Diss. ETH No.8150

MICROBIAL DEATH, LVSIS AND •CRYPTIC• GROWTH: FUNDAMENTAL AND APPLIED ASPECTS

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH for the degree of

Doctor of Natural Sciences

presented by COLIN ANTHONY MASON

BSc (Hons), University of Bath, England Born on 12 October 1959 in London, England

accepted on the recommendation of Prof. Dr. G. Hamer, examiner

Prof. Dr. P. Peringer, co-examiner

Zurich 1986

ADAG Administration & Druck AG

The following sections of this thesis have either been published or

accepted for publication:

Chapter 2: FEMS Microbiology Reviews, 39, 373-401, 1986.

chapter 3: Chemical Engineering. Communications, 4S, 163-176, 1986.

Chapter 5: Dechema Monographs, in press.

Chapter 7: Proceedings 4th Italian-Yugoslavian-Austrian Chemical

Engineering Conference, Grado, Italy, 705-711, 1984.

Chapter 8: Proceedings 5th Yugoslavian-Austrian-Italian Chemical

Engineering Conference, Portoroz, Yugoslavia, 549-556, 1986.

Chapter 11: Bioprocess Engineering, in press.

Chapter 13: Proceedings Pro Aqua Pro Vita, 10th International

Exhibition and Technical Meeting for Engineering in

Environmental Protection and Ecology, Basel, Switzerland,

108, 7.1-7.28, 1986.

The articles in chapters 4,6,9,10 and 12 have been sublllitted for

publication and are currently under review.

TO MY PARENTS

Its a very odd thing -

As odd as can be -

That whatever Miss T. eats

Turns into Miss T.

Porridge and apples,

Mince, muffins and mutton,

Jam, junket, jumbles -

Not a rap, not a button

It matters; the moment

They're out of her plate,

Though shared by Miss Butcher

And sour Mr Bate,

Tiny and cheerful

And neat as can be,

Whatever Hiss T. eats

Turns into Miss T.

Walter De La Mare.

I am especially ·;Jrateful to Professor Geoffrey Hamer for his invaluable

advice, support and encouragement throughout this study.

I would also like to express my thanks to:

- Professor Paul Peringer for acting as co-examiner.

- Dr. James Bryers for his advice and cooperation during the early part

of this work.

- Dr. Thomas Egli for his expert help on many issues.

Thomas Fleischmann and Candid Lang for their expertise in much of the

technical aspects of this work.

- Ruth Meisser for much of the typing of this manuscript.

- Dr. Mario Snozzi for translating the summary.

- Dr. Willi Gujer for assistance in computing.

- Dr. Ernst Wehrli for the electron microscopy.

- Ors Baier for bacteriological identification.

- Heidi Bolliger for the technical drawing.

- Paul Schlup for preparing slides and photographs.

- All the staff and students in the technical biology department for

their encouragement, criticism, laughter and for putting up with me.

- The Director and staff at EAWAG for contributing financial assistance,

materials and help.

- And to the Swiss National Programme 7D for financially supporting this

work.

TABLE OF CONTENTS

Olapter Page

l General Introduction l Background l Microbial death, lysis and "cryptic" growth - Fundamental aspects 3 Microbial death, lysis and "cryptic" growth - Applied aspects 5

2 'l'he death and lysis of microorganisms in environmental processes S Introduction S Physiological classification of a microbial culture S Starvation Autolysis Microbial death Mathematical modelling of death and lysis Concluding remarks References

13 23 24 26 29 29

3 Activity,death and lysis during microbial growth in a chemostat 37 Introduction 37 Model development Experimental Results Discussion Conclusions References

4 •eryptic• growth in Klebsiella pneumoniae Summary Introduction Theoretical Methods Results Discussion References

5 Intra-cellular maintenance requirements and the extra-cellular death/lysis/•cryptic• growth cycle

Introduction Production culture physiology "Cryptic" growth in Klebsiella pneumoniae Discussion

6 Survival and activity of Klebsiella pneWIDniae at super-optimal temperatures

Abstract Introduction Materials and methods Results Discussion Conclusions References

3S 43 44 46 49 49

51 51 52 53 56 5S 59 64

71 71 72 73 73

75 75 76 7S so S3 S7 as

Table of Contents (continued)

Chapter Page

12 Bioparticulate solubilization and biodegradation in semi-continuous aerobic tbermophilic digestion 172

Abstract 172 Introduction 173 Materials and methods 175 Results 176 Discussion 178 Conclusions 180 References 181

13 Some fundamental aspects of two stage waste sewage treatment with aerobic thermophilic pretreatment and anaerobic stabilization 186

Synopsis 186 Introduction 187 Microbial process sequences 189 Bio-oxidation process stoichiometry 191 Process definition 192 Continuous flow process predictions 195 Sludge hygienisation 197 Bacterial survival 198 Reinfection and regrowth 199 Conclu~ing remarks 200 References 201

14 General Discussion 209 Microbial death, lysis and "cryptic" growth - Fundamental aspects 209 Microbial death, lysis and "cryptic" growth - Applied aspects 219 References 225

Suamary 227

ZUsaaaenf assung 229

Appendix Al Structure of INT A2 Computer pro9ramme "Oyneq" A3 Computer programme for "cryptic" 9rowth

231 231 232 238

-1-

CHAPTER 1

GENERAL INTRODUCTION

Background

Microorganisms are an essential part of global ecology. Ignoring for the

moment the negative aspects of disease causing microbes, they play

important roles in global nutrient cycling. The ability for microbes to

degrade matter has long been recognised and was used by sanitary

engineers at the end of the last century, more by chance than by design,

to cope with the increasing problems of sewage and wastewater disposal

in industrialised countries.

In Switzerland, the first wastewater treatment plant was built in St.

Gallen in 1916, and this was followed a few years later in the city of

Zurich in two stages, 1924/1926 and 1932/1934.

Nevertheless, in Switzerland there has been a steady deterioration in

the quality of the water in lakes and rivers since the beginning of the

century. It was only in 1953 that the first article describing the need

for organised water pollution control was introduced into the Swiss

constitution, and this, in turn, led to the first water protection law

published in 1957. The present law concerning water resource protection

was introduced by the Federal government in 1971 and has been enforced

since 1972.

-2-

The effect of this legislation was to see the introduction of wastewater

treatment plants for a major fraction of the domestic and industrial

wastewater produced. At present, 75% of the populations wastewater is

treated by conventional mechanical-biological technology, supplemented

with chemical precipitation of phosphates.

The major by-product of this treatment technology is waste sludge, a

complex mixture of particulate, sorbed and soluble matter.

Microorganisms from both primary and secondary treatment stages are also

present some of which are pathogenic. The present laws concerning

wastewater treatment have resulted in an ever increasing production of

waste sludge and while Switzerland is a land locked country, ultimate

disposal has become a serious problem. It was realised in the 1960s that

the infection cycle of water borne pathogens was not necessarily broken

by water treatment processes alone, but that it was also necessary to

treat the by-product sludge before its disposal. In Switzerland, sludge

is normally disposed of by spreading on agricultural land. In 1981, the

waste sewage sludge ordnance was published (1). Article 1, states that

waste sewage sludge can be regarded as hygienised only when it contains

no more than 100 Enterobacteriaceae per gramme of sludge and no

pathogenic worm eggs.

The necei;sity to reduce the numberi; of microorganisms in wai;te sludge

leads to the central theme of this dissertation, i.e.,an investigation

of the demise of microorganismi;. The thesis is divided into two parti;,

the first deals with the fundamental concepts involved in death and

ly~is and the second with the application of these concepts in practice.

It is clear that effective application of microbiology is only realised

when a sound understanding of the fundamental principles exists. The

fundamental side of this work requires the elucidation of processes such

as death and lysis under conditions where these processes are least

expected, i.e., under optimised growth conditions before effects of

primary and secondary stresses, such as either elevated temperature or

starvation, can be meaningfully investigated. Gray and Postgate (2)

describe very succinctly ~hy this area should be of general interest:

- 3-

"A scientific visitor from another planet, studying the state

of Earth's science, might well conclude that a taboo exists among microbiologists regarding the death of bacteria. Pick

almost any standard microbiology textbook and you will find

discussed growth, multiplication, variation, transformation

and all other varied activities which makes this group of

creatures so fascinating. But you will seek in vain for a

discussion of their death; if it is mentioned at all it will

be almost furtively - 'the phase of decline• - and probably

dismissively, as if there is nothing more to be said except

that they die".

The first half of this dissertation seeks to develop a greater understanding of microbial death, whilst the second half looks at one

means by which death can be assured together with effective means of

disposing of .the residual matter.

Microbial Death, Lysis and "Cryptic" Growth - l'undament:al Aspects

A considerable number of research articles and review papers have been

published in the scientific literature describing aspects of the death

and lysis of microorganisms in the environment. Very few review papers

have been published where the two aspects are brought together; the last

major review appearing in 1976 (3). In order to rectify this situation,

the second chapter of this dissertation comprises a review, entitled

"The death and lysis of microorganisms in environmental processes". This

review covers several important aspects of the death and lysis of

microorganisms, namely;

-4-

1). It establishes suitable definitions for a physiological

classification of microorganisms.

2). A critical examination of the suitability of methods used to

evaluate the presence and abundance of microorganisms is given.

3). The prerequisites for survival, i.e., those properties that a

microorganism must possess to survive adverse conditions are

discussed with emphasis placed on the ability to survive nutrient

deprivation.

4). The physiological basis of lysis is examined from the point of view

of either autolysis or self-induced lysis.

5). The manner in which a microorganism can be forcefully killed, and

the physiological basis for their death are discussed.

6). The application of mathematics to assist in understanding and

solving the intricacies of the area are discussed.

In the third chapter, a

interrelationships between

model

the

is developed which describes

various physiological groups

the

of

microorganisms described in the first section of the review in chapter

two. The model is calibrated using data from both the literature, where

available, and from experimental results. The results indicate that the

processes of death and lysis probably occur simultaneously in unstressed

cultures.

For the fundamental experiments, the bacterium Klebsiella pneumoniae was

chosen as the test microbe due to the fact that a significant amount of

kinetic and physiological data is available for this bacterium growing

in both batch and continuous culture. The bacterium is found in

wastewater and sewage sludge and is potentially pathogenic to man where

it can cause acute pneumonia and diseases of the urinary tract.

In cultures of microorganisms, the utilization of lysis products by

members of the same population as carbon or other nutrient source, is

known as "cryptic" growth. Two papers discussing the concept of

-s-

and of product formation in pure and in mixed cultures are provided in

chapters four and five. Whilst not producing absolute evidence that

"cryptic" growth occurs in growing cultures, the experimental section

provides a strong indiction that this is the case. Verification of

"cryptic" growth processes in cultures would be a natural step in the

development of this area and will have to be approached in the future.

Whilst the sections mentioned so far establish some fundamental concepts

concerning the death, lysis and "cryptic" growth of bacteria under

optimum growth conditions, chapter six looks at the induction of death

and lysis in a culture by subjecting it to super-optimal temperatures.

This effectively links the fundamental aspects of microbial death and

lysis with the applied aspects of aerobic thermophilic waste sludge

treatment which is discussed in the second half of this dissertation. It

is also shown in chapter six that many of the concepts which have been

developed concerning the physiological mechanisms for survival may be

misleading since the units used to describe changes in the population as

a whole do not permit direct extrapolation to describe effects on single

microbes when the consequences of death and lysis are not considered

simultaneously. Experimental data are shown whereby, as a result of

either starvation conditions or of heat shock, the amount of a

macromolecular cellular constituant in the population as a whole

declines but remains constant when expressed on a per cell basis.

Microbial Death, Lysis and "C9ptic" Growth - Applied Aspects

The remainder of this thesis is devoted to the applied aspects of

microbial death, lysis and "cryptic" growth, in particular during

aerobic thermophilic waste sewage sludge treatment process. By means of

general introduction, the short review article presented in chapter

seven was prepared at the start of the experimental period to assess

what was already known about the process and to consider to what extent

microbial solids destruction is dependent on predetermined variables.

-6-

For experimental purposes, aerobic thermophilic sludge treatment was

examined principally with respect to microbial solids destruction. Thus

a reproducable feed of predetermined composition was used. In the

experiments reported here, the feed consisted of pressed bakers yeast,

resuspended in a nutrient solution. In chapters eight, nine and ten the

effects of several process variables are assessed with respect to the

rate and extent of microbial solids removal. In chapter eight the

effects of pH are discussed, in chapter nine, the effect of dissolved

oxygen concentration, and in ten the effect of temperature. The effects

of these three variables are discussed in each case with respect to the

probable biochemical mechanisms involved in microbial solids

destruction.

A more detailed attempt to characterise the mechanisms involved in the

solubilisation and biodegradation of microbial solids in the aerobic

thermophilic treatment process is presented in chapter eleven, where

the amounts of various components of both the substrate microbial cells

and their breakdown products and by-products were followed during a 72 h

batch experiment. This was also combined with a scanning electron

microscope study of the cell destruction porcess. From the data, it was

possible to construct both descriptive and mathematical models to

describe the events involved in microbial solids destruction, and thus

provided a framework to better understand the probable effects of

varying the process operating conditions on process efficiency.

Two operating modes are currently used in technical-scale aerobic

thermophilic waste sludge treatment plants, i.e., continuous flow and

semi-continuous flow (fill and draw), and the data shown in chapters

eight, nine and ten are specific to the continuous flow mode of

operation. However, a study was also made to investigate the factors

influencing microbial solids degradation in the semi-continuous flow

mode of process operation. The results are discussed in chapter twelve.

-7-

Whether aerobic thermophilic digestion can satisfactorily supplement

traditional waste sewage sludge treatment and the limits of effective

operation are dfscussed in chapter thirteen. In this article, it is

argued that aerobic thermophilic digestion cannot operate as a complete

waste sludge stabilization process but nevertheless offers distinct

advantages as a pre-treatment stage when used in conjunction with

anaerobic digestion.

The final chapter summarizes the findings described in the individual

chapters, and discusses their consequences in the fields of

environmental biotechnology and water management.

References

1. Schweizerisches Bundesamt fur Umweltschutz (1981)

Klarschlammverordnung 814.225.23, 20pp, Bern.

2. Gray, T.R.G. and Postgate, J.R. (1976) Editors preface. In: The

Survival of Vegetative Microbes, Proc. Symp. Soc. Gen. Microbiol.,

26, pp ix-x, Cambridge University Press, Cambridge.

3. Gray, T.R.G. and Postgate, J.R. (1976) (Eds.) The Survival of

vegetative Microbes. Proc. Symp. Soc. Gen. Microbial., 26, Cambridge

University Press, Cambridge.

-8-

CHAPTER 2

FEMS Mit:ruhiology Ri:vicws J9 ( 198(,) 37)-401 Puhfo.hi:J hy Els,wii:r

FER tl0044

373

The death and lysis of microorganisms in environmental processes (Death; lysis; survival; starvation: maintenance: 'cryptic' growth: environment)

C.A. Mason, G. Hamer and J.D. Bryers •

ln.~1itu1e of Aquutit' Sciencl!.f, S•·iu f"'etlt:rol butimre of TechnolliXY Ziirtch, Utberlmul~lm.~e 133, ('fl.IJt,00 Dlibem/,,,.f. S••it:trltmJ. t1nd " Depu11men1 cf Cit•il und Ent•lnmmemul Ettgineerintf. Dul.·v U11irer:rl1_r, DurJ1,1m. NC 177fJ6, U.S.A.

Rec~lved 24 March l 986 Accepted 5 May 1986

l. INTRODUCTION

One of the major philosophical stumbling blocks in microbiology relates to the question of ageing. For most macrobes. the processes of birth. growth and death are tangible. observable events. The same cannot be said for the majority of microbes. Exceptions such as some rilamentous microbes (e.g .. Sphaerotilus spp.). some budding bacteria such as llyphomicrobium spp. and yeasts which show effects of ageing by bud scars are well known. However, the question of the ultimate destiny of a particular microbe present at any instantaneous moment in time has as yet to be answered.

Very few papers exist that deal directly with microbial death, ahhough a large amount or cir· cumstantial information is available. Understand· ing more about the processes of ageing. death and lysis in microbes has relevance in the following areas: (I) Public health sector. Erricient testing for the presence or and mechanisms for the destruc-tion of pathogenic organisms in food/ feed and water for human and animal consumption. (2) Biological industries sector. Maximising the per-cemage of active strains with respect to total numbers of microbes. manipulating culture condi-tions to affoct desired optimal system stoichiome· try while maintaining activity and preventing pro-

cess inhibition. (3) Medical sector. Efficacy of antimicrobial agents and qualitative disease as-sessment. In order to deal with such problems, an understanding of the intrinsic physiological mech-anisms involved in (a) induced lysis, (b) autolysis (non-induced lysis). (c) death, (d) resistance. (e) dormancy and (f) survival is necessary.

This review will look at certain aspects of the quantification or microbial death and lysis in terms of both the methods available to investigate the phenomena and the physiological means of defer. ring them. and look at changes which may occur as a result of death or lysis within a population. that may enhance the survival pauern of the re· maining microbes.

2. PHYSIOLOGICAL CLASSIFICATION OF A MICROBIAL CULTURE

2.1. Deft11itio11s

A culture of microbes, either in the laboratory or in their natural environment. is composed of various morphological. biochemical and physio-logical groups. TI1e existence of monocultures is very rare in natural environments, even under conditions where the environment requires special-ised forms such as in some thermophilic and/or

0168·6445/K6/S10.U ~-, l9H6 Fcdi:tution of Eu~1pe.-n Micmhiologicat S;,"-'i~ti~~

-9-

374

acidogenic environments. Biochemical variance arises from the efficient interaction of different microbes in either a food or an energy network, such as in the reduction or various chemical species depending on the redox potential of the environ-ment [lj. Microorganisms have been classified on a physiological basis as: (a) dead microbes; (b) non-viable, active microbes; (c) dormant micro-bes; (d) viable, active microbes.

2.1.I. Dead microbes Strictly defined, these are organisms totally de-

void of metabolic activity, but still possessing a cell wall [2J. Everything failing to meet this description must not and cannot be dead. Defini-tions based on the inability to reproduce [3! are false and lead to embarrassing and dangerous misinterpretations of data. Experimentally, it is almost impossible to determine quantitatively the existence of such cells. Consequently, their pres-ence has to be inferred retrospectively from esti· mates of the total cell numbers and numbets of cells railing into the other categories listed above [ 4) for which more precise methods of quantifica-tion have been developed.

This problem has been compounded by the wide range of different methods available for quantifying microbes. This issue will be addressed in a subsequent section in this review. Dead cells have been treated as inert solids in the considera-tion of particulate degradation in continuous cul-ture systems (5), although they are effectively bio-degradable particulates. As such, dead cells may constitute a very large fraction or the total bio-mass present in trickling filters and activated sludge wastewater treatment systems, in cell re-cycle processes, (i.e., ethanol production) and in semi-continuous (fill and draw) fermentations where a significant portion or the biomass is re-tained as inoculum for each new process cycle.

2.1.2. Non-uiab/e microbes These are organisms which have lost the ability

to reproduce. These result from genetic defects such as absence of a critical enzyme necessary for replication, or lethal breaks in the DNA of the microbe. However, such microbes can carry out substrate transformations when they possess ap-

propriate enzymes. For example. at superoptimal growth temperatures it has been shown that sub-strate energy dissipation can be mediated by non-viable cells (6]. The use of non-viable microbial enzymatic conversion is the basis for immobilised whole cell biocarnlysts [7].

2.1.J. Dormant microbes These can be subdivided into two categories:

(1) Spores; (2) temporarily inactive or resting microbes. These two groups diner in that spores are morphologically differentiated structures. Dorma111 microbes are found in a wide range of environments and may represent the largest class of naturally occurring physiological phenotypes: However, they are frequently confused with non-viable cells, due to the difficulty of promoting their growth using artificial laboratory stimuli. These two types of dormant cells also differ from one another functionally, spores serve distribution and survival functions. whilst resting microbes are only intermediates leading either to active micro-bes or to death or lysis [8).

2.1.4. Active microbes These are cells which can actively assimilate

substrate, increase in mass and replicate. Active microbes are most commonly found in laboratory environments, although they are also frequently the major physiological class found in industrial and technical microbial processes. More informa-tion is available on this category than on any other.

2.2. Analytical met/rods for physiological differentia-tion

The ability to grow and multiply is very often the only criterion used in differentiating between the groups mentioned above. Most working defini-tions for the various types of microbes are restricted by the lack of accurate methods with a sound theoretical basis, together with a fanatical adherence to historically proven inaccurate meth-ods. Use of arcane analytical methods has been perpetuated by public health authorities. despite a wealth of evidence as to the limitations and dangers or such techniques. Some of the current

-10-

methods for assessing cell numbers and activities are discussed in the following section. For more deluils on melhods, the reader should consult spe· cialiscd reviews (9-1 JJ.

!.l. I. Cell c11/rioalio11 mc1hods The use of agar as a solidifying agenl in culture

media was first proposed in 1881 [14]. Sin~'e then, its use has escalated, and today it represents the most universal microbiological technique for the cultivation of microbes (15]. Theoretically the numbers of microbes in a sample can be derived from the numbers of colonies growing on the surface of an agar solidified medium in a Petri dish (16), although it was originally intended that the use of such media should be for the isolation and growth characterisation of microbes. Despite the repeated and often vehement criticism which has appeared in the literature [ 15-19), agar colony counts continue to be used extensively for the enumeration of microbes for estimation of survival (20-23], in stress studies [24-26], and perhaps most surprisingly in public health (27-30], even though alternative accurate techniques are availa-ble (31).

It is now universally accepted that in most natural microbial environments only a very small fraction of the microbes present can be enu· merated using the agar plate technique [18,32-34]. Estimates of this error in the literature vary de-pending on the environment. In soil for example, it has been claimed that only 1-10% of the true number of viable organisms are enumerated [35), whilst in seawater values below 0.1 % are con-sidered normal (33.36]. In the testing of water supplies, as many as 90% of the coliforms, the bacteria used to indicate the presence of patho-gens, may not be enumerated [37).

Some of the problems associated with agar plate counting or microbes include the lack of a single universal medium which will allow growth of all organisms [38), since most organisms are sensitive to the type or medium used [17,39-43]. In the literature 'improved' media for the growth of microorganisms are frequently reported, sug-

. gesting that existing compositions are inadequate. This process or new medium formulation is never-ending, and although improvements are

375

continually being made, an acceptable status will never be reached. The technique has to be rede· fined for use only under those circumstances for which it is suitable. Despite these criticisms it should also be noted that if the method is used, interpretation must embrace all or the limitations of the method, if the results obtained in routine screening work are lo be valid.

Failure to grow on an agar surface has been ascribed to environmental fastidiousness, dorman· cy [21), inhibition by neighbouring cells (16). physicochemical differences between the labora· tory and natural environment [38]. clumping (9) and to the fact that stressed and injured cells may have some difficulty in reproduction (44) since sublethal damage and inactivity are not dis· tinguishable (21,45]. Further comment on the widespread misuse of agar-based cultivation media is superfluous, and the reader may consult the reviews of Buck (17) and Fry (18] for more de-tailed discussion.

1.2.1. Aclioity measurements Metabolic activity, as measured by the rate of

oxygen consumption, has been widely used in microbial ecology for the assessment of microbial activities in different environments [46). Electron transport system (ETS) activity measurements are usually used as a guide to metabolic activity. due to their relative simplicity. The ETS is mediated by the action of several dehydrogenase enzymes such as succinate dehydrogenase, and direct mea-surement of dehydrogenase enzyme activity is as-sumed to be a reliable indication of ETS activity in a specific environment (47,48). A large propor-tion of the total metabolic activity has been shown to be linked to ETS activity (49]. The most fre-quently encountered method for ETS activity as· sessment involves the reduction of the tetrazolium salts 2,3,5-triphenyltetrnrolium chloride (TIC), 2,2' -di·p·nitrophenyl-5,5-diphenyl-3,3' -dimethoxy-4,4' -diphenylene (NBT) or 2-(p-iodophenyl)-3· ( p-nitrophenyl)-5-phenyltetrazolium chloride (INT) to insoluble formazan compounds [50,.51], a technique pioneered by Lenhard (52] for the as-sessment of bacteria in soil. The tetrazolium salts compete with oxygen for electrons [50,53]. Since ETS activity is common to virtually all microbes

-11-

376

(48,49] the technique has provoked increasing in· terest, and since it is applicable under both aerobic and anaerobic conditions [46,49,50) it can thus be used for samples from most environments.

Most of the methods described involve disrup· tion of the cells after incubation with a tetra· zolium salt [54-56] followed by solubilisation of the formazan using. typically, Triton X·lOO (57,58) followed by measuring the absorption at 490 nm {46,47). Direct microscopic e.xamination has also been used to give a more precise estimation of the number of active cells based on the assumption that only living cells will contain the formazan crystal (59-61). However, it has been suggested that not all bacteria present in natural environ· ments are actively metabolising. and may be in a state of dormancy (60,62,63); that not all bacteria are capable of tetrazolium salt reduction [48); and that tetrazolium salts may suppress ETS activity (48). Addition of substrate has also been shown to affect the results, although some disagreement still exists as to the necessity of substrate addition. Tabor and Neihof (49) found no increase in counts of microbes from water samples when succinate was included in the INT·incubation tubes. How· ever, Trevors ·(48) found an increase of 100% at 4°C and 327% at l0°C in water samples when substrate was included, and Bright and Fletcher (64) obtained increased counts when the growth substrate (leucine) was included in their bacterial counts in a series of attachment studies. Work in this laboratory (unpublished results) has also clearly shown an enhancement of active cell num· bers after inclusion of an ETS-activating sub-strate.

Various authors have commented on the prob-lems of detection of small microbes which may be present in natural environments (65) and have suggested modifications to the basic technique [19,49,66) while other authors have chosen differ· ent compounds (67). Much attention has also been given to the other methods for assessment of metabolic activities in natural environments, in-cluding the measurement of uptake rates of ra· diolabelled substrates (681 and measurement of production of metabolic products. A technique involving direct examination of cells whose nucleic acid synthesis has been repressed without ap-

parently affecting their growth has also been sue· cessfully used (36]. The measurement of in situ metabolic activity was the subject of a recent review by Findlay and White (69].

2.2.J. Direct counts Information concerning the physiological state

or a microbe is usually expressed with respect to the total numbers of microbes present (18). De-spite the persistence of some agar based counting procedures. most total counts are presently carried out by microscopy. The use or bright-field mi· croscopy is rare and has limited application [70,71), as has phase contrast microscopy (72). However, epifluorescence (incident light fluorescence) mi-croscopy has been successfully used for enumera-tion purposes [73). Cells are counted on a mem-brane rilter after filtration of a known sample volume. This technique was first developed by Strugger [74) and has since been modified and improved by the introduction or different filters [75,76) and microscope lamp/filter arrangements [77). Various stains have been employed, including fluorescein isothiocyanate (FITC) for soils [12.78) and acridine orange for aquatic samples [21. FITC reacts specifically with proteins and fluoresces green (79). Acridine orange stains nucleic acids, and the technique is very good when used for enumerating total cell numbers, but does not dis-tinguish between live and dead cells (71,80). It has been shown, for example, that acridine orange is still taken up by autoclaved cells [81]. By control-ling the pH it has been suggested that dirrerentia· lion between live and dead cells is possible [82) although experimental verification has yet to be reported. A further restriction or this technique is that acridine orange does not distinguish between dormant and growing cells [2). The accuracy of the method has been examined by comparing counts using epifluorescence and electron microscopy [83J. Using the scanning electron microscope, the counts were found to agree, but epifluorescence gives higher counts than can be obtained with transmis-sion electron microscopy ( 64!.

Other fluorescent compounds have also been used and some claims have been attached to their ability to accurately determine numbers of ac-tively metabolising microbes. Rhodamine 123, for

-12-

example, relies on the existence of a proton motive force for uptake (85] although its use so far has been shown to be restricted to Gram-positive cells. Similarly, fluorescein diacetate (FDA) has been used for enumerating freshwater microbes, where this non-fluorescent compound is attacked within the cells by non-specific esterases [86) with the resultant release of fluorescent products (87]. However, FDA also has dirficulty in penetrating the Gram-negative cell wall [81 l and its effective use is most probably restricted to mammalian cells [88J, yeasts [89) and some cyanobacteria (12j. The stain nuorescamine has also been used success-fully to differentiate between microbial cells and detritus particles in marine samples, due to the high affinity or the stain for amino groups [90].

7~ 6 ·~-·---·-----· 5 4

3

S! 2 en 2

8 e 7

"' 6 ..... ...J 5 w u 4

3 2

I

:::--. .......... ·---·

A

~ ····-· B

0 4 8 12 16 DAYS

Fig. l. Detection of E. <t>li hl0407 i110C11la1ed into (A) aged estuarine water where the pH and salinity were adjusted lO simulate environmental condhiont; and (B) in situ experiments in ~mitropical water using membrane chambers. 0. Acrid inc orange direct coont; C acridine orange direct count control; • fluorescent antibody counts~ •· standard plate count; o, eosin methylene blue agar count; •. dire¢t viable count (nalidixic acid). Redrawn from Jl4) with permission.

7

5

2 ....... ~.---·-· ~4

e -'1'> ...J ...J w u 2

2 4 6 DAYS

371

8

Fig. 2. Detection of V. cholera• CA40t exposed to Pa1uxen1 River water in microcosms. 0, Acridine orange direct count~ "· noorescent antibody count; II. direct viable count (nalidix.ic acid); A, thiosulpate citrate bile $UCCOSC agar coun1~ •. tryptic Soy agar count Redrawn from [34) with permission.

Epifluorescence microscopy using acridine orange has also been combined with autoradiogra· phy for the detection of metabolising cells f91 ). Another approach in total cell enumeration is that of fluorescent antibody labelling. The staining of cells by fluorescent antibodies allows the observa-tion of particular strains in a mixed culture (92,93) thus allowing a direct count of individual species. However, the technique does not provide any in-formation as to the physiological state of the cells. The use of fluorescent antibody technique for public h~alth testing for the presence or patho-genic organisms is currently being investigated. A recent paper clearly shows the advantages of this technique over the presently used standard meth-ods [34]. Their paper reports results from water samples from Bangladesh, tested for the presence

-13-

378

7

4 8 12 16 20 DAYS

Fig. J, ~lection of Shi~/a 1tJ1tnti 530 inoculated into aged l!s.tuarine water where the pH and salinity were adjusted to simulate environmental condittoos. 0. Acrkline orange direct count: •· fluorescent antibody count; a direct viable count; •. tryplic soy agar count; A. MacConkey agar count Redrawn from {341 with permission.

Table I

Evaluation or microbial enumeration methods

of Vibrio dwlcme. Using conventional techniques. 7 or the 52 sample.' tested proved positive. fly the fluorescent antibody technique. 51 out of the 52 were positive. Similar comparisons arc made be-tween acridine orange direcl counts and plalc counts (Figs. 1-3). Clearly. these direct counl methods are beuer suited for the enumeration of organisms, especially where potential public health hazards exist.

A summary of the various conditions for and enumeration possibilities of rhe different methods available is given in Table l.

3. STARVATION

Under laboratory conditions, microbes are gen-erally grown in artificially rich environments the like of which are rarely found in nature. Unlike laboratory culture environments, organisms in natural environments are subjected to variations in substrate availability, (i.e., nitrogen, phos-phorus), in temperature, oxygen, toxic chemicals as well as to spatial variations, and if the cells are attached, differences in substratum composi1ion may exist. Despite these condi1ions, indigenous microbes readily survive in their natural environ-

Act. Non-Rep. and Dorm refer lo the physiologically differentiated cell types discussed in section 2. L Parentheses imply that only .tt part of this population was enumera1ed.

Method Time required Environment Requirement Cell types Differentiation At.wracy for test of test for ccH tnumerated between cell types

PhueCount 24-48 h Modified Yes (Act), (Dor) Yes Low MPN 24-48 h Modified No Act. Non-Rep. (Yes) Low

(Dorm) Activity

ETSONn I h Originol/ No Act. Non·Rep. No High modified (Dorm)

ATP 2h (Cell extract) No Act, Non·Rep No Unknown Oire<:t

AODC/FITC JO min Original/ No All No High modified

FDA 2h Original/ No Act. Non.Rep, Ycs/noo1 Unknown modified (Dorm)

stide culture 24-48 h modified Yes Act. Non .. Rep, Yes/no Low (Chum)

lmmunoOuorescc:nc~ 2h Origimd No All Yes Very high

-14-

ments. Therefore, mechanisms obviously exist conferring the necessary properties to compete and survive. In this section those mechanisms by which microbes compete and survive under varia· ble nutrient conditions (i.e., feast/ famine or starvation) will be discussed with emphasis placed on the modes by which cells are able to defer the death process.

In order lo grow, a microbial cell requires a carbon source, an energy source, and nutrients for biomass synthesis and metabolic regulation. How· ever. it is known that bacteria are able to survive, sometimes for very long periods, in the absence of any or all of these requirements. As pointed out by Morita [65] many publications deal with short time survival (days to weeks) or with survival under specific stress conditions. Stevenson [63] suggested that in most aquatic environments a significant proportion of the bacterial community can be described as being physiologically dormant. Similarly, in soil there is evidence to suggest the dormant bacteria outnumber the active ones [94).

For a cell to survive during starvation, only a very small part of its metabolic potential needs to be expressed. These have been collectively referred to as maintenance functions, and include mainte-nance of osmotic potential, turnover of essential cell materials, and maintenance of the membrane potential. If energy for these processes is not provided, it is said that the cell will irreversibly cease to function. Microbial cells are biochem· ically sophisticated, and many of the. inter· mediates of the chemical reactions have higher free energies than their original substrates. Energy must be supplied to counteract a natural tendency towards disorder and the energy required to main-tain the basic requirements of cellular activity is termed maintenance energy [95). Maintenance en-ergy in the absence of an exogenous energy source has to be derived from the oxidation of either endogenous cellular constituents or storage prod· ucts. This degradation is known as endogenous metabolism and can be defined as the summation of all metabolic reactions which occur when a cell is deprived of either compounds or elements which may serve specifically as exogenous substrates [96}.

The theories resulting in the development of the concept of maintenance are complex. Beauchop

379

and Elsden (97) introduced the concept Y,/ATP relating the mass of cells produced per mol ATP obtained from the energy source in the medium. Theoretical calculations of Y,/ATP can be made under anaerobic growth conditions, since the catabolic pathways for anaerobic breakdown of substrates are known [98]. Despite original theo· fies tO the Contrary f,/ATP is now known not IO be a constant for different microorganisms [99]. Fur· thermore, experimental Y,/ATP values are nearly always much lower than those based on theoreti· cal calculations {100). Much of the early work was carried out in batch culture in a dynamic environ-ment such that yield values were affected by physico-chemical changes {101]. In chemostatic culture, yield values can be obtained under condi· tions without such variations. One method is to use the ratio between the specific growth rate and the specific rate of substrate consumption, i.e.,

r,1,-p./q, (l) Where Y,;. is the microbial biomass yield coef·

ficient (mass of cells produced per mass of sub-strate utilized), JI. is the specific growth rate con· slant (1- 1) and q, the specific substrate consump· tion rate (mass substrate consumed per mass of biomass per unit time). When this is carried out over a range of different growth rates a plot of µ against q, will give a straight line, which when extrapolated fails to go through the origin. The implied assumption is that as the growth rate decreases to zero the value for the specific sub· strate uptake rate tends towards a positive value. Pirt [102) explained this with his theory of mainte· nance energy, and deduced that the consumption of substrates was partly for growth dependent processes and partly for growth independent processes. Stouthamer and Bettenhausen [103) ex· pressed this mathematically as: qATP - µ/f,/ATP

or

qATP - µ/ r~:,.. + m.

(2)

(3)

where qATP is the specific rate of ATP production (mo! ATP· g- 1 dry wt.· r 1), Y,;ATP is the molar growth yield for ATP, and Y,~:,.. is the growth yield per mol ATP corrected for the energy of

-15-

380

maintenance (m,). ·Values for the maintenance energy requirements have been published by Stouthamer (98]. However, Neijssel and Tempest [I 04) question the basic assumption made by Pi rt (102], namely that the maintenance rate does not vary with growth rate. Instead, these authors pro-pose that the maintenance rate does vary with growth rate. This led Pirt (105) to propose a modified model, whereby a growth rate-dependent maintenance term is included in Eqn. I for specific substrate utilisation, i.e.,

q, - µ./ Y,1, + m1 + m'(l - kµ.) (4)

where m 1 is the constant maintenance energy coefficient m', the growth rate dependent mainte-nance energy coefficient when µ. - 0 (mass sub-strate per mass cells· t- 1) and k is a constant. Thus the expression m'(l-kµ.) is the growth rate-dependent maintenance energy, and m' its value when µ. • O {105). This treatment of the mainte· nance energy concept was similar to an approach suggested by Neijssel and Tempest [104). They later clearly expressed their interpretation or maintenance energy terms by stating that the maintenance energy rate and maximum growth yield value derived from linear regression analysis of either yield or metabolic rate versus growth rate are essentially mathematical constants and not biological constants. In a strict physiological sense, both may vary with growth rate (106].

One of the major problems in experimental determinations of maintenance energies is that when growing the cells at a very slow growth rate it is almost impossible to attain steady-state con-ditions, due to non-steady-state medium addition and spent medium removal rates, etc., such that experimental verification that the q,/µ. diagram does indeed tend naturally towards zero at lower growth rate exists and thus the conclusions de· rived cannot be satisfactorily resolved (107).

Microbes possess two mechanisms by which energy (ATP) can be produced Crom the oxidation of an energy substrate. The first of these is sub-strate level phosphorylation where the production of ATP is catalysed by soluble enzyme systems within the cell cytoplasm. The second mechanism is oxidative phosphorylation whereby ATP synthesis is coupled to electron transport reactions

which are driven, in most cases, by the oxidation of either organic compounds or of inorganic ions of negative redox potential, with concomitant re-duction of electron acceptors with higher redox potentials (108). The mechanism by which ATP is produced in oxidative phosphorylation is almost universally agreed to be based upon the chemi-osmotic theory or Mitchell [109). Synthesis of ATP is catalysed by anF0 F1 A TPase enzyme located in microorganisms in the cytoplasmic membrane {108), although it is generally considered that large portions of the membrane itself can be considered to be 'energy transducers' (110). The theory in its simplest form states that these energy transducing systems act as electrogenic proton pumps and translocate protons across the cytoplasmic mem-brane. Since the membrane is cCfectively imper· meable to OH - and H + ions the result or their translocation is the generation of a proton gradi-ent (A pH) and since they are ions, of an electrical gradient (AY,) between the cytoplasm of the cell and its immediate environment. Consequently the cell interior becomes alkaline and electrically negative with respect to the exterior. Since both gradients exert an inwardly directed force on the protons, this force can be expressed by the sum of these two components, namely:

Ailw • M - ZApH (5) where djlw is the proton motive force (mV) and is a measure of the combined chemical and electri· cal forces acting on the protons, 4Y, is the electri-cal potential difference across the membrane, A pH is the pH difference across the membrane and Z - 2.3 RT/ F, where R is the gas constant, T, the absolute temperature and F, the Faraday con-stant. The factor Z converts the pH gradient into m V. The Ailw generated by electron transfer is used to drive an A TP-hyrolysing proton pump in reverse, i.e., in the direction of ATP ·synthesis. Thus, the energy transducing membrane contains two proton pumps, one driven by electron transfer and one driven by ATP hydrolysis.

The energetic activation of the membrane is supposedly a major regulatory mechanism in the physiology of the cell. Its possible influence in the action of autolysing enzymes will be discussed in section 4, but it has also been implicated as a

-16-

regulatory control for other physiological func· tions (111]. The proton motive force supplies the energy for flagella movement [U2), solute trans-port, [113), ppGpp breakdown {114], nitrogen fixa· tion {115], DNA transport (116), and pH homeostasis (117]. The processes driven by ATP are different from those driven by the proton motive force. Since ATP is a general intermediate in biosynthetic processes, supplying energy for the transport of some solutes, it can also be used to generate a proton motive force (118]. The interde-pendence between Ll,P and ATP is one mechanism for metabolic control. In the presence of a high NADH concentration resulting from a high sub· strate flux, Llilw will be enhanced and ATP will be produced. When the ATP content is high. and a low proton motive force is operative, the latter can be regenerated by ATP hydrolysis, thus di· verting ATP away from biosynthesis.

Tempest and Neijssel [106], in a critique of maintenance energy suggest that most of the en· ergy required is necessary for the maintenance of an ionic gradient. This suggestion is based on experiments on the K + ion concentration gradient in Klebsiella aerogenes. It was demonstrated that the transmembrane K + gradient increases with increasing growth rate (since the K + requirement of the cells also increased) which was accompa· nied by concommitant increases in either the specific respiration rate or the oxygen consump-tion rate of the culture (119]. Calculating the amount of 0 2 required to maintain the ionic gradient from the specific respiration rate at each of the different growth rates, the authors demon-strated that a plot of q0 , versus growth rate constant would then indeed go through the origin. Tempest and Neijssel thus conclude that more than 90% or the maintenance energy requirement for glucose-limited cultures of Klebsiella aerogenes is necessary for the maintenance of membrane ionic potential. Since the extracellular K + con-centration diminishes as the growth rate and in-tracellular K + concentration increase, then the maintenance energy must vary with growth rate even in carbon substrate limited cultures with a 10-fold excess or K + [106].

Maintenance energy has also been associated with solute uptake. However, this function can

381

also be related to membrane function, and there-fore, to the chemiosmotic theory. Accordingly, there are several mechanisms by which solutes can be 1ranslocated across the cytoplasmic membrane. Most solutes are transported by the so called secondary transport systems which can be either passive, i.e., without the interaction of specific membrane proteins, or active, whereby such medi· ation occurs [111]. As such a process requires energy, a 'driving force' is derived from the proton motive force of the cells which in general can be described by the expression:

([A]in)

Z log (A)..,, + (n + m)Ll,P- nZLlpH (mV)

(6) where m is the charge of .the solute A and n is the number of protons translocated in the symport.

More recently, Michels et al. (120) have de-scribed an 'energy recycling model' which in es-sence is the reverse process of secondary transport of solutes, whereby the energy of an electrochem-ical product gradient is converted into the energy of an electrochemical proton gradient. Such sys-tems have been detected in the homolactic fermentative Streptococcus cremoris and in Escherichia coli when the microbes are growing fermentatively and therefore have no means for proton extrusion by functional electron transport systems [1 21 ].

Another so called 'maintenance function' is microbial motility, but the motor for flagella mo-tion is not an A TPase as has been demonstrated by the continued motility of microbes even after their ATP pool has been significantly reduced (115), the energy for both flagella motion [122) and the signal mechanism in chemotaxis (123) being derived from the proton motive fo~ (pm!). This has been shown to be the case by using a model system of ghost cell envelopes on which flagella rotation could be initiated by imposing a pH gradient across the envelope membrane (112]. There is also the suggestion that there is a tight stoichiometric coupling between proton transfer and flagella rotation [ 113), although there appears to be a threshold pmf below which llagellar mo-tion is impossible (124).

-17-

)32

Three questions concerning survival need to be answered: (i) How can a microbe in a non-hostile, neutral environment without nutrient source sur-vive for extended periods? (ii) ls energy lost via the non-growth associated membrane functions in microbes with a nutrient nux sufficient to meet its growth requirements? (iii) What effect do such energy losses have on the growth/ survival be· haviour of microbes in which they are occurring? Obviously two different points of reference are indicated. In the first question, natural environ-ments are envisaged, whilst the second and third questions relate predominantly to microbes in technical processes and laboratory systems.

Under starvation conditions, the microbe no longer receives an adequate nutrient flux and is compelled to divert its energy towards specific survival functions. Cellular replication is not pos-sible due to the high energy requirement for synthesis of new cell biopolymers. Energy has to be diverted to the so-called maintenance func-tions. In order to achieve this, the cell has at its disposal several possible mechanisms. Initially if the cell has been under a feast regime, it will contain an excess of polymers such as RNA which are now superfluous for metabolic sustenance. Oxidation of these compounds together with the mobilisation and oxidation of any reserve poly-mers constitutes the observed effect known as endogenous metabolism. However, this step is only the start of the process of survival under starva-tion conditions.

A theory has recently been proposed explaining the relationship between substrate consumption and biomass production at low growth rates {125-127). In order to prevent the 'apparatus ef-fect' whereby substrate addition to slow-growing cultures becomes pulsed as opposed to continuous thereby changing substrate uptake characteristics [107), a recycle reactor was used to grow cells with very long residence times. Il was found that the microbe used ( E. coli) went through various 'phases' before activating the 'stringent response', as the residence time (reciprocal dilution rate or reciprocal growth rate constant) was extended. The stringent response is a series of biochemical adjustments resulting from the limitation of amino-acyl-tRNA. As a result the cell makes a

number of major readjustments of activity in· eluding (I) a reduction in RN A synthesis and accumulation; (2) increased protein turnover; (3) reduced membrane transport; (4) reduced endog-enous synthesis or nucleotides; glycolytic inter-mediates, carbohydrates, lipids, fatty acids, poly-amines and peptidoglycans; (5) increased control ('kinetic proof reading') over protein translation and (6) cAMP accumulation (I 28]. This reaction results from the accumulation of guanosine 5'di-phosphate 3'diphosphate (ppGpp) due to an idling reaction of the ribosomes and uncharged tRNA via a protein known as stringent factor [129). A threshold concentration of ppGpp must be ex-ceeded for the response to manifest. Stouthamer (127] suggests that as the growth rate of the organisms slows down three phases can be dis· tinguished, the first with sufficient tRNA, fol-lowed by a phase in which aminoacyl tRNA be-comes limiting and ppGpp formation begins, and finally a phase where the stringent response is manifested, i.e., where cellular metabolism is put in check at the expense of some limited energy expenditure (from cAMP and ppGpp formation and from the proof-reading of proteins). Whether this energy expenditure is equivalent to the main-tenance energy is not yet clear. This theory is one possible mechanism for survival in nutrient poor environments, whereby the accumulated cAMP may be used by the microbes to activate a range of enzymes when suitable substrates become availa-ble.

During endogenous metabolism and long-term survival experiments, it has as yet been impossible to attribute death to the loss of a specific cellular function (96). However, several theories have been proposed associated with the concept of mainte-nance energy. Correlations between the proton motive force and the metabolic state of the cell have been proposed as well as theories relating the death of the microbe to energy exhaustion (130). Konings and Veldkamp [1201 have also suggested that the gradual decrease in the proton motive force found during starvation experiments could be attributed to the gradual accumulation of •non-viable' cells in the culture.

Zilberstein et al. [l I 7) in a study on the effects of external pH on the proton motive force ob-

served that £. coli shifted its membrane potential

and pH in a homeostatic mechanism to maintain

the internal pH al a constant value. Variations in

the external pH were compensated for by varia-

tions in the membrane potential such that the

Ailw also remained effectively constant. How-

ever, when they subjected a mutant, defective in

its Na• /H • anliporler activity, to a pH change

they found that growth ceased when the A pH

collapsed, although a high membrane potential

was still maintained. The actual death of the mi-

crobes (as measured by lack of colony formation)

did not occur for at least 12 h after the change in

ApH. Recently, 0110 el al. (131} reported that in the

homolactive fermentative S. cremoris the ini-

tiation of lactose starvation caused the membrane

potential lo collapse almost immediately, with an

accompanying collapse in internal ATP concentra-

tion. However, at any time up to 24 h, both can be

restored by addition of lactose, thus allowing the

microbes to survive for short periods in the ab-

sence of a measurable pmf. Unfortunately, no

data were provided as to what happened during

extended periods of starva lion.

In acidophilic bacteria, the pH of the cell cyto-

plasm is maintained al values close to neutrality

despite the extremely low pH values of their growth

:i \6

i 1.2 ~ ~ 0.6 .. ..

0.4 <I A

100 200 300 400 500 TIME (Hrs)

Fig. 4. Effects of starvation on various parameters in T.

acidopltilws. The organism wa.s grown hcterotrophicatly at pH

3.0 in a mineral salts-glucose medium. The cells were harvested

by centrifugation and resuspended at time • 0 in deionized

water at pH 3.0 at 29°C in an incubator shaker. 0, ApH: 0.

~'1-; A, ~~H"; •. cellular ATP level; •. cellular poly-/1-hy-

droxybutyric acid level. The scale for ApH and AP.H• is in

negative mV, that for ~ .. in posilive mV. Redrawn from [1301

with permission.

-18-

)8)

environments. In a starvation study with Thioba-

cillus acidopliilus growing in an environment at pH

3, some decline in ApH cou[d be detected, to-

gether with some minor changes in A ./I during a

starvation period of 530 h (130}. or particular

interest is that as soon as all the internally oxidiz-

able substrates had been exhausted, notably poly-p

hydroxybutyrate, A pH declined, and as soon as

all the available ATP had been used up, Ai/I

collapsed (Fig. 4). The authors were able to show

the dependence of colony-forming abi!ity on the

presence of a measurable Ailw· When the dilw

approached zero, no 'viable cells' could be de-

tected. Ten Brink and Konings (132} found that in

batch cultures with S. cremoris with no pH con-

trol, the proton motive force ( dil H •) collapsed lo

zero as soon as the logarithmic growth phase

ceased (Fig. 5). These researchers concluded that

energy was required lo maintain the pH gradient,

as a result of the changing pH in the external

medium due to lac late production in this organism.

Padan el al. (133} have also shown that in aerobi-

cally grown £. coli, inhibition of respiration by

either anaerobiosis or KCN leads to a collapse in

the ApH. Zychlinslci and Matin [130} showed with

lljiH' 7.0 200

160.! 6.5 ~

~ ~

c

~ 120 z .. s.o: w

I-0

80"'

" j w w u

40 .. : .L

Fig. 5. The electrochemical proton gradient in S. cremoris

during growth in batch culture on a complex medium with

lactose (2 gl- 1) as sole energy source. A, ApH; 0 1 A~; •.

Aji.H•· O, pH; (· - - - - ·), time course of protein synthesis.

Redrawn from (132) with permission.

-19-

384

azide-treated cells of the •!cidophilic T. acidopltilus that the pH gradient could be maintained purely pas.~ively.

A major criticism of much of th.e published material regarding the starvation of microbes is that the effects of death and lysis during the experiments are rarely considered. However, these factors can have serious implications in the inter-pretation of the biochemical changes in the cul· lures under study. A typical example can be seen in a recent study on the survival of the organism Brevibacterium linens during starvation (134J. The authors carried out a very broad range of tests during the starvation period including viability by slide culture, dry weight, oxygen consumption, total sugars, intracellular protein, intracellular amino acids, DNA, RNA, and ATP. They ob· tained what appear at first sight to be interesting changes in the concentration of these parameters with time (Figs. 6-8). ATP was only measured during the first 10 h of the experiment, and showed a rapid decline. However, since the viability, which was unfortunately based on a cell propagation method, decreased to 70% or its initial value, and the total cell mass also decreased together with an increase in extracellular NHZ. lysis processes can· not be ignored. The authors suggest that 'cryptic'

"' z: z

"" :It w

"'

DAYS Fig. 6, Changes in •. turbidity; O. viability; and ;., dry weigh1 in a suspension of BMJibacltrilUtt lintns during 30 days of nutrient 5tarvation. Viability and turbidity AR expressed as Ii of initial values. The ceH1 were grown aerobically at 2J°C in a O.JSI baoto-tryptone, 0.2SI yeast e•t-~ 0.12S'I. glucose medium with mineral .. iu at pH 7.0. The cells were resus-pended in a O.S M Tris-HCI huller solulion (pH 8.0) on day 0 and iPCUbated at 21°C. Redrawn lrom (134) with permission.

