bøgwald (1997) con-cluded that skin is involved as an infection route, while other investigations (Evelyn,1984; Hjeltnes et al., 1987; Baudin Laurencin and Germon, 1987; Svendsen et al.,1999) suggest that gills served as an essential route of penetration. In contrast tothese results, Chart and Munn (1980), Campbell and Buswell (1983), Horne andBaxendale (1983), Muroga et al. (1987), Chair et al. (1994), Olsson (1995), Grisezet al. (1996), Olsson et al. (1996), Romalde et al. (1996), Jöborn et al. (1997), andLødemel et al. (2001) suggest that the alimentary tract is involved in Vibrio andAeromonas infections. Olsson et al. (1992) put forward the hypothesis that the GITis a site of colonization of Vibrio anguillarum as the pathogen could utilize dilutedturbot intestinal mucus as the sole nutrient source. In a later study, Garcia et al.(1997) concluded that Atlantic salmon (Salmo salar L.) intestinal mucus is an excel-lent growth medium of V. anguillarum. This information reveals an important aspectof the pathogenesis of this bacterial species.

Numerous autogenic factors generated by some members of the GIT microbiotaare believed to influence the capacity of other microorganisms to associate withepithelial surfaces. The role of the indigenous microbiota in infection control withregard to exclusion or reduction of pathogen adhesion has been extensively studiedin humans and other endothermic animals (Fuller, 1997; Naidu et al., 1999), and theantimicrobial compounds known to be produced in the GIT and to be inhibitory tomicrobial cells are lactic acid, short-chain fatty acids (SCFA), H2S and macro-molecular substances such as bacteriocins. These proteinaceous antimicrobials areribosomally produced by certain bacterial strains, and may kill or inhibit sensitivestrains of closely related taxonomic groups. During the past two decades in vivoexperiments have clearly demonstrated inhibitory effects of bacteria associated withthe mucosal surfaces of fish and aquatic organisms against both bacterial pathogensand closely related species (for review see Gatesoupe, 1999; Gomez-Gill et al.,2000; Verschuere et al., 2000). Considering the gut microbial ecology, it seems

420 E. Ringø, U. Schillinger and W. Holzapfel

important to take into account the antibacterial effect of the microbiota associatedwith the gills, skin and digestive tract, and to enhance the level of beneficial bacte-ria by antibacterial effects against undesirable bacteria.

Many research laboratories have attempted to improve and modify the microbialbalance by probiotics in the fish digestive tract and larval aquatic organisms, and agrowing number of scientific papers are dealing with this strategy for microbial con-trol (for review, see Ringø and Gatesoupe, 1998; Gatesoupe, 1999; Ringø andBirkbeck, 1999; Skjermo and Vadstein, 1999; Gomez-Gill et al., 2000; Verschuereet al., 2000; Gram and Ringø, 2005, Chapter 17 in this book). In the review by Ringøand Gatesoupe (1998) devoted to LAB in teleost fish, some information was pre-sented on antagonistic activity of LAB, a more specific overview is, however,needed on antagonistic activity, the use of LAB in aquaculture, and the effects ofLAB administration on intestinal microbiota, survival, growth and competition/interactions in vivo.

The term LAB is used to characterize a broad group of Gram-positive, usually non-motile, non-sporulating bacteria, which are generally catalase-negative, usually lackcytochromes, utilize carbohydrates fermentatively and produce lactic acid as a majoror sole product of sugar fermentation. Members of this group are widespread in natureand comprise both rod-shaped (Carnobacterium, Lactobacillus, Weissella) and coccusforms (Aerococcus, Enterococcus, Lactococcus, Leuconostoc, Pediococcus andStreptococcus) as single cells, pairs or chains. Their distribution is typically associatedwith habitats containing high concentrations of soluble carbohydrate, protein break-down products and vitamins. In endothermic animals, lactobacilli are normal inhabi-tants of the intestinal tract (for review see Juven et al., 1991; Naidu et al., 1999), whilstthe same organisms are seldom isolated from the alimentary tract of fish wherecarnobacteria seem to be more common but not numerically dominant (Ringø andGatesoupe, 1998). However, during recent years, some attempts have been made toincrease the level of the lactobacilli and carnobacteria GIT population by nutritionalfactors such as: 1) chromic oxide (Ringø, 1993b,c), 2) dietary polyunsaturated fattyacids (Ringø et al., 1998), 3) level of dietary lipid (Ringø and Olsen, 1999), 4) dietarylipid source (Ringø et al., 2002a) and 5) dietary inulin (Ringø et al., unpublishedresults). Furthermore, Ringø and Gatesoupe (1998) discussed that the population levels of adherent lactobacilli and carnobacteria in the digestive tract are affected byenvironmental factors such as handling stress, hierarchy formation and salinity. Thesefactors described above must certainly be taken into consideration when discussingantagonism of carnobacteria and other LAB isolated from aquatic animals. Today, itis well documented in several investigations that LAB are a part of the native micro-biota of aquatic animals from hatching and onwards (table 1a). Furthermore, severalstudies during the past decade have reported on the isolation of LAB from cold-smoked fish and fermented fish (table 1b).

In their recent review on probiotic bacteria as biological control agents in aqua-culture, Verschuere et al. (2000) discussed that, as LAB normally account only for

Table 1a. Lactic acid bacteria isolated from aquatic animals

Lactic acid bacteria isolated from

Whole

intestinal Small Large

Source tract Stomach intestine intestine Faeces Gills References

Lactobacillus spp.

Arctic charr A + Ringø, 1993a

A + + Ringø, 1993b

A + + Ringø, 1993c

A + Ringø, 1994

A + + + + Ringø et al., 1998

A + + + Ringø and Strøm, 1994

Atlantic cod + Strøm and Olafsen, 1990

L + Strøm and Ringø, 1993

Atlantic salmon + Westerdahl et al., 1998

A + + Ringø et al., 2000

A + + + Ringø and Bendiksen (upd)

Brown trouta Gonzáles et al., 2000

Herring + Kraus, 1961

+ Olsen et al., 1994

Various fish + Kvasnikov et al., 1977

Lb. plantarum-like

Arctic charr A + + + + Ringø et al., 1998

Saithe + Schrøder et al., 1980

Carnobacterium spp.

Arctic charr A + + + + Ringø et al., 1998

A + Ringø and Olsen, 1999

Rainbow trout + Wallbanks et al., 1990

+ + Jöborn et al., 1997

J intestinal content Spangaard et al., 2000

C. divergens-like Ringø et al., 1997

Arctic charr A + Ringø and Olsen, 1999

A + + Strøm, 1988

Atlantic cod J/A + + Strøm, 1988

Atlantic salmon A + + Gonzáles et al., 2000

Brown trouta

Saithe A + + Strøm, 1988

Wolffish F + Ringø et al., 2001a

C. funditum-like

Arctic charr A + + Ringø et al., 2002a

C. inhibens

Atlantic salmon + Jöborn et al., 1999

C. mobile-like

Arctic charr A + Ringø and Olsen, 1999

C. piscicola-like Ringø et al., 1998

Arctic charr A + + + +A + Ringø and Olsen, 1999

Atlantic cod A + + Strøm and Ringø (upd)

A + Ringø (upd)

Atlantic salmon A + + Ringø et al., 2000

A + Ringø and Holzapfel, 2000

J + Ringø (upd)

J Strøm and Ringø (upd)

+ + + Gonzáles et al., 2000

Continued

Table 1a. Lactic acid bacteria isolated from aquatic animals—cont’d

Lactic acid bacteria isolated from

Whole

intestinal Small Large

Source tract Stomach intestine intestine Faeces Gills References

Brown trouta

Fisha + Stoffels et al., 1992a

Rainbow trout + Starliper et al., 1992

Saithe A + + Strøm and Ringø (upd)

Turbot J + Ringø (upd)

Aerococcus spp.

Atlantic salmon + Westerdahl et al., 1998

Enterococcus spp.


Common carp + Cai et al., 1999

Enterococcus

faecium


Lactococcus spp.



Lac. garvieae


Lac. lactis

Rotiferb Harzevili et al., 1998

Leuconostoc spp.

Arctic charr A + Ringø and Strøm, 1994

Leu. mesenteroides-like

Arctic charr A + + + Ringø et al., 1998

Pediococcus acidilactici


Streptococcus spp.

Arctic charr A + + Ringø, 1993b

A + + Ringø, 1993c

A + Ringø, 1994

A + + Ringø and Strøm, 1994

A + + + Ringø et al., 1998

A + + Ringø and Olsen, 1999

Atlantic salmon A + + Ringø et al., 2000

Carp + Sugita et al., 1985

Eel, European + Esteve and Garay, 1991

Eel, Japanese intestinal content Sugita et al., 1996

Goldfish + Sugita et al., 1988

Rainbow trout intestinal content Sugita et al., 1996

Various salmonids + + + Trust and Sparrow, 1974

Turbot J + + + Ringø (upd)

Yellowtail + Takemaru and Kusuda,

1988a,b

Vagococcus spp.


a No further information was given.b Isolated from whole animal.A, adult; F, fry; J, juvenile; L, larvae; upd, unpublished data.


a marginal part of the intestinal microbiota of fish, it can be questioned if bacte-riocins produced by LAB can effectively contribute to the health status of aquatic animals. However, the main reason for not isolating LAB from aquatic animalsmight be that these LAB from fish are generally slow-growing microorganisms, andRingø and Gatesoupe (1998) recommended an incubation time of up to 4 weeks atlow temperature (4 and 12°C). In addition, as for the human GIT, it can also beexpected that a number of LAB representatives are not yet culturable by existingmethods.

The ability of LAB to produce antibacterial substances has historically long beenused to preserve foods. Since the days of Metchnikoff, efforts have been made toimprove the normal microbiota of the intestine of endothermic animals by LAB

Table 1b Lactic acid bacteria isolated from cold-smoked and fermented fish

Sources Lactic acid bacteria References

Carnobacteriavacuum-packed cold-smoked salmon C. piscicola Leroi et al., 1998spoiled cold-smoked salmon Carnobacterium spp. Hansen and Huss, 1998

Lactobacillichilled, stretch-wrap-packed channel Lb. plantarum Fricourt et al., 1994

catfishspoiled cold-smoked salmon Lb. curvatus, Hansen and Huss, 1998

Lb. sakeiLb. plantarum

fermented fish Lb. pentosus, Tanasupawat et al., 1998low-salt fermented fish products Lb. plantarumspoiled, vacuum-packaged, Lb. pentosus, Paludan-Müller et al., 1999

cold-smoked rainbow trout Lb. plantarumLb. plantarum Lyhs et al., 1999

Jeot-gal, a Korean fermented fish Lb. brevis Lee et al., 2000food fermented fish Lb. acidipiscis spp. nov Tanasupawat et al., 2000

Lactococcilow-salt fermented fish products Lac. lactis subsp. lactis Paludan-Müller et al., 1999Jeot-gal, a Korean fermented fish food Lac. lactis Lee et al., 2000

Leuconostocspoiled cold-smoked salmon Leuconostoc spp. Hansen and Huss, 1998low-salt fermented fish products Leu. citreum Paludan-Müller et al., 1999spoiled, vacuum-packaged, Leu. citreum Lyhs et al., 1999

cold-smoked rainbow trout Leu. mesenteroidessubsp. mesenteroides

Pediococcilow-salt fermented fish products P. pentosaceus Paludan-Müller et al., 1999


(Tannock, 1995, 1999; Fuller, 1997; Naidu et al., 1999; Sanders, 1999). During thepast two decades, numerous experiments have evaluated the ability of LAB (lacto-bacilli, carnobacteria, enterococci, lactococci and streptococci) isolated from sev-eral fish species and live food (Brachionus plicatilis) to inhibit growth of obligatepathogenic bacteria, and antagonism seems to be common among LAB. On thebasis of these results, it is suggested that LAB along with other bacteria that belongto the indigenous microbiota of aquatic animals are an important part of the defencemechanism against fish pathogens. In addition to the antagonistic microorganismscolonizing the mucus surface as part of the natural microbial defence mechanisms,it has been shown that the surface mucus also plays a role in the prevention of colonization by parasites, bacteria and fungi. Readers with special interest in theantiviral activity of gastrointestinal mucus and antibacterial activity of epidermalmucus of fish are referred to the papers of Harrell et al. (1976), Austin and McIntosh(1988), Fouz et al. (1990), Takashashi et al. (1992), Magarinos et al. (1995), Cain et al. (1996), Lemaitre et al. (1996), Romalde et al. (1996), Cole et al. (1997),Robinette et al. (1998) and Ebran et al. (1999, 2000). Moreover, it is generallyaccepted that in adult fish local mucosal and secretory immunity is important in pro-tection against bacterial infections (Trust, 1986; Hart et al., 1987, 1988).

A review of the literature shows that there are several papers that report the anti-bacterial abilities of LAB isolated from fish and other aquatic animals, their poten-tial to adhere to the GIT and how LAB administration affects the GIT microbiota.The purpose of this article is to review the current knowledge on this subject and topresent data from challenge tests in vivo, where antagonistic LAB are included inthe feed and to summarize information on purification and characterization of theantibacterial compounds responsible for the inhibitory effect.

2. DEFINITION OF ANTIMICROBIAL COMPOUNDS

Antimicrobial substances of eukaryotes and prokaryotes are produced and functionin entirely different modes and settings (Nissen-Meyer and Nes, 1997). However,apparently diverse biological systems often have many elements in common at amolecular level.

2.1. Bacteriocins

Many bacteria produce proteinaceous agents that inhibit or kill closely relatedspecies or even different strains of the same species. These agents are called bacte-riocins to distinguish them from antibiotics, which generally have a wider spectrumof activity (Madigan et al., 1997) and a different mode of action. Bacteriocins arepeptides or protein complexes with bactericidal activity that are ribosomally syn-thesized in contrast to antibiotics that are synthesized by different mechanisms. Thestructural gene for the bacteriocin and the genes encoding the proteins involved with


processing of the bacteriocin are often carried out by plasmids or transposons. A general feature of most bacteriocins of LAB is the presence of an own dedicatedimmunity protein whose gene is located physically next to the bacteriocin’s struc-tural gene. Bacteriocins are named in accordance with the species of organism thatproduce them. According to the definition of Tagg et al. (1976), the general criteriacharacterizing these molecules are a) a narrow activity spectrum, b) the proteina-ceous nature of the molecule, c) a bactericidal mode of action, and d) receptors onthe cell surface necessary for attachment. Nevertheless, the inhibitory spectrum ofmany bacteriocins produced by LAB includes various more distantly related Gram-positive bacteria such as strains of Listeria monocytogenes, Bacillus cereus andStaphylococcus aureus (Klaenhammer, 1988; Stevens et al., 1991; Vandenbergh,1993; Ouwehand, 1998).

2.2. Antibiotics

Antibiotics are chemical substances produced by certain microorganisms that in smallquantities inhibit or kill other microorganisms (Madigan et al., 1997). They constitutea special class of chemotherapeutic agents, distinguished from growth factor ana-logues because they are natural products often resulting from secondary microbialmetabolism. They are an important class of pharmaceutical substances produced bylarge-scale industrial processes. For instance, many bacilli produce antibiotics, andproduction seems to be related to the sporulation process (Madigan et al., 1997).

Furthermore, it is important to remember that antagonism may be mediated notonly by antibiotics, but also by many other inhibitory substances such as organicacids, hydrogen peroxide (reviewed by Lindgren and Dobrogosz, 1990; Ringø andGatesoupe, 1998) and siderophores (Gram and Melichiorsen, 1996).

3. DETERMINATION OF GROWTH INHIBITION

It is generally believed that the intestinal mucosal barrier, including the epithelialcells, tight junctions controlling paracellular pathways, and a superficial mucouslayer form an effective physical barrier which separates the individual from thecomplex microbial populations that constitute the normal intestinal microbiota(Mims et al., 1995). The factors considered important in mammals in influencingthe gut microflora, such as gastric acid, bile salts, and proteolytic enzymes, will bepresent at various stages of development of fish but little is known on their role inthe natural defence system of fish against invading pathogens.

The presence of bacteria associated with mucus capable of inhibiting the growthof pathogenic bacteria may probably constitute a barrier against colonization andproliferation by pathogens. Several techniques have been proposed as in vitro antag-onism tests such as: 1) the disc diffusion method, 2) the diametric-streak technique,3) the double-layer method, 4) liquid growth medium and 5) the microtitre plate assay.

As the double-layer method and the microtitre plate assay are most frequently usedin the determination of antagonism of LAB, a brief description is presented of thesetwo techniques.

3.1. Double-layer method

Inhibition of a given microorganism can easily be made visible by using a double-layer method as described by Dopazo et al. (1988), and during the past decade thismethod or its modifications have been used in numerous investigations (Westerdahlet al., 1991, 1998; Bergh, 1995; Sugita et al., 1996, 1997, 1998; Jöborn et al., 1999;Ringø, 1999a). Briefly, the method of Dopazo et al. (1988) is as follows: cultures of the producer strains are spotted on the surface of appropriate agar plates. Afterincubation at 20°C for 4 days, the producer cells are killed with chloroform vapour(15 min) and an overlay containing, e.g. the pathogenic strain is poured on the plate.After a diffusion period of 15–30 min at 20°C, plates are incubated for 2−4 days atthe optimum temperature for each species. After incubation, a clear zone of inhibi-tion around the growth of the producer strain indicates antibacterial activity. Readerswith special interest in the modified methods are referred to the investigations presented above.

3.2. Microtitre plate assay

An alternative to the double-layer method is to pre-culture the test bacteria in aflask, withdraw samples, centrifuge the cells, filter-sterilize the supernatant and testpossible inhibition of the extracellular extracts on the growth of, e.g. a fish pathogen.It is recommended to add 50 μl of the sterile supernatant to 150 μl fresh medium inmicrotitre wells and inoculate 10 μl of a dilution of the fish pathogen at CFU valuesof approximately 106 per ml. The use of 50 μl of the sterile supernatant to 150 μlfresh medium is recommended as some LAB are strong acid producers, and as 100 μlof the sterile supernatant to 100 μl fresh medium may give false positive results.Alternatively, the pH of the culture supernatant may be adjusted to a neutral pHbefore using in the microplate assay. In controls, the indicator strain (e.g. the fishpathogen) is inoculated in 200 μl fresh sterile medium. Growth is estimated as opti-cal density at 600 nm (OD600) using a microplate reader. To obtain correct results,triplicates must be used, and e.g strong inhibition is only obtained when completeinhibition is recovered in all triplicates. As the microtitre plate assay seems to bemore convenient and not as time consuming as the double-layer method severalrecent studies have used the microtitre plate assay (Stoffels et al., 1992a,b; Byun et al., 1997; Jöborn et al., 1997, 1999; Gildberg and Mikkelsen, 1998; Gram et al.,1999; Ringø et al., 2000; Ringø and Holzapfel, 2000). Moreover, this assay can beused to quantify the bacteriocin activity by using two-fold dilutions of the super-natant (Holo et al., 1991).



4. ANTAGONISM OF LACTIC ACID BACTERIA FROM AQUATIC ANIMALS

The antimicrobial effect of LAB has been the basis of food preservation by fermen-tation throughout the history of mankind, and the ability of LAB to produce proteinaceous antagonistic substances as one of the mechanisms of antimicrobial activ-ity is well documented (for review see Fernandes et al., 1987, 1988; Klaenhammer,1988; Piard and Desmazeaud, 1992; Parente and Ricciardi, 1994, 1999; Paik and Oh,1996; Ennahar et al., 2000). Since the late 1920s and early 1930s, when the discoveryof nisin initiated the investigation of proteinaceous antimicrobial compounds fromLAB, a large number of chemically diverse bacteriocins have been identified and characterized, particularly in recent years (for review see Piard and Desmazeaud, 1992;Nettles and Barefoot, 1993; Paik and Oh, 1996; Parente and Ricciardi, 1999; Ennaharet al., 2000).

In addition to bacteriocins, LAB can produce other compounds that inhibitgrowth of microorganisms, e.g. H2O2 and organic acids such as lactic acid con-comitantly lowering pH. According to Piard and Desmazeaud (1992), bacteriocinsof most LAB are active against the lactic acid flora itself. On the other hand, moststudies on antagonistic activity of LAB isolated from fish have concentrated onthe capability of LAB to inhibit undesirable microorganisms (table 2). In fishstudies carried out on antagonistic effects of LAB, the Gram-negative species Vibrioanguillarum and Aeromonas salmonicida were most frequently used, as these fishpathogens have caused the heaviest economical losses in the Atlantic salmon (Salmosalar L.) and turbot (Scophthalmus maximus L.) industry in western Europe.Currently, the prophylactic and therapeutic control of, e.g. vibriosis, comprises thefeeding of antibiotics, but an alternative treatment could be the administration ofprobiotic bacteria that exert inhibitory effects against V. anguillarum, and supportthe natural host microbial defence mechanisms. In addition to the fish pathogens astarget organisms, L. monocytogenes is also frequently used.

4.1. Lactobacillus

During the past two decades, five investigations reported on the antibacterial effectsof lactobacilli isolated from the digestive tract of fish. These involved Lactobacillusplantarum UTC 101-11 isolated from saithe (Gadus virens) (Schrøder et al., 1980),Lb. plantarum BF001 isolated from chilled stretch-wrap-packaged channel catfish(Ictalurus punctatus) (Fricourt et al., 1994), Lactobacillus sp. DS-12 isolated fromflounder (Paralichthys olivaceus) (Byun et al., 1997), Lactobacillus-like bacteriaisolated from Atlantic salmon (Salmo salar) (Westerdahl et al., 1998) and twoLactobacillus isolates (8M851 and 8M852) isolated from Atlantic salmon (Ringø,unpublished data) (table 2).

In their early study, Schrøder et al. (1980) demonstrated that Lb. plantarum pro-duced inhibitors against a Vibrio sp. isolated from the gut of saithe, when this strain was

Tabl

e 2.

Ant

agon

istic

act

iviti

es o

f la

ctic

aci

d ba

cter

ia is

olat

ed f

rom

aqu

atic

ani

mal

s an

d fi

sh p

rodu

cts

agai

nst d

iffe

rent

bac

teri

a

Bac

teri

al g

ener

aIs

olat

ed f

rom

Inhi

bito

ry e

ffec

t aga

inst

R

efer

ence

s

Lac

toba

cill

us-l

ike

Atla

ntic

sal

mon

GIT

Vibr

io a

ngui

llar

umH

I 11

360

Wes

terd

ahl e

t al.,

199

8L

b. p

lant

arum

UT

C10

1Sa

ithe

GIT

Vibr

iosp

.Sc

hrød

er e

t al.,

198

0L

b. p

lant

arum

BF0

01C

hann

el c

atfi

shL

acto

baci

llus

;Lac

toco

ccus

; Lis

teri

a; M

icro

cocc

us; L

euco

nost

oc;

Fric

ourt

et a

l., 1

994

Pedi

ococ

cus;

Sta

phyl

ococ

cus;

Stre

ptoc

occu

s; S

alm

onel

la;

Pse

udom

onas

Lac

toba

cill

ussp

. DS-

12Fl

ound

erG

ITA

er. h

ydro

phil

a; E

dwar

dsie

lla

tard

a;Pa

steu

rell

a pi

scid

a;B

yun

et a

l., 1

997

V. a

ngui

llar

umL

acto

baci

llus

sp. 8

M85

1/8M

852

Atla

ntic

sal

mon

Aer

. sal

mon

icid

aA

L 2

020;

V. a

ngui

llar

umR

ingø

(un

publ

ishe

d V.

sal

mon

icid

ada

ta)

Car

noba

cter

ium

sp. s

trai

n K

1A

tlant

ic s

alm

onG

ITV.

ang

uill

arum

HI

1136

0; A

er. s

alm

onic

ida

AT

CC

141

74O

lsso

n, 1

995;

Jöb

orn

et a

l., 1

997,

199

8C

arno

bact

eriu

msp

. str

ain

K1

Atla

ntic

sal

mon

GIT

Aer

. hyd

roph

ila;

Aer

. sal

mon

icid

a; F

lavo

bact

eriu

m p

sych

roph

ilum

Rob

erts

on e

t al.,

200

0A

tlant

ic s

alm

onP

hoto

bact

eriu

m d

amse

lae

subs

p. p

isci

cida

;Str

epto

cocc

us m

ille

riV.

ang

uill

arum

; V. o

rdal

ii

Car

noba

cter

ium

spp.

Atla

ntic

sal

mon

LI

Aer

. sal

mon

icid

a; V

. sal

mon

icid

aR

ingø

et a

l., 2

001b

C. i

nhib

ens

Atla

ntic

sal

mon

LI

V. a

ngui

llar

um; A

er. s

alm

onic

ida

Jöbo

rn e

t al.,

199

9C

. pis

cico

laA

tlant

ic s

alm

onG

ITV.

sal

mon

icid

aSt

røm

, 198

8C

. pis

cico

laU

I49

Fish

aa

6 st

rain

s of

car

noba

cter

ia; 1

9 st

rain

s of

lact

obac

illi

Stof

fels

et a

l., 1

992a

15 s

trai

ns o

f la

ctoc

occi

; 3 s

trai

ns o

f pe

dioc

occi

19 u

nide

ntif

ied

lact

ic a

cid

isol

ates

fro

m f

ish

C. p

isci

cola

UI4

9Fi

sha

aL

acto

baci

llus

lac

tis

LM

G 2

115

Stof

fels

et a

l., 1

992b

C. p

isci

cola

Tro

utG

IT8

stra

ins

of L

iste

ria

incl

udin

g L

.mon

ocyt

ogen

es;

Pile

t et a

l., 1

995

2 st

rain

s of

C. d

iver

gens

; 1 s

trai

n of

C. p

isci

cola

; 2 s

trai

ns o

f la

ctob

acill

i;1

stra

in o

f le

ucon

osto

c; 1

str

ain

of p

edio

cocc

i; 1

stra

in o

f en

tero

cocc

i;C

lost

ridi

um t

yrob

utyr

icum

C. p

isci

cola

V1

Tro

utG

ITL

. mon

ocyt

ogen

esD

uffe

s et

al.,

1999

C. p

isci

cola

9F 2

51A

rctic

cha

rrF

Aer

. sal

mon

icid

aA

L 2

020;

V. a

ngui

llar

um; V

. sal

mon

icid

aR

ingø

(19

99a;

C

. pis

cico

laC

CU

G34

645

unpu

blis

hed

data

)C

. pis

cico

laFi

sha

L. m

onoc

ytog

enes

Duf

fes

et a

l., 2

000

C. p

isci

cola

Arc

tic c

harr

FA

er. s

alm

onic

ida

C. p

isci

cola

Arc

tic c

harr

GA

er. s

alm

onic

ida;

V. a

ngui

llaru

m; V

. sal

mon

icid

aR

ingø

, and

Hol

zapf

el, 2

000

C. p

isci

cola

203

Arc

tic c

harr

LI

Aer

. sal

mon

icid

aA

L 2

020;

Vib

rio

sp. L

FI 5

000;

Vib

rio

sple

ndid

usR

ingø

, (un

publ

ishe

d da

ta)

V. a

ngui

llar

um; V

. sal

mon

icid

a; C

. pis

cico

laC

CU

G34

645

C. p

isci

cola

204

Arc

tic c

harr

LI

Aer

. sal

mon

icid

aA

L 2

020;

Vib

rio

sp. L

FI 5

000;

V. s

plen

didu

sR

ingø

, (un

publ

ishe

d da

ta)

V. a

ngui

llar

um;V

. sal

mon

icid

aC

arno

bact

eriu

m m

obil

eA

rctic

cha

rrL

IA

er. s

alm

onic

ida

AL

202

0; V

ibri

osp

. LFI

500

0R

ingø

, 199

9aC

. div

erge

nsL

ab 0

1A

tlant

ic s

alm

onG

ITVi

brio

spp.

; V. a

ngui

llar

um; V

. sal

mon

icid

a; P

rote

us v

ulga

ris

Strø

m, 1

988

C. d

iver

gens

-lik

eA

tlant

ic s

alm

on,

GIT

Vibr

iosp

p.; V

. ang

uill

arum

; V. s

alm

onic

ida;

Pro

. vul

gari

sSt

røm

, 198

8sa

ithe,

Atla

ntic

cod

C. d

iver

gens

V41

Tro

utG

ITL

. mon

ocyt

ogen

es**

Duf

fes

et a

l., 1

999

C. d

iver

gens

Lab

01

Atla

ntic

sal

mon

GIT

Aer

. sal

mon

icid

aA

L 2

020;

V. s

alm

onic

ida

Rin

gø (

1999

a;

unpu

blis

hed

data

)C

. div

erge

nsW

F220

1W

olff

ish

GIT

Aer

. sal

mon

icid

aA

L 2

020;

V. a

ngui

llar

um; V

. sal

mon

icid

aR

ingø

(un

publ

ishe

d da

ta)

C. d

iver

gens

6251

TA

rctic

cha

rrSI

Aer

. sal

mon

icid

a; V

. ang

uill

arum

; V. v

isco

sus

LFI

500

0R

ingø

et a

l., 2

002b

C. f

undi

tum

Arc

tic c

harr

LI

Aer

. sal

mon

icid

a; V

. ang

uill

arum

; V. s

alm

onic

ida

Rin

gø e

t al.,

200

2aC

arno

bact

eriu

m-l

ike

Atla

ntic

sal

mon

GIT

V. a

ngui

llar

umH

I 11

360

Wes

terd

ahl e

t al.,

199

8A

eroc

occu

s-lik

e (i

sola

te K

1)A

tlant

ic s

alm

onG

ITV.

ang

uill

arum

HI

1136

0; A

er. s

alm

onic

ida;

Aer

. hyd

roph

ila

Wes

terd

ahl e

t al.,

199

8L

acto

cocc

usla

ctis

Bra

chio

nus

plic

atil

isV.

ang

uill

arum

Q19

Har

zevi

li et

al.,

199

8St

rept

ococ

cus

spp.

Japa

nese

eel

,IT

Aer

. cav

iae

AT

CC

154

68; A

er. e

ucre

noph

ilia

AT

CC

233

09Su

gita

et a

l., 1

996

rain

bow

trou

tA

er. j

anda

eiA

TC

C 4

9568

; Aer

. sob

ria

AT

CC

439

79,

Ple

siom

onas

shi

gell

oide

sA

TC

C 1

4029

aN

o fu

rthe

r in

form

atio

n w

as g

iven

.*

Chi

lled

proc

esse

d ch

anne

l cat

fish

; **

vacu

um-p

acke

d co

ld s

mok

ed.

GIT

, gas

troi

ntes

tinal

trac

t; S,

sto

mac

h; S

I, s

mal

l int

estin

e; L

I, la

rge

inte

stin

e; I

T, in

test

inal

con

tent

s; F

, fae

ces;

G, g

ills.

grown in the presence of a culture filtrate of Bacillus thuringiensis. However, noinhibitory effect was observed without the culture filtrate, showing the limitation ofsuch in vitro antagonism tests. Fricourt et al. (1994) demonstrated that Lb. plantarumBF001 produced an antimicrobial substance active against selected strains from thegenera Lactobacillus, Lactococcus, Listeria, Micrococcus, Leuconostoc, Pediococcus,Staphylococcus, Streptococcus, Salmonella and Pseudomonas. The antimicrobialeffect did not result from acidic fermentation products or hydrogen peroxide.Culture extracts showed a bactericidal mode of action, displayed optimal activity atpH 3.5, and retained full activity after 30 min at 100°C (pH 3.5). The molecularweight of the substance was estimated by ultrafiltration to be less than 10 000 daltons. Byun et al. (1997) investigated the antagonistic activity of a Lactobacillustentatively identified as Lactobacillus sp. DS-12 against the fish pathogens Aer. hydrophila, Edwardsiella tarda, Pasteurella piscicida, Pseudomonas fluorescens, Ps. anguilliseptica, Enterococcus faecalis and V. anguillarum. When thetarget organisms were transversely streaked against the lactobacilli grown overnightin MRS agar, the highest antibacterial activity was found against E. tarda and V. anguillarum, moderate activity against Aer. hydrophila and P. piscicida, but noactivity against the Pseudomonas and Ent. faecalis strain. However, when the paperdisc assay with neutralized supernatant was used, no antagonistic activities wereobserved against any of the target organisms. On the basis of their results, theauthors assumed that the antagonistic activity was not due to bacteriocin production,but to the organic acids produced. In a recent study, Westerdahl et al. (1998) usinga double-layer method demonstrated antagonistic activity for strains K2 and K7,characterized as Lactobacillus-like and isolated from Atlantic salmon, against V. anguillarum HI 11360.

Ringø (unpublished data) tested the antagonistic activity of two Lactobacillusisolates from stomach of Atlantic salmon, by the microtitre plate assay. The twostrains were slow growing, and both inhibited growth of Aer. salmonicida subsp.salmonicida, V. anguillarum, V. salmonicida and a pathogenic Carnobacteriumpiscicola strain. The antagonistic activity was considered to be not due to acidproduction, which was low in both exponential and stationary growth phases.

4.2. Carnobacterium

The genus Carnobacterium was described by Collins et al. (1987) as atypical non-aciduric lactobacilli unable to grow on acetate agar at pH 5.6. During the past twodecades, a number of authors have reported antibacterial activities of carnobacteriaisolated from the GIT of fish (Strøm, 1988; Stoffels et al., 1992a,b; Olsson, 1995;Pilet et al., 1995; Jöborn et al., 1997, 1999; Duffes et al., 1999, 2000; Ringø, 1999a;Ringø et al., 2000, 2001a,b; Robertson et al., 2000; Ringø, unpublished data) andvacuum-packed cold-smoked salmon (Duffes et al., 2000; Schillinger, unpublisheddata) (table 2). In most of these studies, the fish pathogenic strains of Vibrio and



Aeromonas and the pathogen L. monocytogenes have been the most abundantly usedtarget organisms.

Strøm (1988) was first to report antagonism of intestinal carnobacteria against fishpathogens. In this study, she proved that the Atlantic salmon bacterium Lb. plantarumLab01, later reclassified as Carnobacterium divergens Lab01 (Ringø et al., 2001a),and eight other strains isolated from the GIT of adult Atlantic salmon, wild-capturedsaithe and wild-captured Atlantic cod (Gadus morhua L.) reared in seawater and characterized as C. divergens-like, inhibited growth of Vibrio spp., V. anguillarum, V. salmonicida and Proteus vulgaris in in vitro culture experiments. Furthermore, shereported that two strains of C. piscicola isolated from Atlantic salmon parr fed a commercial dry pellet, possessed antagonistic activity to V. salmonicida. Later,Stoffels et al. (1992a) investigated the properties of a bacteriocin-producing C. piscicola isolated from fish against six strains of carnobacteria, 37 strains of lacto-bacilli, 21 strains of lactococci, 12 strains of pediococci and 19 unidentified LAB isolates from fish. Of these 95 strains tested, the six strains of carnobacteria, 19 lacto-bacilli, 15 lactococci, three pediococci and all 19 unidentified LAB isolates from fishwere sensitive to carnocin UI49.

In two recent studies, Olsson (1995) and Jöborn et al. (1997) demonstrated thatCarnobacterium K1, later identified as Carnobacterium inhibens (Jöborn et al.,1999), isolated from the gastrointestinal tract of Atlantic salmon, produced non-characterized inhibitory substances against the two common fish pathogens V. anguillarum and Aer. salmonicida. The antagonistic activity was demonstrated in vitro both in mucus and faecal extracts and in laboratory media. Furthermore, thestrain was able to adhere to intestinal mucus, and to survive and proliferate in the gas-trointestinal tract of fish in vivo, indicating a probiotic potential (Jöborn et al., 1997).

