The demonstration and antibodies in the gastrointestinal tract · destruction of immunoglobulin by...

Gut, 1972, 13, 483-499

Progress report

The demonstration and function ofantibodies in the gastrointestinaltractThe presence of antibody activity in intestinal secretions has been known for50 years since Davies1 demonstrated specific antibodies in the faeces ofpatients suffering from dysentery. It is now well documented that the gut ofgerm-free animals2 and the human neonate3 contains very few plasma cells,and after bacterial contamination the normal complement of plasma cellsappears rapidly. These studies have been reviewed recently2. This germ-freecondition is associated with absence of immunoglobulin A in the intestinalsecretions and the immunoglobulin appears as a response to colonization.The importance of IgA in secretions has been emphasized in reviews byTomasi and Bienenstock4, by Tomasi and DeCoteau5, and by Heremans6.Work on salivary IgA has been extensively reviewed by Brandtzaeg, Fjel-langer, and Gjeruldsen7.

It is the purpose of this review to present recent information on gastro-intestinal antibody production, particularly in the stomach and smallintestine, with emphasis on the probable functions of the antibody. There aremany technical difficulties in the detection and measurement of antibodies ingastrointestinal secretions. If the gastroenterologist is to maintain a criticalunderstanding of the many relevant papers in this immunological field, thenhe must appreciate some of the technical problems which are reviewed here.Furthermore, antibody responses within the bowel are important in the studyof oral vaccination and some of the relevant experimental work is discussed.The immunological changes in inflammatory bowel disease have beenreviewed by Kraft and Kirsner8, and the bowel abnormalities in IgA deficiencyare discussed by Bull and Tomasi9. These are not dealt with in the presentreview.

Origin of Intestinal Immunoglobulin

Immunoglobulin may reach the lumen of the gut either by transudation fromserum, or as a result of local production and release. There is evidence fromstudies in animals'0.1' and man'2 that exudation into the gut is a mostimportant route in the catabolism of serum globulins. Currently, however,interest is focused mainly on the local production of immunoglobulinin the tissues of the gut itself. Tissue culture studies show that radioactivelylabelled amino acids are incorporated into IgG, A, and M by tissues takenfrom various levels of the gastrointestinal tract'314. Immunofluorescentstudies, using fluorescein-labelled antisera specific for the various immuno-globulin classes, have shown the presence of immunoglobulin-containing cellsin all gut tissues examined.5-20. The immunoglobulin-containing cells are

483

on February 10, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.13.6.483 on 1 June 1972. D

ownloaded from

http://gut.bmj.com/

D. J. C. Shearman, D. M. Parkin, and D. B. L. McClelland

lymphoid cells, and when examined by conventional microscopy mostly havethe appearance of mature plasma cells, although cells which appear to belymphocytes are also seen to contain immunoglobulin. Immunofluorescencealso shows the presence of interstitial immunoglobulin (mainly IgG), andIgA staining is seen lining the basement membrane, between the epithelialcells and, with certain antisera, in the apical regions of the epithelial cells.

In normal gut tissues, the great majority of fluorescein-staining cellscontain IgA. IgG is present in small numbers of cells, and IgM is inter-mediate. In addition, cells containing IgD21 and IgE22 have been described.The importance of local production of immunoglobulin is emphasized by

studies on the transport of immunoglobulins into secretions. There is evidencethat very small amounts of IgA may be transported into the saliva of hypo-gammaglobulinaemic patients following infusion of fresh plasma23, but thebulk of evidence from infusion studies indicates that although a small amountof IgA may be transferred from serum, this makes a very small contributionto the total amount of IgA in saliva24-26. Although South, Cooper, Wollheim,Hong, and Good23 did not demonstrate the transfer of infused IgG and IgMinto saliva in hypogammaglobulinaemics, there is evidence that IgM maybe preferentially transported into saliva, while IgG may appear in saliva inamounts related to the level in serum27.

Further evidence for the local production of immunoglobulin in the gutcomes from studies discussed below which show dissociation of the antibodyresponse in secretions from that in the serum, in terms of specificity, timecourse, and immunoglobulin class.