0AY5

.. :x: z

Fig, 7. Changes ln the contents or a intrai;cllular protein; A. Cree amino--acids; nnd O. 1he quantity o( ammonium Ions in the exuacellular medium in a suspension of 8. linetU during 30 days of nutrient starvation. The ccUs were grown acrobic:ally at 21°C in a 0.3Sf. baeto-lryptone, 0.2SI yeas1 e•1rac1 and 0.12" glucose medium with mineral salls at pH 7.0. The cells were resuspended in a O.S M Tris-He! buffer solution (pH 8.0) on day 0 and incubated at 21°C. Cells initially contained :no p.g protein and 110 ll& of free amino acids·mg-• dry weight. Redrawn from {1J4J whh permission.

growth was not occurring. based on their failure to observe microscopically either cell wall or cell membrane fragments. The observation of such cell debris. even under conditions where lysis is un-

C) z z "" :It w

"' ... 0 10 20 30

DAY5 Fig. 8. Changes in the intracellular levels of o. DNA; ;., RNA; and D, extracellular material absorbing 11 260 nm. The mean initial quanlity of DNA .., .. 27 1'8 · mg-1 dry wei&ht and tha1 of RNA w .. 70 l'&·mg-1 dry weight The bacterium B. //~ens wti grown aerobically at 21°C in a O.lSi bacto-tryp· tone. 0.251 yeast «traet and O. l2SI glucose medium with mineral ,.ll• at pH 7.0. The cells were resuspended in a O.S M Tris-HCI buffer solution (pH 8.0) on day 0 and incubated al 21°C. Redrawn from [ll4] wilh permi5Sion.

-20-

doubtedly occurring, requires the use of either electron microscopy or of highly specialised mi· croscopic methods such as immunonuorescence, so that this evidence for the lack of lysis is incon· elusive. This aside, if one looks at the data, these are presented in terms or mass of the parameter per unit mass dry weight. However, if death and lysis occur in the culture as suggested by the NHt accumulation and the accumulation of 260 nm absorbing compounds, then the biophysical com· position of what actually constitutes the reference quantity, i.e., unit mass dry weight, must be ques-tioned. H this dry weight were composed entirely of living, active cells, then the data could be accepted as presented. However, after 30 days it is most likely to be predominantly composed of inert cell particulates and dead biodegradable biomass, so that the data have to be recalculated and ex-pressed on a more meaningful basis such as the mass of a particular component per cell or, better still, mass per living/ surviving cell.

The authors measured a reduction in protein content from 310 µg/mg dry weight to 60% of its original value, i.e., 190 µg/mg dry weight after 30 days starvation. However. if the lauer solids were composed of only 60% intact cells then the amount of protein per surviving intact cell is unchanged. This result is rather surprising, since a definitive reduction certainly occurs within the population and intuitively one would expect a cell under starvation to degrade some of its proteins to supply any energy that may be necessary for the survival process. Exactly how such results are best inter· preted is now a problem.

Breuil and Patel (135) have looked at the deple· tion of cellular contents in the anaerobic microbe Methanospirillum hungatei GPl during enforced starvation. Their results were effectively similar to those for the aerobic organism B. linens discussed above, namely that DNA appeared to increase in concentration whilst RNA was degraded, initially rapidly and then somewhat more slowly. ATP was also found to rapidly decline in the system. How-ever, all the results were also expressed on a per unit biomass basis and did not take into account the change in the physiological matrix of the sus-pended solids. A detailed analysis of cell numbers and particulate composition is required before.

JSS

during and after experiments where changes in the intracellular pools of starved cells are being in-vestigated. The foregoing description of the ap-proach to starvation experiments is common. Most of the reported results are expressed in terms of the amount of cell component per ml cell suspen-sion. Unfortunately the use of cell propagation methods is also extremely common in s1arvation experiments, thus reducing the confidence level in the interpretation of most published results.

In a subsequent section, it will be shown 1hat microorganisms possess enzymes capable of de-grading their own cell material. This will be limited to the class of enzymes known as autolysins, which function specifically on the cell walls. However, it has been recognised for some time that other enzymes are active within the cell, modifying in· ternal structures and ensuring substrate availabili-ty during times of exogenous nutrient deprivation. However. these enzymes can also function in catabolism of exogenous nutrients resulting from cellular lysis. For example, Parquet et al. (136) reported on the possible physiological function of the enzyme N·acetylmuramoyl-L-alanine amidase in £. coli K 12. This enzyme was found to be either loosely bound to peptidoglycan or en· trapped in the outer membrane-peptidoglycan complex, and to specifically cleave the bond be-tween N-acetylmuramic acid and alanine in the bacterial peptidoglycan. Paradoxically, such an enzymatic cleavage was found to be extremely rare in normal growing cultures. Hence, it was sug· gested that since the enzyme was active under autolytic conditions; it functioned as a hydrolaSt for growth on peptidoglycan fractions. Parquet et al. (136) .were able to demonstrate growth of £. coli on MurNAc-Lala fractions alone as well as on a MurNAc-lala·Xaa fraction suggesting, there-fore, the presence of other hydrolases. A vast array of enzymes specific for turnover of endog· enous macromolecules are present in microbial cells and under normal growth conditions they function as control mechanisms of the growth and metabolic processes. However, under conditions where exogenous substrate becomes limited. they serve lo provide the cell with .a continued sub· strate supply for energy generation in the absence of resynthesis.

-21-

Growth under a nutrient limitation other than carbon orten leads to the accumulation or an endogenous substrate such as glycogen (for review see Preiss [1371), or poly·P·hydroxy acid polymers. In addition, various intracellular compounds have been identified as substrates ror endogenous metabolism including free amino acids, RNA (138,1391 and proteins [140). The release of these compounds as a result of cell lysis provides the possibility for their uptake by other members of the same population as either carbon energy sub· strate or other nutrient sources and may allow the growth of auxotrophic organisms within mixed populations. This process is known as 'cryptic' growth. The term was first used by Ryan [141 J in a study of the turnover of cells during stationary phase batch growth. Koch [142) also recognised the possibility for cellular turnover during the growth process involving recycling of cellular components in a population. Postgate and Hunter [143) demonstrated 'cryptic' growth in a study on starvation of Aerobar:ter aerogenes ( Klebliella pneumoniae ). After prolonged incubation the capacity for cryptic growth was apparent (Fig. 9) although the authors used plate counts for the

e <fl

"' 0 > ;;: "' :::> "' g 4 "' g

30 20 40 60 80 100 120 140 HOURS

Fig. 9. Survival curve or A. 0,~rugeMJ (K. pMumMi•) showing •cryp1ie' growth. The microbes hanested from continuous cul· turc on glycerol at D • 0.2S h- 1 were washed twice by centri· rugadon. resuspended in distilled water. added to 'saline-Tris' bufler at 20 l'I dry weight organisnu/ml and allowed to starve at 40°C at pH 6.95±0.0S, under lorced aeration. Points up to 24 h by slid< cullure, including plate counlS at 0.7 and 24 h. subsequent points by plate counts only. Redrawn from (143] with permission.

~ 80 ~ ~ "'60 w o5 ~ 40 ..... z: ~ 20 "' &

0 50 INCUBATION TIME AT 37°C

Fig, 10. EUect of nutritiomd status on viabilities of ageing stational)' phase populations of A. o4'togenes (K. pntumoniPt). flasks shaken at J7°C in air in defined minimal media ror~ mulatcd to give b. glycerol~Umited; O. NH; .. \imited; and•. Mg2 ~limited scationary populations. Viability determination by slide cuhure, Redrawn from (144) with permission.

enumeration of starved bacteria. Recycling of potentially limiting nutrients such as Mg2 • or NH.;- (Fig. 10) is thus possible under starvation conditions [144).

During starvation conditions, cells have been shown to respond in a variety of ways including lhe use of their own storage products. Physiologi-cally, it is unreasonable for a cell lo undergo depletion of internal structure if the substrate flux for growth is sufficient. This point will be consid· ered further in the section on the mathematical descriptions of microbial growth and substrate utilisation.

Experiments designed to specifically investigate 'cryptic' growth phenomena are rare. Nioh and Furusaka (145) looked at the growth of bacteria on heat-killed suspensions of the same bacteria. They found that the ability to grow 'cryptically' was not universal and produced figures for cell yields based on the number of cells in the heat killed suspension required to produce unit •new' cells. Unfortunately the nature of the experiment was sufficiently damaging to prevent interpreta· lion of the results under natural conditions. Dur-ing the heat killing process where lysis occurred the heat was probably sufficient to break the

-22-

polymeric particulates up into easily assimilable monomers. However, the heat killing process does not necessarily lead to cell lysis so that most of the soluble material which was potentially available for incorporation into new cells was unavailable, being entrapped in heat-stabilised 'dead cells'. Once again, the enumeration method used was the bacterial plate count technique, and therefore, lit· tic reliance can be placed on the numbers ob· tained. Nioh and Furusaka [145) also found that addition of ammonium chloride led to a consider-able increase in the numbers of cells growing in the heat killed suspensions, suggesting that their suspensions were nitrogen limited, and therefore, the full potential of the 'cryptic' growth process could not be realised.

The growth of bacteria in their heat-killed sus· pensions results in lower biomass yield coeffi· cients than are to be expected from theoretical calculations. Hamer and Bryers [146) used the relationship between maximum biomass yield coefficient and the heat of combustion proposed by Linton and Stevenson [147] to predict a theo-retical maximum yield value of 0.66 g bacterial biomass per g substrate bacterial biomass. assum· ing complete conversion of the substrate carbon to either new cells or CO,. although these authors suggest deoptimization to a yield coefficient of about 0.33 would probably be encountered in real 'cryptic' growth situations. Mason et al. (148] have shown this estimate to be valid with reported yield coefficients around 0.4 (g cell carbon/g substrate carbon) in pure cultures of K p11eumaniae.

In this same study. Mason et al. [1481 grew K pneumoniae in continuous culture in a defined medium with glucose as growth limiting substrate. A volume of the culture was removed and used as the growth medium for the same cells after sonica-tion. centrifugation and sterile filtration. The re-sulting solution, comprised of soluble organics, was used as growth medium for the same cells. The cells for inoculation were derived from the same culture after steady slate conditions were reestablished. With this method there was no ap-parent lag phase before exponential growth oc-curred, suggesting that the microbes were already fully equipped with the necessary battery of en-zymes for growth on the complex nutrient source

0.06 0.05 o.o• 0.03

0.02

387

0.010~-~-~-~-~.-...... -~-...... -~

TIME JMuoJ Fig. 11. •Cryptic· growth curve ot K. pneumoniat NCIB418 growing on a mixture of carbon sourcts derived from cell sonicate preparation. K. pr1eumoniatt was grown in a chemostat at D • O.:S h- 1• The •cryptic' growth medium was prepared by sonicadon follo~d by sterile filtration ol the chcmostat cul§ ture. The inocuium (10 µl in ~ ml) was taken Crom the original culture a(ter steady state conditions had been rccstab,. tished. (From lt48J.)

(fig. 11). Deoptimisation of the maximum theo· retical yield coefficient to values of approx. 0.4 nevertheless represents highly efficient conversion, despite the very large range of potential carbon sources of varying complexity in the sonicated suspension.

15 jg A 14~1

BIOLOGICAL SOLIDS " . 400 13 ~} ~

12 ~ e 300

"' 3 200

100 coo

D m • D B m • MINUTES HOURS

Fig. 12. Metabolism of soluble cell components (prepared by subjecting sludge to sonication) by a.tended aeration activated sludge. The lell·hand graph shows the trend during the first 30 min. whilst lhe extended scale -on the right shows 1hc long~term results. Redrawn from (149) with permission.

-23-

388

A similar approach was adopted by Gaudy et al. (149) who used sonicated sludge as a nutrient source for 'cryptic' growth experiments. By follow-ing the COD reduction arter reinoculating with sludge organisms, it was found that approx. 50% or the COD was removed in the first 15 min, and 90% by the end or the experiment (Fig. 12). and is indicative or the greater potential ror 'cryptic' growth to occur under mixed culture conditions than in pure culture. Obviously a tremendous amount or work still needs to be carried out to assess correctly the impact of 'cryptic' growth in processes such as wastewater and waste sludge treatment and in other industrial processes.

4. AUTOL YSIS

The growth or bacteria requires the interaction of biosynthesis and degradation of various struct-ural polymers, notably cell wall polymers such as peptidoglycan, to allow volumetric expansion and cell division. The class of enzymes responsible for cell wall polymer hydrolysis have been termed autolysins and their action can be broadly divided into (a) constructive or voluntary and (b) destruc· tive or involuntary. Constructive properties are exhibited while the control mechanisms of the autolysins are still effective and include daughter cell separation {150, 15 I j. turnover and expansion of cell wall biopolymers (152-156), and morpho-logical differentiation (151,157). Should this tight control of autolysin activity break down, then the destructive roles of the autolysins are manifested by the rupture of cell walls, leading to loss of cell activity and eventually to total dissolution of cel· lular integrity. While several reviews have been published on the constructive (voluntary) roles of autolysins (158-160J, less auention has been focused on the destructive potential of this class of enzymes and the quantitative realisation of this potential. Some suggestion has been made that the destructive aspect of autolysins may be continu· ously manifested at low levels in populations grown under what are assumed to be ideal condi-tions (146,161,162).

The dual potential or the autolysins demands a highly sensitive and efficient control mechanism.

possibly one of the most demanding in the whole cell, since a single error can result in irreversible damage. A close relationship has been demon-strated between protein synthesis and autolysis (153,163-165) and inhibitors of autolysin activity [164) or activating substances (166-169] have been implicated in the control or autolytic enzymes. Evidence suggesting that cell wall turnover and autolysin activity are controlled by different genes was presented by Vitcovic [ 170) who found no correlation between the two processes.

The location of the autolysins is also unknown. Pooley (171.172) and Glaser and Lindsay [154} have suggested that wall turnover occurs on the outer cell surface, the autolytic enzymes being transported through the cell wall and activated in the region of the older peptidoglycan. New pepti· doglycan is laid down in the inner wall area and protected from degradation by its structural and spatial configuration [156). The matrix in which peptidoglycan is embedded is a highly organised three-dimensional structure and as such should precisely tune the activity of the autolysins. The presence of specific teichoic acids in this matrix have been shown to be necessary for the activity of some of these enzymes in Gram-positive micro-bes (150,173-175]. An alternative explanation was proposed by Joliffe et al. [166) who suggested that the surface chemistry of the cell membrane as determined by the magnitude of the proton motive force, has an inhibitory effect on the activity of the autolytic enzymes. As these move further away from the sphere of influence of the membrane their activity increases. They derived this hypothe-sis from the fact that the addition of agents which were capable of dissipating either electrical or pH gradients resulted in both inhibition of growth and the rapid lysis of exponentially growing cells. They also found that bacteria subjected to starva-tion conditions, such as those suspended in buffer solutions in the absence of an oxidisable carbon substrate, rapidly lyse, but lysis was immediately inhibited by re-energising the membrane by the addition of electron-donating agents.

Membrane de-energisation was also implicated as a possible cause of cell lysis following addition of medium chain fatty acids to cultures or Bacillus subtilis (176). Anomalies are also known where the

-24-

control mechanism over autolysin activity breaks down. Cultures of Myxococcus ccralloides D ex-hibit such behaviour whereby all the cells in a batch culture spontaneously lyse at the end of the exponential growth phase [177}. This behaviour appears to be due to the synthesis of an autolysin activating substance which builds up in the medium and on reaching a certain concentration activates the autolytic enzymes and triggers lysis [167).

Autolytic enzyme activity is not just limited to the bacteria. In fungi, autolysis of mycelia has also been investigated and the same precise mecha· nisms of control are apparent. In Botrytis cinerea, the activity of several different autolytic enzymes varied independently during the autolytic phase in the growth of the mycelium [178). Autolytic activ-ity has also been demonstrated in archaebacteria [179} and in yeasts [180). Microbial autolysis was the subject of a recent symposium [181 ]. It should not be forgotten that anomalies also exist espe-cially in the case of the Mycoplasmas. Since these organisms completely lack a cell wall, they do not succumb to autolysis in the same sense as cell-wall-bound microbes. However, lysis by cell mem-brane rupture in such organisms will lead to Joss of function.

5. MICROBIAL DEATH

Microbes can be killed by various means, al-though in order for a microbe to die it must either lose cellular integrity or incur irreversible damage to its genome. Agents by which one or both of these conditions arise include; chemical damage (including antibiotics), refrigeration, freezing, heating, and irradiation. Very often these results in only temporary damage 10 the microbe, if cellu· lar integrity is not lost or if irreversible damage to 1he genome does not occur. However, if sub-le1hal damage has occurred, it does not necessarily fol-low that the microbe will recover. II has been clearly shown that a damaged microbe is more susceptible to further injury than a healthy mi-crobe {182,183).

Unfortunately, in some instances, damage or injury have become synonymous with failure to

389

illicit growth on an agar surface [45,184). The accuracy and meaning of this interpretation is open to doubt. In some instances the test may indeed be indicative of some form of damage. However. the type and extent of damage varies considerably. Some forms or injury tend to make the microbes more susceptible to the stresses dur-ing either surface or submerged cultivation than others. An example of this was reported by Hurst [184) citing results from (,urran and Evans [185) where one medium used for enumerating the mi-crobes as a test for injury resulted in one set of results. but changing the medium composition re-sulted in a totally different picture. Similarly, in-jury has been described as being the inability of microorganisms to form colonies on a defined minimal medium, while retaining the colony for· ming capability when complex nutrients are pre-sent in the medium or vice versa (45). In experi-ments designed to inves1igate the effects of either physical or chemical maltreatment of microbes, very often no differentiation is made between injured and dead microbes.

There are various possibilities for inflicting in-jury on a microbial cell. The outer layer of a microbe ignoring the capsular material and cell appendages, is the cell wall/membrane complex. Gram-positive microbes possess a simpler wall structure than Gram-negative microbes. In the latter, a three-layered structure exists consisting of an innermost layer of peptidoglycan covalently linked lo lipoprotein molecules. The outermost layer is the outer membrane and is covalently linked to the peptidoglycan middle layer and con· sists of lipopolysaecharides (LPS), phospholipids, and proteins [186). The Gram-positive wall, in contrast, is much simpler and consists of pepli· doglycan chains embedded in a teichoic acid ma-trix. The cell wall serves as a barrier between the internal and external cell environments and con-fers s1ability on the cell. Due to the higher osmotic pressure inside the cell, any damage to the wall/ membrane structure can lead to rupture of this structure and loss of cellular integrity. Strictly speaking the cell does not die, but loss of microbes from a population results from either the disin· tegration or lysis of living individuals.

The cytoplasmic membrane is a metabolically

-25-

390

active, mobile, semipermeable structure which mediates transport between the cell and its en-vironment. Since the elucidation of the chemi-osmotic theory of energy generation, the impor-tance of the membrane and membrane potential have become apparent. Many of the enzyme-me<;li· ated reactions in a cell occur attached to the membrane, and the electron transport chain is an integrated part of the membrane structure. Either rupture or damage to the cell membrane will, therefore. result in injury (if the cell can recover) or death by eventual cessation of metabolic func-tions or by lysis.

Potential sites for damage within the cytoplasm also exist. The DNA, for instance, which encodes the information necessary for the correct func-tioning of a cell and the genetic information for replication is particularly vulnerable. The presence of lesions in !he DNA will lead to the production of nonsense proteins and enzymes and, thereby, to the loss of cellular function or to the inability to reproduce. The ribonucleic acids are also potential sites of damage. They are found as structural components in ribosomes, as well as functioning as messengers between the DNA and the ribo· somes and for transporting amino acids during protein synthesis. Therefore, damage to this com-ponent will also result in non-faithful reproduc-tion of enzymes and other proteins and lead to loss of cellular function.

In the following section a cursory glance will be directed towards the question of methods of in-ducing death in microbes.

Subjecting microbes to sub-lethal concenlra· tions of chemical inhibitors can result in the loss of their ability to either reproduce or grow under certain environmental conditions. Specific agar-based enumeration media used routinely to de-termine the extent of bactericidal activity are thus inappropriate, and may result in unacceptable health risks when used for testing for food- and water-borne palhogens. Chemical agents for kill-ing microbes arc used in water treatment, phar· maceutical therapy, disinfection and food in· duslries. Microbes arc susceptible to attack by chemical agents. In contrast to most physical stresses, chemical agents often interact directly with either the metabolism or the physical struc·

ture of the microbe, and as such, chemical damage occurs at the molecular level. The effects of many chemical agents have still to be elucidated.

TI1e cell wall is the target of many antibiotics, but compounds like EDT A, lysozyme, phenol, and chlorine are also known lo act at the outer boundary of the cell (187) together with various surfactants such as Triton X-100 [188J. Similarly, 1he cell membrane succumbs easily to attack by agents capable of disorganizing the structure, or changing specific activities towards ions, or by inhibiting the membrane-bound proteins involved in transport processes such as by the action of some antibiotics !189), or at superop1imal temper-atures or with strong oxidising agents. The mem-brane is also susceptable to attack by agents capa-ble of uncoupling oxidative phosphorylation reac-tions. Organic solvents such as butanol [190), ethanol !l 91), alkali metal ionophores [192) and detergents [193), have their site of action at the cell membrane.

Cytoplasmic targets for chemicals include cyto· plasmic eniymes, nucleic acids and the ribosomes. Coagulation of the cytoplasm can occur at high drug concentrations [194). A range of different antibiotics affect the biosynthesis and functioning of the nucleic acids. The ribosomes tend to be susceptible to agents which cause instability through removal of the Mg i + necessary for the association of the 30S and SOS subunits (19S,196J.

Often ignored under the heading of chemical injury is the effect of aerobiosis and anaerobiosis in obligate aerobes and obligate anaerobes. The sensitivity of anaerobes exposed to oxygen varies with different strains [197). The time required to deactivate a cell appears 10 be relatively long. Shoesmith and Warsly (198) in a review on the exposure of anaerobes to oxygen, summarised the oxygen tolerance data derived from other authors using various strains and found that despite little agreement in assessment methods, most of the strains were still alive 1 h after exposure.

During refrigeration death occurs as a result of a number of different effects not least of which is the rate of cooling and subsequent thawing processes. Thus, experiments designed to investi· gate the effects of refrigeration on microbes must be very carefully controlled (199). Two types of

-26-

refrigeration are common, namely chilling and freezing. These practices arc commonly used in the preservation of microbial cultures in technical processes, but are equally likely to occur under the innuence of seasonal changes in natural environ· men ts. The effects of refrigeration are also im· portant in the food industry where this practice constitutes a major form of preservation, whereby the destruction and inhibition of microbial growth and activity are desired.

Death often results as a consequence of the loss of control of cell permeability-a phenomenon known as cold shock [199j. One theory for the ensuing death after chilling is the loss of activity of the enzyme DNA ligase as a resull of the depletion of magnesium ions from the cell (200]. At temperatures below 0°C cell damage can occur as a result of either extracellular or intracellular ice crystal formation, depending on the cooling rate [201] and from changes in the ice-crystal structure under different physical (i.e., pressure or temperature) conditions [203]. Death during ex-tended frozen storage occurs at a fast initial rate but slows down with time until a stage is reached where constant numbers remain 'viable' (204). Death is thought to occur as a result of exposure to the raised solute concentration as a conse· quencc of ice crystal formation, as well as from physical damage.

Without knowing the physiological reasons for its effectiveness, heat treatment has been used for over 100 years as a means of either destroying or limiting the growth of microbes. Temperature af-fects different microbes in different ways. The growth of a particular microbe normally takes place within a narrow range of temperatures. It has been suggested that heat principally affects the DNA by inducing the action of exo- and endonucleases resulting in multi-strand breaks in the DNA [204,205). In E. coli, the effect of heat has been suggested to result in the denaturation of the cell wall leaving the peptidoglycan layer weakened at one or two points after which the plasma membrane ruptures [206), Some associa-tion of proteins with the nucleoids also occurs as a result of mild heat treatment (207].

Radiation results in a range of alterations in the DNA, including phosphodiester strand breaks,

)91

nucleic acid-protein cross links, pyrimidine dimer formation. etc. [208).

For detailed information on the mechanisms of destruction of microbes. the reader is recom-mended to consult specialised texts (209-211 ).

6. MATHEMATICAL MODELLING OF DEATH AND LYSIS

Models in general can be physical structures, verbal descriptions, or sets of mathematical equa· tions that aid the user in comprehending complex phenomena. In some instances, a model serves as a hypothesis, other models attempt to predict be. haviour and still others aid in understanding. No model will reflect all aspects of •reality'. Neverthe-less, all models should force the investigator to think in a concise manner and to consolidate what information is known plus what new information is required.

Mathematical modelling or microbial systems has been used to help understand the dynamics of microbial behaviour in laboratory, natural, in· dustrial and man-made environments. Application of such models to scale-up laboratory operations into industrial processes and for extrapolation of laboratory behaviour to natural environments has resulted from the construction of such system modelling. Mathematical models also serve to di-rect and to optimise the research experimentation and to allow prediction of microbial behaviour. Biological systems are extremely complex and the actual system has frequently to be replaced by an imaginary model system which is mathematically tractable [212). As a result, a very large number of models can be proposed for any single system each depending on the assumptions made for sim· plifying the real biological case. Many theoretical dynamic studies have been based on models for-mulated from steady state or equilibrium kinetics and thus inadequately describe dynamic be· haviour [213).

The modelling of death and lysis processes in living cultures requires a broad understanding of the physiology of the organisms concerned. The traditional black box approach, whilst describing the behaviour of a population under a given set or

392

constant conditions, might be totally inadequate when those conditions change. Calibration or the mathematical model with carerul experimentation is also necessary.

One or the earliest mathematical considerations or bacterial growth was presented by Buchanan (214), who presented equations describing the lag, logarithmic, stationary, accelerated death and logarithmic death phases during the batch cycle. Buchanan suggested that cell death begins during the late exponential phase and increases to a level in the stationary phase where it exactly balances the rate or growth or the bacteria, and finally exceeds the rate or growth such that a net decrease in the number or live bacteria present results. He described each of the individual phases using mathematical equations.

The first application of continuous culture growth expressions appeared in 1950 (215] and these were latter assessed by using quantitative experimentation (216]. By 1958 the phenomenon of a decreasing yield coefficient at low growth rates was recognised and this biomass-reducing effect was termed 'endogenous metabolism' (217). Herbert [217) described the constant loss of bio-mass at all growth rates using Eqn. 7.

dx/dt - (,1.1-k.)x (7)

where k, is the endogenous metabolism constant. Powell [218) derived an expression for the steady

state biomass concentration using this notion of endogenous decay

"'"'( s0 D K,D ] x • y•/• D + k - -;;--::::o e rrnax

(8)

A large amount of work during the 1950s and 1960s was concerned with the destruction or mi-crobes, particularly in hostile environments as en-countered in sterilisation and pasteurisation processes. This work arose primarily due to strin-gent hygiene requirements for food for public consumption. A distinct paucity of information exists concerning death and lysis of micro-organisms under conditions where microbes are not exposed suddenly to hostile environments. Without any apparent stress, death must occur by processes such as those of involuntary autolysin activity, as described earlier. In continuous culture

-27-

systems at low growth rates it has long been recognised that a large proportion of tbe microbes is •non-viable' or 'dead' [143,129]. Tempest et al. [219) found that the viability or A. t1eroge11es ( K. pneumoniae) was as low as 40% at a dilution rate or 0.004 h - '. but increased to 90% above 0.5 h - '. Although this data represented evidence or death within a continuous culture, note that the cell enumeration method used was the slide cultivation technique or Postgate et al. [220]. The possibility that such slow-growing organisms (at D • 0.004 h-'. the doubling time= 173 h or 7.2 days) fail to adapt to the agar environment and form colonies does exist and limits extrapolation or these results.

Sinclair and Topiwala [221] presented one or the first models for growth in continuous culture which considers the viability concept and uses the results or Tempest et al. (219) and Postgate and Hunter (143] to verify their model. Two mecha-nisms were assumed to act on the living biomass, x: (1) cell death, which was assumed proportional to the living biomass; and (2) endogenous metabolism, also considered proportional to living biomass. Thus, Sinclair and Topiwala present the mass balance equation for biomass as: dx/dt ~ µ(s)x - k. - yx - Dx (9)

Where y is the specific cell death rate constant (1- 1) and D is the dilution rate (r'). Similarly the mass balance for s appears as:

µ(s)x ds/dt • D(s0 - s) - Y"'"" (10)

•/• where s0 is the influent substrate concentration, s the residual substrate concentration and Y,;:" the yield that would be attained in the absence of decay and cell death. Thus, the steady-state concentration of dead cells (xd) can be written as xd - ( y/D)x (11) and for steady-state soluble substrate, (s),

_ K,(D+K+y) s• l'in-(D+K+y) (12)

The viability can be calculated using Eqn. l3

. b'I' ( ) x x D Vla 11ty P =---=-=--(x+id) x, D+y

(13)

-28-

The specific death rate constant can be de-termined, rearranging Eqn. 13,

l/v=l+y/D (14)

and plotting the inverse of viability versus inverse dilution rate, the slope would equal y.

Using these equations, Sinclair and Topiwala [221 J satisfactorily predicted a decrease in biomass at low dilution rates with concomitant low viabil· ity.

Weddle and Jenkins [222] analysed pilot· and full-scale activated sludge systems for cell viability and activity using various techniques including ATP, dissolved oxygen uptake, cell counts (using enrichment on agar} and electron transport system activity using the compound 2,3,5-triphenyltetra-zolium chloride. They suggested the existence of non-viable organisms and of non-degradable inert biomass fractions. Grady and Roper [223] derived a model for the activated sludge process which predicted cell viability as a function of the mean residence time by incorporating the processes of death, maintenance (endogenous decay) and loss of viability in the absence of death. These authors

393

recognised the existence of those cells which while not dead, were incapable of replication. They fur-ther suggested that these cells could have func· tional enzyme systems and were therefore suscep· table to death and decay by endogenous metabo· lism and lysis. They also commented on the in· fluence of 'cryptic' growth on the system but chose not to model it due a lack of experimental capabilities required for verification. The contri· bution of non-viable cells to the metabolic activity or the system was discussed by Jones (224), who divided biochemical activity in a heterogenous plug now industrial process into the following categories, each group having its own appropriate growth kinetics: (i) Bacteria which grow within the system (Monod kinetics); (ii) bacteria which do not divide within the system but are viable (main-tenance kinetics); (iii) Non-viable cells which still retain some metabolic (biochemical} activity (Mi-chaelis-Menten kinetics). The growth or group (i) microbes maintains suitable concentrations of groups (ii) and (iii) in the system, such that sub· strate removal approaches first order kinetics [223).

The effect of non-viable, metabolically active

-29-

394

cells on the yield coerricient was also shown by van Uden and Madeira-Lopes [6) in !heir model of yeast growth in chemostats at superoptimal tern· peratures.

7. CONCLUDING REMARKS

Despite more than a century of microbiological research, emphasis still tends to be placed on the growth of microbes, and little, if any, attention is focused on their demise. In natural and man-made (including laboratory) environments, microbial 'death' or decline is a process which occurs hand· in-hand with microbial growth. Microbes are neither immortal. nor are they any less accident-prone than other creatures. Since they are not the highly efficient machines they are frequently made out to be, it is time that 1he effect of their frailness be properly considered in process design and evaluation, in public health investigations and in eco·physiological research. In order to salisfacto· rily achieve this, it requires a beuer understanding of the methods available, their limitations and the extent to which deductions can be made from the results they generate.

ACKNOWLEDGEMENTS

C.A.M. was supported by grants from the Swiss Federal Institute for Water Resources and Water Pollution Control (EAWAG) and from the Swiss National Programme 7D.

Within the last 10 years, more accurate meth· ods have evolved to determine the various types or microbe that can exist in a culture. Mason et al. [162) depicted the processes of microbial growth and death in continuous culture, as shown in Fig. 13. They reccgnised the possibility for the pres-ence of each or the various categories or cells and their interrelationship and a11emptted to experi· mentally differentiate between them. Differentia-tion between metabolically active and non-active (dead) cells was possible based on ETS activity. Their results suggested that a very low fraction, if any, of the cells present were indeed dead which appears to disagree with much of the earlier work suggesting the contrary. However, it should be noted, that much or that earlier data were derived using plate count methods which were totally in· adequate for quantitative assessment. The model proposed by Mason et al. [162} suggests that the reduction in biomass yield results from a low level or lysis in the culture and in the model this lysis process and the subsequent growth of the intact organisms on their own lysis products ('cryptic' growth) are represented. Consequently this model may be suited to describe more accurately the behaviour of organisms in natural environments where cell decline may result from cell lysis and actual death occurs rarely, if at all. Hamer and Bryers [146) and Hamer [225} using the same principles, incorporated lysis and 'cryptic' growth in models which satisfactorily describe the effects of these processes in wastewater and waste sludge treatment processes.

REFERENCES

Future prospects in the modelling of net micro-bial growth will have to focus on the utilization of particulate substrates present in the system feed and produced by particulate formation after cell lysis [95). The possible presence of a non·biode· gradable inert biomass fraction accumulating dur-ing microbial growth may also artect the calcu· lation of the yield coefficient [2261 and the assess-ment of the efficiency of microbial biomass pro-duction, product formation and product u1ilisa· ti on.

(!J Stumm. W. and Morgan, JJ. (1981) Aquatic Chemistry. An Introduction Emphasising ChemicaJ Equilibria in Natuta1 Waters. 2nd ed. Wiley. New York.

(2) Daley. R.J. (1979) Direct epi0uorc$Cence enumeration of native aquatic bacteriia-uses, limitations., and compara~ tive accuracy~ in Native Aquatic Bacieria: Enumeration, Activity and Ecology. ASTM STP 69S (Costerton, J.W. and Colwell. R.R .• Eds.) pp. 29-45. Am<riean Society for Testing and Materials, Philadelphia.

131 Stanier. R.Y .• Adelborg. A.E. and Ingraham, J.L. (1976) General Microbio1ogy, 4th ed. MacmiUan~ New Jersey.

141 Postgate. J.R. (1976) Death in macrobes and microbes, in 1ht: Survival of Vegetative Microbes (Gray, T.R.G. and Postgate. J.R .. Eds.) Proc. Symp. Soc. Gen. Microbial. 26. pp. 1-18. Cambridge University Press. Cambridge.

ISi Sykes, R.M. (1976) Microbial product formation and variable yield. J. Water Pollut. Contr. Fed. 411, 2046-2054.

161 van Uden. U. and Madeira-Lopes, A. (1976) Yield and maintenance re1ations or yeast growth in the chemoslal

-30-

at sup.:roptimal temperatures. Biotechnol. Biocng. JR. 791-804,

(7) Chibatu, I. and Tos11. T. (1977) Tri~nsformatlons of organic compounds by immobilised microbial ce11s. Adv, Appl, Microbial. 22. 1-27.

(SJ Koch, A.L. (1971} The adaptive response o( Esd1nidtia C'O/i lo a fast and famine eKistence, Adv. Microh, Phystot. 6. 147-217.

(91 Parkinson. D .. Gray, T.R.G. and Williams, S.T. (1971) Methods (or scudying the ecology of .soil microw organisms. IBP Handbook 19. Olockwell, Oxford,

(IOI Sorokin, Y.I. and Kadota, H. (l 972) Techniques !or the assessnlCnt of microbial production and decomposition in fresh waters. JBP iiandhook 23, Blackwell, 0.-cford.

(Ill Costerton, J.W. and Colwell. R.R. (Eds.> (1979) Native Aquatic B:icteria: Enumer..ition. Activity and Ecology. ASTM STP 695. American Society for Testing and Materials, Philadelphia.

(12) Jones. 1.G. (1979) A Guide to Methods for Estimating Microbial Numbers and Biomass in Fresh Water. Freshwater Biological Association, Scientific Publication 39, Windermere. U.K.

{13) White, O.C. (1983) Analysis of microorganisms in terms of quantity and activity in natural environments, in Microbes in their Natural Environment (Slater, J.H. and Wimpenny, J,W.T,, Eds.) Proc. Symp. Soc, Gen. Mi· crobiol. 34, pp, 17-66. Cambridge University Press, Cambridge.

(14) Koch, fl. (1881) Zur Untersuchung von pathogencn Organismcn, Miuheilungen aus dem Kaiscrlichen Gesundheitsamte 1, 1-48.

(15j Schmldt, E.L (1973) Chairman's summary. Bull. Ecol. Res. Commun. (Stockholm) 17, 453-454.

(16J Hopton, J.W .. Mekhiorri~Santo1ini. U. and Soroldn. Y.l. ( 1972) Enumeration of viable ce:Us or microorganisms by plate coun• technique, in Techniques for the Assessment of Microbial Production and Decomposition in fresh Waters (Sorkin, Y.I. and Kadota. H.. Ed") IBP Handbook 23, pp. 59-64. Blackwell. Oxford.

(17] Buck, J.D. (1979) The plate count in aquatic microbi· ology. in Native Aquatic Bacteria: Enumeration Activity and Ecology ASTM STP 695 (Costerton. J.W. and Col-well, R.R., Eds.) pp. 19-28, American Society for Ten· ing and Materials. Philadelphia.

(18} Fry, J.C. (1982) The analysis or microbial interactions and communities in situ. in Microbial Interactions and Communilies I. (Bull. A.T. and Slater, J,H,, Eds.) pp. 103-152. Academic Press, London.

(19j Dutton. R.J.. Bitton. G. and Koopman. B. (1983) MaJachite green~lNT (MINT) method for determining active bacteria in sewage. App1. Environ. Microbiol. 46, 1263-1267.

f201 Breznak. J.A,, Potrikus. C.S.. Pfennig. N. and Ensign, J.C. {1978) Viability and endogenous substrates used during starvation survival of Rlwdmprrillum rubrum. 1. Bacteriol. 134, 381-388.

(2lJ Allen-Austin, D,, Austin. B. and Colwell. R.R. (1984)

39S

Survival of A"'1monuJ solmotticido in river w:uer. FEMS Miorobiol. Leu. 21. 143-146,

(221 Bell, R.G. and 0.Lacy, K.M. (1984) Heal injury and recovery of Streptorocc1u fattiunt associated wiih the souring of chub-p~ked lundlC'On meat. J. Appl Bact. 57. 229-236,

(231 Kumada. K., Koik<, K. and Fujiwara. K, (19115) Th< survival of bac1eria urtder · starvation e<>nditions: 4 mathema1ical expr~ssion or microbial death. J. Gen. Mi~ orobial. DI, 2309-2312.

(24) Reichardt, 0. (1979) A new eJtperimental method for the determination of \he heat destruction parameters of mi~ croorganlsms. Acta Atimentaria 8, 131-155.

(2SJ Arnold. W,N., Pringle. A.T. and Garriwn, R.G, (1980) Amphotericin b-induccd changes in K +content, viability and uhrastructure or yeast phase llistopfa.rma ('1Jpsult:t-1um. J, Bacteriol. 141. 350-358,

[261 Hyun, H.H,, Zeikus. J.G,. Longin. R,. Millet. J. and Ry1cr, A. (1983) Ultrustrutture and extreme heal resis-tance ot spores from thermophillc Clastridium species. J. Bacteriol. IS6. 1332-1337.

(271 Graumlich, T.R, and Stevenson, K.E. (1978) Rcoovery of thermally injured Saccharomyces eerevisiae: cCCects of media and storage conditions. J. Food Sci. 43. 1865-1870.

f28] Rippey, S.R. and Cabelli, VJ. (1979) Membrane filter procedure ror enumeration of AerDnWltas AJdrophifa in freshwaters. Appl, Environ, Microbial. 38, 108-113.

(291 Farber. J.M. and Sharpe, A.N. (19&4) Improved ba<tcrial recovery by membrane filters in the presence ur food debris. Appl, Environ. Microbial. 48. 441-443.

(301 Tse, K.·M, and Lewi" C.M. (1984) Membrane filter staining method: bacterial p1ate counts in 24 h. Appl. Environ. Microhiot 48, 433...434.

IJll Cundell. A.M, (1981) Rapid counting methods for coli· form bacteria. Adv, Appl. Microbiol. 27. 169-183,

[32j Brock, T.D. (1971) Microbial growth rates in nature. Bacterio1. Rev. 35. 39-58.

(33} Jones. J.G. (1977) The crfect of environmental factors on estimated viable and total populations ot plank.tonic bacteria in lakes and experimental enclosures.. Frnh-water. Biol. 7, 67-91.

(34) Colwell. R.R., Brayton, P.R., Grimes, D.J.. Rosuk, D.B,. Hug. S.A. and Palmer. L,M, (198S) Viable but non-cultufablc Vibrio dwleae and related pathogens in the environment: implications for release of genetically en-gineered microorganisms. Bio/Technol, 3. 817-820.

f351 Stotzky, G. (1972) Activity, ecology and population dy· namics of microorganisms in soil. CRC Crit. Rev. MiR crobiol. 2. S7-D7.

(36) Kogure. K,, Simldu. U. and Taga. N. (1979) A tentative direct microsropic me1hod for counting living marine bacteria, Can J, Microbiol. 25. 415-420.

(37] LeC!tcvallier. M.W,, Jekanoski, P.E.. Camper. A.K. and McFeters. G.A. {1984) Evaluation of m~n agar as a fecal coliform medium. Appl. Environ. Microbiot 48, 37l-37S.

[38) Carlwxi. A.F. and Pramer, D, (19$7) Factors innuencing

-31-

396

the plate method for determining abundance or bacteria in ... watet Proc. Soc. E..p. Biol. Med. 96, 392-394.

(391 St•ples. D.G. and Fry. J.C. (1973) Factors which in-n'-leflce the enumeration of Btkllooibrio boderiooorous in -aae and river water. J. Appl. Bacteriol. 36. 1-ll.

(401 Ramsay. A.1. and fry, J.C. (1976) Response of epiphytic bacteria to the treatment of two aquatic macrophyte:s with the herbicide paraquat. Water Res. 10, 453-459.

(41) Vlltllnen, P. (1977) Elfect• of composition of substrate and inoculation technique on plate counts of bacteria in the nonhern Bailie .... J. Appl. Baccerial. 42, 437-443.

(42) Erickson. J.E. and Deibel, R.H. (1978) New medium for rapid screening and enumeration of Clastridium perfrfn .. ''"' in foods. Appl. Environ. Microbiot. 36, 567-571.

(431 Wachenheim, D.E. and Hespell. R.B. (1984) lnhibicory effects of 1itanium (III) citrate on enumeration of bacteria from rumen contents. Appl. Environ. Microbio1. 48. 444-445.

(441 BusUI, F.F. (1978) Introduction to injury and repair ol microbial cells. Adv. Appl. Microbiol. 23, 195-201.

(451 Reichard~ W. (1979) lnnuence of tempcracure and sub-strate 1hocb on survival and suecinic dehydrosenase activity of heterotrophic fruhwater bacleria. Water Res. 13, 1149-1154.

1461 Olalu:zuk-Neyman. K.M. and Vosjan, J.H. (1977) Mca· suring respiratory ctectron-tranlpOrt•systcm activity ln marine udimcnt. Netherlands J. Sea Res. 11, 1-13.

(471 Packard, T.T. (1971) The measurement of respiratory ekctron·transpon activity in marine phy1oplank1on. J. Mar. Res. 29, 235-244.

(481 Trcvors. J.T. (1984) The measurement ol eleccron crans-port system (ETS) activity in freshwater sediment. Water Ra. 18, 581-584.

(491 Tabor, P.S. and Neihol, R.A. (1982) Improved method for determination of respiring individual miCfOOfgAnisms in natural waters. Appl. Environ. Microbio1. 43. 1249-1255.

(:Mii Curl Jr_ H. and Sandberg. J. (1961) The measurement of dchydrogcnase activity in marine organisms. J. Mar. Res. 19, 123-138.

ISi) luuriaga. von R. and Rheinheimer. G. (1975) Einc einlache Methode zur Auszllhluns von Bakterien mil aktivem Elektroncruransportsystem in was.scr.. und sedimentprobcn. lnst. Meercskunde. Univcnitlt Kiel 31. 83-116.

(521 Lenhard, G. (1956) Die 0.hyd_.,..ktivitlt de• Bodens als Mus l\\r die Mikroorpnismentltigkeit im Boden. Z. Pftanzenemllhrung. Dunaung. Bodenkunde 73. 1-11.

(531 Logue. C, Koopman, B. and Bitton, 0. (1983) INT-re· dllClion iSUys and control of sludge bulking. J. Env. EllJ. 109, 915-923.

(54) Ohle, W. (1972) Meuuring che dehydrogcnasc activity in bottom sediments by using uurazotium chloride. in Tech~ niquet for the A1JeUment or Microbial Production and Oecomposilion in Fresh Waters (Sorokin, Y.I. and KadoUI, H., Eds.) pp. 27-28. IBP Handbook 23. Black· well. Oxford.

(SSJ Klnpwijk. A.. Drent, J. and S1eenwoorden, J.H.A.M. (1974) A modified procedure for the lTC-dchydrosenase test in acd\'at.ed studgc. Waler Res. 8, 121-ll!L

156J Christensen. J.P. and Packard, T.T. (1979) Respiratory electron transport activhies in phytoplankton and bacteria: comparison of methods, Limnol. Oceanogr. 24, 576-583.

(S7J Owens. T.G. and King, f.D. (1975) The measurement of rupir11,ory electron·transport~system activity in marine z.ooplankton. Mar. Biol. 30, 27-36.

(S8J Kenner. R.A. and Ahmed, S.I. (1975) Mcaluremcnt of electron transport aciivitiCs in marine phycoplankton. Mar. Biol. 33, 119-127.

(S9J Zimmermann. R.. hurriaga. R. and Becker-Birck, J, ( 1973) Simultaneous determination or the total number of aquatic bacteria and the number thereof involved in respiration. Appl. Environ. Microbial. 36, 926-935.

160) Maki. J.S. and Remson. C.C. (1981) Comparison of two dtrect·coun1 methods for determining metabo1Wng bacteria in fresh water. Appl. Environ, MJcrobiot. 41, 1132-1138. '

1611 Trevors, J.T. (1983) A note on viability measurements in Socdraromy.:es spp. Biotechnal. Lett. 5, 363-364.

(621 Novitsky. J.A. and Morita, R.Y. (1977) Survival ol a psychrophiUc marine vibrio under long·term nutrient starvation. Appl. Environ. Microbial. 33, 63S-641.

(631 S1evenson. L.H. (1978) A case for bacterial dormancy in aquatic systems. Microb. Ecol. 4, 127-133.

1641 Bright. J.J. and Fletcher, M. (1983) Amino acid auimila· don and electron transport system activity in attached and rree living marine bacteria. Appl. Environ. Mi~ crobial. 45. 818-825.

(651 Morica. R.Y. (1982) Starvation-survival of hcterOlroph> in the marine environment. Adv. Microb. Ecol. 6, 171-197. .

1661 Newell. S.Y. (1984) Modificacion of the selatin·matrix method for enumeration or rcspirina bacterial cells lor use with salt·marsh water samples. Appl. Environ. Mi~ crobiol. 47.873-873.

[671 Liu. D. (1983) Resazurin reduction method tor activated sludse process control. Environ. Sci. Technol. 17, 407-411.

(68) Wright. R.T. and Burnison, B.K. (1979) HeterOlrophic acli'vity measured with radiolabcllcd organic substrates, in Native Aquatic Batteria: Enumeradon. Activity and Ecology. ASTM STP 695 (Costcnon. J. W. and Colwell, R.R., Eds.) pp. 140-155. American Society for Tescing and Materials. Philadelphia.

(691 Findlay, R.H. and While. D.C. (1984) In situ determina-tion of me1abolic~activity in aquatic environments. Mi· crobiot. Sci. I, 90-95.

(101 Cassida Jr., L.E. (1971) Microorganisnu in unamend~ soil u observed by various forms or microscopy and stainillj. Appl. Microbiol. 21, 1040-1045.

(711 fry, J.C. and Humphrey, N.C.B. (1978) Techniques lor the siudy or bacteria epiphytic on aquatic macrophytes in Techniques for the S1udy of Mixed Populations (Lovelock, D.W. and Davies. R.. Eds.) pp. 1-29. Sociocy

-32-

for Applied Bacteriology, Technical Series 11. Academic Pn.'!i5, London,

1721 Saloncn, K. (1977) The e~timarion of bacterioplankton numbers and biomass by phase contrast microscopy. Ann. llot. Fenn. 14, 25-28.

f73J Jones, J.G. (1974) A method for observation and enu~ meration of epilithic algae directly on the surfacf!' or stones. Oecologia 16, 1-8.

f74J Strugger, S. (1948) FJuorcscence microscope eumination of bacteria in 50il. Can. J. Res. Ser. C. 26, 183-193,

(751 Hobbie. J.E .. Daley, R.J. and JasJl<r. S. (1977) Use of Nudepore filters for counting bacteria by nuoresccnce microscopy, Appl. Environ. Microbiol. 33, 1225-1228.

{761 Ramsay. A.J. (1978) Direct counts of bacteria by a modified acridine orange method in relation to their hcterotrophic activity. N.Z. J. Mar. freshwater Res. 12, 265-269.

(77} Daley, R. and Hobbie. J.E. (1975) Direct counts ol aquatic bacteria by a modified epiOuorescence technique. Limnol. O<;eanogr. 20. 875-g82.

(781 Babiuk. L.A. and Paul. E.A. (1969) The use ol Ouo· telcein isothiocyanate in the determination of the bacterial biomass of grassland soil. Can J. Microbiot 16, 57-62.

[79] Grea"s. M.P .. Cooper. S.L.. Davies, H.A., Marsh. J.A.P. and Wingfield, G.I. (1978) Methods of analysis for de-termining the effects or herbicides on soil micro~

organisms and their activities, Technical Report A.R.C. W~ Research Organisa1ion 4.S. Oxford.

180) Trevor;;, J.T .. Merrick, R.L .. Russell. I. and Stewart, O.G. (1983) A compari50lt of methods for assessing yeast viability. Biotechnol. Lett. 5, 131-134.

(81) Jones, J.0. and Simon, B.M. (197.S) An investigation of errors in direct counts of aquatic bacteria by epinuores-cenee microscopy with reference to a new method for dyeing membrane filters. J, Appl. Bactcriol. 39, 317-329.

[821 Buhrer. H. (1977) Verbesserte Acridinorangemethode zur Direktzii.hlung von Bakterien aus Scescdiment. Schwciz. Z. Hydro!. 39, 99-103.

J83) Bowden, W.B. (1977) Comparison ol two direct count techniques for enumerating aquatic bacteria. Appt En-viron. M1crobiol. 33, 1229-1232.

[g4] Watson, S.W .. Novitsky, T J .. Quinby. H.L. and Valoi-. F.W. (1971) Determination of bJKteriaJ number and bio-mass in the marine environment Appl. Environ. Mi-crobiof. 33, 940-946.

(851 Matsuyama, T. (1984) Staining of living bacteria with rhodamine 123. FEMS Microbiol. Lei!. 21. 153-157.

!861 Cbnanowski, T.H .. Crotty, R.D., Hubbard. J.D. and Welch, R.P. (1984) Applicability of the fluoresoein di· acetate method of detecting active bacteria in freshwater, Microl>. Ecol. IO, 179-lgs.

lg7J Snyder. A.P. and Greenberg. D.B. (1984) Viable micro-organism detection by induced fluorescence. Biotechn-01. Bioeng. 26. 1395-1397.

(88) Rotman, B. and PaJl<rrnaster, 8.W. (1966) Membrane properties of living mammalian cells as studied by en~

397

z.ymatic byJro1ysis of Ouorogenic esters. Proc. Natl. Acad. Sci. USA SS. 134-141.

f891 Paton. A.M. and Jones, S.M. (1975) The observation and enumeration of microorganisms in fluids using mem-brane lihrarion and lncidcnt fluorescence micr~y, J, Appl. Bt1C1eriol. 38, 199-200.