Pilet et al. (1995) showed activity of two bacteriocins produced by C. piscicolaand C. divergens isolated from fish which were active against eight strains ofListeria including five strains of L. monocytogenes. However, the bacteriocins piscicocin V1 and divercin V41 differed in their spectra of activity against otherLAB as piscicocin VI inhibited growth of two strains of C. divergens, one strain ofC. piscicola, two strains of lactobacilli, one strain of Leuconostoc and one strain ofpediococci, while divercin V41 only inhibited growth of one strain of C. divergensand two strains of C. piscicola (table 2). MRS broth was chosen for production stud-ies because its higher carbohydrate level provided better growth and better bacterio-cin production than Elliker broth. The inhibitory substances were produced in theearly exponential growth phase and maximum production of the bacteriocinsoccurred at the beginning of the stationary growth phase and was also higher whenthe producer strain was grown at 20°C than at 30°C (Pilet et al., 1995). In contrastwith piscicocin V1, maximum activity of carnocin UI49 was observed at 34°C withcompletely abolished activity at 15°C (Stoffels et al., 1992a).

In view of the finding by Ringø et al. (1998) that dietary fatty acids affect thepopulation level of LAB associated with the GIT, these intestinal LAB were


screened for the presence of antibacterial activity (Ringø, unpublished data). Of the153 strains of C. piscicola examined, 121 strains were able to inhibit growth of thepathogen Aer. salmonicida subsp. salmonicida in an agar diffusion assay, indicatinga high frequency of antagonists. However, it was interesting to note that the abilityof C. piscicola to inhibit Aer. salmonicida subsp. salmonicida strain LFI 4038 washighest (19 out of 19) in isolates isolated from fish fed 4% dietary α-linolenic acid(18:3 n-3) and highly unsaturated fatty acids (HUFA) (25 out of 27), compared tofish fed linoleic acid (18:2 n-6) where only 11 out of 21 inhibited the pathogen.On the basis of these results it is recommended that greater attention should be givento the subject of how to increase the level of intestinal carnobacteria with inhibitoryeffect against fish pathogens by dietary manipulation. The results obtained from fishfed dietary 18:3 (n-3) may lead to the conclusion that it is desirable to increase levelof dietary 18:3 (n-3) in commercial diets in order to obtain a higher population levelof intestinal strains of C. piscicola able to inhibit growth of Aer. salmonicida.However, in this respect it is worthwhile to note that feeding the charr high levels(>15%) of dietary 18:3 (n-3) increased accumulation of lipid droplets in the entero-cytes and cell damage which may increase the risk of microbial infections (Olsen et al., 1999, 2000).

Under conditions of artificial rearing of teleost fish, it is not uncommon that fishmay be subjected to situations of excessive stress. It is well known that teleosts placedunder stress show a set of physiological reactions, and stress (excessive handling orcrowding) is assumed to be one of several factors contributing to the increased susceptibility to infectious diseases in the fish farming industry (Wedemeyer, 1970;Snieszko, 1974; Mazeaud et al., 1977; Peters et al., 1988; Espelid, 1991). The possibleinvolvement of the GIT as a possible route of infections has mostly been overlooked,although several investigations have suggested the GIT as an infection route ofpathogens (Chart and Munn, 1980; Campbell and Buswell, 1983; Horne and Baxendale,1983; Muroga et al., 1987; Chair et al., 1994; Olsson, 1995; Grisez et al., 1996; Olssonet al., 1996; Romalde et al., 1996; Jöborn et al., 1997; Lødemel et al., 2001). It is wellknown that chronic stress alters the intestinal microbiota of endothermic animals(Tannock and Savage, 1974; Tannock, 1983). The same seems to apply to fish (Leseland Sechet, 1982; Ringø et al., 1997) although reports are scarce. However, in arecent work, it was demonstrated that population levels of C. piscicola-like isolatesin the GIT of Atlantic salmon reared in seawater were affected neither by starvationnor by daily handling stress (Ringø et al., 2000). Out of the 196 C. piscicola isolates,139 showed the ability to inhibit growth of Aer. salmonicida subsp. salmonicida LFI4038 (Ringø et al., 2000), while Vibrio viscocus LFI 5000, the causative agent ofwinter ulcers, was only inhibited by 21 C. piscicola-like isolates using the double-layer method (Ringø, unpublished data). As the frequency of C. piscicola-like isolatesable to inhibit Aer. salmonicida subsp. salmonicida LFI 4038 (Ringø et al., 2000) andV. viscocus LFI 5000 (Ringø, unpublished data) decreased during the experiment(11 days of regular handling stress), it was suggested that the antagonistic strains are


more loosely associated to the epithelial mucosa of the GIT than other C. piscicola-like strains. This controversial hypothesis implies that the loss of these beneficial C. piscicola-like strains may lead to colonization of pathogens if present and if thestress occurs over several days. This hypothesis is controversial and calls for furtherinvestigations.

In a recent study, Ringø and Holzapfel (2000) evaluated the ability of 26 isolatesof carnobacteria from gills of Atlantic salmon to inhibit growth of Aer. salmonicida,V. anguillarum and V. salmonicida using the microtitre plate method. Nine strainscharacterized as C. piscicola by 16S rDNA and amplified fragment length poly-morphism (AFLP™), produced an antagonistic compound active against all threefish pathogens. The nature of the inhibitory compound was not further investigated.As the gills may be an infection route of pathogenic bacteria, these results illustratethat carnobacteria associated with the gills may play a role in the disease defence offish through contact with infected fish or contaminated water (Evelyn, 1984; BaudinLaurencin and Germon, 1987; Hjeltnes et al., 1987; Svendsen et al., 1999).

Duffes et al. (2000) recently showed that C. piscicola SF668 isolated fromNorwegian cold-smoked salmon was able to produce bacteriocins with a strongantilisterial activity and to inhibit L. monocytogenes growth in a simulated cold-smoked fish system at 4°C.

4.3. Aerococcus

Antagonistic activity of five Aerococcus-like isolates from the GIT of Atlanticsalmon against fish pathogens has been reported in one recent study (Westerdahl et al., 1998). One of these isolates (strain K1) inhibited growth of V. anguillarum HI11360, Aer. salmonicida and Aer. hydrophila in liquid media (table 2). The authorspresent further information on growth inhibition of V. anguillarum HI 11360 in liquid media supplemented with 50% sterile supernatant from strain K1. However,as growth of V. anguillarum HI 11360 depended on the pH of the media, it may beconcluded that the inhibitory effect not only was the result of antagonist productionbut may also be related to acid production by strain K1.

4.4. Lactococcus

Lactococcus strains are rarely isolated from aquatic animals. Yet in a recentstudy, Harzevili et al. (1998) isolated a Lactococcus lactis strain AR21 from therotifer Brachionus plicatilis (Müller) which exhibited an inhibitory effect againstV. anguillarum Q19 (table 2).

4.5. Streptococcus

During the past decade, several authors have reported on strains of the genusStreptococcus as indigenous to the intestine of fish (for review see Ringø and


Gatesoupe, 1998), but less research has been done on their antibacterial activity.In a recent study, Sugita and co-authors (1996) isolated 12 strains of Streptococcus spp.from intestinal contents of Japanese eel (Anguilla japonica) (11 isolates) and rainbowtrout (Oncorhynchus mykiss) (one isolate). Of these 12 strains, three showed antag-onistic activity against Aer. caviae, two against Aer. eucrenophilia, one againstAer. jandaei and five against Aer. sobria, while Plesiomonas shigelloides was sensitiveto two Streptococcus spp. (table 2).

4.6. Aggregation with pathogens

Aggregation may be an interesting feature of strains for probiotics, as the aggregationby LAB may expel pathogens not adhering to the mucus or surface of enterocytesfrom the gut. Spencer and Chesson (1994) reported that five strains out of 43 lacto-bacilli were able to aggregate E. coli 0129-K88+. Similarily, Kmet and Lucchini(1999) isolated 20 Lactobacillus strains from the oesophagus and vagina of farrowingsows, and aggregation activity was observed between six homofermentative auto-aggregative strains and three strains of pathogenic E. coli with F4, F5 and F6 fimbriae.At present, no information on this topic is available on probiotic candidates isolatedfrom fish, and this topic deserves further study.

4.7. Purification and characterization of inhibitory substances from LAB

Information on purification and characterization of bacteriocins produced by LABassociated with food systems is provided in numerous reports (for review see Piard andDesmazeaud, 1992; Nettles and Barefoot, 1993; Paik and Oh, 1996; Parente andRicciardi, 1999; Ennahar et al., 2000). Yet, there is a paucity of such information onbacteriocins of LAB from aquatic animals. Knowledge on the identity of inhibitorysubstances produced by intestinal LAB is crucial for assessing the potential of differentisolates and their significance as antagonists in the digestive tract of fish. Informationon bacteriocin purification and characterization is only available for Carnobacteriumstrains from aquatic animals; these results are summarized in table 3. Still, knowledgeof bacteriocins produced by this relatively new group of bacteria is limited.

Stoffels et al. (1992a,b, 1993) purified a bacteriocin designated carnocin UI49from C. piscicola UI49 using a purification protocol including XAD chromatogra-phy and cation exchange chromatography. It is a small peptide molecule containinglanthionine and therefore belongs to the class of bacteriocins termed lantibiotics.It was produced during the mid-exponential phase of growth at temperatures between15 and 34°C. As known for other bacteriocins, its bactericidal mode of action resultsin the lysis of sensitive cells of closely related LAB.

The bacteriocins named piscicocin V1 and divercin V41 were primarily charac-terized in a study of Pilet et al. (1995). They were shown to be heat resistant for upto 30 min at 100°C, but the activity was completely lost after autoclaving for 15 min

Tabl

e 3.

Puri

fica

tion

and

char

acte

riza

tion

of b

acte

rioc

ins

from

car

noba

cter

ia is

olat

ed f

rom

fis

h

Num

ber

Hea

t of

Am

ino

Prod

ucer

stab

ility

pHSe

nsiti

vity

toPu

rifi

catio

nM

olec

ular

amin

oac

idB

acte

rioc

inst

rain

Med

ium

at 1

00°C

stab

ility

prot

ease

ssc

hem

em

ass

(Da)

acid

sse

quen

ceR

efer

ence

s

Car

noci

n C

. pis

cico

laM

RS

Part

ial

2–10

Try

psin

: +A

SP, d

esal

t on

4635

35–3

7N

DSt

offe

ls e

t al.,

199

2aU

I49

UI4

9in

activ

atio

nα-

Chy

mot

ryps

in:+

CFC

, CE

C, X

AD

Stof

fels

et a

l., 1

993

afte

r 60

min

Peps

in: ±

chro

mat

ogra

phy,

CE

C (

larg

e sc

ale)

Pisc

icoc

in

C. p

isci

cola

MR

SN

o in

activ

a-N

DT

ryps

in: +

ASP

, des

alt w

ith44

16 (

V1a

)44

pisc

icol

inPi

let e

t al.,

199

5V

1V

1tio

n af

ter

Prot

eina

se K

:+Se

p-Pa

ck, H

PLC

4526

(V

1b)

4312

6*30

min

Pron

ase

E: +

carn

obac

-B

huga

loo-

Via

l te

rioc

inet

al.,

199

6B

M1*

Div

erci

n C

. div

erge

nsM

RS

No

inac

tiva-

ND

Try

psin

: +A

SP, d

esal

t with

4509

43si

mila

r to

Pile

t et a

l., 1

995

V41

V41

tion

afte

rPr

otei

nase

K:+

Sep-

Pack

, HPL

Cen

tero

cin

Met

ivie

r et

al.,

199

830

min

Pron

ase

E: +

A a

nd

pedi

ocin

PA

1

*The

am

ino

acid

seq

uenc

es a

re id

entic

al w

ith th

ose

of p

isci

colin

126

fro

m C

. pis

cico

laJG

126

isol

ated

fro

m h

am (

Jack

et a

l., 1

996)

and

ca

rnob

acte

rioc

in B

M1

from

C. p

isci

cola

LV17

B f

rom

mea

t (Q

uadr

i et a

l., 1

994)

, res

pect

ivel

y.A

SP, a

mm

oniu

m s

ulph

ate

prec

ipita

tion;

CFC

, gel

filt

ratio

n ch

rom

atog

raph

y; C

EC

, cat

ion

exch

ange

chr

omat

ogra

phy;

XA

D, a

dsor

ptio

n m

edia

, pol

yaro

mat

ic;

HPL

C, h

igh-

perf

orm

ance

liqu

id c

hrom

atog

raph

y.

ND

, no

data

ava

ilabl

e.


at 120°C. They exhibited an antilisterial activity and were sensitive to various pro-teolytic enzymes (pronase E, proteinase K and trypsin) suggesting their proteina-ceous nature. A later study (Bhugaloo-Vial et al., 1996) revealed that theantibacterial activity of piscicocin V1 was due to two different bacteriocins (piscico-cin V1a and piscicocin V1b) showing identical inhibitory spectra. Both substanceswere purified to homogeneity and were found to be non-lantibiotic small heat-stable bacteriocins with molecular masses of 4416 dalton and 4526 dalton. Thepurification and determination of the amino acid sequence of divercin V41 revealedthat this bacteriocin shows a high homology with pediocin PA-1 and enterocin Aproduced by strains of Pediococcus and Enterococcus (Metivier et al., 1998).

Jöborn (1998) described inhibitory substances greater than 1000 dalton from LABstrain K3 isolated from Atlantic salmon and probably a representative of C. piscicola(Westerdahl et al., 1998). By contrast, other strains of LAB including Lactobacillus-like, Aerococcus-like and Carnobacterium-like isolates produced inhibitory substances less than 1000 dalton when tested in an in vitro assay (Jöborn, 1998).

5. USE OF LACTIC ACID BACTERIA IN AQUACULTURE

Specific bacterial pathogens can be an important cause of mortalities in fish hatch-eries, as intensive husbandry practices often result in breakdown of the natural hostbarriers. Research laboratories and commercial hatcheries have attempted to over-come this problem by disinfection of water supplies and food, stimulation of hostresistance, and the prophylactic or therapeutic use of antibiotics. However, the indis-criminate use of antibiotics in disease control in many sections of the aquacultureindustry has led to selective pressure of antibiotic resistance in bacteria, a propertywhich may be readily transferred to other bacteria (Aoki et al., 1985; Amàbile-Cuevas et al., 1995; Towner, 1995; Sørum, 1999). An alternative approach by whichopportunistic infections of fish pathogens may be reduced is by manipulation of thegut flora either by adding antagonistic bacteria to the diet or by dietary manipulation,in order to increase the proportion of health-promoting bacteria in the gut microflora.An advantage of these methods is that they can be implemented during the earlystages of development when vaccination by injection is impractical. In this regard,the stability of the antagonistic feature is a very important trait of probiotic LAB.According to Olsson (1995) several turbot (Scophthalmus maximus L.) and Atlanticsalmon LAB isolates lost their capacity to inhibit growth of V. anguillarum afterbeing subcultured a limited number of times and stored at −70°C. Similar observa-tions were made by Westerdahl et al. (1998) who described that antagonistic activityof several fish intestinal bacteria was rapidly lost after storage and subculturing.

5.1. Effects of LAB administration on intestinal microbiota

Several studies on endothermic animals have demonstrated that administration ofLAB affects intestinal microbiota (for reviews see Gorbach, 1990; Juven et al., 1991;

Chateau et al., 1993; Naidu et al., 1999). However, a number of factors determine thetype and extent of colonization and continued presence of bacteria within the digestivetract. These factors have been extensively reviewed by Savage (1983) and include:1) gastric acidity (Gilliland, 1979), 2) bile salts (Floch et al., 1972), 3) peristalsis,4) digestive enzymes (Marmur, 1961), 5) immune response, and 6) indigenous(autochthonous) microorganisms and their antibacterial compounds. Attempts to controlthe intestinal microbiota by LAB administration in aquatic animals are rather scarce.

It is well known that LAB under normal circumstances are not numerically dom-inant in the digestive tract of fish (Ringø and Gatesoupe, 1998). In order to increasethe proportion of LAB, some investigations have attempted to increase their popu-lation level by dietary factors such as: 1) chromic oxide (Ringø, 1993b,c), 2) differ-ent oils (Ringø et al., 1998, 2002a), 3) high and low dietary lipids (Ringø and Olsen,1999), and 4) inulin (Ringø et al., unpublished results). Another important criterionfor the use of LAB in commercial aquaculture is the colonization potential of LABin the fish gut, as Vibrionaceae may also persist for days or weeks in fish (Austin et al., 1995; Munro et al., 1995; Ringø and Vadstein, 1998). Some recent studieshave demonstrated that carnobacteria strains are able to survive for several days inthe intestine of larval and juvenile fish (Strøm and Ringø, 1993; Jöborn et al., 1997;Gildberg and Mikkelsen, 1998; Ringø, 1999b). Three of these studies (Jöborn et al.,1997; Gildberg and Mikkelsen, 1998; Ringø, 1999b) have suggested that there isapparently no host specificity with regard to colonization of the fish gut withcarnobacteria, in contrast to endothermic animals where adhesion of LAB appearsto be complicated by host specificity (Lin and Savage, 1984; Fuller, 1986; Conway,1989). However, the colonization site in the fish gut is also an important criterion.In a recent study, Gildberg and Mikkelsen (1998) administered two C. divergensstrains originally isolated from the intestine of mature Atlantic cod and Atlanticsalmon, to Atlantic cod juveniles via the food. When the Atlantic cod isolate wasused, the authors only detected LAB in pyloric caeca, while the concentration of thebacteria was approximately ten-fold higher in the pyloric caeca than in the intestinewhen the salmon isolate was used.

Transient bacteria may also be efficient if the cells are introduced at high dose.Moreover, as LAB may exert antibacterial effects against undesirable microbes,some investigators have attempted to increase the proportion of LAB associatedwith the fish digestive tract. In a study with 4-day-old Atlantic cod larvae, Strøm andRingø (1993) used an antagonistic LAB strain which, when added to the rearingwater, favourably influenced the intestinal microbiota of the larvae by increasing theproportion of LAB from approximately 5 to 70% and by a subsequent decrease inthe proportion of the bacteria genera Pseudomonas, Cytophaga/Flexibacter andAeromonas (table 4). These results indicate that the LAB were able to colonize andmay comprise a major part of the autochthonous microbiota in the gut of the larvae.A similar increase in intestinal LAB was also found in Atlantic cod fry fed a dietcontaining C. divergens (Gildberg et al., 1997) (table 4). In a study with Atlantic


Tabl

e 4.

Eff

ect o

f L

AB

adm

inis

trat

ion

on in

test

inal

mic

robi

ota

LA

BB

acte

rial

gen

era

isol

ated

and

pro

port

ion

of m

icro

flor

a po

pula

tion

Fish

spe

cies

used

Bef

ore

adm

inis

trat

ion

(con

trol

)A

fter

adm

inis

trat

ion

Aft

er c

halle

nge

Ref

eren

ces

Atla

ntic

cod

–C

. div

erge

nsP

seud

omon

as42

.5; C

ytop

haga

/Fle

xiba

cter

42.5

C. d

iver

gens

70;

Pse

udom

onas

20*

Strø

m a

nd R

ingø

,la

rvae

Aer

omon

as 1

0; C

. div

erge

ns 5

1993

Atla

ntic

cod

–C

. div

erge

nsN

o in

form

atio

n w

as g

iven

No

info

rmat

ion

was

giv

enC

. div

erge

ns75

Gild

berg

et a

l.,

fry

Pse

udom

onas

-lik

e 25

1997

Atla

ntic

C

. div

erge

nsP

seud

omon

as, E

nter

obac

teri

acea

eC

. div

erge

ns10

0A

er. s

alm

onic

ida

90G

ildbe

rg e

t al.,

sa

lmon

– f

ryG

ram

-pos

itive

coc

ciC

. div

erge

ns10

1995

Tur

bot –

larv

aeC

. div

erge

nsC

. div

erge

nsn.

dC

. div

erge

ns(8

×10

3 )*

Rin

gø, 1

999b

Flou

nder

aL

acto

baci

llus

sp.

Ent

erob

acte

riac

eae

4.3

(5/5

); G

(+)

Ent

erob

acte

riac

eae

*B

yun

et a

l., 1

997

4.6

(5/5

)4.

8 (5

/5);

G(+

) 4.

3 (5

/5)

DS-

12Y

east

4.6

(5/

5); h

aem

olyt

ic b

acte

ria

5.8

(2/5

) L

acto

baci

llus

sp. D

S-12

7.0

(3/

5)M

ucoi

d co

lony

for

m 4

.8 (

1/5)

; aer

obes

C

lost

ridi

um4.

3 (1

/5);

yea

st 4

.3 (

1/5)

8.5

(5/5

)A

naer

obes

7.6

(5/

5)H

aem

olyt

ic b

acte

ria

5.1

(1/5

)A

erob

es 7

.3 (

5/5)

; ana

erob

es 6

.6 (

5/5)

Car

pbE

nt. f

aeci

umE

nter

obac

teri

acea

e 6.

2; E

. col

i4.

2E

nter

obac

teri

acea

e 6.

2; E

. col

in.

d*

Bog

ut e

t al.,

199

8E

nt. f

aeca

lis

3.3;

Sta

ph. a

ureu

s3.

7E

nt. f

aeca

lis

3.5;

Sta

ph. a

ureu

s4.

0B

acil

lus

spp.

7.0

; Clo

stri

dium

spp.

2.9

Bac

illu

ssp

p. 7

.0; C

lost

ridi

umsp

p. 2

.7Sh

eat f

ishc

Ent

. fae

cium

Esc

heri

chia

col

i3.

1; E

nter

obac

teri

acea

e 3.

0E

sche

rich

ia c

oli

1.1;

*B

ogut

et a

l., 2

000

Ent

erob

acte

riac

eae

1.9

Stap

h. a

ureu

s4.

7St

aph.

aur

eus

1.4

Bac

illu

s6.

0; C

lost

ridi

um2.

1B

acil

lus

5.6;

Clo

stri

dium

n.d

aD

ata

are

pres

ente

d as

log

10 a

nd f

requ

ency

is s

how

n in

par

enth

eses

.b

Dat

a ar

e pr

esen

ted

as lo

g 10

aft

er 4

wee

ks o

f fe

edin

g.c

Dat

a ar

e pr

esen

ted

as lo

g 10

aft

er 5

8 da

ys o

f fe

edin

g.n.

d, n

ot d

etec

ted;

*, c

halle

nge

test

not

don

e.


salmon fry, Gildberg et al. (1995) demonstrated that administration of LAB reportedas Lb. plantarum, but later reclassified as C. divergens (Ringø et al., 2001a) increasedthe proportion of adherent LAB to intestinal wall from nil to 100% (table 4).

Recently, Byun et al. (1997) evaluated the effect of LAB (Lactobacillus sp. DS-12)administration via the feed on the intestinal microbiota of flounder (Paralichthys olivaceus) after 1 month of feeding (table 4). Lactobacillus sp. DS-12 was notdetected in the intestine of the control group, but 107/g LAB were found in the GITwhen the fish were fed a LAB supplemented feed.

In a recent study, Bogut et al. (2000) evaluated the effect of Ent. faecium on theintestinal microbiota of Sheat fish (Silurus glanis). In this study, the fish were exposedto Ent. faecium by including the bacteria in the diet. After approximately 2 months offeeding, some interesting differences in the intestinal microbiota were observedbetween the two rearing groups. Ent. faecium administration decreased the populationlevel of Staph. aureus, E. coli and other bacteria of the family Enterobacteriaceae, andresulted in complete elimination of Clostridium spp. (table 4).

Only one investigation has evaluated the influence of a commercial LAB prepa-ration on the allochthonous intestinal microbiota. Supplementation of 1 gram of Ent.faecium M74 per 100 kg feed influenced the intestinal microbiota of 0+ Israeli carp(Cyprinus carpio) to some extent (Bogut et al., 1998). While Escherichia coli disap-peared from the intestinal microbiota of the fish after 14 days and onwards by feed-ing the probiotic preparation (table 4), the population levels of Enterobacteriaceae,Ent. faecalis, Staph. aureus, Bacillus spp. and Clostridium spp. were not reduced asa result of including Ent. faecium into the diet (Bogut et al., 1998). The authors suggested a high adhesive ability in the epithelium of carp digestive tract for Ent. faecium. However, as they isolated the allochthonous intestinal microbiota, convinc-ing experimental evidence was not provided.

When dealing with the potential of probiotics (for example LAB) in aquaculturethe fundamental question arises whether it is possible to colonize and maintain theprobiotic bacteria within the digestive tract. This is particularly important whenlong-term exposure may be required for the probiotic effect. In this respect, electronmicroscope investigations are a useful tool (Ringø et al., 2003). Furthermore, read-ers with special interest in prospects of fish probiotics, are referred to the recentreview on this topic by Gram and Ringø (2005, Chapter 17 in this book).

During the past decade some reports have been published on the nutritional con-tribution of LAB to the production rate of rotifer Brachionus plicatilis (Gatesoupeet al., 1989; Gatesoupe, 1990, 1991a), while the control of the microbiota of rotifercultures has received less attention.

5.2. Effects of LAB administration on survival and growth of fish

Some research has been conducted on the effect of LAB administration on survival(Garcia de la Banda et al., 1992; Hamácková et al., 1992; Gildberg et al., 1995;


Ringø, 1999b; Ottesen and Olafsen, 2000) and growth of fish (Hamácková et al.,1992; Noh et al., 1994; Gildberg et al., 1995; Byun et al., 1997; Bogut et al., 1998)(table 5). In their study with turbot larvae, Garcia de la Banda et al. (1992) claimedthat administration of commercial preparations of live LAB (Lactococcus lactis andLb. bulgaricus) via enrichment of rotifer and Artemia increased survival of the larvae 17 days after hatching. Accelerated growth of Sheat fish after Ent. faeciumM-74 administration was also reported by Hamácková et al. (1992). Ottesen andOlafsen (2000) working with Atlantic halibut (Hippoglossus hippoglossus L.) larvaedemonstrated that Lb. plantarum (originally isolated from Atlantic cod) administra-tion to the culture water improved larval survival. At 12 days post-hatching (the firstcritical stage of initial feeding), larval survival was approximately 96% compared to 81.5% survival in the control group (unsupplemented with bacteria). This posi-tive trend in the LAB rearing group was also observed after 32 days, as survival was significantly higher in the larval group incubated with Lb. plantarum (68.4%)compared to the control group (58.2%). Contrary to these results, Ringø (1999b) didnot observe any positive effect on survival of turbot larvae exposed to C. divergens(originally isolated from Atlantic salmon) compared to larvae not exposed to bacteria(control).

Commercial preparations of Ent. faecium improved the growth and feed efficiencyof Israeli carp (Noh et al., 1994; Bogut et al., 1998) and Sheat fish (Hamácková et al.,1992). Gildberg et al. (1995) included a C. divergens strain (originally isolated fromAtlantic salmon) to the diet, but the authors did not observe increased growth ofAtlantic salmon fry as a result of LAB administration. Byun et al. (1997) checked thefeeding effects of Lactobacillus sp. DS-12 on body weight of flounder (Paralichthysolivaceus) in two feeding trials. In the first trial, a significant effect of LAB adminis-tration was observed. However, the second experiment showed no significant differ-ences between the groups although there was a tendency of greater increase in bodyweight as a result of LAB administration.

No increase in the growth rate of the rotifers was observed after addition of Lac.lactis AR21 through the diet under optimal conditions (Harzevili et al., 1998). Undera suboptimal feeding regime where the food was reduced to 45%, Lac. lactis coun-teracted the growth inhibition of rotifers due to V. anguillarum in two of the threeexperiments performed. However, the authors recovered neither Lac. lactis norV. anguillarum from the rotifer after 24 h.

5.3. Challenges in vivo

The major factors involved in the biocontrol of bacterial pathogens in the gastro-intestinal tract are primarily those regulating the composition, functions and inter-actions of indigenous microbial populations with the animal tissues. This concept issupported by repeated observations that strains of transient enteropathogens can col-onize intestinal habitats of endothermic animals. The fact that fish contain intestinal

Tabl

e 5.

Eff

ect o

f L

AB

adm

inis

trat

ion

on s

urvi

val a

nd g

row

th

LA

B is

olat

e us

edH

ost

Way

of

adm

inis

trat

ion

Eff

ect

Ref

eren

ce

Surv

ival

Lac

. lac

tis

and

Tur

bot l

arva

eE

nric

hmen

t of

rotif

erIn

crea

sed

surv

ival

of

larv

aeG

arci

a de

la B

anda

et a

l.,19

92

Lb.

bul

gari

cus

and

Art

emia

17 d

ays

afte

r ha

tchi

ng

Ent

. fae

cium

M-7

4Sh

eat f

ish

fry

Add

ition

to th

e di

etIn

crea

sed

surv

ival

Ham

ácko

vá e

t al.,

199

2L

b. p

lant

arum

a*A

tlant

ic h

alib

ut la

rvae

Add

ition

to th

e cu

lture

wat

erIn

crea

sed

surv

ival

of

larv

aeO

ttese

n an

d O

lafs

en, 2

000

2 w

eeks

aft

er h

atch

ing

C. d

iver

gens

bT

urbo

t lar

vae

Add

ition

to th

e cu

lture

wat

erN

o si

gnif

ican

t eff

ect o

n la

rval

Rin

gø, 1

999b

surv

ival

Gro

wth

Str.

ther

mop

hilu

s,L

b. h

elve

ticu

sT

urbo

t lar

vae

Enr

ichm

ent o

f ro

tifer

Enh

ance

d gr

owth

Gat

esou

pe, 1

991a

orL

b. p

lant

arum

Ent

. fae

cium

M-7

4Sh

eat f

ish

fry

Add

ition

to th

e di

etE

nhan

ced

grow

thH

amác

ková

et a

l., 1

992

Ent

. fae

cium

Isra

eli c

arp

adul

tA

dditi

on to

the

diet

Enh

ance

d gr

owth

Noh

et a

l., 1

994

Ent

. fae

cium

Isra

eli c

arp

juve

nile

Add

ition

to th

e di

etE

nhan

ced

grow

th a

nd f

eed

Bog

ut e

t al.,

199

8co

nver

sion

Ent

. fae

cium

Shea

t fis

h ju

veni

leA

dditi

on to

the

diet

Enh

ance

d gr

owth

Bog

ut e

t al.,

200

0C

. div

erge

nsb

Atla

ntic

sal

mon

fry

Add

ition

to th

e di

etN

o si

gnif

ican

t eff

ect o

n gr

owth

Gild

berg

et a

l., 1

995

Lac

toba

cill

ussp

. DS

-12

Flou

nder

juve

nile

Add

ition

to th

e di

etV

arie

d re

sults

Byu

n et

al.,

199

7

aIs

olat

ed f

rom

Atla

ntic

cod

by

Strø

m, 1

988.

bIs

olat

ed f

rom

Atla

ntic

sal

mon

by

Strø

m, 1

988.

* R

ecla

ssif

ied

from

Lac

toba

cill

us p

lant

arum

to C

arno

bact

eriu

m d

iver

gens

by R

ingø

et a

l., 2

001a

.


microbiota with antagonistic effects against fish pathogens has prompted investiga-tors to conduct challenge experiments with LAB during the past decade (Gatesoupe,1994; Gildberg et al., 1995, 1997; Gildberg and Mikkelsen, 1998; Harzevili et al.,1998). However, in these studies some conflicting results on the mortality werereported when the control group was compared with probiotic treatment (table 6).

Gatesoupe (1994) suggested that in vivo experiments with turbot larvae usingrotifers grown on LAB strains (resembling those of Lb. plantarum orCarnobacterium sp.) improved the disease resistance in challenge tests with patho-genic vibrio (V. splendidus strain VS11). However, the results reported in this studywere registered after 48 and 72 h, beyond which the mortality pattern was not dis-cussed. In three papers, Gildberg and Mikkelsen (1998) and Gildberg et al. (1995,1997) have used two LAB strains originally isolated from Atlantic salmon andAtlantic cod by Strøm (1988). These two isolates were recently identified by 16SrDNA and AFLPTM fingerprinting as C. divergens (Ringø et al., 2001a). In challengetrials with co-habitants with Aer. salmonicida, Gildberg et al. (1995) in contrast toexpectations, registered the highest mortality of Atlantic salmon fry with fish giventhe diet containing C. divergens, originally isolated from Atlantic salmon intestine.In their study with Atlantic cod fry, Gildberg and Mikkelsen (1998) observed thesame cumulative mortality independent of whether the C. divergens isolates supple-mented to the commercial feed were originally isolated from the digestive tract ofAtlantic cod or Atlantic salmon, when the fish were bath exposed to V. anguillarum.On the other hand, an improved disease resistance of Atlantic cod fry was observedby supplementing a commercial dry feed with a strain of C. divergens originally isolated from the cod (Gildberg et al., 1997). The explanation for these conflictingresults has not been elucidated. Gildberg and Mikkelsen (1998) put forward ahypothesis that bacteriocin production can be inducible and may not occur if thebacteria are not frequently challenged with inhibitors as previously demonstrated bySchrøder et al. (1980). Furthermore, a recent study by Nikoskelainen et al. (2001)used the human probiotic Lb. rhamnosus in a challenge test with Aer. salmonicidawith promising results (table 6). These results should stimulate fish microbiologiststo use human probiotic LAB in future studies.

If the intestine is involved in infection, the fundamental question arises ofwhether there are differences between the posterior part of the intestine and thehindgut region of the intestine. It is well established that the intestine in an imma-ture or inflammatory state has an enhanced capacity to absorb intact macromole-cules (for review see Olsen and Ringø, 1997). Furthermore, some studies reportendocytosis of bacteria by enterocytes in the epithelial border of hindgut of herring(Clupea harengus) larvae (Hansen et al., 1992; Hansen and Olafsen, 1999), herringand Atlantic cod larvae (Olafsen and Hansen, 1992) and 36-day-old juvenile turbot(Grisez et al., 1996). It is generally accepted that mature and non-inflammatoryintestines of adult salmonids are not permeable to microparticulates, in contrast tothe mammalian GIT where M cells are active in phagocytosis. However, a recent

Tabl

e 6.

Cha

lleng

e te

sts

of f

ish

with

LA

B

Eff

ect i

nW

ay o

f ch

alle

nge

Sugg

este

d m

ode

LA

B is

olat

e us

edH

ost

Path

ogen

adm

inis

trat

ion

test

of a

ctio

nR

efer

ence

Car

noba

cter

ium

spp.

aT

urbo

t lar

vae

V. s

plen

didu

sE

nric

hmen

t of

rotif

ers

+A

ntag

onis

m a

nd/o

rG

ates

oupe

, 199

4im

prov

ed n

utri

tiona

lva

lue

of th

e ro

tifer

s

C. d

iver

gens

bA

tlant

ic s

alm

on f

ryA

er. s

alm

onic

ida

Add

ition

to th

e di

et–

Not

spe

cifi

edG

ildbe

rg e

t al.,

199

5C

. div

erge

nsc

Atla

ntic

cod

juve

nile

sV.

ang

uill

arum

Add

ition

to th

e di

et+

Ant

agon

ism

Gild

berg

et a

l., 1

997

C. d

iver

gens

bA

tlant

ic c

od f

ryV.

ang

uill

arum

Add

ition

to th

e di

et+

eG

ildbe

rg a

nd M

ikke

lsen

, 199

8C

. div

erge

nsc

Atla

ntic

cod

fry

V. a

ngui

llar

umA

dditi

on to

the

diet

–G

ildbe

rg a

nd M

ikke

lsen

, 199

8L

b. r

ham

nosu

sd

Rai

nbow

trou

tA

er. s

alm

onic

ida

Add

ition

to th

e di

et+

Nik

oske

lain

en e

t al.,

200

1

+, I

mpr

oved

dis

ease

res

ista

nce;

–, n

o si

gnif

ican

t eff

ect.

aIs

olat

ed f

rom

rot

ifer

.b

Isol

ated

fro

m in

test

ine

of A

tlant

ic s

alm

on (

Strø

m, 1

988)

.c

Isol

ated

fro

m in

test

ine

of A

tlant

ic c

od (

Strø

m, 1

988)

.d

A p

robi

otic

for

hum

an u

se.

e12

day

s af

ter

infe

ctio

n si

gnif

ican

t red

uced

cum

ulat

ive

mor

talit

y w

as r

ecor

ded

in f

ish

give

n fe

ed s

uppl

emen

ted

with

C. d

iver

gens

isol

ated

fro

m A

tlant

ic s

alm

on,

but n

o ef

fect

was

det

ecte

d 4

wee

ks a

fter

infe

ctio

n.


study demonstrated endocytosis of bacteria by enterocytes in the epithelial border ofhindgut of adult salmonid fish (fig. 1a), as well as in the posterior part of the intes-tine (pyloric caeca) (fig. 1b) (Ringø et al., 2001b). These results are in accordancewith observations made by Vigneulle and Laurencin (1991) and Tamura et al. (1993)who measured phagocytosis of fixed V. anguillarum in the posterior intestine ofrainbow trout (Oncorhynchus mykiss), sea bass (Dicentrarchus labrax), turbot(Scophthalmus maximus) and eel (Anguilla anguilla).