Special Characteristics of Intestinal Inmunoglobulins

STRUCTUREThere has been particular interest in secreted immunoglobulins since theobservations that the amounts of IgA relative to the other immunoglobulinsis much greater than in serum28,29, and this IgA in secretions was in adifferent form to the IgA in serum18. IgA in human secretions has been shownto have a higher molecular weight than serum IgA, and to possess an ad-ditional component, known as the secretory piece. This is a glycoprotein,molecular weight about 76000, which appears to bejoined to two molecules ofIgA by covalent and non-covalent forces, giving a molecule with an electronmicroscopic appearance of a double 'T30. The secretory piece can be dis-sociated from the IgA molecules by reduction and alkylation, and the resultingIgA molecules closely resemble serum IgA, but are mainly of a form of theIgA subclass IgA231, which is unusual among immunoglobulins in having nodisulphide bonds linking the heavy and light chains32. The significance ofthis is not yet known.

Antisera can be raised against the intact secretory IgA molecule, and theseantisera detect the presence of an additional determinant, absent from serumIgA, and due to the presence of the secretory piece. Antisera against thesecretory piece can also be prepared by immunizing animals with freesecretory piece, and by suitable absorptions a serum can be prepared whichwill react only with free secretory piece, and not with secretory piece bound toIgA. Using this antiserum it has been shown that normal secretions (salivaand colostrum) contain significant amounts of free secretory piece7. Secre-tions deficient in IgA also contain free secretory piece23. Using fluorescent

484



j.com/


ownloaded from

http://gut.bmj.com/

The demonstration andfunction of antibodies in the gastrointestinal tract

labelled antisera to the secretory piece, fluorescence is localized to surfaceepithelial cells, goblet cells, and duct-lining epithelial cells of glands. Sincethe secretory piece does not appear to exist in the lamina propria of the gut,where IgA-containing lymphoid cells are seen, it has been suggested that IgA,synthesized by these cells in the lamina propria, moves across the basementmembrane and complexes with the secretory piece before entering the gutlumen, possibly by transport through the epithelial cells20. However, it isextremely difficult to produce antisera specific for secretory piece which donot cross-react with lactoferrin. It is likely that an IgA dimer is formed beforethe secretory piece is incorporated, and that the addition of the secretory piecestabilizes the structure of the dimer7.

THE EFFECTS OF PEPSIN AND TRYPSIN ONIMMUNOGLOBULINSThe presence of the secretory piece appears to alter the susceptibility of theIgA molecule to enzymatic digestion, and so may be important in preservingthe function of IgA antibody in the gut. Secretory IgA is resistant to digestionby trypsin33-35. When it is dissociated from secretory piece, the free lgAloses this resistance. The effects of digestion can be assessed by gel filtration,which shows that in the case of secretory IgA there is little breakdown intolow molecular weight fragments. It can also be assessed by studying antigen-binding by methods involving radiolabelled antigen. These methods show thatthe antigen-binding capacity of IgA and IgG is not greatly affected by trypticdigestion. However, in the case of IgG, the antigen is bound by a fragment ofthe immunoglobulin molecule, whereas the secretory IgA antibody appears toremain intact34. Further evidence for the trypsin resistance of secretory IgAcomes from observations that the ability of secretory IgA to agglutinatebacteria is not affected by tryptic digestion3ff, and since it has been shown thatpolymer IgA is a considerably more efficient haemagglutinating antibodythan monomer IgA37 '38, it seems likely that secretory IgA remains in itsdimeric form following trypsin digestion, otherwise its agglutinating abilitywould be reduced.

Secretory IgA also has a greater resistance to peptic digestion than do IgAmyeloma proteins39 despite an earlier report which indicated otherwise40. It isinteresting that when IgA myeloma proteins were studied, the IgA subclasswhich possesses disulphide bonds between heavy and light chains was shownto be considerably more resistant to peptic digestion than the IgA2 proteinswhich do not have these disulphide bonds. Since, as already mentioned, thegreat majority of secretory IgA is of IgA2 type, there is a strong indicationthat secretory piece provides protection against peptic digestion39. It shouldbe pointed out that in these studies peptic digestion was carried out at pHvalues around 4. When secretory IgA is exposed to pepsin at pH23fi or to acidgastric juice41, degradation is very rapid indeed, so it seems unlikely that evensecretory IgA could survive in the lumen of the normal, acid-secretingstomach.