(901 Poglazov. M.N. and Mitskevich. l.N. (1984) Use of nu.,.. rescamin for determining the number or microorganisms in s~ water by 1hc epiOuorescence method. Mikrobio-logiya 53, 850-858.

(91) Meyer-Reil. L.·A. (1978) Autoradiography and e~ifluo­rcscem.-e microscopy combined for 1hc determination of numbers and spectrum of actively metaboliz.ing bBCteria in natural waters. Appl. Environ, Microbiol. 36. S06-S12.

!921 Schmidt. E.L. (1974) Quantitative autecological study of microorgnni~ms in soil by immunofluorw:ence, Sot! Sci 118. 141-149.

1931 Bolhool. 8.D. and Schmidt. E.L. (1980) The immuno· Ouoresccncc approach in microbial e<:ology. Adv. Mi-crob. Ecol. 4. 203-241.

(94) Gray. T.R.G. and Williams. S.T. (1971) Microbial pro-ductivity in soi1. in Microbes and Biological Productivity (Hughes. D.E. and Rose, A.H .. Eds.) Proc. Symp. Soc. Gen. Microbiol. 21, pp. 255-2g6, Cambridge University Press, Cambridge.

(951 Dryers. J.D. and Mason, C.A. (1986) 8iopolymer par-ticulate turnover in wastewater treatment systems: a review. Bloproc. Eng,. in press.

1961 Daw.,, E.A. (1976) Endogenous metabolism and the sun·ival of staf\-ed procllf)'otes. in The Survival Of Vegetative Microbes (Gray. T.R.G. and Poatgate, J.R .. Eds.) Proc. Symp. Soc. Gen. Microbiol. 26, pp. 19-53. Cambridge University Press, Cambridge.

1971 Bauchop. T. and Elsdon, S.R. (1960) The growth of micro-organisms in relation to their energy supply. J. Gen. Microbiol, 23. 457-469.

[981 Stouthamer. A.H. (1977) Energetic aspects of the growth of microorganisms, in Microbial Energe'lics (Haddock. 8.A. and Hamilton, W.A., Eds.) Proc. Symp. Soc. Gen. Microbiot. 27. pp. 28S-31S. Cambridge University Press, Cambridge.

!991 Stouthamer. A.H. (1976) Yield Studies in Micro· organisms. Putterns of Progress. Meadowtield Press. U,K.

[1001 Stouthamer. A.H. (1979) The search for correlation be· twC'fn theore1icat and C1'perimentaJ grow1h yield. Int. Rev. Biochem. 21. 1-45.

UOlJ Tempest. O.W, (1978) The biochemical significance of microbial growth yidds: a reamssment. Trends Bio· chem. Sci. 3. 180- I 84.

(1021 Pirt. SJ. (1965) The maintenance energy of bac1eria in growing cultures. Proc. R. Soc. Lond. Ser. B 163, 224-231,

(1031 Stoutbamer. A.H. and B<ttenbaussen, C. (1973) Utilisa· tion of energy for growth and maintenance in continuous and batch (:uhure of mictoorganisms. A reevaluation of the method for the determination of ATP production by measu!'ing molar growth yields. Piochim. Biophys. Acla 301. 53-70.

-33-

(1041 Ncijssel, O.M. and Tempest. D.W. (1976) Biocnergeiic aspects of aerobic growth of Kleb.riella ueroJ:t'nes NCTC418 in carbon~limited and carhon·s:ufficient cheroostal cultu"'. An:h. Miorobiol. 107, 215-221.

(IOSJ Pin. SJ. (1982) Mainl<nance energy a general model for energy-Umi~ and energy.suWcient growth. Arch. Mierobiol. 133. 300-302.

(1061 Tempa1, D.W. and Neijssel. O.M. (19114) The Sta1us of Y>.TP and maintenance energy as bio.logica:lly incerpreta· hie phenomena. Annu. Rev. Miorobiol. 38, 459-486.

(1071 Tempe$1, D.W. and Neijssd. 0.M. (1980) Comparative aspects or microhial growth yields with specfol reference to C1 utilizers. in Microbial Growth on C 1 Compounds (Dalton. H .. Ed.) pp. 325-334. Heyden. London.

(108] Haddock. B.A. and Jones, C.W. (1977) &cterial respira-tion. Bacterial. Rev. 41, 47-99.

11091 Mitchell, P, (1966) Chemiosmotic coupling in oxidntive and photosynthctio phosphorylation. Bacterial. Rev. 41. 445-502.

11101 Nioholls. D.G. (1982) Bioenergetics-an Introduction 10 the Chemiosmotic Theory. Academic Press, London.

(1111 Konings, W.N. (1985) Generation ol metabolic energy by end-product crnu .. Trends Biochem. Sci. 10, 317-319.

(1121 MacNab, R.M. and Aizawa. S-1. (1984) Bacterial motili-ty and tht bacterial nageUa.r motor. Annu, Rev. Biophys. lliocng. 13, Sl-83.

(Ill( Konings. W.N., Hellingwerf, K.J. and Robillard. G.T. (1981) Transpon across bacterial membranes, in Mem~ branc Transport (Bonting. S.C. and de Pont. J.J.H.H.M., Eds.) pp. 251-283. Elsevier/North Holland, Amsterdam.

(114] n1u, C,, Dasoa, E. and Boquet. P·L (1980) The energy dependent degradation of guanosine S '...Ji phosphate 3' ~ diphosphate in Esclterl<hia coli. Eur J. Biochem. 103, 117-124.

(1151 Laanc. C .. Krone. W., Konings, W.N .. Haaker. H. and Veeger. C. (1979) The inY(!lvement of the membrane potential in nitrogen fixation by bacteroids. of Rhitobium kgumlltOUlrwn. FEBS Lett. 103. 328-332.

(1161 Santos, E. and Kaback, H.R. (1981) Involvement of the proton electrochemical gradient in genetic transfarma* tion in £Jcltuidtia coli. Biochem. Biophys. Res. Com~ mun. 99, 1153-1160.

(1171 Zilbmtcin. D .. Agmon. V., Schuldiner. S. and Padan. E. (1984) Ei<h<rldila coli intracellular pH, membrane pol<ntial and cell growth. J. Bacteriol. 158. 246-2S2.

(1181 Konings, W.N. and Veldkamp. H. (1983) Energy trans.. duction and solute transport mechanisms in relation to environment& occupied by microorganis~ in Microbes in their Natural Environment (Slater. J.H. and Wimpenny, J.W.T., Eds.) Proc. Symp. Soc. Gen. Micro-biot 34, pp. ISJ-IS6, Cambridge University Pttss. Cam-bridge.

(1191 Hucting. S .. dclange. T.D. and Tempest. D.W. (1979) Energy rcquin:ment for maintenance of the lransmem~ brane potassium gradient in K.lebsitlla aerogenes NCTC418: a continuous culture study. Arch. Microbiol. 123, 183-188.

(120! Mich<ls, P.A.M., Michel,, J.P.J .. Boonstra. J. and Kon· ings. W.N. (1979) Generation oC an electrochemical pm· ion gradiem in bacteria by 1he excretion of me1aholk end products.. FEMS Mkrohiol. Lett, S, 357-364,

(121! Kooings, W.N .. Hellingwerr. K.J. and Elferink, M.G.L. (1984) The inter3Clion between eJeetron transfer. proton motive rorce and solute transport in bacteria. Antonie van Leeuwenhoek SO, 545-555.

(1221 Glagolcv, A.N. and Skulachev, V.P. (1918) The proton pump is a molecular engine of mo1ile bacteria. Nature 272. 280-282.

(123J Ordal. G. (1985) Bacterial ch<momi>: biochemistry of behaviour in a single ce:Jt. CRC Crit. Rev, Microbiol. 12. 95-130.

(1241 Khan. S. and MacNab. R.M. (1985) Proton chemical potential. proton electrical potencial and bacterial motili-ty. J. Mot. Biol. 138. 599-614.

(125] Chesbro. W., Evans. T. and Eifert, R. (1979) Very slow growth of E1cherichia coli. J. Bacteriot 139. 625-638.

(l26J Van Verseveld, H.W .. Chesbro. W.R .. Brester. M. and Stouthamer. A.H. (1984) Eubacteria have 3 growth modes keyed to nutrient flow-consequences for the concept o( maintenance and maximal growth yield. Arch. Micro .. biol. ll7. 176-184.

1127) Stouthamer. A.H. (1984) The relation between biomass production and substrate consumption at very low growth ra1es, in Innovations in Bio1echno1ogy (Houwink. E.H. and van der Meer, R.R .. Eds.) pp. 517-529. Elsevier, Amsterdam.

(128) Gallant. J.A. (1979) Stringent control in £. coli. Annu. Rev. Genet. lJ, 393-415.

IJ29( Cozzone. A.J. (1981) How do bacteria synthesise pro1eins during amino acid starvation? Trends Biochem. Sci. 6. 108-110.

(130] Zyc:hlinski. E. and Matin, A. (1983) Effect of 51arvation on cytoplasmic pH. proton motive force and viability of an acidophilic bacterium Thicbocillus acidophi/u.t, J. Bacteriol. 153, 371-374.

(1311 Otto. R., Vije. J .. ten Brink. B. and Konings. W.N. (1985) Energy me1aboJi11m in Str~p.t«occu.t c"111(>ris during lac1ose starvation. Arch. Mictobiol. 141. 348-352.

ll321 ten Brink. 8. and Konings, W.N. (1982) Electrochemical proton gradient and lactate concentration gradient in S1nptOC'OC'nlS cremoris cells grown in batch cuhure. J. Bacteriol. I S2. 682-686.

(1331 Padan, E., Zilberstcin, D. nnd Rottenberg. H. (1976) The protot1 electrochemical gradient ln Escherichia coli cel1s, Eur. J. Biochem. 63, 533-541.

(134] Boyaval. P .. Boyaval, E. and Dcsma:eaud. M.J. (1985) Survival of Brevibacteriwn liMns during nutrient starva-tion and intraceltu1ar changes. Arch. Microbiol. 141, 128-132.

(135) Brc:uil. C. and Patel. G.B. (19!!0) Composition of Metl«m01pitillum ltungaltii OPl during growlh on dlfter~ ent media. Can. J. Microbiol. 26. 571-582.

11361 Parquet. C .• Flourct, II .• Leduc. M., Hirota. Y. and van Heijenoort. J. (1983) N-acetylmuramoyl~L·a1kaline

-34-

amiJase or £scMrichfr1 C"oli K12. Eur. l Biochem, 133. 371-377.

(137) Preiss. J, (1984) Bacterial glycogen syn1hesis and its regulation, Annu. Rev. MicrobioJ. 38. 419-458.

(138) Ramsay. H.H. (1%2) Endogenous respiradon or $taphy1««cus Qurekf. J. Bacteriol. 8), 507-514.

(139) Krzcmihski. Z .. Mikucld. J. and Szarapihska~Kwas~

uwska, J. (1972) Endogenous metabolism ol Staphy/o-ct><cus aureus. Folta Microbiol. 17, 46-54.

(1401 Su:wciyk, E. and Mickucki, J. (1983) Protein A as a :!'ubstratc of endogenous metabolism in staphylococci. FEMS Microbial. Leu. 19. 55-58.

11411 Ryan, FJ. (1959) Bacterial mutation in a stationary phase and the question of cell turnover. J. Gen. Micro. biol. 21. 530-549.

1142) Koch. A.L. (1959) Death ol bacteria in growing culture. J. Bactcriol. 77, 623-627.

1143) Postgate, J.R. and Hunter, J.R. (1962) The survival ol starved bacteria. J. Gen. Microbiol. 29. 2)3-263.

(144) Postgate, J.R. and Hunter, J.R. (1963) The survival of starved bacteria. J. Appl. Bacmiol. 26, 295-306.

fl45) Niob. I. and Furusaka, C. (1968) Growth or bacteria in the hcat~killed s1upensions of the same bacteria. J, Gen, Appl. Microbial. 14. 373-385.

1146) Hamer. G. and Brycrs, J.D. (1985) Aerobic thermophillic sludge treatment. Some biotechnological concepts, in Waste Treatment and UtiHsation, Proc. 3rd IWTUS Int. Waste Treatment and Utilisation Symposium (Dryers. J.D .. Hamer, G. and Moo-Young. M., Eds.) Conserv. Rccycl. 8, 267-285.

1147) Linton. J.D. and Stephenson, R.J. (1978) A preliminary study on growth yields in relation to the carbon and energy content of various organic growth substrates. FEMS Microbial. Lett. 3, 95-98.

1148) Mason, C.A. and Hamer, G. Cryptic growth in KJ.l>siel/o pMwnonioe, submitted for publication.

(149) Gaudy Jr., A.F., Yang. P.Y. and Obayashi, A.W. (1971) Studies on the total oxidation of activated sludge with and without hydrolylie pretreatment. J. Water PoHut. Contr. Fed. 43, 40-54.

(I SQ) Tomasz. A. (1968) Biological conscqucnccs or the replacement of choline by ethanolaminc in the cell wan of pncumococcus. Chain formation. loss of transforma-bility and loss of autolysis. Proc. Natl. Acad. Sci. USA 59, 86-93.

11511 Fein. J.F. and Roge .... H.J. (1976) Autolytic cntyme-defi-cicnt mutants of Bacillus sulnllis 168. J. Bacterio1. 127, 1427-1442.

(1521 Rogers. H.J. (1970) Bacterial growth and the cell en· velope. Bacteriol. Rev. 34. 194-214.

(153! Wong. W., Young. F.E. and Chatterjee, A.N. (1974) Regulation of bacterial ccl1 walJs: turnover or cell wall in Stoph.>'IO<OCcus aureus. J. Bacterial. 120, 837-843.

1154) Glaser. L. and Lindsay. B. (1977) Relation between cell wall turnover and cell growth in IJat'illus &ubtllis. J, Bacteriol. 130. 610-619.

(155) Frehel. C. and Ryter, A. (1979) Peptidoglycan turnover

399

during growch of a Bacillu.1 nr~f(otrrlum Oap - lys-mutant. J. Bacteriol. 137, 947-955.

fl56j DeBoer. W.R .. Meyer. P.O .. Jordans. C.G .. Kruyssen, F.J. and Wouters. J.T.M. {1982) Cell wall turnover in growing and nongrowing c1,dtures or Batillus subtilis. J. Bacterial. 149. 977-984.

11571 Fein. J.F. (1979) Possible involvement or bacteri•I auto-tytie enzymes in Oage11ar morphogenesis. J. Bacteriol. 137, 933-946.

1158) Danco-Moore, L. and Shockman. G.D. (1977) The bacterial cell surface in growth and division, in The Synthesis. Assembly and Turnover of CeU Surface Com~ ponents (Poste. G. and Nicolson, G.L., Eds.) Cell Surface Reviews 4. pp. 597-715. Elsevier/North Holland. Amsterdam.

[1591 Rogers, HJ. (1979) The function ol bacterial autolysins. In Microbial Polysaccharides and Polysaccharascs (Berkley. R.C.W., Gooday. G.W. and Ellwood, D.C., Eds.) pp. 237-268. Society for General Microbiology. Academic Press, London.

1160) Shockman. G.D. and Barran, J.F. (1983) Structure. func-tion and assembJy or ceU walls of Gram~poshive bacteria. Annu. Rev. Microbial. 37, 501-527.

11611 Drozd, J.W .. Linton, J.D .• Downs. J. and StcphcnlOn, R.J. (1978) An in situ assessment of the specir.c lysls rate in continuous cultures of Merhylocomn sp. (NCIB11083) grown on methane. FEMS Microbial. Lett. 4, 311-314.

11621 Mason, C.A.. Bryers. J.D. and Hamer. 0. (1986) Activ-ity, death and Jysis during microbial growth in a chem .. ostat. Chem. Eng. Commun. 45, 163-176.

!163) Rogers. H.J. and Forsberg. C.W. (1971) Role ol auto-1ysins ln the kitling or bac1eria by some bactericidal antibiotics. J. Bacteriol. 108, 1235-1243.

1164! Sayare. M .. Dane<>-Moorc, Land Shockman, G.D. (1972) Innuence of macromolecular biosyn1hesis on cellular au· tolysis in S1rep1«0«us fa«olis. J. Bacteriol. 112. 337-344.

(165] Leduc. M., Kasra, R. and van Heijenoort, J. (1982) Induction and control of the autolytic system of Eschtric/lia coli. J. Bacterial. 152, 26-34.

[166) Jolirre, L.K .• Doyle, R.J. and Strcips. U.N. (1981) The energjsed membrane and cellular autolysis in Bodi/us sub1ili1. Cell 25, 1S3-76J.

(167) Arias. J.M .. Fcrnandet·Vivas. A. and Montoya, E. (1983) Evidence for an activating substance relaced to auiolysi' in Myxo<IJ«us corolloides D. Arch. Microbiol. 134. 164-166.

[168) Kawamura, T. and Shockman, G.D. (1983) Purification and some propcrtia of the endogenous autoJytie N~acetylmuramoylhydrolasc of Str<pt«o«us /attium. a bacterial glycocnzyme. J. Biol. Chem. 258. 9514-9521.

(169) Sbockman. G.D .. Kawamura, T., Barrett, J.F. and Dolinger. D.L. (1985) The autolytic peptidoglyean hy· drotases of Streptoe«cus faeciJPtt. Ann. Inst. Pasteur 136. 6J-66.

(170) Vitl<ovi~. L.. Cheung, H.·Y. and Freese, E. (1984) Af>. sence of correlation bet~n rates of cell wall turnover

-35-

and autolysls shown ':>}' Bacillus subtiliS mut4nts. J. Bacteriol. 157, 318-320.

(171) Pooley, H.M. (1976) Turno- and spreading or old wall during surrace grow1h of Bacillus subtilis.. J. Bncteriol. 125, 1127-1138.

(1121 Pooley, H.M. (1976) Layered distribution according to cdl agef within the cell wall of Bac;/lu.r subrifiJ. J, Bacterio!. 125, 1139-1147.

(173) Brown. W.C. and Young. F.E. (1970) Dynamic interac· tions between cell waH polymers, C>ttraceUular proleascs arul autolytic enzymes. Biochem. Biophys. Res. Com· mun. JS, 564-568.

(174) Herbold, D.R. and Glaser. L. (1975) Bacillus subrllis N-acetylmuramic add L·alaninc amidase, J. Blot. Chem. 2:111. 1676-1682.

(17') Giudiodli, S. and Tomasi, A. (1984) Atlftchmcnt or pncumocoecal autolysin to wall teichoic acids, an esscn .. tial step in enzymatic wall degradation. J. Bactcriol. lSS. 1188-1190.

(176) Tsuchido, T .. Hiraoka. T., Takano, M. and Shibasaki, I. (lffS) lnvolvanent or autolysin In <ollular lys!s of a.,.;t. /us sr.brillt indU«d by llilott· and medium-ehain fatty acida. J, Bacterial. 162, 42-46.

(177) Arias, J.M. arul Montoya, E. (1978) Dispersed growth and cell lysis in Myxocoa:us corolloitkJ. Microbios Leu. 5, 81-84.

(1781 Manincz. MJ .. Reyes. F., Lah0<, R. and Pem·Leblic, M.I. (198J) Lytic enzymes in autolysis or Boiry1is c/Mrea. FEMS Microbial. Lett. 19, 157-160.

{179) Konig, H., Semmler. R .. Lerp, C. and Winter, J. (1985) Evidence for the occurrence of autolytic ... ymes in Mttltanobat:t<rium wol/el. Arch. Microbial. 141. 177-180.

(1801 Babayan, T.L a.nd Beuukov, M.G. (1985) Autolysis in yea .... Acta Biotechnol. S, 129-136.

(1811 Nombda. C. (Ed.) (1984) Microbial Cell Wall Synthesis arul Autolysis. Proceedings FEMS Symposium. Elsevier, Amsterdam.

(1821 Wu, S.Y. and Klein. D.A. (1976) Swrvation effects on Euli~rldria coli and aquatic bacterial tap0nses to nutfi,. ent additlon and secondary warming stresses. Appl. En .. viron. Microbial. Jl, 216-220.

(IBJ) Ru ... n, A.D. (1984) Potential sites of damage in micro-organisms exposed to chemical or physical qents, in The Revival or Injured Microbes (Andrew, M.H.E. and RUS• sell, A.D" Eds.) Symp. Soc. Appl. Bacterial. 12. pp. 1-111. Academic Pras, London.

(1841 Hunt, A. (1977) Bacterial ittjury: a review. Can. J. Microbiol. 23, 936-944.

(llSI Curran, H.R. and Evans, F.R. (19J7) The importan« of el'lrichmenta in the cultivation Of bacterial spores previ· ously exposed to lethal agents. J. Bacteriol. J4, 179-189.

(11161 Rose. A.H. (1978) Chemical MicrobiotosY, An lntroduc· tion to Microbial Physiology, Jrd ed. Butterworth .. London.

1117] Hugo, W.B. (1976) Survival ol microbes ellpOled to chemical str..., in The Survival of Vegetative Microbes (Gray, T.R.G. and Postgate, J.R., Eds.) Proc. Symp. Soc.

Gen. Microbio1. 26, pp. 383~413. Cambridge University Press. Cambridge.

{1881 Cornell, J.B. and Shockman, G.D. (1978) Cellular lysis of SlreptfXO«US fattalis induced with triton X~JOO. J, Bactoriol. 135, 153-160.

(1891 Gale, E.F .. Cundlille, E .• Reynolds. P.E •• Richmond. M.H. and Waring. M.J. (1972) The Molecular B35is or Antibiotic Ac1ion. Wiley. New York.

(1901 P..thica, B.A. (1958) Bacterial lysis. Lysis by physiologi· cal and chemical methods. J. Oen. Microbiol. 18. 413-480.

(191) Salton, M.R.J. (196J) The relationship between the na-ture of the cell waU and the Gram slain. J. Gen. Micro~ biol. JO, 223-235.

(1921 Harold, F.M. (1910) Antimicrobial a3ents and mom· brane function. Adv. Microb. Physiol. 4. 45-lOJ.

(J9JJ Razin, S. and Argaman. M. (1963) Lysis or mycoplasma. bacterial protoplasts. spheroplasts and L~forms by vari· ous agent>. J. Gen. Micsobiol. 30, 155-172.

(1941 Hugo. W.B. (1967) The mode of action of antibacterial agents. J. Appl. Bacterial. JO, 17-SO.

(195] Russell, A.O. (1971) Ethylenediaminetetraacctic acid, in Inhibition and Destruction ol the Microbial Cell (Hugo, W.B., Ed.) pp. 209-224. Academic Press, London.

(196) Nakamura, K. arul Tamaolci. T. (1968) Reversible dis· socialion of Esclttrichla coli ribosomes by hydrogen per~ oltide. Biochim. Biophys. Acta. 161. 368-J76.

(1971 Carlsson, J., F~lander, F. and Sundquist, G. (1917) Oxygen tolerance of anaerobic bacteria isolated from necrotic dental pulps. Acta Odont. Scand. JS, IJ9-145.

(198) Shoesmith, J.O. and Warsley, B. (1984) Anaerobes and exposure to oxygen, in The Revival of Injured Microbes (Andrew, M.H.E. and Russell, A.D .. Eds.) Soc. Appl. Bacttriol. Symp. Ser. 12, pp. 127-146. Academic Press, London.

{199) Macl.eod. R.A. and Calcou, P.H. (1976) Cold shock and freezing damage to microbes, in The survival of Vegeta· tive Microbes (Gray, T.R.G. and Postgate, J.R .. Eds.) Proc. Symp. Soc. Gen. Microbiol. 26. pp. 81-109. Cam· bridge Unive,.ity Press. Cambridge.

(200) Sato, M. and Takahashi, H. (1970) Cold shock or bacteria, 1\1. Involvement or DNA·ligasc reaction in recovery of EsdterU:hla <oli from cold shock. J. Gen. Appl. Micro· biol. u. 217-229.

(201) Edebo. L. and Magnusson, K.·E. (lll7J) Disintegration of cells and protein recovery. Puro Appl. Chem. 36, J2S-338.

(202) Maxur, P. (1966) Physical and chemical basis for injury ln single celled microorganisms subjected to frec:zing and thawing. in Cryobiology (Meryman, H.T., Ed.) pp. 213-JIS. Academic Pres .. London.

(2031 Mackey, P. (1984) Lethal and &ublethal effects or ,.. frigeration. free:r.ing and frcez.c drying on micr<r organisms. in The Revival of lnjured Microbes {Andrew, M.H.E. and Russell, A.O., Eds.) Soc. Appl. Bacteriol. Symp. Ser. 12. pp. 45- 75. Academic Pras. London.

1204) Sedgwick, S.O. and Bridges, B.A. (1912) Evidence for

indirect production of DNA strand scissions during mild heating of Escherichia coli. J. G~n. Microbiol. 71. 191-193.

(20S) Pellon, J.R .. Ulmer. K.M. and Gomez. R.F. (1980) Heat damage to the folded chromosome o( Escherichia coli Kil Appl. Environ. Microbial. 40, 3S8-364.

(2061 Scheie, P. and Ehrensp<ck, S. (1973) Larsc surface blebs on Esc:heridtia c:oll heated to inactivating temperatures. J. Bac1criol. 114, 814-818.

(207] Pellon, J.R. and Gomez. R.F. (1911) Repair of thermal damage to the Eschen'chia toli nucleoid. J. Bacteriol. 145, 1456-1458.

(208] Moseley. B.E.B. (1984) Radiation damage and its repair in non~sporu1ating bacteria. in The Revival of Injured Microbes (Andrew, M.H.E.. and Russell. A.O .. Eds.) Soc. Appl. Bacterial. Symp. Ser. 12, pp. 147-174. Academic Press. London.

f209l Hugo. W.B. (Ed.) (1971) Inhibition and destruction of the microbial cell. Academic Pre5s, London.

1210) Gray, T.R.G. and Postgate, J.R. (Eds.) (1976) The Sutvivat of Vegetative Microbes. Proc. Symp. Soc. Gen Microbiol. 26. Cambridge University Prw. Cambridge.

121!] Andrew. M.H.E. and Rusacll, A.O. (EdJ.) (1984) The Revival of 1'1iurcd Microbe•. Soc. Appl. Bacterial. Symp. Ser. 12. Academic Press, London.

1212] Topiwala. H.H. (1973) Mathematical models in microbi-ology. in Methods In Microbiolol)I (Norris, J.R. and Ribbon., D.W .. (Eds.) Vol. 8, pp. 35-59. Academic Press, London.

(213] Harrison, D.E.F. and Topiwala. H.H. (1974) Transient and oscillatory states or continuous culture. Adv. Bio.. chem. Eng. 3. 167-219.

12141 Buchanan. R.E. (1918) Lile phases in a bacterial culture. J. Infect. Dis. 23, 109-125.

(21Sf Monod, J. (1950) la technique de culture continue. thCorie ct applications. Ann. Inst. Pasteur 79. 390-410.

(216] Herbert. 0., Elsworth, R. and Telling. R.C. (1956) The

-36-

401

rontinuous culture of bac1cri•. a theoretical and expcri .. mental study. J, Gen. Microbial. 14, 601-622.

1217] Herbert. 0. (1958) Some principles of continuous cul· ture, in Recent Progress in Mic:robio1ogy (Tuncvall. G., Ed.) 7th Int. Congr. Microbial., pp. 381-396. Blackwell, Oxford.

[2181 Powell, E.O. (1967) The growth rate of micJ'OO!!lnisms IS a function or substrate concentration, in Microbial Physiolog)I and Continuous Culture (Powell, E.O., Evans, C.G.T., Strange, R.E. and Tempest, D.W., Eds.) Proc. 3rd Int. Symp., pp. 34-56. HMSO, London.

12191 Tempest. D.W .. Herbert, D. and Phipps, PJ. (1967) Studies on the growth of Atr0b«1er OefoSeMl •• low dilution rates in a chemostat, in Microbial Physiology and Continuous Culture (Powell, E..O., Evans, C.G.T., Strange, R.E. and Tempe5~ D.W., Eds.) Proc. 3rd Int. Symp. pp. 240-2S3. HMSO, London.

[2201 Postgate, J.R., Crumpton, J.E. and Hunter. J.R. (1961) The measurement of bacterial cultures by slide culture. J. Gen. Microbial. 24, I S-24.

12211 Sinclair. C.G. and Topiwala, ff.ff. (1970) Model for continuous culture which considcn the viability concept. Biotechnol. Bioeng. 12, 1069-1079.

1222] Weddle, C.l. and Jenkins, D. (1971) Th• 10ability and activity of activated &Judge. Water Res. S, 621-640.

(223] Grady Jr_ C.P.l. and Roper Jr., R.E. (1974) A model for the biooxidation process which incorporates the viability concept. Water Res. 8, 471-483.

(224) Jones. Q.L. (1973) Bacterial growth kinetics; mca&ure· mcnt and significance in the activated stvdp process. Water Res. 7. 1475-1492.

f225J Hamer. G. (1985) lysis and 'cl')lptic' growth in waste· water and sludge treatment processes. Acta Biotechnot 5, 117-127.

12261 Oujer, W. (1980) The effect of particulate organic llijlterial on activated sludge yield and oxygen require· ment. Proc. Water Technol. 12. 79-95.

-37-

CHAPTER 3

Chem. t:11g. Cl1mmw1. Vol. 45 pp. 163-176 (~J'J8-M45/86/450(1-0l<13$20.<Xl/0 © l'JH6, Gordon and Drench Science Publishers $.A. Printed in 1he United Stales of America.

ACTIVITY, DEATH AND LYSIS DURING MICROBIAL GROWTH IN A CHEMOSTAT

C.A. MASON, J.D. BRYERSt and G. HAMER

Institute for Aquatic Sciences, Swiss Federal lnstitllle of Technology Ziirich,

Ueberla11dstrasse 133, CH-8600 Diibendorf, Switzerland

(Received April 12, 1985;;,, fi11a/ form June 17, 19115)

The significance of death and lysis processes during microbial growth in chemostat cultures is discussed. A structured model is developed to describe such processes and 1he model is partially verified for carbon limited chemostat eul1ures of Klcbsiella p11eumo11iac. It seems probable that death and lysis arc coincident, when non-viable cells are not considered 10 he dead. KEYWORDS Growth Death Lysis Chemoslal Microbe

INTRODUCTION

Viability is a microbiological term which is rarely correctly interpreted. The main reason is that any definition is system specific and subject to the limitations of the methods used. Such operational definitions frequently lead to misinterpretation once they are extrapolated outside of the regime of their applicability. Recently, however, the consequences of inadequate quantification of viable and dead microorganisms have been realised in terms of public health hazards and in the inability to optimize processes orientated towards production of microbial cells and their products. Thus, there is a need to establish functional definitions and accurate methods to describe the physiological status of microorganisms and thereby contribute more meaningful information towards design criteria for industrial processes.

Nearly a quarter of a century ago Humphrey and Nickerson (1961) remarked that in spite of almost sixty years of intensive research, little understanding of the exact nature of bacterial death had been gained. This comment was in reference to sterilization operations' where the work published by Humphrey and his co-workers during the 1960's did much to elucidate the kinetics and mechanisms of death at elevated temperatures. However, the nature of bacterial death at temperatures either coincident with or close to the optimum for growth still requires resolution and one objective of this communication is to contribute to such resolution.

t Present address: Dept. of Civil und Environmental Engineering, Duke University, Durham, NC 27706, USA

163 -

-38-

164 C.A. MASON, J.D. BRYERS AND G. HAMER Hinshelwood (1951) discussed bacterial death, viability and lysis in terms of

their physiological basis in the cell. He proposed a simple model in which he derived equations similar to Volterra's, based on fluctuations in lysis and synthesis within the cell. To test these equations a vast range of methods for viability assessment have been developed and are well documented (Postgate 1967, Costerton & Colwell, 1979).

Several studies have shown the presence of non-viable microbes (e.g. Tempest et al., 1967, van Uden and Maderia-Lopes, 1976), and several existing models have incorporated the concepts of viability and death into traditional continuous culture theory. For example, Sinclair and Topiwala (1970) modelled the natural death of bacteria in steady state continuous culture systems, using the ratio of viable cell mass to total bacterial cell mass as an index of viability.

Their model considered two mechanisms by which viable cell mass could be lost; namely death and endogenous metabolism. Grady and Roper (1974) presented a model of the bio-oxidation process which incorporates a third physiological type of cell, in addition to the viable and dead, which they designated as cells which had lost the ability to divide. Of the processes resulting in the loss of various cell types from their system, endogenous metabolism, autolysis, lysis by other organisms and predation, only the first two processes were considered relevant to pure culture studies and these were grouped into a single term, decay.

Since lysis and endogenous metabolism effectively represent growth on the cells own constituents these can be separated from the lumped parameters of decay and considered as individual processes.

MODEL DEVELOPMENT

Bacteria, growing in continuous culture, are assumed for purposes of this study, to participate in the processes depicted in Figure 1. Based on this rationale, biomass dry weight, the common indicator of bacterial concentration (and activity), is replaced by a more "structured" description which necessitates the following groupings of cellular material within a culture:

Viable bacteria (Xv) are aerobic cells that consume dissolved organic compounds as carbon and energy substrates. Viable cells respire and replicate at a growth rate, µ, which is a function of dissolved carbon substrate concentration.

Non-viable, respiring bacteria (X • .,) are those cells that do not replicate but still moderate (perhaps enzymatically) substrate oxidation and are thus actively respiring. x ... cells arise only by inactivation of viable cells, X, ..

Dead bacteria (Xd) are those cells that are no longer able to respire and replicate. )(1 cells arise from both the "death" of viable and non-viable, respiring cells.

Biopolymeric particulates (P) are colloidal fragments of cellular origin .that arise due to the lysis of viable, non-viable and intact dead cells. Combined X,,, Xm,, X,1 and P would analytically comprise the total biomass concentration, X, in a

-39-

MICROBIAL GROWTH 165

soluble and particulate debris FIGURE 1 Fundamental processes considered to oceur in a chemostat.

culture. A more precise distinction between these biological particulates, based upon specific analytical methods, is detailed subsequently.

Mathematical Derivation

Accumulation of compounds in the bulk liquid of a completely mixed reactor can be described by mass balances of the general form:

net rate of = net rate of + net rate of accumulation transport transformation (1)

Material balances are organized here in matrix notation as described by Roels (1980).

The seven components of interest in this study are primary substrate con-centration, S., dissolved organic carbon compounds released during lysis, S2 ,

oxygen, 0 2 , viable cells, x;,, non-viable respiring cells, X.,.,. dead cells, Xd, and biopolymer particulates, P. The component vector, a, is defined as:

S1 S2 02

a= x., (2) xntl X,i p

Thus the left hand side of Eq. (I) is daldt.

-40-

166 C.A. MASON, J.D. BRYERS AND G. HAMER Jn a chemostat, receiving a sterile feed of substrate SV and being aerated by

sparging with air, the transport vector a• becomes:

a*=

D(S~- S1)

-DS2 K1.,..(0i - 02) - D02

-DXv -DXnu -DXd -DP

(3)

where D is the dilution rate, K1• the overall liquid side oxygen transfer coefficient and a, the specific gas-liquid interfacial area.

Processes in Figure 1 can be translated into the irreversible reactions, written in the unit mass based stoichiometry outlined by Irvine and Bryers (1985) as shown in Table I. Each reaction is normalized with respect to the mass of the individual

TABLE I

Stoichiometric reactions for growth, death, and lysis processes in a chemostat

Process .

Growth on S1 by viable cells

Growth on S2 by viable cells

Inactivation Death of viable

cells Death of non·

viable cells Lysis of non·

viable cells Lysis of dead

cells Lysis of viable

cells Hydrolysis Enzymatic oxi·

dation by non-viable cells

Enzymatic oxi-dation by non-viable cells

Reaction

(-I)S, -os2 - Y,_,o, + Y,_.x .. +ox ... + oxd +OP= o

os, + c-1)s,- Y,,,o,+ Y,,.x .. +ox ... +oxd+OP=O os, + os, - Y,.302 + (-l)X,. + (1".,,,)X.,,. + oxd +OP= 0

os, +os, - Y.,,,02 + (-l)X., +OXm, + (Y.,.)x. +OP =O

os, +os,- }". .. ,O, +ox .. + (-l)X.,.. + (Y, .• )X. +OP =O

05 1 + (Y,,_ 2)S2 + (0)0,+0X,. + (-l)Xm, + OXa+ (Y6 ,1)P=O

OS 1 + (Y1 . .JS2 + (0)02 +OX,,+ OX •• + (-l)Xd + (Y1,1)P =O

os, + (Y.,,Js, + (0)0,+ (-l)X,. +ox .... +oxd + (Y •. 1)P=O OS,+ (Y •. 2)S2 + (O)O, +ox .. + ox ... + oxd + (-l)P = 0

( - l)S1 + OS2 + (- Y10.;)02 +OX.. + ( Yoo,,)X.,. +OX.,+ OP = 0

OS,+ (-l)S2 + (-Y11,;)02+ ox..+ (Y11,,)X.,, +ox., +OP =O

Reaction rate

r,

f2 r,

r,

r,

r,,

r1

'• '•

rrn

'"

Equation

(4)

(5) (6)

(7)

(8)

(9)

(10)

(11) (12)

(13)

(14)

Y;,1 the stoichiometric cocflicicnt for the jth component in the ith reaction and is a unit mass stoichiometric value having been normalized by the limiting reactant. Units of Yi.; arc (mass of component j per mass of limiting reactant in reaction I). Consequently, Yi.1 values for the limiting reactant arc (-1). Note: rcactnnts arc assigned negative coefficients and products are positive.

-41-

MICROBIAL GROWTH 167

reactions limiting reactant. Reaction rates, r;, where i = I- I I, arc also in terms of the limiting reactant of the specific reaction. Stoichiometric coellicients in Tahle I are ratios between the various products formed or reactants consumed and the reactions rate limiting reactant. Thus these ratios are fundamentally characteristic for the experimental system selected. Observed "yields", which represent ratios between product and total reactant consumed, do not have the fundamental significance of the intrinsic stoichiometry and the two should not be confused. Intrinsic stoichiometry is inherent to the processes involved, whilst observed yields are artifacts of the reactor system and its operating conditions.

Reaction rate expressions for the eleven reactions in Table I are listed in Table II. Growth of viable cells, X.,. is presumed to be the sum of two Monod-type rate expressions; one based upon primary substrate, St> and the other based upon dissolved organics produced via lysis and hydrolysis, S2 • For lack of information and simplicity, all other rate expressions are presumed to be first order in their limiting reactant.

For simplicity, at this stage of model development oxygen consumption coefficients for the death and inactivation processes (Yu, Y... 3 and }".,, 3 ) (Eqns. 6-8) are assumed to be zero such that }".,, 5, Y4•6 and l';,6 are thus equivalent to unity, although in reality this is unlikely to be the case.

Combining stoichiometry and kinetics in a transformation vector, a*, results in: -r1 -r10

-r2 + (Y6,2)r6 + (Y1,2)r1 + (Ys.2)rs + (Y •. 2)r9- r11 ( - Y1,3)r, - ( Yi.»r2 - (l".1,3)r3 - ( Yu)r4 - (Y,,3)rs - (Yw,3)ru1 - ( Y11,.1)r1,

a*= ( + Y1•4)r1 + (Y2, 4)r2 - r3 - r4 - '" ( + Y1,s)r, - r.s - r6 + (Yio.s)rw + (Y11,s)rn ( + Y4,6)r4 + ( l".s,6)rs - r1 ( + Y6,1)r6 + (Y1,1)r1 + (Y ... 1)r8 - r9

TABLE II

Rate expressions for processes illustrated in Figure l

De.scription

Growth of viable cells on S1 Growth of viable cells on S2 Inactivation of viable cells Death of viable cells Death of non-viable cells Lysis of non-viable cells Lysis of dead cells Lysis of viable cells Hydrolysis of particulates Enzymatic conversion of S 1 by X.,., Enzymatic oxidation of S2 by X.,,,

Rate expression

r 1 = [µj · S1X.,l(K, 1 + S1)!/Yi.4 Tz = (µ1 · S2X..l(K,, + S2) IY,,, T3 =" k"J_)(, T4 = k.X,, r5 =k5Xm, ft) =k0m .. • r1= k1Xa r8 = kw\;., r9 =k9P

r10= !VioS1X .. ,.l(KM, + S,)]/Ym.5 '11 = [V;,s,x ... t(KM, + S:,)}IY11,,

Equation

(15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25)

(26)

µ: .. o is the appropriate maximum specific growth rate constant, K,w., the appropriate saturation constant. K .. m1,, the appropriate first.order rate constant~ V :uh• the appropriate maximum reaction rate and KM., .. the appropriate Michaelis-Mcntcn constant.

Parameter

s, s, o, x.. X.., xd p

TABLE !II

Material balances for a chemostat incorporating concepts of growth, death, lysis, and paniculate hydrolysis

Net rate of accumulation

dS 11dt dS21dt dO,tdt dX,.ldt dX,.,ldt dXdldt dPldt

Net rate of transport out

of reactor

D(S?-S,) -Ds, -D02 + K1.cr(Oi - 0 0) -DX,. -DXm. -DXd -DP

-r1 - 'to

Net rate of traruformati.on

-r1 + Y6.?'t. + Y1,2r1 + Yg.zrs + Y9.2'•J- ru -Y1,J'1 Y2.Jh - r,,3,J - Y4,Y'4 - r_..,.3rs - Y10.J'10 - Y11.Y1 t + Y1.4't + Y:?,4'2 - T3 - T4 - Tg

+Y3_5F3- rs- r6+ Y10.sr10+ Y11.s'f1 + Y.i,6'.t + Ys.6"s - r, + Y6,7'ti + Y1.7T7 + Y8,7,8 - T9

Equation

(28) (29) (30} (31) (32) (33) (34)

(j

> ::: > VI 0 z ... b tc :;i:i ....,: tT1 :;i:i VI

> z 0 Cl

= > ::: tT1

"

I ~

"' I

-43-

MICROBIAL GROWTH Rewriting Eq. (1) mathematically,

daldt =a*+ a*

169

(27)

Complete material balances for the seven components are listed in Table III. Unsteady state solution of the material balances required numerical integration.

EXPERIMENT AL

Organism Klebsiella pneumoniae NCIB 418, maintained by monthly subculture on plate count agar slopes.

Growth conditions Bacteria were grown in a 2.51 fermenter (MBR Bioreactor AG, Wetzikon, CH) on a simple salts medium (Evans et al., 1970) modified by substituting EDT A (100mg1- 1) for citrate as chelating agent. Polypropylene glycol (20mg1- 1) was also added as an antifoaming agent. Glucose (4.8g1- 1) was the sole carbon energy substrate and limited growth. Pure oxygen was mixed with influent air to maintain constant dissolved oxygen tension over the entire range of residence times investigated. The pH of the culture was automatically maintained at 6.80 by controlled addition of a mixture of 1.5 M NaOH and 1.5 M KOH. The temperature was maintained constant at 35"C and the inlet nutrient flow rate was measured by weight displacement.

Analyses

Total Biomass concentration was measured as dry weight (DW) by a filtration/ gravimetric procedure using tared 0.4 µm Nuclepore filters. Filters were dried at 105"C for 1 h.

Glucose was assayed enzymatically using the GOD-PeridR (Boehringer Mann-heim GmbH, D) according to the procedure described by Egli et al. (1983).

Dissolved Organic Carbon (S2) was measured in cell filtrates using a Tocor 2 (Maihak GmbH, Hamburg, D) dissolved organic carbon analyser.

Cell activity Experimental differentiation between living and dead cells was carried out using the INT-reduction assay of Zimmermann et al. (1978). Cells with active dehydrogenase enzymes (i.e., actively respiring cells) reduce the compound 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyltetrazolium chloride (INT) to an insoluble opaque formazan crystal, which is internally deposited in the cell. Cells were diluted using 0.02 M·phosphate-0.11 M saline buffer (pH 6.80). INT was added to the final dilution tube together with glucose (0.25g1- 1,

in 1 ml Evans mineral solution), which served as an electron source. Total cell counts were determined using an acridine orange stain, by counting green and orange fluorescing cells with a Zeiss SM-lux epifluoroescence microscope. Transmitted light was used for enumerating INT-formazan containing cells.

-44-

170 C.A. MASON, J.D. BRYERS AND G. HAMER

Non-respiring cells are determined by difference between the total epi- and INT-cell counts.

RESULTS

Figures 2 to 5 show the results of both the computer predictions and the experiments. The predictions are represented by the Jines and the experimental data as points. The kinetic constants used for the model predictions are listed in Table IV. It should be noted that total biomass comprises viable, dead and non-viable cells and biopolymeric particulates released from cell lysis. The predicted relative proportions of each to the total biomass is shown in Figure 3. The prediction for active cells (Fig. 4) is calculated from the proportion of viable and non-viable respiring cell mass to the total cell biomass. Total biomass agrees well with experimental results as does cell activity. However, no experimental data were available for comparison of particulate lysis prOduct concentrations. Residual glucose concentrations were below the detection limit (2mg1-1

) of the method used. The curves were not extended to include washout because of the interplay of additional factors, i.e., wall growth (Bryers, 1984) at such dilution rates.

3.0~------------------.

2.5

' ;;, 2.0

~ 1.5 <t: ~

0 1.0 O'.l

-' <t:

6 0.5 I-

EXPERIMENT

PREDICTION

0

0 ...... --...---...---~--~--...----1 0 0.2 0.4 0.6

DILUTION RATE 0.8

W') 1.0 1.2

FIGURE 2 Steady state total biomass conccn1ra1ioo as a function of dilution rate.

3.0

2.5

{:' 2.0

"' 1.5

ll) <.()

< :::E 1.0 0 OJ

0.5

-45-

MICROBIAL GROWTH

/ /

I I

I I i i I

0.2

..... //

-·-·--·- -·-·-

VIABLE CELLS NON-VIABLE RESPIRING CELLS DEAD CELLS POLYMERIC PARTICULATES TOTAL BIOMASS

0.4 0.6 0.8 DILUTION RATE (h"1J

1.0

171

1.2

FIGURE 3 Predicted concentrations of viable cells, non-viable respiring cells, dead cells, polymeric particulates and total biomas.' as a function of dilution rate.

100 oo 0 -0.- 0 0 0 0

80 0 EXPERIMENT

-- PREDICTION

>- 60 I-> I-u < 40 I-z: w (.) a: w 20 a.

0 0 0.2 0.4 0.6 0.8 1.0 1.2

DILUTION RATE (h"') FIGURE 4 Percent active Klebsiella pneum,miae in steady state continuous culture as a function of dilution rate.

-46-

172 C.A. MASON, J.D. BRYERS AND G. HAMER

JOO-.------------------. u 0: ~-250 '.:::i-0!... a. "" gj .!:. 200 z 0 CD 0:: 6 150

0

PREDICTED POLYMERIC ------- PARTICULATES

--- PREDICTED } DISSOLVED o EXPERIMENTAL g~~~~~

0

0.2 0.4 0.6 0.8 1.0 DILUTION RATE W1J

1.2

FIGURE 5 Steady state concentrations of dissolved organic carbon (S,) and polymeric particulate producrn (P) as a function o( dilution rate.

DISCUSSION

Endogenous metabolism has been considered to be a significant factor affecting the microbial biomass concentration and hence, the overall biomass yield coefficient, particularly in continuous culture systems operating at low growth rates. Traditionally, microbiologists have explained endogenous metabolism on the basis of the maintenance functions of microbes, i.e., turnover of cell materials, osmotic work to maintain concentration gradients between the cell and its exterior and cell mobility. Strictly, endogenous metabolism embraces several additional phenomena including death, lysis and "cryptic growth" (Hamer, 1985).

Very few microbiological studies have sought to differentiate between tradi-tional maintenance energy requirements and energy dissipation by the other phenomena. All of the contributory components of endogenous metabolism impact on the biomass yield coefficient. Biomass yield coefficients depend on the nature of the limiting substrate and on environmental conditions and hence, are not constant. The maximum theoretical yield coefficient for aerobic growth on glucose lies between 0.69 and 0.73 (Stouthamer, 1979) depending on the number of phosphorylation sites. Relatively few experimental studies with K. pneumoniae indicate that such a biomass yield coefficient is obtainable. Paca and Gregr (1977) report achieving biomass yield coefficients of 0.65 at low growth rates, but most

-47-

MICROBIAL GROWTH 173

TABLE IV

Kinetic constants used for computer prediction

Parameter Value Units

Y,,. 0.597 Y,,4 0.25 µj l.22 (h-') K,, 8 <msr'> µ~ 0.05 (h' ) K,, 200 (mgl-1) k, O.OJ (h-') k. 0.04 (h-') k, 0.04 (h-') ko 0.05 (h-') k7 0.8 (h-') k. 0.05 (h-') k. 0.1 (h-') vro 0.J (h-') KM, 8 (mgl- 1)

vr1 0.05 (h-') K.,, 200 (mg 1- 1)

Y10.s 0.1 Y11.s 0.25 Y•.z 0.65 Y •. 1 0.35 Y,,, 0.65 y7,7 0.35 Y,,, 0.65 Y.,, 0.35 Y•.2 I S'i 4800 mg1-•

of the authors report values between 0.43 and 0.54 (Sinclair and Topiwala, 1970, Topiwala and Sinclair, 1971, Harrison and Loveless, 1971, Neijssel and Tempest, 1975). Many reponed experimental biomass yield coefficients fail to consider endogenous metabolism which results in significant reductions in observed, but not necessarily true yield.

Because of endogenous activities, the observed biomass yield coefficients will vary with dilution rate. In view of the fact that the endogenous metabolism concept incorporates other physiological processes, it is proposed that a major contribution results from death and lysis and it is hypothesized that the true maintenance requirement, i.e., utilization of cellular material for energy, is small relative to cell breakdown and death.

Hinshelwood's (1951) theory of cell ageing and death includes the hypothesis that growth involves balanced lysis and synthesis of major structural polymers which results in a net removal from the cell. Sometimes, lysis of the whole cell results when the control over this process is lost. Either way, particulate and/or soluble organic matter is released by the cell into the culture medium.

Whether the products of intracellular lysis remain in the cell to be further hydrolysed and reused (yielding energy or being reincorporated) or whether they

-48-

174 C.A. MASON, J.D. BRYERS AND G. HAMER

are extruded from the cell is not known. However, since new cell wall material is laid down on its inner surface after transport through the cell membrane, it is hypothesized that the older material is lost entirely from the cell.

Whole cell lysis releases products including hydrolyzing enzymes. Conse-quently, much of the polymeric fraction is further broken down into low molecular weight substances which are more readily utilised by the cell. In monocultures this will be a minor activity, but takes on major importance in mixed cultures (Hamer, 1985). The model presented here differentiates between living viable, living non-reproducing, and dead cells as was suggested by van Uden and Madeira-Lopes (1976). It is not known whether death precedes lysis or whether lysis of viable cells (and non-reproducing cells) is possible before death occurs. This work suggests that dead cells do not remain intact for very long, a result supporting an assumption by Drozd et al. (1978) in their consideration of lysis and "cryptic" growth. Physiologically, this makes sense, since a dead cell must be a cell devoid of all metabolic activity. Such cells are difficult to imagine and a more likely scenario is that cell lysis and cell death are synonymous such that the cell dies as a result of lysis. Thus, the concept of dead cells may well be false.

In many cases where viability is considered, anything less than 100 percent viability suggests the presence of cells with genetic defects that are irreversibly non-viable. Under normal conditions (e.g., in water samples) the probability of this is low and any large deviation from 100 percent viability suggests the presence of reversibly non-viable cells which can further propagate given favourable environments.