The observations of Vigneulle and Laurencin (1991), Tamura et al. (1993) andRingø et al. (2001b, 2002a, 2003) indicate that the intestine is involved in bacterialtranslocation. Yet no clear evidence is available on possible differences between dif-ferent parts of the intestine with regard to bacterial infection.

It is well known that rotifers are often suspected of being a vector for bacterialinfections to the predatory organisms (Muroga et al., 1987; Perez-Benavente andGatesoupe, 1988; Tanasomwang and Muroga, 1988; Nicolas et al., 1989). It is there-fore surprising that studies dealing with the proliferation of larval pathogens inrotifer cultures are so scarce (Gatesoupe, 1991a; Harzevili et al., 1998). Gatesoupe(1991a) reported that the proliferation of Aer. salmonicida that accidentallyappeared in the experimental rotifer culture was inhibited by treatment withLb. plantarum. Harzevili et al. (1998) reported that administration of the probioticstrain Lac. lactis AR21 under a suboptimal feeding regime counteracted the growthinhibition of the rotifers owing to V. anguillarum.

Fig. 1. Endocytosis of bacteria demonstrated in enterocyte in the hindgut region (a) (arrowhead) andbacteria between the microvilli (arrows) (Ringø et al., 2002a), and pyloric caeca (b) (arrow) of Arctic charr(Ringø, unpublished data). Bar = 1.0 μm.


6. CONCLUSIONS

The intestine and its associated microbiota can be considered as a complex ecosystem.Interactions and competition within the resident population, as well as dietary inputs,environmental conditions, and possibly the host immune system, influence the com-position of the enteric community and its ability to inhibit pathogens. On the basis ofthe results presented in this paper, it is suggested that LAB with antagonistic activityagainst fish pathogens are potential candidate probiotics in future aquaculture, but further studies on how LAB can improve the disease resistance in challenge tests withpathogenic bacteria and the influence of LAB on the GIT microbiota are necessary.Future studies on the effect of LAB administration on gut microbiota of fish shouldinclude molecular approaches to analyse bacterial communities as described byRaskin et al. (1997), Wallner et al. (1997) and Hugenholtz et al. (1998).


As most probiotic studies on fish have not seen any clear effect in disease studies, wesuggest a new strategy. We must think as the military and use military strategy. If weare going to defeat our enemies, the pathogens, we have to be at the same place andtime as them. For example, if probiotic bacteria mostly colonize the pyloric caeca,the probionts will have no effect if the pathogen mostly colonizes the mid or hindgutregions and translocates in these regions. In this respect, electron microscopical stud-ies may be a useful tool in evaluating which part of the gastrointestinal tract the pro-bionts colonize and where the main translocation of the pathogen occurs.

REFERENCES

Amàbile-Cuevas, C.F., Gàrdenas-Garcia, M., Lundgar, M., 1995. Antibiotic resistance. Amer. Sci. 83,320–329.

Aoki, T., Kanazawa, T., Kitao, T., 1985. Epidemiological surveillance of drug resistant Vibrioanguillarum strains. Fish Pathol. 20, 199–208.

Austin, B., McIntosh, D., 1988. Natural antibacterial compounds on the surface of rainbow trout, Salmogairdneri Richardson. J. Fish Diseases 11, 275–277.

Austin, B., Stuckey, L.F., Robertson, P.A.W., Effendi, I., Griffith, D.R.W., 1995. A probiotic strainof Vibrio alginolyticus effective in reducing diseases caused by Aeromonas salmonicida, Vibrioanguillarum and Vibrio ordalii. J. Fish Diseases 18, 93–96.

Baudin Laurencin, F., Germon, E., 1987. Experimental infection of rainbow trout, Salmo gairdneri R.,by dipping in suspensions of Vibrio anguillarum: ways of bacterial penetration: influence of temper-ature and salinity. Aquaculture 67, 203–205.

Bergh, Ø., 1995. Bacteria associated with early life stages of halibut, Hippoglossus hippoglossus L.,inhibit growth of a pathogenic Vibrio spp. J. Fish Diseases 18, 31–40.

Bhugaloo-Vial, P., Dousset, X., Metivier, A., Sorokine, O., Anglade, P., Boyaval, P., Marion, D., 1996.Purification and amino acid sequences of piscicocins V1a and V1b, two class IIa bacteriocins secretedby Carnobacterium piscicola V1 that display significantly different levels of specific inhibitory activ-ity. App. Environ. Microbiol. 62, 4410–4416.

Bogut, I., Milakovic, Z., Bukvic, Z., Brkic, S., Zimmer, R., 1998. Influence of probiotic (Streptococcusfaecium M74) on growth and content of intestinal microflora in carp (Cyprinus carpio). Czech.J. Anim. Sci. 43, 231–235.


Bogut, I., Milakovic, Z., Brkic, S., Novoselic, D., Bukvic, Z., 2000. Effects of Enterococcus faecium onthe growth rate and content of intestinal microflora in Sheat fish (Silurus glanis). Vet. Med. Czech.4, 107–109.

Byun, J.W., Park, S.C., Benno, Y., Oh, T.K., 1997. Probiotic effect of Lactobacillus sp. DS-12 in floun-der (Paralichthys olivaceus). J. Gen. Appl. Microbiol. 43, 305–308.

Cai, Y.M., Suyanandana, P., Saman, P., Benno, Y., 1999. Classification and characterization of lactic acidbacteria isolated from the intestines of common carp and freshwater prawns. J. Gen. Appl. Microbiol.45, 177–184.

Cain, K.D., LaPatra, S.E., Baldwin, T.J., Shewmaker, B., Jones, J., Ristow, S.S., 1996. Characterizationof mucosal immunity in rainbow trout Oncorhynchus mykiss challenged with infectious hematopoi-etic necrosis virus: identification of antiviral activity. Dis. Aquatic Org. 27, 161–172.

Campbell, A.C., Buswell, J.A., 1983. The intestinal microflora of farmed Dover sole (Solea solea) atthree different stages of fish development. J. Appl. Bacteriol. 35, 215–223.

Chair, M., Dehasque, M., van Poucke, S., Nelis, H., Sorgeloos, P., de Leenher, A.P., 1994. An oral chal-lenge for turbot larvae with Vibrio anguillarum. Aquacult. Int. 2, 270–272.

Chart, H., Munn, C.B., 1980. Experimental vibriosis in the eel (Anguilla anguilla). In: Ahne, W.T.C.-S.(Ed.), Fish Diseases. Springer-Verlag, Berlin, pp. 39–44.

Chateau, N., Castellanos, I., Deschamps, A.M., 1993. Distribution of pathogen inhibition in theLactobacillus isolates of commercial probiotic consortium. J. Appl. Bacteriol. 74, 36–40.

Cid, D., Sanz, R., Marin, I., De Greve, H., Ruiz-Santd-Quiteria, J.A., Amils, R., De la Fuente, R. (1999).Characterization of nonenterotoxigenic Escherichia coli strains producing F17 fimbriae isolatedfrom diarrheic lambs and goat kids. J. Clin. Microbiol. 37, 1370–1375.

Cole, A.M., Weis, P., Diamond, G., 1997. Isolation and characterization of pleurocidin, an antimicrobialpeptide in the skin secretions of winter flounder. J. Biol. Chem. 272, 12 008–12 013.

Collins, M.D., Farrow, A.E., Phillips, B.A., Ferusu, S., Jones, D., 1987. Classification of Lactobacillusdivergens, Lactobacillus piscicola and some catalase-negative, asporogenous, rod-shaped bacteriafrom poultry in a new genus Carnobacterium. Int. J. Sys. Bacteriol. 37, 310–316.

Conway, P.L., 1989. Lactobacilli: fact and fiction. In: Grubb, R., Midtvedt, T., Norrin, E. (Eds.), The Regulatory and Protective Role of the Normal Microflora. Macmillan Press Ltd., London, pp. 263–282.

Dopazo, C.P., Lemos, M.L., Lodeiros, C., Bolinches, J., Barja, J.L., Toranzo, A.E., 1988. Inhibitory activ-ity of antibiotic-producing marine bacteria against fish pathogens. J. Appl. Bacteriol. 65, 97–101.

Duffes, F., Corre, C., Leroi, F., Dousset, X., Boyaval, P., 1999. Inhibition of Listeria monocytogenes byin situ produced and semipurified bacteriocins of Carnobacterium spp. on vacuum-packed, refriger-ated cold-smoked salmon. J. Food Prot. 62, 1394–1403.

Duffes, F., Leroi, F., Dousset, X., Boyaval, P., 2000. Use of a bacteriocin producing Carnobacterium piscicola strain, isolated from fish to control Listeria monocytogenes development in vacuum-packedcold-smoked salmon stored at 4°C. Sci. Alim. 1, 153–158.

Ebran, N., Julien, S., Orange, N., Saglio, P., Lemaitre, C., Molle, G., 1999. Pore-forming properties andantibacterial activity of proteins extracted from epidermal mucus of fish. Comp. Biochem. Physiol.A 122, 181–189.

Ebran, N., Julien, S., Orange, N., Auperin, B., Molle, G., 2000. Isolation and characterization of novelglycoproteins from fish epidermal mucus: correlation between their pore-forming properties and theirantibacterial activities. Biochem. Biophys. Acta 1467, 271–280.

Ennahar, S., Sashihara, T., Sonomoto, K., Ishizaki, A., 2000. Class IIa bacteriocins: biosynthesis, struc-ture and activity. FEMS Microbiol. Rev. 24, 85–106.

Espelid, S., 1991. Immune responses in fish against pathogenic vibrios. Dr. Sci. Thesis, The NorwegianCollege of Fishery Science, University of Tromsø, Norway.

Esteve, C., Garay, E., 1991. Heterotrophic bacterial flora associated with European eel Anguilla anguillareared in freshwater. Nip. Sui. Gak. 57, 1369–1375.

Evelyn, T.P.T., 1984. Immunization against pathogenic vibrios. In: de Kinkelin, P. (Ed.), Symposium onFish Vaccination. Off. Int. Epizooties (IOE), Paris, pp. 121–149.

Fernandes, C.F., Shahani, K.M., Amer, M.A., 1987. Therapeutic role of dietary lactobacilli and lacto-bacillic fermented dairy products. FEMS Microbiol. Rev. 46, 343–356.


Fernandes, C.F., Shahani, K.M., Amer, M.A., 1988. Effect of nutrient media and bile salts on growth andantimicrobial activity of Lactobacillus acidophilus. J. Dairy Sci. 71, 3222–3229.

Floch, M.H., Binder, H.J., Filburn, B., Gesshengoren, W., 1972. The effect of bile acids on intestinalmicroflora. Amer. J. Clin. Nutr. 25, 1418–1426.

Fouz, B., Devesa, S., Gravningen, K., Barja, J.L., Toranzo, A.E., 1990. Antibacterial action of the mucusof the turbot. Bull. European Ass. Fish Pathol. 10, 56–59.

Fricourt, B.V., Barefoot, S.F., Testin, R.F., Hayasaka, S.S., 1994. Detection and activity of plantaricin-Fan antibacterial substance from Lactobacillus plantarum BF001 isolated from processed channel cat-fish. J. Food Prot. 57, 698–701.

Fuller, R., 1986. Probiotics. J. Appl. Bacteriol., Suppl. 1S–7S.Fuller, R., 1997. Probiotics 2. Applications and Practical Aspects. Chapman and Hall, London, pp. 212.Garcia, T., Otto, K., Kjelleberg, S., Nelson, D.R., 1997. Growth of Vibrio anguillarum in salmon intes-

tinal mucus. Appl. Environ. Microbiol. 63, 1034–1039.Garcia de la Banda, I., Cherequini, O., Rasines, I., 1992. Influence of lactic bacterial additives on turbot

(Scophthalmus maximus L.) larvae culture. Bol. Instit. Esp. Oceano. 8, 247–254.Gatesoupe, F.J., 1990. The continuous feeding of turbot larvae, Scophthalmus maximus, and control of

bacterial environment of rotifers. Aquaculture 89, 139–148.Gatesoupe, F.J., 1991a. The effect of three strains of lactic acid bacteria on the production rate of rotifers,

Brachionus plicatilis, and their dietary value for larval turbot, Scophthalmus maximus. Aquaculture96, 335–342.

Gatesoupe, F.J., 1991b. Bacillus sp. spores as food additive for the rotifer Brachionus plicatilis:improvement of their bacterial environment and their dietary value for larval turbot, Scophthalmusmaximus L. In: Kaushik, S. (Ed.), Fish Nutrition in Practice, Proceedings of the 4th InternationalSymposium on Fish Nutrition and Feeding. INRA (Les colloques, No. 61), Paris, pp. 561–568.

Gatesoupe, F.J., 1994. Lactic acid bacteria increase the resistance of turbot larvae, Scophthalmusmaximus, against pathogenic vibrio. Aquatic Living Res. 7, 277–282.

Gatesoupe, F.J., 1999. The use of probiotics in aquaculture. Aquaculture 180, 147–165.Gatesoupe, F.J., Arakawa, T., Watanabe, T., 1989. The effect of bacterial additives on the production rate

and dietary value of rotifers as food for Japanese flounder, Paralichthys olivaceus. Aquaculture83, 39–44.

Gildberg, A., Mikkelsen, H., 1998. Effects of supplementing the feed to Atlantic cod (Gadus morhua) fry withlactic acid bacteria and immunostimulating peptides during a challenge trial with Vibrio anguillarum.Aquaculture 167, 103–113.

Gildberg, A., Johansen, A., Bøgvald, J., 1995. Growth and survival of Atlantic salmon (Salmo salar) frygiven diets supplemented with fish protein hydrolysate and lactic acid bacteria during a challenge trialwith Aeromonas salmonicida. Aquaculture 138, 23–34.

Gildberg, A., Mikkelsen, H., Sandaker, E., Ringø, E., 1997. Probiotic effect of lactic acid bacteria in the feed on growth and survival of fry of Atlantic cod (Gadus morhua). Hydrobiologia 352,279–285.

Gilliland, S.E., 1979. Beneficial interrelationships between certain microorganisms and humans: candi-date organisms for use as dietary adjuncts. J. Food Prot. 42, 164–167.

Gomez-Gill, B., Roque, A., Turnbull, J.F., 2000. The use and selection of prebiotic bacteria for use in theculture of larval aquatic organisms. Aquaculture 191, 259–270.

González, C.L., Enicinas, J.P., García López, M.L., Otero, A., 2000. Characterization and identificationof lactic acid bacteria from freshwater fishes. Food Microbiol. 17, 383–391.

Gorbach, S.L., 1990. Lactic acid bacteria and human health. Ann. Med. 22, 37–41.Gram, L., Melichiorsen, J., 1996. Interaction between fish spoilage bacteria Pseudomonas sp. and

Shewanella putrefaciens in fish extracts and on fish tissue. J. Appl. Bacteriol. 80, 589–595.Gram, L., Ringø, E., 2005. Prospect of fish probiotics. In: Holzapfel, W.H., Naughton, P.J. (Eds.),

Microbial Ecology of the Growing Animal. Biology in Growing Animals Series, Pierzynowski, S.G.,Zabielski, R. (Eds.). Elsevier, Oxford, Chapter 17.

Gram, L., Melichiorsen, J., Spanggard, B., Huber, I., Nielsen, T.F., 1999. Inhibition of Vibrio anguillarumby Pseudomonas fluorescens strain AH2, a possible probiotic treatment of fish. Appl. Environ.Microbiol. 65, 969–973.


Grisez, L., Chair, M., Sorgeloos, P., Ollevier, F., 1996. Mode of infection and spread of Vibrio anguillarumin turbot Scophthalmus maximus larvae after oral challenge through live feed. Dis. Aquatic Org.26, 181–187.

Hamácková, J., Párová. J., Vachta, R., Kumprecht, I., 1992. Effects of additive of microbioticsStreptococcus faecium M-74 to the feedstuff on the Sheat fish (Silurus glanis) fry growth. Bull.VÚRH Vodnany 28, 10–15.

Hansen, G.H., Olafsen, J.A., 1999. Bacterial interactions in early life stages of marine cold water fish.Microb. Ecol. 38, 1–26.

Hansen, G.H., Strøm, E., Olafsen, J.A., 1992. Effect of different holding regimes on the intestinalmicroflora of herring (Clupea harengus) larvae. Appl. Environ. Microbiol. 58, 461–470.

Hansen, L.T., Huss, H.H., 1998. Comparison of the microflora isolated from spoiled cold-smoked salmonfrom three smokehouses. Food Res. Int. 31, 703–711.

Harrell, L.W., Etlinger, H.M., Hodgins, H.O., 1976. Humoral factors important in resistance of salmonidfish to bacterial disease. II. Anti-Vibrio anguillarum activity in mucus and observations on comple-ment. Aquaculture 7, 363–370.

Hart, S., Wrathmell, A.B., Harris, J.E., Dogett, T.A., 1987. Gut associated lymphoid tissue (GALT) in thecommon dogfish Scyliorhinus canicula L. J. Mar. Biol. Ass. UK 67, 639–645.

Hart, S., Wrathmell, A.B., Harris, J.E., Grayson, T.H., 1988. Gut immunology in fish: A review. Develop.Comp. Immunol. 12, 453–480.

Harzevili, A.R.S., Van Duffel, H., Dhert, P., Swings, J., Sorgeloss, P., 1998. Use of a potential probioticLactococcus lactis AR21 strain for the enhancement of growth in the rotifer Brachionus plicatilis(Müller). Aquacult. Res. 29, 411–417.

Hjeltnes, B., Andersen, K., Ellingsen, H.M., Egedius, E., 1987. Experimental studies on pathogenicity ofa Vibrio sp. isolated from Atlantic salmon, Salmo salar L., suffering from Hitra disease. J. FishDiseases 10, 21–27.

Holo, H., Nilssen, O., Nes, I.F., 1991. Lactococcin A, a new bacteriocin from Lactococcus lactis subsp.cremoris: isolation and characterization of the protein and its gene. J. Bacteriol. 173, 3879–3887.

Horne, M.T., Baxendale, A., 1983. The adhesion of Vibrio anguillarum to host tissues and its role inpathogenesis. J. Fish Diseases 6, 461–471.

Hugenholtz, P., Pitulle, C., Hershberger, K.L., Pace, N.R., 1998. Novel division level bacterial diversityin a Yellowstone hot spring. J. Bacteriol. 180, 366–376.

Jack, R.W., Wan, J., Gordon, J., Harmark, K., Davidson, B.E., Hillier, A.J., Wettenhall, R.E.H., Hickey, M.W.,Coventry, M.J., 1996. Characterization of the chemical and antimicrobial properties of piscicolin 126,a bacteriocin produced by Carnobacterium piscicola JG126. Appl. Environ. Microbiol. 62, 2897–2903.

Jöborn, A., 1998. The role of the gastrointestinal microbiota in the prevention of bacterial infections infish. Fil. Dr. thesis, Gothenburg University, Sweden. ISBN 91-628-2596-8.

Jöborn, A., Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1997. Colonization in the fishintestinal tract and production of inhibitory substances in intestinal mucus and faecal extracts byCarnobacterium sp. strain K1. J. Fish Diseases 20, 383–392.

Jöborn, A., Larsson, T., Karlsson, H., Karlsson, K.-A., Kahnberg, P., Sterner, O., Conway, P.L.,Kjelleberg, S., 1998. Inhibition of the growth of bacterial fish pathogens by low molecular weightcomponents produced by Carnobacterium inhibens isolated from the intestine of Atlantic salmon(Salmo salar). In: Jöborn, A., The role of the gastrointestinal microbiota in the prevention of bacte-rial infections in fish. Fil. Dr. thesis, Gothenburg University, Sweden. ISBN 91-628-2596-8.

Jöborn, A., Dorsch, M., Olsson, J.C., Westerdahl, A., Kjelleberg, S., 1999. Carnobacterium inhibens sp.nov. isolated from the intestine of Atlantic salmon (Salmo salar). Int. J. Sys. Bacteriol. 49, 1891–1898.

Juven, B.J., Meinersmann, R.J., Stern, N., 1991. Antagonistic effects of lactobacilli and pediococci to con-trol intestinal colonization by human enteropathogens in live poultry. J. Appl. Bacteriol. 70, 95–103.

Kanno, T., Nakai, T., Muroga, K., 1990. Mode of transmission of vibrios among ayu Plecoglossusalvelitis. J. Aquat. Anim. Health 1, 2–6.

Klaenhammer, T.R., 1988. Bacteriocins of lactic acid bacteria. Biochimie 70, 337–349.Kmet, V., Lucchini, F., 1999. Aggregation of sow lactobacilli with diarrhoeagenic Escherichia coli.

Zentralbl. Veterinarmed. (B) 46, 683–687.Kraus, H., 1961. Kurze mitteilung über das vorkommen von laktobazillen auf frischen heringen. Arch.

Lebensmittelh. 12, 101–102.


Kvasnikov, E.I., Kovalenko, N.K., Materinskaya, L.G., 1977. Lactic acid bacteria of freshwater fish.Microbiol. 46, 619–624.

Lee, N.K., Jun, S.A., Ha, J.U., Paik, H.D., 2000. Screening and characterization of bacteriocinogenic lac-tic acid bacteria from Jeot-gal, a Korean fermented fish food. J. Microbiol. Biotechnol. 10, 423–428.

Lemaitre, C., Orange, N., Saglio, P., Saint, N., Gagnon, J., Molle, G., 1996. Characterization and ionchannel activities of novel antibacterial proteins from the skin mucosa of carp (Cyprinus carpio). Eur.J. Biochem. 240, 143–149.

Leroi, F., Joffaud, J.J., Chevalier, F., Cardinal, M., 1998. A study of the microbial ecology of cold-smokedsalmon during storage at 8°C. Int. J. Food Microbiol. 39, 111–121.

Lesel, R., Sechet, J., 1982. Influence d’un stress de manipulation sur le transi de la microflore bacteriennedigestive chez la truite arc-en-ciel Salmo gairdneri Richardson. Acta Oecol. Oecol. Appl. 3, 23–33.

Lin, J.H.C., Savage, D.C., 1984. Host specificity of the colonization of murine gastric epithelium bylactobacilli. FEMS Microbiol. Lett. 24, 67–71.

Lindgren, S.E., Dobrogosz, W.J., 1990. Antagonistic activities of lactic acid bacteria in food and feedfermentations. FEMS Microbiol. Rev. 87, 149–164.

Lødemel, J.B., Mayhew, T.M., Myklebust, R., Olsen, R.E., Espelid, S., Ringø, E., 2001. Effect of threedietary oils on disease susceptibility in Arctic charr (Salvelinus alpinus L.) during cohabitantchallenge with Aeromonas salmonicida ssp. salmonicida. Aquacult. Res. 32, 935–945.

Lyhs, U., Björkroth, J., Korkeala, H., 1999. Characterisation of lactic acid bacteria from spoiled, vacuum-packaged, cold-smoked rainbow trout using ribotyping. Int. J. Food Microbiol. 52, 77–84.

Madigan, M.T., Martinko, J.M., Parker, J., 1997. Biology of Microorganisms. Prentice-Hall, EnglewoodCliffs, NJ, p. 986.

Magarinos, B., Pazos, F., Santos, Y., Romalde, J.L., Toranzo, A.E., 1995. Response of Pasteurella piscicidaand Flexibacter maritimus to skin mucus of marine fish. Dis. Aquatic Org. 21, 103–108.

Marmur, J., 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol.Biol. 3, 208–218.

Mazeaud, M.M., Mazeaud, F., Donaldson, E.H., 1977. Stress resulting from handling in fish: primary andsecondary effects. Trans. Am. Fish. Soc. 106, 201–212.

Metivier, A., Pilet, M.-F., Dousset, X., Sorokine, O., Anglade, P., Zagorec, M., Piard, J.-C., Marion, D.,Cenatiempo, Y., Femaux, C., 1998. Divercin V41, a new bacteriocin with two disulphide bonds pro-duced by Carnobacterium divergens V41: primary structure and genomic organization. Microbiology144, 2837–2844.

Mims, C.A., Dimmock, N.J., Nash, A., Stephen, J., 1995. Mims’ Pathogenesis of Infectious Disease,4th edition. Academic Press, London, p. 414.

Mol, O., Fokkema, H., and Oudega, B. (1996a) The Escherichia coli K99 periplasmic chaperone FanEis a monomeric protein. FEMS Microbiol. Lett. 138, 185–189.

Mol, O., Oudhuis, W.C., Fokkema, H., and Oudega, B. (1996b) The N-terminal β-barrel domain of theEscherichia coli K88 periplasmic chaperone FaeE determines fimbrial subunit recognition and dimer-ization. Mol. Microbiol. 22, 379–388.

Munro, P.D., Barbour, A., Birkbeck, T.H., 1995. Comparison of the growth and survival of larval turbotin the absence of culturable bacteria with those in the presence of Vibrio anguillarum, Vibrioalginolyticus, or a marine Aeromonas sp. Appl. Environ. Microbiol. 61, 4425–4428.

Muroga, K., De La Cruz, M.C., 1987. Fate and location of Vibrio anguillarum in tissues of artificiallyinfected ayu (Plecoglossus altivelis). Fish Pathol. 22, 99–103.

Muroga, K., Higashi, M., Keetoku, H., 1987. The isolation of intestinal microflora of farmed redseabream (Pagrus major) and black seabream (Acanthopagrus schlegeli) at larval and juvenile stages.Aquaculture 65, 79–88.

Naidu, A.S., Bidlack, W.R., Clemens, R.A., 1999. Probiotic spectra of lactic acid bacteria (LAB). Crit.Rev. Food Sci. Nutr. 38, 13–126.

Nettles, C.G., Barefoot, S.F., 1993. Biochemical and genetic characteristics of bacteriocins of food-associated lactic acid bacteria. J. Food Prot. 56, 338–356.

Nicolas, J.L., Robic, E., Ansquer, D., 1989. Bacterial flora associated with a trophic chain consisting ofmicro-algae, rotifers and turbot larvae: influence of bacteria on larval survival. Aquaculture 83, 237–248.

Nikoskelainen, S., Ouwehand, A.C., Salminen, S., Bylund, G., 2001. Protection of rainbow trout(Oncorhynchus mykiss) from furunculosis by Lactobacillus rhamnosus. Aquaculture 198, 229–236.


Nissen-Meyer, J., Nes, I.F., 1997. Ribosomally synthesized antimicrobial peptides: their function, struc-ture, biogenesis, and mechanisms of action. Arch. Microbiol. 167, 67–77.

Noh, S.H., Han, K., Won, T.H., Choi, Y.J., 1994. Effect of antibiotics, enzyme, yeast culture and pro-biotics on the growth performance of Israeli carp. Korean J. Anim. Sci. 36, 4809–4816.

Olafsen, J.A., Hansen, G.H., 1992. Intact antigen uptake in intestinal epithelial cells of marine fish larave.J. Fish Biol. 40, 141–156.

Olsen, M.A., Aagnes, T.H., Mathiesen, S.D., 1994. Digestion of herring by indigenous bacteria in theminke whale forestomach. Appl. Environ. Microbiol. 60, 4445–4455.

Olsen, R.E., Ringø, E., 1997. Lipid digestibility in fish: A review. In: Pandalai, S.G. (Ed.), Recent ResearchDevelopments in Lipids Research, Vol. 1. Transworld Research Network, Trivandrum, pp. 199–265.

Olsen, R.E., Myklebust, R., Kaino, T., Ringø, E., 1999. Lipid digestibility and histological changes in theenterocytes of Arctic charr (Salvelinus alpinus L.) fed linseed oil and soybean lecithin. Fish Physiol.Biochem. 21, 35–44.

Olsen, R.E., Myklebust, R., Ringø, E., Mayhew, T.M., 2000. The influence of dietary linseed oil and sat-urated fatty acids on caecal enterocytes of Arctic charr (Salvelinus alpinus L.): a quantitative ultra-structural study. Fish Physiol. Biochem. 22, 207–216.

Olsson, C., 1995. Bacteria with inhibitory activity and Vibrio anguillarum in fish intestinal tract. Fil.Dr. Thesis. Gothenburg University, Sweden, ISBN 91-628-1850-3, p.141

Olsson, J.C., Westerdahl, A., Conway, P.L., Kjelleberg, S., 1992. Intestinal characterization potential ofturbot (Scophthalmus maximus) and dab (Limanda limanda)-associated bacteria with inhibitoryeffects against Vibrio anguillarum. Appl. Environ. Microbiol. 58, 551–556.

Olsson, J.C., Jöborn, A., Westerdahl, A., Blomberg, L., Kjelleberg, S., Conway, P.L., 1996. Is the turbot,Scophthalmus maximus L., intestine a port of entry for the fish pathogen Vibrio anguillarum? J. FishDiseases 19, 225–234.

Ottesen, O.H., Olafsen, J.A., 2000. Effects on survival and mucous cell proliferation of Atlantic halibut,Hippoglossus hippoglossus L., larvae following microflora manipulation. Aquaculture 187, 225–238.

Ouwehand, A.C., 1998. Antimicrobial components from lactic acid bacteria. In: Salminen, S., vonWright, A. (Eds.), Lactic Acid Bacteria. Marcel Dekker, Inc., New York, pp. 139–159.

Paik, H.-D., Oh, D.W., 1996. Purification, characterization, and comparison of bacteriocins. J. Microbiol.Biotechnol. 6, 151–161.

Paludan-Müller, C., Huss, H.H., Gram, L., 1999. Characterization of lactic acid bacteria isolated from aThai low-salt fermented fish product and the role of garlic as substrate for fermentation. Int. J. FoodMicrobiol. 46, 219–229.

Parente, E., Ricciardi, A., 1994. Influence of pH on the production of enterocin 1146 during batch fer-mentation. Lett. Appl. Microbiol. 19, 12–15.

Parente, E., Ricciardi, A., 1999. Production, recovery and purification of bacteriocins from lactic acidbacteria. Appl. Microbiol. Biotechnol. 52, 628–638.

Perez-Benavente, G., Gatesoupe, F.J., 1988. Bacteria associated with cultures of rotifers and Artemia aredetrimental to larval turbot, Scophthalmus maximus L. Aquacult. Eng. 118, 289–293.

Peters, G., Faisal, M., Lang, T., Ahmed, I., 1988. Stress caused by social interaction and its effect on suscep-tibility to Aeromonas hydrophila infection in rainbow trout Salmo gairdneri. Dis. Aquatic Org. 4, 83–89.

Piard, J.C., Desmazeaud, M., 1992. Inhibition factors produced by lactic acid bacteria. 2. Bacteriocinsand other antibacterial substances. Lait 72, 113–142.

Pilet, M.-F., Dousset, X., Barr, R., Novel, G., Desmazeaud, M., Piard, J.-C., 1995. Evidence for two bac-teriocins produced by Carnobacterium piscicola and Carnobacterium divergens isolated from fishand active against Listeria monocytogenes. J. Food Prot. 58, 256–262.

Quadri, L.E.N., Sailer, M., Roy, K.L., Vederas, J.C., Stiles, M.E., 1994. Chemical and genetic characteriza-tion of bacteriocins produced by Carnobacterium piscicola LV17B. J. Biol. Chem. 269, 12 204–12 211.

Raskin, L., Capman, W.C., Sharp, R., Poulsen, L.K., Stahl, D.A., 1997. Molecular ecology of gastroin-testinal ecosystems. In: Mackie, R.I., White, B.A., Isacson, R.E. (Eds.), Gastrointestinal Microbiology,Vol. 2, Gastrointestinal Microbes and Host Interactions. Chapman and Hall Microbiology Series,International Thompson Publishing, New York, pp. 243–298.

Ringø, E. The ability of carnobacteria isolated from fish intestine to inhibit growth of fish pathogenicbacteria. Unpublished data.

Ringø, E., 1993a. Does dietary linoleic acid affect intestinal microflora in Arctic charr, Salvelinus alpinus (L.)?Aquacult. Fish. Manage. 24, 133–135.


Ringø, E., 1993b. Does chromic oxide (Cr2O3) affect faecal lipid and intestinal bacterial flora in Arcticcharr, Salvelinus alpinus (L.)? Aquacult. Fish. Manage. 24, 767–776.

Ringø, E., 1993c. The effect of chromic oxide (Cr2O3) on aerobic bacterial populations associated withthe epithelial mucosa of Arctic charr, Salvelinus alpinus (L.). Can. J. Microbiol. 39, 1169–1173.

Ringø, E., 1994. The effect of chromic oxide (Cr2O3) on faecal lipid and intestinal bacterial flora of seawater reared Arctic charr, Salvelinus alpinus (L.). Aquacult. Fish. Manage. 25, 341–344.

Ringø, E., 1999a. Lactic acid bacteria in fish: antibacterial effect against fish pathogens. In: Krogdahl, Å.,Mathiesen, S.D., Pryme, I. (Eds.), Effects of antinutrients on the nutritional value of legume diets.Vol. 8. COST 98, EEC Publication, Luxembourg, pp. 70–75.

Ringø, E., 1999b. Does Carnobacterium divergens isolated from Atlantic salmon (Salmo salar L.) colonisethe gut of early developing turbot (Scophthalmus maximus L.) larvae? Aquacult. Res. 30, 229–232.

Ringø, E., Birkbeck, T.H., 1999. Intestinal microflora of fish larvae and fry. Aquacult. Res. 30, 73–93.Ringø, E., Gatesoupe, F.-J., 1998. Lactic acid bacteria in fish: a review. Aquaculture 160, 177–203.Ringø, E., Holzapfel, W., 2000. Identification and characterisation of carnobacteria associated with the

gills of Atlantic salmon (Salmo salar L.). Sys. Appl. Microbiol. 23, 523–527.Ringø, E., Olsen, R.E., 1999. The effect of diet on aerobic bacterial flora associated with intestine of

Arctic charr (Salvelinus alpinus L.). J. Appl. Microbiol. 86, 22–28.Ringø, E., Strøm, E., 1994. Microflora of Arctic charr, Salvelinus alpinus (L.); gastrointestinal microflora

of free-living fish, and effect of diet and salinity on the intestinal microflora. Aquacult. Fish. Manage.25, 623–629.

Ringø, E., Vadstein, O., 1998. Colonization of Vibrio pelagius and Aeromonas caviae in early develop-ing turbot (Scophthalmus maximus L.) larvae. J. Appl. Microbiol. 84, 227–233.

Ringø, E., Myklebust, R., Mayhew, T.M., Olsen, R.E. The influence of dietary inulin on aerobic bacteriaassociated with the gastrointestinal tract of Arctic charr (Salvelinus alpinus L.). unpublished results.

Ringø, E., Bendiksen, H.R., Gausen, S.J., Sundsfjord, A., Olsen, R.E., 1998. The effect of dietary fattyacids on lactic acid bacteria associated with the epithelial mucosa and from faecalia of Arctic charr,Salvelinus alpinus (L.). J. Appl. Microbiol. 85, 855–864.

Ringø, E., Olsen, R.E., Øverli, Ø., Løvik, F., 1997. Effect of dominance hierarchy formation on aerobicmicrobiota associated with epithelial mucosa of subordinate and dominant individuals of Arctic charr,Salvelinus alpinus (L.). Aquacult. Res. 28, 901–904.

Ringø, E., Bendiksen, H.R., Wesmajervi, M.S., Olsen, R.E., Jansen, P.A., Mikkelsen, H., 2000. Lacticacid bacteria associated with the digestive tract of Atlantic salmon (Salmo salar L.). J. Appl.Microbiol. 89, 317–322.

Ringø, E., Wesmajervi, M.S., Bendiksen, H.R., Berg, A., Olsen, R.E., Johnsen, T., Mikkelsen, H.,Seppola, M., Strøm, E., Holzapfel, W., 2001a. Identification and characterization of carnobacteriaisolated from fish intestine. Sys. Appl. Microbiol. 24, 183–191.