CONTRIBUTION OF THE GASTROINTESTINAL TRACT TO SERUMIMMUNOGLOBULINSThe immunoglobulin-producing cells of the gut probably make a significantcontribution to the serum immunoglobulins. IgA synthesized by isolatedperfused segments of human gut appears in part in the secretions, and in part

485



j.com/


ownloaded from

http://gut.bmj.com/


in the venous effluent perfusion fluid42. Studies in the dog show that IgAappears in the efferent lymph from the intestine43. Furthermore, secretoryIgA can be detected in the serum of normal individuals, and in higher con-centrations in disease states, particularly those in which mucosal abnormali-ties are present, such as coeliac disease and ulcerative colitis, and also inpregnancy44-47. The secretory IgA in serum presumably reflects the escape ofexocrine IgA into the circulation via the lymphatics from its sites of pro-duction. Although the levels in the serum of normal individuals are verylow, it has been pointed out4 that because secretory IgA has a very shortlife in the serum48, there must be quite an appreciable input of secretory IgAinto the serum to maintain these levels.

Measurement of Immunoglobulins in Secretions

PROBLEMS IN DIFFERENT SECRETIONSQuantitative measurements of immunoglobulin concentrations in secretionsof the gut are bedevilled by technical problems, and many results publishedto date may be misleading. The most precise results have been obtained withsaliva7, which has the great advantage for the investigator that it can becollected direct from the parotid duct by use of the Curby cap, free frommucus, debris, and proteolytic enzymes, and easy to work with in immuno-precipitation systems. Gastric juice poses problems because of the very rapiddestruction of immunoglobulin by acid-pepsin41 which probably explainsreports which indicate that there is no immunoglobulin in acid gastricjuice49. Achlorhydric gastric juice also poses problems, because althoughthere may be high concentrations of immunoglobulin, non-immunoglobulinprecipitates may interfere with quantitative measurement41.

Small intestinal aspirates have been fairly extensively studied both quali-tatively and quantitatively42'50 -53. Although secretory IgA is resistant todigestion by trypsin, this does not apply to IgG and IgM and in any report itshould be demonstrated that the sampling and storage methods used provideadequate inhibition of proteolytic activity. A further problem is the formationof protein precipitates, unrelated to specific antiserum-immunoglobulinprecipitates50 54. These are mainly due to the reaction of trypsin or chymo-trypsin with a non-immunoglobulin component of certain animal sera, andcan be avoided either by choice of suitable antisera or preferably by using theglobulin fraction of the antiserum rather than the whole serum55'56.Measurement of IgA levels in stool requires an extensive preparation of the

stool to obtain a clear globulin-containing fraction and the preparationlosses may be considerable so that quantitative levels quoted can only beconsidered to be estimates.

PROBLEMS WITH STANDARDSA further difficulty in the assessment of IgA levels in the gut is that the pro-portion of monomer, dimer, and higher molecular weight forms is not known.The standards used in quantitating IgA in secretions should contain IgA inthe same form as in secretions or results will be misleading (the use of a 7Sserum IgA standard to quantitate 11S secretory IgA will lead to an under-estimate of approximately threefold7). Estimates have been made of themolecular weight distribution of IgA in gut secretions42'57'58 and these have

486



j.com/


ownloaded from

http://gut.bmj.com/


been taken into account when levels have been measured, but even withthese refinements the results can only be considered as estimates.

PROBLEMS WITH ANTISERAAs Brandtzaeg has shown very fully7'59, antisera to IgA and secretory piececan be produced in a number of ways using different forms of immunogen,and thus producing antisera with markedly different specificities. In additionto determinants specific for the IgA heavy chain, he has demonstrated adeterminant associated with polymers of IgA but absent from monomers, adeterminant unique to free secretory piece, and a determinant specific forbound secretory piece but concealed on free secretory piece. He has alsoshown that the use of different antisera may produce significantly differentresults when quantitating the same sample. It is therefore important that thespecificity and method of preparation of antisera should be described whenresults are quoted.

PROBLEMS WHEN SAMPLES ARE CONCENTRATEDBecause concentrations of IgA may be low, especially in saliva, concentrationof samples may be necessary. Significant protein loss can occur with some ofthe procedures used7 and it is obviously desirable to use sensitive methodswhich can be used on unconcentrated samples. The method of radioactivesingle radial immunodiffusion60 should be particularly suitable for use on gutsecretions.