The biomass of truly non-viable respiring cells in some systems may be small relative to the total biomass. The mass balance assumes that these cells can increase in mass by growth linked enzymatic oxidation. However, after a certain time period they will lyse. Microscopic examination of K. pneumoniae cultures revealed the occasional presence of elongated cells which may be representative of this category. Experimentally, the non-viable and viable respiring cells were not distinguishable. However, on the assumption that non-viable respiring cells only account for a very minor fraction in this system it is considered that measuring cell activity provides a far more accurate a'nd meaningful parameter than using propagation methods. The question must also be raised as to the applicability of using agar propagation methods especially in systems with very long residence times.

Extension of the model to very long residence times suggests that the proportion of active cells still remains >85 percent at dilution rates as low as 0.002 h- 1• This contradicts the findings of Tempest et al. (1967) who report low viabilities (40 percent) at dilution rates of 0.004 h- 1 (residence time of 250 h). Such a result was obtained using a slide culture technique and doubt must be expressed concerning the possibility for individual very slow growing, effectively nutrient starved cells to replicate on solid medium in the short time span over which the test was carried out. More work is obviously needed to confirm this prediction.

The kinetic constants used to evaluate the model were selected, where possible,

-49-

MICROBIAL GROWTH 175

from the literature. Experimental verification of the assumed constants and rate expression (first order, unless growth is involved, when Monod kinetics were used) must follow and will probably account for the discrepancies between prediction and experimental results.

CONCLUSIONS

The structured type of model that has been developed to describe simultaneous growth, death and lysis in chemostat cultures provides an effective framework for the development of a more detailed understanding of the relative importance of cellular death phenomena. Although the experimental data generated so far are insufficient for any major conclusions of a general nature to be drawn, they do support for the concepts proposed by Drozd et al. (1978). Clearly, a great deal still remains to be investigated with respect to microbial death and lysis in non-hostile environments.

ACKNOWLEDGEMENTS

We wish to express our thanks to Dr. W. Gujer for his help in computerizing the model and to the Swiss National Programme 7D for financial support.

REFERENCES

Beyers, J.D., "Biofilm Formation and Chcmostat Dynamics: Pure and Mixed Culture Considera-tions", Biotechnol. Bioeng. 26, 948-958 (1984).

Costerton, J.W. and Colwell, R.R. (eds). Native Aquatic Bacteria: Enumeration, Ac1iui1y and Ecology, A.S.T.M., Philadelphia (1979).

Drozd, J.W., Linton, J.D., Downs, J. and Stephenson, R.J., "An in situ Assessment of the Specific Lysis Rate in Continuous Cultures of Methylocoecus sp. (NCIB 1!083) Grown on Methane", FEMS Microbiol. Lellers 4, 311-314 (1978).

Egli, Th., Lindley, N.D., and Quayle, J.R., "Regulation of Enzyme Synthesis and Variation of Residual Methanol Concentration During Carbon Limited Growth of Kloekera sp. 2201 on Mixtures of Methanol and Glucose", J. Gen. Microbiol. 129, 1269-1281 (1983).

Evans, C.G.T., Herbert, D. and Tempest, D.W., The Com/nuous C11/tiua1ion of Micro-organisms. 2 Construction of a Chemo.<tal. In: Methods in Microbiology, Vol. 2, J.R. Norris, D.W. Ribbons, eds., pp. 277-327, Academic Press, London (1970).

Grady, C.P.L. and Roper, R.E., "A Model for the Biooxidation Process which Incorporates the Viability Concept", Water Res. 8, 471-483 (1974).

Hamer, G., "Lysis and "Cryptic" Growth in Wastewater and Sludge Treatment Processes", Acta Biotechno/. 5, 117-125 (1985).

Harrison, D.E.F. and Loveless, J.E., "The Effect of Growth Conditions on Respiratory Activity and Growth Efficiency in Facultative Anaerobes Grown in Chcmostat Culture", J. Gen. Microbial. 68, 35-43 (1971).

Hinshelwood, C., "Decline and Death of Bacterial Populations", Nature 167, 666-669 (1951). Humphrey, A.E., and Nickerson, J.T.R., "Testing Thermal Death Data for Significant Nonlogarith-

mic Behavior", Appl. Microbiol. 9, 282-286 (1961). Irvine, R.L., and Bryers, J.D., Stoichiometry and Kinetics. In: Comprehensive biotechnology Vol. 4.

M. Moo-Young, J.A. Howell (eds). Pergamon Press, Oxford (in press). Neijssel, O.M. and Tempest, D.W., "The Regulation of Carbohydrate Metabolism in Klebsiel/a

Aerogenes NCTC 418 Organisms, Growing in Continuous Culture", Arch. Microbial. Ul6, 251-258 (1975).

-so-

176 C.A. MASON, J.D. BRYERS AND G. HAMER

Paca, J, and Grcgr, V., "Growth and Metabolic Activity of Klebsielfa Aerogenes in a Chemostal Culture", J. Fermf'nt. Technol. SS, 213-223 (1977).

Postgate, J.R., "Viability Measurements and the Survival of Microbes under Minimum Stress", Adv. Microbial. Phy.<iol. I, 2-23 (1967).

Roels, J.A .. "Simple Model for the Energetics o( Growth on Substrates with Different Degrees ol Reduction", lliotecluwl. llioeng. 22, 33-53 ( 1980).

Sinclair, C.G. and Topiwala, H.H., "Model for Cominuous Culture which Considers the Viability Con~pt", lliotedmol. Bioe11g. 12, 1069-1079 (1970).

Stouthamcr. A.H .. The Search for Correlation between Theoretical and Exp<'rimental Growth Yields, International Review of Biochemistry, Microbial Biochemistry, Vol. 21 J.R. Quale (ed). University Park Press, Baltimore (1979).

Tempest, D. W., Herbert, D. and Phipps, P.J., Studies on the Growth of Aerobacter Aerogenes at Low Dilution Rates in a Chemostal. In: Microbial Physiology and Continuous Culture, E.O. Powell, C.G.T. Evans, R.E. Strange, D.W. Tempest (eds). H.M.S.O., London (1967).

Topiwala, H. and Sinclair, C.G., Temperature Relationships in Continuous Culture", Biotechnol. Bioeng. 13, 795-813 (1971).

van Udcn, N. and Madeira-Lopes, A .. "Yield and Maintenance Relations of Yeast Growth in the Chemostat at Superoptimal Temperatures", Biotech110/. Bioeng. 18, 719-804 (1976).

Zimmermann, R., lturriaga, R. and Becker-Birck, J., "Simultaneous Determination of the Total Number of Aquatic Bacteria and the Number thereof Involved in Respiration", Appl. Environ. Microbial. 36, 926-935 (1978).

-51-

CHAPTER 4

"CRYPTIC" GROWTH IN KLEBSIELLA PNEUMONIAE

Summary

The ability of Klebsiella pneumoniae to grow on its own soluble lysis products is shown in a series of batch growth experiments. Maximum specific growth rate coefficients ranging from 0.69 to 1.46 h-l were obtained with experi-mental "cryptic" yield coefficients ranging from 0. 42 to 0.52 (mg-cell-C/mg-substrate-C). These kinetic data are used to calibrate a model which demonstrates that the depression of theoreti.cal maximum yield coefficents to experimentally obtained values can be explained by "cryptic" growth pheno-mena without the need to resort to the use of physiolo-gically undefined, mathematical constants. Growth of K. pneumoniae on sonicated cells derived from steady-state chemostat cultures was followed in batch culture and obser-ved to occur with no lag phase. Batch growth curves did not indicate either diauxic or polyauxic growth, suggesting simultaneous utilization of the complex organic substrate mixture. These data suggest that "cryptic" growth is probably a real event occurring in growing chemostat cultures under ideal growth conditions and most probably also under starvation conditions.

-52-

Introduction

Increasing awareness of the importance of lysis and death in technical

and environmental processes such as in wastewater treatment and in

routine environmental monitoring programmes (Mason !!_ al., 1986a) has

stimulated interest in "cryptic" growth phenomena. The term "cryptic"

growth was introduced by Ryan (1959) and describes the re-utilization of

lysis and leakage products by intact cells of the same population as

carbon ene.rgy substrates and nutrient sources. The phenomenon was also

recognised in cultures of Escherichia coli by Koch (1959). I\ recent

model describing growth of Klebsiella pneumoniae in continuous culture

(Mason !!_ al., 1986b) shows that the processes of lysis and "cryptic"

growth could account for the observed depression of maximum biomass

yield coefficients normally attributed to maintenance energy

requirements (Pirt, 1965).

Whilst "cryptic" growth in monospecies cultures can be economically

important in many biotechnological production processes, the efficiency

and relevance can be much greater in mixed culture systems such as in

wastewater treatment processes (Hamer,19851 Hamer et!'!.!_, 1985) and other

processes where biomass concentration subjects the organisms to extended

periods of starvation within a recycle loop. Furthermore, in activated

sludge processes, recycle frequently involves a change from aerobic to

anaerobic conditions, during which lysis is induced in part of the

population enabling growth and metabolism by the remainder of the

population on the released products. Similarly, during the extended

operation of immobilised cell systems, an initial gradual loss followed

by a precipitous drop in performance occurs; the latter probably as a

result of death/lysis phenomena.

Previous investigations of "cryptic" growth have tended to focus on the

ability of cells to grow on heat killed suspensions of the same strain

(Nioh and Furusaka, 1968). This approach is so unlike the situation

likely to prevail in either growing or in starving cultures that the

conclusions derived from it cannot be generally applied. Therefore, it

is necessary to examine the process of "cryptic" growth using more

strictly defined conditions, closer akin to the processes occurring in

-53-

real culture environments. The purpose of this communication is to

de1110nstrate some of the fundamental principles of "cryptic" growth and

provide some kinetic data.

Theoretical

The processes involved in the growth of a microbial culture can be

described by the scheme shown in Fig. 1. A soluble carbon energy

substrate is used for the production of microbial biomass, soluble

products and C02 • Processes such as maintenance energy detract from

product formation via intra-cellular carbon recycling. However, energy

dissipation can also occur via an external cycle involving lysis of the

microbes followed by "cryptic" growth on the soluble lytic products,

thereby reducing the potential for both product and microbe formation.

This can be described in mathematical terms by establishing equations

for each of the various components present in the culture. For reasons

discussed by Mason !.!:_ al. (1986b), dead cells, Le., microbes totally

devoid of enzymic activity, will be assumed to be absent in continuous

cultures. The rate of accumulation of a component in a continuous flow

culture can be described by the generalised mass balance expression:

net rate of net rate of net rate of + (1)

accumulation transport transformation

Subsequent material balances are expressed in matrix notation as applied

by Roels (1980) for microbial processes.

The vector containing the components of interest, namely S, soluble

carbon energy substrate in feed, SL soluble organic lysis product, X,

the bacterial biomass, P, biopolymeric particulates from lysis and o2 ,

oxygen is defined as:

-54-

s

a • (2)

The transport vector will take the form

D(S -S) 0

-DSL a* -DK (3)

-DP

Ke(02*+02)-002

where D is the dilution rate (equivalent to the flow rate per unit

reactor volume), S0

the soluble carbon energy substrate concentration in

the feed, KL the overall liquid side oxygen transfer coefficient, ex the

specific gas-liquid interfacial area per unit

equilibrium saturation concentration of oxygen.

The· most concise means for representation

volume and

of the

0 * 2 the

reactions

(transformations) occurring in the culture is by means of a

stoichiometric matrix:

Component

Growth

cryptic

Growth

Lysis

Hydrolysis

s

-1

y

SL x

+l

-1 +l

Y*

f L -1

+l

-55-

l? 02

Reaction rate

-1-Y )lX y

-1-Y* p*X Y*

fl? K1X

-1 KhP

where Y is the theoretical maximum biomass growth yield on s, Y*, the

theoretical maximum biomass growth yield on SL, fL and f p' the fraction of dissolved and particulate lysis product, p, the specific

growth rate constant for growth on s, p*, the specific growth rate

constant for growth on SL, K and K * the saturation constants s s corresponding to S and SL, respectively, Kh, the hydrolysis rate

constant and K1 , the lysis rate constant.

Combining rate expressions and stoichiometry gives the transformation x vector a :

:1!!. y

-p*X + fLKlX + ~p Y*

It px + p*X - K1X (4) a

fpK1X-~P

·..:::: (JlX) - .};::!_* (p*X) y Y*

-56-

Thus, equation 1 can be rewritten:

da

a* + a"' (5)

dt

Complete material balances are given in table 1.

Methods

Organism. Klebsiella pneumoniae NClB 418 was maintained by monthly

sub-culture on plate count agar slopes incubated at 35°c and stored at

4°c.

Cell cultivation. K. pneumoniae was grown in a 2.5 1 bioreactor (MBR

Bioreactor AG, Wetzikon, CH) in a mineral salts medium modified from the

recipe of Evans .!!,! al. (1974), such that it contained EDTA (100 mg 1-l)

in place of citrate and used only 50% of the concentration of trace

elements. It was also necessary to add polypropylene glycol (20 mg 1-l)

as an antifoaming agent. Glucose (4.8 g 1-1) was the sole carbon energy

source limiting growth. The medium was added by means of a peristaltic

pump and the flow rate monitored by weight displacement from the

substrate resevoir bottle. Spent broth was removed, and the volume

maintained by an overflow weir. The cultivation temperature was

maintained constant at 35.o0 c and the pH at 6.80 by controlled addition

of a mixture of 1.5 M KOH and 1.5 M NaOH. At least five retention times

were allowed for attainment of a steady state when process conditions

were altered. Dissolved oxygen concentration was measured using an

oxygen electrode (Ingold AG, Urdorf, CHJ and was maintained at> 30%

saturation by varying the oxygen/air ratio in a constant volume of gas

sparged to the bioreactor. Verification of monoculture operation was

carried out by periodic microscopic observation and by cell cultivation

on plate count agar followed by identification of selected single

colonies using the API 20E test kit (API System S.A., Montalieu-Vercieu,

F).

-57-

Preparation of "cryptic" growth medium: Approximately 200 ml of a steady

state culture was withdrawn from the bioreactor into pre-cooled flasks

immersed in ice. The cells were lysed/broken by sonicating 100 ml

volumes in an ice bath for 10 minutes at an amplitude of 28 microns

using a MSE Soniprep 150 Ultrasonicator (MSE, Crawley, GB). The soluble

lysis products were separated by centrifugation at 4°c at 30,000 x g for

10 minutes and pre-filtered under pressure through 0. 2 )Im cellulose

nitrate filters (Sartorius GmbH, Gottingen, D). Phosphate buffer (final

concentration 0.02 M) was added so that the pH was 6.80. The batch

growth experiments were carried out in 250 ml, baffled Erlenmeyer

flasks. These were pre-sterilized and the medium added in a laminar flow

cabinet by sterile filtration through 0.2 )Im Zetapor filters (Zeta,

Meriden, Connecticut, USA).

Batch Growth Experiments: Each growth experiment was carried out in

duplicate with duplicate uninoculated controls. 100 )11 of a steady state

culture from the same chemostat from which the substrate microbes were

derived and growing at the same dilution rate, was used as inoculum. The

cultures were incubated in a water bath with shaking at 100 rpm at 35°c.

Samples {lml) were taken at regular intervals and the optical density

measured at 540 nm using a Uvikon 860 split beam spectrophotometer

(Kontron AG, Zilrich, CH) against the control.

Organic Carbon Determination: Total organic carbon was measured using a

Tocor 2, Total Organic Carbon analyser (Maihak AG, Hamburg, D). The

samples were suitably diluted using double distilled water, acidified

using concentrated HCl and the excess co2 removed by sparging with N2 for 12 minutes.

Dissolved organic carbon was determined using the filtrate obtained

after passing the samples through 0.2 pm Nuclepore filters. The

filtrates were suitably diluted, acidified and stripped of co2 as

described above before measuring in the Total Organic Carbon analyser.

Chemicals: All chemicals were of analytical grade and manufactured by

either Fluka AG (Buchs/SG, CH) or Merck AG (Darmstadt, Dl .•

-58-

"Cryptic" growth - batch studies: - The results of the experiments

designed to determine whether "cryptic" growth occurs in cultures of K.

pneumoniae are shown in Fig. 1. Batch growth studies were carried out on

sonicated cell suspensions using cells derived from cultures growing at

a range of different dilution (growth) rates. Typical "batch" growth

curves resulted in all instances, but with either an extremely short or

an undetectable lag phase. During the course of the experiment, a

significant amount of precipitation occurred in the culture flasks (and

in the uninoculated controls). This was most pronounced in the

experiments with cells derived from low growth rate continuous cultures.

The precipitate which contained 8-12 % protein was less evident in the

experiments derived from fast growing continuous cultures and was

markedly reduced by the end of the batch growth cycle (as judged

optically). In all cases, the exponential growth phase ceased after 5 h

and was followed by an essentially stationary phase which lasted up to

24 h (data not shown), during which a very slight increase in optical

density occurred.

Kinetic data: - The maximum specific growth rate constants measured for

the growth of ~· pneumoniae on its own lytic products in this series of

experiments are shown in Table 2. The values are all high, 0.69-1.46

h-l, reflecting very fast growth in this complex mixed substrate. -1 Similar batch experiments with !· pneumoniae grown in 2 g 1 glucose in

Evans mineral salts medium, resulted in a maximum specific growth rate

constant of 1.06 h-l with exponential growth also being complete after

approximately 5 h.

Table 3 shows the amounts of total organic carbon (TOC) and dissolved

organic carbon (DOC) released from ~· pneumoniae following sonication

and filtration, and the remaining concentration at the end of the

experiments, as a function of the original growth rate. In each case,

the final DOC was taken after 24 h batch growth.

-59-

The amount of carbon released from the cells by the sonication procedure

is dependent on the growth rate and increases with increasing growth

rate. Similarly, the amount of dissolved organic carbon remaining at the

end of the expel!"imental period is also higher at the higher growth

rates, suggesting that the amount of either non-biodegradable or slowly

biodegradable organic material released from the cells is a function of

growth rate. However, the proportion of dissolved organic carbon

remaining of the total initial quantity decreases with increasing growth

rate from 51% for "cryptic" growth using cells grown at O.l h-1 , to only

37% for cells grown at 0.8 h-1•

The values for total organic carbon in the cultures after batch growth

include both that for cellular carbon and the fraction entrapped in the

precipitate. In order to calculate yield values, the amount of

particulate organic carbon was determined in the control flasks and

deducted from the TOC present at inoculation. The yield coefficients

obtained (mg-cell-C per mg-C in original cell lysate) are shown in Table

4. These values vary from 0.41 to 0.52 and appear to be dilution rate

independent.

Discussion

Growth of microbial cultures has been investigated by microbiologists

for 100 years. During this time, the concepts of growth in continuous

and batch culture have been developed and applied in many technical

fields. Both negative growth and growth inhibitory effects have also

been considered, but remain less well understood and questionable

concepts involving such terms as maintenance energy have been developed.

The discrepancy between the fundamentally derived biomass yield

coefficients, i.e., the amount of biomass expected to be produced by

calculation (Stouthamer, 1979) and the observed values for biomass yield

coefficients have been explained in various ways under the term

endogenous metabolism in which the maintenance energy contribution is

just one factor. The processes of death and lysis and potential

"cryptic" growth have been largely ignored despite the fact that such

processes obviously can play a role in yield coefficient reduction.

-60-

The series of "cryptic" growth experiments presented here suggests, on

the basis of a negligable lag phase, that the cells are already in

possession of the enzymes necessary for utilization of the "cryptic"

growth substrates and thus the process is likely to be a real one

occurring within growing cultures of microbes that exhibit a broad

spectrum of carbon energy substrate utilization even under optimum

growth conditions.

Calibration of the proposed mathematical model is possible using the

kinetic data derived from the experiments. The factor fL, the fraction

of soluble lysis products appears, on the basis of the DOC and TOC data,

to be dilution rate dependent. Equations 11 and 12 were thus used to

describe the values for fL and fp.

f p

(11)

(12)

Lysis was also the subject of an investigation of Drozd ~ al. (1978) in

cultures of a Methylococcus sp. The advantage of using this bacterium

for such experiments is that, due to the obligate methanotrophic nature

of its metabolism, "cryptic" growth on the lysis products is not

possible, allowing the lysis rate constant to be determined by measuring

the amount of DOC produced with respect to the growth rate for a range

of different growth rates. Drozd et al. (1978) found that the lysis rate

was dilution (growth) rate dependent such that:

y D + z (13)

Using this information together with appropriate values of 11• m' the

maximum specific "cryptic" growth rate and y•, the "cryptic" growth

yield, an X-D diagram can be predicted (Fig. 3). The form of this

diagram is typical of those obtained in experiments and shows that

microbial growth can be adequately described employing the concepts of

-61-

lysis and "cryptic" growth. Observed biomass yields in the model would

have had a maximum of 0.52 at D ~ 0.8 h-1 , when a theoretical maximum

value of 0.65 is used in the prediction. "Cryptic" growth is clearly not

the only procesa that reduces the theoretical maximum biomass yield

coefficient, but it offers a physiologically plausible explanation as to

the fate of biomass.

"Cryptic" growth has been considered spasmodically in the scientific

literature. Postgate and Hunter (1962) found a yield coefficient of 0.02

(i.e. 50 cells were required to lyse for the production of one new

cell). Similarly, Martin and Washington (1964) developed a highly

intricate model for bacterial growth in continuous culture where they

included lysis and "cryptic" growth, but assumed a yield of one new

organism for 500 "dead" ones. Hamer and Bryers (1985) used the

relationship developed by Linton and Stephenson (19"/8) between the heat

of combustion per g substrate carbon and the maximum biomass yield

coefficient to estimate the maximum biomass yield coefficient for

"cryptic" growth. They obtained a value of 0.66 g dry biomass per g dry

substrate biomass but also stated that deoptimisation should produce

experimental values as low as 0.33. The experimentally obtained values

of 0.42-0.52 ratify their estimate.

Much of the literature concerning "cryptic" growth phenomena has

appeared in the field of wastewater and waste sludge treatment where a

reduction in biomass is a desired effect. Such situations of mixed

culture "cryptic" growth are likely to be more efficient due to the fact

that many mixed cultures are better adapted to deal with complex mixed

substrates (Hamer !:! !_!., 1985). Gaudy !:! !.!· (1971) looked at "cryptic"

growth in an activated sludge system using similar methods to those used

here. Feeding sonicated activated sludge microbes back to the remaining

population they found extremely rapid growth on and utilisation of the

lysate.

-62-

The use of a complex mixed substrate for growth by !· pneumoniae brings

into focus the concepts of simultaneous nutrient utilization Such

phenomena have been described by Egli ~ al. (1983) for cultures of the

yeast Kloeckera sp.2201 growing on mixtures of glucose and methanol. As

to whether truely mixed substrate assimilation is occurring in the

reported "cryptic" growth experiments cannot be conclusively stated,

However, it is highly probable because of the short duration and

rapidity of growth. The large amount of non- or slowly- biodegradable

DOC found. at the end of the experiment can be accounted for by protein.

According to Schlegel (1981) K. pneumoniae is incapable of protein

degradation, and this is indeed true when substrates such as either

bovine serum albumin or casein serve as the sole substrate. However,

under stressed situations such as occur under c- or N-limitation,

protein derived form !· pneumoniae cells is degraded (U.Wanner,

unpublished results). This suggests that protein degradation is likely

to occur very slowly in the batch "cryptic" growth experiments but

probably much faster in the nutrient stressed chemostat environment.

The data are still inconclusive, since large particulate cell wall

matter was separated from the substrate before growth. Obayashi and

Gaudy (1973) showed that the capsular material from !· pneumoniae, as

well as from Arthrobacter viscosum, Azotobacter vinelandii, Xanthomonas

compestris and Zooglea ramigera, can serve as the sole carbon and energy

source for a mixed population derived from municipal wastewater

treatment, with yield coefficients ranging from 0.55 to 0.35 depending

on the source of the polysaccharide.

The consequences of "cryptic" growth phenomena are immense. Postgate and

Hunter (1963) showed that "cryptic" growth can account for the survival

pattern in cultures under starvation (non-growth) situations. They

presented data for extended survival of magnesium limited batch cultures

(stationary phase) despite extremely rapid initial "death" rates. It is

indeed ramarkable that in most survival experiments "cryptic" growth is

st i 11 ignored and omitted in the formulation of models to describe

growth and non-growth situations. Postgate (1976) stated that in

survival studies in which "cryptic" growth is possible, it should be

-63-

expected to be the rule rather than the exception. From the mathematical

model presented here it can be seen that the maintenance concept is

clearly too clumsy to deal with the intricate physiological changes

which occur in cultures expecially under either starvation or non-growth

stresses. Moreover, the application of the lysis/"cryptic" growth

concept is more realistic than the redefinition of maintenance as an

ill-defined multiphase phenomenon (Van Verseveld et.!!:·• 1984).

-64-

Ref er enc es

Drozd JW, Linton JD, Downs J, Stephenson RJ (1978) An in situ assessment of the specific lysis rate in continuous cultures-ofMethylococcus sp. (NCIB 11083) grown on methane. FEMS Microbiol Letters 4: 311-314

Egli Th, Lindley ND, Quayle JR (1983) Regulation of enzyme synthesis and variation of residual methanol concentration during carbon-limited growth of Kloeckera sp. 2201 on mixtures of methanol and glucose. J Gen Microbiol 129:1269-1281

Evans CGT, Herbert D, Tempest DW (1974) The continuous cultivation of Micro-organisms. 2: construction of a chemostat. In: Norris JR, Ribbons DW (eds) Methods in Microbiology, vol. 2, Academic Press, London, pp 277-327

Gaudy AF, Yang PY, Obayashi AW (1971) Studies on the total oxidation of activated sludge with and without hydrolytic pretreatment. J Water Poll Contr Fed 43: 40-54

Hamer G (1985) Lysis and "cryptic" growth in wastewater and sludge treatment processes. Acta Biotechnol 2: 117-127

Hamer G, Egli Th, Mechsner Kl (1985) Biological treatment of industrial wastewater: a microbiological basis for process performance. J Appl Bacteriol Syrop Supp 127S-140S

Hamer G, Bryers JD (1985) Aerobic thermophilic sludge treatment - some biotechnological conepts. Conserv Recycl 8: 267-284

Koch AL (1959) Death of bacteria in growing culture. J Bacteriol 77: 623-627

Linton JD, Stephenson RJ (1978) A preliminary study on growth yields in relation to the carbon and energy content of various organic growth substrates. FEMS Microbiol Letters 3: 95-98

Martin EJ, Washington DR (1964) Kinetics of the steady-state bacterial culture. 1. Mathematical model. Proc. Purdue Industrial Waste Conference, Lafayette, Indiana, part 2, 724-737.

Mason CA, Hamer G, Bryers JD (1986a) The death and lysis of microorganisms in environmental processes. FEMS Microbiology Reviews 39:373-401

Mason CA, Bryers JD, Hamer G (1986b) Activity, death and lysis during microbial growth in a chemostat. Chem Engng Commun 45:163-176

Nioh I, Furusaka c (1968) Growth of bacteria in the heat-killed suspensions of the same bacteria. J Gen Appl Microbiol 14: 373-385

Obayashi AW, Gaudy AF (1973) Aerobic digestion of extracellular microbial polysaccharides. J Water Poll Contr Fed 45: 1584-1594

-65-

Pirt SJ (1965) The maintenance energy of bacteria in growing cultures. Proc Roy Soc Lon Ser B 163: 224-231.

Postgate JR, Hunter JR (1962) The survival of starved bacteria. J Gen Microbiol 29: 233-263

Postgate JR, Hunter JR (1963) The survival of starved bacteria. J Appl Bact 26: 295-306

Postgate JR (1976) Death in macrobes and microbes. In: Gray TRG, Postgate JR (eds) The survival of vegetative microbes. Proc Symp Soc Gen Microbiol vol 26, Cambridge University Press, Cambridge, pp 1-18.

Roels JA (1980) Simple model for the energetics of growth on substrates with different degrees of reduction. Biotechnol Bioeng 22: 33-53

Ryan FJ (1959) Bacterial mutation in a stationary phase and the question of cell turnover. J Gen Microbiol 21 530-549

Schlegel HG (1981) Allgemeine Mikrobi.ologie (5th Edn) George Thieme-Verlag, Stuttgart.

Stouthamer AH (1979) The search for correlation between theoretical and experimental growth yield. Int Rev Biochem 21: 1-45

Van Verseveld HW, Chesbro WR, Braster M, Stouthamer AH (1984) Eubacteria have 3 growth modes keyed to nutrient flow - consequences for the concept of maintenance and maximal growth yield. Arch Microbiol 137: 176-184

-66-

Table 1. Material balances for microbial growth in a chemostat

incorporating the concept of "cryptic" growth.

Para-meter

s

x

p

Net Rate of

Accumulation

dS/dt

dX/dt

dP/dt

Net rate of

Transport

-DX

-DP

Net Rate of

Transformation

-µX y-

Table 2. "Cryptic" growth of !• pneumoniae - batch growth rates.

Cells derived from

culture with dilution

rate (growth rate)

0.10

0.195 0.30

0.50 0.70

0.81

Maximum specific growth

rate constant (pm) in batch simulated "cryptic" growth study

0.69

0.77

0.85 1.10

1. 46

1. 22

( 6)

( 7)

(8)

( 9)

( 10}

-67-

Table 3. Total organic carbon released by sonication of !· pneumoniae and residual total and dissolved concentrations after batch "cryptic" growth

Cells derived Total organic Total organic Dissolved organic Dissolved organic fran culture carbon relea- carbon in the carbon in the carbon at the end with dilution sed by the culture after culture after of the experiment rate (growth cells after batch growth batch growth as percent of initi-rate) sonication al concentration

(h-1) (m::J 1-1) (m::J 1-1) (m::J 1-1) (%)

0.10 241 175 123 51.04 0.195 350 285 180 48.6 0.30 460 348 207 45 0.50 471 375 201 42.7 0.70 650 486 303 46.6 0.81 690 501 258 37.4

Table 4. "cryptic" growth yields for K. pneumoniae in batch culture.

Cells derived from chemostat with dilution rate (growth rate)

0.1 0.195 0.30 0.50 0.10 0.81

Experimental yield (cell carbon produced per carbon used in lysate)

0.44 0.42 0.48 0.41 0.42 0.52

-68-

SOLUBLE CARBON ENERGY SUBSTRATE

. .... PRODUCTS

' .. . . CARBON DIOXIDE

MAINTENANCE ~

" (Growth rate ...

independent?)

t ... ~ VIABLE ANO .

NON - VIABLE , CRY PT IC , . MI C RO· B ES <ri " :0

0 :'IE -i :r:

LYSIS (Growth rate dependent ? l

~ SOLUBLE LYTIC . PRODUCTS

·~ ... •r

DEAD CELLS? f---* PARTICULATE RESIDUES

Figure 1 Carbon and energy fluxes for heterotrophic growth and product production from soluble carbon energy substrates.

-69-

1.0 a) E 1.0 b) E c: 0.6 c 0.6 0

0 0.4 "'" 0.4 "'" LO LO 0.2 0.2 z z ~ 0.1 0 0.1 ~ 0.06 0.. 0.06 ~ 0.04 ~ 0.04 gj 0.02 ~ 0.02 <( <(

0.01 0.01 23456789 0 23456789 0

TIME ( h) TIME (h)

1.0 c) E 1.0 d) E 0.6

c: 0.6 c 0 0 0.4 "'" 0.4 "'" LO LO 0.2 0.2 z z

0.1 0 0.1 0 j:: ~0.06 0.. 0.06 ~0.04 ~ 0.04 ~ 0.02 ~ 0.02

<(

<( 0.01 0.01 23456789 0 23456789 0 1

Tl ME { h) TIME (h)

E 1.0 E 1.0 f) ~ 0.6 c 0.6 c: 0 0 0.4 "'" 0.4 "'" LO LO 0.2 z 0.2

z Q '0.1 0 0.1 j:: h: 0.06 0.. 0.06 ~ 0.04 0:: 0.04 0

~ 0.02 gj 0.02 <( <(

0.01 0.01 9 0 I 2 3456789 0 1 2 345678

Tl ME (h) Tl ME (h)

Figure 2 The growth potential of !· pneumoniae on its own lytic products,

after sonicating cells from a steady-state chemostat culture and

using sterile filtrate as growth substrate for the batch growth of

cells taken from the same chemostat culture after steady state

operation had been re-established. Various chemostat growth

(dilution) rates were used: a) 0.10 h-1 , b) 0,195 h-l, c) 0.30 h-1 ,

d) 0.50 h-1 , e) 0.70 h-1 , f) 0.81 h -l.

3.0

2.5

2.0

~ 1.5

-g 1.0 "' -

<fl_ 0.5 ><

-70-

x

I i js i

r------------ j ----------1'----------=-":::::::-·-::-:.· __ 0-1-~-.-~-.-~--.~~...---...-~...-~....--=::.::;::::.;___,..;;.:::__.,

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 DILUTIO.N RATE (h-1)

Figure 3 Computer prediction from the model for bacterial growth in continuous culture. SL, soluble lytic product concentration, never exceeded 3.7 mg 1-1 • Other symbols: X - biomass, P - particulate biomass and S -

soluble carbon energy -1

substrate. Kinetic constants: Y = 0.65 mg mg ,

-1 p*m = 1.0 h , ks a = 0.95, b = 0.2,

s = 4.8 g 1-1 • 0

Y*= 0.42 mg = 8 mg 1-1 ,

-1 -1 mg , µ = 1.2 h , * m -1 K = 20 mg 1 ,

s -1 Y = 0 o 4 I Z 0 • 0 2 I kh = 0 o 0 5 h t

-71-

CHAPTER 5

INTRA-CELLULAR MAINTENANCE REQUI.REMENTS AND THE

EXTRA-CELLULAR DEATH/LYSIS/"CRYPTIC" GROWTH CYCLE.

Introduction

Microbially mediated biotechnological processes are dependent on the activities of either growing microbes, in the case of biomass and growth associated product production, or non-growing microbes, which are potentially superior for many transformations. In both cases endogenous activity will detract from overall process yields and productivities. In processes involving growth, cell production will be reduced, whilst in processes employing non-growing cells an initial gradual decline followed by a precipitous decline in productivity occurs with extended operation.

The requirements for effective process cultures for commercial exploitation are:

(1) Maximum yield coefficients for the product or products; (2) Rapid growth rates and/or product production rates; (3) High affinity for carbon energy substrates; (4) High stability, robustness and lack of fastidiousness.

Many natural, mutant, and engineered strains fail to meet these criteria.

-72-

It has long been postulated that microbes require energy for both growth

and so called maintenance requirements. The latter is generally

expressed as a growth rate independent factor which depresses yield

coefficients and productivities.

Production Culture Physiology.

In all microbial production processes characterisation of the process

culture requires strict definition of its physiological status. The

consequences of inadequate characterisation are an inability to optimise

production processes for either biomass or products. Heterotrophic

processes can be represented by the scheme shown in Fig. 1., where

traditional maintenance requirements such as turnover of intra-cellular

materials, osmotic work to maintain transmembrane concentration

gradients and motility involving internal carbon energy fluxes are

differentiated from processes such as death/lysis/"cryptic" growth that

involve external carbon energy fluxes.

The death/lysis/"cryptic" growth cycle is an undesirable feature in all

commercial microbial processes and is a function of culture robustness.

Furthermore, unlike maintenance requirements which are traditionally

modelled as process rate independent, the death/lysis/"cryptic" growth

cycle is process rate dependent. In the case of those cultures with

broad spectrum capacity for complex substrate utilization, the effects

of the cycle are most marked at low growth rates, but where substrate

specificity is restricted, as with methanotrophs and obligate

methylotrophs, manifestation is as growth inhibtion.

"Cryptic• Growth in Klebsiella pneumoniae

The growth potential of Klebsiella pneumoniae NCIB 418 on its own lytic

products was investigated by sonicating cells from a steady-state

chemostat culture and using sterile filtrate as growth substrate for the

batch growth of cells taken from the same chemostat culture after

-73-

steady-state operation had been re-established. "Cryptic" growth was

shown to occur (Fig. 2), without lag and with a high specific growth

rate constant, 1.04 h-1 , which approaches the maximum value of 1.22 h-l

when glucose was used as sole carbon energy substrate. The amount and

type of dissolved organic carbon released from ~· pneumoniae, grown at

different dilution rates, following sonication is shown in Fig. 3.

Discussion

Reduction of the biomass yield coefficient in microbial growth

processes, concentration of active microbes in non-growing (immobilized)

systems and of product production in all microbial processes can be

largely accounted for using the death/lysis/"cryptic" growth concept

(1). This concept involves extra-cellular processes in contrast to the

traditional maintenance requirement concept which involves only

intra-cellular processes. In view of the fact that the former concept is

generally ignored, over-emphasis has frequently been given to the latter

concept, such that it has largely degenerated into a convenient

mathematical factor with little real physiological meaning, when, in

fact, both processes occur. Under optimized process conditions the

contribution of the extra-cellular process will probably markedly exceed

that of the intra-cellular process. The adoption of two different,

simultaneous process mechanisms offers a more plausible explanation of

the process physiology in all types of microbial production processes.

1) Mason,C.A., Bryers,J.D. and Hamer,G. (1986) Activity, death and lysis

during microbial growth in a chemostat. Chem. Engng. Comm. 45, 163-176.

-74-SOLUIU C.lftl<Htl-----------~ £frilRGY SUl$TRATI

PAOOUCT$ 1----

NAtNTeNANCI fGnwu ttlt

J ............ ,

'I !ABU: HD

LYStS ... ..,. ..... ,}

CARIOH OIOllOI

" " ~ .

OEAO CEl.l$1 PAITICUUT£ I----' Af'!IOUES

Fi:gure 2.

'

•cryptic• growth of !!· eneumoniae on it• own lysis products. natch growth was carried out using cells derived from a continuous cult\Jre (D • O.S h-ll at pH 6.80 at

Js0 c.

1.0.--...--..-----~--------0.8 0.6 0.5 0.4 0.3

0.2 Ms

540nm 0.1

0.08 0.06 0.05 0.04 0.03

0.02

o.oi o;--:--.,1;--*°3 --....4 --:---:---+--!a TIME lh .. ra)

Figure L Carbon and enerqy tluxee

...

... , ..

. ..

for beterotrophic qrowth .and product prOduction from eoluble carbon en~uqy aub•tratea.

.1 ·" .• .• OikJlion Rate (If')

~ ---le - r:e.leued fr<n !!· I!!!!!!!!!!!!!! aa a functian of dilutian rate. 'Die lar<JI! particulate fr-Ian !oell ...Uls etc.) - ,,.,,.._, bf contrifugatian. 'Die ol<lwly (or non-) biodagradl!lble fr-=tion ia

- bf the li9'ter aree end the - area ._.....,,. tile ooc .n!eh could be dogr-.S "1thin 24 h in "cryptic" 9f<lWtl\

""""""'""" ..

-75-

CHAPTER 6

SURVIVAL AND ACTIVITY OF KLEBSIELLA PNEUMONIAE

AT SUPER-OPTIMAL TEMPERATURES

Abstract

The effect of temperature on a population of Klebsiella pneumoniae was examined together with the imposition of mild starvation conditions at 35°, 41°, 49°, 55° and 6o0 c. Results for changes in biomass, protein and metabolic activity are presented in terms of gross population changes and show that these parameters decline with increasing temperature. Increases in the amount of dissolved organic carbon together with a decrease in the number of cells present with increasing temperature suggest that death and lysis processes are occurring. Regrowth in the bioreactor occurred after returning the temperature to 35°c and starting a flow of carbon and other nutrients. This was

probably due to reinoculation from head space wall growth drainage rather than reversion of heat-stressed microbes. The consequences of this for thermophilic sludge treatment processes are discussed. The concept of endogenous metabolism is questioned with respect to its being a realistic description of the survival process.

-76-

Introduction

The use of heat for rendering a population of microbes inert is commonly

practised. Sterilization processes in the public health sector, and

pasteurization processes in the food and beverage industries have been

used with considerable effect to achieve either reduction or total

inhibition of microbes and their activity. However, the use of heat

treatment has limited application in the wastewater and waste sludge

treatment \,industries. Introduction of pasteurization processes in the

treatment of sewage sludge have proved unsuccesful where inefficient or

insufficient biodegradation of the organic material in the sludge allows

the possibility of post- pasteurization reinfection either from

organisms in aerosols or from other sources in the vicinity of the

treated sludge.

At present, several novel treatment technologies are being assesssed for

their ability to achieve adequate removal of pathogenic organisms from

waste sludge during treatment, but at the same time achieve high levels

of biodegradation and stabilization during exposure to high

temperatures. Such processes operate in the thermophilic temperature

range (55-7o0 c) under either aerobic or anaerobic conditions. The target

of such treatment processes with respect to hygienisation aspects of

sludge are the bacteria, viruses, worm eggs and protozoa which can

potentially result in serious diseases in humans and animals.

Bacteria of enteric origin such as Escherichia coli and Klebsiella

pneumoniae have been used as indicators of more serious pathogens. Thus,

E.coli and .!5_.pneumoniae, which are both themselves potentially

pathogenic, are monitored together with other coliforms to indicate the

possible presence of more seriously pathogenic microbes such as

Salmonella spp. and ~ spp. . Bacteria are found that can grow or

survive over a wide range of temperatures. Although most are found only

at temperatures below 45°c, others can survive and even grow at

temperatures in excess of loo0 c (l). The pathogenic microbes are of the

-77-

former group and are "killed" or inactivated by temperatures above so0 c,

thus, the potential for heat treatment technologies in the treatment of

waste sludges.

Mow heat treatment affects microbes has been the subject of considerable

attention. In !'.·~· physical changes in the cell surface have been

described whereby a weakening of the structure of the peptidoglycan

results in extrusion of cell membrane-bound "blebs" following transfer

to temperatures in excess of so0 c (2). Pellon~ al. (3) and Pellon and

Gomez (4) found extensive damage to the DNA of !!.coli following a shift

of temperature from 37°c to so0 c, with concommitant loss of viability.

They showed single and double strand breaks occurred in the DNA together

with substantial physical association of protein to the DNA molecule,

the DNA scissions resulting from the action of endo- and exo-nuclease

activity (SJ. Returning the microbes to 37°c resulted in regrowth of the

population following a 40 min lag period during which time the DNA was

repaired (6).

The ability to survive heat shock has been shown to be dependent on the

growth rate (7,8), the nutritional status of the culture (9,10) and the

water activity in the bulk liquid (ll). l'iu and Klein (9) found that a

mixed bacterial culture isolated from water showed a decreasing

sensitivity to mild warming stress in relation to increasing time of

nutrient starvation, whilst the reverse was true of E.coli.

The production of !!.! ~ enzymes and other proteins as a result of heat

stress has also been described in considerable detail for a wide range

of organisms (12,13,14), although the function of these proteins has yet

to be elucidated.

In the aerobic thermophilic sludge biodegradation process, hygienisation

requires the destruction of the pathogenic microbes. It is known that

microbial solids removal can occur at a high rate using this process

(15,16,17) but the fate of pathogenic microbes in such environments is

largely unknown, especially when one examines only those effects caused

-79-

of single colonies using the API 20E test kit (API System S.A.,

Montalieu-vercieu, F).

Heat Treatment: The effect of temperature on !· pneumoniae was invest-

igated by continuosly feeding cells from the cultivation bioreactor into

a second, similar bioreactor maintained at various temperatures. The

hydraulic residence time in the second bioreactor was maintained

constant at 32 h (D = O.Ollh -l). The pH was maintained constant. The

second bioreactor received no nutrient flow. Before sampling, at least

five residence times were allowed so as to establish steady state

conditions. The reactor was aerated by sparging with air.

Analyses

Total Biomass Concentration: was measured as dry weight by a

filtration/gravimetric procedure using tared O. 4 pm Nuclepore filters.

The filters were dried at lOS0 c for lh before reweighing.

Dissolved Organic carbon was determined in the filtrate obtained from

dry weight determination after acidifying using concentrated HCl and removing the inorganic carbon by sparging with N2 for 12 min, in a TOCOR

2, total organic carbon analyser (Maihak AG, Hamburg, D).

~ in the cells and in the extracellular medium was measured after

centrifuging at 30,000 x g using the Biuret method with bovine serum

albumin as the standard as described by Herbert.!!:!!_. (20).

Metabolic Activity was measured using 2-(p-iodophenyll-3-(-

p-nitrophenyl)-5-phenyl tetrazolium chloride (INT). One ml INT (2.0g 1-l

aqueous solution) was added to a 10 ml sample immediately after removal

from the bioreactor after suitable dilution in a O.llM NaCl, 0.02M

phosphate buffer (pH 6.80). One hundred pl glucose in Evans mineral salts medium (glucose concentration 2 .Sgl -l) was also added to provide

an electron source. After 20 min incubation, 100 pl 37\ formaldehyde was

added to stop the reaction. The solution was centrifuged at 30,000 x g

-ao-

for 25 min and the solids resuspended in 10 ml tetrachloroethylene (40

v%) / acetone (60 v%) solution. After 30 min incubation at room

temperature in the dark, the solution was centrifuged at 30,000 x g for 10 min and the absorption of the extracted INT-formazan measured at

490nm.

Total Cell Number was assayed by means of the acridine orange direct

count method. Suitable dilutions of the microbes were made using O.llM

NaCl, 0.02~ phospahte buffer (pH 6.80). Five ml were filtered through pre-stained (1:15000 sudan black in 50% absolute ethanol) 0.2 pm

Nuclepore filters. One ml acridine orange solution (1:5000 in 6.6 mM

phosphate buffer, pH 6.80) was added to the filter for 2 min, and then

removed by filtration. Total cell counts were made by counting green and orange fluorescing cells with a Leitz SM-lux epifluorescence microscope

(Leitz GmbH, Wetzlar, D).

Chemicals: All chemicals were of analytical grade and supplied by either

Fluka (Buchs/SG, CH) or Merck (Darmstadt, D).

The effects of heat on a population of !· pneumoniae were examined

together with the effect of mild starvtion conditions by feeding a

steady state culture of the bacterium from a continuously fed bioreactor into a second bioreactor which received no additional nutrient flow, but

was maintained at pH 6.80 and aerated. The second bioreactor was maintained at each of a series of temperatures; 35°c (equivalent to the

growth temperature in the first bioreactor), 41°, 49°, 55° and 6o0 c.

Samples were removed from the second bioreactor once steady state

conditions had been established, and examined with respect to dry weight, dissolved organic carbon concentration, metabolic activity,

protein content, cell number and the accumulation of 260nm absorbing compounds in the culture medium. The results of these analyses are shown

in Figures. 1-3.

-81-

As can be seen from Figure 1., the dry weight decreases as a result of

only the starvation conditions, between 1.09 gl-l in the first

bioreactor and 0.99 gl -l in the bioreactor maintained at the same

temperature as the first (35°cJ but with no nutrient flow. Noticeable

also is the further decrease in dry weight as a function of increasing

temperature such that at 60°c only 72.4% of the concentration of

bacteria being fed into the bioreactor, on a dry weight basis, remained

in the bioreactor at its outflow.

Simultaneous with the decrease in dry weight, there was a decrease in

the number of cells present as determined by the acridine orange direct

count. This reduction from the number present in the first bioreactor

and the feed to the second bioreactor, occurred under all conditions of

starvation and heating. The extent of reduction was a function of

temperature.

Attempts to enumerate stressed microbes have tended to use techniques

requiring cell replication on an agar surface. This technique

drastically underestimates the true number of microbes present since

failure to reproduce can occur purely through the physico-chemical

characterisics of the agar medium, which differ drastically from

conditions in submerged continuous cultures. Moreover, a time limit is

usually enforced on such tests, thus, further negating the application

of this technique to stress studies where a period of adaptation is to

be expected before growth of those microbes capable of repairing damage

caused by heat is likely to occur. As a consequence, a technique was

used which measures metabolic activity of the culture. This technique is

based on the ability of the compound INT to be reduced by dehydrogenase

enzymes to a water insoluble salt, INT-formazan, which is deposited

internally in the cell. Several dehydrogenase enzymes are active in the

electron transport system where INT competes with oxygen for electrons.

Previous studies have shown that a large proportion of the total

metabolic activity can be linked to electron transport system activity

(21). Using this technique, whereby, the INT-formazan salt is extracted

from the cells and its concentration measured by absorption at 490 nm,

-83-

flow (with a dilution rate of 0.08 h-ll containing glucose (2.5 g 1-ll

in Evans mineral salts medium to the previously heat stressed cells in

the bioreactor after cooling the culture to 35°c. In order to ensure all

bacteria in the bioreactor were subjected to the heat stress the

procedure followed was to stop the flow of bacteria from the growth

reactor, and only four hours later to reduce the temperature and start

the nutrient flow. The ensueing washout and regrowth was monitored by

absorption at 546 nm (see Fig 4). This experiment was carried out after

treatment at 49°, 55° and 60°c and in each case regrowth occurred. After

treatment at 49°c, there was a long lag phase lasting approximately 7 h

whilst for the bacteria treated at 55° and 60°c, the lag phase was much

shorter. The identity and purity of the regrown culture was checked

using the API 20E identification system and !· pneumoniae was identified

as the only microbe present in all regrowth experiments.

Discussion

The results with respect to changes in various parameters as a result of

heat shock, shown in Figs. 1-3 are presented in terms of mass per unit

volume. In studies such as this, where the physiological matrix of the

reference quantity is drastically different under different conditions,

the use of this basis is questionable. Every indication suggests that

death/lysis events are occurring in the culture, and therefore, release

of cell contents into the extracellular medium is occurring. The

acridine orange direct count also shows a clear decrease in the number

of cells per ml and thus expression of the amount of protein found on a

mg per ml basis does not take into account that successively fewer cells

are present per ml with increasing temperature. Thus, as examples, the

results for dry weight and cellular protein have been recalculated and

expressed on a mg per cell basis (Fig. 5). The trend here is the reverse

of that described earlier. Whilst in terms of total cellular protein per

-84-

ml, a reduction was measured, the amount per cell is higher under

starvation conditions with a 32 h bioreactor residence time and during

starvation/temperature treatment. Similarly, whilst the dry weight of l

ml culture fluid decreases with increasing temperature, an increase in

the dry weight of individual cells could be measured when calculated on

a different basis.

Historically, the decrease in biomass found as a result of starvation

has been ,,scribed to endogenous metabolism. Endogenous metabolism is

defined as 'the summation of all metabolic reactions which occur when a

cell is deprived of either compounds or elements which may serve spec-

ifically as exogenous substrates (23). Comparing the data for cells

grown at 35°c with nutrient flow, and those maintained under starvation

(32 hour residence time) in the second bioreactor at 35°c would suggest

that metabolism of endogenous substrates has brought about a reduction

in the biomass. That a reduction at 41° and 49°c occurs can be justified

to some extent using this same hypothesis based on the fact that a

significant level of metabolic activity can still be measured. However,

the extremely low level of metabolic activity found at 55° and 6o0c,

resulting from bacteria which had been in the bioreactor for an

extremely short time, is insufficient to justify the biomass reduction

based on the endogenous metabolism hypothesis. The foregoing discussion

taken together with the fact that protein in individual cells is not

declining as suggested by Fig. 3, but accumulating, leads to increasing

reluctance to ascribe the observed effects to endogenous activity.

Certainly, the production of heat shock proteins has been well

documented and may account for some of the increased values for protein

per cell (24). However, as in the example of endogenous degradation of

protein and other cellular components, the low level of metabolic

activity in the culture at 55° and 6o0 c prohibits the extrapolation of

this concept to the higher temperature conditions.

-as-

Evidence for lysis of the cultrue as a consequence of both starvation

conditions and from temperature effects are provided by the increase in

UV absorbing material (protein, nucleic acids) and from the reduction in

cell number. Both of these variables are unaffected by the difficulties

described above such that more reliance can be placed on these data.