Ringø, E., Lødemel, J.B., Myklebust, R., Kaino, T., Mayhew, T.M., Olsen, R.E., 2001b. Epithelium asso-ciated bacteria in the gastrointestinal tract of Arctic charr (Salvelinus alpinus L.). An electron micro-scopical study. J. Appl. Microbiol. 90, 294–300.

Ringø, E., Lødemel, J.B., Myklebust, R., Jensen, L., Lund, V., Mayhew, T.M., Olsen, R.E., 2002a. The effectof soybean, linseed and marine oils on aerobic gut microbiota of Arctic charr (Salvelinus alpinus L.)before and after challenge with Aeromonas salmonicida ssp. salmonicida. Aquacult. Res. 33, 591–606.

Ringø, E., Seppola, M., Berg, A., Olsen, R.E., Schillinger, U., Holzapfel, W., 2002b. Characterization ofCarnobacterium divergens strain 6251 isolated from intestine of Arctic charr (Salvelinus alpinus L.).Sys. Appl. Microbiol. 25, 120–129.

Ringø, E., Olsen, R.E., Mayhew, T.M., Myklebust, R., 2003. Electron microscopy of the intestinalmicroflora of fish. Aquaculture 227, 395–415.

Robertson, P.A.W., O’Dowd, C., Burells, C., Williams, P., Austin, B., 2000. Use of Carnobacterium sp.as probiotic for Atlantic salmon (Salmo salar L.) and rainbow trout (Oncorhynchus mykiss,Walbaum). Aquaculture 185, 235–243.

Robinette, D., Wada, S., Arroll, T., Levy, M.G., Miller, W.L., Noga, E.J., 1998. Antimicrobial activity inthe skin of the channel catfish Ictalurus punctatus: characterization of broad-spectrum histone-likeantimicrobial proteins. CMLS Cell. Mol. Life Sci. 54, 467–475.

Romalde, J.L., Magarinos, B., Nunez, S., Barja, J.L., Toranzo, A.E., 1996. Host range susceptibility ofEnterococcus sp. strains isolated from diseased turbot: Possible routes of infection. Appl. Environ.Microbiol. 62, 607–611.


Saarela, S., Taira, S., Nurmiaho-Lassila, E-L., Makkonen, A., Rhen, M. (1995) The Escherichia coliG-fimbrial lectin protein participates both in fimbrial biogenesis and in recognition of the receptor N-acetyl-D-glucosamine. J. Bacteriol. 177, 1477–1484.

Saarela, S., Westerlund-Wikström, B., Rhen, M., Korhonen, T.K. (1996) The GafD protein of the G (F17)fimbrial complex confers adhesiveness of Escherichia coli to laminin. Infect. Immunity 64, 2857–2860.

Sanders, M.E., 1999. Probiotics. Food Technol. 53, 67–77.Savage, D.C., 1983. Mechanisms by which indigenous microorganisms colonize gastrointestinal epithe-

lial surface. Prog. Food Nutr. Sci. 7, 65–74.Schrøder, K., Clausen, E., Sandberg, A.M., Raa, J., 1980. Psychotrophic Lactobacillus plantarum from

fish and its ability to produce antibiotic substances. In: Connell, J.J. (Ed.), Advances in Fish Scienceand Technology. Fishing News Books, Farnham, Surrey, pp. 480–483.

Skjermo, J., Vadstein, O., 1999. Techniques for microbial control in intensive fish larval rearing.Aquaculture 177, 333–343.

Snieszko, S.F., 1974. The effect of environmental stress on outbreaks of infectious diseases of fish. J. FishDiseases 6, 197–208.

Sørrum, H., 1999. Antibiotic resistance in aquaculture. Acta Vet. Scand. Suppl. 92, 29–36.Spanggard, B., Huber, I., Nielsen, J., Nielsen, T., Appel, K.F., Gram, L., 2000. The microflora of rainbow

trout intestine: a comparison of traditional and molecular identification. Aquaculture 182, 1–15.Spencer, R.J., Chesson, A., 1994. The effect of Lactobacillus spp. on the attachment of enterotoxigenic

Escherichia coli to isolated porcine enterocytes. J. Appl. Bacteriol. 77, 215–220.Starliper, C.E., Shotts, E.B., Brown, J., 1992. Isolation of Carnobacterium piscicola and an unidentified

Gram-positive bacillus from sexually mature and post-spawning rainbow trout Oncorhynchus mykiss.Dis. Aquat. Org. 13, 181–187.

Stevens, K.A., Sheldon, B.W., Klapes, A., Klaenhammer, T.R., 1991. Nisin treatment for inactivation ofSalmonella species and other Gram-negative bacteria. Appl. Environ. Microbiol. 57, 3613–3615.

Stoffels, G., Nes, I.F., Gudmundsdottir, A., 1992a. Isolation and properties of a bacteriocin-producingCarnobacterium piscicola isolated from fish. J. Appl. Bacteriol. 73, 309–316.

Stoffels, G., Nissen-Meyer, J., Gudmundsdottir, A., Sletten, K., Holo, H., Nes, I.F., 1992b. Purificationand characterization of a new bacteriocin isolated from Carnobacterium spp. Appl. Environ.Microbiol. 58, 1417–1422.

Stoffels, G., Sahl, H.-G., Gudmundsdottir, A., 1993. Carnocin U149, a potential biopreservative producedby Carnobacterium piscicola: large scale purification and activity against various gram-positive bac-teria including Listeria sp. Int. J. Food Microbiol. 20, 199–210.

Strøm, E., 1988. Melkesyrebakterier i fisketarm. Isolasjon, karakterisering og egenskaper. Master`s thesis.The Norwegian College of Fishery Science, University of Tromsø, Norway (in Norwegian).

Strøm, E., Olafsen, J.A., 1990. The indigenous microflora of wild-captured juvenile cod in net-pen rear-ing. In: Lesel, R. (Ed.), Microbiology in Poecilotherms. Elsevier, Amsterdam, pp. 181–185.

Strøm, E., Ringø, E., 1993. Changes in the bacterial flora in early developing cod, Gadus morhua (L.), larvaeafter an inoculation of Lactobacillus plantarum in the water. In: Walter, B., Fyhn, H.J. (Eds.), Physiologicaland Biochemical Aspects of Fish Development. University of Bergen, Norway, pp. 226–228.

Sugita, H., Tokuyama, K., Deguchi, Y., 1985. The intestinal microflora of carp Cyprinus carpio, grass carpCtenopharynogodon idella and tilapia Sarotherodon niloticus. Bull. Jap. Soc. Sci. Fish. 51, 1325–1329.

Sugita, H., Tsunohara, M., Ohkoshi, T., Deguchi, Y., 1988. The establishment of an intestinal microflorain developing goldfish (Carassius auratus) of culture ponds. Microb. Ecol. 15, 333–344.

Sugita, H., Shibuya, K., Shimooka, H., Deguchi, Y., 1996. Antibacterial abilities of intestinal bacteria infreshwater cultured fish. Aquaculture 145, 195–203.

Sugita, H., Shibuya, K., Hanada, H., Deguchi, Y., 1997. Antibacterial abilities of intestinal microflora ofthe river fish. Fish. Sci. 63, 378–383.

Sugita, H., Hirose, Y., Matsuo, N., Deguchi, Y., 1998. Production of the antibacterial substance byBacillus sp. strain NM 12, an intestinal bacterium of Japanese coastal fish. Aquaculture 165, 269–280.

Svendsen, Y.S., Bøgwald, J., 1997. Influence of artificial wound and non-intact mucus layer on mortalityof Atlantic salmon (Salmo salar L.) following a bath challenge with Vibrio anguillarum andAeromonas salmonicida. Fish Shellfish Immunol. 7, 317–325.

Svendsen, Y.S., Dalmo, R.A., Bøgwald, J., 1999. Tissue localization of Aeromonas salmonicida inAtlantic salmon, Salmo salar L., following experimental challenge. J. Fish Diseases 22, 125–131.


Tagg, J.R., Dajani, A.S., Wannamaker, L., 1976. Bacteriocins of Gram-positive bacteria. Bacteriol. Rev.40, 722–756.

Takashashi, Y., Itami, T., Kajiwaki, T., 1992. Bacteriolytic substances in the skin mucus of yellowtail.Gyobyo Kenkyu 27, 171–172.

Takemaru, I., Kusuda, R., 1988a. In vitro antibacterial activity of josamycin against Streptococcus sp. andvarious pathogenic bacteria. Nip. Sui. Gak. 54, 1527–1531.

Takemaru, I., Kusuda, R., 1988b. Effect of josamycin administration on intestinal microflora of culturedyellowtail. Nip. Sui. Gak. 54, 837–840.

Tamura, S., Shimizu, T., Ikegami, S., 1993. Endocytosis in adult eel intestine: immunological detectionof phagocytic cells in the surface epithelium. Biolog. Bull. 184, 330–337.

Tanasomwang, V., Muroga, K., 1988. Intestinal microflora of larval and juvenile stages in japanese floun-der (Paralichthys olivaceus). Fish Pathol. 23, 77–83.

Tanasupawat, S., Okada, S., Komagata, K., 1998. Lactic acid bacteria found in fermented fish inThailand. J. Gen. Appl. Microbiol. 44, 193–200.

Tanasupawat, S., Shida, O., Okada, S., Komagata, K., 2000. Lactobacillus acidipiscis sp. nov. andWeissella thailandensis sp. nov., isolated from fermented fish in Thailand. Int. J. Sys. Evolution.Microbiol. 50, 1479–1485.

Tannock, G.W., 1983. Effect of dietary and environmental stress on the gastrointestinal microbiota. In:Hentgens, D.J. (Ed.), Human Intestinal Microflora in Health and Disease. Academic Press, New York,pp. 517–539.

Tannock, G.W., 1995. Normal Microflora. An Introduction to Microbes Inhabiting the Human Body.Chapman and Hall, London.

Tannock, G.W., 1999. A fresh look at the intestinal microflora. In: Tannock, G.W. (Ed.), Probiotics.A Critical Review. Horizon Scientific Press. Norfolk, UK, pp. 5–14.

Tannock, G.W., Savage, D.C., 1974. Association of microorganisms with epithelium in the alimentarytract of Aspicularis tetraptera. Infect. Immun. 9, 475–476.

Tatner, M.F., Johnson, C.M., Horne, M.T., 1984. The tissue localization of Aeromonas salmonicida inrainbow trout, Salmo gairdneri Richardson, following three methods of administration. J. Fish Biol.25, 95–108.

Towner, K.J., 1995. The genetics of resistance. In: Greenwood, D. (Ed.), Antimicrobial Chemotheraphy.Oxford University Press, Oxford, pp. 159–167.

Trust, T.T., 1986. Pathogenesis of infectious diseases of fish. Ann. Rev. Microbiol. 40, 479–502.Trust, T.J., Sparrow, R.A.H., 1974. The bacterial flora in the alimentary tract of freshwater salmonid

fishes. Can. J. Microbiol. 20, 1219–1228.Vandenbergh, P., 1993. Lactic acid bacteria, their metabolic products and interference with microbial

growth. FEMS Microbiol. Rev. 12, 221–238.Verschuere, L., Rombout, G., Sorgeloos, P., Verstraete, W., 2000. Probiotic bacteria as biological control

agents in aquaculture – a review. Microbiol. Mol. Biol. Rev. 64, 655–671.Vigneulle, M., Laurencin, F.B., 1991. Uptake of Vibrio anguillarum bacterin in the posterior intestine of

rainbow trout (Oncorhynchus mykiss), sea bass (Dicentrarchus labrax) and turbot (Scophthalmusmaximus) after oral administration or anal intubation. Dis. Aquatic Org. 11, 85–92.

Wallbanks, S., Martinez-Murcia, A.J., Fryer, J.L., Phillips, B.A., Collins, M.D., 1990. 16S rRNAsequence determination for members of the genus Carnobacterium and related lactic acid bacteriaand description of Vagococcus salmoninarum sp. nov. Int. J. Syst. Bacteriol. 40, 224–230.

Wallner, G., Fuchs, B., Spring, S., Beisker, W., Amann, R., 1997. Flow sorting of microorganisms formolecular analysis. Appl. Environ. Microbiol. 63, 4223–4231.

Wedemeyer, G., 1970. The role of stress in the disease resistance of fishes. Special Pub. Amer. Fish. Soc.5, 30–35.

Westerdahl, A., Olsson, C.J., Kjelleberg, S., Conway, P.L., 1991. Isolation and characterization of turbot(Scophthalmus maximus)-associated bacteria with inhibitory effects against Vibrio anguillarum.Appl. Environ. Microbiol. 57, 2223–2228.

Westerdahl, A., Jöborn, A., Olsson, C.J., Kjelleberg, S., Conway, P.L., 1998. Classification and antag-onistic activity of fish intestinal bacteria with inhibitory effects against fish pathogens. In: Jöborn, A.,The role of the gastrointestinal microbiota in the prevention of bacterial infections in fish. Fil. Dr.thesis, Göteborg University, Sweden. ISBN 91-628-2596-8.

454

Probiotics may represent an effective alternative to the use of synthetic substances innutrition and medicine. The data concerning the efficacy of probiotics are often contradictory. It is therefore important to search for ways to improve the efficacy ofprobiotic microorganisms. In order to improve the selection and enhance the efficacyof probiotics, additional knowledge is required on the mode(s) of action. The efficacyof probiotics may be potentiated by several methods: the selection of more efficientstrains; genetic modification; combination of several strains; and the combination ofprobiotics and synergistically acting components. Combination with synergisticallyacting components of natural origin seems to be one of the best ways of potentiatingthe efficacy of probiotics and is widely used in practice. By this method, more effec-tive probiotic preparations (potentiated probiotics) may be developed.

1. INTRODUCTION

Probiotics are biopreparations containing living cells or metabolites of stabilizedautochthonous microorganisms that optimize the colonization and composition ofgut microflora in both animals and humans, and have a stimulatory effect on diges-tive processes and the immunity of the host (Fuller, 1992). From the viewpoint ofthe practical use of probiotics, it is of particular importance that probiotics have bothlocal and general biomedical effects, an inhibitory effect against pathogens, an opti-mizing effect on digestive processes, an immunostimulative effect, and possiblyeven anti-tumour and cholesterol reducing activities.

19 Enhancement of the efficacy ofprobiotic microorganisms in nutritionand prevention of diseases of the young animal

A. Bomba, R. Nemcová and D. Mudronová

University of Veterinary Medicine, Komenského 73, 041 81 Kosice, SlovakRepublic


Enhancement of efficacy of probiotics 455

Probiotics are being widely used in the food industry, agriculture, and human andveterinary medicine (O’Brien et al., 1999; Shortt, 1999). Their applications includeuses in farm animal nutrition, for feed preservation, for improving the conversion offeed nutrients, and for improving production (Nousiainen and Setälä, 1993). They arealso applied for modulating the functional development of the digestive tract of younganimals (Wallace and Newbold, 1992), and in the prevention and therapy of diseasesin humans and farm animals (Watkins et al., 1982; Bomba et al., 1997). Lactobacillusbulgaricus, L. acidophilus, L. casei, L. helveticus, L. lactis, L. salivarius, L. plantarum,Streptococcus thermophilus, Enterococcus faecium, Enterococcus faecalis,Bifidobacterium spp., and particular strains of E. coli are most frequently used for probiotic purposes. All the above-mentioned microorganisms, except L. bulgaricusand Streptococcus thermophilus, which are starter cultures of yoghurt, form naturalcomponents of the gut microflora (Fuller, 1989).

Despite extensive knowledge obtained in recent years, the mode of action of pro-biotics has not been fully elucidated yet. The mode of inhibitory action of probioticsagainst pathogens may be mediated by competition for receptors on the gut mucosa,competition for nutrients (Freter, 1992), the production of antibacterial substances(Piard and Desmazeaud, 1991), and the stimulation of the immune system (Perdigonand Alvarez, 1992). Probiotics influence digestive processes by the improvement ofthe microbial population beneficial for the macroorganism by enhancing its enzymeactivity, and by improving digestibility and feed utilization (Burgstaller et al., 1984).The optimization of digestion can be exhibited by growth, stimulation effect, andgreater weight increase. The anti-tumour activity of probiotics may be realized inthree ways: a) the inhibition of tumour cells, b) the suppression of bacteria produc-ing beta-glucosidase, beta-glucuronidase, and azoreductase, which catalyse the con-version of procarcinogens to proximal carcinogens, and c) by the destruction ofcarcinogens such as nitrosamines, and by the suppression of nitroreductase activitywhich is involved in their synthesis. Probiotics may influence the blood cholesterollevel by the inhibition of cholesterol synthesis, or decrease its level directly byassimilation (Zacconi et al., 1992).

2. METHODS OF POTENTIATING THE EFFICACY OF PROBIOTICS

Probiotics as natural bioregulators help to maintain the balance of the digestive tractecosystem by a variety of mechanisms and prevent the colonization of the digestivetract by pathogenic bacteria (Ávila et al., 1995). The data concerning the efficacy ofprobiotics in practice are often contradictory (Simmering and Blaut, 2001). In theapplication to pigs of probiotics based on lactobacilli, many authors have recordeda growth and stimulation effect (Pollmann et al., 1980; Nousiainen and Setälä,1993). Some authors, on the other hand, did not observe growth improvement withthe administration of probiotics (Hines and Koch, 1971; Kornegay and Thomas,1973). The data concerning the efficacy of probiotics in the prevention of diarrhoeic

diseases in young animals are contradictory too. The preventive effect of lactobacilliand bifidobacteria against diarrhoea in pigs was confirmed by Maeng et al. (1989),Depta et al. (1998), and Bomba et al. (1998). Some authors, however, have not confirmed the preventive effect of probiotics against diarrhoeic diseases of piglets(Wu et al., 1987; De Cupere et al., 1992; Bekaert et al., 1996).

The variation in efficacy of probiotics under different conditions may be attrib-utable to the probiotic preparation itself or may be caused by external conditions.The most frequent reasons and factors that may lead to variability in the efficacy ofprobiotics are the following: low survival rate and instability of the strain, the use ofa non-specific strain relative to the host, low dose and frequency of administrationof the probiotics, interactions with some medicines, the health and nutritional statusof the animal, stress, genetic differences among animals, age and type of animal.Research experience points to the fact that probiotics are most effective in animalsin the period of microflora development or in cases when physiological stability isimpaired (Stavric and Kornegay, 1995). A production strain of a probiotic should beable to tolerate the conditions of the digestive tract, and, preferably, to adhere inhigh numbers to the digestive tract mucosa; it should maintain high viability in processing, lyophilization, and storage, and should re-vitalize rapidly in the diges-tive tract; it should be able to produce inhibitory substances against pathogens andstimulate the immune system (Chesson, 1993). The production strain must be non-pathogenic. Some of the above-mentioned criteria for the selection of microorgan-isms for probiotic purposes can be tested in vitro, but most of them must be verified in vivo. Some properties of microorganisms observed under laboratory conditionshave not been confirmed in animal trials (Chateau et al., 1993; Bomba et al., 1996).

Despite increasing knowledge gained, the mode of action of probiotics has not beenfully explained yet. In order to enhance the efficacy of probiotics, it is necessary toobtain additional important knowledge on the mechanisms mediating their effect in thedigestive tract (Stavric and Korgenay, 1995). The antibacterial effect of each probioticmicroorganism or its beneficial effect on the host may be mediated by one or a numberof mechanisms that may be expressed to different degrees. This is the starting point forpotentiating the efficacy of probiotics that may be realized either by intensifying one ofthe functions, or by extending the range of functions of the probiotic organism.

The efficacy of probiotics may be enhanced by the following methods:the selection of more efficient strains of a particular microorganism/species

genetic modifications,the combination of a number of strains of one or more species,the combination of probiotics and synergistically acting components.More effective microbial strains can be selected under field conditions. This

procedure is, however, very time consuming, costly, and impracticable. The numberof microorganisms tested under conditions of practice may be reduced by using laboratory tests (Fuller, 1989). This method uses natural properties of the strains.

Gene manipulations can be another way of enhancing the efficacy of probiotics.Using techniques of gene manipulations would make it possible to connect the

456 A. Bomba, R. Nemcova and D. Mudronova

ability of microorganisms to survive in the digestive tract with the ability to producemetabolites that are carriers of probiotic effects (Fuller, 1989).

The advantage of probiotics based on a number of strains consists in their efficacy under a range of conditions and in a variety of animal types. These multiple-strain preparations can exhibit different characteristics (health, productive)typical of these bacteria and, thus, offer a wider spectrum of biomedical effects.Bacteria of the genus Bacillus in multiple-strain preparations may have an effect as potential growth stimulators for Streptococcus and Lactobacillus species(Pollmann et al., 1980). Although Bacillus spp. do not adhere to the intestinalmicrovilli, these bacteria do grow in the mucous biofilm over the intestinal villi andrender the mucus more suitable as a nutrient source for other probiotic bacteria(Porubcan, 1990).

The combination of probiotics with synergistically acting components of naturalorigin seems to be the best way of enhancing the efficacy of probiotic preparationsfrom the practical point of view. It seems that in order to potentiate the effect of probiotics, a number of suitable components may be used such as oligosaccharides,phytocomponents, nutrients and growth factors, proteins, polyunsaturated fattyacids, organic acids and bacterial metabolites (Pollmann et al., 1980; Gibson andRoberfroid, 1995; Yadava et al., 1995). The enhancement of the efficacy of probi-otics by their combination with synergistically acting components is frequently usedin practice.

3. SYNBIOTICS

Future challenges include the incorporation of one or more probiotics together or incombination with suitable prebiotic substrates to enhance the efficacy of the prepa-rations for clinical use (Salminen et al., 1998). A prebiotic is a non-digestible foodingredient that beneficially affects the host by selectively stimulating the growthand/or activity of one or a limited group of bacteria in the colon. In order for a foodingredient to be classified as prebiotic, it must 1) be neither hydrolysed nor absorbedin the upper part of the gastrointestinal tract, 2) be a selective substrate for one or alimited number of beneficial bacteria commensal with the colon, which are stimu-lated to grow and/or are metabolically activated, 3) consequently, be able to alter thecolonic flora in favour of a healthier composition, and 4) induce luminal or systemiceffects that are beneficial to the host’s health (Gibson and Roberfroid, 1995).

Some oligosaccharides comply with all criteria for prebiotics. Oligosaccharidesof various origin are found as natural components in many common foods includingfruit, vegetables, milk and honey. Oligosaccharides can be obtained by: a) extractionfrom plant sources, b) controlled enzymatic hydrolysis of polysaccharides, and c) enzy-matic synthesis. A wide variety of oligosaccharides (fructo-oligosaccharides, gluco-oligosaccharides and galacto-oligosaccharides) is commercially available as feed additives.The supplementation of feed with oligosaccharides improves feed conversion,increases weight gains, and beneficially affects the health of animals (Bastien, 1990).


The mode of action of oligosaccharides may be explained as follows:1. Partial or full resistance to the effects of the host’s digestive enzymes. Not

absorbed or metabolized in the upper part of the digestive tract, and therefore,able to reach the colon, where they may react with microflora and enterocytes.

2. Used as specific growth substrates by beneficial bacteria (lactobacilli,bifidobacteria, streptococci) and not by pathogens or potential pathogens(coliform bacteria, salmonellae, clostridia). Induce the enzyme systems ofuseful bacteria that break them down into organic acids and gases. In this way,good conditions are created for the colonization of the digestive tract by usefulbacteria, while the growth of undesirable microflora is suppressed, which mayreduce the occurrence of diarrhoea (Modler et al., 1990). Higher weight gainsin animals may also be explained by the induction of the enzyme activity of gutbacteria, which may be caused by feeding a limited amount of oligosaccharides.

3. Binding to the surface of bacterial cells thereby inhibiting the adhesion ofvarious bacterial pathogens to epithelial cells (Aniansson et al., 1990) orbinding to protein receptors on the surface of immunocompetent cells of theintestinal mucosa activating the immune response. It is also thought thatoligosaccharides, acting as a soluble fibre, may decrease bacterial translocationand thus promote the preservation of systemic immunity.A way of potentiating the efficacy of probiotic preparations may be the combi-

nation of both probiotics and prebiotics to synbiotics. These may be defined as amixture of probiotics and prebiotics that beneficially affects the host by improvingthe survival and implantation of live microbial dietary supplements in the gastro-intestinal tract, by activating the metabolism of one or a limited number of health-promoting bacteria and/or by selectively stimulating their growth improving thehost’s welfare (Gibson and Roberfroid, 1995). Nemcová et al. (1999) confirmedthe synergistic effect of Lactobacillus paracasei and fructo-oligosaccharide combi-nation on faecal microflora of weaned pigs. This effect was demonstrated byincreased total anaerobes, aerobes, lactobacilli, and bifidobacteria counts, as well asby decreased clostridia, Enterobacteriaceae, and E. coli counts. Kumprecht andZobac (1998) showed that biological preparations containing stabilized livingmicroorganisms, mainly lactic acid bacteria (LAB) as well as extracts of the cellularwall of Saccharomyces cerevisiae (mannan-oligosaccharides) could have positiveeffects on pig growth and feed conversion in addition to a positive effect on health.The combination of probiotics and non-digestible carbohydrates may be a way ofstabilizing and/or potentiating the effect of probiotics.

4. POTENTIATED PROBIOTICS

Synbiotics appear to be preparations whose potentiated protective and stimulativeeffects occur only in the colon. Taking into account the pathogenesis of diarrhoeicdiseases in young animals there is a need for protecting the digestive tract mucosa


throughout its length, i.e. also in the small intestine so that the adhesion of patho-genic microorganisms can be prevented. In various disorders of the gastrointestinaltract, there occur conditions for the translocation of conditioned pathogens orpathogens from the colon into the front part of the digestive tract. Potentiated probiotics are defined as biopreparations containing production strains of micro-organisms and synergistically acting components of natural origin which amplifytheir probiotic effect on both the small intestine and the colon, and their beneficialeffect on the host by intensifying a mechanism or by extending the range of theirprobiotic action. Potentiated probiotics must comply with the criteria as follows: a) they must be more effective than their components separately, b) their potentiatedprotective and stimulative effects must be expressed in all parts of the digestive tract(table 1).

On the basis of the above-mentioned criteria, a synbiotic could be regarded as apotentiated probiotic, provided a component potentiating the probiotic effect on thesmall intestine is added. From this it also follows that potentiated probiotics willprobably be multicomponent preparations.

5. COMBINATION OF PROBIOTIC MICROORGANISMS WITH NON-SPECIFIC SUBSTRATES

Prebiotics are specific substrates selectively fermented in the colon. It has beendemonstrated that to enhance the efficacy of probiotics, non-specific substrates canbe used as well.

Hawley et al. (1969) suggested that large quantities of lactose are necessary forlactobacilli to become established in the gut. Pollmann et al. (1980) studied theeffect of Lactobacillus acidophilus on starter pigs fed a diet supplemented withlactose. Pigs receiving lactose in combination with the Lactobacillus inoculum hadthe highest Lactobacillus counts and the best average daily gain in comparison withother groups. The effect of caecal flora, cultured in lactose-based broths, againstSalmonella was enhanced by adding lactose to chick diets. The reduction in caecalcolonization was accompanied by an increase in the concentration of volatile fattyacids and a decrease in the caecal pH (Corrier et al., 1990).


Table 1. The localization of protective and stimulative effects in the digestive tract

Preparation Small intestine Colon

Probiotic + +Prebiotic − +Synbiotic + + +Potentiated probiotic + + + +

+, protective and stimulative effect.+ +, potentiated protective and stimulative effect.−, no effect.

The combination of peptides and lactobacilli reduced mortality following diar-rhoea, halved the incidence of digestive disorders and improved animal growth significantly. While peptides or LAB alone improved animal productivity, theircombination resulted in a synergy of action (Lyons, 1987). Mordenti (1986) foundthat the growth promoting effect which he obtained by feeding Enterococcusfaecium to pigs could be improved synergistically by the addition of whey peptides.

The use of whey and bovine blood plasma as non-specific substrates could be analternative processing route for probiotic production and/or for potentiating theireffect. Whey, an abundant by-product of the dairy industry, contains lactose, solubleproteins, lipids, vitamins and mineral salts. It is used directly in animal feedmixtures (Burnell et al., 1988). Supplementation of whey in a diet for turkey poultsenhanced the effect of Lactobacillus reuteri by increasing body weight gain andresistance to salmonellae (Edens et al., 1991).

Bury et al. (1998) reported the use of whey protein concentrate as a nutrient supplement for LAB. The addition of this protein significantly increased both thecell numbers and lactic acid production by lactobacilli and streptococci. The wheyproteins, α-lactalbumin and β-lactoglobulin, were found to be excellent growth promoters of bifidobacteria (Petschow and Talbott, 1990). The proteolytic enzymesof the lactic acid bacteria are of great importance in the generation of free aminoacid and vitamins in fermented whey (Law and Haandrikman, 1997). More recently,it has been recognized that several of the whey proteins confer antibacterial andimmuno-associated protection to the neonate against disease and that these andother whey proteins also have putative biological effects when ingested, includingan anti-cancer action (McIntosh et al., 1998). Romond et al. (1998) suggested thatconsumption of cell-free whey from a selected strain of bifidobacteria is capable ofmodifying the ecosystem of the human intestine.

The effect of a preparation containing nutrients and growth factors which stimu-late rumen bacteria was investigated in suckling lambs. The preparation containeddried skim milk, glucose, ascorbic acid, nutritious broth, liver hydrolysate, andenzymatic casein hydrolysate. A higher molar proportion of propionate and valerateand a higher alpha-amylase activity in rumen fluid were observed in the lambs ofthe experimental groups. The counts of streptococci and lactobacilli in the rumencontent and of those adhering to the rumen epithelium were significantly higher inthe experimental animals, too. Scanning electron microscopy showed longer andmore differentiated rumen papillae in an experimental lamb at the age of 4 weeks incomparison with lambs of the control group (Bomba and Zitnan, 1993).

Bomba et al. (1999) investigated the influence of the preventive administrationof Lactobacillus casei subsp. casei and maltodextrin KMS X-70 on Escherichia coli08:K88 adhesion in the gastrointestinal tract of conventional and gnotobiotic piglets. L. casei alone had almost no inhibitory effect on the adherence of E. coli tothe jejunal mucosa of gnotobiotic and conventional piglets while the lactobacilliadministered together with maltodextrin decreased the jejunal mucosa E. coli counts


of gnotobiotic piglets by 1 logarithm (4.95 log/cm2) in comparison with the controlgroup (5.96 log 10/cm2) (fig. 1). L. casei administered in combination with malto-dextrin decreased the number of E. coli colonizing the jejunum of conventional pigletsby more than two and a half logarithms (4.75 log 10/cm2, P < 0.05) in comparison withthe control (7.42 log 10/cm2, fig. 2). The pH values of jejunum contents were lowestin the group of gnotobiotic and conventional piglets administered with L. casei andmaltodextrin KMS X-70.

6. PROBIOTICS AND METABOLIC PRODUCTS OF MICROORGANISMS

Lactic acid bacteria – among them many probiotics – have been found to produceantimicrobial substances (Ouwehand et al., 1999). They include toxic metabolites ofoxygen, the lactoperoxidase-thiocyanate system, organic acids, the effect of pH, andbacteriocins (Nemcová, 1997).

The ability to generate organic acids, particularly lactic and acetic acids, presentsone of the mechanisms by which lactobacilli perform their inhibitory effect uponpathogens (Piard and Desmazeaud, 1991). Organic acids together with probiotics


Fig. 1. The numbers of E. coli 08:K88adhered to the jejunal mucosa in 7-day-old gnotobiotic pigs after administration ofLactobacillus casei and maltodextrin KMS X-70.(■) Group E: E. coli 08:K88.( ) Group L: Lactobacillus casei + E. coli

08:K88.(� ) Group M: Lactobacillus casei + maltodex-

trin KMS X-70 + E. coli08:K88.

Fig. 2. The numbers of E. coli 08:K88adhered to the jejunal mucosa in 7-day-old conventional pigs after administration ofLactobacillus casei and maltodextrin KMS X-70.(■) Group E: E. coli 08:K88.( ) Group L: Lactobacillus casei + E. coli

08:K88.(�) Group M: Lactobacillus casei + maltodex-

trin KMS X-70 + E. coli 08:K88.

and specific carbohydrates (yeast cell walls) are often suggested as alternatives tothe use of antibiotic growth promoters (Jensen, 1998). A few studies have attemptedto correlate changes in the gastrointestinal ecosystem in response to diet acidifica-tion. Sutton et al. (1991) reported that fumaric acid decreased E. coli numbers in thestomach. Gedek et al. (1992) observed a decrease of several enteric bacteria includingE. coli in the gastrointestinal tract of weaned piglets fed a diet supplemented withfumaric acid. Bolduan et al. (1998) and Kirchgessner et al. (1992) found that formicacid decreased the population of coliform bacteria in the gastrointestinal tract ofweaned piglets. Propionic acid reduced the population of coliform bacteria in thestomach (Bolduan et al., 1988) and in the small intestine (Cole et al., 1968).Thomlinson and Lawrence (1981) found that the addition of 1% lactic acid to thedrinking water of creep-fed piglets significantly decreased the gastric pH, delayedthe multiplication of enterotoxigenic E. coli and lowered the mortality rate.

An alternative to the use of organic acids in combination with probiotics in pigdiets is the use of fermented feed. Under certain conditions probiotic strains may beused as the sole fermenting agent in milk. However, in many cases use of a supportculture is preferable. The combination of the probiotic culture and the support culture enhanced the acidification rate (Saxelin et al., 1999). Fermented liquid feedis characterized by high LAB and yeast counts, and a high concentration of lacticacid. Fermented milks are claimed to contain a number of biologically active components. These include bacteria used for fermentation, their metabolic products,and components derived from milk. Milk contains a large variety of proteins andpeptides. These bioactive peptide properties include antimicrobial, anti-cancer, antihypertensive, immunomodulatory, and mineral carriers (Meisel, 1997). Feedingmilk fermented with Lactobacillus casei and Lactobacillus acidophilus to miceresulted in an increased resistance to Salmonella typhimurium. Perdigon et al.(1991) and Nader de Macías et al. (1993) described the immunostimulant effect offermented milk (lymphokines and macrophages) responsible for the elimination ofthe pathogens from the liver and spleen in E. coli and Listeria challenged mice. LABused in fermented milk alter the immunogenic properties of milk proteins (Perdigonet al., 1986). The administration of milk fermented with Lactobacillus acidophilusLA-2 caused a remarkable decrease (71.9% on the average) in faecal mutagenicityand increased Lactobacillus spp. and Bifidobacterium spp. populations in the humanintestine (Hosoda et al., 1996).

Several investigations have shown fermented liquid feed to improve growth performance in pigs and to establish a prophylactic barrier against gastrointestinaldisorders. Another way of administering organic acids with probiotics would be touse water as a vehicle. It appears to have given more consistent advantages than dietacidification. Because the volume of water intake is about 2.5-fold higher than feedintake, and especially because newly weaned piglets take in high amounts of water,water acidification provides a greater scope for delivering higher and more regulardoses of acids to piglets (Jensen, 1988).


The in vitro bactericidal activity of certain fatty acids has been known for a longtime (Kanai and Kondo, 1979), and may be of importance in vivo in preventing,eliminating, and, in some cases, treating infections. The number of LAB associatedwith the epithelial mucosa and from faeces was highest in the digestive tract of fishthat were fed diets with added 7% linolic acid (18:3 a-3) or 4% of a polyunsaturatedfatty acids mix. It is suggested that dietary fatty acids affect the attachment sites forthe intestinal microbiota, possibly by modifying the fatty acid composition of theintestinal wall (Ringø and Gatesoupe, 1998).