Thus the process of "cryptic" growth is expected to be occurring in

those cultures where the temperature does not inhibit the specific

matabolic pathways involved. "Cryptic" growth has been described in

detail by Mason and Hamer (25) for the microbe !:· pneumoniae and may

well be the reason for the apparent absence of accumulation of

extracellular protein at 35°, 41° and 49°c. At 55° and 60°c it is also

assumed that "cryptic" growth is not occurring.

A more likely explanation of the results reported here is that the

permeability of the cell changes as a result of both the imposition of

starvation conditions and from the effects of temperature. As a result,

the contents of the cell become concentrated due to water efflux, thus

resulting in the higher concentrations of protein per cell. An effect of

temperature on the microstructure of the cell contents was suggested

many years ago by Heden and Wyckoff (26) who showed that heating of ~·

coli to temperatures between 50° and 60° C resulted in granulation of

the cytoplasm. This effect began when cells were heated to 5o0 c and was

extensive in cells heated to 60°c. But even in cells heated to 40°c

(from a growth temperature of 37°c), some loss of protoplasmic

homogeneity had already occurred. This change in the protoplasmic matrix

was found to be irreversible above 5o0 c, and partially reversible at 45°

and 50°c when the cells were returned to their growth optimum

temperature. A similar effect was observed more recently where under

starvation conditions only, a marked decrease in the size of the

protoplasm could be discerned with increasing time of starvation (27).

This change in permeability of the membrane may well be accompanied by a

change in the ash content, thus accounting for an increase in the dry

weight of ~· pneumoniae on a per cell basis. At temperatures above 55°c

either the cell membrane or the cell wall may become heat stabilized and

-86-

prevent additional loss of internal contents and this may be the reason

why the amount of DOC released and the amount of 260 nm absorbing

material decreases slightly between 55° and 60°c.

Given these negative aspects with respect to the survival of K.

pneumoniae at high temperatures, the data shown in Fig. 4 still need to

be explained. The most likely reason for the regrowth of ~· pneumoniae

is that microbes were able to reinoculate the cooled culture from head

space splash. Wall growth in the head space is commonly encountered in

continuous culture vessels. Hamer (28) suggested that in nonfoaming

cultures some degree of microbial entrainment by bubble flotation

followed by entrainment of microorganism containing droplets from bubble

bursting at the liquid surface and collisions between the droplets and

the vessel walls can lead to head space wall growth. Such microbes will

be in an atypical environment compared to that in the bulk liquid and

may, therefore, be exposed to much lower temperatures than prevalent in

the bulk liquid. Hamer (28) calculated that in a fermenter operating at

35°c with a sparged air flow of l volume per volume of medium per minute

the volume of head-space drops formed as a result of bubbles bursting on

the surface would be equal to 1.64 litres per litre of medium each hour.

A certain fraction of the droplets produced will collide with the

fermenter walls and can result in a layer (often a thick layer) of cells

in the head space wall area.

The consequences of this finding is that in any incompletely filled

bioreactor, the head space will contain a source for reinoculation, by

head space drainage, even under conditions where careful control is

carried out. In this series of experiments, foaming was avoided by

careful operation, including gradual changes in bioreactor temperature

for enhanced temperature treatment. Moreover, the control of a small 2.5

litre fermenter is better than that of a 10 m3 industrial bioreactor.

Thus in the treatment of waste sludge, reinfection and regrowth may

arise in "treated" sludge as a result of head space reinoculation in the

bioreactor, particularly where only partial biodegradable organic matter

-87-

stabilization has ocurred. The low level of metabolic activity found at

temperatures of ss0 and 60°c also suggests possibilities for regrowth in

partially treated sludge when organisms short-circuit the thermophilic

bioreactor in continuously fed processes, a feature that tends to be

enhanced as system heterogeneity increases.

Conclusions

The use of temperature to kill pathogenic organisms in processes such as

waste sewage sludge treatment requires effective stabilisation and

sensible bioreactor design to minimise the danger of reinfection. In

stress studies on microbes, the use of specific defined parameters to

describe effects on individual microbes as opposed to overall parameters

reveals different trends and casts doubts on the concept of endogenous

metabolism as a realistic explanation of the survival process mechanism.

Such processes can be better described in terms of death/lysis and

"cryptic" growth phenomenon, together with changes in the physical

properties of the cells.

-88-

References

1. Sonnleitner, B. 1 Fiechter, A.: Advantages of using thermophiles in biotechnological processes: expectations and reality. Trends in Biotechnol. 1 (1983) 74-80.

2. Scheie, P; Ehrenspeck, S.: Large surface blebs on Escherichia coli heated to inactivating temperatures. J. Bacteriol. 114 (197'3) 814-818.

3. Pellon, J.R.; Ulmer, K.M. and Gomez, R.F.: Heat damage to the folded chromosome of Escherichia ~ K-12. Appl. Environ. Microbial. 40 (1980) 358-364.

I 4. Pellon, J.R.; Gomez, R.F.: Repair of thermal damage to the

Escherichia coli nucleoid. J. Bacterial. 45 (1981) 1456-1458.

5. Sedgwick, S.G.; Bridges, B.A.: Evidence for indicrect production of DNA strand scissions during mild heating of Escherichia coli. J. Gen. Microbial. 71 (1972) 191-193.

6. Pellon, J.R.: A note on the repair of the Escherichia coli nucleoid structure after heat shock. J. Appl. Bacterial. 54 (1983) 437-439.

7. George, T.K.; Gaudy, A.F.Jr.: Transient response of continuously cultured heterogenous population to changes in temperature. Appl. Microbial. 26 (1973) 796-803.

8. NG, H.: Effects of growth conditions on heat resistance of Arizona bacteria grown in a chemostat. Appl. Environ. Microbial. 43 (1982) 1016-1019.

9. Wu, S-Y.; Klein, D.A.: Starvation effects on Escherichia coli and aquatic bacterial responses to nutrient addition and secondary warming stresses. Appl. Environ. Microbial. 31 (1976) 216-220.

10. Paris, S.; Pringle, J.R.; Saccharomyces cerevisiae: heat and gluculase sensitivities of starved cells. Ann. Microbial. (Inst. Pasteur). 1348 (1983) 379-385.

11. Verrips, C.T.; Glas, R.; Kwast, R.H.: Heat resistance of Klebsiella pneumoniae in media with various sucrose concentrations. Eur. J. Appl. Microbial. Biotechnol. 8 (1979) 299-308.

12. Neidhardt, F.C.; Vanbogelen, R.A.; Vaughn, V.: regulation of heat shock proteins. Annu. Rev. 295-329.

The genetics and Gen. 18 (1984)

13. Streips, u.N.; Polio, F.W.: Heat shock proteins in bacilli. J. Bacteriol. 162 (1985) 434-437.

14. Gross, T.; Schulz-Harder, B.: Induction of a proteinase by heat-shock in yeast. FEMS Microbiology Letters 33 (1986) 199-203.

-89-

15. Mason, C.A.; Hamer, G.; Fleischmann, Th.; Lang, C.: Aerobic thermophilic biodegradation of microbial cells: 1. Some effects of dissolved oxygen concentration. Manuscript submitted.

16. Mason, C.A.; Hamer, G.1 Fleischmann, Th.; Lang, C.: Aerobic thermophilic biodegradation of microbial cells.: 2. Some effects of temperature. Manuscript submitted.

17. Hamer, G.; Mason, C.A.: Fundamental aspects of waste sewage sludge treatment: microbial solids biodegradation in an aerobic thermophilic semi-continuous system. Manuscript submitted.

18. Mason, C.A.; Hamer, G.; Bryers, microorganisms in enviornmental Reviews 39 (1986) 373-401.

J .o.: The death processes. FEMS

and lysis of Microbiology

19. Evans, C.G.T.; Herbert, D.; Tempest, D.W.; The continuous cultivation of micro-organisms. 2 Construction of a chemostat. In: Methods in Microbiology 2 (J.R. Norris, o.w. Ribbons, eds.) pp 277-327. Academic Press, London (1970).

20. Herbert, D.; Phipps, P.J.; Strange, R.E.: Chemical analysis of microbial cells. In: Methods in Microbiology 7B (J .R. Norris and D.W. Ribbons, eds.) pp. 209-344. Academic Press, London (1971).

21. Tabor.. P.S.; Neihof, R.A.: Improved method for determination of respiring individual microorganisms in natural waters. Appl. Environ. Microbiol. 43 (1982) 1249-1255.

22. Lopez, J.M.; Koopman, B.; Bitton, activated sludge process control. 1080-1085).

G.: INT-dehydrogenase test for Biotechnol. Bioeng. 28 (1986)

23. Dawes, E.A.: Endogenous metabolism and the survival of starved prokaryotes. In: The Survival of Vegetative Microbes (T.R.G. Gray and J.R. Postgate, eds). Proc. Symp. Soc. Gen. Microbiol. 26 (1976) 19-53. Cambridge University Press, Cambridge.

24. Groat, R.G.; Matin, A.: Synthesis of unique proteins at the onset of carbon starvation in Escherichia coli. J. Ind. Microbiol. 1 (1986) 69-73.

25. Mason, C.A.: Hamer, G.: Cryptic growth in Klebsiella pneumoniae. Manuscript submitted.

26. Heden, c. -G.; Wyckoff, R. W .G.: The electronmicroscopy of heated bacteria. J. Bacteriol. 58 (1949) 153-160.

27. Reeve, C.A.; Amy, P.S.; Matin, A.: Role of protein synthesis in the survival of carbon-starved Escherichia coli K-12. J. Bacteriol. 160 (1984) 1041-1046.

28. Hamer, G.: Discussion on entrained droplets in fermenters used for the cultivation of single-celled microorganisms. Biotechnol. Bioen9. 14 (1972) 1-12.

2.0

>-0

~

-0 c :. 1.0 e g >< :o.5 ... .Q

E "' z

=iii {.) 2.0

Fi2ure

100

80 .,,, ..§. c 60 0 .Q ::a

{.)

u ·;: "' ~ 40

0 -0 ... .. 0 "' "' 20 i5

0

-90-

t:.._ ......

0.8

.:;;..... - ... A... --

0

..I. I

Growing Culture

35

1. Changes and cell

I

40

in dry number

a culture of K.

0

"'

..I. Growing 35 40 Culture

---- ... _ "t>...._

I

45 Temp(°C)

.... _ .......... _6... -- .... _

-a.

weight ( 0) I metabolic activity ([J

(A)' as a result of heat treating

pneumoniae at various temperatures.

/'""- .... 3

/ --/ /

/ 2.5 /,e_._

/:I' /

/ 2 E // c:

/ 0 / <D

/ N

/ 15 .2 Q. 15 .,, .Q <t

05

45 50 55 60 0 Temp(°C)

Fi2ure 2. Production of dissolved organic carbon (o) and

260 nm absorbing material (A) in steady-state

heat treated cultures of K. 2neumoniae.

-! c ·;;; 0 a: ~ :i u

3.5

3.2

2.9

2.6

2.3

2.0

0

A

.L Growing Culture

-91-

1::1:------~-_..

35 40

---- ,,."' __ a

. 45 50 Temp(°C)

0.4

55 60 0

Figure 3. Changes in the cellular (o) and extracellular (A) protein concentrations in cultures of K. pneumoniae exposed to super-optimal temperatures.

5 4

E 3 c:

<D 2 -:t tn c: 0 ·.;:; 0.. 1 .... 0 0.9 .,,

D QS <( 0.7

Q6 Q5

4 E 3 c:

<D -:t tn 2 c: .2 ..... 0.. .... 0

1 .,, D 0.9 <( (18

0.7 0.6 0.5

4

E 3 c:

<.D 2 -4" I()

c: g 0.. 1 .... 0 0.9 .,,

D 0.8 <( 0.7

0.6 0.5

-4

A

8

c

·2

-92-

0 2 4 6 Time(li)

8 10 12

Figure 4; Regrowth of heat treated cultures of K. pneumoniae. Continuous feed of microbes was halted four hours before returning the temperature to 35°c and starting nutrient flow. A, 49°c; B, ss0 c1 c, 60°c. The expected washout curves are also shown (---) •

0.8

\... al v ~ 0.6

0

~ ~

~ 0.4 ~ -0

0.2

A

0 .!. Growing Culture

-93-

35 40 45 50 55 60 Temp(0 C)

3

2.7 ~ a; v en E

2.4:;: Q

c a;

2.1 0 ct

1.5

Figure 5. Expression of protein and dry weight data in units of per cell.

-94-

CHAPTER 7

THE LIMITS OF EFFECTIVE PERFORMANCE OF

THERMOPHILIC AEROBIC SLUDGE TREATMENT.

Abstract

Increasing application of effective sewage and wastewater treatment technology results in an increasing production of sludge for ultimate disposal. In addition, the continued disposal of sludge on agricultural land requires the introduction of improved sludge stabilization processes. Aerobic thermophilic digestion is one such process, and here, it is examined in the light of established microbial process information.

-95-

Introduction

The widespread application of sewage and wastewater treatment processes

to satisfy increasingly stringent legislation concerning aquaeous

discharges into surface waters has resulted in increased sludge

production and has exacerbated problems of sludge disposal. In

conventional treatment processes, sludge is produced from the

preliminary, primary and secondary process stages. It comprises a dilute

putrefactive suspension of biodegradable and non-biodegradable matter

containing pathogenic organisms and potentially toxic chemicals.

Traditional methods of sludge disposal include; (1) dumping at sea, an

option that is unacceptable to land locked countries; (2) incineration

which frequently requires supplementary fuel and results in air

pollution and, ultimately, extensive land and surface water pollutioni

(3) burial with urban garbage, which can result in leachate problems,

(4) composting with either solid biodegradable residues, where the

balance of the materials processed is critial with respect to compost

quality; (5) disposal by spreading on agricultural land after

stabilization. The last approach involves digestion of the sludge to

reduce the fraction of rapidly biodegradable matter present prior to

spreading, and is usually the preferred solution for sludge disposal in

land locked regions. However, increasing concern, on the grounds of both

public/veterinary health aspects and nuisance problems arising from land

application, is becoming evident and more effective and reliable sludge

digestion processes are urgently required if the practice of land

disposal is to continue as an acceptable option.

The most widely used sludge stabilization process presently in use is

mesophilic anaerobic digestion. Alternative processes, with potentially

superior performance and operating economics are the following:

thermophilic anaerobic digestion, mesophilic aerobic digestion, and

thermophilic aerobic digestion. In this paper, factors affecting the

effective performance of the thermophilic aerobic digestion process will

be evaluated.

-96-

'l'herinophilic Microbes

Microbes are a universally available natural resource, but their

effective application in industrial processes depends on a comprehensive

understanding of their physiology and biochemistry. Temperature is a

very important factor affecting microbial kinetics, physiology and

biochemistry, and any species or strain can be broadly classified

according to its optimum temperature for growth. The psychrophiles have

temperature optima for growth of <:10°c, the mesophiles, of between 15° and 40°c, and the thermophiles, of >4s0 c. Generally, the thermophiles

are further sub-divided into; thermotolerant microbes that grow

optimally between 40° and so0 c, a group that comprizes predominantly

facultative thermophiles; moderately thermophilic microbes with growth

temperature optima between so0 and 65°c and exhibit a minimum

temperature for growth of ca. 45°c; extremely thermophilic or

caldoactive microbes that have growth optima )65°c. Caldoactive microbes

are relatively fastidious in their growth requirements (ll, and hence,

are poorly suited for practical sludge digestion processes, whilst

thermotolerant microbes can be considered to offer only minor advantages

over mesophilic microbes. Therefore, as far as improved sludge digestion

processes are concerned, it is moderate thermophiles that offer genuine

practical process potential, and will be discussed here.

Moderate thermophiles are optimally adapted for growth at elevated

temperatures and do not struggle to survive at such temperatures (2). By

analogy with chemical reactions, where increasing temperature usually

increases the reaction rate, it is generally assumed that increasing

temperature will increase the rate of microbially mediated reactions.

Inspection of the relationship between the maximum specific growth rate

constant and temperature for any particular microbial strain shows an

exponential dependency of the maximum specific growth rate constant on

temperature, over a range of ca. 15°c until a maximum value is reached.

This is maintained for a few degrees and then, within a further 4° to

s0 c, a precipitous drop in the maximum specific growth rate constant

occurs.

-97-

In sludge digestion processes, the primary substrates are particular in

nature, and hence, the environment in which the microbes function is

markedly heterogenous, and this is of obvious significance in any

analysis of temperature effects particularly when the rate controlling

step concerns a mass transfer resistance rather than one of a sequence

of biochemical reactions.

Clearly identifiable effects of temperature on microbes include those

related to; (1) the maximum specific growth rate constant; (2) the

affinity of the microbes for their growth limiting substrate; (3)

endogenous metabolism or maintenance requirements; (4) the specific

death and lysis rate constants. In addition, temperature can also be a

selective pressure with respect to metabolic pathways employed, Hence,

the response of either a microbial strain or an association of strains

to temperature will be both complex and varied. However, as far as

effective sludge mass reduction and heat production are concerned, it is

the factors that affect overall yield coefficient that are most

important, and as far as a hygienic final product is concerned, it is

the factors that control thermal death and deactivation processes that

are critical.

Yield Coefficients and Beat Production

The yield coefficient for microbial biomass formation from any substrate

or nutrient is defined as the weight of biomass produced per unit weight

of substrate or nutrient utilized. In the case of substrates that are

insoluble, immiscible and/or partially non biodegradable, a conversion

coefficient, defined as the weight of biomass produced per unit weight

of substrate supplied, is frequently more appropriate. Digestion

processes seek to minimise the fraction of biodegradable matter present

in the sludge by eith<?r utilization of the biodegradable matter under

conditions where the biomass yield coefficient is minimised or by a

sequence of utilization, lysis and reutilization steps where the overall

biomass yield coefficient is minimised.

-98-

The heat produced in any well-defined, microbial bio-oxidation process

can be calculated by application of thermodynamics to the particular

process (3). However, such an approach is much more difficult when

ill-defined, complex substrates, such as waste sludge, are considered.

Microbial biomass constitutes a significant fraction of sewage and

wastewater treatment process sludge and the average heat of combustion

of microbial biomass has been reported to be 5.32 Kcal g-l on an ash and

moisture free basis (4). For the aerobic growth on and utilisation of

any carbon energy substate, Linton and Stephenson (5) have demonstrated

that the heat of combustion per g substrate carbon is related to the

maximum biomass yield coefficient based on unit weight of carbon in the

substrate. Assuming microbial biomass has a carbon content of 47\, the

heat of combustion on a unit weight of carbon basis will be 11.32 Kcal

g -l C. The maximum biomass yield coefficient for a "cryptic" growth

process can then be estimated from the relationship proposed by Linton

and Stephenson (5). This is 0.66g dry biomass per g dry substrate

biomass, assuming complete incorporation of substrate carbon into new

cells and carbon dioxide. De-optimization of the biomass yield

coefficient for "cryptic" growth will probably result in operating yield

coefficients as low as ca. 0.33.

several empirical formulae for microbial biomass, on an ash and moisture

free basis have been proposed, including CH1. 8o0 • 43N0 . 23 (6). For a

carbon content of 47%, this gives corresponding nitrogen and ash

contents of 12.6\ and 6.4%, respectively. Using this empirical formula

and assuming no nitrification, theoretical equations that describe

"cryptic" growth processes, operating at different yield coefficients

can be devised and the respective oxygen demands calculated.

For aerobic growth processes, Cooney .!!.!: al. (7) obtained a linear

correlation between the rates of heat production and of oxygen

consumption. This correlation can be applied to aerobic "cryptic" growth

processes, operating at various productivities for biomass formation on

the basis of oxygen demands calculated from the theoretical equations,

for various produced biomass yield coefficients.

relationships are shown in Figure 1.

Some predicted

-99-

Death, Lysis and •eryptic• Growth

In all digestion processes, one group of bacteria utilizes other

bacteria as their carbon energy substrate. For such processes to occur,

the bacteria that act as the substrate loose viability and subsequently

lyse. The mechanisms involved in such a process are undoubtedly complex.

In thermophilic aerobic digestion processes the substrate bacteria are

first subjected to mild temperature shock that either deactivates or

even kills them and then to the effects of either autolysis or to the

action of lytic enzymes produced by the thermophilic process culture.

Only after lysis is it possible for the process culture to utilize the

lytic products for growth, as direct ingestion of one bacterium by

another cannot be postulated as a utilization mechanism.

For the mixed, continuous flow digestor, operating without recylce,

shown diagramatically in Figure 2, material balance equations can be

established for the lysis of the substrate bacteria, lytic product

formation and growth of the product bacteria on the lytic products.

Substrate bacteria balance

V.dx s (1)

where V is the digestor liquid volume, F, the inlet and outlet liquid

flow, xso and xs' the concentration of substrate bacteria in the inlet liquid and the digestor, respectively, k

1 , the specific lysis rate

constant and, t, time. Rearrangement of equation 1 gives

dx s

dt

where D is the dilution rate.

(2)

-100-

Product bacteria balance

V.dx V.p.x.dt - F.x.dt (3)

where x is the concentration of product bacteria in the digestor.

Rearrangement of equation 3 gives

dx (p-D) x (4)

dt

Lytic product balance

V.dp -F.p.dt - dt + (5)

where p is the lytic product concentration in the digestor, Yx/p' the

yield coefficient for product bacteria formation from lytic product and

Yp/s' the yield coefficient for lytic product formation from substrate

bacteria, which in a first approximation can be assumed to be 1.

Rearrangement of equation 5

dp p.x -Dp - + (6)

dt

For steady state conditions, dxs/dt, dx/dt and dp/dt will be zero, and

hence equations 2,4 and 6 can be rewritten, for steady state conditions

be equating to zero and substituting x5

, i and p for xs, x and p,

respectively. If it is assumed that the Monod relationship applies to

product bacteria growth on lytic products under steady state conditions,

then

(K + p) p

-101-

where Pm is the maximum specific growth rate constant and KP,

saturation constant.

From the steady state version of equation 2

x s

D.x so

(7)

the

(8)

Substituting for p in the steady state version of equation 4 in equation

7 gives

K .D p

Substituting for p, J'• and

gives

kl.xso x Yx/p

(D + K1 )

x

(9)

in the steady state version of equation 6 s

K .D p (10)

(J.lm - Dl

Typical values of the several constants can be inserted in eguations 8,

9, and 10 and the effect of D on x8

, x, and p can be evaluated.

Obviously, the value of k1 is critical as far as digestor performance is

concerned. Specific lysis rate constants for mesophilic bacte~ia in the

thermophilic temperature range under aerobic conditions are unavailable.

However, specific lysis rate constants, during the growth of a

methanotrophic bacterium in continuous culture at 42°c of between O.Ol

and 0.07 h-1 , depending on the dilution rate, have been reported (8).

Therefore, it can be assumed that the k1 values encountered in aerobic

-102-

thermophilic digestors for the bacteria undergoing destruction will

either be similar to or exceed those quoted. Predictions based on k1 values of 0.04 and 0.14 h-l are shown in Figures 3 and 4, respectively.

concluding Remarks

In order to develop competitive aerobic thermophilic sludge

stabilization processes, it is essential to produce sufficient heat to

permit autothermal operation, and simultaneously, achieve effective

destruction of putrefactive matter, particularly pathogenic organisms

present in either an active or a resting state in the sludge undergoing

stabilization. Effective oxygen conversion and short residence times in

the digester are also desirable. The approach presented in this

communication clearly indicates that established microbial process

information can be directly applied to aerobic thermophilic sludge

digestion in order to identify some of the process performance criteria

that will have to be met by such processes.

References

1. B. Sonnleitner, A. Fiechter, Trends Biotechnol., !• 74, 1983.

2. T.D. Brock, Proc. 19th Symp. Soc. Gen. Microbial., London, 15, 1969.

3. P.L. McCarty, Int. J. Air Wat. Poll., ~. 621, 1965.

4. J.J. Powers, A.J. Howell, S.J. Vacinek, Appl. Microbial.,~· 689,

1973.

5. J.D. Linton, R.J. Stephenson, FEMS Microbiol. Lett., 1• 95, 1978.

6. G. Hamer, D.E.F. Harrison, Proc. Inst. Petroleum Symp. Cantebury, GB,

59, 1979.

7. C.L. Cooney, D.I.C. Wang, R.I. Mateles, Biotechnol. Bioeng., 11, 269,

1968.

8. J.W. Drozd, J.D. Linton, J. Downs, R.J. Stephenson, FEMS Microbial

Lett.,~· 311, 1978.

-103-

0.33 u: 200 u.

0.44 w 0

0.55 u

7..: 100 0 _, \.... 0.66 \!::! ;;; >-..... 50 .... z 0 j::: u ;:::) 0 0 10 0:: Q_

!;( 5 w :c

2 0.1 0.2 0.5 1 2 5

BIOMASS PRODUCTIVITY (g 1"1 h'1 )

Fiqure· l. Estimate!; of heat production during "cryptic" growth

with respect to biomasi; productivity at a series of

yield coefficients.

F --

v x x p ii. ...

Fi<JUre 2. Digestor configuration.

-

F x.s x p

' -OI

10. .... 0

I><

' -en

!><

-104-

12 48

k, = 0.04 h-1

8 32 ><I .. '° -:.

4 16

oL--1...~.1.=====::::t:::=:::::x::::::=i_--L----L-.l.-J....Jo

12

a

4

0 0.04 0.16

Figure 3. The effect of dilution rate on steady state bacteria {i5

),

product bacteria (i} and lytic product (i)) for a specific

lysis rate constant (k1 ) of o.04 h-l when µm = 0.17 h- 1 ,

-1 -1 • 40 gl-1 Yx/p = 0.33 h , kp • 0.2 gl and x50

0.14 h"1 kl =

24

><I 16 ..

0 ... ..;;,.. j5 "01

'----""" x s '° :. 8

OL__.1..~-'-~"----======:r:::=::::::r::___,_..J_J_JO

0

1'i9ure 4.

0.04 0.16

The effect of dilution rate (0) on steady state substrate

bacteria (i ) , s for a specific

-1 I' • 0.17 h • m

product bacteria (i) and lytic product (f>) -1

lysis rate constant (kl) of 0.14 h when

\;p • 0.33, kp • 0.2 gl·l and x50

• 40 gl•l

-105-

CHAPTER 8

AEROBIC THBRMOPHILIC BIODEGRADATION OF MICROBIAL CELLS

Abstract

Aerobic thermophilic treatment can be used either as a pretreatment step prior to anaerobic mesophilic treatment or

as a total treatment process for waste sewage sludge. In this contribution a laboratory-scale process research study

that has been undertaken to investigate the effects of some process variables on microbial solids solubilization is described. The operating parameters investigated are bioreactor residence time and pH at 60° c and high dissolved oxygen concentrations.

-106-

Introduction

The major by-product of conventional aerobic biological wastewater

treatment processes is waste sludge, a putrefactive, aqueous suspension

of biodegradable, partially biodegradable and essentially

non-biodegradable solid, dissolved and sorbed matter. Waste sludge

presents a serious disposal problem, particularly in regions remote from

the sea, where frequently the policy for ultimate disposal involves

spreading treated sludge on agricultural land so as to cause minimum

nuisance. Conventional waste sludge treatment technology involves

anaerobic, mesophilic digestion, but such technology is no longer

considered entirely satisfactory for the removal of either pathogenic

organisms or toxic chemicals from sludge. Further, the physical charac-

teristics that affect the dewatering of conventionally treated sludges

are frequently such as to result in high dewatering process costs, a

feature of obvious economic concern.

Anaerobic thermophilic, sludge digestion processes have been proposed as

cost effective alternatives to conventional, anaerobic mesophilic,

processes, but plants constructed for the latter technology, and

comprizing pre-stressed concrete reactors, cannot easily be converted

for thermophilic process operation. Most compounds that are

biodegradable under aerobic conditions are also biodegradable under

anaerobic conditions, usually at markedly different rates, but certain

important pollutants commonly found in waste sludges, including

hydrocarbons and some synthetic organic compounds, particularly

detergent residues (Giger ~ al., 1984) are recalcitrant to anaerobic

biodegradation. In view of both this and of the question of recently

installed anaerobic mesophilic, digestion capacity in many treatment

plants, the most realistic approach for more effective waste sludge

treatment would seem to be the introduction of either pre- or

post-treatment processes operating in conjunction )'lith

anaerobic, mesophilic, sludge digestion processes,

conventional,

that enhance

treatment effectiveness with respect to the criteria discussed.

-107-

Aerobic sludge treatment was first proposed in 1959 (Husmann & Malz, 1959). Essentially four distinct processing concepts exist 1

(i) Aerobic mesophilic (15-35°C) digestion;

(ii) Aerobic autothermic (40-50°C) digestion;

(iii) Aerobic thermophilic (50-70°C) digestion;

(iv) Aerobic thermophilic (50-70°C) I Anaerobic mesophilic (30-40°C) digestion.

Combined aerobic thermophilic /anaerobic mesophilic treatment processes

offer the greatest potential for effective waste sludge treatment.

However, bioprocess data concerning aerobic thermophilic treatment are strictly limited. In order to rectify this, data for microbial solids

destruction are reported.

Materials and Methods

A laboratory-scale bioreactor (Chemap AG, Volketswil, CH) with an

operating volume of 6 litres was used for the experiments. The

bioreactor was operated in a continuous flow mode with a feed stream

consisting of bakers yeast, as the sole biodegradable carbon and nitrogen source, dispersed in a solution of K2HP04 (5.7 g/l) and KH2Po4 (8 g/ll. No other nutrients were provided. Air or air/oxygen mixtures

were sparged into the bioreactor and the temperature of the bioreactor

was maintained constant at 6o0 c.

Samples of the bioreactor liquor and feed were centrifuged at 30,000 x g

and the supernatants decanted. The solids were transferred into a

crucible and analysed for total suspended solids (TSS) by overnight

drying at ios0 c, and for volatile suspended solids (VSS) by heating at 0 + - -600 c. NH4 -N, N02 -N, and N03 -N were determined in the supernatant

using an automated nitrogen analyser (Skalar, lli:"eda, NL). carboxylic

acids wei:"e analysed by gas chi:"omatogi:"aphy.

-108-

Results and Discussion

In the experiments, the effects of pH and residence time on the

operation of the aerobic thermophilic biodegradaton process were

investigated. Biodegradation of particulate organic matter proceeds

initially by its solubilization thereby enabling utilization of the

lytic products by the aerobic thermophilic bacteria. Particulate

biodegradation is a complex process which has received little attention

when compared to the utilization of soluble substrates in wastewater

treatment processes. Clearly, the rate limiting step in the overall

process is this initial solubilization and, therefore, the various

factors regulating this process must be understood for optimisation of

any sludge treatment technology.

Hamer et al. (1984) and Hamer & Bryers (1985) discussed the limitations

in the performance of the process with respect to the thermophilic

process bacteria and the effects of thermophilic operating temperatures

on mesophilic (substrate) microbes.

To complete their approach a discussion of enzyme production/action by

thermophilic bacteria and of the pattern of biodegradation is required.

Particulate biodegradation can proceed by two mechanisms;

a) physical contact between process bacteria and bioparticulates,

b) exoenzyme production by the process bacteria, and destruction

without direct contact.

In the former case, digestion is more likely to occur by entrapment in

floes rather than by incidental contact within the bioreactor. The

enzymes responsible for biodegradation will be either bound to the cell

surface or released within the floe such that enhanced concentration

will occur within the microenvironments in the trapped

particulates/floe.

-109-

More likely in a well mixed system, is the second mechanism whereby

exoenzymes (e.g. protease, lipase etc.) specific for the polymeric

components of the bioparticulate material are produced. Since

microorganisms account for the major fraction of the solid material in

sludge, their biodegradation is essential. The cell wall forms a

physical barrier between a pool of readily degradable soluble material

and the process bacteria. Thus initial attack by the enzmes produced by

the thermophilic process bacteria, on the cell wall biopolymers enables

the release and utilisation of the intracellularly available or9anic

compounds as carbon and/or ener<JY sources for growth and for autothermal

heat production.

The enzymes attacking and digesting the yeast cells will function

optimally at a particular pH. However, if more than one enzyme type is

acting, it is conceivable that there will be a range of pH values over

which biodegradation can occur - although with varying rates and

efficiencies. From the data in Fig. l where the percentage total

suspended solids degraded are shown, it is clear that most effective

biodegradation occurs around pH 6. 5. At pH 7. 5 the amount of total

solids removed is not as great, whilst at pH 5.7 biodegradation occurs

to an even lesser degree. These results represent steady state values

durin9 continuous flow operation of the process. The foregoin9 comments

are generalised for the two residence times investi9ated ( 3 and 1. 5

days). At the longer residence time biodegradation proceeds to a greater

de9ree. However, if the total solids removal rates are compared (Fi9. 2)

it is obvious that the shorter residence time is more efficient and

operation at pH 6.5 achieved the best results.

The pH also has a stron9 influence on the ash content of the biomass in

the bioreactor (Fi9. 3). At pH 7.5 the total suspended solids are

composed of a lar9er amount of ash than present in the feed, whilst at

pH 6. 5 and 5. 7 the ash content is reduced. Volatile suspended solids

(VSS) are also degraded optimally at pH 6.5 (Fi9. 4) and despite the

varyin9 ash residue contents, the rate of vss biodegradation, in

-110-

agreement with that for total suspended solids, is optimum at 1.5 day

residence times (Fig. 5). Under all conditions examined, insignificant

amounts of carboxylic acids were found, the highest concentration being

45 mg/l acetic acid at pH 5. 7/1.5 day. This is not surprising since

oxygen was always in excess during these experiments.

In the yeast cells, significant amounts of proteinacious material is

present, the biodegradation of which will release the nitrogen necessary + for the thermophilic bacteria. Accumulation of NH 4 -N was found at

extended residence times (3 days) but not at 1.5 days (Fig. 6).The

dependence of pH on the extent of accumulation, and therefore of

degradation, is clearly demonstrated in these data. No2 -N and N03 -N

concentrations were below detection limits (1 mg/l).Some loss of NH 3 in

the exhaust gasses is to be expected and therefore the presented data

probably underestimates the true amount released.

The results presented here are for a feed sludge composed entirely of

yeast cells in the absence of other soluble carbon energy sauces,

bioparticulates of other microbial origin and of inert particulates.

Naturally, real sludges are likely to behave somewhat differently from

the model study conducted here. Nevertheless, the majority of the

secondary sludge solids fraction, and some of the primary sludge solids,

is microbial in nature and, therefore, extrapolation to operating

conditions with real sludges is probably valid. The effect of residence

time is merely prolongation of contact between enzyme and substrate and

it is questionable as to whether the increased level of biodegradation,

which can be as much as 64% higher at the same pH, under this set of

operating conditions, warrant the increase in operating costs, part-

icularly when the aerobic thermophilic treatment step is intended as a

precursor to anaerobic rnesophilic digestion so that pretreatment of

complex biological particulates enables subsequent anaerobic digestion

to be more rapid. In the data presented here, aerobic treatment is shown

to be both pH and residence time dependent, reflecting the probable pH

optima of the enzymes involved in cell lysis and subsequent growth on

lytic product.

-lll-

References

Giqer, w., P.H.Brunner and C.Schaffner, Science 225, 623 (1984)

Hamer, G., J.D.Bryers, J.Berqer and C.A.Mason, Proc. 4th C.E.C., Grado,

I. 705 (1984).

Hamer, G. and J.D.Bryers, Conservation and Recyclinq ~. 267 (1985)

Husmann, w. and F.Malz, GWF-Wasser/Abwasser 100, 189 (1959).

-112-

.. .. 1.5

, . ..

••

Figure 1. Total suspended solids biodegraded as a percentage

of the influent total solids concentration at

various values of pH and for residence times of

i 1! •

J ft ' .., :g 1. l a·

3 and 1. 5 days.

1.5

Figure 2. Total suspended solid removal rate as a function

of pH and residence time in the aerobic thermophilic

biodegradation process.

i s'

J '

Figure 3.

..

Figure 4.

-113-

!.5

Ash content in the feed (left hand bar) and in the bioreactor (right hand bar) during aerobic thermophilic digestion of yeast cells.

3 1.5 ..

.. ••

pH

Volatile suspended solids biodegraded as a percentage of the input volatile suspended solids concentration at various residence times of 3 and 1. 5 days •

i • "

j· .. l

I Figure 5.

l •

~ % I . " :i: .. %

Figure 6.

-114-

3 1.5

Volatile suspended solids removal rate as a function of pH and residence time in the aerobic thermophilic biodegradation process.

3 1.5 ...

..

pH

NH 4-N in the feed (left hand bar) and in the bioreactor (right hand bar) during aerobic thermophilic digestion of yeast cells,

-115-

CHAPTER 9

AEROBIC THERMOPHILIC BIODEGRADATION

OF MICROBIAL CELLS:

1. SOME EFFECTS OF DISSOLVED OXYGEN CONCENTRATION

Results are presented comparing the extent of solubilization/biodegradation of yeast cells by aerobic thermophilic bacteria under conditions of oxygen excess and oxygen limitation. The process was most effective at low dissolved oxygen concentration as judged by solids removal data and by the production of often considerable quantities of carboxylic acids.

-116-

Introduction

Effective legislation by the governments of most West European countries

with respect to pollution of the aquatic environment has resulted in the

widespread installation of both municipal sewage and industrial

wastewater treatment during the past 10-15 years (Hamer, 1985).

Conventional treatment technology comprizes the mechanical and physical

separation of pollutants during preliminary and primary treatment and

the bio-oxidation of pollutants in secondary aerobic biotreatment

processes, most commonly of the activated sludge type. In some cases, an

additional tertiary treatment stage is also employed. The principal

by-product of treatment is waste sludge, comprizing a complex mixture of

suspended biodegradable, partially biodegradable and non-biodegradable

solid matter and associated sorbed and dissolved pollutants. A

significant fraction of waste sewage sludge comprizes microbial biomass,

including some potentially pathogenic microbes present in the process

feed and large quantities of process microbes from secondary treatment.

The conventional procedure used for waste sewage sludge treatment prior

to ultimate disposal, frequently by spreading on agricultural land, is

anaerobic mesophilic digestion/stabilization. Increases in the

quantities of sewage undergoing treatment has resulted in corresponding

increases in waste sewage sludge production and attendant difficulties

in finding disposal sites, particularly as some authorities responsible

for public and veterinary health have expressed reservations with

respect to the efficiency of conventional sludge treatment processes for

the destruction of both pathogenic organisms and removal of toxic

chemicals. To solve the problem of sewage sludge hygienization, it has

been suggested that anaerobic treatment be carried out at thermophilic

temperatures, ss-6o0c, in order to ensure the destruction of pathogenic

organisms. However, pre-stressed concrete anaerobic digestors, designed

for mesophilic operation, 30-37°c, and installed at many treatment

plants, cannot be operated in the thermophilic temperature range. Hence,

-117-

in Switzerland, West Germany and Austria, retrospective installation of

relatively short residence time, aerobic thermophilic pre-treatment/-

hygienization p.~ocesses prior to anaerobic mesophilic digestion/

stabilization processes is occurring (Hamer and Zwiefelhofer, 1986).

Such processes have a multiple role1 the destruction (deactivation) of

pathogenic organisms, the solubilization of particulate biodegradable

matter, particularly microbial biomass and the bio-oxidation of sorbed

pollutants, removed with solids during primary treatment, that are

essentially recalcitrant to anaerobic biodegradation (McEvoy and Giger,

1986) •

In this contribution, the question of the effect of dissolved oxygen

concentration on the solubilization and biodegradation of microbial

cells during aerobic thermophilic treatment will be addressed.

Solubilization of the biodegradable particulate fraction of waste sewage

sludges is a prerequisite for complete treatment. Eastman and Ferguson

(1981) recognized that the hydrolysis of particulate matter to soluble

substrates was the real rate limiting step in anaerobic digestion

processes and recent kinetic models for anaerobic digestion (Gujer and

Zehnder, 1983, Bryers ~al., 1985) incorporate this concept. Because of

the variability of real waste sewage sludges, it is necessary to select

a source of microbial biomass that is of standardized and reproducable

quality for experimental process studies. Here, pressed baker's yeast

was used.

llaterials and llethods

Bioreactor and Operating Conditions: Two laboratory scale bioreactors

with operating volumes of 1.3 1 and 9 1 (Bioengineering AG, Wald, CH)

each with full measurement instrumention, i.e., pH, dissolved oxygen

concentration, temperature and impellar speed monitoring were used for

these experiments. The bioreactors were operated in the continuous flow

mode and were continuously sparged with air at 32 1 h -l for oxygen

-118-

sufficient conditions in the smaller bioreactor and 300 1 h-l in the

larger. Oxygen limited conditions were achieved by reducing the air flow

to 6 and 30 1 h-l respectively. The temperature was maintained constant

at 60° c.

Aerobic Thermophilic Culture and Feed: Aerobic thermophilic bacteria

were obtained from an operating municipal waste sewage sludge aerobic

thermophilic digestor (UTB, Urnwelttechnik Buchs AG, Buchs/SG, CH).

Pressed bakers yeast as the sole biodegradable particulate carbon source

was suspended in a nutrient solution containig NH4Cl,4gl-\ KH2Po4 , 8 -1 -1 -1 -1 gl ; K2HP04 , 5. 7 g l ; ZnO, 3.26 mg l 1 , FeC136H20, 43.2 mg l ;

-1 -1 -1 MnC1 24H2o, 16mg 1 ; CuC122H20, 1.36 mg l ; CoC1 26H2o, 3.8 mg 1 -1 -1 H3B04, 0.5 mg l ; MgC12 , 0.2 g l • The bioreactor feed was stored at

4°c for no longer than 5 days.

Analytical

Dry weight: 5 ml samples were centrifuged at 30,000 x g and the

supernatant removed. The solids were resuspended in water, poured into

tared crucibles and heated overnight at l05°c. They were reweighed after

cooling in a desiccator.

Ash/volatile suspended solids: After drying overnight, the crucibles

used to determine dry weight were placed in an oven at 6oo0 c for ca. 5

h, allowed to cool in a desiccator and then reweighed.

Dissolved Organic Carbon: Appropriate dilutions of the supernatent were

made and the DOC assayed directly with a TOCOR 2 total organic carbon

analyser (Maihak AG, Hamburg, D). Inorganic carbon was removed by

acidification using concentrated HCl and the excess co2 removed by

sparging with N2 for 12 minutes.

NH +-N: This was measured in the supernatants using an automated -4-nitrogen analyzer (Skalar, Breda, NL).

-119-

carboxylic Acids: Low molecular weight carboxylic acids were analysed by

acidifying the supernatants with formic acid and injecting into a

Shimadzu GC-RIA gas chromatograph (Shimadzu, Kyoto, J).

Scanning Electron Microscopy: samples were freeze dried at -ao0 c on

mica, spatter coated with Au/Pd and viewed in a Hitachi s-700 (Nissai,

Sangyo, J) scanning electron microscope.

Chemicals: All chemicals were of analytical grade and supplied by either

Fluka (Buchs/SG, CH) or Merck (Darmstadt, DJ •

The solubilization and subsequent biodegradation of whole yeast cells by

a thermophilic aerobic mixed culture was investigated with respect to

the effects of oxygen. The results are shown in bar charts in Figs. 2 to

8, and represent steady-state means of samples taken daily over at least

7 days. The results under oxygen sufficient conditions have been grouped

together (residence times 1.2, 1.5 and 5 days) as have those for oxygen

limited conditions (residence times 0.6, 1.0, 1.5, 2.1 and 5 days).

Both process microbes (aerobic thermophiles) and substrate yeast cells

can be seen in the scanning electron micrograph shown in Fig. 1. Some

indication of morphological variation in the bacterial population

together with partially lysed (ghost) yeast cells can be discerned in

this electron micrograph. A combination of the process microbe biomass

and the substrate ye~st biomass in the bioreactor give the results for

the total suspended solid concentrations shown in Fig. 2. The left hand

bar of each pair represents the input substrate concentration and the

right hand bar, the output concentration. in most cases, the input yeast

cell concentration was between 30 and 40 g 1-l A lower concentration

(20 g 1-l) was used for the 5 day residence times.

-120-

Using this data, the net percentage of microbial solids biodegraded can

be calculated and this is shown in Fig. 3. Some fluctuation in the

solubilisation/biodegradation efficiency can be seen by comparing the

results for 1. 2 and 1. 5 day residence times under oxygen sufficient

conditions. This is likely to have been due to a fault in the pH control

during operation at 1.2 days. The data for 1.5 days were taken at a pH

of 7.0 whilst those for 1.2 days were taken under uncontrolled pH. This

data clearly demonstrates the sensitivity of the process to variables

such as pH. Comparison between the extent of solubilization/bio-

degradation under oxygen sufficient conditions (Fig. 3a) with that

achieved under oxygen limited conditions (Fig. 3b) shows that for most

residence times studied, considerably greater solids removal is possible

under the latter set of conditions. No definite trend in the variation

of solubilization /biodegradation with residence time can be seen from

these data.

Rationalising the variance in the amount of solubilization/-

biodegradation with different residence times provides the data

presented in Fig. 4. Here, the total suspended solids removal rate is

shown as a function of both dissolved oxygen concentration and residence

time. The oxygen sufficient data indicate very low removal rates with

the exception of that for the 1.2 day residence time. In contrast, the

total suspended solids removal rate under oxygen limited conditions

demonstrates, in general, much higher removal rates.

Analysis of the ash content of both the feed and the bioreactor liquor

revealed several features. Under oxygen sufficient conditions (1.5 and 5

days) the ash content of the biomass was actually higher than in the

feed (Fig. Sa), whilst a net decrease in the ash content was recorded

for 1.2 days and for all conditions of oxygen limitation (Fig. Sb). The

percentage change in ash content is shown in table 1. There it can be

seen that the greatest change occurs with a 5 day residence time under

both oxygen limited and oxygen sufficient conditions.

-121-

The changes with respect to volatile suspended solids are shown in Fig.

6. The trend for volatile suspended solids is exactly the same as for

total suspended solids, with 5 day residence times and oxygen limited

conditions producing the greatest amount of solubilisation/bio-

degradation of volatile suspended solids.

More distinct differences as a result of varying the dissolved oxygen

concentration can be recognised in the dissolved organic carbon (DOC)

content (Fig. 7). Very little change in DOC occurs under oxygen

sufficient conditions but significant increases in DOC occurred at all

residence times when oxygen was limiting. The amount of DOC produced as

a percentage of the input DOC concentration is shown in table 2. The

highest DOC production (386\) occurred under oxygen limited conditions

with a l day residence time. Fig. 8 shows the carboxylic acid

concentrations under oxygen limited conditions. No carboxylic acids were

found under oxygen sufficient condition. The amount of each acid

produced varies with residence time, although, clearly, acetate is

produced in the highest concentration. NH4+-N concentrations are shown

in Fig. 9 for both the feed (NH4cl was present in the feed) and in the + bioreactor. NH4 -N was produced under both conditions, although greater

release was found during oxygen limitation.

Discussion

The results of the comparison of oxygen limited conditions and oicygen

sufficient conditions on the solubilisation/biodegradation of yeast

cells under thermophilic conditions clearly shows that oxygen limited

conditions are more effective, both in terms of the amount of solids

removed and in the rate of removal. Indeed, at economically feasable

residence times, 0.6-1 days, extremely high solids removal rates were

demonstrated. The same complex mixed culture was used for both oicygen

sufficient and oxygen limited conditions. Since process conditions were

-122-

changed at random, the culture must have been dominated by facultative

anaerobes. The growth of thermophilic aerobic bacteria on yeast cells

requires the production of specific yeast cell wall degrading

exo-enzymes by the process microbes. In the absence of these specific

exo-enzymes no degradation can take place. The outermost layer of the

cell wall of Saccharomyces cerevisiae (bakers yeast) is composed mainly

of polysaccharides (typically glucan-mannan polymers) together with

small quantities of protein and lipids. Once this physical barrier has

been disrupted, a pool of soluble organic compounds (proteins, amino

acids, nucleic acids etc.) are made available for utilisation by the

thermophilic aerobic bacteria. These soluble compounds are, after

further hydrolysis, the preferred substrate for the bacteria and whilst

they are available for assimilation, residual cell wall material remains

largely intact (see Fig. l).

The production of specific exo-enzymes depends upon the environment to

which the process microbes are subjected. Temperature, pll, dissolved

oxygen concentration and growth rate play a significant role in the

control of production. Truely exo-enzymes are defined as those which are

actively secreted by viable microbes, rather than those which arise from

either damaged or lysed cells (Burns, 1983). Due to the tremendous

diversity of soluble substrates released by cell breakage in aerobic

thermophilic sludge treatment, a complex mixed microbial community must

be expected to develop. This then implies that the process culture will

have an extensive potential biodeqradative capacity and also, the

possibility exists that positive interactions such as the removal of

toxic by-products and the establishment of nutrient gradients can occur.

Detailed information concerning mixed substrate utilisation in

continuous flow bioreactors is largely restricted to the case of

monospecies cultures and only few exceptions exist (Wilkinson and !lamer,

1979). In the process system under examination here, it is conceivable

that different process microbes are attacking different released

substrates from yeast cell lysis, and only limited multiple substrate

utilization by specific microbes is occurring. In a study on bacteria

isolated from an aerobic thermophilic treatment process for waste sewage

-123-

sludge, Sonnleitner and Fiechter (1983) found a range of thermophilic

strains of which they identified most as Bacillus spp. Of these, less

than 10\ were unable to degrade starch and less than 5\ were unable to

degrade casein. Extracellular enzyme production, particularly protease

and polysaccharase production, are well known features of Bacillus spp.

(Norris~ 21·• 1981).

Given the importance of exo-enzyme production for the solubilisation and

biodegradation processes for microbial solids in waste sludge treatment,

the factors regulating their production must be considered. Microbes

have evolved to occupy diverse environmental niches and a spectrum of

affinities by microbes for oxygen exists. Under anaerobic conditions,

microbial biomass yield coefficients are typically lower than under

aerobic conditions, a feature which obviously benefits anaerobic sludge

treatment by minimising biomass production. The results indicate that

enhanced exo-cellular enzyme production occurs under oxygen limited

conditions. This conclusion is supported by the observation that amylase

production is induced in oxygen limited shake flask cultures of

thermophilic ~ derived from thermophilic sludge (Grueninger et

21·. 1984).

During oxygen limited growth, organic compounds frequently serve as

electron acceptors in the metabolic pathways of facultative anaerobic

microbes and this results in considerable carboxylic acid and other

product formation and a low biomass yield coefficient. Excess production

of carboxylic acids· is not an l(\lndesirable feature in aerobic

thermophilic sludge treatment processes when such processes are coupled

with an anaerobic treatment process as a second stage.

-124-

Conclusions

The primary implication of the results is that aerobic thermophilic

pre-treatment processes for waste sludges function optimally under

oxygen limited conditions. Whilst it is probable that some aerobic

thermophilic sludge pretreatment/hygienisation plants already func-

tioning do indeed run under oxygen limited conditions, as suggested by

high oxygen conversion data, process designers still seek to achieve

oxygen excess conditions in aerobic thermophilic bioreactors (e.g. Match

and Drnevich, 1977) thereby ignoring the process microbiology of such

systems and its effect on efficient operation.

-125-

References

Bryers JD, Berger J and Hamer G (1985) Interpretation of thermophilic anaerobic digestion experiments using a dynamic structured model. Proc IWTUS-3 Conserv and Recyc 8:251-266.

Burns RG (1983) Extracellular enzyme-substrate interactions in Slater JH, Whittenbury R, and fiimpenny JWT (eds) Microbes natural environments. Proc 34 t symp SOC Gen Microbial. University Press, Cambridge, p 249.

soil. In: in their

Cambridge

Eastman JA and Furguson JF (1981) Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. J Wat Poll Cont Fed 53:352-366.

Hamer G (1985) The impact of government legislation on industrial effluent treatment. Proc IWTUS-3 Conserv and Recyc 8:25-43.

Hamer G and Zwiefelhofer HP (1986) Aerobic thermophilic hygienization -a supplement to anaerobic mesophilic waste sludge digestion. IChemE symposium series 96: 163-180.

Grueninger H, Sonnleitner B and Fiechter A (1984) Bacterial diversity in thermophilic aerobic sewage sludge. III A source of organisms producing heat-stable industrial useful enzymes, e.g., Ot-amylases. Appl Microbial Biotechnol 19: 414-421.

Gujer W and Zehnder AJB (1983) Conversion processes in anaerobic digestion, Wat Sci Tech 15:127-167.

Matsch LC and Drnevich RF (1977) Autothermal aerobic digestion. J Wat Poll Cont Fed 49:296-310.

McEvoy J and Giger W (1986) Determin<1.tion of line<1.r alkylbenzenesulfonates in sewage sludge by high-resolution gas chromatography/mass spectrometry. Environ Sci Technol 20:376-383.

Norris JR, Berkely RCW, Logan NA and O'Donnell AG (1981) The Genera Bacillus and Sporolactobacillus. In: Starr MP, Stolp H, TrUper HG, Balonos A and Schlegel HG. The Prokaryotes - a h<1.ndbook on habitats, isolation and identification of b<1.cteria. Springer, Berlin, Heidelberg, New York p 1711.

Sonnleitner B and Fiechter A (1983) Bacterial diversity in thermophilic aerobic sewage sludge - II. Types of organism and their capacities. Appl Microbial Biotechnol 18: 174-180.

Wilkinson TG and Hamer G (1979) The microbial oxidation of mixtures of methanol, phenol, acetone and isopropanol with reference to effluent purification. J Chem Tech Biotechnol 29:56-67.

-126-

Table l Change in ash concentration of the total suspended solids after aerobic thermophilic treatment as a percentage of the amount present in the feed yeast cell suspension.

oxygen sufficient

oxygen limited

Time

(d)

1.2 1.5

0.6 1.0 1.5 2.1 5

in ash content after aerobic thermophllic biotreatment

+ 6.83 10.00

- 33. 07

11.48 7. 22 1.88 9.64

21.22

Table 2 Change in dissolved organic carbon concentration after aerobic thermophilic treatment as a percentage of the amount present in the original feed yeast cell suspension.

oxygen sufficient

oxygen limited

Figure 1.

-127-

Scanning electron micrograph showing the process

microbes and the feed yeast cells during

aerobic-thermophilic biotreatment. Several lysed

(ghost) yeast cells are present composed only of

cell wall debris.

'°1 "'

' :§'

~JO 0

"' " .. I! ; 20

Jl ~ ......

a

-128-

' Si

:l!! ~

..

••

b

Residence Time (d)

Figure 2.. Total suspended solids concentration in feed (left hand bar of

each pair) and in the bioreactor (right hand bar) as a function

of residence time under ·a) oxygen sufficient conditions and

b) reduced oxygen concentrations.

a b .. .. "'

.,, ~

.. 'O

:~ f .. "8 iii ., i '° ~

~AO ,:;: i 'O .. "li .. ? .. ~ ~ JI Jl

~ ao .. ~ JO

~ e ll' !! g $0 ii ..

~ l

Figure J. Total suspended solids biodegraded as a percentage of the

influent total solids concentration for various residence times.,

under a) Oxygen sufficient conditions and bl low oxygen concentrations.

i :g. 5 <

-129-

a b 20 20

i " i :g. .! 15 ::.

j .... l

L .

Piqure 4.

a

•

Figure 5.

Residence Time (d)

Total suspended solids removal rate as a function of residence

time and a) oxygen sufficient conditions and b) reduced oxygen

concentrations.

b

~ ~ •

Residence Time (di

Ash content in feed yeast cell suspension (left hand bar of each

pair) and in the bioreactor (right hand bar) as a function of

residence time under a) oxygen sufficient conditions and

b) low oxygen concentrations.

i eo a

I 0 .. iii ~

31 •• "ll ...

'" "' .!! j 20 !$ .. "' .\!! ti •• li "-

F:j.gure 6.

a

i §

8 " 3 " 'i ~ ... ~

j

' Figure 7.

-130-

i •• b ,, i ,, 0 .. iii

j ~o a: 1 lh• ill .!! "' j It) .. :ii' ! .. ;f

Residence Time (d) Resldenee Time (d)

Volatile suspended solids biode9raded as a percentage of the

influent(feedl volatile suspended solids concentration for

various residence times under a) oxygen sufficient conditions

and b) low oxygen concentrations.

b

L. s

i 0

I I

• " Residence nme (d) Residence Tl ... (d)

Dissolved organic carbon in feed yeast cell suspension

(left hand bar of each pair) and in the bioreactor (right hand

bar) as a function of residence time under a) high dissolved

oxygen concentrations and b) low dissolved oxygen concentrations.

3

-I I s

.5

0

Piqure 8.

a ...

..

Figure 9.

-131-

Residence Time (d) Carboxylic acids measure in the bioreactor as a function of

residence time under reduced oxygen conditions. No carboxylic

acids could be detected u.nder oxygen sufficient conditions.

. .. b

~ ... i .

..

Rt•ldenct Time (d)

+ NH4 -N present in the feed yeast cell suspension (left hand bar

of each pair) and in the bioreactor (right hand bar) as a

function of residence time under a) oxygen sufficient conditions

and b) low oxygen concentrations.

-132-

CHAPTER 10

AEROBIC THERMOPHILIC BIODEGRADATION

OF MICROBIAL CELLS.:

2. SOME EFFECTS OF TEMPERATURE

Summary

The effect of temperature on solubilization and biodeg-radation in a laboratory scale bioreactor under aerobic thermophilic operating conditions was assessed using a substrate consisting of whole yeast cells and inorganic nutrients. Under oxygen limited conditions, consistently better results were obtained for operation at 65°c, reflecting the true thermophilic nature of the process microbes. An operating temperature of 70°c probably exceeded the optimum for effective functioning of the thermophilic microbes and resulted in a less efficient process, whilst an operating temperature of 6o0 c was intermediate with respect to its effectiveness.

-133-

Introduction

i During recent years, the process of aerobic thermophitl.ic waste sludge

treatment has been introduced at several municipal sewaf e and industrial

wastewater treatment plants in Austria, Swi t.:erland nd West Germany

prior to conventional anaerobic treatment. Such pre-tr atment processes

offer both hygienization, i.e., the destruction of pat ogenic organisms

pres.ent in the feed, and solubilization, i.e., partiall degradation of

some of the organic particulate material in the s1u9ge. In addition,

some pollutants sorbed on primary sludge which are essentially

recalcitrant under anaerobic treatment conditons are b1' degraded.

Thermophilic operation can be economically achieved from heat-energy

generation by the process microbes during aerobic bi egradation. Only

minor external heat input is required provided that possibilities for

effective heat exchange between processed and Jaw sludges are

incorporated and heat loss to the surroundings is min~mized (Hamer and

Zwiefelhofer, 1986).

Whilst some aerobic thermophilic treatment plants have been reported to

operate between 44°c and 55°c (Gould and Drnevich, !1978; Jewell and

Kabrick, 1980; Wolinski, 1985) temperatures between ~o0c and 10°c are i

also used (Zwiefelhofer, 1985). At these higher teml;>eratures, truely

thermophilic process microbes, i.e., microbes with ph~siological optima

at elevated temperatures, as opposed to the thermot,lerant mesophiles

which are physiologically adapted for survival only al temperatures up

to 55°c, above which they are unlikely to be found, arj able to develop

and assert their potential.

The object of this investigation was to determine the temperature

optimum for the solubilization and degradation of whol+ microbial cells,

as a basis for rational design of microbial solit destruction in

aerobic thermophilic waste sludge pre-treatment procesres.

-134-

Materials and Methods

Bioreactors and operating conditions: Two laboratory scale bioreactors

with operating volumes of 6 litres (Chemap At;;, Volketswil, CH) and 9 1

{Bioengineering At;;, Wald, CH) each with extensive measurement

instrumentation, i.e., pH, dissolved oxygen concentration, temperature

and impellar speed monitoring, were used for these experiments. The

bioreactors were operated oxygen limited by continuously sparging air at

21 and 31,5 1 h-l respectively. The pH was maintained constant at 7.0 by

controlled addition of a mixture of IM KOH and IM NaOH.

Aerobic thermophilic culture and feed. Aerobic thermophilic bacteria

were obtained from an operating municipal waste sewage aerobic

thermophilic digestor {OTB, Omwelttechnik Buchs AG, Buchs/SG, CH).

Pressed bakers yeast, as the sole biodegradable carbon source was -1 suspended in a nutrient solution containing:- NH4cl, 4 g 1 ; KH2Po_4 ,

-1 -1 -1 l 8 g 1 I K2HP04, 5.7 g 1 I zno, 3.26 mg 1 I FeC136H20' 43.2 mg 1 I

MnC12 4 H20, 16 mg 1-1 ; CuC12 2H20, 1.36 mg 11 ; CoC126H20, 3.8 mg 1-11; H

3B0

4, 0.5 mg 1-l MgC1 2 , 0.2 g 1-1 . The bioreactor feed was stored at

4°c for no longer than 5 days.

Analytical.

Dry weight: 5 ml samples were centrifuged at 30,000 x g and the

supernatant removed. The solids were resuspended in water, poured into

tared crucibles and heated overnight at 105°c. They were re-weighed

after cooling in a desiccator.

Ash/volatile suspended solids: After drying overnight, the crucibles

used to determine dry weight were placed in an oven at 60o0 c for ca. 5

h, allowed to cool in a desiccator and then re-weighed.

Dissolved Organic carbon: Appropriate dilutions of the supernatant were

made and the DOC assayed directly with a TOCOR 2 total organic carbon

analyser (Maihak AG, Hamburg, D) • Inorganic carbon was removed by

-135-

acidification using concentrated HCl, and the excess co2 removed by

sparging with N2 for 12 minutes.

+ NH4 ~: This was measured in the supernatants using an automated

nitrogen analyser (Skalar, Breda, NL).

Carboxylic Acids: Low molecular weight carboxylic acids were analysed by

acidifying the supernatants with formic acid and injecting into a

Shimadzu GC-RlA gas chromatograph (Shimadzu, Kyoto, J).

Chemicals: All chemicals were of analytical grade and supplied by either

Fluka (Buchs/SG, CH) or Merck (Darmstadt, D).

Results and Discussion

Three operating temperatures, 60°c, 65°c and 7o0 c, were selected for

examination of the solubilization and biodegradation of whole microbial

cells. The reason for this selection was that autothermal operation

could be used to achieve such temperatures during technical scale

operation and that such temperatures would allow rapid and effective

hygienization. Bioreactor residence times were varied between 1.5 and 5

days at each temperature with the exception of 70°c, where only 1.5 and

3.0 day residence times were examined. The combined results are shown in

Figs. l to 6, where the various measured parameters, total suspended

solids (Fig. + 1). DOC (Fig. 2) and NH4 -N (Fig. 3) are shown in bar

charts together with the percentage solids removal (Figs. 4 and 5) and

the soldis removal rate (Fig. 6) obtained by calculation.

Consistently better results were achieved for operation at 65°c compared

with the other temperatures used. A residence time of 1.5 days at 70°c

resulted in very poor solubilization/biodegration of the microbial

cells, whilst extreme solubilization/biodegradation occurred with a 5

-136-

day residence time at 60°c. However, the solids removal rate under the

latter set of conditions was low when compared to removal rates at

shorter residence times. The maximum substrate removal rate (5.4 g 1-l

d-l) was achieved using 65°c and a 1.5 day residence time, a slightly

lower substrate removal rate resulted at the same residence time at 60°c

(4.9 g 1-l d- 1). Operation at 3 days and 65°c also resulted in a high

rate of substrate removal (4.0 g 1-l d-l).

+ NH4 -N accumulated under all conditions, although at short residence

times this was hardly detectable. Ammonia production is indicative of

deamination of the particulate nitrogen containing fraction of the + biomass and thus an increase in NH

4 -N content should reflect the extent

of solubilization. This process probably occurs slower than the overall

solubilization of the organic matter, hence the low level of + + accumulation of NH4 -N at short residence times. Some NH4 -N will also

be lost by stripping in the exhaust gases as a result of non-optimal

condenser operation and thus the measured concentrations can only be

considered indicative of effective solubilization.

An increase in the level of dissolved organic carbon was found under all

process operating conditions. This is the primary parameter indicative

of the extent of solubilization. Accumulation of dissolved organic

carbon occurs as a consequence of carboxylic acid production from the

oxygen limited metabolism of the process microbes. Table 1 shows the

amounts of the various acids produced. Acetate was consistently found in

the highest concentrations. The very low production of carboxylic acids

for 60°c and 3 day residence time indicates that during this set of

operating conditions oxygen was probably not limiting growth.

The solubilization/biodegradation of whole yeast cells has been shown to

be optimal under conditions of low dissolved oxygen (Mason et ~· 1986).

From the results reported here, operation at a temperature of 65°c would

also appear desirable for an effective process.

-138-

The rate limiting step for particulate substrate biodegradation is the

solubilization of the organic matter, a fact recognised by Eastman and

Ferguson (1981) for aerobic sludge treatment. The optimisation of

temperature could thus be only necessary for enhancing the capacities of

those process microbes producing the yeast cell wall hydrolysing

enzymes, with the remaining components of the mixed population adopting

the role of opportunists.

The physical effects of increasing temperature in bioreactors are also

important. Higher temperatures increase the vapour pressures,

diffusivity and ionization of many compounds, and it decreases viscosity

and gas solubility (Zeikus, 1979). The influence of temperature on

diffusion coefficients for gases was expressed by Scheibel (1954) as

D= T/Fp

where T, is the absolute temperature, P• the liquid viscosity and F, the

diffusion factor. Hamer and Bryers (1985) showed that the oxygen

diffusion coefficient in water increases markedly with increasing

temperature and the increase largely compensates for the reduction in

oxygen solubility with increasing temperatures. Thus, both the

physiological properties of microbes and the variability of the physical

properties of their environment dictate the optimum conditions for

effective process operation in systems of a heterogenous nature.

-139-

References

Brock TD (1969) Microbial growth under extreme conditions. In: Meadow P and Pirt SJ (eds) Microbial growth. Proceedings 19th Symp Soc Gen Microbial. Cambridge University Press, Cambride, p 15.

Eastman JA and Ferguson JF (1981) Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. J. Wat Poll Cont Fed 53: 352-366.

Gould MS and Drnevich RF (1978) Autothermal thermophilic aerobic digestion. Env Engng Div Proc of Am Soc Civ Engnrs 104: 259-270.

Hamer G and Bryers JD some technological 267-284.

(1985) Aerobic thermophilic <;ludge treatment -concepts. Proc IWTUS-3 Conserv and Recyc 8:

Hamer G and Zwiefelhofer HP (1986) Aerobic thermophilic hygienization -a supplement to anaerobic mesophilic waste sludge digestion. IChemE symposium series 96: 163-180.

Jewell WJ and Kabrik RM (1980) Autoheated aerobic thermophilic digestion with aeration. J Wat Poll Cont Fed 52: 512-523.

Mason CA, Hamer G, Fleischmann Th and Lang C (1986) Aerobic thermophilic biodegradation of microbial cells: 1. Some effects of dissolved oxygen concentration. Appl Microbial Biotechnol.

Scheibel EG (1954) Liquid Diffusivities. lndustr Engng Chem 46: 2007-2008.

Sonnleitner B and Fiechter A (1983) Advantages of using thermophiles in biotechnological processes: expectations and reality. Trends in Biotechnol 1:74-80.

Wolinski WK (1985) Aerobic thermophilic sludge stabilisation using air. Wat Poll Cont 84: 433-445.

Zeikus JG (1979) Thermophilic bacteria: ecology, physiology and technology. Enzyme Microb Technol 1:243-252.

Zwiefelhofer HP (1985) Aerobic-thermophilic/anaerobic mesophilic two stage sewage sludge treatment: practical experiences in Switzerland. Proc IWTUS-3. Conserv and Recyc 8: 285-302.

Temp Res.Time Acetate Propionate Isa-Butyrate Butyrate 2Me-Butyrate Iso-valerate

(OC) (d) -1 (mg l ) -1 (mg 1 ) -1 (mg l ) -1 (mg l ) -1 (mg 1 ) (mg -1 l )

60 1.5 2365 322 83 130 19 72

60 3.0 00 27 23 22 NFa NF

60 5.0 2700 347 84 138 NF NF

65 1.5 2813 846 158 164 79 130

65 3.0 1552 419 208 160 96 144

65 5.0 1300 521 183 144 85 235

70 1.5 1121 181 61 28 25 31

70 3.0 2869 410 168 44 53 47

(a - none found)

Tab1e 1. Carboxylic acid production during oxygen limited aerobic thermophilic digestion of

whole yeast cells as a function of temperature and bioreactor residence time.

Valerate -1 (mg l )

22

NF

NF

87

42 I

24 ..... ... 0 I

23

7

••

.. i :g

~ 30

i ... & IO

~ !!! ~

10

L , s

a

Figure 1.

a

Figure 2.

-141-

b c

Residence Time (d) Residence Time ldl

Total suspended solids concentration in feed (left hand bar

of each pair and in the bioreactor during aerobic thermophilic

biodegradation at a) 60°c, b) 65°c, c) 10°c.

b c

Dissolved organic carbon in feed yeast cells suspension (left

hand bar of each pair) and after solubilization/biodegradation

(right h~nd bar) at a) 60°c, b) 65°c, and c) 10°c.

a

..

Figure 3.

a

Figure 4.

-142-

b c

Residence Time (di Residence Time (di

+ NH4

- N concentration in the feed (left hand bar of each pair)

and after aerobic thermophilic biotreatment (right hand bar) at

a) Go0 c, bl 65°c, c) 10°c.

b c

Residence Time (di Rtsldenco Time (di Resi"- Time (di

Total suspended solids removal rate as a function of both residence

time and temperature during aerobic thermophilic treatment at

low oxygen concentrations. a) 60°c, b) 65°c, c) 10°c.

.. a i i Bo"° " ~ ~ •o ~ i ]~ .!!IO "' . ~

i·· ~ ~ 0

Figure 5.

00 a " .. " f .. ii5

i .. i l~ ~ !I .. ~ • J i 10

l

Figure 6.

-143-

b c

Residence Time (d) Residence Time (d) Residence Time (d)

Volatile suspended solids degraded as a percentage of the influent

(feed) volatile suspended solids concentration following

aerobic thermophilic biotreatment at a) 60°c, b) 65°c, c) 10°c.

b c

Reslda.C. Time (d) Residence Time (d)

Amount of total suspended solids solubilised/biodegraded

as a percentage of the feed yeast cell concentration during

aerobic thermophilic biotreatment at a) 60°c, b) 65°c and

c) 10°c.

-144-

CHAPTER 11

FUNDAMENTAL ASPECTS OF WASTE SEWAGE SLUDGE TREATMENT:

Abstract

MICROBIAL SOLIDS BIODEGRAOATION IN AN AEROBIC THERMOPHILIC SEMI-CONTINUOUS SYSTEM

The solubilisation and biodegradation of whole microbial cells by an aerobic thermophilic microbial population was investigated over a 72 h period. Various parameters were followed inlcuding total suspended solids reduction, changes in the dissolved organic carbon, protein and carbohydrate concentrations, and carboxylic acid production and utilisation. From the rates of removal of the various fractions a simple model for the biodegradation processes is proposed and verified with respect to acetic acid production and utilisation, and total suspended solids removal. The process is initiated by enzymic degradation of the substrate microbe cell walls followed by growth on the released soluble substrates at low dissolved oxygen concentration with concommitant carboxylic acid production. Subsequent utilization of the unbranched, lower molecular weight carboxylic acids allows additional energy supply following exhaustion of the easily utilisable soluble substrate from microbial cell hydrolysis.

-145-

Introduction

The enforcement , of increasingly stringent environmental legislation in

many countries has resulted in increased capacity for both municipal

sewage and industrial wastewater treatment by combinations of

mechanical, biological and physico-chemical process technology. The

major result of this situation is an ever increasing quantity of waste

sludge, which presents a serious ultimate disposal problem. Policies

which allow the dumping/spreading of untreated sludge both on land and

at sea are increasingly subject to criticism, necessitating the devel-

opment of effective sludge treatment technologies, Effective sludge

treatment requires the stabilization of biodegradable matter in the

sludge and the removal of potentially pathogenic organisms and toxic

chemicals from the sludge. Conventional waste sludge treatment involves

mesophilic anaerobic digestion, a technology which poses questions with

respect to its effectiveness for the removal of pathogens and some toxic

chemicals (1) • Microbes, including pathogens originally present in raw

sewage and wastewater, are a major constituent of waste sludge and the

biodegradation of such particulate matter is the key to effective

stabilization and/or hygienization. Aerobic thermophilic processes have

been proposed either as a pre-treatment stage prior to conventional

anaerobic mesophilic digestion (2) or as a complete sludge treatment

(3).

The work reported here concerns the process biology of a semi-con-

tinuous, laboratory-scale bioreactor in which microbial solids are

undergoing solubilization and biodegradation under aerobic thermophilic

conditions. In order to have a standardized feedstock, pressed bakers

yeast was used.

-146-

Theoretical

The dissolution of solids in agitated batch reactors has, since the

theoretical analysis (4) and experimental verification thereof (5) by

Hixson and Crowell, been considered to follow the cube root law. In the

case of cellulose particles, which are solubilized by enzymic

hydrolysis, Humphrey and co-workers (6, 7 ,8) have proposed a complex

shrinking-site model, which incorporates both inhibition and repression,

but ultima~ely predicts modified cube root law type dissolution. In the

case of the enzymic hydrolysis of microbial cells prior to their

utilization as carbon energy substrates by the microbes responsible for

their hydrolysis, the cube root law is inapplicable, because here the

enzymic hydrolysis involves either puncturing or bursting of the

substrate microbe cell walls, processes that depend on point attack and

point strength of the substrate microbes.

For purposes of modelling the system under investigation here, the

following processes are assumed to be occuring:

1. A feed cell population consisting of intact yeast cells is added to

a culture containing thermophilic process microbes.

2. The thermophilic process microbes produce extracellular enzymes which

are capable of cleaving the cell wall of the yeast. Enzymes attach to

the yeast cells in a non-specific distribution over the entire surface.

3. Cell wall lysis occurs at random locations on the yeast cells

resulting in release of the soluble cytoplasmic components of the

substrate particulates. Lysis results after a minimum of one site

cleavage in the wall but may require multiple site cleavage before the

wall is sufficiently weak to lyse.

4. Thermophilic process microbes utilise the released soluble substrates

as carbon and energy source and in so doing produce acetate as a result

of the low dissolved oxygen concentration. Further lysis of remainin9

-147-

whole yeast cells supplies a constant source of soluble nutrients for

the thermophilic process microbes. Enzymic degradation of the yeast cell

wall polymers supplements this supply.

5. After a significant period, the rate of supply of soluble substrates

from yeast cell lysis slows down and the thermophilic microbes begin to

utilize the now accumulated acetate to support growth and energy

requirements. Additionally, any remaining wall particulates are

hydrolysed to soluble energy substrates.

Mathematical description

The mathematics of this process are based on the following molar

stoichiometry.

Yeast Cells

Acetate

Thermophilic microbes

Thermophilic microbes

Acetate

The elemental composition of yeast cells is assumed to be the same as

that for thermophlic microbes and is expressed above on a nitrogen and

ash free basis. For the gramme yield coefficients a molecular weight of

21 g was assumed for l mole of cells inclusive of nitrogen and ash. Thus

from the above coefficients, yield values (g/g) of thermophlilic

microbes and acetate from yeast cells would be 0.3 and 0.52,

respectively. In order to describe growth of thermophilic microbes,

Monad kinetics were assumed for growth on yeast cells and Monod kinetics

with an inhibition term for growth on acetate, i.e.,

(2)

f (SS) µmax (Ss} • Ss Ks +Ss SS

µ(Ac) ~ Pmax(Ac) ' Ac.Ki (Ss+Ki) (Ac+Ksl\c)

-148-

The nomenclature is defined in Table 4.

(3)

(4)

The remaining rate expressions i.e., for yeast cell lysis, for process

microbe death/lysis, and for particulate hydrolysis, have been assumed

to be first order due to lack of information regarding the kinetics of

such processes. In the case of the particle hydrolysis rate constant the

use of first order kinetics can be justified for reasons discussed by

Eastman and Ferguson (9).

From this basic consideration, a stoichiometric matrix can be

constructed for the five parameters, x p (process microbes) x s (feed

yeast cells), Ss (soluble substrate concentration from lysed yeast

cells), P (particulate materials released from cells as a result of

lysis, e.g., cell wall fractions) and Ac (acetate) The stoichiometric

matrix is shown in Table 1.

From this, the following material balances can be described:

X : dX (5) p ____£.

dt

X : dX (6) s __ s_ dt

Ss: (7)

P: dP

dt

Ac: dAc

-149-

= Y .µ(Ss).X - p(Ac) .X Ac/Ss p p -y~~ y

Xp/Xs Xp/Ac

Materials and Methods

(8)

(9)

Bioreactor and operating conditions. A l.5 1 bioreactor (Bioengineering

AG, Wald, CH) with full measurement instrumentation, i.e. pH, dissolved

oxygen concentration, temperature, impellar speed monitoring, was used

for the experiments. Air (6 l/h) was sparged into the bioreactor which

was operated with an impellar speed of 900 rpm. The operating

temperature was maintained constant at 60°c and the working volume was

1300 ml. The pH was not controlled.

Aerobic thermophilic culture and feed. Aerobic thermophilic microbes

were obtained from an operating municipal waste sludge thermophilic

aerobic digester (UTB, Umwelttechnik Buchs AG, Buchs/SG, CH). Pressed

bakers yeast (40g/ll as the sole biodegradable carbon and nitrogen

source was suspended in a nutrient solution containinq KH2PO 4 , 8 g/l 1

K2HP04 , 5.7 g/l; ZnO, 3.26 mg/l; FeC1 36H20, 43.2 mg/l; MnC124H20, 16

mq/l; Cucl 22H20, 1.36 mg/11 CoC12 6H20 3.8 mq/l; H3B04, 0.5 mg/l; MgC1 2 0.2 g/l. The stock solution was stored at 4°C for no lonqer than 5 days.

Start-up was carried out by inoculating 500 ml aerobic thermophilic

sludqe into 800 ml of preheated (60°ci yeast suspension. The reactor was

operated on a fill and draw cycle of 50\ withdrawl/50' addition. Yeast

addition was carried out slowly so as not to cause a drop in

temperature. A cycle time of 3 days was operated during the period

immediately preceeding the experiment reported here.

-150-

Analytical

Dry weight. 5 ml samples were centrifuged at 30,000 x g and the

supernatant removed. The solids were resuspended in water, poured into

tared crucibles and left overnight at lo5°c. They were reweighed after

cooling in a dessicator.

!!!!!· After drying overnight, the crucibles used to determine dry weight

were placfd in an oven at 60o0 c for at least 5 h, allowed to cool and

then reweighed.

Dissolved organic carbon (DOC). Appropriate dilutions of the supernatant

were made and the DOC assayed directly on a TOCOR total organic carbon

analyser (Maihak AG, Hamburg, D). Inorganic carbon was removed by

acidification and sparging with nitrogen.

Yeast cell numbers were assayed by direct microscopic enumeration in a

haemocytometer.

Microbial activity was assayed using 2-(p-iodophenyl)- 3-{p-nitro-

phenyl) -5-phenyltetrazoliumchloride (INT) . 1 ml INT ( 2. o g/l aquaeous

solution) was added to a 5 ml sample immediately after removal from the

bioreactor and placed in a shaking water bath at 6o0 c. After 20 min

incubation, l ml 37'11 formaldehyde was added to stop the reaction. The

solution was centrifuged at 10,000 x g and the solids resuspended in 10

ml tetrachloroethylene (40 v'il)/ acetone (60 v'il) solution. After 30 min

incubation in the dark, the absorbtion of the extracted formazan was

measured at 490 nm.

Protein in cells and in the extracellular medium was measured using the

Biuret method with bovine serum albumin as the standard.

carbohydrate concentration in cells and in the extracellular medium was

measured using the anthrone method with glucose as the standard

according to Herbert !! !:!..!.· (10).

-151-

~ -NH4 -N, N02-N and No3 .:! were measured in the extracellular medium using

an automated nitrogen analyser (Skalar Analytical, Breda, NL).

Carboxylic acids. The extracellular medium was assayed for low molecular

weight carboxylic acids by acidifying using formic acid and injecting

into a gas chromatograph (Shimadzu Corp., Kyoto, J). The column was a GP

carbopack c/o 3\ carbowax 20 m/O.l\ H3Po4 , 2 1 m x 2.6 mm int o glass

column. Column temperature-programmed 145°C/2min, Rate: s0 c;min

Detector, FID.

Scanning Electron Microscopy

50 ml samples were fixed in glutaraldehyde (4% v/v) washed with

distilled water and frozen in a thin layer between two copper plates

according to the method of Miiller !! al (11). They were then fractured

at -1os0 c and dried at -ao0 c for 2 h in a Balzers BAF 300 freeze etching

device (Balzers AG, Lichtenstein) as described by Walther !! !! (12).

The specimens were rotary shadowed (5nm) with carbon and overlayed with

platinum carbon (5nm) and viewed in a Hitachi s-700 scanning electron

microscope (Nissai, Sangyo, J). As a consequence of this preparation

method some fracturing of surface cells occurred.

Chemicals. All chemicals used were of analytical grade and supplied by

either Fluka AG, (Buchs/SG, CH) or Merck AG (Darmstadt, D).

In this study, the biodegradation of particulate microbial solids by a

mixed thermophilic aerobic microbial culture was investigated. This

step, typical of the rate-limiting step in waste sludge biodegradation,

was followed in the semi-continuous operation of the process where whole

yeast cells were used as a stand<1rdized. substitute for the complex

biopolymeric particulates present in actual waste sludge feeds. The

thermophilic microbes in the bioreactor were preconditioned by

maintaining the same operating conditions for the immediate period prior

-152-

to the start of this investigation. At the start of the experiment 50\

of the reactor contents (650 ml) were removed and 650 ml of a yeast

suspension in the phosphate and trace element solution were added. The

addition was sufficiently slow so as to prevent any decrease in

temperature of the bioreactor. The temperature during the entire

experimental period, including the change over, was controlled at, and

never deviated from, 60°c. The pH was not controlled during the

experiment and remained constant for the first 24 hat pH 5.9 (Fig. 1).

However, after 24 h the pH increased and attained a final value of 7.68

after 72 h. Oxygen was supplied to the bioreactor in air but at a

sufficiently slow flow rate as to achieve a low dissolved oxygen

concentration. The amount of dissolved oxygen in the bioreactor (\

saturation, where saturation at was 3.03 g/l) during the

biodegradation process is shown in Figure l. The dissolved oxygen

concentration was initially very low (2\) and decreased further to an

undetectable level between 18 and 26.5 h. There followed a slight rise

to l\ after 48 h and a levelling out at 4% after 65 h.

Biodegradation was followed with respect to the decrease in total and

volatile suspended solids and the results are shown in Fig. 1. The total

suspended solid removal rates are shown in Table 2. The rate

approximately doubles after 21 h operation to 0.64 g/l.h, subsequently

decreasing to a lower value after 48 h. The quantity of solids removed

at the end of the experiment represents 77\ of the feed concentration

(41.4 g). Concommitant to the decrease in total and volatile suspended

solids, there was an increase in the amount of dissolved organic carbon

(DOC). This reached a plateau value of ca. 5.8 g/l between 24 and 48 h

(see Fig. 1) after which time the level declined.

Since the solids were being biodegraded at a fast rate, especially

during the priod between 18 and 26.5 h, it was of interest to look at

the fate of the particulate feed cells .f!!. ~ to see whether either

whole cells were being biodegraded or cell lysis was occurring leaving

behind partially intact ghosts composed primarily of cell walls. This

was conducted by carrying out light microscope direct counts of the

-153-

yeast cells. The results of these counts are shown in Fig. 2. Noticeable is the very sharp decline in yeast cell numbers during the lB to 26.5 h

period, paralleling the decrease in total and volatile suspended solids shown in Fig. 1. This decrease in whole yeast cell numbers occurred at a

high rate (Table 3) particularly between 18 and 26.5 h where rates 3.5 to 7 x the initial rate were observed. By the end of the experiment, 98\

of the whole yeast cell concentration in the feed had been either partially or totally biodegraded (Table 3) . This compares to a residual

total suspended solids concentration equivalent to 23\ of the substrate solids. This difference can be accounted for by the presence of

particulate lysed cell fractions and by the growth of the thermophilic aerobic microbes. Due to their size and thus ease of direct microscopic

quantification, yeast cell numbers are easily determined. On the other hand, bacterial cell numbers are not so easily determined due to their

morphological diversity and the imposition of a spectrum of nutrient specificities and temperature requriements for accurate quantification

(13). Therefore, a method was used which followed the activity of the

bacteria which, whilst being an indirect means of quantification,

nevertheless provides the more important information as to what is happening w,ithin the bioreactor. As shown in Fig. 2, the activity rises

significantly between 18 and 26.5 h and remains at or above this maximum value between 26.5 and 48 h until the yeast cells have been almost

totally biodegraded. The start of the rise in thermophilic microbial activity occurs simultaneous with the increases in the rate of solids removal and in pH (Fig. ll. The decline in activity after 48 h is reflected in the almost insignificant further removal of solids after

this time and is also accompanied by a decrease in dissolved organic carbon and an increase in dissolved oxygen concentration.

Direct examination of the biodegradation of yeast cells was carried out using scanning electron microscopy (Fig. 3). It should be noted that the

micrographs present qualitative rather than quantitative information as

a result of the concentration effects in their preparation procedure. Fig. 3(a) shows the substrate for the biodegradation process. This is a

discretely dispersed suspension of yeast cells with negligable bacterial

-154-

contamination. After introduction into the bioreactor, the cells are

mixed with the thermophilic aerobic microbes (Fig. 3b). The ensuing

degradation of yeast cells and growth of process microbes is shown in

Fig. 3 c-e. These electron micrographs indicate that a variety of

morphological types of microbes are present. Direct evidence for lytic

effects are difficult to discern with any degreee of confidence, but

whilst there is some suggestion that lysis is occurring (see Fig. 3e)

this implies that certainly some of the yeasts seen under the light

microscoE'Ej may be ghosts rather than intact cells. As time progresses in

the biodegradation process the yeast cells appear much more irregular

and shrunken.

If lysis is occurring it is unlikely to be as a result of direct contact

between the process microbes and the yeast cells: more likely is the

release of lytic enzymes from the thermophilic aerobic process microbes.

Therefore, it was of interest to examine the variation in the levels of

cellular and extracellular protein (Fig. 4). Cellular protein levels

(i.e. protein contained in both thermophilic process microbes and yeast

cells) decreases constantly during the course of the experiment, whilst

the level of extracellular protein remains effectively constant and even

exceeds the intracellular concentration after 72 h. The extracellular

protein was not differentiated into structural and enzymic fractions and

therefore direct evidence of enzymic activity was not obtained. However,

as the level of cellular protein decreases, - -

+ the amount of NH4 -N

released increases (Fig. 4). Both N02 -N and N03 -N concentrations were

below detection limits (1 mg/l).

The cell wall of bakers yeast (Saccharomyces cerevisiae) is composed of

approximately equal quantities of glucan and mannan, both of which are

carbohydrates which can often account for about 85\ of the dry weight of

the wall (14). Fig. 5 showe the levels of cellular and extra-cellular

carbohydrate during the biodegradation of yeast cells. In contrast to

the suspended solids and whole yeast cell numbers, the bulk of the

carbohydrate is degraded in the first 18 h (49\), with very little

activity during the 18 to 26.5 h period. Between 26.5 and 48 h further

-155-

degradation occurs resulting in a final carbohydrate concentration

equivalent to only 9' of the initial. Analogous to the extracellular

protein level, the extracellular carbohydrate remained effectively

constant during the time course.

In order to identify some of the components of the dissolved organic

carbon fraction, assays of low molecular weight carboxylic acids were

carried out during the biodegradation process. The results, shown in

Fig. 6, indicate that most of the DOC was in the form of acetate, which

reached a maximum concentration of 5.9 g/l after 26.5 h. From the data

shown in Fig. 6 it appears that the lower the molecular weight and the

less branched the molecule the earlier both production and

biodegradation occurs. Propionate also peaked at 26.5 h but at a lower

concentration (0.88 g/l). Similarly, butyrate was produced at a still

lower concentration and probably reached a maximum between 24 and 48 h,

but like acetate and propionate was fully biodegraded after 72 h. Both

iso-butyric and iso-valeric acids were present in significant quantities

after 72 h.

Discussion

The foregoing results clearly indicate the apparent complexity of the

process under consideration. They suggest a sequence of events:

1. Initial degradation of cell wall polymers resulting in the release

of soluble cell components, thus the initial slow disappearence in

whole yeast cells but fast rate of carbohydrate biodegradation.

2. Accumulation of carbo><ylic acids, particularly acetic acid, as a

result of the low dissolved oxygen concentration, and thus an

increase in the dissolved organic carbon concentration.

•156-

3. Once suitable soluble substrates become available, the process

thermophilic microbe activity is enhanced and soluble substrate is

preferentially utilized.

4. Exhaustion of the preferred substrate followed by utilisation of low

molecular weight carboxylic acids.

5. Hydrolysis of remaining cell wall fragments and utilisation of

higher molecular weight carboxylic acids.

If one simplifies this sequence of events by omitting the production of

all carboxylic acids other than acetic acid, it should be possible to

describe the process by the model presented earlier, provided

appropriate values of the constants and yield coefficients in the

material balances are selected. On the basis of the values shown in

Table 4, the predictions resulting from the model are shown in Figures 7

and 8 as complete lines. Experimental data where available, are

superimposed as points in these figures. As can be seen, a relatively

good fit is achieved, indicating that the simple model proposed is

capable of describing the overall process provided a slight degree of

simplification is introduced.

The tendancy for those carboxylic acids with higher molecular weights

than acetic acid to accumulate and subsequently to only degrade towards

the end of the batch cycle could be a disadvantage for a sludge

treatment process if used in isolation and without subsequent anaerobic

stabilization, because the presence of such carboxylic acids would

impart malodour to both treated sludge and associated supernatent.

-157-

Conclusions

Introduction of ·new technologies to achieve more efficient process

operation require a sound understanding of the fundamental aspects

involved. Process optimisation of the thermophilic aerobic sludge

treatment technology is now possible knowing the sequence of events involved in microbial solids destruction described here.

References

1. McEvoy, J.; Giger, W.: Determination of linear alkylbenzene-sulfonates in sewage sludge by high-resolution gas chromatography/mass spectrometry. Environ. Sci. Technol. 20 (1986) 376-383.

2. Hamer, G.; Zwiefelhofer, H.P.: Aerobic thermophilic hygienization -a supplement to anaerobic mesophilic waste sludge digestion. Inst. Chem. Eng. Symposium series 96 (1986) 163-180.

3. Smith, J.E. Jr.; Young, K.W.; Dean, R.B.: Biological oxidation and disinfection of sludge. Wat. Res. 9 (1975) 17-24.

4. Hixson, A.W.; Crowell, J.H.: Dependence of reaction velocity surface and agitation: I Theoretical consideration. Engng. Chem. 23 (1931) 923-931.

upon Ind.

5. Hixson, A.W.; Crowell, J ,H.: Dependence of reaction velocity upon surface and agitation: II - Experimental procedure in study of surface. Ind. Engng. Chem. 23 (1931) 1002-1009.

6. Humphrey, A.E.; Armiger, W.B.; Zabriskie, D.W.; Lee, S.E. 1 Moreira, A.; Joly, G.: Utilization of waste cellulose for the production of single cell protein. In: Continuous Culture 6: Applications and New Fields (A.C.R. Dean, D.C. Ellwood, C.G.T. Evans and J. Melling, Eds). (1976) 85-99.

7. Humphrey, A.E.; Moreira, A.1 Armiger, W.; Zabriskie, D.: Production of single cell protein from cellulose wastes. Biotechnol. Bioeng. Symp. 7 (1977) 45-64.

8. Moreira, A.R.; Phillips, J.A.; Humphrey, A.E.: Utilization of carbohydrates by Thermanospora Sp. grown on glucose, cellobiose and cellulose. Biotechnol. Bioengng. 23 (1981) 1325-1338.

-158-

9. Eastman, J.A.; Ferguson, J.F.: Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. J. Wat. Poll. Cont. Fed. 53 (1981) 352-366.

10. Herbert, D. 1 Phipps, P.J; Strange, R.E.: Chemical analyisis of microbial cells. In: Methods in Microbiology 7B (J.R. Norris and D.W. Ribbons, eds) (1971) 209-344.

11. Mliller, M.; Meister, N.; Moor, H.: Freezing in a propane jet and its application in freeze-fracturing. Mikroskopie (Wien) 36 (1980) 129-140.

12. Walther, P.; Mliller, M.; Schweingruber, M.E.: The ultrastructure of the cell surface and plasma membrane of exponential and stationary phase cells of Schizosaccharomyces pombe, grown in different media. Arch. Microbiol. 137 (1980) 128-134.

13. Mason, C.A.; Hamer, G.; Bryers, J. D.: The death and lysis of microorganisms in environmental processes. FEMS Microbiology Reviews 39 (1986) 373-401.

14. Rose, A.H.: Chemical Microbiology, 3rd Edn. (1976) Butterworths, London.

-159-

~ Stoichiornet r ic matrix for aerobic thermophilic biodegredat ion process.

xP x. Ss p Ac

~ Reaction Rate

l. Substrate Cell 0 -1 Y Ss/Xs YP/Xs KL.XS Lysia

2. Porticle 0 0 +l ·l Kh.P Hydrolysis

J. Growth on +l 0 -1 0 •Y Ac/SS ~(Ss).Xp

Lysis Products YXp/Ss YXp/Ss

4. Process Microbe -1 0 +YSs/Xp +YP/Xp Kd.Xp Oeath/Lysis

s. Growth on +l 0 0 -1 µ(Ac) .xP Acetate Yxp/Ac

Note:- As the stoichiometry is incomplete, i.e., co2 and o2 are not considered, the matrix

is not balanced. The symbols used are explained in Table 4.

~ Solids removal data.

Time after addition

of substrate cells

lh)

18

21

24

2(l.5

48

72

-160-

Solids removal

rate

lq/l.h)

0. 33

0. 38

0. 64

o. 64

0. 20

0. OJ

Notes l. Solids concentration at time Oh

Solids removed

\t 1 ll \feed 0

concentration

2712)

26 46

31 50

39 56

46 61

65 74

68 77

2. Reduction of solids due to dilution of feed by bioreactor contents

Table J. Yeast cell removal data.

'I'ime after addition Whole yeaHt cell Yeast cells Removed

of substrate cells removal rate

lh) 1106/1.h) \t ill % feed 0

28 12 )

18 1. 63 22 44

21 23. 02 33 52

24 50. 54 57 69

26. 5 lJ. 23 63 13

48 6. 38 85 89

72 o. 68 98 98

Notes:- I.Yeast cell number at time Oh

2. Reduct ion of yeast ce 11 number due to dilution of feed by bioreactor

contents.

-161-Table 4. Values for yield coefficients and kinetic constants used for

calculation using the mathematical model

Stoichiometric Coefficient or Kinetic Constant

Description

YSs/Xp

Pmax(Ss)

Yield process microbes on substrate yeast cells

Yield process microbes on acetate

Yield acetate produced by process microbes on substrate yeast cells growing

Yield soluble substrate from lysis of yeast cells

Yield soluble substrate from lysis of process microbes

Yield particulates from lysis of yeast cells

Yield particulates from,.lysis of process microbes

Maximum specific growth rate constant for ,growth of process microbes on soluble substrate

Maximum specific growth rate constant for growth of process microbes on acetate

Value

0.30

0.17

0.52

0.80

0.80

0.20

0.20

0.60

0.60

Saturation coefficient for growth of process 0.80 microbes on soluble substrate

K. 1

Saturation coefficient for growth of process microbes on acetate

Death/lysis rate constant for process microbes

Inhibition constant for growth of process microbes on acetate

Lysis rate constant for whole yeast cells

Hydrolysis rate constant for particulate biomass

0.01

0.02

0.00012

0.04

0.065

Units

g/g

g/g

g/g

g/g

g/g

g/g

g/g

g/l

g/l

-1 h

g/1

-1 h

-"' 35 8

30 7.5 <I)

0 ::::; 5l 25

0 w 0 z UJ 0.. <I)

20 7

iil 15

~ ~ g 0 z <

6.5

10

6 5

0 5.5

-162-

.... ~·p···········

0 24 Tl ME (HOURS)

46

..

z 0 I-< 0:: :::J

10 6

5 -"' z 0

4~ "" u u

':;;: 5 3 z <I) "" I.!) 02 /, ........ o ;;fl 0::

0

72

N 0

2

0 0

0 UJ > --' 0 <I) <I)

0

Firo>re 1. Changes in the concentration of total suspended solids (TSSJ, volatile suspended solids (VSS), dissolved organic carbon (DOC), pH and dissolved oxygen concentration during semi-continuous aerobic thermophilic biodegradation of microbial solids with a cycle time of 3 days.

10 2.5

ACTIVITY 2.0 B ' e ...... , "' ----..... E

!2 ........... c:

" 0

" 6 1.5 "' \ ~

0:: ~ z w I 0 IXl \ :E I-:::J 4 1.0 IXl z !>._ 0::

0 --' b....,..._,YEAST

<I) --' co UJ < u ........... .... ........... 0.5 "' ....... ....... < "O........, w >- .......... .........

0 ............ o-__

0 0 24 48 72

TIME {HOURS)

Figure 2. Changes in process thermophilic aerobic microbe activity and substrate yeast cell number during semi-continuous aerobic thermophilic biodegradation of microbial cells.

-163-

Figure 3a. Scanning electron micrograph of bakers yeast

cells used as the feed for the experimental

studies on the aerobic thermophilic treatment

process.

-164-

Figure 3b. Scanning electron micrograph of bakers yeast cells immediately after addition to a culture

containing aerobic thermophilic process bacteria

at 60°c.

-165-

Fi9ure Jc. Scanning electron micrograph of the aerobic

thermophilic solubilization/biodegradation

process 24 h after the addition of the substrate

microbes to the bioreactor.

-166-

Figure Jd. Scanning electron micrograph of the aerobic thermophilic solubilization/biodegradation process 48 h after the addition of the substrate microbes to the bioreactor. A significant quantity of cell debris can be seen, together with the various morphological forms of process thermophilic microbes present.

-167-

Figure Je. Scanning electron micrograph of the aerobic thermophilic solubilization/biodegradation process 72 h after the addition of the substrate microbes to the bioreactor. Most of the yeast cells now have the appearance of being "ghosts" whereby partially intact cell walls are still present but the contents have been lost by lysis.

-168-

Figure 3f. Scanning electron micrograph of the aerobic

thermophilic solubilization/biodegradation

process 144 h after the addition of the

substrate microbes to the bioreactor.

-169-

12 2.0

10

c:n

NH(·-~---·-"""··-... _, ........... .c ........... ~ 1.5

8 -z c:n 0 s 6 Cl: I-

1.0 z I

z w u z 4 0

.,,, :i:: z

u z 0.5

EXTRACELLULAR w

2 I-0 Cl: (l.

__)\ ....... o..... ./ \ .................. '--. /

---- j)' \ .............. ................. / --~ ~- '-<I"

0 -+-~~~~~~~~~~~~~~~~~~~~~~-+0 0

Fi51ure 4.

6

:::::-5 -... z 4 0 ;:: < Cl: I-z 3 w u z 0 u

w 2 !;;: a: 0 >-:i:: 0 ID a: < u

0

Figure 5.

24 48 72 TIME (HOURS)

Changes in cellular and ~xtracellular protein concentrations and of NH 4-N during semi-continuous aerobic therm~philic biodegradation of microbial solids. No NH 4-N was a~ded with the yeast cell substrate, released NH 4-N being derived from deamination of the particulate nitrogen containing fraction of the biomass.

,...,,,,...0--0...~ ,.,. '0--------/ ........ EXTRACELLAR

-~--------0----0

0 24 48 TIME (HOURS)

72

Changes in cellular and extracellular concentration of carbohydrates during the solubilization and biodegradation of whole yeast cells by aerobic thermophilic microbes.

Cl E

z 0 I-<{ a: I-z w u z 0 u

1000

750

500

250

-170-

.· ,•

•'

.. o· 0 .. --

.·

... ~----•• · ••• •• Isa- Butyrate

.c( er.er · . . ,n-·

.· o-... · *"'"

•·· *""

Butyrate

.· .·

.·· ,,, 2-Me-Butyrate -::J o··--··--··

.··

-o·· o--o··--··- .. --··--.. -- ·· o- Valer ate .

.·· 0 ..

6

5.::: Cl

z 0

4 i= <{ a: 1-z w

3 ~ 0 u

w 2 ':;:

1-w u <{

o~:.:;...~~~~o--<e>-<~·o-=·~-=·~~=·~:i=·~-=·..=.:·:.=::o-~·=-~·=-~·=.:.·=-~·~~~o 0 24 48 72

TIM E (HOURS)

Figure 6. Variation in the concentration of low-molecular

weight carboxylic acids during the biodegradation

of whole yeast cells by aerobic thermophilic

microbes.

25 35

-.!?]

20 .. >< 25~ If) "' If)

15 <( 20~ ::E

0 :::::; ffi 0

If) _J 15 Cl _J w 10 w u Cl w z I- 10~ <( a:: If)

I- 5 :::;)

If) If) [IJ

5 _J. :::;) If) ~

0 I-

0 0 0

-1 71-

12 24 36 TIME

...

48 ( h)

60 72

6

5

4~

w 3 I-<(

I-w u <(

2

0

Fi~re 7. Comparison of predicted values for acetate concentration, total suspended solids (TSS) and yeast cell biomass (X ) with experimental values (o, acetate; ,TSS; ,yeast cel~s)w

25

-"' 20

If) If) 15 <( ::E 0 [IJ

w 10 [IJ 0 a:: u ~

If) If)

5 w u 0 a:: a..

0

35

30 -.!?] 20

20~ :::::; 0 If)

15 Cl w

..... "',., ..........

~ ··-• .rss

'•, .. Xp

··. .......... .................

............ ............

10 ~ .......... -·2 ·-·-·- ·················· ~ /. ·-·-·-~-~ If)/ ·-·-·-5 _J ·-·-

0

~ 0 I- ,---

0 Ss -..,.---=---

12 24 -==.----'-;--':..- I I

36 48 60 72 TIME (h)

~ a:: 6 t:>

[IJ ::::> If)

5~ [IJ ::::> _J -

~~ 4 Cl~

z <( z -o 3~~

If) a:: If) I-<( z ::Ew

2 Q ~ CIJo u

w_ I- "' <( If) _J ~ ::::> u i== a::

o~

Fi~re 8. Predicted values for process microbe biomass(X ) and for particulate biomass (P). No comparison with exl:erimental values was made due to the lack of data. The contribution of these biomass fractions to the total suspended solids is also shown.

-1 72-

CHAPTER 12

BIOPARTICULATE SOLUBILIZATION AND BIODEGRADATION

IN SEMI-CONTINUOUS AEROBIC THERMOPHILIC DIGESTION

Abstract

The effects of charge size, added nitrogen on the degradation of microbial semi-continuous operation digestion process. Using

cycle time and the presence of extent of solubilization/bio-solids were investigated in of the aerobic thermophilic

a charge size of 50% of the bioreactor operating volume resulted in the removal of more than 59% of the feed total suspended solids concentration, and showed enhanced performance compared to 25% bioreactor volume charge size. The addition of a supplementary nitrogen source had pronounced effects on dissolved organic carbon concentrations, with the larger charge size but did not affect overall solids removal, whereas using the smaller charge size, addition of a supplementary nitrogen source resulted in ,improved process performance.