Bacteriocins, which are proteinaceous compounds produced by some strains ofLAB, inhibit the growth of Gram-positive bacteria including food spoilage organ-isms and pathogens and are expected to be used increasingly as natural food preser-vatives (Dykes, 1995). Several methods for the control of diarrhoeagenic E. colihave been described (Giese, 1994). The use of biopreservatives, including bacteri-ocins, has been proposed for the control of many food-borne pathogens. Nisin, a bacteriocin from Lactococcus lactis subsp. lactis, which acts exclusively againstselected Gram-positive bacteria, has served as a model for the application of otherbacteriocins as food biopreservatives. Colicins are the classical bacteriocins produced by E. coli, which specifically inhibit E. coli and closely related strains.There is potential for using colicins in foods and agriculture to inhibit sensitive diarrhoeagenic E. coli strains (Murinda et al., 1996). Colicin-sensitive cells havecolicin-specific receptors located in the outer membrane of their cell envelope andare killed by colicins that attach to these receptors. It is suggested that combiningprobiotic microorganisms with bacteriocins could improve their positive effecton the host.

7. COMBINATION OF PROBIOTIC MICROORGANISMS AND PLANT EXTRACTS

Mitchell and Kenworthy (1976) considered that LAB might prevent coliform diarrhoea by interaction with the enterotoxins. Such interaction might be indirect byinfluencing E. coli populations or metabolism or might be more direct by neutralizingthe enterotoxin itself. Eleven species of lactic bacteria have been investigated for theability to neutralize cell-free enterotoxins, in a bio-assay. Action against the E. coli enterotoxin was found in two species, L. bulgaricus and L. faecalis. Part ofthe activity in the former species was in the cell-free fraction. Further investigationof the cell-free anti-enterotoxic action from L. bulgaricus showed that it had a lowmolecular weight, probably less than 103, was not very stable, and was independentof the anti-E. coli activity. Broths containing L. bulgaricus fermented to produce highlevels of anti-enterotoxin were beneficial when added to diets for early weaned pigs.It was inferred that this effect was likely to be caused by the anti-enterotoxic action.

Plant extracts seem to have a similar ability, which might be used in potentiatingthe neutralization effect of lactobacilli against enterotoxin-producing E. coli.


A crude alcohol extract of Coleus froskohlii Briq. roots showed a marked inhibitoryaction against an E. coli toxin-induced secretory response at a dose of 300 mg/loopin ileal loops of rabbits and guinea pigs. Coleonol A and B obtained by column chro-matography of a benzene fraction of the alcohol extract of Coleus froskohlii rootsexhibited antisecretory (antidiarrhoeal) action at a dose of 1 mg/loop. Coleonol Ashowed nearly 79% inhibition against heat-labile (LT) toxin and 67% against heat-stable (ST) toxin-induced secretions. Coleonol B exhibited less activity against LT(66.2%) but slightly higher activity against ST (68.3%) enterotoxins as comparedwith the widely used antidiarrhoeal (antisecretory) drug Loperamide (an opiate ago-nist) which showed 77 and 65% inhibition against LT and ST toxins, respectively,at a similar dose (Yadava et al., 1995).

It is interesting to consider potentiating the efficacy of probiotics in the controlof the post-weaning diarrhoea syndrome of piglets. Weaned piglets usually show amalabsorption syndrome known as non-infectious diarrhoea which is characterizedby increased excretion of fatty acids and carbohydrates in the faeces, watery stools,and degenerative changes in villi of the small intestine (Kyriakis, 1989). In themajority of these cases, opportunistic pathogens, particularly enterotoxigenicEscherichia coli (ETEC) strains and rotaviruses, take advantage of the presence ofthis diarrhoea and cause the post-weaning diarrhoea syndrome (PWDS). The essen-tial oils derived from the plant Origanum were proved to have in vitro antimicrobialaction against various bacteria, including E. coli (Sivropoulou et al., 1996). Kyriakiset al. (1998) studied the possible effect of two different dosages in feed applicationsof 5% of etherous oils of flower and leaf of Origanum on the control of post-weaning colibacillosis in piglets. The medication with the Origanum essential oilsseemed to be effective in the control of PWDS, resulting in a mild atypical illnessin the animals combined with very good growth performance. The unique conclu-sion of this study is the discovery of the possibility of PWDS control without the useof “classical” antibacterials. Within the first days after weaning, the E. coli popula-tion in weaned pigs increases and the lactobacilli population in the digestive tractdecreases. That is why the application of probiotics in combination with plant components exhibiting antimicrobial effects may be useful.

Babic et al. (1994) demonstrated that purified ethanolic extracts of carrots had anantimicrobial effect against a range of food-borne microorganisms: Leuconostocmesenteroides, Listeria monocytogenes, Staphylococcus aureus, Pseudomonasfluorescens, Candida albicans and Escherichia coli. The antimicrobial activity wasnot linked to phenolic compounds, but was presumably due to apolar components.Free dodecanoic acid and methyl esters of dodecanoic and pentadecanoic acids wereidentified in the purified active extracts of carrots and could be responsible for theantimicrobial activity. An extract from the cortex of the African Okoubaka tree hasbeen successfully used for the treatment of enterotoxin-induced diarrhoea in horses.

It has been shown that some phytochemical components stimulate the productionof lactic acid by lactobacilli. Nakashima and Yoshikai (1980) have shown that


phytins, including phytic acid (a naturally occurring compound formed during thematuration of seeds and cereal grains) stimulated the growth of LAB in a skim milkmedium as measured by the number of live bacteria and the amount of acid produced. Nakashima (1997) reported that the stimulating effect of the phytin prepa-ration on acid production by Lactobacillus casei is not attributable to phytin, but isexerted mainly by the Mn in the preparation, while the presence of other inorganicmaterials also augments the effect to some extent. Similar results were reported forgreen laver and sea lettuce.

8. PROBIOTICS AND TRACE ELEMENTS

Some microbes have the ability to bind metal ions present in the external environ-ment at the cell surface or to accumulate them in the cell. These properties have beenexploited also in probiotic preparations.

Selenium: yeasts and lactobacilli (Lactobacillus delbrueckii subsp. bulgaricus,L. casei subsp. casei, L. plantarum) (Calomme et al., 1995; Ingledew, 1999) are ableto concentrate selenium from their growth media into their cells in high concentra-tions and both produce an organic form of Se from inorganic Se. Yeasts produceabove all selenomethionine, as well as a wide variety of Se compounds, many ofwhich are unidentified (Wolffram, 1999). Lactobacilli contain a mixture of Se invarious chemical forms too, but their main product is selenocysteine (Calomme et al., 1995). Recently, Se-enriched yeasts and lactobacilli have been commercial-ized as a Se supplement, which is much more bio-available for useful metabolismand storage than an inorganic salt (Calomme et al., 1995; Ingledew, 1999).

Chromium: trivalent chromium is the active constituent of the glucose tolerancefactor (GTF), which is a cofactor needed to potentiate insulin. Yeast is able to pro-duce GTF and thus to serve as a feed supplement of the bioactive and non-toxicform of chromium (Ingledew, 1999).

Iron: the position of iron in bacterial metabolism is specific. Iron is an essentialnutrient for all microbes, except the genus Lactobacillus, which uses manganese andcobalt instead. On the other hand, iron can react with reduced forms of oxygen,leading to the production of free radicals responsible for the peroxidative alterationof cell membranes, and therefore iron homeostasis has to be strictly controlled.Bifidobacteria are capable of accumulating large amounts of Fe2+ from a mediumand they contain an intracellular ferro-oxidase, which in the presence of O2

catalyses the oxidation of Fe2+ to insoluble and less available Fe3+. Lactobacillus delbrueckii spp. bulgaricus and L. acidophilus also have the ability to oxidize Fe2+

and to bind Fe(OH)3 to their cell surfaces (Kot et al., 1997). The binding of iron bybifidobacteria and lactobacilli, used extensively as probiotics, can serve to reducethe availability of iron to pathogenic microorganisms.

Manganese: certain species of lactobacilli (e.g. L. plantarum, L. casei subsp.casei) are able to concentrate high levels of manganese, which supports their


pseudocatalase and superoxide dismutase activity. These strains are more oxygen-and hydrogen peroxide-radical tolerant. The addition of Mn to the growth mediumincreases the viable count of L. casei spp. casei and L. plantarum compared tounsupplemented cultures (Calomme et al., 1995).

Zinc: zinc is important for both the stability and action of alcohol dehydrogenaseas well as for many other microbial Zn-metallo-enzymes. However, zinc’s antimi-crobial effects are well known. Zinc has inhibited the growth of Escherichia coli,Streptococcus faecalis, some species of soil bacteria, and fungi (Collins and Stotzky,1989). This inhibiting effect of zinc has been used successfully in the treatment ofE. coli diarrhoea in post-weaning piglets (Holm and Poulsen, 1996).

9. COMBINATION OF PROBIOTIC MICROORGANISMS AND ANTIBIOTICS

Probiotics are often considered as “natural” substitutes for feed antibiotics. On thebasis of the results of some studies, it may be feasible to combine the probiotic andantibiotic treatments to obtain an additive advantage (Stavric and Kornegay, 1995).It has been suggested that the natural microflora resist the invasion of both harmfuland probiotic bacteria. Therefore, the probiotic bacteria may be established moreeasily in the digestive tract of animals if the natural flora is weakened by the use ofan antibacterial feed additive (Nousiainen and Setälä, 1993). Pollmann et al. (1980)found an additive effect in the combination of lincomycin and Lactobacillus spp.

The disruption of the ecological balance of the gastrointestinal tract by antibioticsapplied per os and by other immunosuppressive drugs enables bacteria to over-multiply, and facilitates the translocation of these bacteria from the gastrointestinaltract into other organs. Prophylactic use of lactobacilli-containing preparations mayprotect against the side effects of antibiotics. The principal risk of the therapeutic useof antibiotics consists in bacterial translocation with potential subsequent septicaemiacaused by an increase in the number of resistant species and a decrease in the numberof more sensitive bacteria. Each antibiotic is associated with an increase in the number of specific microorganisms (Lambert-Zechovsky et al., 1984). The applicationof L. casei, L. acidophilus and L. bulgaricus in conjunction with antibiotics preventedboth the increase in the number of ampicillin-resistant bacteria and their translocationinto the liver. Wasson et al. (2000) suggested that oral administration of aLactobacillus-containing product is ineffective in preventing clinical disease in guineapigs administered clindamycin phosphate. The gastrointestinal tract microflora greatlyinfluences the numbers and distribution of lymphoid cells from lymphatic tissue associated with the digestive tract. The impairment of this microflora interferes withthe regulation of the gut mucosa’s immune response (Porter and Allen, 1989).Lactobacilli, when applied with antibiotics, may eliminate the impairment of the gutmicroecosystem by maintaining the equilibrium between microbial types in the gut,thus preventing the impairment of the mucosa’s immune response.


10. COMBINATION OF PROBIOTICS AND VACCINES

Methner et al. (1999) tested the combination of competitive exclusion (CE) and vaccination against Salmonella infection in chickens of different ages. The results oftheir model study demonstrated principally that a combination of CE and immunizationagainst Salmonella infection results in a degree of protection considerably beyond thatafforded by the use of one of the two methods. The administration of the live Salmonellavaccine strain prior to or simultaneously with the CE culture revealed the best protec-tive effect, because these combinations ensure an adequate persistence of the vaccinestrain as a prerequisite for the expression of inhibitory effects in very young chicks andthe development of strong immune protection in older birds. The combination ofLactobacillus acidophilus-, Enterococcus faecium-, and S. typhimurium-specific anti-bodies when administered via spray to broiler chicks at 1 day of age, and administeredorally for the first 3 days after placement, significantly reduced S. typhimurium colo-nization of the caecum and colon in market-aged broilers (Promsopone et al., 1998).


If probiotics are to represent a real and effective alternative to antibiotics andchemotherapeutics, it is absolutely necessary to ensure their consistently high effi-cacy. To ensure the high efficacy of probiotic preparations requires a complex solu-tion aimed at the product and its mode of application.

Regarding the product itself, future research should be aimed at the selection ofstrains with strong probiotic effects that will comply with the main criteria of selection.It will be important to search for ways to potentiate the efficacy of probiotic micro-organisms in all parts of the digestive tract. In addition to prebiotics which potentiatethe effect of probiotics in the colon, there should be components that, in combinationwith probiotic preparations, will ensure their high efficacy in the small intestine also.

With regard to the form of application, research and development should beaimed at methods that will ensure the maximum efficacy of probiotics at the time oftheir consumption. Current knowledge confirms that probiotic preparations as wellas fermented products are most effective in a fresh state when given together with aparticular medium or substrate. On the basis of current knowledge it may beexpected that the developments in the field of biotechnological research will resultin a very simple “fermentor” which will enable customers to prepare probioticpreparations and fermented products directly.

REFERENCES

Aniansson, G., Andersson, B., Lindstedt, R., Svanbrog, C., 1990. Anti-adhesive activity of human caseinagainst Streptococcus-pneumoniae and Haemophilus-influenzae. Microb. Pathog. 8, 315–324.

Ávila, F.A., Paulillo, A.C., Schocken-Iturrino, R.P., Lucas, F.A., Orgaz, A., Quintana, J.L., 1995. A com-parative study of the efficiency of a probiotic and the anti-K99 and anti-A14 vaccines in the controlof diarrhea in calves in Brazil. Revue Élev. Méd. Vét. Pays trop. 48, 239–243.


Babic, I., Nguyen-the, C., Amiot, M.J., Aubert, S., 1994. Antimicrobial activity of shredded carrotextracts on food-borne bacteria and yeast. J. Appl. Bacteriol. 76, 135–141.

Bastien, R., 1990. New regulators of flora. Rech. Alim. Anim. 434, 56–57.Bekaert, H., Moermans, R., Eeckhout, W., 1996. Effects of a live levure culture (Levucell SB2) in a feed-

ing mixture for growing piglets upon animal performance and the frequency of diarrhea. Ann.Zootech. 45, 369–376.

Bolduan, G., Jung, H., Schneider, R., Block, J., Klenke, B., 1998. The effects of propionic and formicacid in pig rearing. J. Anim. Physiol. Anim. Nutr. 59, 72–78.

Bomba, A., Zitnan, R., 1993. Development and manipulation of the digestion in young ruminants.Veterinárství 43, 13–14.

Bomba, A., Kastel’, R., Gancarcíková, S., Nemcová, R., Herich, R., Cízek, M., 1996. The effect oflactobacilli inoculation on organic acid levels in the mucosal film and the small intestine contents ingnotobiotic pigs. Berl. Munch. Tierärztl. Wschr. 109, 428–430.

Bomba, A., Kravjansky, I., Kastel’, R., Herich, R., Juhásová, Z., Cízek, M., Kapitancík, B., 1997.Inhibitory effects of Lactobacillus casei upon adhesion of enterotoxigenic Escherichia coli K 99 tothe intestinal mucosa in gnotobiotic lambs. Small Ruminant Res. 23, 199–209.

Bomba, A., Gancarcíková, S., Nemcová, R., Herich, R., Kastel’, R., Depta, A., Demeterová, M., Ledecky, V.,Zitnan, R., 1998. The effect of lactic acid bacteria on intestinal metabolism and metabolic profile ofgnotobiotic pigs. Dtsch. Tierärztl. Wschr. 105, 384–389.

Bomba, A., Nemcová, R., Gancarcíková, S., Herich, R., Kastel’, R., 1999. Potentation of the effective-ness of Lactobacillus casei in the preventation of E. coli induced diarrhea in conventional andgnotobiotic pigs. In: Paul, P.S., Francis, D.H. (Eds.), Mechanisms in the Pathogenesis of EntericDiseases 2. Kluwer Academic/Plenum Publishers, New York, pp. 185–190.

Burnell, T.W., Cromwell, G.L., Stahly, T.S., 1988. Effects of dried whey and copper sulfate on the growthresponses to organic acid in diets for weanling pigs. J. Anim. Sci. 66, 1100–1108.

Burgstaller, G., Ferstl, R., Apls, H., 1984. Contribution to the addition of lactic acid bacteria(Streptococcus faecium SF-68) in milk replacers for fattening calves. Zuchtungskunde 56, 156–162.

Bury, D., Jelen, P., Kimura, K., 1998. Whey protein concentrate as a nutrient supplement for lactic acidbacteria. Int. Dairy J. 8, 149–151.

Calomme, M.R., Van Den Branden, K., Van Den Berghe, D.A., 1995. Selenium and Lactobacillusspecies, J. Appl. Bacteriol. 79, 331–340.

Chateau, N., Castellanos, I., Deschamps, A.M., 1993. Distribution of pathogen inhibition in theLactobacillus isolates of a commercial probiotic consortium. J. Appl. Bact. 74, 36–40.

Chesson, A., 1993. Probiotics and other intestinal mediators. In: Cole, D.J.A., Wisiman, J., Varley, M.A.(Eds.), Principles of Pig Science. Loughborough, Nottingham University Press, pp. 197–214.

Cole, D.J.A., Beal, R.M., Luscombe, J.R., 1968. The effect on performance and bacterial flora of lacticacid, propionic acid, calcium propionate and calcium acrylate in the drinking water of weaned pigs.Vet. Res. 83, 459–464.

Collins, Y.E., Stotzky, G., 1989. Factors affecting the toxicity of heavy metals to microbes. In: Beveridge, T.J.,Doyle, R.J. (Eds.), Metal Ions and Bacteria. Wiley, New York, p. 461.

Corrier, D.E., Hinton, A., Ziprin, R.L., Beier, R.C., De Loach, J.R., 1990. Effect of dietary lactose on cecalpH, bacteriostatic volatile fatty acids and Salmonella typhimurium colonization of broiler chickens.Avian Dis. 34, 617–625.

De Cupere, F., Deprez, P., Demeulenaere, D., Muylle, E., 1992. Evaluation of the effect of 3 probioticson experimental Escherichia coli enterotoxaemia in weaned piglets. J. Vet. Med. B 39, 277–284.

Depta, A., Rychlik, R., Nieradka, R., Rotkiewicz, T., Kujawa, K., Bomba, A., Grabowska-Swiecicka, G.,1998. The influence of alimentary tract colonisation with Lactobacillus sp. strains on chosen meta-bolic profile indices in piglets. Polish J. Vet. Sci. 1–7.

Dykes, G.A., 1995. Bacteriocins: ecological and evolutionary significance. Tree 10, 186–189.Edens, F.W., Parkhurst, C.R., Casas, I.A., 1991. Lactobacillus reuteri and whey reduce salmonella colo-

nization in the ceca of turkey poults. Poultry Sci. 70, Suppl. 1, 158.Freter, R., 1992. Factors affecting the microecology of the gut. In: Fuller, R. (Ed.), Probiotics: The

Scientific Basis. Chapman and Hall, London, pp. 111–144.Fuller, R., 1989. Probiotics in man and animals. J. Appl. Bacteriol. 66, 365–378.


Fuller, R., 1992. The effect of probiotics on the gut microbiology of farm animals. In: Wood, B.J.B. (Ed.),The Lactic Acid Bacteria. Elsevier Applied Science, London, pp. 171–192.

Gedek, B., Roth, F.X., Kirchgessner, M., Wiehler, S., Bott, A., Eidelsburger, U., 1992. The effects of furmaric and hydrochloric acid, sodium formiate, tylosine and toyocerine upon germ counts of themicroflora and its composition in different segments of the gastrointestinal tract. J. Anim. Physiol.Anim. Nutr. 68, 209–217.

Gibson, G.R., Roberfroid, M.B., 1995. Dietary modulation of the human colonic microbiota: introducingthe concept of prebiotics. J. Nutr. 125, 1401–1412.

Giese, J., 1994. Antimicrobials assuring food safety. Food Technol. 48, 102–110.Hawley, H.B., Shepherd, P.A., Wheater, D.M., 1969. Factors affecting implantation of lactobacilli in the

intestine. J. Appl. Bacteriol. 22, 360.Hines, R.H., Koch, B.A., 1971. Response of growing and finishing swine to a dietary source of

Lactobacillus acidophilus. Kans. Agr. Exp. St. Prog. Rep. Manhattan, Kansas 181, 32–36.Holm, A., Poulsen, H.D., 1996. Zinc oxide in treating E. coli diarrhea in pigs after weaning. Comp. Cont.

Educ. Pract. Veter. 18, 26–30.Hosoda, M., Hashimoto, H., He, F., Morita, H., Hosono, A., 1996. Effect of administration of milk

fermented with Lactobacillus acidophilus LA-2 on fecal mutagenicity and microflora in the humanintestine. J. Dairy Sci. 79, 745–749.

Ingledew, W.M., 1999. Yeast – could you base a business on this bug? In: Lyons, T.P., Jasques, K.A.(Eds.), Biotechnology in the Feed Industry, Proceedings of Alltech’s 15th Annual Symposium.University Press, Nottingham, pp. 547–566.

Jensen, B.B., 1988. Effect of diet composition and Virginiamycin on microbial activity in the digestive tractof pigs. In: Buraczewska, L., Buraczewski, S., Pastuszewska, B., Z

.ebrowska, T. (Eds.), Proceedings of

the 4th Symposium on Digestive Physiology in the Pig. Institute of Animal Physiology and Nutrition,Jablonna, Poland, pp. 392–400.

Jensen, B.B., 1998. The impact of feed additives on the microbial ecology of the gut in young pigs.J. Anim. Feed Sci. 7, Suppl. 1, 45–64.

Kanai, K., Kondo, E., 1979. Antibacterial and cytotoxic aspects of long-chain fatty acids as cell surfaceevents: selected topics. Japan J. Med. Sci. Biol. 32, 135–174.

Kirchgessner, M., Gedek, B., Wiehler, S., Bott, A., Eidelsburger, U., Roth, F.X., 1992. The effects offormic acid, calcium formiate and sodium hydrogen carbonate upon germ counts of the microflora andits composition in different segments of the gastrointestinal tract. 10th communication: Investigationinto the nutritive effectivity of organic acids in the rearing of piglets. J. Anim. Physiol. Anim. Nutr.68, 73–81.

Kornegay, E.T., Thomas, H.R., 1973. Bacterial and yeast preparations for starter and grower rations. V.P.I.and S.U. Res. Div. Rep. Blacksburg, Virginia, 151, 1–9.

Kot, E., Furmanov, S., Bezkorovainy, A., 1997. Binding of Fe(OH)3 to Lactobacillus delbrueckii subsp.bulgaricus and L. acidophilus: Apparent role of hydrogen peroxide and free radicals. J. Agr. FoodChem. 45, 690–696.

Kumprecht, I., Zobac, P., 1998. Study of the effect of a combined preparation containing Enterococcusfaecium M-74 and mannanoligosaccharides in diets for weanling piglets. Czech. J. Anim. Sci. 43, 477–481.

Kyriakis, S.C., 1989. New aspects of the prevention and/or treatment of the major stress induced diseasesof the early weaned piglet. Pig News Info. 2, 177–181.

Kyriakis, S.C., Sariss, K., Lekkas, S., Tsinas, A.C., Giannakopoulos, C., Alexopoulos, C., Saoulidis, K.,1998. Control of post weaning diarrhoea syndrome of piglets by in feed application of Origanumessential oils. In: Proceedings of the 15th International Pig Veterinary Society Congress, Vol. III:Poster Presentations, Birmingham, UK, p. 218.

Lambert-Zechovsky, N., Bingen, E., Bourrillon, A., Aujard, Y., Mathieu, H., 1984. Effects of antibioticson the microbial intestinal ecosystem. Dev. Pharmacol. Ther. Suppl. 1, 150–157.

Law, J., Haandrikman, A., 1997. Proteolytic enzymes of lactic acid bacteria. Internat. Dairy J. 7, 1–11.Lyons, T.P., 1987. Probiotics: an alternative to antibiotics. Pig News Info. 8, 157–164.Maeng, W.J., Kim, C.W., Shin, H.T., 1989. Effect of feeding lactic acid bacteria concentrate (LBC,

Streptococcus faecium Cernelle 68) on the growth rate and prevention of scouring in piglets. KoreanJ. Anim. Sci. 31, 318.


McIntosh, G.H., Royle, P.J., Le Leu, R.K., Regester, G.O., Johnson, M.A., Grinsted, R.L., Kenward, R.S.,Smithers, G.W., 1998. Whey proteins as functional food ingredients? Int. Dairy J. 8, 425–434.

Meisel, H., 1997. Biochemical properties of bioactive peptides derived from milk proteins: potentialnutraceuticals for food and pharmaceutical applications. Livest. Prod. Sci. 50, 125–138.

Methner, U., Barrow, P.A., Berndt, A., Steinbach, G., 1999. Combination of vaccination and competitiveexclusion to prevent salmonella colonisation in chickens: experimental studies. Int. J. FoodMicrobiol. 49, 35–42.

Mitchell, I. De G., Kenworthy, R., 1976. Investigations on a metabolite from Lactobacillus bulgaricuswhich neutralises the effect of enterotoxin from Escherichia coli pathogenic for pigs. J. Appl.Bacteriol. 41, 163–174.

Modler, H.W., McKellar, R.C., Yaguchi, M., 1990. Bifidobacteria and bifidogenic factors. Can. Inst. FoodSci. Technol. J. 23, 29–41.

Mordenti, A., 1986. Probiotics and new aspects of growth promoters in pig production. Informat.Zootechnol. 32(5), 69.

Murinda, S.E., Roberts, R.F., Wilson, R.A., 1996. Evaluation of colicins for inhibitory activity againstdiarrheagenic Escherichia coli strains, including serotype O157:H7. Appl. Environ. Microbiol. 3196–3202.

Nader de Macías, M.E., Romero, N.C., Apella, M.C., González, S.N., Oliver, G., 1993. Prevention ofinfections produced by Escherichia coli and Listeria monocytogenes by feeding milk fermented withlactobacilli. J. Food Protect. 56, 401–405.

Nakashima, A., 1997. Stimulatory effect of phytin on acid production by Lactobacillus casei. J. Nutr. Sci.Vitaminol. 43, 419–424.

Nakashima, A., Yoshikai, F., 1980. On the growth stimulating activities of various animal and vegetablematerials. Fukuoka Joshi Tandai Kiyó (Fukuoka Women’s Junior Coll Stud.) 20, 11–19.

Nemcová, R., 1997. Selection criteria of lactobacilli for probiotic use. Vet. Med.-Czech. 42, 19–27.Nemcová, R., Bomba, A., Gancarcíková, S., Herich, R., Guba, P., 1999. Study of the effect of

Lactobacillus paracasei and fructo-oligosaccharides on the faecal microflora in weanling piglets.Berl. Munch. Tierärztl. Wschr. 112, 225–228.

Nousiainen, J., Setälä, J., 1993. Lactic acid bacteria as animal probiotics. In: Salminen, S., von Wright, A.(Eds.), Lactic Acid Bacteria. Marcel Dekker, New York, pp. 315–356.

O’Brien, J., Crittenden, R., Ouwehand, A.C., Salminen, S., 1999. Safety evaluation of probiotics. TrendsFood Sci. Technol. 10, 418–424.

Ouwehand, A.C., Isolauri, E., Kirjavainen, P.V., Salminen, S.J., 1999. Adhesion of four Bifidobacteriumstrains to human intestinal mucus from subjects in different age groups. FEMS Microbiol. Lett. 172,61–64.

Perdigón, G., Alvarez, S., 1992. Bacterial interactions in the gut. In: Fuller, R. (Ed.), Probiotics: TheScientific Basis. Chapman and Hall, London, pp. 145–180.

Perdigón, G., Nader de Macías, M.E., Alvarez, S., Medici, M., Oliver, G., de Ruiz Holgado, A.A.P., 1986.Effect of a mixture of Lactobacillus casei and L. acidophilus administered orally on the immune sys-tem in mice. J. Food Prot. 49, 986–989.

Perdigón, G., Nader de Macías, M.E., Alvarez, S., Oliver, G., de Ruiz Holgado, A.A.P., 1991. Preventionof gastrointestinal infections using immunobiological methods with milk fermented withLactobacillus casei and Lactobacillus acidophilus. J. Dairy Res. 57, 255–264.

Petschow, B.W., Talbott, R.D., 1990. Growth promotion of Bifidobacterium species by whey and caseinfractions from human and bovine milk. J. Clin. Microbiol. 28, 287–292.

Piard, J.C., Desmazeaud, M., 1991. Inhibiting factors produced by lactic acid bacteria. 1. Oxygenmetabolites and catabolism end-products. Lait 71, 525–541.

Pollmann, D.S., Danielson, D.M., Wren, W.B., Peo, E.R., Shahani, K.M., 1980. Influence ofLactobacillus acidophilus inoculum on gnotobiotic and conventional pigs. J. Anim. Sci. 51, 629–637.

Porter, P., Allen, W.D., 1989. The relationship between intestinal flora and the immunity of the intestinaltract. Rev. Sci. Tech. Off. Int. Epiz. 8 (2), 465–489.

Porubcan, R.S., 1990. Probiotics in the 1990s. Compend. Cont. Ed. 12, 1353–1359.Promsopone, B., Morishita, T.Y., Aye, P.P., Cobb, C.W., Veldkamp, A., Clifford, J.R., 1998. Evaluation

of an avian-specific probiotic and Salmonella typhimurium-specific antibodies on the colonization ofSalmonella typhimurium in broilers. J. Food Protect. 61, 2, 176–180.


Ringø, E., Gatesoupe, F.-J., 1998. Lactic acid bacteria in fish: a review. Aquaculture 160, 177–203.Romond, M.-B., Ais, A., Guillemot, F., Bounouader, R., Cortot, A., Romond, Ch., 1998. Cell-free whey

from milk fermented with Bifidobacterium breve C50 used to modify the colonic microflora ofhealthy subjects. J. Dairy Sci. 81, 1229–1235.

Salminen, S., Bouley, C., Boutron-Ruault, M.-C., Cummings, J.H., Franck, A., Gibson, G.R., Isolauri, E.,Moreau, M.C., Roberfroid, M., Rowland, I., 1998. Functional food science and gastrointestinal phys-iology and function. Brit. J. Nutr. 80, Suppl. 1, S147–S171.

Saxelin, M., Grenov, B., Svensson, U., Fondén, R., Reniero, R., Mattila-Sandholm, T., 1999. The tech-nology of probiotics. Trends Food Sci. Technol. 10, 387–392.

Shortt, C., 1999. The probiotic century: historical and current perspectives. Trends Food Sci. Technol. 10,411–417.

Simmering, R., Blaut, M., 2001. Pro- and prebiotics – the tasty guardian angels? Appl. Microbiol.Biotechnol. 55, 19–28.

Sivropoulou, A., Papanikolaou, E., Nikolaou, C., Kokkini, S., Lanaras, T., Arsenakis, M., 1996.Antimicrobial and cytotoxic activities of Origanum essential oils. J. Agr. Food Chem. 44, 1202–1205.

Stavric, S., Kornegay, E.T., 1995. Microbial probiotic for pigs and poultry. In: Wallace, R.J., Chesson, A.(Eds.), Biotechnology in Animal Feeds and Animal Feeding. Weinheim, Germany, VCHVerlagsgesellschaft mbH, pp. 205–231.

Sutton, A.L., Mathew, A.G., Scheidt, A.B., Patterson, J.A., Kelly, D.T., 1991. Effects of carbohydratesources and organic acids on intestinal microflora and performance of the weanling pig. In:Verstegren, M.W.A., Huisman, J., den Hartog, L.A. (Eds.), Proceedings of the 5th InternationalSymposium on Digestive Physiology in Pigs. Wageningen Press, Wageningen, The Netherlands,pp. 442–447.

Thomlinson, J.R., Lawrence, T.L.J., 1981. Dietary manipulation of gastric pH in the prophylaxis ofenteric disease in weaned pigs: Some field observations. Vet. Rec. 109, 120–122.

Wallace, R.J., Newbold, C.J., 1992. Probiotics for ruminants. In: Fuller, R. (Ed.), Probiotics: TheScientific Basis. Chapman and Hall, London, pp. 317–353.

Wasson, K., Criley, J.M., Glabaugh, M.B., Koch, M.A., Peper, R.L., 2000. Therapeutic efficacy of oralLactobacillus preparation for antibiotic-associated enteritis in guinea pigs. Contemp. Topics Lab.Anim. Sci. 39, 32–38.

Watkins, B.A., Miller, B.F., Neil, D.H., 1982. In vivo inhibitory effects of Lactobacillus acidophillusagainst pathogenic Escherichia coli in gnotobiotic chicks. Poultry Sci. 61, 1298–1308.

Wolffram, S., 1999. Absorption and metabolism of selenium: differences between inorganic and organicsources. In: Lyons, T.P., Jasques, K.A. (Eds.), Biotechnology in the Feed Industry, Proceedings ofAlltech’s 15th Annual Symposium. University Press, Nottingham, pp. 547–566.

Wu, M.C., Wung, L.C., Chen, S.Y., Kuo, C.C., 1987. Study on the feeding value of Streptococcus faeciumM-74 for pigs: I. Large scale feeding trial of Streptococcus faecium M-74 on the performance of wean-ing pigs. Research Report, Animal Industry Research Institute, Taiwan Sugar Corp., Chuman Miaoli,Taiwan, pp. 11–12.

Yadava, J.N.S., Gupta, S., Ahmad, I., Varma, N., Tandon, J.S., 1995. Neutralization of enterotoxins ofE. coli by coleonol (forskolin) in rabbit and guinea pig ileal loop models. Indian J. Anim. Sci. 65,1177–1181.

Zacconi, C., Bottazzi, V., Rebecchi, A., Bosi, E., Sarra, P.G., Tagliaferi, L., 1992. Serum cholesterol levelsin axenic mice colonized with Enterococcus faecium and Lactobacillus acidophilus. Microbiologica15, 413–418.


472

Attachment of disease-causing E. coli to the tissues of the host is an initial but essen-tial first step in pathogenesis. Consequently, any strategy that prevents attachmentwill interfere with the development of disease.

Diarrhoea is a common result of the colonization of the small intestine by entero-toxigenic E. coli (ETEC). In this chapter, several strategies that can prevent colo-nization of the gut by these microorganisms will be discussed and compared. Wewill first describe how E. coli, by virtue of fimbriae,2 attach to the intestinal wall,and which molecules are involved, both at the bacterial surface and on the intestinalepithelium. Then, attention will be paid to several approaches to prevent attachmentof enterotoxigenic E. coli to the small intestinal mucosa, i.e. passive and active vaccination, the use of receptor analogues, modification of intestinal receptors, andfinally the potential of probiotics will be described.

1. INTRODUCTION

Escherichia coli is a well-known member of the complex colonic microflora ofmammals and birds. However, some E. coli strains have the possibility to colonizeother tissues as well, and this causes severe and often life-threatening diseases such

20 Strategies for the prevention of E. coliinfection in the young animal1

E. Van Driessche and S. Beeckmans

Laboratory of Protein Chemistry, Institute of Molecular Biology andBiotechnology, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium

1The Belgian “Ministerie van Landbouw en Middenstand”, VEOS n.v. (Zwevezele, Belgium), the Flemish FWO (Fonds voor Wetenschappelijk Onderzoek), VLIR (Vlaamse Interuniversitaire Raad), and DGIS (Directoraat-Generaal voor Internationale Samenwerking) are gratefully acknowledged for financially supporting the authors’ research related to the topic covered by this chapter.

2In this chapter, the F-terminology has been followed to designate fimbriae: F1, F2, F3, F4, F5 and F6 correspond respectively with type-1, CFA/I, CFA/II, K88, K99 and 987P.


Prevention of E. coli infection 473

as diarrhoea. The relation between diarrhoea and E. coli was noticed soon after thefirst description of this bacterium by Theodore Escherich in 1885. If not treatedproperly and adequately after the onset of the first symptoms, diarrhoea is a devas-tating disease.

In western countries, E. coli infections in humans and livestock are generallytreated with antibiotics. The occurrence of resistance and even multiresistancetowards the available antibiotics is a very serious problem we will have to face inthe future. In the tropics, diarrhoea, especially in children, remains life threateningand is a major cause of death, often because adequate treatment is lacking. All overthe world, colibacillosis causes important economic losses because of death of animals, veterinary costs, reduced performance, etc. Modern husbandry practicescreate optimal conditions for the fast spreading of any infectious agent and especially newborn animals and animals at the age of weaning are very sensitive toinfection by bacteria and viruses. The overuse of antibiotics during past decades hasresulted in the selection of multiresistant E. coli that can no longer be tackled withthe available antibiotics. Moreover, recent changes in the attitude of the consumernecessitate husbandry practices that avoid, or at least strongly reduce the use ofantibiotics. For these reasons, alternatives for the treatment of colibacillosis havebecome a priority in research efforts.

The understanding of the principles that govern the attachment of E. coli to hosttissues opens new avenues to treat diseases caused by these microorganisms. In thischapter, we will restrict ourselves to enterotoxigenic E. coli that cause diarrhoea bycolonizing the small intestine, and that produce heat-stable (ST) and/or heat-labile(LT) enterotoxins (Gyles, 1992; Spangler, 1992) that will cause massive water andelectrolyte losses, resulting in watery diarrhoea. As well as enterotoxigenic E. coli(ETEC), other types such as enteroinvasive (EIEC), enteropathogenic (EPEC),enterohaemorrhagic (EHEC), enteroaggregative (EAEC) and uropathogenic(UPEC) strains are also known as important virulent agents (Nataro and Kaper,1998; Nagy and Fekete, 1999).

In view of the importance of enterotoxigenic E. coli in provoking disease inhumans and animals, including important livestock species, ETEC have been investi-gated very intensively for more than two decades, and hundreds of research papershave been published on this topic. Consequently, it cannot be expected that this vastliterature can be completely covered in this chapter. Rather we will focus on some general principles of biosynthesis, purification and application of fimbrial lectins ofETEC for the elaboration of protective vaccines and on possible attachment–inhibitionstrategies that have been developed based on the understanding of the mechanismsETEC use to colonize the small intestine. Although ETEC are important causativeagents of diarrhoea in humans and several animals, emphasis will be put on calves andpigs. When appropriate, reference will also be made to ETEC from human origin,especially when strategies to prevent attachment of these micoorganisms to the intes-tinal mucosa are concerned.

2. E. COLI SURFACE LECTINS AS VIRULENCE FACTORS

Since the beginning of the 20th century it has been known that some E. coli strainsare able to agglutinate human and/or animal erythrocytes. In the mid 1950s, Collierand De Miranda (1955) showed that this agglutination could be specifically inhib-ited by D-mannose and some mannose derivatives. In 1957, Duguid and Gillies(1957) observed that the mannose-sensitive agglutination is correlated to the pres-ence of long proteinaceous filaments at the surface of the bacteria, and that these E. coli can also attach in a mannose-sensitive way to enterocytes. Ofek and co-workers (1977, 1978) showed that destruction of oligosaccharides at the surfaceof enterocytes, or incubation of these cells with the mannose-specific lectin ConA,prevents binding of the E. coli that provoke mannose-specific haemagglutination.Moreover, these researchers showed that yeast mannan is a strong inhibitor ofattachment, and that yeast cells are agglutinated by E. coli that display mannose-sensitive haemagglutination. From the observations mentioned, it was hypothesizedthat the E. coli under investigation express at their surface filamentous mannose-specific lectins (fimbriae) that recognize complementary receptors at the surface oferythrocytes or enterocytes. Definitive proof that fimbriae are lectins responsible for haemagglutination and attachment was provided by Salit and Gotschlich (1977),who were able to isolate these structures in a highly purified form, and could showthem to cause mannose-inhibitable haemagglutination as well as binding to epithelialcells in a mannose-inhibitable fashion. Finally Krogfelt and co-workers (1990) succeeded in identifying the protein molecule within these fimbriae that is respon-sible for the mannose-specific adhesion (i.e. the adhesin or lectin subunit, see below).

Apart from mannose-sensitive haemagglutination, some E. coli strains wereshown to provoke mannose-resistant haemagglutination, which is also associatedwith fimbriae. The former fimbriae are referred to as F1 fimbriae or common fimbriae because they are expressed by both commensal and pathologic E. coli,while the latter are called host-specific, are produced by pathogenic E. coli and areresponsible for host and tissue specificity. In table 1, some important properties ofboth types of fimbriae are compared.

Although haemagglutination is the most straightforward and easiest test for thedetection of surface lectins on E. coli, some lectins may not be detected by thismethod. For example for F6, F18 and CS31A lectins, no erythrocytes are known tobe agglutinated by E. coli expressing these antigens. Also, haemagglutination doesnot allow us to discriminate between fimbrial and non-fimbrial lectins. This differ-ence can, however, be readily made by electron microscopy after negative stainingof the E. coli cells with uranyl acetate (see Van Driessche et al., 1995, for a review).Electron microscopy can also be used for the unequivocal identification of surfaceadhesins when monospecific antisera against known and highly purified adhesinsare available. However, if specific antibodies are available, a simple slide aggluti-nation test is easier to perform when a newly isolated E. coli strain is assayed for the

474 E. Van Driessche and S. Beeckmans

Tabl

e 1.

Cha

ract

eris

tics

of f

imbr

iae

of e

nter

otox

igen

ic E

. col

i(E

TE

C)

stra

ins

that

col

oniz

e liv

esto

ck, a

nd th

at a

re a

ssem

bled

thro

ugh

the

chap

eron

e/us

her

path

way

Alte

rnat

ive

Cha

pero

ne/

Maj

or s

ubun

it G

ene

Ery

thro

cyte

s In

hibi

ting

Fim

bria

ena

me

Nat

ural

hos

tM

orph

olog

yus

her

MM

(D

alto

n)lo

caliz

atio

nag

glut

inat

edM

S/M

Rsu

gars

FGS

chap

eron

eF1

Ty

pe-1

Com

mon

fim

bria

e R

igid

, Fi

mC

/Fim

DFi

mA

C

hrom

osom

eG

uine

a pi

gM

SM

anno

se,

(not

hos

t-sp

ecif

ic)

ø 7

nm17

000

man

nosi

des

F4*

K88

*Pi

gFl

exib

le,

FaeE

/Fae

DFa

eG

Plas

mid

Gui

nea

pig,

M

RG

alac

tosi

des

ø 2.

1 nm

2300

0–27

000

chic

ken

F5K

99Pi

g, c

alf,

lam

bFl

exib

le,

FanE

/Fan

DFa

nC

Plas

mid

Hor

se, s

heep

MR

Sial

ic a

cid

ø 5

nm18

500

F698

7PPi

gR

igid

, Fa

sB/F

asD

FasA

Pl

asm

idN

one

MR

Not

kno

wn

ø 7

nm20

000

F17*

*va

riou

sC

alf,

lam

b, g

oat

Flex

ible

F17-

D/F

17-C

F17-

A

Chr

omos

ome

Bov

ine

MR

Glc

NA

c17

500–

20 0

00F1

8***

F107

Wea

ned

pig

Rat

her

rigi

d,

FedC

/Fed

BFe

dA

Plas

mid

Non

eM

RN

ot k

now

nø

3–4

nm15

000

F41

Pig,

cal

f, la

mb

Flex

ible

, Fi

m41

aC

hrom

osom

eH

orse

, she

ep,

MR

Gal

NA

cø

3.2

nm29

500

guin

ea p

ig,

hum

anFG

L c

hape

rone

CS3

1AC

alf,

lam

b, g

oat

Flex

ible

, C

lpE

/Clp

DC

lpG

Plas

mid

Not

kno

wn

MR

Not

kno

wn

ø 2

nm30

000

MM

, mol

ecul

ar m

ass;

MS,

man

nose

sen

sitiv

e; M

R, m

anno

se r

esis

tant

.*D

iffe

rent

var

iant

s ar

e de

sign

ated

K88

ab, K

88ac

, and

K88

ad.

**F1

7 re

fers

to a

fam

ily o

f hi

ghly

rel

ated

Glc

NA

c-sp

ecif

ic f

imbr

iae,

incl

udin

g F1

7a (

from

bov

ine

ET

EC

), F

17b

(iso

late

d fr

om s

eptic

aem

ic a

nd d

iarr

hoei

c ca

lves

an

d la

mbs

), F

17c

(pre

viou

sly

calle

d 20

K f

imbr

iae,

cau

sing

bov

ine

sept

icae

mic

dia

rrho

ea, o

r G

-fim

bria

e, in

itial

ly is

olat

ed f

rom

hum

an U

PEC

), a

nd F

17d

(pre

viou

sly

calle

d F1

11 a

nd is

olat

ed f

rom

cal

f E

TE

C)

(see

Lin

term

ans

et a

l., 1

988a

,b; B

erte

ls e

t al.,

198

9; E

l Maz

ouar

i et a

l., 1

994;

Saa

rela

et a

l., 1

995,

199

6;

Ber

tin e

t al.,

199

6; M

artin

et a

l., 1

997;

Cid

et a

l., 1

999)

.**

*Dif

fere

nt v

aria

nts

are

desi

gnat

ed F

18ab

and

F18

ac.

expression of known surface lectins. When a new E. coli strain is to be investigated,several techniques can thus be used for the detection and identification of surfacelectins, i.e. haemagglutination, attachment to isolated villi, isolated enterocytes orcell lines such as Caco-2, or attachment to glycoproteins covalently immobilized ona solid support (Van Driessche et al., 1995). Attachment studies on villi or entero-cytes require fixation, a process that might affect the lectin receptors and preventthem from interacting with lectins. An easy to use in vitro attachment system usingEupergit-C beads, which can be easily derivatized with glycoproteins (including solubilized brush border membranes or mucus) has been developed by VanDriessche et al. (1988). These beads can be stored for years at 4°C and allow theinvestigation of attachment as well as attachment-inhibition. For example, Eupergit-C-glycoprotein beads were successfully used to investigate the expression by E. colistrains of F17 fimbriae for which initially no erythrocytes were known that were recognized by these fimbriae (Lintermans et al., 1991). Similarly, with Eupergit-glycoprotein beads we were able to demonstrate the carbohydrate-binding hetero-geneity of F17 fimbriae expressed by different E. coli strains (Van Driessche et al.,1988). Today, multiplex polymerase chain reaction (PCR) has developed into a fastand reliable technique for diagnosis and characterization of fimbrial lectin genes, aswell as for the genes encoding LT and ST toxins (see Osek, 2001, as an example).

Not only is the expression of lectins at their surface important for E. coli to beable to attach to host tissues, but also lectins have been shown to be involved inattachment of Gonococcus species, Salmonella species, Yersinia species, Proteusmirabilis, Bordetella pertussis, Klebsiella pneumoniae, and others (see Soto andHultgren, 1999, for a review). Moreover, there is a close correlation between theexpression of surface lectins, the in vitro binding properties to cells, membranes,and glycoconjugates, and in vivo infectivity of the strains expressing surface lectins.Obviously, E. coli expressing surface structures that allow them to colonize host tis-sues have quite some advantages over those that do not express adhesion molecules.Indeed, attached E. coli are able to withstand the mechanical cleansing mechanismsin, for example, the intestine or the urinary tract, they have considerable growthadvantage, and they display increased resistance to deleterious agents.

Whether an E. coli strain will be able to colonize a host tissue is dependent notonly on the expression of surface adhesins, but also on the presence of tissue recep-tors that can be recognized by the lectins. This statement is best exemplified by thesusceptibility of piglets to infection by E. coli expressing F4 fimbriae. Investigationsof Sellwood in 1980 have shown that sensitive animals possess F4 receptors, whileresistant animals are lacking them. Several E. coli adhesin receptors have beenisolated and characterized, and were shown to be glycoproteins and/or glycolipidspresent in the enterocyte membrane and/or mucus covering the intestinal epithelium(Dean and Isaacson, 1985; Teneberg et al., 1993; Erickson et al., 1994; Khan andSchifferli, 1994; Khan et al., 1996; Billey et al., 1998; Francis et al., 1998, 1999;Al-Majali et al., 2000; Fang et al., 2000; Jin and Zhao, 2000; Jin et al., 2000a;


Mouricout and Védrine, 2000; Van den Broeck et al., 2000; Van Driessche et al.,2000). Age-dependent variation of the oligosaccharide composition and structurecan explain why susceptibility often depends on the age of the animals. Similarly,tissue-specific glycosylation explains the tissue tropism displayed by pathogens.

In the case of F6 fimbriae, Dean et al. (1989) demonstrated that the age-dependentsensitivity of piglets to colonization by E. coli F6 is correlated to the presence of F6receptors in the mucus layer. Indeed, both neonates (sensitive) and older piglets (notsensitive) express F6 receptors at the enterocyte surface. However, mucus receptors inolder piglets proved to be low molecular weight glycoproteins, trace amounts of whichare only detectable in newborn animals. These results show that resistance to E. coliF6 resides in the presence of attachment blockers secreted in the small intestinalmucus layer in older piglets. Besides glycoproteins, Khan et al. (1996) showed that thebrush border membranes also contain two glycolipids that act as F6 fimbrial receptors.Age-dependent expression of F5 fimbrial receptors has been investigated by Yuyamaet al. (1993). These investigators found a good correlation between the postnatalchanges of the F5 glycolipid receptor and the susceptibility to E. coli F5 infection.

3. STRUCTURAL FEATURES OF ENTEROTOXIGENIC E. COLI FIMBRIAL LECTINS

By now, the structure of several fimbriae has been unravelled. Upon isolation (seebelow), SDS-PAGE analysis of purified fimbriae generally shows only one band.Initially, it was thought that fimbriae are very simple oligomeric structures, builtup of a single subunit. However, genetic analysis of several gene clusters encod-ing fimbriae shows that several genes are involved, and their presence is necessaryin order to get the expression of functionally active fimbriae, i.e. able to bind car-bohydrates. Whether or not the encoding genes will eventually be expressed maydepend on different physical factors such as temperature, pH, composition of themedium, etc. (De Graaf, 1988, 1990). This finding also implies that, in vivo, E. coli do not necessarily express the same repertoire of fimbriae as in in vitroconditions.

It was shown that fimbriae biosynthesis requires a complex and meticulouslyregulated interplay of several genes and the polypeptides they encode (De Graaf,1988, 1990; Lintermans et al., 1988a,b, 1991, 1995; Moon, 1990; Krogfelt, 1991;Mouricout, 1991, 1997; Mouricout et al., 1995; Pohl et al., 1992, 1995; Imberechtset al., 1992, 1997a; Hultgren et al., 1993; Van Driessche and Beeckmans, 1993; Khanand Schifferli, 1994; Van Driessche et al., 1995, 2000; Gaastra and Svennerholm,1996; Mol and Oudega, 1996; Smyth et al., 1996; Nataro and Kaper, 1998; Nagyand Fekete, 1999; Soto and Hultgren, 1999; Sauer et al., 2000; Van den Broeck et al., 2000). From the information available today it becomes obvious that at leastthe following polypeptides are required. 1) A large outer membrane “pore” protein,called the “usher”, which is implicated in the translocation of fimbrial subunits


Fig. 1. Genes involved in fimbriae biosynthesis form a cluster, which is generally composed of regulatorygenes (stippled), and genes encoding a variety of polypeptide chains. In this figure, the final polypeptides are coloured in the same way as the genes encoding them. The number of genes in the cluster, as well as their order and length, is different for different ETEC strains.

The ultimate fimbrial structure is composed of a large number of identical major subunits (white ovals),minor subunits (lightly hatched spheres), and special minor subunits recognizing carbohydrates (the lectinsubunits, heavily hatched). The fimbrial structure is attached to the bacterial surface through a large, pore-forming outer membrane (OM) usher protein (vertically striped cylinder). Another essential component of thesystem is the boomerang-shaped chaperone (grey).

During biosynthesis of the above-mentioned series of polypeptides in the bacterial cytoplasm, the emergingchains cross the cytoplasmic membrane (CM) through the general transport apparatus, See. Within the periplasmicspace, each of them is captured by a single chaperone molecule. This results in fimbriae subunits reaching anassembly-competent state, also preventing them from premature aggregation. The chaperone–subunit complexestravel towards an assembly site comprising an usher protein. Once arrived there, the chaperone–subunit complexfalls apart, the subunit being transported through the pore and being incorporated in the growing fimbrial structure,and the chaperone being recycled. Minor subunits are required to initiate fimbriae formation, they also control thelength of the fimbrial structure, and they are implicated in growth termination. Usually, carbohydrate recognitionis due to the presence of another class of minor subunits (the lectin subunits). Lectin subunits often are located atthe tip of the fimbrial structure, in other ETEC strains they occur at regular distances along the whole fimbriae.Occassionally, however (as is the case in K88 fimbriae), all major subunits display lectin activity.

Subunits that are not successfully trapped by chaperone molecules misfold and aggregate in the periplasm.These non-functional entities activate the Cpx system that is composed of a CpxA kinase embedded in thecytoplasmic membrane and a CpxR response regulator. The latter responds to the trigger from the misfoldedchains in several ways, one of which is the induction of expression of periplasmic proteases such as DegP,which will degrade the misfolded subunits.

The fimbrial biosynthetic pathway is indicated with dotted arrows. The activation pathway involing Cpxand the degradative pathway are respectively shown wth dashed and full line arrows.

across the outer membrane, and also serves as a mould upon which fimbriae polymerization occurs. 2) A multifunctional periplasmic protein involved in stabi-lizing non-polymerized fimbrial subunits and in transporting subunits from the innerto the outer membrane. This protein obviously acts as a chaperone (see e.g. Edwardset al., 1996; Mol et al., 1996a,b; Soto and Hultgren, 1999; Normark, 2000). 3) Themajor fimbrial subunits that build up the “body” of the fimbriae. These subunits arethe most prominent polypeptide seen on sodium dodecyl sulphate (SDS)-gels afterelectrophoresis of purified fimbriae. In some fimbrial systems, such as F2, F4, F5and others, the major subunits display carbohydrate-binding activity. 4) Minor fim-brial subunits, which may be involved in initiation and termination of subunit poly-merization, in regulation of the extent of fimbriation, in determining the length offimbriae, in carbohydrate-binding, etc.

Assembly of the fimbriae structures described in table 1 follows the highly conservedso-called “chaperone–usher pathway”, which is schematically depicted in fig. 1 (see e.g.Soto and Hultgren, 1999; Normark, 2000; Sauer et al., 2000, and references therein). Themodel is essentially based on investigations performed with both F1 and uropathogenicPap fimbriae. It can be used as a working hypothesis for the other systems.

Newly synthesized subunits are picked up by the chaperone as soon as theyemerge from the inner cytoplasmic membrane through Sec, the general secretionapparatus. Two subfamilies of chaperones have been described in literature (Hunget al., 1996), named FGS and FGL, the former being involved in the assembly offimbriae with a rod-like architecture (e.g. the thick and rigid F1 and F6 fimbriae, aswell as the thinner and more flexible F4, F5, and F17 fimbriae), the latter beinginvolved in biogenesis of atypical structures such as very thin fimbriae or afimbrialadhesins. The chaperone molecules facilitate the release of nascent fimbrial subunitsfrom the inner membrane, they mediate folding of these subunits into an assembly-competent shape, and they protect them from premature oligomerization in theperiplasmic space. A unique mechanism of so-called donor strand complementation,whereby the chaperone donates a β-strand that fits within a groove of the nascentsubunit, was shown to form the basis of the action of these chaperone molecules(Barnhart et al., 2000). The chaperones further deliver the subunits to the usher,which is embedded within the outer membrane of the bacterium. The usher is apore-forming molecule built up of several protein subunits, presumably presentingextensive regions towards the periplasm ready to interact with incoming chaperone–subunit complexes. All fimbriae grow from top to bottom, as more and more subunits are being delivered and handed over to the “chaperone–usher pathway”.The incoming subunits pass through the usher’s pore, and get packaged into theirfinal structure when reaching the bacterial surface. This packaging event was shownto involve donor strand exchange, whereby the N-terminal extension of the incom-ing subunit (previously capped by the chaperone) displaces the chaperone’s β-strandand inserts itself within the groove of the subunit that was most recently incorpo-rated in the fimbrial structure (Barnhart et al., 2000).


Besides being a pore, the usher is also supposed to play a more active role in fim-briae formation. Indeed, the usher seems to be able to discriminate between variousincoming chaperone–subunit complexes. This would result in trafficking theseincoming subunits in the right order. This order is assumed to be dictated both bykinetic parameters and by subunit–subunit surface complementarity. In this respect,distinct fimbrial minor subunits are thought to have unique structural determinantsrequired to either initiate or terminate fimbrial growth.

In the absence of chaperone molecules, fimbrial nascent subunits get misfoldedand aggregate in the periplasmic space, thereby activating the Cpx signalling systemthat consists of two components, CpxA (a membrane-bound kinase) and CpxR(a response regulator) (Sauer et al., 2000). Phosphorylation of the latter results inthe induction of a variety of genes encoding periplasmic folding factors on the onehand, and periplasmic proteases (e.g. DegP) on the other hand.

4. THE POTENTIAL USE OF ENTEROTOXIGENIC E. COLIFIMBRIAE AS VACCINE COMPONENTS

It is now generally accepted that attachment of ETEC to the intestinal mucosa is aninitial but essential step in pathogenesis. Consequently, any strategy that preventsattachment will interfere with the development of disease. Because of their size,architecture and extracellular expression, fimbriae can easily be obtained in a highlypurified state and in appreciable quantities to be used as vaccine components. The purification procedure generally consists of a solubilization step, ammoniumsulphate precipitation, and, if necessary, eventually followed by gel filtration or ion-exchange chromatography (Van Driessche et al., 1993, 2000; Van den Broeck et al.,1999a) (see scheme 1). It should be noticed that the experimental conditions usedfor the solubilization may critically affect the following steps required to achieve


Scheme 1. Flow sheet describing the purification procedure for enterotoxigenic E. coli fimbriae.

successful purification. For example, sonication of the bacterial suspension com-pletely disrupts the cells and sets free all cytoplasmic proteins as well, some ofwhich may later be difficult to remove. Similarly, heat shock of the bacterial sus-pension can give quite different degrees of homogeneity of fimbrial lectins afterammonium sulphate precipitation, even for fimbriae isolated from the same strainbut grown in different conditions. A gentle way to solubilize fimbriae is by shearingforce using a Waring blender or a similar device. Upon using this technique, fimbriae released from the cells are mostly only slightly contaminated by other proteins, and the fimbriae can easily be recovered in very pure form using low con-centrations of ammonium sulphate that will precipitate fimbriae from solution.Using this procedure, we succeeded, for example, in obtaining F5, F17, F41, andF18 fimbrial preparations that are completely devoid of contaminating proteins (VanDriessche et al., 1993). This procedure could even be used to isolate separately F5and F41 fimbriae from an E. coli strain expressing both types of fimbriae at the sametime, using differential ammonium sulphate precipitation.

Purified fimbriae have proven to be strong immunogens, and humoral antibodiesare readily generated in rodents, chickens and farm animals. Already in the mid1970s, Rutter et al. (1976) described the antibacterial activity of colostrum of sowsthat had been vaccinated with F4 fimbriae. These authors further showed that thisactivity is due to the presence of anti-adhesive F4 antibodies. Similarly, Sojka et al.(1978), Nagy et al. (1986), and Acres et al. (1979) showed that respectively newbornlambs, piglets and calves are successfully protected by colostral antibodies raised byvaccination of dams with F5 fimbriae during pregnancy. Colostral antibodiesdirected against F6 fimbriae were found by Lösch et al. (1986) to passively protectpiglets against infection by F6 positive E. coli. These examples convincinglydemonstrate that newborns can successfully be protected from being colonized bypathogens in the small intestine by maternal antibodies secreted in colostrum andmilk upon vaccination of dams, during pregnancy, with fimbriae isolated from thechallenging E. coli. It should be kept in mind, however, that many strains expressmore than just one type of fimbriae. The investigations of Contrepois and Girardeau(1985) and Runnels et al. (1987) showed that in these cases newborns will only besuccessfully protected upon suckling colostrum and milk when the dams had beenvaccinated with a mixture of all fimbriae involved. These studies thus clearlydemonstrate the urgent need for retrieving new adhesins, or variants of existing fimbriae that might appear in the population.

According to Moon and Bunn (1993), the successful protection of newborns thatis achieved upon parenteral vaccination of pregnant cattle, sheep and pigs can beexplained by the facts that: 1) most ETEC infections occur soon after birth, whencolostral and milk antibody titres are high; 2) only a limited number of fimbrial typesare important in farm animals; 3) fimbriae are good immunogens, being present at thebacterial surface; 4) fimbriae are implicated in an initial but essential step in the devel-opment of disease, i.e. in the attachment of the pathogens to the mucosal surfaces.


The prophylactic effect of various inactivated E. coli vaccines in the control ofpig colibacillosis has been investigated under large-scale farm conditions by Oseket al. (1995). These investigators immunized 2472 pregnant sows with eight differ-ent vaccines containing E. coli fimbrial adhesins and adjuvants. Upon consideringthe general health status of the newborn pigs, the percentage of piglets with diar-rhoea and of dead piglets, as well as the mean body weight gain of weaned piglets,it was found that vaccination had an overall positive effect, although the vaccinestested differed in their protective effect. Best results were obtained when pregnantsows were immunized with a vaccine containing F4, F5 and F6 fimbriae and theB-subunit of the LT enterotoxin.

At the time of weaning, a new period of high sensibility to ETEC starts becauseof the lack of protective maternal antibodies, changes in the diet composition, stress,new environmental conditions, etc. Two approaches have been used to treat or pre-vent post-weaning diarrhoea immunologically, namely by active immunization in anattempt to provoke the secretion of protective antibodies at the level of the intestinalmucosa, and by passive immunization using egg yolk antibodies. Protection ofpiglets against post-weaning diarrhoea was described by Alexa et al. (1995) whoused a combination of parenteral and oral vaccination against ETEC expressing F18fimbriae. When piglets received an intramuscular injection the day before weaningand were treated perorally with a live but non-toxic E. coli strain expressing thesame F18 fimbriae one day after weaning, they were protected against a subsequentchallenge with the virulent strain. Field trials confirmed the protective effect of thisimmunization scheme. Bianchi et al. (1996), on the other hand, reported that par-enteral immunization of mice or piglets with E. coli expressing F4 fimbriae or withpurified F4 fimbriae is ineffective in inducing protective immunity at the mucosallevel against a subsequent challenge with live bacteria. These authors reported thatthis immunization procedure might even be detrimental by inducing a state of suppression. Oral vaccination with live bacteria expressing F4 fimbriae induces anenteric immune response, while killed bacteria were without effect.

Using rabbits as a model, Reid et al. (1993) observed that the colonization fac-tor F3 isolated from ETEC, when incorporated in biodegradable polymer capsules,is immunogenic when administered intradermally. After vaccination, Peyer’s patchcells responded by lymphocyte proliferation to in vitro challenge with F3, and B cellssecreting specific anti-F3 antibodies were detected in the spleen of vaccinated animals. In human volunteers, the studies of Tacket et al. (1994) showed that F3encapsulated in biodegradable microspheres stimulated the secretion of s-IgA anti-F3 in the jejunal fluid. Some volunteers were found to be protected againstchallenge with a pathogenic strain expressing F3 antigens. In pigs, however, neithermicroencapsulated ETEC nor their isolated fimbriae were found to induce serumantibodies, or to reduce the intestinal colonization (Felder et al., 2000).

In a series of elegant investigations, Van den Broeck and associates (1999a,b,c,2000, 2002) showed that purified F4 fimbriae induce both a systemic and an intes-tinal mucosal immune response in just-weaned pigs that express F4 receptors, and


consequently are susceptible to E. coli F4 infections. No response to the orallyadministered F4 antigen could be evidenced in F4 receptor-negative animals. Theseobservations made these researchers conclude that the expression of specific F4receptors by the enterocytes is a prerequisite for purified F4 fimbriae to induce animmune response upon oral immunization. The secretion of IgA antibodies by theintestinal mucosa of F4 receptor-positive animals prevented a subsequent coloniza-tion by E. coli F4. The results obtained open new avenues to use oral vaccination forthe protection of animals from being infected by E. coli (and probably otherpathogens that colonize the intestine following initial binding to mucosal receptorsvia surface adhesins) that induce post-weaning diarrhoea.

Intestinal infection with F18-expressing E. coli was shown by Imberechts et al.(1997a) to induce protection against repeated infections with E. coli expressing eitherthe homologous or the heterologous fimbrial variant of F18 (F18ab or F18ac). Thisprotection was ascribed by Sarrazin and Bertschinger (1997) to IgA secretion at thelevel of the intestinal mucosa. Further studies by Bertschinger et al. (2000) revealedthat oral vaccination of piglets with a fimbriated E. coli F18 strain 10 days beforeweaning successfully protected the animals against a subsequent challenge with ahomologous strain, but not with a heterologous strain. When non-fimbriated E. coliF18 were used for the vaccination, no protective immunity developed. Upon study-ing the antibody responses in newly weaned pigs following infection with an ETECexpressing F4 or a verotoxigenic E. coli (VETEC) expressing F18 fimbriae,Verdonck et al. (2002) reported that F4+ ETEC rapidly colonized the intestine andprovoked a fast F4 specific mucosal immune response, while the VETEC express-ing F18 fimbriae colonized the intestinal mucosa slower, and also made the animalsrespond slower in developing an F18 specific mucosal immune response. Theauthors argued that, in view of the susceptibility of piglets to infection soon afterweaning, a quick response at the level of the mucosa is required. In the case of infec-tion by F4+ E. coli, the immune response already occurred 4 days after infection,possibly as the result of the adjuvant effect of LT produced by the challenging strain.Because of the slow colonization and the retarded immune response provoked by theVETEC F18-producing strain that lacks LT, the same authors suggested that theimmune response might be accelerated by using LT enterotoxin as an adjuvant.

From the examples given above it becomes clear that protective immunity at thelevel of the intestinal mucosa can be achieved by oral vaccination with E. coliexpressing fimbriae, and/or by purified fimbriae (as is the case for F4) on conditionthat the intestinal receptors that are recognized by the fimbriae are present at the surface of the enterocytes. In order to deal with post-weaning diarrhoea, instead ofusing active immunization schemes, passive immunization has also proven to successfully prevent animals from being infected by ETEC. In particular, yolk fromthe eggs of immunized laying hens was shown on several occasions to be a mostappropriate source of protective antibodies. As will be shown below, egg yolk anti-bodies can be used to tackle neonatal diarrhoea in cases where colostrum or milkcontain too low specific antibody titres, or no protective antibodies at all. Indeed, it


is known that upon immunization, laying hens produce high titres of specific anti-bodies in the egg yolk. These antibodies are called IgY, and can readily be isolatedfrom the yolk if required. Passive immunization experiments were already being per-formed in 1992 by Yokoyama and co-workers (1992). They showed that colostrum-deprived piglets were successfully protected against fatal enteric colibacillosis usingpowder preparations of specific antibodies obtained by spray-drying the water-solublefraction of yolk from the eggs of laying hens that had been immunized against F4,F5 and F6 fimbriae. Scanning electron microscopy of intestinal sections showed thatanimals fed with egg yolk powder containing specific anti-fimbriae IgY were devoidof adhering E. coli, while in control animals adherent bacteria could be evidenced.Moreover, when the specific yolk anti-fimbriae antibodies were removed from thepreparations using a fimbrial immunosorbent, the protective activity of the yolkpreparations was significantly reduced, proving that the protective effect is indeeddue to the anti-fimbriae antibodies. Similarly, O’Farrelly et al. (1992) showed thatrabbits were protected from developing diarrhoea when fed egg yolk from hens previously immunized with E. coli that produced heat-labile enterotoxin and colo-nization factor antigen-I. Ikemori et al. (1992) reported that neonatal, colostrum-deprived calves could be protected against ETEC-induced diarrhoea by the proteinfraction (containing anti-fimbriae antibodies) of egg yolk from hens vaccinated withheat-extracted antigens from F5-fimbriated E. coli. Whereas control calves that onlyreceived milk with egg yolk powder from non-immunized hens suffered from severediarrhoea and died within 72 h after challenge, the calves fed milk containing eggyolk powder from immunized hens all survived and had good weight gain, althoughtransient diarrhoea occurred. Independently, Imberechts et al. (1997b) and Yokoyamaet al. (1997) described the protective effect of hen egg yolk containing specific anti-bodies against F18 fimbriae. It was also shown that the yolk antibodies interferewith the attachment of the E. coli to the intestinal mucosa. Similar results werereported by Zuniga et al. (1997), and these authors also concluded that the oralapplication of egg antibodies is a promising approach for the prevention of infec-tious diseases of the digestive tract. The prophylactic effect of yolk antibodiesdirected against F4, F5, F6 and rotavirus has also been reported by Erhard et al.(1996). When egg yolk containing these antibodies was included in the diet ofpiglets, a significant decrease was observed in the number of piglets with diarrhoea,as well as a decrease in the severity of the symptoms, when compared to piglets fedon a diet containing yolk of non-immunized hens, or without any egg yolk at all.

Using in vitro competition and displacement tests, Jin et al. (1998) showed thategg yolk antibodies inhibit the attachment of E. coli F4 to piglet small intestinalmucus, but that displacement of attached E. coli is not possible. The in vivo experi-ments of Marquardt et al. (1999) on the other hand revealed that E. coli F4 caninduce neonatal diarrhoea in 3-day-old piglets, and that animals that were treatedwith egg yolk antibodies were cured within 24 hours, while animals that receivedegg yolk powder of non-immunized hens continued to suffer from diarrhoea, most of


them dying from the infection. Also 21-day-old weaned piglets, fed with egg yolkcontaining specific fimbriae antibodies, successfully survived the infection, whereascontrol animals developed severe diarrhoea, suffered from dehydration, and someanimals eventually died. These authors further reported that, in a field trial, the incidenceand severity of diarrhoea of 14–18-day-old weaned piglets fed egg yolk antibodieswere much lower than in piglets fed a commercial diet containing an antibiotic.From this study, it is convincingly clear that egg yolk antibodies are able to protectboth neonatal and weaned piglets from being infected by ETEC.

Carlander et al. (2000) favourably advocated the use of IgY to establish passiveimmunity against gastrointestinal pathogens such as bacteria and viruses in bothhumans and animals, because these antibodies do not activate mammalian comple-ment or interact with mammalian Fc receptors that can mediate an inflammatoryresponse in the gastrointestinal tract. These authors also considered the risk of toxicside effects very limited, since eggs are part of the diet.

Although it might be expected that, because of the intensive use of fimbrial vaccines, a rapid selection would occur in favour of previously rather infrequentlyoccurring fimbrial antigens, this does not seem to be the case (Moon and Bunn, 1993).

5. ANTI-ADHESION THERAPY BASED ON RECEPTOR ANALOGUES

As mentioned above, the susceptibility of animals and humans towards the colonizationof the small intestinal mucosa depends on the presence of mucosal receptors (glyco-proteins and/or glycolipids) that are recognized by surface lectins of E. coli. In severalinstances, the temporal susceptibility of the host to infection could be correlated withthe presence of soluble receptors in the intestinal lumen and mucus. Since the oligosac-charides of the bacterial lectin receptors are the structures that are recognized,researchers have attempted to use receptor analogues to prevent binding of E. coli tothe intestinal surface by blocking the carbohydrate-binding sites of the bacterial lectins.Already in 1979, Aronson and co-workers (1979) showed the potential of this approachby demonstrating that methyl-α,D-mannopyranoside is able to prevent the in vivo col-onization of the urinary tract of mice by uropathogenic E. coli expressing F1 fimbriae.