-173-

Introduction

Microbial growth theory and a significant amount of research in microbial biodegradation processes is only directly relevant to soluble

and/or miscible carbon energy substrates despite the fact that particulate biodegradable substrates are encountered in wastewaters and

waste sludges subjected to biotreatment and in many other situations. Waste sludge biodegradation processes require initial solubilisation of

the particulate matter present prior to biodegradation of the released soluble lysis products. Despite the fact that this constitutes the rate

limiting step in waste sludge biotreatment, relatively little research has been directed towards this aspect of the process.

Recently, several west European governments have shown concern about the

effectiveness of existing waste sludge treatment technologies, particularly with regard to the removal of pathogenic organisms and of

toxic chemicals. Sludge pre-treatment under aerobic thermophilic

operating conditions has been proposed and is being investigated as a

means of improving current technology, especially when such processes are operated in conjunction with conventional anaerobic mesophilic

sludge treatment. Aerobic operation in the first stage is preferable when toxic compounds which are essentially recalcitrant under anaerobic

conditions but which can be biodegraded under aerobic are present in waste sludges (McEvoy and Giger, 1986).

Several municipal aerobic thermophilic waste sludge treatment plants

operate in the semi-continuous mode (Zwiefelhofer, 1985; Wolinski, 1985) as opposed to continuous flow mode described by most researchers (Smith

.!!!:, al., 1975, Keller and Berninger, 1981, Kabrick and Jewell, 1982). The former have the advantage of a defined minimum retention time, a factor

of critical importance with respect to the hygienisation potential of

the process.

-174-

Our understanding of the solubilization/biodegradation of microbial

solids in aerobic thermophilic biotreatment systems remains scant. Kamer

and Mason (in press) have proposed a model for the biodegradation of

microbial solids in a semi-continuous system whereby growth of the

aerobic thermophilic population occurs on the released soluble compounds

contained in the cell cytoplasm following enzymatic degradation of a

part of the feed microbe cell wall. The cytoplasm is under a high

osmotic pressure such that lysis of the feed microbe cell wall easily

causes release of the cell contents. Growth on this soluble substrate

occurs simultaneously with the production of acetate when the system is

run under oxygen restricted conditions. As time progresses, the

remainder of the cell wall fractions are hydrolysed into utilisable

soluble substrates whilst the accumulated acetate is used as an

additional growth substrate. The effects of varying the charge size and

cycle time (minimum particle residence time) on the performance of this

process cycle has never been investigated in detail. Moreover, most

treatment plants operate only on weekdays whereby on Mondays to Fridays

a 24 h cycle is normal, whereas charging the reactor between Friday and

Monday allows a 3 day cycle time. Therefore, it was also of interest to

examine the effects of cycle time on process performance.

The final variable investigated in these experiments was the question as

to whether there was a requirement for addition of further nutrients.

Whilst phosphate is most unlikely to be limiting in such processes, the

availability of nitrogen could well be a problem. Nitrogen, when not

present in excess would have to be obtained from organic-N components of

the feed microbial cells. The differences in behaviour of process

systems in the presence of excess inorganic-N and under reduced availa-

bility of such sources were assessed.

-175-

Materials and Methods

Bioreactor and operating conditions. A l.5 l bioreactor (Bioengineering

AG, Wald, CH) with full measurement instrumentation, i.e. pH, dissolved

oxygen concentration, temperature and impellar speed monitoring, was

used for the experiments. Air (6 l h -l) was sparged into the bioreactor

which was operated with an impellar speed of 900 rpm. The operating

temperature was maintained constant at 6o0 c and the working volume was

1300 ml. The pH was not controlled.

Aerobic thermophilic culture and feed. Aerobic thermophilic microbes

were obtained from an operating municipal waste sludge aerobic

thermophilic digester (UTB, Umwelttechnik Buchs AG, Buchs/SG, CH).

Pressed bakers yeast (at concentrations between 30 and 60 g l -l) as the

sole biodegradable carbon and nitrogen source was

nutrient solution containing KH2

Po4 , 8 g l-1 , K2HP04 ,

suspended in a

5. 7 g l -l 1 zno, -l -l -l 3.26 mgl I FeCl36H20, 43.2 mgl I MnC124H20, 16 mgl I CUC122H20'

-l -1 -l -l mg l I CoC12 6H20 3.8 mg l ; H3B04, 0.5 mg l I MgC12 0.2 g l

l. 36

The

stock solution was stored at 4°c for no longer than 5 days. Start-up was

carried out by inoculating 500 ml aerobic thermophilic sludge into 800

ml of preheated (60°c) yeast suspension. The bioreactor was operated on

a fill and draw cycle of either 25% or 50% bioreactor volume withdrawal

of spent broth, followed by addition of the same volume of fresh yeast

cell suspension. Yeast addition was carried out slowly so as not to

cause a drop in temperature.

Analytical

Ory weight. 5 ml samples were centrifuged at 30,000 x g and the

supernatant removed. The solids were resuspended in water, poured into

tared crucibles and left overnight at lOs0 c. They were reweighed after

cooling in a desiccator.

Ash. After drying overnight, the crucibles used to determine dry weight

were placed in an oven at 600°c for at least 5 h, allowed to cool and

then reweighed.

-176-

Dissolved organic carbon (DOC). Appropriate dilutions of the supernatant

were made and the DOC assayed directly on a TOCOR total organic carbon

analyser (Maihak AG, Hamburg, D). Inorganic carbon was removed by

acidification and sparging with nitrogen.

+ NH4 ~ was measured in the extracellular medium using an automated

nitrogen analyser (Skalar Analytical, Breda, NL).

Chemicals. All chemicals were of analytical grade and supplied by either

Fluka (Buchs/SG, CH) or Merck (Darmstadt, Dl.

Results

The extent of solubilization and biodegradation of microbial cells in

semi-continuous culture in which aerobic thermophilic bacteria are

growing was investigated to examine the effects of charge size,

bioreactor residence time and inorganic-N availability. Differences were

obtained with respect to the amount of solids biodegraded at different

residence times and inorganic-N source concentrations in the bioreactor

feed. The results are shown in Figures 1-5. Figure 1 shows the amounts

of total suspended solids in the feed and after 7 and 24 h in the

bioreactor. From the data for a 25\ fill and draw volume and l day cycle

time it can be seen that the amount of solubilisation and biodegradation

oc,;urring between 7 and 24 h is almost negligable, Le., most effects

occur during the first 7 h in the bioreactor. Some increase in the

amount of solids degraded between 24 and 72 h can be seen from the data

from 3 day cycle times. Significantly better biodegradation in terms of

the percentage of total solids removed can be achieved with the larger

fill and draw volume (50\), where the extent of total solids removal was

consistently greater than 59\ (of feed microbial cell concentration)

irrespective of cycle time and the presence or absence of inorganic-N in

-177-

the feed (Fi9. 2). When a 25% char9e size was used, the de9radation

efficiencies were lower and affected by the presence or absence of

nitro9en in the feed. The volatile suspended solids were found to behave

in the same manner as the total suspended solids with respect to the

effects of char9e size and nitro9en, but showed hi9her removal

efficiences at lon9er cycle times for the lar9er char9e size (Fi9. 3).

For both volatile and total suspended solids a si9nif icant increase in

the amount biode9raded could be measured after 72 h compared to the

amount biode9raded after 7 h where data is available.

The results obtained from dissolved or9anic carbon analysis are shown in

Fi9. 4. The data for a 25% char9e size show some consistency while that

for 50% char9e size shows very different results in the presence of and

in the absence of added inorganic-N. Under conditions of nitrogen excess

with a 50% bioreactor volume char9e size, the dissolved organic carbon

concentration at the end of the charge period (24 or 72 h) is only

sli9htly higher than the influent DOC, and there is some su9gestion as

to an actual reduction in DOC during the biodegradation process. Very

much hi9her concentrations of dissolved organic carbon could be found in

the bioreactor at both l day and 3 day cycle times in the absence of

added nitrogen. When 25% bioreactor volume charge size was used, the DOC

remained constant between 7 and 24/72 h in the absence of added nitrogen

whilst sli9ht increases in the levels of DOC were found when nitrogen

was added to the feed.

The data for inor9anic-N (Fig. 5) indicate that for the smaller charge

size, inorganic-N accumulates in the absence of an added inor9anic-N + source, such that the concentrations of NH4 -N found at the end of the

+ 24 h period are nearly as high as those found when NH4 -N is included in

the feed. Use of the 50% bioreactor volume charge size resulted in a . + much lower accumulation of NH4 -N in the bioreactor when nitro9en was

ommitted from the feed compared with when nitrogen was present. It + should be noted that some loss of NH4 -N was undoubtedly occurring by

stripping from the bioreactor due to non-optimal outlet 9as condenser

operation and thus resulted in ammoniacal odours around the bioreactor.

-178-

The pH in the bioreactor was not controlled and the variation in pH

under different sets of conditions can be seen in Figs. 6 and 7. The

initial values are not important since they will depend on the

circumstances prevailing during the time preceeding the start of

operation. The pH values for a charge size of 50\ of the bioreactor

volume (Fig. 7) clearly shows that the greatest degree of variation

occurs with a 1 day cycle time both with and without added nitrogen. The

effect of fresh feed addition on the pH can be clearly seen in this

figure.

In the presence of nitrogen, operation during the weekend (3 days)

allows a significant rise in the pH value, whilst in the absence of

nitrogen the value at the end of the experiment is little different from

that at the start.

With a charge size of 25\ of the bioreactor volume (Fig.6), less effects

on pH can be seen and a gradual rise in pH occurred in all cases, with

the exception of 1 day cycle times in the absence of nitrogen where the

pH ended each 24 h only slightly lower than at the start of the cycle.

Discussion

One of the key factors controlling the rate of microbial growth is the

availability of nutrients. When a pool of soluble nutrients is

available, both metabolism and growth can occur optimally provided that

no nutrient is present at either growth limiting or inhibitory

concentrations. When microbes are presented with organic-N, this must

first be converted into a readily assimilable form. The thermophlilic

process microbes in the system described here are faced with such a

problem. Clearly, when nitrogen is provided as ammonia, the need to

release nitrogen from proteins, RNA and DNA, and from feed microbe cell

wall organic polymers is not growth restricting. However, when excess

soluble inorganic-N is unavailable, the growth and metabolic efficiency

-179-

of the therrnophilic process microbes is reduced. Egli and Quayle (1986)

showed that for nitrogen-sufficient growth of a rnethylotrophic yeast, a

carbon to nitrogen ratio of 12 : l was necessary. For the bacterium

Klebsiella pneurnoniae, Egli and Fiechter (1981) showed that this ratio

was 8.75 : 1. If it is assumed that the empirical formula for microbial

biomass on an ash and moisture free basis is: C Hl. 800•43 110 •23 (Harner

and Harrison, 1979) and a typical carbon content for microbial biomass

to be 47\ then this corresponds to a nitrogen content in biomass of

12.6\. Thus the theoretical carbon to nitrogen ratio in microbial

biomass is 4.35 l, a ratio which suggests sufficient nitrogen

availability when microbial cells are used as a substrate for growth of

therrnophilic process microbes assuming the nitrogen requirement of such

microbes is not excessively high. From the results presented here it is

apparent that in the aerobic therrnophilic digestion process, one of the

major rate limiting steps could be the release of nitrogen from complex

polymeric particles. Eastman and Ferguson (1981) discussed the fact that

hydrolysis of particulate material constituted the rate limiting step in

anaerobic digestion processes. In aerobic thermophilic digestion

processes where soluble nitrogen resources are limited, the further

hydrolysis of nitrogen bound in organic polymers may further limit the

overall rate of digestion.

The system investigated here was a semi-continuous process such that the

effect of reduced nitrogen availability in the feed was alleviated by

accumulation which took place as a result of only limited removal of the

bioreactor contents at the end of each cycle time. The effect of

removing a large proportion of the bioreactor contents (50\ as opposed

to 25\ of the bioreactor volume) on nitrogen accumulation can also be

clearly seen (Fig. 5).

The advantages of operating with larger charge sizes can be related to

the amount of readily utilizable carbon and energy substrates in the

system. Given that lytic products from lysis of the feed microbial cells

will be readily utilized, the remaining cell wall debris is likely to be

effectively only slowly biodegradable. Since, with a 50\ bioreactor

-180-

volume charge size more of the readily utilisable soluble lytic product

will be present a higher degreee of biodegradation is possible. In

addition, use of the larger charge size prevents high level accumulation

of slowly biodegradable material.

The variations in pll also demonstrate advantages for the process as a

result of five day (Monday to Friday) operation. During the one day

cycle time a progressive decrease in pH occurs, it is only due to the

fact that 3-day (weekend) cycle time allows a regeneration of the pH to

higher values that a collapse of effective operation as a result of low

pH conditions, causing inhibition, is prevented.

Hamer and Mason (in press) presented a model for semi-continuous

operation of the aerobic thermophilic process. This clearly showed that

variation in thermophilic process microbe activity and thus in the rates

of substrate utilization could be measured as a function of time. The

system investigated had a repeated three day cycle time. In systems with

a one day cycle time, these variations in process microbe activity are

unlikely to have any significance since the bacterial activity should

remain consistently high due to repeated substrate addition. When the

one day cycling is interrupted by a 3 day cycle (Friday-Monday) the

effects of process thermophilic microbe inactivation may reduce process

efficiency.

Conclusions

Semi -continuous operation of the aerobic thermophilic process for the

biodegradation of microbial solids results in high substrate removal.

Addition of inorganic-N results in enhanced removal of solids especially

under conditions where low charge sizes are used. Operation with SO'fs

bioreactor volume charge size results in slightly enhanced performance

over 25% bioreactor volume charge size. The effects of 3 day cycle time

(weekends) are effectively only seen in pH effects and offer only

advantages in this respect for overall process efficiency.

-181-

References

Eastman, J.A. and Ferguson, J.F.: 1981, :!· Wat. Poll. Cont. Fed. 53, 352-366.

Egli, Th. and Fiechter, A.: 1981, :! . ~· Microbial. 123, 365-369.

Egli, Th. and Quayle, J.R.: 1986, J. ~- Microbial. 132, 1779-1788.

Hamer, G. and Harrison, D.E.F.: 1979, ~: ~· Petroleum ~-Canterbury, G.B., 59-73.

Kabrick, R.M. and Jewell, W.J.: 1982, ~ ~· 16, 1051-1060.

Keller, u. and Berninger, I.: 1984, Gas-Wasser-Abwasser 64, 215-224.

McEvoy, J. and Giger, W.: 1986, Environ. Sci, Technol. 20, 376-383.

Smith, J.E. Jr., Young, K.W. and Dean, R.B.: 1975, ~ ~- 9, 17-24.

Wolinski, w.K.: 1985, !!!!:!· ~· f.2!!!· 84, 433-445.

Zwiefelhofer, H.P.: 1985, ~· Recyc. 8, 285-302,

-182-

251 501 ., +N -N +N -N +N +N

.. ..

" .. "

Cycle Time (d)

Figure l. Total suspended solids in the feed (first bar), after 7 h (middle bar) and after 24/72 h during semi-continuous aerobic thermophilic solubilization/biodegradation of microbial solids, as a function of cycle time, the presence (+N) or absence (-N) of inorqanic-N in the feed and of the charge size (25% and 50\ of bioreactor operating volume).

251 501 .. .. +N -N +N -N +N +N

" ••

..

.. "

Cycle Tim• (d)

-N

-N

Figure 2. The amount of total suspended solids biodegraded as a percentage of the concentration in the feed after 7 h (left hand bar, where data available) and after 24/72 h during semi-continuous aerobic thermophilic solubilization/biode9radation of microbial solids, as a function of cycle time, the presence (+N) or absence (-N) of inorganic-N in the feed and of the charge size (25\ and 50\ of bioreactor operating volume) .

..

J! ..

I r· i ..

Figure 3.

-183-

2U SOX .. -N +N -N +N -N +N -N

,. ,.,,

T :::::::

"'· I .. ·;:: f::i / .. :::~::

.... :. ·.-:• ::.~ .. ::~::: :·:·::· ~i ".'/

~~r .. ::·:. 0 .....

it .. .... ..... :~:::

•• I ~~~~~ ~~i~

~~ 71 •• .. ,

·~{~~ ~ ::i::

....,

,, TI' ... (d) Cycit ..

The amount of volatile suspended solids biodegraded as a percentage of the concentration in the feed after 7 h (left hand bar, where data available) and after 24/72 h during semi-continuous aerobic thermophilic solubilization/biode9radation of microbial solids, as a function of cycle time, the presence (+N) or absence (-N) of inorganic-N in the feed and of the charge size (25• and 50~ of bioreactor operating volume).

+N -N +N -N

Figure 4. Dissolved organic carbon concentration in the feed (first bar), after 7 h (middle bar) 'and after 24/74 h during semi-continuous aerobic thermophilic solubilization/biodeqradation of microbial aolids, as a function of cycle time, the presence (+N) or absence (-N) of inorganic-N in the feed and of the charge size (25\ and 50• of bioreactor operating volume).

-184-

25% 50%

-N

• Figure 5. NH4

-Nin the feed (first bar), after 7 h (middle bar) and after 24/72 h during semi-continuous aerobic thermophilic solubilization/biodegradation of microbial solids, as a function of cycle time, the presence (+N) or absence (-N) of inorganic-N in the feed and of the chaxge size (25% and 50% of bio~eactor operating volume}~

:r Q.

i

-185-

8.1

7.8

7.9 3d +N

' . 1d +N

8.9 " ./' 3d -N

6.6 1d -N ..... ____ .......... / a.3

&.1-I-----~------.------..-----~-----.----.....,...-----...------,

8.!

7.8

7.9

7.2

8,9

e.s

8.3

0 111 •• eo 72

Time (h) Variation in pH during aerobic thermopbilic stabilization/• biodeqradation of microbial solids, as a function of the cycle time and of the presence (+N) or absence (·N) of inorganic-N in the feed using a charge size of 25• of the bioreactor operating volume.

3d +H

I I

I ~" I

I

I I

I

I I

I

r·

1d +N

8.7'-l-------~----...... ------...-----...... ----...... ------------...... ------~ 0 •• ••

Time (h) 72

!!l!!!!..1• Variation in pll during aerobic thermophilic stabilization/-biod•tradation ot llticrobia1 •olida, •• a function of the cycle time and of the preaenca (+Nl or absence (-Nl of inorganic-N in the feed uaing a char9e size of SO• of the bioreactor operating volume.

-186-

CHAPTER 13

SOME FUNDAMENTAL ASPECTS OF TWO STAGE WASTE SEWAGE

SLUDGE TREATMENT WITH AEROBIC THERMOPHILIC PRETREATMENT

ANO ANAEROBIC STABILIZATION

Synopsis

The design and effective operation of aerobic thermophilic waste sewage sludge pretreatment processes requires a detailed understanding of the process microbiology. In this paper the mechanisms of microbial solids biodegradation are defined in relatively simple terms and a realistic process model is developed. The problems associated with pathogen survival are examined in detail and the facets involved in both the reinfection and regrowth of pathogens in treated sludges are discussed.

-187-

J.. Introduction

Increasingly stringent environmental legislation in most West European

countries has resulted in increased capacity for municipal sewage and

industrial wastewater treatment by combinations of mechanical,

biological and physico-chemical process technology. The major by-product

of both mechanical and aerobic biological treatment processes is waste

sludge, a putrefactive, aqueous suspension of biodegradable, partially

biodegradable and essentially non-biodegradable solids and similarly

degradable dissolved and sorbed matter. Waste sludge presents a serious

disposal problem, particularly in regions remote from the sea, where

frequently the policy for ultimate disposal involves spreading treated

sludge on agricultural land so as to cause minimum nuisance.

Conventional waste sludge treatment technology involves mesophilic,

anaerobic digestion, but such technology is no longer considered

entirely satisfactory for the removal of either pathogenic organisms or

toxic chemicals from sludge. Further, the physical characteristics that

affect the dewatering of conventionally treated sludges are frequently

such as to result in high dewatering process costs, a feature of

increasing economic concern.

Thermophilic, anaerobic sludge digestion processes have been proposed as

cost effective alternatives to conventional, mesophilic, anaerobic

processes, but plants constructed for the latter technology cannot

easily be converted for thermophilic process operation. Although

thermophilic, anaerobic sludge digestion might seem attractive on the

basis of both reactor residence times and hygienization effect, the

spectrum of apparently recalcitrant pollutants is most unlikely to vary

for anaerobic processes, irrespective of their temperature of operation.

Whilst most compounds that are biodegradable under aerobic conditions

are also biodegradable under anaerobic conditions, usually at markedly

different rates, some important pollutants commonly found in waste

sewage sludges, including hydrocarbons and some synthetic organic

compounds, particularly detergent residues (1), are recalcitrant to

- urn-

anaerobic biodegradation. In view of both this and of the question of

recently installed mesophilic, anaerobic digestion capacity in many

treatment plants, the most realistic approach for more effective waste

sludge treatment would seem to be the introduction of either pre- or

post-treatment processes operating in conjunction with conventional in-

stalled, anaerobic, mesophilic, waste sludge digestion processes, that

enhance treatment effectiveness with respect to the criteria discussed.

In Switzerland an ordinance, issued in revised form by the Federal

Government in 1981 (2), requires treated sludge that is to be spread on

agricultural land to have a count of Enterobacteriaceae, essentially

pathogen indicator bacteria, of less than 100 per gramme of wet treated

sludge and no virulent worm eggs. To achieve this, attention has been

focused on the hygienization potential of aerobic, thermophilic

pre-treatment processes. The self-heating and hygienization potentials

of aerobic composting processes used for solid wastes is widely

recognized, but the realization that suspended carbonaceous solids

destruction and/or solubilization could be undertaken as effectively

under aerobic as under anaerobic conditions has been latent.

A major objective of all biological fluid phase waste treatment

processes is minimization of the microbial biomass yield coefficient.

For aerobic processes this automatically results in maximization of heat

production. For low concentration waste streams, heat production during

aerobic biotreatrnent is relatively insignificant and the structure of

typical wastewater treatment plants, i.e., shallow open aerated basins,

optimizes heat losses to the surroundings and equilibration of the

process temperature with that of the surroundings. For the aerobic

biotreatment of concentrated fluid waste streams such constraints on the

process operating temperature need not apply provided the treatment

process is undertaken in either high or medium aspect ratio, insulated,

enclosed bioreactors under operating conditions where the dispersed air

(oxygen) flow to the bioreactor is minimized and the potential for

autothermal process heating, and hence, both process rate and

hygienization, enhanced.

-189-

Aerobic sludge treatment was first proposed in 1959 (3). Essentially

four distinct processing concepts exist;

(i) Aerobic mesophilic (15-35°C) digestion;

(ii) Aerobic autothermic (40-S0°CJ di9estion;

(iii) Aerobic thermophilic (50-70°CJ di9estion;

(iv) Aerobic thermophilic (50-70°C) I Anaerobic mesophilic (30-40°C) digestion.

Combined aerobic thermophilic/anaerobic mesophilic treatment processes

offer the greatest pctential for effective waste sludge treatment.

However, fundamental bioprocess data concerning aerobic thermophilic

pre-treatment are strictly limited. In this paper, rectification of this

unacceptable status has been sought with respect to two important

process features; the definition of process mechanisms and to the

question of pathogenic organism survival.

2. Microbial 'Process Sequences

Traditional microbiology has focused on monospecies cultures and

fostered dispropcrtionate emphasis on soluble carbon energy substrates

and batch culture systems, suppcsedly operatin9 under constant physical

and environmental conditions. Bull and Quayle (4) have ar9ued that such

an approach has resulted in a restricted perspective of microbial

behaviour. Fortunately, the recent past has seen studies involvin9

multispecies cultures and multiple and insoluble substrates. System

inhomogeneities, gradients and transients and their possible effects are

also becomin9 more widely reco<Jnized. However, the microbiological basis

for slud9e treatment processes remains scant.

The utilization of mixed particulate and soluble carbon ener9y

substrates by multispecies cultures is highly complex. A simplified

representation of such a process is shown in Figure 1.

-190-

Particle hydrolysis is the least studied and least understood process

step in sewage sludge treatment, even though it is probable that the

hydrolysis of particulates to form soluble substrates is the rate

limiting step in the overall process (5). Major factors affecting the

rate of hydrolysis of non-microbial solids are particle size and size

distribution, composition of the solids and the rate of and source for

appropriate enzyme production for hydrolysis. In the case of the

decomposi,tion of microbial solids, several different process mechanisms

occur, but data concerning these mechanisms are negligible, although the

processes that are involved in microbial solids degradation are most

important for the production of hygienized and stable residual sludge.

The key processes in the biodegradation of microbes are lytic product

formation and "cryptic" growth. Microbial cell lysis can be attributed

to various mechanisms, but in the context of sewage sludge digestion,

autolysis, i.e., the action of endo-enzymes within the cell undergoing

lysis, and lysis by exo-enzymes, produced by other microbial species

present in the digester, are probably the most important. Either

mechanism will result in the production of soluble lytic products and

submicron size cell wall fragments.

The dissolution of solids has, since the theoretical analysis (6) and

experimental verification thereof (7) by Hixson and Crowell, been

considered to follow the cube root law. In the case of cellulose

particles, which are solubilized by enzymic hydrolysis, Humphrey and

co-workers (6,9,10) have proposed a complex shrinking-site model, which

incorporates both inhibition and repression, but ultimately predicts

modified cube root law type dissolution. In the case of the enzymic

hydrolysis of microbial cells prior to their utilization as carbon

energy substrates by the microbes responsible for their hydrolysis, the

cube root law is inapplicable, because here enzymic hydrolysis involves

either puncturing or bursting of the substrate microbe cell walls,

processes that depend on point attack and point strength of the

substrate microbes.

-191-

!.Bio-0.Xidation Process Stoichiometry

When biodegradable matter is aerobically degraded by microbes, in the

presence of all other nutrients essential for the aerobic growth of the

microbes, a fraction of the biodegradable matter i5 converted into

carbon dioxide and the energy generated is used by the microbes

performing the oxidation to fix a further fraction of the substrate

carbon to form additional microbial biomass. Depending on both the

nature of the original substrate and oxygen availability, biodegradable

and/or non-biodegradable products can also be produced together with a

correspondingly reduced production of energy.

Mixed culture systems are obviously more complex, in every respect, than

are mono-cultures of individual microbial species, but for effective

sludge treatment the former systems are clearly both more versatile and

more efficient. Both aerobic and facultative anaerobic microbes function

effectively in aerobic process environments and in such process

environments excess oxygen availability frequently results in complete

oxidation, with predominantly microbial biomass and carbon dioxide

production, whilst oxygen restricted growth conditions result in

significant levels of soluble product formation and lower biomass yield

coefficients.

The microbial biomass yield coefficient is defined as the weight of dry

biomass produced from unit weight of carbon energy substrate. Analogous

definitions exist for the production of growth associated extracellular

products from both the carbon energy substrate and from other nutrients.

Yield coefficients depend on process operating conditions, and

therefore, are not constant. Optimization of process operating

conditions allows either maximization or minimization of yield

coefficients as required, particularly in the case of aerobic processes

involving facultative anaerobes as process microbes (ll). Most yield

coefficients data quoted refer to the growth of microbial monocultures

on essentially pure carbonaceous compounds under conditions designed to

maximize the biomass yield coefficient. In contrast, sludge treatment

-192-

processes involve the growth of complex mixed microbial cultures on

ill-defined mixtures of carbonaceous compounds under process conditions

designed to minimize the overall biomass yield coefficient, For

realistic process design, it is essential to be able to predict biomass,

carbon dioxide and soluble product production. The basis for this is the

establishment of appropriate equations that describe the overall

process. In order to achieve this for microbial solids degradation in a

continuous flow, non-recycle, aerobic thermophilic process with

restricted oxygen supply, it is first necessary to define the several

process steps involved.

4. Process Definition

The process steps involved in microbial solids biodegradation in a

continuous flow bioreactor, shown diagrammatically in Figure 2, are:

(i) A suspension of intact microbial cells is fed into the

bioreactor, where the thermophilic process microbes in the

bioreactor produce exo-enzymes which cleave the cell walls of

the feed microbes after attachment to their surface in a

non-specific distribution.

(ii) Cell wall lysis occurs at random locations on the feed

microbes resulting in release of the soluble cytoplasmic

components (lysis results after a minimum of one site cleavage

in the wall but may require multiple site cleavage before the

wall is sufficiently weak to lyse).

(iii) Thermophilic process microbes utilize the released soluble

substrates as carbon and energy sources and in so doing

produce acetate since the dissolved oxygen concentration is

restricted.

-193-

(iv) Lysis of remaining whole feed microbes supplies a continuous

source of soluble nutrients for the thermophilic process

microbes 11nd enzymic degradation of the feed microbe cell

wall polymers further supplements this supply.

(v) Autolysis of some thermophilic process microbes also

supplements the pool of soluble nutrients.

(vi) Acetate, produced under conditions of dissolved oxygen

restricted growth of the thermophilic process microbes, can

also be utilized by the thermophilic process microbes, but

as the least preferentially utilized substrate.

Fc;r a process system in which the oxygen supply is restricted, the

overall process can be described by the following molar stoichiometry:

Feed microbes

Acetate

Thermophilic process microbes

Thermophilic process microbes

Acetate

The elemental composition of the feed microbes is assumed to be the same

as that for thermc;philic process microbes and is expressed above on a

nitrogen and ash free basis. For the dry weight based yield

coefficients, a molecular weight of 27 g is assumed for 1 mole of cells

inclusive of nitrogen and ash. Thus from the above coefficients, yield

values (g/g) of thermophlilic microbes and acetate from the feed

microbes will be 0.3 and 0.52, respectively.

(2)

-194-

In order to describe the growth of thermophilic process microbes, Monod

kinetics can be assumed for growth on feed microbe lysate and Monod

kinetics with an inhibition term for growth on acetate, i.e.,

J1 (Ss) t'max (Ss) • Ss Ks +Ss Ss

µ(Ac) = pmax(Ac) • Ac.Ki (Ss+Ki l (Ac+KsAc)

The nomenclature is defined subsequently.

(3)

(4)

The remaining rate expressions i.e., for feed microbe lysis, for process

microbe lysis, and for particulate hydrolysis, have been assumed to be

first order due to lack of information regarding the kinetics of such

processes. In the case of the particle hydrolysis rate constant the use

of first order kinetics can be justified for reasons discussed by

Eastman and Ferguson (5).

From this basic considl!ration, a stoichiometric matrix can be

constructed for the five para111eters,

microbes), Ss (soluble substrate XP (proc;:ess microbes) Xs

concentration from lysed

(feed

feed

microbes), p (particulate materials released from cells as a resµlt of

lysis, e.g., cell wall fractions) and Ac (acetate). The stoichiometric

matrix is shown in Table 1.

From this, the following material balance equations, for the process

system shown in Figure 2 result:

x : p

x : s

Ss:

P:

dX __£ dt

dX __ s dt

dSs

dt

dP

dt

-195-

D(X-X)-K.X 5., s L s

= -DSS + YSs/Xs.KL.xs + Kh.P - µ(Ss) .xp + YSs/Xp.Kd.xp

YXp/Ss

(5)

(6)

(7)

(8)

Ac: dAc -DAc + YAc/Ss.p(Ss) .Xp - µ(Ac).Xp (9)

dt YXp/Xs YXp/Ac

~- Continuous Flow Process Predictions

The predictions resulting from the material balance equations are shown

in Figure 3. The numerical values attributed to the various constants

are shown in the figure caption. The predicted trends are:

(i) Feed microbe concentration in the completely mixed aerobic

thermophilic bioreactor and the outlet thereof decreases

with increasing residence times in the bioreactor.

(ii) Process microbe concentrations increases with increasing

residence times and reaches a maximum.

(iii) Particulate matter (cell wall debris) initially increases

and then slowly decreases with increasing residence times.

-196-

(iv) Total suspended solids (feed and process microbes plus

particulate matter) decreases gradually with increasing

residence times.

(v) Soluble substrate (residual lytic products from feed and

process microbes) concentration initially drops

precipitously and remains at a very low level with

increasing residence times.

(vi) Acetate concentration first increases and then decreases

with increasing residence times.

These predictions have been experimentially verified in a model process

system, discussed elsewhere (12), where an aerobic thermophilic mixed

process culture was grown on a standardized microbial solids substrate

(bakers yeast) to simulate secondary sludge in a completely mixed,

aerated bioreactor maintained at 6S0 c and operated at various residence

times. Some results for total suspended solids and acetate

concentrations from this investigation are shown as points in Figure 3.

These experimentally determined values confirm the trends indicated by

the theoretical analysis, but the fit between theory and experiment is

by no means perfect; a feature which undoubtedly results frOlll the high

degree of simplification particularly ignoring all cartioxylic acids

other than acetic acid, used to describe such a highly complex process.

It should also be noted that the values attributed to two constants

involved in the process model, the lysis rate constant for process

microbes and the inhibition constant for acetate utilization, markedly

affect the predicted level of acetate accumulation.

The predictions clearly indicate that aerobic thermophilic treatment

processes are best used as pretreatment processes. To achieve complete

stabilization of microbial solids solely by aerobic thermophilic

treatment would require residence times well in excess of S ~ys, and it

-197-

seems probable that such processes would not compete effectively, on

either technical or economic grounds, with conventional anaerobic

mesophilic stabilization processes. However, the predictions clearly

indicate the high levels of activity that occur in such processes when

operated with residence times of less than two days, particularly where

high rates of biodegradation are coupled with effective hygienization

which is so when they are used as pretreatment processes. Acetate

accumulation during pretreatment can only be regarded as beneficial for

the rate of the subsequent anaerobic stabilization process.

6. Sludge eygienization

The objectives of all sludge hygienization processes is the destruction

of pathogenic bacteria, viruses, worm eggs and protozoa which can

potentially result in serious diseases in humans and animals after

disposal of treated waste sewage sludge.

Pathogenic organisms have optimum temperatures for growth below 45°c and

most are either inactivated or "killed" at temperatures in excess of

so0 c, hence the susceptibility of pathogens to thermophilic treatment

processes.

Bacteria of enteric origin such as Escherichia ££!.!:. and Klebsiella

pneumoniae are used as indicators of more serious pathogens. Thus, ~·

coli and K. eneumoniae, which are both themselves potentially

pathogenic, are monitored together with other coliforms to indicate the

possible presence of more seriously pathogenic microbes such as

Salmonella spp. and Vibrio spp. such an approach is embodied in the

Swiss Federal Ordinance concerning sewage sludge treatment and disposal

(2), as mentioned earlier. The ability of bacteria to survive heat

treatment is a function of the exposure time and the temperature used.

However, the physiological state of the bacteria is also significant.

-198-

The realization of effective sewage sludge hygienization involves two

major facetsi the death/irreversible inactivation of pathogens and

prevention of subsequent reinfection with and regrowth potential of

pathogens. In order to investigate these facets, experiments haVE< been

undertaken in which the bacterium !· pneumoniae was grown aerobically at

35°c and pH 6.80 under steady state conditions in a bioreactor operated

as a chemostat and the spent culture fed to a second bioreactor where

the bacteria were subjected to heat treatment under asceptic conditions,

constant pH and aeration. The experimental system ill shown

diagrammatically in Figure 4.

1· Bacterial survival

The survival of!· pneumoniae cells exposed to temperatures of 35°, 41°,

49°, 55° and 60°c with a mean residence time of 32 h in the second

bioreactor was determined using measurement of metabolic activity in an

INT [2-(p-iodophenyl)-3-(p-nitropnenylJ-S-phenyltetrazolium chlorid~

reduction assay (13), rather than bY means of inac:;curate and

inappropriate agar plate count techniques. A significant decrease in

overall metabolic activity was found between microbes growin9 in the

chemostat at 35°c and those present in the second bioreactor when

maintained at 3S0c. Increasing the temperature of the second bioreactor

resulted in further decreases in metabolic activity at 41° and 49°c,

although the levels found were significant in that they indicated the

bacteria were still capable of metabolic reactions despite the

super-optimal temperatures. At 55° and 60°c the measured metabolic

activity was extremely low, but not zero. Thus, despite temperatures in

excess of the minimum for effective deactivation, metabolic activity

could still be detected.

-199-

This low level of metabolic activity found after apparently prolonged

heat treatment at either 55° or 60°c in a continuous flow bioreactor

suggests a potential for regrowth under conditions where short

circuiting of feed in the bioreactor can occur and where appropriate

substrates and nutrients are available, as would be the case in sewage

sludge treatment processes where overall stabilization was incomplete.

In sewage sludge treatment, the heterogeneity of the process feed could

enhance short circuiting in continuous flow hygienization processes.

8. Reinfection and R!growth

In order to investigate reinfection and regrowth after heat treatment of

K. pneumoniae cells at 49°, 55° and 60°c, the flow of spent culture

fluid from the chemostatically operated bioreactor was stopped, but the

operating temperature of the second bioreactor was maintained. After 4 h

the temperature of the second bioreactor was reduced to 35°c and a flow

of fresh sterile medium added such that the mean residence time was 12.5

h. The ensuing washout and regrowth of the culture in the second

bioreactor was monitored and the results are shown in Figure 5, where

regrowth after a distinct lag, can clearly be seen after treatment at

all three temperatures. Checks were carried out to confirm that the

growing bacteria were in fact, K. pneumoniae. The most plausible

explanation for the regrowth is reinoculation of the cooled culture by

head space splash drainage rather than survival of bacteria in the

liquid culture, particularly as the bioreactor head space would never

have reached the heat treatment temperature. Wall growth is, of course,

commonly encountered in the head space of continuous flow bioreactors

(14).

The consequences of this, is that in any incompletely filled bioreactor,

the head space will contain a source for reinoculation, by head space

drainage, even under conditions where careful control is carried out.

-200-

!· Concluding Relnarks

Aerobic thermophilic pretreatment processes of fer a realistic technology

for achieving effective hygienization of waste sewage sludges. However,

the experimental results presented clearly indicate that ~th process

design and operating strategies must be based on an adequate

understanding of the process microbiology of such systems. The results

also suggest that thermophilic hygienization processes must be coupled

with fully effective stabilization processs if treated sludge is to meet

stringent standards prior to ultimate disposal on agricultural land,

It is unlikely that the use of aerobic thermophilic processes for

complete sewage sludge stabilization, as a single stage process, will be

able to effectively compete with conventional anaerobic mesophilic

treatment processes, on either an economic or a technological basis,

although their use as a pretreatment process, could, in addition to

achieving effective hygienization, also increase process ratee during

subsequent stabilization.

-201-

References

1. Giger, W., Brunner, P.H. and Schaffner, c. (1984). 4-Nonylphenol in sewage sludge: Accumulation of toxic metabolites from nonionic surfactants.· Science, 225, 623-625.

2. Schweizerisches Bundesamt fUr Umweltschutz (1981). Klarschlammverordnung 814.225.23, 20pp Bern.

3. Husmann, W. and Malz, F. (1959). Untersuchung zur biologischen Abwasserreinigung auf aerober thermophiler Grundlage. GWF-Wasser/Abwasser !.Q.2. 189-193.

4. Bull, A.T. and Quayle, J.R. (1982). New dimensions in microbiology: an introduction. Phil. ~· .!'!£l· ~· ~· ~· ~ 297, 447-457.

5. Eastman, J.A. and Ferguson, J.F. (1981). Solubilization of particulate organic carbon during the acid phase of anaerobic digestion. l· .!!!!!· ~· ~· ~· 22_, 352-366.

6. Hixson, A.W. and Crowell, J.B. (1931). Dependence velocity upon surface and agitation: I consideration. Ind. Engng. ~· 23, 923-931.

of reaction Theoretical

7. Hixson, A.W. and Crowell, J .H. (1931). Dependence of reaction velocity upon surface and agitation: II- Experimental procedure in study of surface. ~· Engng. Chem. !]_, 1002-1009.

8. Humphrey, A.E., Armiger, W.B., Zabriskie, D.W., Lee, S.E., Moreira, A. and Joly, G. (1976). Utilization of waste cellulose for the production of single cell protein. In: Continuous Culture 6: Applications and New Fields (A.C.R. Dean, D.C. Ellwood, C.G.T. Evans and J. Melling, Eds), 85-99.

9. Humphrey, A.E., Moreira, A., Armiger, w. and Zabriskie, o. (1977).

10.

Production of single cell protein from cellulose wastes. Biotechnol. Bioeng. !r.!!!f· z, 45-64.

Moreira, A.R., Phillips, J.A. and Humphrey, Utilization of carbohydrates by Thermanospora glucose, cellobiose and cellulose. Biotechnol. 1325-1338.

A. E. (1981). Sp. grown on Bioengni. !]_,

11. Harrison, D.E.F. and Loveless, J.E. (1971). The effect of growth conditions on respiratory activity and growth efficiency in facultative anaerobes grown in chemostat culture. l· ~· Microbiol. ~· 35-43.

-202-

12. Mason, C.A., Hamer, G., Fleischtmlnn, Th. and Lanc,J. c. Aerobic thermophilic biodegradation of micr0bial cells; 2. Some effects of temperature (Manuscript submitted for publication).

13. Lopez, J.M., KooJ?1114n, B. and Bitton, G. (1986). INT~dehydroqenase

test for activated sludge process control. Biotechnol. Bioen9. ~· 1080-1085.

14. Hamer, G. (1972). Discussion used for the cultivation Biotechno1. Bioen9. 14, 1-12.

on entrained droplets in fermentere of single-celled microorganisms.

Nomenclature

YXp/Xs

YA<;/Ss

Ks Ss

Ks Ac

Yield process thermophile microbes on feed microbial cells

Yield process microbes on acetate

Yield acetate produced by process thermophiles growing on feed microbial cells

Yield soluble substrate from lysis of feed microbes

Yield soluble substrate from lysis of process thermophile microbes

Yield particulates from lysis of feed microbes

Yield particulates from lysis of process thermophile microbes

Maximum specific growth rate con-stant for growth of process microbes on soluble substrates

Maximum specific growth rate con-stant for growth of process microbes on acetate

saturation coefficient for growth of process microbes on soluble substrate

Saturation cqef ficient for growth of process microbes on acetate

-203-

l(d Death/lysis rate constant for process T-1 microbes

I(. Inhibition constant for growth of -1 1 process microbes on acetate ML

KL Lysis rate constant for feed microbes T -l

l(h Hydrolysis rate constant for parti- -1 culate biomass T

D Dilution rate T -1

x So Microbial cell concentration in feed ML -1

F, F2 Hydraulic inflow and outflow rates LT -l

x,x• Steady state bacterial concentrations ML-1

-204-

Table l. - StoichiQMetric matrix for aerobic ther1110phi lie b iode9radation proceaa.

x x. p Sa p Ac

~ 11eac:U9n !l!t•

I. Substrate Cell 0 -1 YSs/Xo YP/Xs 0 ltL,xs Ly sis

2. Particle 0 +l •l 0 '), ,p

Hydrolysis

J. Growth on +l 0 -1 0 +YAc/Ss f(Sal .Xp Lysis Products YXp/S• Yxp/So

4. Process Microbe -1 0 •Yso/xp +YP/Xp 0 ICd,Xp Death/Lysts

5. Growth on +l 0 0 -l )l(Ac).Xp

Acetate YXp/N:

Note: .. J\• the stoichiometry 1• incomplete, i.e., co2 and o 2 a.re not considered, the utrix

is not balanced.

-20s-

PARTICULATE BIO-

DEGRADABLE POLLUTANTS

hydrolytic partial enzymes degradation

• SOLUBLE PRODUCTS •

SOLUBLE/IMMISCIBLE BIODEGRADABLE

1 POLLUTANTS

J ' . ------+ GROWING MIXED CARBON DIOXIDE MICROBIAL CULTURE + ..........

ENERGY 1-------------------+ MAINTENANCE ~

cryptic ·growth • I I I l YTIC PRODUCTS L DEAD MICROBES I

----...... ENERGY

--- CARBON Figure 1. Carbon and energy flows for the heterotrophic

Figure 2.

growth of multispecies cultures on soluble and particulate carbon energy substrates.

F F Xso Xp

Xs Ac.

Xp Ac p Ss

Xs p Ss

Continuous flow stirred tank bioreactor operating under thermophilic aerobic conditions (55-7o0 ci showing steady state conditions.

10 ......

1... -8

.. 6 ... .a

0 .. ... E

4 .. .. ... " e a..

2 .... c:: ... ... ~

o~

-206-

"" "" \ ,, 0

\ \.o '"'-...::·······~ ........ .

\ '· .... '·· ·· ... '. ........ .. ~' ..... ~.. ·· . I .,;..-.·,-:--... -··-::·----£ .... 0

I /1 '· ·-.. -.::::_ · ---.................. rss I I / ...... ~.~·-... ·:·-1·---. ....... ·. -··· I : -.. ·. Parr-··· I l : ·-.. ..... \ tcu1a1es-,.....-·-

0

: ' ·-.Fee . ! \ - . .!!.._Microbes ~, ! "'°'\·- • L-.,;:

: ~ ---': ...... -.-.... I. : ····· .. "'l ...__, ""

. ".----::.. '~ ' ..... ~,. '--.1~~!..22!!!.!!!..J::"" ____ _::: .•.

1 2 3 4 Residence Time ''"

5

2.0

1.5

1.0

0.5

0

Figure J. Steady state solution of mathematical model for aerobic thermophilic sludge tretment. Stoichiometric and kinetic constants used: YX /X , 0.3 g g-1 ,

-1 p -J: -1 Pmax(Ss)' 0,6 h ; Ksss' 0.8 g l ; Kd, 0.004 h ;

-1 -1 Y~~/Ac' 0.17g g ; ~Tax(Ac)' 0.6 h_

1; Ki, 0.00012 g

1-1 KsAc' 0,01 :11 i KL, 0.02S h -i YSs/Xs' 0.8 g g-l Kh' 0.06S h ; Y§!/Xp' 0.8 g g -i YP/~s' 0.2 g 9 YAc/Xs' 0.52 g g ; x50 , 40 g l • Points represent experimental values for total suspended solids (TSS) (A) and for carboxylic acid concentrations (o).

1... "" -., ... -:;; ::I .~ .. ... a.. .... c ... ... ~ .. .a ::I II)

... :a " 0

II)

F Glucose

+ Nutrients

-207-

F

pH 6.80 35°C

x'

~2 x

pH 6.80 35-60°C

Figure 4. Two continuous flow stirred tank bioreactors in series showing steady state operating conditions. The outflow from the first bioreactor was used as the sole feed to the second bioreactor which was maintained at temperatures of 35°, 41°, 49°, 53° and 6o0 c with a residence time of 32 hours.

-208-5 4 A

E 3 c

<.O 2 -.:!' t.n c 0 :.;:: c.. 1 ... 0 U)

D <(

E c;

<.O ...;:t c.o c 0 :.;::; c.. ... 0 U) .D <(

E c:: <.O ...;:t t.n c:: .2 -c.. ._ 0 U) .D <(

81 0.7 0.6 0.5

4

3

2

1 o.9 0.8 0.7 0.6 0.5

4

3

2

1 09 0.8 0.7 0.6 0.5

-4

8

c

-2 0 2 4 6 Time(h)

8 10 12

~egrowth of ~· pneumqniae cu1turee at 35°c after heat-treatment

at; A, 49°ci B, ss0c and c, 60°c. Tl\e e~pected washout curves are

also shown(---). Before initiating a flow of carbQn and

nutrients, the culture was ""'intained at the elevated temperature

for four hours without a flow of fresh cells from the first

bioreactor.

-209-

CHAPTER 14

GENERAL DISCUSSION

Whilst each chapter contains

overall discussion presented

consequences and implications

Cldditional information, where

drawn.

discussion of the results presented, the

here concentrates on the more general

of the work and further, provides

necessary, to enforce the conclusions

Microbial death, lysis and "cryptic" growth - Fundamental Aspects

Microbial solids destruction, whether in technical-scale wastewater

treatment processes, during optimised growth, during unbalanced growth

or as a result of externally applied stresses is a topic which has been

largely ignored and is, therefore, little understood when compared with

the available knowledge concerning microbial growth and growth related

phenomenon. That microbial growth and microbial death/lysis are

processes which effectively occur coincidentally in most real systems is

often overlooked, and has frequently led to either over simplification

or to erroneous data interpretation.

Much of the data concerning microbial death is found in the food and

pharmaceutical literature where contaminating microbes are subjected to

E!Xtreme stresses so that products meet public health safety

requirements. However, as mentioned in the conclusions to chapter two,

microbes are neither immortal nor are they the infallable machines that

they are often made out to be. Even under idealised growth conditions,

mistakes can occur - resulting in death/lysis phenomena.

-210-

At the start of the experimental programme, it was hypothesised that the

processes of death, lysis and "cryptic" growth were all events which

could occur in a growing culture of microbes. To test this hypothesis,

major problems were encountered in both defining what was meant by these

terms and in the selection of satisfactory methods which could be used

to measure the rates of the respective processes.

As a result of the model presented in chapter three and the use of the

INT-reduction assay as a means of quantifying active microbe

concentrations (see appendix for structure of INT) it became clear that

certainly in bacterial mono-cultures, death and lysis are synonymous, in

that death is more likely to occur by lysis than by failure of metabolic

processes. Even in situations where a gradual decline in the metabolic

activity of individual cells occurs, because of the finely balanced

control mechanisms required to maintain the integrity of the cell wall

and membrane structure, and hence, the individual entity, lysis is

likely to precede total metabolic shutdown due to an imbalance in

homeostatic mechanisms. In multicellular organisms the definition ol:

death presents additional problems since a part of the organism might

still maintain metabolic potential whilst the whole fails to function as

an integrated entity. However, such problems in multicellular organisms

are outside the scope of this discussion and, the theories presented.

here are therefore, only applicable to unicellular microbes. In

bacteria, the absence of compartmentalisation allows death to be defined

as a total absence of metabolic activity in an otherwise intact cell.

The situation pertaining to laboratory cultures of microorganisms may

also apply to natural environments where complete shutdown of metabolic

activity, resulting in the production of intact dead cells may be a rare

event such that it is likely that here also actual death is preceded by·

lysis.

Lysis has been a recognised fate of bacteria for a long time. Stolp and

Starr (1965) described the fate of an individual bacterium undergoing

lysis as a kind of death which has no parallel in higher organisms. They

suggested that, beyond the loss of vital functions (metabolism, growth

-211-

and reproduction) death by lysis is characterised by the concurrent and

abrupt dissolution of the individual. Involuntary autolysis, or the

spontaneous lysis of a cell without the aid of an exogenous agent can

occur whenever the metabolism of the cell is drastically disturbed. For

example, in cultures of Bacillus subtilis growing exponentially,

autolysis can be induced either by a sudden interruption of oxygen

supply in aerated cultures or by the addition of metabolic inhibitors

such as KCN, NaN3 or dinitrophenol (Meyer, 1985). In the natural

environment, opportunists, microbes which are capable of inducing

bacterial lysis, have evolved and are able to grow on microbial lysis

products. This is typical of the behaviour of the Myxobacterium spp.

which are commonly found in soil and in waste sludge. Attempts have been

made to harness such behaviour by using these bacteria, and others with

similar potential, in the biological control of cyanobacterial blooms in

fresh surface waters (Burnham, 1981; Wright and Thompson, 1985). Such

processes have also been shown to apply to bacterial growth on the lysis

products of green algae in the epilimnion of lakes and could thus be of

importance in nutrient cycling in such environments (Uehlinger, 1986).

Chapman and Grey (1986) recently discussed the importance of the lysis

of dying microbes with respect to recycling of nutrients in the soil

environment via the process of "cryptic" growth, and "cryptic" growth

has also been recently recognised as a feature in animal. cell culture

where the effects of shear in bioreactors containing hybridoma cells can

result in the lysis of these fragile, shear-sensitive, wall-less cells

thus providing aditional substrate availability (D.F. Ollis, Personal

Communication) .

The model presented in chapter three can be simplified if all the

processes resulting in the production of dead cells are omitted from the

scheme. The result describes a culture in which the viable biomass is

reduced predominantly as a result of lysis phenomena, but with a

hitherto unproven loss by genetic inactivation. This latter process is

certainly relevant when a culture is subjected to super-optimal but not

growth inhibitory temperatures and capitalisation of the sustained

metabolic potential of such cells could have benefits in

-212-

biotechnological production processes requiring high product yields with

concommitant low biomass yield coefficients. Similar applications of

this concept could be achieved by selective irradiation of a culture

such that low level genetic damage results, thereby preventing cell

replication but not inhibiting metabolic activity. Obviously in

continuous culture processes such action could be directed towards a

significant fraction of a culture, simultaneously ensuring that a

sufficient fraction of "replication-positive" cells remains to prevent

washout. Experimentation must follow the development of techniques which

can distinguish DNA injury with respect to replication, such as by

specific enzyme assays for DNA polymerase, or by flow cytometry.

Since lysis is the most likely cause of biomass reduction in continuous

cultures, two product classes result; soluble organic material i.e.,

proteins, nucleic acids, carbohydrates etc. and particulate organic

material. Whilst measurement of the dissolved fraction is relatively

simple, experimental quantification of the particulate fraction is more

difficult. Differentiation between particulate matter and soluble matter

is usually artificial, i.e., when does particulate matter cease to be

particulate and be considered soluble? In the experimental procedure

described in chapter three, a particulate was artificially defined as

being anything that fails to pass through a 0.4pm membrane filter. To

demonstrate that this definition is not necessarily adequate, the

following experiment was conducted:

From a steady state chemostat culture of !· pneumoniae (D=0.077 h-1),

200 ml culture broth was removed and centrifuged at 15 1 000 x g for 10

min. The supernatant was decanted and filtered through a series of

different pore-size filters and the DOC in the resulting filtrates

determined. The results are shown below in Table 1.

A 14% difference in the amount of DOC present can be demonstrated when

the DOC in the filtrates are compared after filtration through very

small and very large pore size filters. The quantification of

particulate species in a heterogenous liquid environment is, therefore,

-213-

complex and detection may require the application of highly

sophisticated techniques, e.g., immunofluorescence microscopy for the

detection of cell wall fragments, to clearly differentiate between

soluble and particulate matter.

Table 1. The effect of filter pore size on particulate

retention.

Filter DOC

Pore Diameter

(pml -1 (mg 1 )

1.0 79

0.8 79

0.6 78

0.4 79

0.2 74

0.1 74

0.05 69

0.015 68

The experimental results for the soluble organic lytic fraction, when

compared with the computer predictions (Chapter 3, Fig.5) are

unsatisfactory. Unfortunately during the preparation of the data for

this publication, the carbon content of the lytic products was not

considered, such that the predicted line shown in this figure is

incorrect. Additional experimental points were obtained after submission

of the paper and these are shown below in Figure l together with the

recalculated data for the predicted DOC concentration assuming that 50'

I O>

c: 0 ..c ,_ <U u 0 c: <U O> ,_

0

"C (I.> > 0 (/) (/)

0

-214-

of s 2 (dissolved organic lytic product) is carbon. The carbon present in

EDTA and in polypropylene glycol have been taken into account in the

calculation of the experimental data points. The trend appears from the

additional data points to be effectively constant with increasing

dilution rate. The model predicts, more or less, the same general trend

but disagrees with respect to the absolute quantities. One possible

reason for this discrepancy is the assumption made in the model that all

the soluble lysis products released are indeed biodegradable. The

experimental data for DOC su9gest that this might not be the case and

that a part might be either non-biodegradable or only very slowly

biodegraded.

150

0 120

0

90 0 0 0

0 0 00 0 0 0 0 0 0 0

~ 0 0 60 0 0

30

0 0 0.2 0.4 0.6 0.8

Dilution Rate ( h-1)

Figure 1. Steady state concentrations of dissolved organic carbon (points) and the computer prediction (line) from the model described in chapter 3, and corrected for the carbon content of the released lytic product.

-1.0

-215-

To investigate this, and to determine experimentally the lysis rate

constant, it would be necessary to characterise the lysis products of K.

pneumoniae and to identify one component which is either recalcitrant or

slowly biodegraded. It would then be possible to determine lysis

kinetics by quantifying this fraction at different dilution rates.

The computer programme used for the mathematical calculations of the

model is shown in the appendix (A2). To examine the sensitivity of the

model to the choice of constants and coefficients a sensitivity analysis

was carried out whereby the effects of increasing and decreasing

individual constants were determined on the predictive response of the

model. The following coefficients and constants were particularly

sensitive:

Yield of non-viable cells by enzymic conversion of primary

substrate. Clearly whilst the non-viable cells are incapable of

replication they could, from incorrect choice of this coefficient,

conceivably continuously increase in mass. The only mechanisms limiting

this are lysis and washout from the system. When this constant was

increased the model predicted that the overall biomass would increase,

particularly that of the viable cell fraction. The reverse was the case

for a decrease in the value of this coefficient.

p• 2

: The maximum specific growth rate constant an dissolved organic

carbon. This constant was also sensitive, affecting most markedly, the

quantity of DOC accumulating in the system. Thus, after examining the

data in Fig. l above, a change in this constant could result in better

agreement between model and experiment. The value for this coefficient

was selected at random. However, the results describing "cryptic" growth

kinetics in chapter four clearly show that the value used in the

original model was unsatisfactory.

K4 , K8 and K3 : The death rate constant of viable cells, the lysis rate

constant of viable cells and the inactivation rate constant of viable

cells were also sensitive parameters affecting the predictions of the

model. It would probably have been justified, in the light of the

results, to have had a single term for K4 and K8

when K7 (the lysis rate

constant of dead cells) is assumed to be unity.

-216-

The modified model, similar to that described above, is presented in

chapter four where particular attention is focussed on "cryptic" growth.

In this model only K1 , the lysis rate of viable cells is sensitive, any

variation in this constant radically changes the quantitative prediction

of the model. Thus, it is absolutely essential and unfortunate that it

was not possible to adequately assess this rate constant. The computer

programme for calculating the predictions for this modified model is

given in appendix A3,

Some limited attention has been focussed on this question in the

literature. For example, Drozd .:.!: !_!. (1978) reported on the lysis of a

Methylococcus sp. in continuous culture. Since this bacterium is

incapable of assimilating lytic products, steady state measurements of

the dissolved organic carbon concentration in the culture supernatant

could be directly related to the specific lysis rate, the metabolic

pathway in this obligate methanotrophic bacterium resulting in the

production of co2 and biomass only. Drozd and co-workers found that

there was a direct relationship between dilution rate and the specific

lysis rate, with the rate of lysis increasing with increasing dilution

rate.

Meyer and Wouters (1984) examined the lysis of Bacillus subtilis var

niger WM in continuous culture. They were able to induce an autolytic

response in this bacterium by interfering with energy generation, and

could measure the extent of lysis by monitoring the decline in optical

density. In contrast to the work by Drozd and co-workers, Meyer and

Wouters found no relationship between lysis rate and growth rate nor

between lysis rate and growth limitation when comparing cells derived

from potassium-, phosphorous-, carbon- and magnesium-limited chemostat

cultures.

These two papers clearly demonstrate different facets of lysis. The

paper by Drozd .:.!: al., is relevant to in vivo lysis whilst Meyer and

Wouters are more interested in the potential activity of autolytic

enzymes when induced. Whether the two are necessarily related remains to

-217-

be shown. Thus, there is some justification, until more data becomes

available, to assume that the in vivo lysis rate is dilution rate

dependent as was assumed in the case in the model describing "cryptic"

growth.

The levels of dissolved organic carbon found in chemostat cultures of ~·

pneumoniae were lower in the experiments reported here, where the pH was

maintained at 6.80 and the temperature at 35.o0 c, than the

concentrations found in cultures of the same organism when cultured in

oxygen saturated, glucose limited chemostat cultures at pH 6.50 and at

30.o0 c. Harrison and Loveless (1971) report a DOC concentration of 255 ! 50 mg 1-l at a dilution rate of 0. 2 1 h-l which is much higher than

values obtained under the different culture conditions used in the

experiments reported here. Thus, there is suggestion of the potential

for increased lytic activity on changing from optimised to non-optimum

growth conditions.

It is possible also that in this current series of experiments the

observed DOC could have arisen either as a result of mechanical damage

or as a result of shear stresses from the impeller. These possibilities

were discounted after quantifying proportionately similar levels of DOC

accumulation at slower impeller speeds, when using lower concentrations

.~of glucose in the feed to ensure oxygen sufficient operation with lower

impeller speeds.

In the fundamental section of this thesis, there has been some comments

on the concept of endogenous metabolism with respect to starvation

survival. Such concepts are now relatively old. Lamanna (1963) described

a starving cell as one in which a gradual decline in the amount of

substance occurs such that it gradually wastes away and dies. He

suggests that it would be much more efficient if the cell, when deprived

of external resources, could automatically cease endogenous metabolism

and rest in a state of suspended animation. Clearly some 23 years later,

whilst knowledge concerning endogenous metabolism has advanced little,

it is now also known that some cells can ,exist in either a resting or a

-218-

dormant state (Novitsky and Morita, 1983; Stevenson, 1978).

Nevertheless, some of the current concepts are also somewhat odd. For

example, in an article by Kjelleberg et al. (1983), the behaviour of a

Vibrio sp. and of a Pseudomonas sp. were examined at interfaces

following the induction of starvation. These investigators found that

the initial events could be characterised by rapid fragmentation of the

cells followed by a continuous size reduction. The fragmentation gives

rise to an dramatic increase in the number of cells during the first 1

to 2 h, and the subsequent dwarfing apparently occurs over the first 4

to 5 h, during which time there is no loss in viability and measurable

metabolic activity. A similar sequence of events for another microbe has

been reported by Morita (1982). This series of stages of adaptation to

starvation prompts several questions; i.e., does this rapid division of

a mother cell into two daughter cells at the onset of starvation occur

such that a full compliment of DNA is present in each daughter cell? If

so, why does the bacterium invest valuable energy into rushed completion

of DNA replication instead of using the partially formed DNA as a source

of energy? How does the microbe "know" that it is going to have to

survive for a long period in the absence of exogenous nutrients such

that it initiates this sudden replication and rapid dwarfing effects

seen at the onset of starvation?

Clearly our understanding of starvation-survival is at present very

limited and many conflicting ideas continue to co-exist. However, whilst

these fundamental studies were undertaken to provide a framework to

understanding the effects involved in the destruction of microbes in

aerobic thermophilic sludge treatment, attention was directed towards

the question of microbial decline in general and the last paper in this

section has specific relevence to heat stress as well as starvation and

shows some of the problems involved in data interpretation in

starvation-type studies.

-219-

Microbial death, lysis and •crypt.ic• growth - applied aspects.

The utilisation of particulate substrates by microorganisms has been

neglected in the development of microbial growth theory and in much of

the work concerned with understanding the processes involved in waste

sewage sludge treatment. It is unfortunate that many reports concerning

anaerobic treatment processes depict it as being a sequence of stages

begining with simple soluble substrates e.g., proteins, lipids,

carbohydrates, and ignore the process stages involved in the production

of these simple substrates from more comlex polymeric and particulate

matter.Therefore, it was of considerable importance in this study that

the same deficiency should not occur. This was the main reason for using

a defined reproducable feed composed of whole yeast cells for the

process research programme. As a consequence, it was possible, through a

series of fundamental studies, to propose a mechanism for the

biodegradative process and thus appreciate the efects of varying

parameters such as temperature, pH, dissolved oxygen etc. on overall

process performance. The results clearly show that an ideal set of

conditions for microbial solids degradation would consist of low

dissolved oxygen conditions, a temperature between 6o0 c and 65°c and a

pH between 6. 5 and 7. 5. At present, where this process is used in

practise, only pH and temperature are routinely examined, but

measurement and control of dissolved oxygen concentration is rare.

Nevertheless, of the treatment plants which have been looked at in

experimental studies, the often high levels of carboxylic acids produced

indicate that they must operate at low dissolved oxygen concentrations

(Hamer and Zwiefelhofer, 1986). That this process can function optimally

with repect to both hygienisation and stabilisation has been shown very

nicely in practise at a sewage treatment plant in Marstatten, Kanton

Thurgau, Switzerland, where, despite the fact that the sludge storage

tank is open to the atmosphere, no recontamination or regrowth of

Enterobacteriacae has been found to occur during prolonged storage of

the treated sludge (H.P Zwiefelhofer, Personal Communication).

-220-

The Swiss Federal Ordinance of 1981, which states the hygiene

requirements of treated sludge makes no reference to the survival of

viruses. However, in a project conducted by Wyler and colleagues at the

University of Zilrich, the question of viral susceptability in the

aerobic thermophilic sludge treatment process has been examined

(Traub,1985). A technique was used which involved entrapping viruses

between polycarbonate membranes and suspending this in an aerobic

thermophilic bioreactor. It was found that viral inactivation occurred

under the thermophilic aerobic process conditions. However, some

viruses, notably Parvovirus, were resistant to short term exposure

( < 2h) at 60°c and would thus be more susceptable under semi-continuous

process operation than under continuous flow operation where washout

from the bioreactor could occur before deactivation. The pH was also

shown to be of relevence for the rate of viral inactivation with alkali

pH values, i.e., pH 8.0 being inducive for viral destruction. However,

the results of the present investigation into the effects of pH in this

study showed that operation at such high pH values would not necessarily

be conducive to optimum solids removal.

Similar results with respect to viral inactivation were described

recently by Dizer ~ al. (1986) and by Filip ~ al. (1986) where they

also investigated the fate of Enterobacteriaceae and of various other

groups of bacteria during aerobic and anaerobic thermophilic sludge

treatment processes. ~· coli and Salmonella spp. were effectively

destroyed at temperatures above 50°c under both aerobic and anaerobic

conditions, whilst some bacteria, notably sulphate-reducing bacteria and

Clostridium spp., were not destroyed even at 60°c under anaerobic

thermophilic conditions. No results were reported for the survival of

this group of bacteria during aerobic thermophilic treatment. The

possibility that these were thermophilic strains was not discussed

although it was suggested that the Clostridium spp. were probably spore

forming strains.

-221-

The presence of pathogenic spore-forming bacteria in feeds to aerobic

thermophilic waste sludge pre-treatment processes is a potential problem

for effective hygienisation. Representatives of this group are both

anaerobic and aerobic, e.g., £· botulinum and Bacillus anthracis, and

whilst spores are typically resistant survival structures to stresses

such as temperature, the use of either aerobic thermophilic or anaerobic

thermophilic treatment is unlikely to have any effect on such organisms.

In this context, it is probable that effective sludge stabilisation,

i.e., the degradation of all the potentially biodegradable matter

present, is as important for sludges containing spore-forming pathogenic

bacteria as with respect to the possible regrowth of Enterobacteriaceae.

Should stabilisation be inadequate, both aerobic and anaerobic

spore-forming bacteria could conceivably revert to their vegetative

forms and propagate in large numbers during the subsequent process

stages, a feature of definite concern with respect to the practise of

sludge disposal on agricultural land. It might be advantageous to

undertake routine counts of spore-forming bacteria in a similar manner

as for Enterobacteriaceae in the evaluation of safety for disposal of

the treated sludge on agricultural land.

Whilst the results and discussions presented in this dissertation are

part of a process research programme designed to investigate the

mechanisms involved in particulate degradation in aerobic thermophilic

sludge pre-treatment processes, there remains the question as to its

applicability to real feeds. Investigations to understand this problem

remain to be carried out both under controlled laboratory conditions and

during small pilot-scale operation using a feed of microbial cells

together with hydrocarbons, cellulose and lipids. This would be a

natural progression for this work, and only then could reliable

predictions be made with respect to the effective treatment of real

waste sludges .• What this work has shown, is the basis for efficient and

effective operation. Many of the results have already been seen to have

parallels under real process operating conditions. Without further

controlled investigations, i.e., using a supplemented feed, the effects

of feed variability under real process conditions cannot be either

properly predicted or controlled.

-222-

With the increasing availability of sensitive and reliable tests for

bacterial activity assessment, e.g., the INT-reduction assay, it would

be of interest to monitor the activity of the thermophilic process

microbes under different process operating conditions. Particularly in

the case of semi-continuous process operation, optimisation of cycle

times can be investigated with respect to the variation in activity of

the process microbes. Thus it should be possible to define conditions

whereby microbial activity can be maintained high and solids removal

maximized by operating at a particular cycle time. Investigations of

this kind should also provide information as to the effects of weell.-

day /week-end operation particularly with respect the length of time

required for the activity to recover after week-end operation.

Only very limited microbiological investigations were conducted on the

aerobic thermophilic system, but these showed that a mixed culture was

certainly present. Whilst these microbes were selectively enriched under

laboratory conditions, it remains to be seen as to whether the relative

numbers of the different species are the same as found in the

technical-scale plant from which the laboratory culture was originally

derived. However, the culture showed a considerable degree of robustness

as judged by the fact that the physical conditions, e.g., temperature,

pH, dissolved oxygen concentration, bioreactor operating mode,

bioreactor residence time, were frequently changed .without the need to

reinoculate the bioreactors. High stability, robustness and lack of

fastiousness are one criteria discussed in chapter five for the choice

of cultures for biotechnological process exploitation.

To evaluate the effects of the variations in physical conditions on

optimum process performance, some further investigations into the

bacteriology, physiology and biochemistry of the process are necessary.

Accurate evaluation of the different species present, their

physiological classification and their relative numbers, should be

carried out. It is not known how many of the species present are merely

opportunists, nor is it known which species produce the lytic enzymes

for either cell wall lysis or the hydrolysis of other macromolecules

-223-

prior to assimilation. What is the spectrum of enzymes produced? Are the

enzymes either produced constitutively or are they in any way inducable.

In real systems in which bacterial cells are the targets for such

enzymes, what mechanisms prevent action of the enzymes on the process

microbes themselves, i.e., a suicide mechanism? What are the inhibitors

of the various enzymes? What are the temperature and pH.optima of these

enzymes? Are the kinetics and mechanisms derived for the degradation of

yeast cells also applicable to bacterial cells?

The answers to such questions are required for a fully comprehensive

understanding of aerobic thermophilic process control and optimisation

and must be addressed in the future.

It has been assumed that one fraction of the thermophilic aerobic

process microbial culture is facultatively anaerobic whilst another

fraction is probably aerobic and uses up the oxygen preferentially when

it is present at low concentrations. Obviously, operating under oxygen

restricted conditions has advantages with respect to minimising

potential heat loss in the exhaust gases. Nevertheless, the true nature

of the oxygen requiring fraction of the biomass remains undefined. In

the equations used in the models for both semi-continuous and continuous

flow process operation, the production of both new cells and of acetate

are presented in a single equation. This is, of course, an over

simplification and it is more likely that acetate and other carboxylic

acids are produced by the fa cul tatively anaerobic population whilst

direct biomass production is a strictly aerobic process.

The process of "cryptic" growth was shown to be potentially important in

the regulation and deoptimisation of microbial product formation in the

fundamental section of this thesis. Its effect is not known in the

aerobic thermophilic process although from the sensitivity of the

mathematical constant used for the death/lysis of process microbes in

the models presented in chapters eleven and thirteen, the effect must

also be more closely evaluated. Moreover, the effects of lysis and

"cryptic" growth are likely to be advantageous in this system, resulting

-224-

in lower overall biomass yield coefficients and would, therefore,

represent a desirable process feature.

Differentiation between the effectiveness of semi-continuous and

continuous flow operation has not been directly assessed in this work.

Nevertheless, several features have become apparent. Semi-continuous

flow offers the advantage that a guaranteed minimum residence time,

i.e., the time available for the interaction between process bacteria

and substrate microbes is predetermined and can be guaranteed. This

allows greater confidence in the hygienisation potential of such systems

than in the case of continuous flow processes where the minimum

residence time in heterogeneous, incompletely mixed systems can approach

zero, thus permitting the possibility for short-circuiting of treatment

with respect to both pathogenic microorganisms and toxic chemicals. The

complete utilisation of both rapidly biodegradable and slowly

biodegradable substrates is more likely to occur in semi-continuous

systems than in contionuous flow operation due to the preferential

utilisation and continuous supply of the rapidly biodegradable

substrates under the latter mode of process operation. Furthermore, in

semi-continuous operation there is an enforced contact of the slowly

biodegradable fraction as the sole nutrient source after exhaustion of

the rapidly biodegradable substrates. The continuous flow process has

the advantage that either any potentially toxic or inhibitory metabolic

product will be continuously diluted in the system whilst under

semi-continuous process operation markedly higher concentrations of such

compounds can occur. Accumulation of high molecular weight carboxylic

acids occurred under both modes of process operation and is

disadvantageous because of the mal-odour these compounds impart to the

treated sludge. • However, since the aerobic thermophilic process is

proposed as a pre-treatment, the use of effective post-aerobic,

mesophilic anaerobic digestion should eliminate this problem, and in.

effectively functioning, technical-scale, waste sludge treatment plants

this is seen to be the case.

-225-

Clearly aerobic thermophilic treatment is inappropriate as a complete

treatment process and effective removal of the biodegradable matter from

waste sludges requires a second treatment stage. Nevertheless, an

aerobic thermophilic process stage conditions the sludge for faster and

more effective subsequent mesophilic anaerobic digestion and thus offers

advantages over conventional single stage mesophilic anaerobic treatment

in both this respect and in exhibiting higher hygienization potential.

References

Burnham,J.C. (1981) Entrapment and lysis of the cyanobacterium Phormidum luridium by aqueous colonies of Myxococcus ~- Arch. Microbiol. 129, 285-295.

Chapman,S.J., and Gray,T.R.G. (1986) Importance of cryptic growth, yield factors and maintenance energy in models of microbial growth in soil. Soil Biol. Biochem. 18, 1-4.

Dizer,H., Leschber,R., Lopez Untersuchungen zur anaerob-thermophile 703-709.

Pila,J.M. and Seidel,K. (1986) Entseuchung von Klarschlamm durch

Behandlung. Korrespondenz Abwasser JJ,

Drozd,J.W., Linton,J.D., Downs,J. and Stephenson,R.J. (1978) An in situ assessment of the specific lysis rate in continuous cultures of Methylococcus sp. (NCIB 11083) grown on methane. FEMS Microbiol. Lett. 4, 311-314.

Filip,Z., Dizer,H., Leschber,R. and Seidel,K. (1986) Untersuchungen zur Entseuchung von Klarschlamm durch aerob-thermophile Behandlung. Zbl. Bakt. Hyg. B 182, 241-253.

Harrison,O.E.F., and Loveless,J.E. (1971) The effect of growth conditions on respiratory activity and growth effeiciency in facultative anaerobes grown in chemostat culture. J. Gen. Microbiol. 68, 35-43.

Kjelleberg,s., Humphrey,B.A. and Marshall,K.C. (1983) Initial phases of starvation and activity of bacteria at surfaces. Appl. Environ. Microbiol. 46, 978-984.

Lamanna,c. (1963) Studies on endogenous metabolism in bacteriology. Ann. NY. Acad. Sci. 102, 517-520.

-226-

Meyer,P.D. (1985) Cell wall autolysis and turnover in Bacillus subtilis. PhD thesis, University of Amsterdam.

Meyer,P.D. and Wouters, J.T.M. (1984) Is autolysis of Bacillus subtilis subject to phenotypic variation? In: Microbial Cell Wall synthesis and Autolysis. (C. Nombela, Ed.) pp 219-223, Elsevier Science Publishers, Amsterdam.

Morita,R. Y. (1982) starvation-survival of heterotrophs in the marine environment. Adv. Micro. Ecol. 6, 171-198.

Novitsky,J.A. and Morita,R.Y. (1977) Survival of a psychrophilic marine vibrio under long term nutrient starvation. Appl. Environ. Microbial. 33, 635-641.

Stevenson,L.H. (1978) A case for bacterial dormancy in aquatic systems. Microb. Ecol. 4, 127-133.

Stolp,H. and Starr,M.P. (1965) Bacteriolysis. Annu. Rev. Microbial. 19, 79-104.

Traub,F. (1985) Virusinaktivierung bei der Klarschlammbehandlung. Verband Schweizerischer Abwasserfachleute Bericht Nr 298.

Uehlinger,U. (1986) Bacteria and phosphorous regeneration in lakes. An experimental study. Hydrobiologia 135, 197-206.

Wright,S.J.L. and Thompson,R.J. (1985) Bacillus volatiles antagonise cyanobacteria. FEMS Microbial. Lett. 30, 263-267.

-227-

SUMMARY

The processes involved in and the factors affecting death, lysis and

"cryptic" growth have received little or no attention in microbiological

research. An understanding of these processes is particularly necessary

in research programmes designed to investigate the mechanisms involved

in and optimisation of microbial solids destruction in waste sludge

treatment processes.

In the work presented here, research at the fundamental level was

undertaken to investigate the processes of death, lysis and "cryptic"

growth in the absence of any form of stress, i.e., under optimal growth

conditions. The results suggest that death only results from either

genetic injury or by the application of stresses e.g., heat. Biomass

reduction during cultivation is suggested to occur as a result of lysis

events rather than through inactivation of metabolic functions. As a

result of the release of cellular components following lysis, the

process of "cryptic" growth, i.e., the utilisation of the lytic products

by members of the same population as carbon, energy, or other nutrient

sources can occur. Whilst this results in a complex series of events

during the growth process, a mathematical model is derived to aid in the

understanding of these various reactions and raises questions as to the

validity of using poorly-defined concepts such as maintenance energy to

describe physiological events in dynamic culture systems.

The application of heat is frequently used in the deactivation of

microorganisms. In a simple study, the physiological changes which occur

to a population of bacteria as a result of mild starvation together with

heat treatment were examined and also evaluated with respect to the

effects on individual cells. At temperatures slightly above the optimum

for growth, anomalous changes with respect to macromolecular

constituents per cell were obtained suggesting that the concepts of

endogenous metabolism alone are inappropriate to describe starvation or

other stress effects when changes in physical properties, e.g., wall and

membrane permeability are not simultaneously considered.

-228-

In the applied part of this study, the mechanisms involved in and

factors affecting the destruction of microbial solids in aerobic

thermophilic waste sludge treatment were investigated. This process is

designed to supplement conventional anaerobic mesophilic digestion, such

that beth pathogen and toxic chemical removal efficiencies are enhanced.

Whilst most work conducted on sludge digestion has concentrated either

on soluble substrate utilisation or on the degradation of substrates of

non-microbial origin, this study was designed to investigate the

degradation of microbial solids. This fraction represents a major

constituent of the total suspended solids in sludges and is derived from

both primary and secondary wastewater treatment processes. The effects

of physical process parameters such as temperature, pH, dissolved oxygen

concentration and bioreactor residence time were examined together with

effects of nutrient supplementation. The process proved to be highly

effective with respect to microbial solids destruction at pH 6.5, 6S0 c and when the air flow into the bioreactor was regulated such that the

residual dissolved oxygen concentration was very low.

The mechanism of microbial solids destruction was investigated in a

batch type experiment where the fate of particular components of the

feed microbial solids such as protein, carbohydrates and cell number

were monitored during the degradative process. Coupling this with

scanning electron microscopy allowed the proposal of a mechanism for the

biodegradation process whereby the cell walls of the feed microbes are

attacked by exo-enzymes produced by the thermophilic process microbes.

Lysis of the cell wall results in the release of readily utilisable

substrates whilst residual wall biodegradation occurs more slowly. Under

conditions where the dissolved oxygen concentration is low, considerable

amounts of carboxylic acids are produced which are readily utilised when

the aerobic thermophilic process stage is used as a pre-treatment to

anaerobic mesophilic digestion. Moreover the rate of anaerobic

mesophilic digestion is likely to be enhanced as a result of the

extensive particle solubilization in the pre-treatment process,

-229-

ZUSAMMENFASSUNG

Prozesse im Zusammenhang mit Tod, Lyse und kryptischem Wachstum haben in

der bisherigen mikrobiologischen Forschung wenig bis keine Beachtung

gefunden. Ein gesichertes Wissen liber diese Vorglfoge ist besonders

wichtig im Hinblick auf Programme, die eine Untersuchung und Optimierung

des mikrobiellen Abbaus von Feststoffen im Klarschlamm zum Ziele haben.

In der vorliegenden Arbei t wurden die grundlegenden Prozesse von Tod,

Lyse und kryptischem Wachstum in Abwesenheit von Stressfaktoren, d.h.

unter optimalen Wachstumsbedingungen, untersucht. Die Resultate deuten

darauf bin, dass Tod nur infolge genetischer Schaden oder durch den

Einfluss von Stress, z.B. Hitze, eintritt. Eine Reduktion der Biomasse

wahrend der Kultivation scheint das Resultat von Lyse zu sein und nicht

auf einer Inaktivierung der metabolischen Funktionen zu beruhen. Als

Folge der Freisetzung von Zellkomponenten durch Lyse kann kryptisches

Wachstum auftreten, d. h. die Verwendung von Lyseprodukten als

Kohlenstoff-, Energie- oder sonstige Nahrstoffquelle durch Vertreter

derselben Population. Daraus resultiert eine komplexe Folge von

Vorgangen wahrend des Wachstums. Zurn besseren Verstandnis wurde daflir

ein mathematisches Model! entwickelt. Es stellt sich in der Folge die

Frage, in wie we it ungeniigend definierte Begriffe wie "maintainance

energy" brauchbar sind zur Beschreibung der physiologischen Vorgi!nge in

einem Kultursystem.

Zur Inaktivierung von Mikroorganismen wird oft Hitze eingesetzt. Die

physiologischen Veranderungen einer Bakterienkultur als Folge einer

leichten Aushungerung zusammen mit einer Hitzebehandlung, wurden in

einer weiteren Studie untersucht bei gleichzeitiger Beobachtung von

Einzelzellen. Bei Temperaturen wenig oberhalb der optimalen wachstums-

temperatur wurden Veranderungen in den makromolekularen Inhaltsstoffen

von Zell en beobachtet. Das bedeutet, dass das Konzept des endogenen

Metabolismus alleine nicht geniigt um den Hungerzustand oder andere

Stressfolgen zu erklaren. Es mlissen gleichzeitige Veranderungen in den

physikalischen Eigenschaften der Zelle, wie z.B. bei der Permeabilitat

der Zellwand und der cytoplasmatischen Membran, angenommen werden.

-230-

Im angewandten Teil der Arbeit wurden die Faktoren, welche den

mikrobiellen Abbau von Feststoffen beeinflussen, wie auch der

Abbaumechanismus als solcher in aerob thermophilen Behandlungsreaktoren

von Kliirschlamm untersucht. Dieser Prozess dient als Zusatz zur

mesophilen biologischen Faulungsstufe und soll garantieren, dass der

Kliirschlamm am Ausgang der Anlage frei von pathogenen Organismen ist.

Gleichzeitig wird auch eine Verbesserung in bezug auf die Entfernung

umwelttoxischer Substanzen erwartet. Wiihrend der grosste Teil von

Arbeiten Uber den Abbau von Kliirschlamm sich entweder auf die Entfernung

geloster Stoffe oder auf den Abbau von Feststoff en nicht

mikrobiologischen Ursprungs beschriinkte, wurden in der vorliegenden

Arbeit der Abbau der mikrobiellen Feststoffe untersucht. Diese stellen

den Hauptteil der suspendierten Feststoffe im Kliirschlamm der ersten,

wie auch der zweiten Abwasserbehandlungsstufe dar. Der Einfluss der

physikalischen Prozessparameter wie Temperatur, pH, Konzentration des

gelosten Sauerstoffs und Aufenthaltszeit im Bioreaktor wurden zusammen

mit den Veriinderungen, die durch Zugabe von weiteren Niihrstoffen

hervorgerufen wurden, gepriift. Es zeigte sich, dass der Abbau

mikrobieller Feststoffe sehr effizient erfolgte bei einem pH-Optimum von

6.5 und einer Reaktortemperatur von 65°c sowie einer Beliiftungsrate, die

zu sehr kleinen Konzentrationen an gelostem Sauerstoff fiihrte.

Der Abbau von mikrobiellen Feststoffen wurde in 'Batch'-Versuchen

abgekliirt, so dass die einzelnen partikuliiren Komponenten wie Protein,

Kohlenwasserstof f e und Zellzahl wiihrend der Abbauprozesse verfolgt

werden konnte. Dies, kombiniert rnit Rasterelektronen-Mikroskopischen

Aufnahrnen erlaubt den Vorgang wie folgt zu charakterisieren: zuerst

werden durch Exoenzyme der thermophilen Organisrnen die Zellwl!nde der

gefUtterten Mikroben angegriffen, die darauf folgende Lyse setzt leicht

verwertbare Substrate frei, wiihrend der biologische Abbau der

Restzellwand langsamer erfolgt. Bei kleiner Konzentration an gelostem

Sauerstoff werden beachtliche Mengen von Carbonsiiuren produziert. Diese

konnen in einer anschliessenden, mesophilen, anaeroben Faulstufe leicht

verwertet werden. Zudem wird der anaerobe Abbau dieser Faulstufe

verbessert durch den Aufschluss der partikull!ren Stoffe irn

vorbereitenden therrnophilen Prozess.

-231-APPENDIX

Al

Iodonitrotetrazolium Chloride (INT)

Properties

Ivory white crystalline powder, m.p. 24S0 c. Slightly soluble in methanol

and water, very slightly soluble in acetone,ethyl acetate and

tetrahydrfuran, insoluble in ethyl ether. Reduction potential about

-0.09 volts. Yields a red violet monoformazan pigment in the presence of

dehydroqenases (LOH, SOH, etc.). Monoformazan melts at 185.6°C.

Uses

As a histochemical and cytochemical reagent for detection or

determination of dehy?roqenases, NAO and NAOP.

(Reference: Fluka Technical Bulletin).

-232-

APPENDIX

A2.

PYNEO 10 ON ERROR GO TO 2000 SO REH PROGRAM BY G,A,H, KINGr J, ECOL. MODELLING Sr 259-268 (1978) 100 REH CHEMICAL EQUATIONS CHEMED 110 REH PREPARE LIST OF CHEMICAL SPECIES 120 REH FIRST SPECIES IS A BLANK 'BOX' 130 REH PREPARE LIST OF EQUATIONS 140 REH CONVERT EQUATIONS TO NUMBER FORM 150 REH LIST RATE COEFFICIENTS 160 REH LIST INITIAL CONCENTRATIONS FOR SPECIES 170 REH CONCENTRATION OF FIRST SPECIES IS PUT TO ONE 180 REH SPECIES AND EQUATIONS DIMENSIONED TO 20 190 REH ---------------------------------------------200 DIH E140r4lrRl40r3loVl40lrWl40l 201 DIH J(8lrNC4014lrEOl40)rEll40lrE2140l 202 DIH RICISlrR211SlrRSl15) 203 DIH Pl4) 204 FS='' 205 GOSUB 1100 207 INPUT 'INTEGRATION CRITERION='rX 210 PRINT'ENTER NUMBER OF VARIABLES -ENTRY PT I' 220 INPUT NI 230 PRINT'ENTER NUMBER OF PROCESSES -ENTRY PT 2' 240 INPUT II 250 PRINT'ENTER EQUATIONS - HAX. 8 NUMBERS -ENTRY PT 3' 251 PRINT'INPUT NlrV1rN2rV2r••••' 252 PRINT'Vl,,V4 = VARIABLESr Nl,,N4 =STOICHIOMETRIC FACTORS' 253 PRINT'NI<O FOR CONSUHPTIONr NI>O FOR PRODUCTION OF VARIABLE' 260 FOR I=I TO II 262 HAT J=ZER 264 HAT INPUT J 266 Kl=I 268 K2=INTC0,5+Kl/2l 270 NCirK2>=JIKI) : EIIrK2l=JCKl+I> 273 IF K2>=4 GOTO 277 274 IF JIK1+3><>0 THEN Kl=Kl+2 : GOTO 268 277 E11Il=K2 280 NEXT I 290 PRINT 300 PRINT'ENTER PROCESS FUNCTIONS -ENTRY PT 4' 301 INPUT'DO YOU WANT LIST IYES/N0l'IL$ 305 L$=LEFTIL$1IX) 310 IF L$<>'N' THEN GOSUB 1000 311 R8=0 312 PRINTlPRINT'PROCESS t'r'TYPE t' 315 FOR I=I TO II 320 PRINT I1 325 INPUT K : EOIIl=K : E21I>=R21K) 326 IF EllI><RllKl THEN PRINT'NOT ENOUGH VARIABLES FOR PROCESS'lGOTO 320 328 NEXT I 330 PRINT'ENTER RATE CONSTANTS -ENTRY PT S' 332 FOR I=I TO II 334 K=E21Il : PRINT'PROCESS 'III' NEEDS 'IKI' CONSTANTISl'I 336 HAT J=ZER : HAT INPUT J 337 FOR Kl=I TO K : RIIrKl)=J(KI) : NEXT Kl 338 IF EO<I>=8 THEN R8=0,0S*K3 339 NEXT I 340 PRINT'ENTER INITIAL CONCENTRATIONS -ENTRY PT 6' 350 FOR N=I TO NI 360 PRINT No : INPUT VIN) 370 NEXT N

-233-

380 T9~0 390 PRINT'SPECIFY WHICH VARIABLES PRINTED <HAX,4) -ENTRY PT 7' 392 HAT J=ZER : HAT INPUT J I P1=1 395 IF JCPll•O GOTO 403 397 PIP1l•JIP1> : P1•P1+1 : GOTO 395 403 Pl•Pl-1 : IF P1<1 GOTO 390 405 IF P1>4 THEN PRINT'TOO HANY OUTPUT-VARIABLES'! GOTO 390 410 PRINT'SPECIFY PRINT INTERVALL AND LIMIT -ENTRY PT 8' 420 INPUT T1 • T2 421 IF Ft<>"' THEN PRINT'DATA FILE OPEN ALREADY' I GOTO 955 422 PRINT'OUTPUT ON FILE (YES/NOlT -ENTRY PT 9' 424 INPUT Xt : XS•LEFTCXl1lX> 426 IF XS•'N' GOTO 432 428 INPUT 'NAHE OF FILE'I Fl 430 OPEN Ft FDR OUTPUT AS FILE 432 PRINT!PRINTIPRINT 440 PRINT'T'• 441 FOR P2=1 TD Pl l PRINT P<P2lr NEXT P2 443 IF P1<3 THEN PRINT 450 REH 460 T•T9 470 T4=0 474 PRINT Tr 475 IF P1=1 GOTO 495 480 FOR P2•1 TO Pl-1 483 KcP(P2l 485 PRINT VIK l 1

490 NEXT f'2 495 K•P1 : PRINT V<Kl 512 IF X••'N' GOTO 520 514 PRINT11rT'r'TO'•'I 515 IF N1s1 GOTO 518 S16 FOR N=l TO Nl-1 517 f'RINTtlrV!Nl'r'I l NEXT N 518 f'RINTl1rVIN1l 520 IF T>•T2 THEN 940 540! BEGIN LOOP <TO LINE 900)

·541 HAT W•ZER 550 FOR I•l TO I1 560 ON EO!Il GOTO 610r620r630164016S0r660r670r680 610!PROCESS 11 612 Y•R<I, 1) 615 GOTO 800 620!PROCESS t2 623 V1•E1Ir1l 625 Y•R(Ir1l*V<V1) 627 GOTO 800 6JO!PROCESS U 632 Vl•EIIrll 635 Y•R(I,1l*VIV1!*VIV1l 637 GOTO 800 640!PROCESS 14 642 Vt-E<Ir1) : V2•E(I,2) 645 Y•R<Ir1l*VIV1l*V(V2l 648 GOTO 800

650!PROCESS tS 652 V!=E<I.tl

-234-

655 Y=R<I•l>*V(Vlll<R<Ir2ltV(VI)) bSB GOTO BOO 660!PROCESS 16 662 Vl=E(lrll : V2=E<I•2> 66S Y=R<l•1l*V(Vl)*V(V2)/(R{lr2)tV{V1)) 667 GOTO BOO 670!PROCESS 17 672 Vl=E<I•ll : V2=E!It2l 675 Y=RCl•1l*V<V1l*V<V2l/CRCl•2>+V<Vll)/(R(It3ltV(V2ll 670 GOTO BOO 6BO!PROCESS IB 692 Y=R(l,ll+RCit2l*SIN<6.283105*T/R!l•3>> 604 GOTO aoo 800 FOR Kl=! TO EICil BOS K=ECltKU BIO W<Kl=WCKltN<I•Kll*Y BIS NEXT Kl 820 NEXT I BSO!CALCULATE TIHE STEP 051 TO=! 955 FOR N=I TO NI 060 IF W<Nl=O OR V(Nl=O GOTO 075 965 TB=A9SCX*V(Nl/W<N>> 870 IF TB<TO THEN TO=TB B75 NEXT N 880 IF TO><Tl-T4l THEN TO=Tl-T4 BBi IF RB>O AND TO>RB THEN TO=RB B92 T=T+TO : T4=T4tTO BBS FOR N=I TO NI 890 V(Nl=V<Nl+TO*W<N> 900 NEXT N 910 IF T4>=TI GOTO 470 920 GOTO 540 940 T9=T 942 IF X$='N' GOTO 970 944 ~ NEXT 5 LINES VALID FOR TONY ONLY ~===s==••=~==•••••••••=~•m•• 945- T2=T2+100 946 PRINT : INPUT 'D='•D : PRINT 947 IF D=O THEN GO TO 955 948 R<I•l>=D FOR I=12 TO 10 949 GOTO 540 955 INPUT'CLOSE DATA FILE CYES/NOTl'IXS 960 Xt=LEFT<XS11Xl 969 IF Xt='Y' THEN PRINTll•CHRt<26) l CLOSE 1 : FS='' 970 INPUT'CHOOSE ENTRY PT FOR NEXT RUN - USE 0 TO END SET'IH 900 IF H<O OR H>9 GOTO 970 985 IF H=O GOTO 2000 990 ON H GOTO 210.230,2so.300.330,J40.390.410•421 lOOO!LIST OF PROCESS EOUATIONS 1001!************************* 1010 I9=8 101S Rf(ll='Kl' 1016 RS<2l='Kl*VI' 1017 R$(3l='K1•V1*V1" 1016 Rt(4l='K1*Vl*V2' 1019 Rt(5)•'K1*V1/(K2+V1)' 1020 RS!6l="K1*Vl*V2/!K2tV1l" 1021 R$<7l='K1*CV1/(~2tV1ll*CV2/(K3+V2ll 1022 Rt(Sl='K1tK2*SIN<2*PI*T/K3l" 1050 PRINTlPRINT'TYPES OF EQUATIONS AREl' 105S PRINT'NUHBER'

1060 FOR 1=1 TO 19 1065 PRINT loR$CI) 1070 NEXT I 1075 PRINT 1080 RETURN

-235-

1100'HINIHUM NUM8ER OF VARIA~LES AND NUH8ER OF COEFFICIENTS FOR DIFFERENT 11011TYPES OF RATE EQUATIONS 1102!******************************************************************** 1110 HAT Rl=CON l MAT R2=CON 1115 RJC4)=2 1116 R2<5)=2 1117 R1C6l=2 R216l=2 1118 RIC7l=2 R2<7l=3 1119 R2(8)=3 1190 RETURN 2000 ENI.I

-236-

~ ! 'l'IHS ~ Jl,I:: SUllNl'l'!i 'l'llE l'llOGR/\f-1 UYNl::Q t'OR 'l'Oi;YS MOOEL s ! U>it: 'l'l!J:: coml/1NU suu • • • • • ( lllSl::HT PllOGi\ANMt: N/\Mr:) S! l"Ol< OU'l'l'U'1' USt: PlP t'lLt: •••.LOG 01! l<UNSll/\LlST ~·11,E ••••.LOG $ ! lJI;POIU:: SUUM1'1'1'ING II JOll l<t:PL/\Cr: 1\1.L Vl>lll/\1.11,E !11\Ml::S wn-11 l\USOLU'l'l:: S ! Vl\LUl:;S $! +++++++•~·+~++++++++•+++++++++++++++++++++++++++++++++++++++++++

SJOil/PlHOIU'l"i: -i4 $HUN L>'iNl::Q ~ ! IN'rl::GRJ\'l'ION CIH'l'El<ION

S ! NUMill::l< Of Vl\!Ui\DLES b S ! NUMBl::R 01" PROCESSES 18 $ ! IW'l'ER PROCESSl::S -1/'i(l,4),Sl, 'i(l.4),Xv -1 /Y ( 2, 4) , S2, Y ( 2, 4) , Xnv -1,Xv, l,Xnv -1,Xv, l,Xd -1,Xnv~ 1,xn -l,l\nv, 'i(6,2!,S2, 'i(b,7),P -1,Xd, 'i(7.2),S2, Y(7,7),P -1,Xv, Y(U,2),S2, 'i(B,7),P -1.P, 'i(9,2),S2 -1/Y(l0,5),Sl, Y(lO,S),Xnv -l/Y(ll,5),52, 'i(ll,5),Xnv So,Sl, S2o,S2 -1. Sl -l,S2 -1 ,Xv -1,Xnv -1,Xd -1,P $ ! l::NTl::R PROCl::SS !"UNCTIONS S l DO YOU NAN'r Ll ST y $ ! TYPI:: OP PROCESSl::S 6 0 2 2 2 6 6 2 2 2 2

.. 2 2 2 2 2

$I HA'l'I:: CONS'l'AN1'S u• L Ksl u•2,Ks2 kl k4 kS k& k7 k8 k8 v• 10, Kml V*ll ,Km2 0 [)

0 0 0 [) 0 $J INITIAL CONC Sl S2 Xv Xnv Xd p $1 PRINTED VARIABLES 1.2,J,4 $I INTERVAL /\NU LIMIT ••OELT , .. TENO $1 OUTPUT y ****.OUT

-237-

$I NEX'r LINES INTRODUCE NEW VALUES FOR 0 AS MANY TIMES AS DESIRED $I JUS1' REPEA'l' ON NEW LINE, ONCE ONLY f'OR EACH CYCLE. ••o ••o 0 y 0 $RUN $HELDE BA'l'CH TONY r'INISHEO $EOJ

-238-

APPENDIX

AJ.

PRINT '•====~=~#=•c=~==c~::uzc~:=m~==~~==z•~#=~=•=~=====~===~#~====~~$•

10 PRINT 20 PRINT 'MATHEMATICAL MODEL FOR GROWTH IN CONTINUOUS CULTURE' 30 PR INT 40 PRINT ·===•=:===~==a=====~====~,=~===·==·=~~·=·=======•s•~a==~====~~u~·

:;o f'P.INT 60 Pf<HIT 70 PR INT SO F'f:INT 90 EXTENL> 100 l•IH FEEDl 4) 110 HAT NUE=ZER(4r41 120 INPUT 'OUTPUT ON DATA FILE IYES/NOI 130 IF X$•'N' DOTO 190 140 INPUT 'NAHE OF DATA FILE 150 OPEN A$ FOR OUTPUT AS FILE 160 PRINT 170 INPUT 'DH HIN !SO INPUT 'DH HAX 190 INPUT 'DH STEP 200 INPUT 'OUTPUT ON SCREEN 210 ITCRIT•0,0001 220 F'flINT 230 INPUT 'YIELDl <YIELD 240 INPUT 'YIELD2 <YIELD 330 INF'UT 'HUE! 340 INPUT •KS! 350 INPUT 'HUE2 CXv-S21 360 INPUT 'KS2 130 !NF'UT 'KH 480 INPUT 'SO(! I 485 FOR !tH•!IHH!N TO [tHHAX 4ro K3•0.4*DH+0.02

Xv FROM Sil XV FROH S2l =· diUE 1

=' 11\Sl •"HUE2 :• 1KS:! = 1 rK4

'tSO(I) S TEF' STEPI•H

506 !F DH•O THEN GOTO 32767 510 NUEC!r!l•-1/Yl 520 NUEC!13)•1 540 NUE(3r3l•-l 550 NUE(3,2)=0.95*DH~O.l5 560 NUEC3r4l=!-NUE<3•2l 570 NUE<2r21•-l/Y2 580 NUE(2,3>=1 sro NUEC4r4l=-t 600 NUE(4r31=1 610 FOR !=1 TO 4 620 FEEDIIl•-SOCll*DH 630 NEXT I 640 Ll=HUEl/KS! 650 L2•11UE2/KS2

• rAf

'•DHMIN • 1DHl'!AX '' STEF'DH

•rot

840 I --------------------------------------------------------------------850 ! BEGIN OF ITERATION LOOP FOR STEADY STATE 860 ! --------------------------------------------------------------------870 LOOF'•() 880 HAT s~zER<1·4> 890 FOR I•! TO 4 900 S<I•ll•Ll*NUE!l•Il 910 S(J,2}2L2t.NUE(2,r> ?20 SCl•3l•K3*NUE<31Il 930 SCir4l•K4*NUEl41Il

960 HEY.T I 970 FOR I~l TO 4 980 S<I•ll•SII•I>-DH 990 NEXT I 1001 HAT S•INVISI 1010 HAT CSS•S*FEED 1020 L10Lll•L1 1030 L20Lll•L2

-239-

1060 L1•HUE1*CSS<Zl/CKS!tCS511)1 1070 L2~HUE2*CSS(J)/CKS~+CSSC2JJ 1100 LOOP•LOOP+l 1110 IF LOOP>2000 THEN PR!NT 'WASHOUT ! !••: GOTO 32767 1120 IF A8S(L10LD/L1-11>ITCRIT OR A8S<L20LO/L2-1l>ITCRIT 1170 RT•1/DH 1180 Y08S•CSSl3)/IS0(1)-CSS<lll 1190 as•ItH/YOBS 1200 IF Ot•'N' GOTO 1440 1210 PRINT 'NUHSER OF ITERATION LOOPS •'•LOOP 1220 PRINT 1230 PRINT 'STEADY STATE IS ' 1~40 PRINT •===~===========· 1250 PRIHT 1260 PRINT USIHG "DILUTION RATE • ttttttt,tttt'oDH 1270 PRINT 1260 PRINT USING 'X 1290 PRINT USING 'S 1295 PRINT USING 'SL 1340 PRINT USING 'P 1350 PRINT

tttfttt,tttt'oCS5(31 ttttttt.tttt•,css111 ttttttt.tttt•.cssc2> ttttttt.tttt·.csS<41

THEN GOTO eeo

1410 PRINT USING 'ODSERUED YIELD 1420 PRINT USING 'SPECIFIC SU85TRATE UPTAKE RATE 1HO PRINT

•tttttt.•tt',YOBS ~tt11tt.Ht'1a,;

1450 PRINT tloDH'1'CSSCl>'•'CSSC2J','CSSl31'•'C5S<41 1470 NEXT DH 32767 END

12 October 1959

1965 - 1973

1973 - 1977

1978 - 1982

1983 - 1984

1983 - 1986

CURRICULUM VITAE

Born in London, England.

Attended Dulwich College Preparatory School, London,

England.

Attended Westminster School, London, England.

Undergraduate study at the University of Bath,

England.

Batchelor of Science Sandwich Course on Applied

Biology with Microbiology Specialisation.

Eidgenossiche Technische Hochschule ZUrich, (ETH),

Nachdiplomstudium in Sanitary Engineering and Water

Protection.

Eidgenossiche Technische Hochschule ZUrich, Institut

fUr Gewasserschutz und Wassertechnologie, Doctoral

studies in the Department of Technical Biology in

the Swiss Federal Institute for Water Resources and

Water Pollution Control (EAWAG) .