Similarly, Neeser et al. (1986) used short oligomannoside-type glycopeptides of ovalbumin and oligomannoside-type glycopeptides derived from legume storage proteins to inhibit the agglutination of erythrocytes caused by E. coliexpressing F1 fimbriae. The studies of Mouricout and co-workers (Mouricout, 1991,1997; Mouricout and Julien, 1987; Mouricout et al., 1990, 1995) boosted the investi-gations on the applicability of receptor analogues by showing that glycoprotein glycansobtained from bovine plasma protect colostrum-deprived newborn calves against lethalchallenge by E. coli F5. In vitro investigations by Sanchez et al. (1993a,b) conclusivelydemonstrated that glycoproteins such as fetuin, ovomucoid, submaxillary gland mucin,hen egg white glycoproteins as well as plasma glycoproteins, especially those of cow,prevent E. coli F17 from attaching to mucus and brush border membranes isolated from


different parts of intestines from calves. Subsequent in vivo studies of Nollet et al.(1996) revealed that non-immune plasma powders from cows protected newborn andcolostrum-deprived calves from developing diarrhoea upon infection with E. coliexpressing F5 or F17 fimbriae. All control calves infected with E. coli expressing eitherF5 or F17 fimbriae developed signs of lethargy, loss of appetite, fever, severe waterydiarrhoea and dehydration from the first day after infection, and eventually diedbetween 1 and 7 days after infection. In one control calf infected with E. coli F17+, thechallenge strain could be isolated not only from the small and the large intestine, butalso from the spleen and liver, indicating that the strain used was septicaemic.Septicaemia was never observed with E. coli F5. Unlike the controls, animals that werefed milk supplemented with 25 g/litre plasma powder remained completely healthy,showed no loss of appetite and did not develop diarrhoea during the first 5 days afterinfection with either E. coli F17 or E. coli F5. When the dose of plasma powder wasreduced to10 g/litre milk, mild disease symptoms developed in animals infected withE. coli F17 cells, but none of the animals developed septicaemia. At this lower dose ofplasma supplementation, the E. coli F5 infected calves on the other hand got fever andalso developed more severe diarrhoea. Remarkably, unlike E. coli F5 infected calves,animals challenged with E. coli F17 continued to excrete the pathogen, indicating thatthese bacteria might colonize the colon, which is not inhibited by plasma powder. Aftersome time, this strain behaves like a commensal that is no longer harmful for thecarrier. All these studies show that plasma glycoproteins are perfectly able to act asreceptor analogues for both F17 and F5 fimbriae, since the plasma preparations usedwere shown to be devoid of antibodies directed against these fimbriae. Similar resultswere obtained by Deprez et al. (1996), who used swine plasma components as adhesininhibitors in the protection of piglets against E. coli enterotoxaemia. It was shown thatthe amount of plasma powder to be administered can be reduced when using plasmafrom pigs that had been vaccinated with fimbriae isolated from the challenge strain. Theinvestigations of Nollet et al. (1999) also conclusively show that non-immune porcineplasma powder protects just-weaned piglets against infection with an E. coli strainexpressing F18 fimbriae isolated from a clinical case of oedema disease. The bestresults were obtained when the piglets received a daily supplementation of feed with 45 g plasma powder. When the amount of plasma powder was increased to 90 g perday, the animals developed diarrhoea, which was ascribed to biogenic amines releasedfrom excessive protein in the diet.

As well as plasma, hen egg white and milk have been shown to be rich sources ofoligosaccharides that can be applied in receptor analogue therapy. In 1983, Wadströmand co-workers (1983a,b) reported milk to be an excellent source of oligosaccharidestructures that mimic intestinal receptors for E. coli adhesins. Also milk fat globulemembranes were shown by Schroten et al. (1992, 1993) to have great potential foruse as a source of receptor analogues. It should also be kept in mind that, besidesoligosaccharides and milk fat globules, milk also contains other antimicrobial activecomponents such as lysozyme, the lacto-peroxidase system, macrophages, as well as


secretory IgA. The protective action of IgA is not exclusively attributed to its specificantigen-binding properties, but also to its covalently linked oligosaccharides that mayinteract with bacterial surface lectins (Wold et al., 1990, 1994). Consequently, IgAdisplays antibacterial activity not only because of its action as an antibody, but alsoas a glycoprotein that can prevent intestinal mucosal colonization by blocking sur-face lectins of bacteria. Other glycoproteins present in milk, such as lactoferrin, maybe acting as attachment inhibitors or as inhibitors of toxins that display carbohydrate-binding properties (Shida et al., 1994; Newman, 1995).

Lindahl (1989) showed that E. coli expressing F41 and F5 fimbriae bind glycopro-teins present in porcine and bovine colostrum, the former being a richer source. Non-immunoglobulin fractions of human milk and colostrum were shown by Ashkenazi andMirelman (1987) to inhibit enterotoxigenic E. coli expressing F2 and F3 from attach-ing to the intestinal tract of guinea pigs. No effect was noticed on the binding of E. coliexpressing F1 fimbriae. Since the inhibitory activity was not affected by boiling or bytrypsin digestion, but was sensitive to metaperiodate oxidation, the authors concludedthat carbohydrate residues of the inhibitor were involved. Similar results were obtainedby Holmgren et al. (1981), who showed that the non-immunoglobulin fraction ofhuman milk and colostrum inhibited the haemagglutination provoked by E. coliexpressing F2, F3, or F4 fimbriae, as well as the haemagglutination by Vibrio cholerae.Moreover, these authors reported that the non-immunoglobulin fraction interfered withbinding of cholera toxin and the LT enterotoxin to GM1 ganglioside. Consequently, thenon-immunoglobulin fraction of milk and colostrum may contribute to the protectiveeffect of milk against enteric infections. Giugliano et al. (1995) identified lactoferrinand free secretory components (fsc) in human milk to be inhibitors of E. coli F2 adhe-sion, as monitored by haemagglutination. At least in the case of F1 fimbriae, Teraguchiet al. (1996) showed that inhibition of haemagglutination is due to the glycan part ofbovine lactoferrin. In addition to inhibiting bacterial attachment, some glycoconjugatesfrom milk may also interfere with binding of enterotoxins to their intestinal receptors.

From the examples given above, it has been shown that receptor analogues cansuccessfully be used to protect calves and pigs against neonatal and post-weaning diar-rhoea by interfering with the colonization of the intestinal mucosa by these pathogens.On the basis of the results obtained in animals, the use of carbohydrates and oligo-saccharide drugs to prevent the attachment of pathogens to human tissues, and/or torevert attached microorganisms, opens new perspectives for combating diseases thatare initiated as a result of lectin-mediated attachment. The major advantage of usingcarbohydrate-based anti-adhesion drugs, when compared to antibiotic or chemothera-peutic approaches, is to be found in the fact that saccharides are not bactericidal and,consequently, selection of resistant strains is unlikely to occur (Zopf and Roth, 1996;Sharon and Ofek, 2000). At present, high production costs of obtaining the requiredoligosaccharides may be a limiting factor, but the further development of techniquesthat allow the enzymatic tailoring of anti-adhesive oligosaccharides could make themavailable at reasonable costs (Sharon and Ofek, 2000).


6. IN VIVO MODIFICATION OF BACTERIAL LECTIN RECEPTORS

An alternative to blocking the E. coli surface lectins might consist of interfering withthe lectin-binding properties of host receptors for bacterial lectins. To achieve this goal,Pusztai and co-workers (1993a,b,c) investigated the effect of the non-toxic mannose-specific lectin from Galanthus nivalis, and showed that it prevented the overgrowth ofthe small intestine of rats by F1 fimbriated E. coli. Plant lectins are indeed potentialcandidates to block intestinal mucosal E. coli lectin receptors, because most plantlectins are rather resistant to proteolytic degradation in the stomach and small intestine,and, as such, can bind to glycoconjugates present in the intestinal mucosa. It should bekept in mind, however, that many plant lectins, when present in appreciable quantities,can adversely affect the structure and functional properties of the small intestine(Pusztai et al., 1993a,b,c). Also, because many plant lectins themselves are glycopro-teins, they could act as “neo-receptors” for bacterial surface lectins. Moreover, in viewof the fast turnover of the small intestinal epithelium, and also because plant lectins,once bound to intestinal receptors, are often internalized, a constant supply of hugeamounts of plant lectin molecules, for instance present in food and feed, would berequired, making this approach hardly feasible for prophylactic or therapeutic purposes.

A more promising approach to, at least temporarily, modify intestinal glycocon-jugates might be offered by the studies of Mynott et al. (1996) and Chandler andMynott (1998). These investigators showed that bromelain, a proteolytic enzymepresent in pineapple juice, can prevent attachment of E. coli F4 to the small intes-tinal mucosa, without causing adverse effects. Although this approach deserves fur-ther investigation, for other E. coli fimbrial systems, possible adverse effects shouldbe carefully looked for. Indeed, the glycoconjugates that we designate as “E. colilectin receptors” have of course other roles to fulfil in the intestine, bacteria just ben-efit from their presence to attach to and colonize the intestinal lining. Consequently,proteolytic modification of these glycoconjugates could eventually interfere withthe endogenous function(s) of these intestinal receptors.

7. PROBIOTICS: TOWARDS THE USE OF HEALTH PROMOTINGBACTERIA TO COMBAT PATHOGENS

As mentioned earlier in this chapter, because of the ever increasing explicit demandof the consumer for meat from animals that have not been treated with antibiotics,but mainly because of the concern of the appearance of bacteria that are resistant oreven multiresistant towards currently used antibiotics, the use of probiotics toachieve “organic food” has gained a lot of popularity during recent years. Probioticscan be described as microorganisms, often Lactobacillus spp., Streptococcus spp.,and Bacillus spp., that are administered either on their own, or included in food orfeed, in order to achieve a favourable intestinal flora and to prevent digestive disor-ders and/or to increase performance.


Kyriakis and co-workers (1999) reported that viable spores of Bacillus licheniformisor Bacillus toyoi added to the feed can be used to successfully control post-weaningdiarrhoea caused by enterotoxigenic E. coli. Moreover, these probiotics improved thegeneral health status and performance of the piglets considerably. The beneficial effectof probiotics may be due to competition with pathogens for available receptors to estab-lish themselves in the small intestine, but also to their possible immunomodulatoryeffects. Herias and co-workers (1999) reported that in gnotobiotic rats, a mannose-binding Lactobacillus plantarium strain not only competes with F1-fimbriated E. colifor colonization sites, especially in the small intestine, but also influences both systemicand intestinal immunity.

Several groups investigated the effect of Lactobacillus spp. on the colonizationof the small intestine by ETEC. Spencer and Chesson (1994) reported that someLactobacillus strains isolated from the gastrointestinal tract of piglets at the time ofweaning attach to enterocytes. On the basis of the number of bacteria that attachedto the enterocytes, strongly and non-to-weakly attaching isolates could be discrimi-nated. However, even the strongly adherent strains had no effect on the attachmentof ETEC. Fimbriated E. coli F4 were shown to co-aggregate with someLactobacillus strains. The same authors concluded that the Lactobacilli under inves-tigation could have a beneficial in vivo effect by aggregating pathogens and pre-venting them from binding to the mucosa. Rojas and Conway (1996) showed thatLactobacilli colonize the mucus layer of the small intestine. It was further shown(Ouwehand and Conway, 1996) that Lactobacillus fermentum 104R releases a com-pound (or compounds) in the culture fluid that inhibits (inhibit) the adhesion of E. coli F4. On the basis of their properties, these compounds were identified as cellwall fragments released from dead cells in the culture fluid. Upon examining theeffect of including Lactobacillus reuteri BSA131, isolated from pig faeces, in thefeed of 1-month-old piglets, Chang et al. (2001) reported a beneficial effect onliveweight gain and feed conversion, as well as a decrease in the number of entero-bacteria in the faeces, and they considered the strain as a potential probiotic forpiglets after weaning.

In addition to Lactobacillus species, also Enterococcus faecium 18C23 has beenshown by Jin et al. (2000b) to inhibit in vitro the attachment of E. coli F4 to themucus of the small intestine of piglets. These authors showed that the mucus recep-tor for Enterococcus faecium is a glycoprotein, since treatment of the mucus withmetaperiodate decreased the adhesion of the bacteria. The inhibitory effect on E. coli F4 attachment is most probably not the result of competition for the samemucus receptors, since treatment of mucus with pronase reduced adhesion of E. coliF4 but increased the adhesion of the probiotic bacteria. From these results, theauthors concluded that the inhibition of adhesion of E. coli F4 by E. faecium is dueto steric hindrance. According to Jin et al. (2000b), also the spent culture super-natant contains some substances that might contribute to the inhibitory effect displayed by the E. faecium cells.


In conclusion, the beneficial effect of probiotics in protecting the host from intes-tinal disorders and/or colonization of the intestine is supposed to be due to one or acombination of the following effects (Rolfe, 2000): 1) stimulation of the immunesystem, possibly by cell wall fragments that act as adjuvants; 2) competitive inhibi-tion for bacterial adhesion sites on the intestinal surface; 3) degradation of intestinaltoxin receptors; 4) production of inhibitory substances such as bacteriocins, hydro-gen peroxide or organic acids. Some organic acids, e.g. lactic acid and others, wereindeed shown by Tsiloyiannis et al. (2001) to reduce both the incidence and sever-ity of post-weaning diarrhoea caused by ETEC in piglets. Bacteriocins produced bylactic acid bacteria, on the other hand, are a large group of small antimicrobial peptides varying in spectrum of activity and molecular structure (De Vuyst and Vandamme, 1994; Nissen-Meyer and Ness, 1997).

8. CONCLUSIONS

Enterotoxigenic E. coli are an important group of pathogens that cause diarrhoea inhumans and livestock. These bacteria are able to colonize the small intestinal epithe-lium by virtue of expressing adhesins (lectins) in the form of long proteinaceousappendages that protrude from their surface, and that are called fimbriae. Fimbriaeare multimeric structures consisting of several types of subunits, referred to as majorand minor subunits. In most instances, it is a special type of minor subunit that displays carbohydrate-binding activity. Sometimes, also the major subunits can bindsaccharides, such as is the case in F4 fimbriae. On the basis of their carbohydratespecificity, fimbriae are classified as F1 (type-1) fimbriae that cause haemaggluti-nation inhibitable by mannose and mannosides, and host-specific fimbriae thatcause haemagglutination not inhibitable by mannose or mannosides. For some fimbriae such as F6, no erythrocytes have been described that are agglutinated eitherby isolated fimbriae or by the E. coli expressing them. The expression of non-haemagglutinating fimbriae can be investigated by in vitro adhesion systems usingfor example enterocytes, Eupergit, or another solid support to which glycoproteinsor glycolipids have been linked or adsorbed.

Whereas the overall structure, the expression and its regulation of several impor-tant fimbriae, as well as the genes that encode them, have been studied in great detail,our knowledge of the structure of the fimbriae receptors present at the surface of theenterocytes, and often in the mucus layer that covers mucosal surfaces, is lacking.However, whether mucosae will be colonized by ETEC depends not only on theexpression of fimbriae, but also on the presence of receptors, i.e. glycoproteins orglycolipids. It is the oligosaccharides of the latter macromolecules that are recognizedby the fimbriae. Tissue-specific, developmentally regulated and species-specific glyco-sylation explains the tissue-dependent, age-dependent and species tropism of ETEC.

During the past decades, antibiotics have successfully been used to combat ETECinfections. However, the intensive use of these chemicals, not only as therapeutics but


also as growth promoters, has finally cumulated in the selection of E. coli strainsthat are resistant to the action of these compounds. As such, we will have to face inthe future the situation where it will be impossible to cure some bacterial infectionswith antibiotics and, consequently, alternatives should become available. Also, theattitude of consumers has changed quite dramatically. The demand for “organic food”is not just a romantic reflex to go back to the practices of “good-old-days”, but is ratherthe consequence of our growing concern about the quality and safety of food. Both theappearance of resistant and even multiresistant bacteria, and consumer demand formeat and meat products derived from animals not fed antibiotics are the driving forcein the search for alternatives to reduce antibiotic use as much as possible.

As mentioned earlier in this chapter, modern husbandry practices create the opti-mal conditions to make animals susceptible to all kinds of infections and allow diseaseto spread quickly. Besides management measures, several strategies are availabletoday that allow for the protection of neonates as well as animals at the age ofweaning and thereafter, from infection by ETEC-causing diarrhoea. Nature itselfhas come up with solutions to prevent the colonization of the small intestine byETEC, where they find the best environment to reproduce massively and to releasetheir toxins in the immediate vicinity of the toxin receptors. In this chapter, severalstrategies have been described, each of them illustrated with examples that showhow the colonization of the small intestine by ETEC can be prevented.

Neonates can successfully be protected by maternal antibodies directed againstE. coli fimbriae and secreted in colostrum and milk. This type of passive immu-nization can easily be achieved by vaccinating dams during pregnancy with purifiedfimbriae, sometimes in combination with the heat-labile toxin that acts as an adju-vant. At the age of weaning and soon thereafter, animals can be passively protectedby egg yolk antibodies mixed in the feed. Eggs from hens immunized with fimbriaeare indeed an excellent source of protective antibodies for animals at an age whereactive immunization does not provide adequate protection in the short term. Morerecently, oral immunization with purified fimbriae such as F4 has been shown toprotect animals from post-weaning diarrhoea by stimulating an immune response atthe level of the small intestinal mucosa.

Since a successful attachment of ETEC to the mucosa is required for these bacteria to cause disease, blocking the fimbrial carbohydrate sites by receptor ana-logues is another alternative to combat ETEC-induced diarrhoea. This strategy,called “receptor analogue therapy”, is still in its infancy, and further progress can beexpected only when we get a more detailed picture of the structure of fimbrial recep-tors, and more particularly of the oligosaccharide sequences involved. Tailored syn-thetic oligosaccharides should make it possible to obtain attachment inhibitors thatoptimally fit the carbohydrate-binding sites of the fimbrial lectins, and consequentlybind with much higher affinity than the receptor analogues available today. Sincereceptor analogues are not toxic for the bacteria, selection for resistance is highlyimprobable.


Still another alternative to antibiotics is probiotics that can compete for ETECreceptors, prevent pathogens from binding by steric hindrance, and/or stimulate themucosal immune system. Finally, temporal modification of fimbrial receptors byproteolytic enzymes such as bromelain might prevent ETEC from binding to thesmall intestinal mucosa.

Although this chapter has mainly dealt with diarrhoeagenic ETEC in pigs andcalves, it should be kept in mind that in the tropics ETEC diarrhoea can still be afatal disease in humans, mainly affecting young children. It is hoped that the know-ledge gained from investigating ETEC in farm animals will be used to deal withhuman ETEC.


Especially during the past decade, much progress has been made in understandingthe mechanisms that govern the attachment of ETEC to the intestinal mucosa andseveral investigations have shown new ways to combat ETEC-induced diarrhoea.Promising strategies that have been developed to prevent colonization of the smallintestine have to be proven valid and useful for more types of fimbriae than thoseused until now. In particular oral vaccination to achieve local mucosal immunity, asshown for F4 fimbriae, should be extended to other fimbrial systems. Optimal dosesof fimbriae to be used for vaccination, the effect of adjuvants, optimal time of vacci-nation, etc., should further be analysed. A continuous search for possible new fim-briae or non-fimbrial adhesins should be carried out in order to improve theprotective effect of existing vaccines. Also elucidation of the fine structure of fim-briae receptor and the changes in the glycosylation pattern during development, andfurther development of enzymatic synthesis procedures that allow the production oftailored receptor analogues at a reasonable cost, will make the use of receptor ana-logues as therapeutics more feasible. Further investigations are also required to findout whether temporal enzymatic receptor modification is a safe and general way toprevent intestinal infection. Finally, further proof of the often only presumed bene-fits of probiotics, and their mode of action, deserve more attention.

REFERENCES

Acres, S.D., Isaacson, R.E., Babink, L.A., Kapitany, R.A., 1979. Immunization of calves against entero-toxigenic colibacillosis by vaccinating dams with purified K99 antigen and whole cell bacterins.Infect. Immunity 25, 121–126.

Al-Majali, A.M., Asem, E.K., Lamar, C.H., Robinson, J.P., Freeman, M.J., Saeed, A.M., 2000. Studieson the mechanism of diarrhoea induced by Escherichia coli heat-stable enterotoxin (Sta) in new-born calves. Vet. Res. Commun. 24, 327–338.

Alexa, P., Salajka, E., Salajkova, Z., Machova, A., 1995. Combined parenteral and oral immunizationagainst enterotoxigenic Escherichia coli diarrhea in weaned piglets. Vet. Med. (Praha) 40, 365–370.

Aronson, M., Medalia, O., Schori, L., Mirelman, D., Sharon, N., Ofek, I., 1979. Prevention of coloniza-tion of the urinary tract of mice with Escherichia coli by blocking of bacterial adherence with methyl-α,D-mannopyranoside. J. Infect. Dis. 139, 329–332.


Ashkenazi, S., Mirelman, D., 1987. Nonimmunoglobulin fraction of human milk inhibits the adherence ofcertain enterotoxigenic Escherichia coli strains to guinea pig intestinal tract. Pediat. Res. 22, 130–134.

Barnhart, M.M., Pinkner, J.S., Soto, G.E., Sauer, F.G., Langermann, S., Waksman, G., Frieden, C.,Hultgren, S.J., 2000. PapD-like chaperones provide the missing information for folding of pilinproteins. Proc. Natl. Acad. Sci. USA 97, 7709–7714.

Bertels, A., Pohl, P., Schlicker, C., Van Driessche, E., Charlier, G., De Greve, H., Lintermans, P., 1989.Isolation of the F111 fimbrial antigen on the surface of a bovine Escherichia coli strain isolated outof calf diarrhea: characterization and discussion of the need to adapt recent vaccines against neonatalcalf diarrhea. Vlaams Diergeneesk. Tijdschr. 58, 118–122.

Bertin, Y., Girardeau, J.-P., Darfeuille-Michaud, A., Contrepois, M., 1996. Characterization of 20K fimbriae,a new adhesin of septicemic and diarrhea-associated Escherichia coli strains, that belongs to a familyof adhesins with N-acetyl-D-glucosamine recognition. Infect. Immunity 64, 332–342.

Bertschinger, H.U., Nief, V., Tschape, H., 2000. Active oral immunization of suckling piglets to preventcolonization after weaning by enterotoxigenic Escherichia coli with fimbriae F18. Vet. Microbiol. 71,255–267.

Bianchi, A.T., Scholten, J.W., van Zijderveld, A.M., van Zijderveld, F.G., Bokhout, B.A., 1996.Parenteral vaccination of mice and piglets with F4+ Escherichia coli suppresses the enteric anti-F4response upon oral infection. Vaccine 14, 199–206.

Billey, L.O., Erickson, A.K., Francis, D.H., 1998. Multiple receptors on porcine intestinal epithelial cellsfor the three variants of Escherichia coli K88 fimbrial adhesin. Vet. Microbiol. 59, 203–212.

Carlander, D., Kolberg, H., Wejaker, P.E., Larsson, A., 2000. Peroral immunotherapy with yolk antibodiesfor the prevention and treatment of enteric infections. Immunol. Res. 21, 1–6.

Chandler, D.S., Mynott, T.L., 1998. Bromelain protects piglets from diarrhoea caused by oral challengewith K88 positive enterotoxigenic Escherichia coli. Gut 43, 196–202.

Chang, Y.H., Kim, J.K., Kim, H.J., Kim, W.Y., Kim, Y.B., Park, Y.H., 2001. Selection of a potentialprobiotic Lactobacillus strain and subsequent in vivo studies. Antonie Van Leeuwenhoek 80, 193–199.

Collier, W.A., De Miranda, J.C., 1955. Microbial serology. Antonie van Leeuwenhoek 21, 133–140.Contrepois, M.G., Girardeau, J.P., 1985. Additive protective effects of colostral antipili antibodies in

calves experimentally infected with enterotoxigenic Escherichia coli. Infect. Immunity 50, 947–949.De Graaf, F.K., 1988. Fimbrial structures of enterotoxigenic E. coli. Antonie van Leeuwenhoek 54, 395–404.De Graaf, F.K., 1990. Genetics of adhesive fimbriae of intestinal Escherichia coli. Curr. Topics

Microbiol. Immunol. 151, 29–53.De Vuyst, L., Vandamme, E.J., 1994. Bacteriocins of lactic acid bacteria. In: De Vuyst, L., Vandamme, E.J.

(Eds.), Bacteriocins of Lactic Acid Bacteria: Microbiology, Genetics and Applications. Blackie Acad.Prof., London, pp. 1–11.

Dean, E.A., Isaacson, R.E., 1985. Purification and characterization of a receptor for the 987P pilus ofEscherichia coli. Infect. Immunity 47, 98–105.

Dean, E.A., Whipp, S.C., Moon, H.W., 1989. Age-specific colonization of porcine intestinal epitheliumby 987P-piliated enterotoxigenic Escherichia coli. Infect. Immunity 57, 82–87.

Deprez, P., Nollet, H., Van Driessche, E., Muylle, E., 1996. The use of swine plasma components asadhesin inhibitors in the protection of piglets against Escherichia coli enterotoxemia. Proceedings ofthe 14th IVPS Congress, Bologna, Italy, pp. 276.

Duguid, J.P., Gillies, R.R., 1957. Fimbriae and adhesive properties in dysentery bacilli. J. Path. Bacteriol.74, 397–411.

Edwards, R.A., Cao, J., Schifferli, D.M., 1996. Identification of major and minor chaperone proteinsinvolved in the export of 987P fimbriae. J. Bacteriol. 178, 3426–3433.

El Mazouari, K., Oswald, E., Hernalsteens, J.-P., Lintermans, P., De Greve, H., 1994. F17-like fimbriaefrom an invasive Escherichia coli strain producing cytotoxic necrotizing factor type 2 toxin. Infect.Immunity 62, 2633–2638.

Erhard, M.H., Bergmann, J., Renner, M., Hofmann, A., Heinritzi, K., 1996. Prophylactic effect ofspecific egg yolk antibodies in diarrhea caused by Escherichia coli K88 (F4) in weaned piglets.Zentralbl. Veterinarmed. A 43, 217–223.

Erickson, A.K., Baker, D.R., Bosworth, B.T., Casey, T.A., Benfield, D.A., Francis, D.H., 1994.Characterization of porcine intestinal receptors for the K88ac fimbrial adhesin of Escherichia coli asmucin-type sialoglycoproteins. Infect. Immunity 62, 5404–5410.


Fang, L., Gan, Z., Marquardt, R.R., 2000. Isolation, affinity purification, and identification of piglet smallintestine mucosa receptor for enterotoxigenic Escherichia coli K88ac+ fimbriae. Infect. Immunity 68,564–569.

Felder, C.B., Vorlaender, N., Gander, B., Merkle, H.P., Bertschinger, H.U., 2000. Microencapsulated entero-toxigenic Escherichia coli and detached fimbriae for peroral vaccination of pigs. Vaccine 19, 706–715.

Francis, D.H., Grange, P.A., Zeman, D.H., Baker, D.R., Sun, R., Erickson, A.K., 1998. Expression ofmucin-type glycoprotein K88 receptors strongly correlates with piglet susceptibility to K88+ entero-toxigenic Escherichia coli, but adhesion of this bacterium to brushborders does not. Infect. Immunity66, 4050–4055.

Francis, D.H., Erickson, A.K., Grange, P.A., 1999. K88 adhesins of enterotoxigenic Escherichia coli andtheir porcine enterocyte receptors. Adv. Exp. Med. Biol. 473, 147–154.

Gaastra, W., Svennerholm, A.M., 1996. Colonization factors of human enterotoxigenic Escherichia coli(ETEC). Trends Microbiol. 4, 444–452.

Giugliano, L.G., Ribeiro, S.T., Vainstein, M.H., Ulhoa, C.J., 1995. Free secretory component and lacto-ferrin of human milk inhibit the adhesion of ETEC. J. Med. Microbiol. 42, 3–9.

Gyles, C.L., 1992. Escherichia coli cytotoxins and enterotoxins. Can. J. Microbiol. 38, 734–746.Herias, M.V., Hessle, E., Telemo, E., Midtvedt, T., Hanson, L.Å., Wold, A.E., 1999. Immunomodulatory

effects of Lactobacillus plantarum colonizing the intestine of gnotobiotic rats. Clin. Exp. Immunol.116, 283–290.

Holmgren, J., Svennerholm, A.M., Ahren, C., 1981. Nonimmunoglobulin fraction of human milk inhibitsbacterial adhesion (hemagglutination) and enterotoxin binding of Escherichia coli and Vibriocholerae. Infect. Immunity 33, 136–141.

Hultgren, S.J., Abraham, S., Caparon, M., Falk, P., St. Geme, J.W. III., Normark, S., 1993. Pilus and non-pilus bacterial adhesins: assembly and function in cell recognition. Cell 73, 887–901.

Hung, D.L., Knight, S.D., Woods, R.M., Pinkner, J.S., Hultgren, S.J., 1996. Molecular basis of two sub-families of immunoglobulin-like chaperones. EMBO J. 15, 3792–3805.

Ikemori, Y., Kuroki, M., Peralta, R.C., Yokoyama, H., Kodama, Y., 1992. Protection of neonatal calvesagainst fatal enteric colibacillosis by administration of egg yolk powder from hens immunized withK99-piliated enterotoxigenic Escherichia coli. Amer. J. Vet. Res. 53, 2005–2008.

Imberechts, H., De Greve, H., Schlicker, C., Bouchet, H., Pohl, P., Charlier, G., Bertschinger, H., Wild, P.,Vandekerckhove, J., Van Damme, J., Van Montagu, M., Lintermans, P., 1992. Characterisation ofF107 fimbriae of Escherichia coli 107/86, which causes edema disease in pigs, and nucleotidesequence of the F107 major fimbrial subunit gene, fedA. Infect. Immunity 60, 1963–1971.

Imberechts, H., Bertschinger, U., Nagy, B., Deprez, P., Pohl, P., 1997a. Fimbrial colonisation factorsF17ab and F17ac of Escherichia coli isolated from pigs with postweaning diarrhea and edema disease.Adv. Exp. Med. Biol. 412, 175–183.

Imberechts, H., Deprez, P., Van Driessche, E., Pohl, P., 1997b. Chicken egg yolk antibodies against F18abfimbriae of Escherichia coli inhibit shedding of F18 positive E. coli by experimentally infected pigs.Vet. Microbiol. 54, 329–341.

Jin, L.Z., Zhao, X., 2000. Intestinal receptors for adhesive fimbriae of enterotoxigenic Escherichia coli(ETEC) K88 in swine – a review. Appl. Microbiol. Biotechnol. 54, 311–318.

Jin, L.Z., Baidoo, S.K., Marquardt, R.R., Frohlich, A.A., 1998. In vitro inhibition of adhesion of entero-toxigenic Escherichia coli K88 to piglet intestinal mucus by egg-yolk antibodies. FEMS Immunol.Med. Microbiol. 21, 313–321.

Jin, L.Z., Marquardt, R.R., Baidoo, S.K., Frohlich, A.A., 2000a. Characterization and purification ofporcine small intestinal mucus receptor for Escherichia coli K88ac fimbrial adhesion. FEMSImmunol. Med. Microbiol. 27, 17–22.

Jin, L.Z., Marquardt, R.R., Zhao, X., 2000b. A strain of Enterococcus faecium (18C23) inhibits adhesionof enterotoxigenic Escherichia coli K88 to porcine small intestine mucus. Appl. Environ. Microbiol.66, 4200–4204.

Khan, A.S., Schifferli, D.M., 1994. A minor 987P protein different from the structural fimbrial subunit isthe adhesin. Infect. Immunity 62, 4233–4243.

Khan, A.S., Johnston, N.C., Goldfine, H., Schifferli, D.M., 1996. Porcine 987P glycolipid receptors onintestinal brush borders and their cognate bacterial ligands. Infect. Immunity 64, 3688–3693.


Krogfelt, K.A., 1991. Bacterial adhesion: genetics, biogenesis, and role in pathogenesis of fimbrialadhesins of Escherichia coli. Rev. Infect. Dis. 13, 721–735.

Krogfelt, K.A., 1995. Adhesin-dependent isolation and characterization of bacteria from their naturalenvironment. Methods Enzymology 253, 50–53.

Krogfelt, K.A., Bergmans, H., Klemm, P., 1990. Direct evidence that the FimH protein is the mannosespecific adhesin of Escherichia coli type 1 fimbriae. Infect. Immunity 58, 1995–1999.

Kyriakis, S.C., Tsiloyiannis, V.K., Vlemmas, J., Sarris, K., Tsinas, A.C., Alexopoulos, C., Jansegers, L.,1999. The effect of probiotic LSP 122 on the control of post-weaning diarrhoea syndrome of piglets.Res. Vet. Sci. 67, 223–228.

Lindahl, M., 1989. Binding of F41 and K99 fimbriae of enterotoxigenic Escherichia coli to glycoproteinsfrom bovine and porcine colostrum. Microbiol. Immunol. 33, 373–379.

Lintermans, P., Pohl, P., Bertels, A., Charlier, G., Vandekerckhove, J., Van Damme, J., Schoup, J.,Schlicker, C., Korhonen, T., De Greve, H., Van Montagu, M., 1988a. Characterization and purifica-tion of the F17 adhesin on the surface of bovine enteropathogenic and septicemic Escherichia coli.Amer. J. Vet. Res. 49, 1794–1799.

Lintermans, P., Pohl, P., Deboeck, F., Bertels, A., Schlicker, C., Vandekerckhove, J., Van Damme, J., VanMontagu, M., De Greve, H., 1988b. Isolation and nucleotide sequence of the F17-A gene encodingthe structural protein of the F17 fimbriae in bovine enterotoxigenic Escherichia coli. Infect. Immunity56, 1475–1484.

Lintermans, P.F., Bertels, A., Schlicker, C., Deboeck, F., Charlier, G., Pohl, P., Norgren, M., Normark, S.,Van Montagu, M., De Greve, H., 1991. Identification, characterization, and nucleotide sequence ofthe F17-G gene, which determines receptor binding of Escherichia coli F17 fimbriae. J. Bacteriol.173, 3366–3373.

Lintermans, P., Bertels, A., Van Driessche, E., De Greve, H., 1995. Identification of the F17 gene clusterand development of adhesion blockers and vaccine components. In: Pusztai, A., Bardocz, S. (Eds.),Lectins: Biomedical Perspectives. Taylor & Francis, London, pp. 235–292.

Lösch, U., Schranner, I., Wanke, R., Jürgens, L., 1986. The chicken egg, an antibody source. J. Vet. Med.33, 609–619.

Marquardt, R.R., Jin, L.Z., Kim, J.W., Fang, L., Frohlich, A.A., Baidoo, S.K., 1999. Passive protectioneffect of egg-yolk antibodies against enterotoxigenic Escherichia coli K88+ infection in neonatal andearly-weaned piglets. FEMS Immunol. Med. Microbiol. 23, 283–288.

Martin, C., Rousset, E., De Greve, H., 1997. Human uropathogenic and bovine septicaemic Escherichiacoli strains carry an identical F17-related adhesin. Res. Microbiol. 148, 55–64.

Mol, O., Oudega, B., 1996. Molecular and structural aspects of fimbriae biosynthesis and assembly inEscherichia coli. FEMS Microbiol. Rev. 19, 25–52.

Moon, H.W., 1990. Colonisation factor antigens of enterotoxigenic Escherichia coli in animals. Curr.Topics Microbiol. Immunol. 151, 147–165.

Moon, H.W., Bunn, T.O., 1993. Vaccines for preventing enterotoxigenic Escherichia coli infections infarm animals. Vaccine 11, 213–220.

Mouricout, M., 1991. Swine and cattle enterotoxigenic Escherichia coli-mediated diarrhoea. Developmentof therapies based on inhibition of bacteria-host interactions. Eur. J. Epidemiol. 7, 588–604.

Mouricout, M., 1997. Interactions between the enteric pathogen and the host. An assortment of bacteriallectins and a set of glycoconjugate receptors. Adv. Exp. Med. Biol. 412, 109–123.

Mouricout, M.A., Julien, R.A., 1987. Pilus-mediated binding of bovine enterotoxigenic Escherichia colito calf small intestinal mucins. Infect. Immunity 55, 1216–1223.

Mouricout, M., Védrine, B., 2000. Lectin-carbohydrate interactions in bacterial pathogenesis. In: Caron, M.,Sève, A.-P. (Eds.), Lectins and Pathology. Harwood Acad. Publ., Amsterdam, pp. 157–171.

Mouricout, M., Milhavet, M., Durié, C., Grange, P., 1995. Characterization of glycoprotein glycan recep-tors for Escherichia coli F17 fimbrial lectin. Microb. Pathogenesis 18, 297–306.

Mouricout, M., Petit, J.M., Carias, J.R., Julien, R., 1990. Glycoprotein glycans that inhibit adhesion ofEscherichia coli mediated by K99 fimbriae: treatment of experimental colibacillosis. Infect.Immunity 58, 98–106.

Mynott, T.L., Luke, R.K., Chandler, D.S., 1996. Oral administration of protease inhibits enterotoxigenicEscherichia coli receptor activity in piglet small intestine. Gut 38, 28–32.


Nagy, B., Fekete, P.Z., 1999. Enterotoxigenic Escherichia coli (ETEC) in farm animals. Vet. Res. 30,259–284.

Nagy, L.K., Mackenzie, T., Pickard, D.J., Dougan, G., 1986. Effects of immune colostrum on the expres-sion of a K88 plasmid encoded determinant: role of plasmid stability and influence of phenotypicexpression of K88 fimbriae. J. Gen. Microbiol. 132, 2497–2503.

Nataro, J., Kaper, J.B., 1998. Diarrheagenic Escherichia coli. Clin. Microb. Rev. 11, 142–201.Neeser, J.R., Koellreutter, B., Wuersch, P., 1986. Oligomannoside-type glycopeptides inhibiting adhesion

of Escherichia coli strains mediated by type 1 pili: preparation of potent inhibitors from plant glyco-proteins. Infect. Immunity 52, 428–436.

Newman, J., 1995. How breast milk protects newborns. Scientific Amer., December, 76–79.Nissen-Meyer, J., Nes, I.F., 1997. Ribosomally synthesized antimicrobial peptides: their function, struc-

ture, biogenesis, and mechanism of action. Arch. Microbiol. 167, 67–77.Nollet, H., Deprez, P., Muylle, E., Van Driessche, E., 1996. The use of non-immune plasma powder in

the prophylaxis of neonatal E.T.E.C. diarrhoea in calves. Proceedings of the 19th World BuiatricsCongress, Edinburgh, pp. 117–120.

Nollet, H., Deprez, P., Van Driessche, E., Muylle, E., 1999. Protection of just weaned pigs against infec-tion with F18+ Escherichia coli by non-immune plasma powder. Vet. Microbiol. 65, 37–45.

Normark, S., 2000. Anfinsen comes out of the cage during assembly of the bacterial pilus. Proc. Natl.Acad. Sci. USA 97, 7670–7672.

O’Farrelly, C., Branton, D., Wanke, C.A., 1992. Oral ingestion of egg yolk immunoglobulin from hensimmunized with an enterotoxigenic Escherichia coli strain prevents diarrhea in rabbits challengedwith the same strain. Infect. Immunity 60, 2593–2597.

Ofek, I., Mirelman, D., Sharon, N., 1977. Adherence of Escherichia coli to human mucosal cells medi-ated by mannose receptors. Nature (London) 265, 623–625.

Ofek, I., Beachey, E.H., Sharon, N., 1978. Surface sugars of animal cells as determinants of recognitionin bacterial adherence. Trends Biochem. Sci. 3, 159–160.

Osek, J., 2001. Multiplex polymerase chain reaction assay for identification of enterotoxigenic Escherichiacoli strains. J. Vet. Diagn. Invest. 13, 308–311.

Osek, J., Truszczynski, M., Tarasiuk, K., Pejsak, Z., 1995. Evaluation of different vaccines to control ofpig colibacillosis under large-scale farm conditions. Comp. Immunol. Microbiol. Infect. Dis. 18, 1–8.

Ouwehand, A.C., Conway, P.L., 1996. Purification and characterization of a component produced byLactobacillus fermentum that inhibits the adhesion of K88 expressing Escherichia coli to porcine ilealmucus. J. Appl. Bacteriol. 80, 311–318.

Pohl, P., Van Muylem, K., Lintermans, P., Imberechts, H., Mainil, J., 1992. Known and new adhesins ofEscherichia coli isolated from diarrhoegenic calves. Ann. Méd. Vét. 136, 479–481.

Pohl, P., Contrepois, M., Imberechts, H., Van Muylem, K., Jacquemin, E., Oswald, E., Mainil, J., 1995.In vitro investigation on attachment and virulence factors of Escherichia coli that adhere to calf intes-tinal villi, but do not produce F5 (K99), F17 or Att111 fimbriae. Ann. Méd. Vét. 139, 421–425.

Pusztai, A., Grant, G., Bardocs, S., 1993a. Lectins as metabolic and growth signals: an overview. In: VanDriessche, E., Franz, H., Beeckmans, S., Pfüller, U., Kallikorm, A., Bøg-Hansen, T.C. (Eds.), Lectins:Biology, Biochemistry, Clinical Biochemistry, 8. Textop, Hellerup, Denmark, pp. 245–249.

Pusztai, A., Ewen, S.W.B., Carvalho, A. de F.F.U., Grant, G., Stewart, J.C., Bardocs, S., 1993b. Lectinsas metabolic signals for the gut and body. In: Van Driessche, E., Franz, H., Beeckmans, S., Pfüller, U.,Kallikorm, A., Bøg-Hansen, T.C. (Eds.), Lectins: Biology, Biochemistry, Clinical Biochemistry,8. Textop, Hellerup, Denmark, pp. 250–257.

Pusztai, A., Grant, G., Spencer, R.J., Duguid, T.J., Brown, D.S., Ewen, S.W., Peumans, W.J., VanDamme, E.J.M., Bardocz, S., 1993c. Kidney bean lectin-induced Escherichia coli overgrowth in thesmall intestine is blocked by GNA, a mannose-specific lectin. J. Appl. Bacteriol. 75, 360–368.

Reid, R.H., Boedeker, E.C., McQueen, C.E., Davis, D., Tseng, L.Y., Kodak, J., Sau, K., Wilhelmsen, C.L.,Nellore, R., Dalal, P., 1993. Preclinical evaluation of microencapsulated CFA/II oral vaccine againstenterotoxigenic E. coli. Vaccine 11, 159–167.

Rojas, M., Conway, P.L., 1996. Colonization by lactobacilli of piglet small intestinal mucus. J. Appl.Bacteriol. 81, 474–480.

Rolfe, R.D., 2000. The role of probiotic cultures in the control of gastrointestinal health. J. Nutr. 130,396S–402S.


Runnels, P.L., Moseley, S.L., Moon, H.W., 1987. F41 pili as protective antigens of enterotoxigenicEscherichia coli that produce F41, K99, or both pilus antigens. Infect. Immunity 55, 555–558.

Rutter, J.M., Jones, G.W., Brown, G.T.H., Burrows, M.R., Luther, P.D., 1976. Antibacterial activity incolostrum and milk associated with protection of piglets against enteric disease caused by K88-positiveEscherichia coli. Infect. Immunity 13, 667–676.

Salit, I.E., Gotschlich, E.C., 1977. Hemagglutination by purified type 1 Escherichia coli pili. J. Exp.Med. 146, 1169–1181.

Sanchez, R., Kanarek, L., Koninkx, J., Hendriks, H., Lintermans, P., Bertels, A., Charlier, G., VanDriessche, E., 1993a. Inhibition of adhesion of enterotoxigenic Escherichia coli cells expressing F17fimbriae to small intestinal mucus and brush-border membranes of young calves. Microb.Pathogenesis 15, 407–419.

Sanchez, R., Hendriks, H., Koninckx, J., Lintermans, P., Kanarek, L., Van Driessche, E., 1993b. Studieson attachment-inhibitors of Escherichia coli strains expressing F17 fimbriae. In: Van Driessche, E.,Franz, H., Beeckmans, S., Pfüller, U., Kallikorm, A., Bøg-Hansen, T.C. (Eds.), Lectins: Biology,Biochemistry, Clinical Biochemistry, 8. Textop, Hellerup, Denmark, pp. 223–228.

Sarrazin, E., Bertschinger, H.U., 1997. Role of fimbriae F18 for actively acquired immunity againstporcine enterotoxigenic Escherichia coli. Vet. Microbiol. 54, 133–144.

Sauer, F.G., Mulvey, M.A., Schilling, J.D., Martinez, J.J., Hultgren, S.J., 2000. Bacterial pili: molecularmechanisms of pathogenesis. Curr. Opinion Microbiol. 3, 65–72.

Schroten, H., Hanisch, F.G., Plogmann, R., Hacker, J., Uhlenbruck, G., Nobis-Bosch, R., Wahn, V., 1992.Inhibition of adhesion of S-fimbriated Escherichia coli to buccal epithelial cells by human milk fatglobule membrane components: a novel aspect of the protective function of mucins in the non-immunoglobulin fraction. Infect. Immunity 60, 2893–2899.

Schroten, H., Plogmann, R., Hanisch, F.G., Hacker, J., Nobis-Bosch, R., Wahn, V., 1993. Inhibition ofadhesion of S-fimbriated E. coli to buccal epithelial cells by human skim milk is predominantly medi-ated by mucins and depends on the period of lactation. Acta Paediatr. 82, 6–11.

Sellwood, R., 1980. The interaction of the K88 antigen with porcine intestinal epithelial cell brush borders.Biochim. Biophys. Acta 632, 326–335.

Sharon, N., Ofek, I., 2000. Safe as mother’s milk: carbohydrates as future anti-adhesion drugs for bacterialdiseases. Glycoconjugate J. 17, 651–656.

Shida, K., Takamizawa, K., Nagaoka, M., Osawa, T., Tsuji, T., 1994. Enterotoxin-binding glycoproteinsin a protease-peptone fraction of heated bovine milk. J. Dairy Sci. 77, 930–939.

Smyth, C.J., Marron, M.B., Twohig, J.M., Smith, S.G., 1996. Fimbrial adhesins: similarities and varia-tions in structure and biogenesis. FEMS Immunol. Med. Microbiol. 16, 127–139.

Sojka, W.J., Wray, C., Morris, J.A., 1978. Passive protection of lambs against experimental entericcolibacillosis by colostral transfer of antibodies from K99 vaccinated ewes. J. Med. Microbiol. 11, 493–499.

Soto, G.E., Hultgren, S.J., 1999. Bacterial adhesins: common themes and variations in architecture andassembly. J. Bacteriol. 181, 1059–1071.

Spangler, B.D., 1992. Structure and function of cholera toxin and the related Escherichia coli heat-labileenterotoxin. Microb. Rev. 56, 622–647.

Spencer, R.J., Chesson, A., 1994. The effect of Lactobacillus spp. on the attachment of enterotoxigenicEscherichia coli to isolate porcine enterocytes. J. Appl. Bacteriol. 77, 215–220.

Tacket, C.O., Reid, R.H., Boedecker, E.C., Losonsky, G., Natora, J.P., Bhagat, H., Edelman, R., 1994.Enteral immunization and challenge of volunteers given enterotoxigenic E. coli CFA/II encapsulatedin biodegradable microspheres. Vaccine 12, 1270–1274.

Teneberg, S., Willemsen, P.T.J., De Graaf, F.K., Karlsson, K.-A., 1993. Calf small intestine receptors forK99 fimbriated enterotoxigenic Escherichia coli. FEMS Microbiol. Lett. 109, 107–112.

Teraguchi, S., Shin, K., Fukuwatari, Y., Shimamura, S., 1996. Glycans of bovine lactoferrin function asreceptors for the type 1 fimbrial lectin of Escherichia coli. Infect. Immunity 64, 1075–1077.

Tsiloyiannis, V.K., Kyriakis, S.C., Vlemmas, J., Sarris, K., 2001. The effect of organic acids on thecontrol of porcine post-weaning diarrhoea. Res. Vet. Sci. 70, 287–293.

Van den Broeck, W., Cox, E., Goddeeris, B.M., 1999a. Receptor-specific binding of purified F4 to isolatedvilli. Vet. Microbiol. 68, 255–263.

Van den Broeck, W., Cox, E., Goddeeris, B.M., 1999b. Induction of immune responses in pigs followingoral administration of purified F4 fimbriae. Vaccine 17, 2020–2029.


Van den Broeck, W., Cox, E., Goddeeris, B.M., 1999c. Receptor-dependent immune responses in pigsafter oral immunization with F4 fimbriae. Infect. Immunity 67, 520–526.

Van den Broeck, W., Cox, E., Oudega, B., Goddeeris, B.M., 2000. The F4 fimbrial antigen of Escherichiacoli and its receptors. Vet. Microbiol. 71, 223–244.

Van den Broeck, W., Bouchaut, H., Cox, E., Goddeeris, B.M., 2002. F4 receptor-independent priming ofthe systemic immune system of pigs by low oral doses of F4 fimbriae. Vet. Immunol. Immunopath.85, 171–178.

Van Driessche, E., Beeckmans, S., 1993. Fimbrial lectins from enterotoxigenic E. coli: a mini-review. In:Van Driessche, E., Franz, H., Beeckmans, S., Pfüller, U., Kallikorm, A., Bøg-Hansen, T.C. (Eds.),Lectins: Biology, Biochemistry, Clinical Biochemistry, 8. Textop, Hellerup, Denmark, pp. 207–216.

Van Driessche, E., Schoup, J., Charlier, G., Lintermans, P., Beeckmans, S., Zeeuws, R., Pohl, P., Kanarek, L.,1988. The attachment of E. coli to intestinal calf villi and Eupergit-C-glycoprotein beads. In: Bøg-Hansen, T.C., Freed, D.L.J. (Eds.), Lectins: Biology, Biochemistry, Clinical Biochemistry, 6. SigmaChem. Comp., St. Louis, MO, pp. 55–62.

Van Driessche, E., Sanchez, R., Beeckmans, S., De Cupere, F., Charlier, G., Pohl, P., Lintermans, P.,Kanarek, L., 1993. A general procedure for the purification of fimbrial lectins from Escherichia coli.In: Gabius, H.J., Gabius, S. (Eds.), Lectins and Glycobiology. Springer-Verlag, Heidelberg, pp.47–54.

Van Driessche, E., Sanchez, R., Dieussaert, I., Kanarek, L., Lintermans, P., Beeckmans, S., 1995.Enterotoxigenic fimbrial Escherichia coli lectins and their receptors: targets for probiotic treatmentof diarrhoea. In: Pusztai, A., Bardocz, S. (Eds.), Lectins: Biomedical Perspectives. Taylor & Francis,London, pp. 235–292.

Van Driessche, E., De Cupere, F., Beeckmans, S., 2000. Protein-carbohydrate interactions in the attach-ment of enterotoxigenic Escherichia coli to the intestinal mucosa. In: Caron, M., Sève, A.-P. (Eds.),Lectins and Pathology. Harwood Acad. Publ., Amsterdam, pp. 135–156.

Verdonck, F., Cox, E., Van Gog, K., Van der Stede, Y., Duchateau, L., Deprez, P., Goddeeris, B.M., 2002.Different kinetics of antibody responses following infection of newly weaned pigs with an F4 entero-toxigenic Escherichia coli strain or an F18 verotoxigenic Escherichia coli strain. Vaccine 20,2995–3004.

Wadström, T., Trust, T.J., Brooks, D.E., 1983a. Bacterial surface lectins. In: Bøg-Hansen, T.C., Spengler, G.A.(Eds.), Lectins: Biology, Biochemistry, Clinical Biochemistry, 3. W. deGruyter, Berlin, pp. 479–494.

Wadström, T., Faris, A., Lindahl, M., Lönnerdahl, B., Ersson, B., 1983b. Mannose-specific lectins onenterotoxigenic E. coli recognizing different glycoconjugates. In: Bøg-Hansen, T.C., Spengler, G.A.(Eds.), Lectins: Biology, Biochemistry, Clinical Biochemistry, 3. W. deGruyter, Berlin, pp. 503–510

Wold, A.E., Mestecky, J., Tomana, M., Kobata, A., Ohbayashi, H., Endo, T., Svanborg Eden, C., 1990.Secretory immunoglobulin A carries oligosaccharide receptors for Escherichia coli type 1 fimbriallectin. Infect. Immunity 58, 3073–3077.

Wold, A.E., Motas, C., Svanborg, C., Mestecky, J., 1994. Lectin receptors on IgA isotypes. Scand. J.Immunology 39, 195–201.

Yokoyama, H., Peralta, R.C., Diaz, R., Sendo, S., Ikemori, Y., Kodama, Y., 1992. Passive protective effectof chicken egg yolk immunoglobulins against experimental enterotoxigenic Escherichia coli infectionin neonatal piglets. Infect. Immunity 60, 998–1007.

Yokoyama, H., Hashi, T., Umeda, K., Icatlo, F.C., Kuroki, M., Ikemori, Y., Kodama, Y., 1997. Effect oforal egg antibody in experimental F18+ Escherichia coli infection in weaned pigs. J. Vet. Med. Sci.59, 917–921.

Yuyama, Y., Yoshimatsu, K., Ono, E., Saito, M., Naiki, M., 1993. Postnatal change of pig intestinalganglioside bound by Escherichia coli with K99 fimbriae. J. Biochem. 113, 488–492.

Zopf, D., Roth, S., 1996. Oligosaccharide anti-infective agents. Lancet 347, 1017–1021.Zuniga, A., Yokoyama, H., Albicker-Rippinger, P., Eggenberger, E., Bertschinger, H.U., 1997. Reduced

intestinal colonisation with F18-positive enterotoxigenic Escherichia coli in weaned pigs fed chickenegg antibody against the fimbriae. FEMS Immunol. Med. Microbiol. 18, 153–161.


499

Adhesin receptors, 159, 163Adhesins, 159, 168Adhesion, 474, 476, 485, 487, 489, 490Adhesion-receptor, 360Advances in age, 106Aeromonas hydrophila, 209, 222Aeromonas salmonicida, 209, 211, 214, 216,

221, 226, 227S-layer, 211

AFA (afimbrial adhesin), 174, 176Age, 112Agglutination, 474, 485Agrin, 296, 308Ammonia, 60, 65, 69, 70Amplified fragment length polymorphism (AFLP), 126Amplified ribosomal DNA restriction analysis

(ARDRA), 126Anaerobes, 25Anguilla japonica, 225Antibiotic, 473, 484, 487, 488, 490, 491Antibiotics, 277, 278Antibody (-ies), 474, 481–484, 486, 494Antibody-dependent cell-mediated cytotoxicity

(ADCC), 298Antigen presenting cells (APC), 296, 299, 362Antimicrobial activity, 427Antwerp database, 130Apoptosis, 307Appearance of ciliates, 54, 62, 63Aquatic animals, 418, 423, 424, 427, 433, 434, 437ARB, 130Arctic, 75, 78–81, 86, 88, 90Arctic charr, 216Associated immune cells, 361Attachment, 472–477, 480, 481, 484, 487, 488,

491, 492Autoantigens, 357Autochthonous strains, 358Autochtonic microflora, 296

B cells, 357, 359, 369, 370tB cells (lymphocytes), 298Bacillus, 488

Bacillus spp., 382, 383, 386t, 388, 391, 396t, 400,402, 404, 405, 408

Bacteria, 57, 60, 65, 69, 70, 79, 81, 83, 85, 86,88–96, 119

Bacterial adhesins, 210, 215Bacterial adhesion, 211, 359, 360

environmental factors, 210mechanisms, 212

Bacterial floraautochthonous, 217stability, 217

Bacterial invasion, 218Bactericidal activity, 302Bacteriocins, 419, 423, 424, 426, 427, 431, 433, 434Bacteroides, 124, 126, 130, 131

thetaiotaomicron, 130vulgatus, 124

Batch culture, 144Bifidobacteria, 274–275Bifidobacterium, 126, 131Biochemical fingerprinting, 32Biochip, 136Biosynthesis (fimbriae), 473, 477, 479Blocker (attachment), 477B-lymphocytes, 318, 319Bovine milk, 302, 305Brevoortia patronus, 225Bromelain, 488, 492Bronchus-associated lymphoid tissues (BALT), 295Brush border, 476, 477, 485

Caco-2, 151Caco-2 cells, 303Caecum, 86, 90, 91Calf, 473, 481, 484, 485, 487, 492Calves, 54, 56, 61–63, 65, 69, 70Campylobacter, 114, 265–267Campylobacter jejuni, 211, 16Capsule, bacterial, 210Carbohydrate-binding, 476, 477, 479, 485, 487,

490, 491Carbohydrate fermentation, 55, 56, 65Carbohydrates, 60, 65

Index

Carnobacterium, 418, 420, 430, 431, 434, 442Carnobacterium piscicola, 209Carnobacterium spp., 390, 391, 393, 394, 396t,

403, 407Celiac disease, 299Cellulose, 60Chaperone, 479Chemokines, 296, 299Chlamydia, 127

psittaci, 127CHO cells, 300Chromophore, 135Ciliate fauna, 54, 55, 62, 63, 70Ciliates, 61, 79, 87, 89Clostridia, 268, 269Clostridium, 130

coccoides, 130leptum, 130perfringens, 127

Clostridium difficile, 114Clostridium perfringens, 108CNF (cytotoxic necrotizing factor), 173Colibacillosis, 473, 482, 484Coliforms, 27, 32, 33, 36, 42, 44Colon, 111Colonization, 258, 259, 472, 477, 482–485, 487,

489, 490, 491, 492Colonization factor, 482, 484Colostrum, 295, 305, 306, 481, 483–485, 487, 491Commensal microflora, 302Common mucosal antigen, 360Competitive exclusion, 275–277ConA, 250Continuous model, 142Continuously Stirred Tank Reactor (CSTR), 145Cryptdins, 297Cytochalasin D, 220, 222, 224Cytokines, 296, 299–303, 367, 368Cytostatic factor, 306Cytotoxic effect (CTE), 300Cytotoxic T lymphocytes (CTL), 296Cytotoxins, 300Dam, 481, 491Dam’s milk, 104Delivery, 328–330, 333, 335, 341, 343, 346Denaturing gradient gel electrophoresis (DGGE),

121, 128Dendritic cells, 367DGGE, 151, 152Diagnosis, 177Diarrhoea, 41, 42, 114, 472, 473, 482, 483, 484,

486, 487, 489–492Dicentrarchus labrax, 222Diet, and disease resistance, 226Digestive tract, 4–6Disease control, 381, 383, 385, 404, 409DNA, 120–125, 129, 134

Dynamicconditions, 146, 149model, 142, 144, 148

E. coli, 260–263EAF (EPEC adherence factor), 169, 171Edwardsiella ictaluri, 209, 211, 222, 226

pili, 211Edwardsiella tarda, 209, 222Egg yolk, 482–484, 491EHEC (enterohaemorrhagic E. coli), 169Electron microscopy, scanning (SEM), 211, 214Electron microscopy, transmission (TEM), 211Elemental ration, 107EMBL, 130Enterobacteriaceae, 126Enterobacteriaceae, 353, 355Enterococcus, 272, 420, 436, 489Enterococcus sp., 225Enterocyte, 474–477, 483, 489, 490Entero-mammary link, 302, 303Enteropathogenic, 363Enterotoxigenic, 473, 477, 480, 487, 489, 490Enterotoxins, 160, 300, 473, 482–484, 487EPEC (enteropathogenic E. coli), 169Epithelial barriers, 361Epithelial cells, 361–363Epithelium, 472, 474, 476, 488, 490Epitopes, 369Erysipelothrix rhusiopathiae, 272Escherichia

coli, 125, 127, 133Escherichia coli, 210, 214, 218

attaching and effacing, 108enteroinvasive, 215enteropathogenic (EPEC), 214enterotoxigenic, 108

Escherichia coli (E. coli), 472, 474–477, 480–492

Establishment of ciliates, 54, 61, 63, 65, 70ETEC, 472, 473, 480, 481–485, 489, 490, 491ETEC (enterotoxigenic E. coli), 158Eubacterium, 120

rectale, 130Eupergit-C, 476, 490Expression vector, 337, 338, 345

Factors affecting establishment of ciliates, 63FAE cells, 364, 365Family Ophryoscolecidae, 57, 60Fat globule, 486Fc receptors (FcR), 305Fermenter, 150Fermentative capacity, 34–36, 38, 39Fibre degradation, 60Fibrinogen, 214Fibronectin, 214, 222

Index500

Fimbriae, 158Fimbriae, 210, 262, 472–492

E. coli type 1, 208, 210E. coli type 4, 210, 214E. coli type 5, 210

Fimbrial lectin, 473–477, 480, 485, 487, 488, 490, 491

Fish, 314–325, 386t, 389t, 395t, 396tFISH, 151Fish larvae, 385, 394, 398, 399, 405Fish pathogens, 424, 431–433, 436, 442, 445Fish, See Chapter 17Flagella, 210, 211, 262Flexibacter maritimus, 211Fluorescence-in situ-hybridization (FISH), 121, 132Fluorophore, 135Food conversion efficiency, 67, 68tFry, 383, 385, 388, 391, 393, 395t, 399, 406Fungi, 55, 57, 61, 79, 89Fusobacterium, 130

prausnitzii, 130

Gastrointestinal (GI) tract, 294, 296, 297, 299, 302, 303

Gastrointestinal tract, 119–121, 123, 125, 136Gastro-Intestinal Tract (GIT)

metabolites/products, 142, 144, 145, 149, 152analysis of, 11

Genbank, 130Genome sequencing, 227Genomic fingerprint, 125GIT, 142

bile, 144, 146, 149conditions, 143crop, 143digestive processes, 144gastric emptying, 147, 149gizzard, 143monogastric animals, 144morphology, 143parameters

composition, 150digestive juices

composition, 142residence time, 142secretion, 142

pH, 142residence time, 143secretion, 146, 149

rumen, 144ruminants, 143, 144

Glycolipid, 476, 477, 485, 490, 491Glycoprotein, 476, 477, 485–490GNA, 250Gnotobiotic animals, 3–8Goblet cells, 296Green fluorescent protein, 222, 227

Growth rate, 67, 69Gut-associated lymphoid tissues (GALT), 297, 299

Haemagglutination, 474, 487, 490Helicobacter infection, 103, 106Helper oligonucleotides, 133Hippoglossus hippoglossus, 221Human, 473, 482, 485, 487, 492Humic compounds, 123Hybridization, 121

dot-blot, 121, 123, 131, 132slot-blot, 121, 124touch-blot, 124whole cell-in situ, 121, 132

Ictalurus punctatus, 222, 226IgA, 297–299, 302IgA, 482, 487IgG1, 302IgM, 298IgY, 484, 485Immune complexes (Ic), 305Immune escape, 302Immune memory, 324Immune paradox, 305Immune response, 354, 358, 361Immunity, 5, 314–316Immunization, 481–485, 491Immunoglobulin A, 362Immunosuppression, 116“In-feed” vaccines, 180In vitro model, 143, 144, 150–152Infection routes, 209, 223Infectious diseases, 108Infectious haematopoietic necrosis (IHN)

virus, 225Infectious pancreatic necrosis (IPN) virus, 225Infectious salmon anaemia (ISA) virus, 225Inhibitor (attachment), 474, 486, 487, 489, 491Integrin, 220Intergenic spacer regions (ISR), 122Internal transcribed spacers (ITS), 127Intestinal microflora, 363, 370, 371Intestinal mucosa, 472, 473, 480, 481–484, 485,

487–489, 491, 492Intestine

large, 144, 150small, 144, 146, 149

Intimin, 169, 214Intracellular bacteria, 218Intracellular killing, 300Intraepithelial lymphocytes (IELs), 297, 300, 302Invasion of eukaryotic cells, 220, 222Ity, 241Lactation, 303Lactic acid bacteria (LAB), 382, 383, 390–392,

399, 400, 403, 407, 418, 427, 436–445

Index 501

Lactic acid bacteria (Lactobacilli), 273, 356, 363,364, 367, 370t

Lactobacillus, 33, 39, 46, 127, 131, 328–330,332–336, 338, 340–346, 418, 420, 427, 430, 434, 436, 439, 440, 488, 489

Lactobacillus casei, 218Lactococcus garviae, 209Lambs, 6, 7, 8, 56, 61, 62, 65, 67, 69, 70Lectin, 473, 474, 476, 477, 481, 485, 487, 488,

490, 491Lectins, 250LEE (locus of enterocyte effacement), 169Length heterogenicity PCR (LH-PCR), 126Listeria monocytogenes, 272Lymphoid system, 317Lysozyme, 209

M cells, 218, 361, 363, 364, 369Macrophages (MΦ), 300, 303, 357, 360, 362, 367Major histocompatibility complex (MHC), 299,

300, 302, 359, 363Major subunit (fimbriae), 477, 490Mammary gland, 295, 302–304Mannose, 214, 224, 474, 488, 490Mastitis, 303, 304Maternal antibodies, 481, 482, 491Maternal influences, 315, 316Metabolic fingerprinting, 29Methanogens, 79, 81Microbiota, 302Micro-flora, 258, 260Microflora

composition, 142, 150–152Milk, 106, 481, 483, 484, 486, 487, 491Minor subunit (fimbriae), 480, 491Molecular beacons, 135Monogastric animals, 295, 302Moraxella, 271Moritella marina, 209Moritella viscosa, 209Morone saxatilis, 225Mucosa, 472, 473, 480, 481–485, 487–489,

490–492Mucosal immune system (MIS), 293, 307Mucosal immunization, 329Mucosal lymphoid system, 318Mucosal mast cells (MMC), 298Mucosal-associated lymphoid tissues (MALT),

295, 297Mucus, 215, 216, 296, 476, 477, 484, 485, 489, 490

as bacterial nutrient source, 216composition, 215functions, 215

Mugil cephalus, 225Mycobacterium, 270Mycoplasma, 269–270Nasal-associated lymphoid tissues (NALT), 295

Native flora, 272, 273Neonatal, 477, 483, 484, 487, 491Neurotoxins, 300NK cells, 298, 302Nodavirus, 225Non-specific defence, 316Normal microflora, 25, 28NTEC (necrotoxigenic E. coli), 173

Oligonucleotide probes, 29, 122, 131–133Oligosaccharide, 474, 477, 485–487, 490, 491Oncorhynchus keta, 224Oncorhynchus kisutch, 224Oncorhynchus tshawytscha, 224Ontogeny, 321, 324Oral cavity, 104Oral immunization, 328, 329, 333Oral tolerance, 358, 360, 367Oral vaccination, 482–484, 490, 491Outer membrane proteins, bacterial, 210

P (Pap) fimbriae, 174paa (porcine attaching effacing associated)

factor, 172Paneth cells, 296, 297Panleukopenia virus, 116Pap fimbriae, 479Parvovirus, 108Pasteurella, 271Pasteurella piscicida, 212Pathogens, 92, 93, 191, 196, 197, 202, 203PCR, 151, 476PE (phosphatidylethanolamine), 171Peyer’s patches (PP), 297, 357, 358, 362, 364, 369pH, 54, 63Phagocytic system, 316Phagocytosis, 300, 305Photobacterium damselae subsp. piscicida, 212,

214, 223PhPlates, 30Phylogeny, 130

Phylogenetic markers, 120Pig, 44, 473, 476, 477, 481–484, 486, 487, 489, 492Piglets, 7, 13, 14Piscirickettsia salmonis, 209, 223, 227Plant lectin, 488Plasma powder, 486Plasmid, 122, 123Polygastric animals, 302Polymerase chain reaction (PCR), 121, 122, 125,

127, 130, 134Arbitrarily primed PCR (AP-PCR), 125Length heterogenicity PCR (LH-PCR), 126Repetitive extragenic palindromic sequences

(REP-PCR), 127Reverse transcription PCR (RT-PCR), 134

Polymorphism, 125

Index502

Polymorphonuclear neutrophils (PMNs), 300, 303, 304

Postweaning, 22, 23, 23, 35, 37–39, 41–43, 45, 46Post-weaning, 482, 483, 487, 489, 490Potentiated probiotics, 454, 458–459Prebiotics, 217, 457–458, 459, 467Prevotella, 126Probiotics, 2, 23, 24, 25, 28, 30, 351, 382–383,

384t, 385, 390–394, 399, 400, 402, 407,409, 454–457, 467

Protection, 323Protein tyrosine phosphatase, 222Protozoa, 119Pseudomonas spp., 382, 386t, 387, 388, 389t, 392,

395t, 396t, 401, 404Pulse-field gel electrophoresis (PFGE), 123Purification (fimbriae), 473, 480

Radiotoxicity, 5Receptor, 472, 474, 476, 482, 483, 485, 486,

486–492Receptor analogue, 472, 485–487, 491, 492Receptors, 159, 162Rectum, 111Regulation of expression, 161Renibacterium salmoninarum, 209Repetitive extragenic palindromic sequences

(REP-PCR), 127Resistance, 472, 476, 477, 491Restriction fragment length polymorphism

(RFLP), 125Ribonuclease protection assays (RPA), 134Ribosomal Database project (RDP), 130Ribosomal intergenic spacer analysis

(RISA), 127Rickettsia, 221, 223Rickettsia prowazekii, 227RNA, 120–123, 127, 129–131, 133, 136Rumen, 54–56, 60, 61, 62, 63, 65, 70, 76, 79–83,

85–87, 89, 90, 92, 95Rumen ciliates, 58, 61Rumen fauna, 55, 61Rumen microflora, 6, 11Rumen protozoa, 57, 59, 70Ruminant(s), 54, 55, 56, 60, 63, 65, 67, 69, 70Ruminants, 76, 78, 79, 81, 82, 84, 86–89, 93–95

S fimbriae, 175Salmo trutta, 221Salmonella, 220, 263–265Salmonella, 127

enterica, 127Salmonella and malnutrition, 249Salmonella enteritidis, 236Salmonella fimbriae, 241Salmonella flagellin, 241Salmonella in calves, 247

Salmonella in the mouse, 241, 243Salmonella in the pig, 246, 247Salmonella in the rat, 238–241Salmonella typhi, 236Salmonella typhimurium, 116, 237Salmonellosis, 103, 114Salmonid rickettsial septicaemia (SRS), 223Salvelinus alpinus, 214, 216, 221Scophthalmus maximus, 216, 221, 222, 224Seasonal changes in body mass, 78, 85, 87, 94Sec, 479Seriola spp. (yellowtail), 225Shigella, 220

adhesins, 220invasin, 220

Single-strand-conformation polymorphism (SSCP), 127

Skin immune system (SIS), 293S-layer, bacterial, 211Small intestine, 91, 108, 472, 473, 476, 481, 483,

485, 488–492Sparus auratus, 222SPF pigs, 39Spirochaetes, 267, 268Staphylococcus, 271Staphylococcus aureus, 105, 214Static model, 146Stomach, 105, 144, 146, 147

kittens, 106puppies, 105

Streptococcosis, 225Streptococcus difficile, 209Streptococcus iniae, 209Structural carbohydrates, See also cellulose and

hemicellulose, 54, 55, 60Substrate-tracking autoradiographic fluorescent-

in situ-hybridization (STARFISH), 133

Suckling, 22, 23, 33, 35, 36, 38, 40, 42, 44, 46Synbiotics, 457–458

T cells, 367, 369, 370tT lymphocytes, 318T-cells (lymphocytes), 298, 299, 301Temperature gradient gel electrophoresis

(TGGE), 128TIM-1, 148, 149TIM-2, 148Tir (translocated intimin receptor), 169Tolerance, 321Toxin, 473, 476, 483, 484, 487, 490, 491Turbot, see Scophthalmus maximus, 216Type 3 secretory system, 214Tyramide System Amplification, 133Tyrosine kinase, 220, 222

Usher, 477, 479, 480

Index 503

V. anguillarum, 211Vaccinations, 179, 325, 472, 481–484, 491, 492Vaccine, 329, 330, 333, 337, 341, 342, 346, 493,

480, 481, 485, 492Verotoxigenic, 483VETEC, 483Vibrio anguillarum, 209, 211, 214, 216, 224, 226Vibrio cholerae, 211, 218Vibrio harveyii, 224Vibrio ordalii, 209, 224Vibrio salmonicida, 209, 211, 214, 226Vibrio spp., 381, 382, 387, 391, 394, 395t,

398–400, 402–405Vibrio viscosus, 209, 211Vibrio vulnificus, 224

Viral Encephalopathy and Retinopathy (VER), 225

Viral haemorrhaghic septicaemia (VHS) virus, 225Virulence factor, 474Volatile fatty acids (VFA), 6, 11, 12, 56, 60, 65, 67tVTEC (verotoxigenic E. coli), 167

Weaning, 472, 482, 483, 487, 489–491

Yersinia, 271Yersinia sp., 215, 220Yersinia ruckeri, 209α-defensins, 297α-lactalbumin (α-LA), 303, 305β-lactoglobulin (β-LG), 303

Index504

Microbial Ecology in Growing Animals

Documents

Transcript of Microbial Ecology in Growing Animals