Antibody Activity in the Gastrointestinal Tract

METHODS OF DEMONSTRATING ANTIBODY ACTIVITYEarly studies of antibody in stool produced inconsistent results and thereasons for this have been discussed extensively61. The antibody activity in aspecimen may show marked variations according to how it is assessed61'62.The most reliable results are produced by methods which directly measureantigen-antibody combination62 such as the radioimmunoassay systemdescribed by Farr or its modifications. Freter has demonstrated that antibodyto Vibrio cholera can be consistently demonstrated in the stool of volunteersgiven oral vaccine when the Farr test is used, but when tested by agglutinationresults were inconsistent and titres low63. Among the reasons suggested forthis discrepancy are partial degradation of antibody by proteolytic enzyme;the presence of agglutination inhibitors in stool extracts; and formation ofnon-specific precipitates in the extracts.

Other direct methods of detecting antigen-antibody combination includeimmunofluorescence and radioimmunoelectrophoresis or radioimmunodiffu-sion. Immunofluorescence has been used to detect antibody in saliva64f6f andin colonic washings67 with apparently satisfactory results, but other workershave found the method to be unsatisfactory with intestinal secretions67, andour own observations confirm these problems. The technical difficulty ofusing this technique with unfractionated gut secretions demands extensivecontrols. Radioimmunodiffusion61'68 and radioimmunoelectrophoresis69'70present problems due to the binding of labelled antigen to non-specificprecipitates, and therefore it is necessary to demonstrate that precipitateswhich appear to bind antigen do in fact contain immunoglobulin and thatnon-specific binding does not occur54. Although tests depending on the effects

487



j.com/


ownloaded from

http://gut.bmj.com/


ofantigen-antibody combination may pose technical problems, they should notnecessarily be rejected, as such tests may provide information related topossible physiological function of antibody. A recent review by Minden andFarr62 emphasizes the difference in results which may be obtained usingdifferent methods to study the antibody activity in any one sample.

THE RANGE OF ANTIBODY ACTIVITYTomasi,Tan, Solomon, and Prendergast demonstrated antibody activity againstblood group substances (isohaemagglutinins) inhuman parotid saliva and col-ostrum, and showed that these were of IgA class18. The development of serumisohaemagglutinins is thought to be due to exposure to antigenically cross-reacting bacteria and viruses7' and presumably the same is true for exocrineisohaemagglutinins. Certainly gastrointestinal secretions have antibodyactivity to a wide range of commensal organisms: these are frequentlytermed 'natural' antibodies to distinguish them from antibodies which followinfection or immunization, though this distinction is somewhat meaningless.Antibodies directed against bacteria can be found in saliva64'65'72 -75,achlorhydric gastric juice76, and small intestinal secretions64'77.

There is evidence from animal work which suggests that antibodies toparasites are present in intestinal secretions78 -80. In man there is littleevidence for the existence of secretory antibody to intestinal parasites,although it has been observed that patients with panhypogammaglobulin-aemia have a high incidence of infection with Giardia lamblia81'82.Antibody activity in intestinal secretions is also present to viruses83 84 as

well as dietary constituents85'86. The development of such natural antibodiesis possibly related to the antigenic stimulus provided by the normal microbialflora and dietary proteins.

THE DISSOCIATION OF GASTROINTESTINAL AND SERUM ANTIBODYRESPONSESThe clinical studies referred to below show clearly that the secreted antibodyresponse differs from the serum response, and studies of the immune responseto ingested antigens in animals confirm this strikingly different pattern.When ferritin is fed to rats in drinking water, large numbers of immunocytesin the gut lamina propria can be shown to produce antibody to ferritin. Thesecells are all of the IgA type. In the mesenteric nodes and spleen, a fewantiferritin-producing cells can be seen, and again these are mainly of IgAtype. By contrast, when ferritin is given parenterally, the great majority ofantiferritin cells are in the lymph nodes and spleen. These cells are of IgMtype initially but after further immunization they are mainly of IgG type87.Similar findings have been reported for the response to ingested red cells andalbumin88'89. These studies provide a clear indication that in the speciesstudied the antibody response to ingested antigen consists of IgA antibodywhich is produced mainly by cells in the gut wall.

Burrows, Elliot, and Havens" and Burrows and Havens9' showed thatexperimental cholera in the guinea pig is followed by a rise in antibody titre inboth serum and faeces. Coproantibody appeared earlier, reached peak titressooner, and declined before serum antibody. Resistance to infection appearedto correlate better with faecal than serum antibody.

Eddie, Schulkind, and Robbins92 studied the effect of various immunizationschedules on rabbits using Salmonella typhimurium and studying antibody

488



j.com/


ownloaded from

http://gut.bmj.com/


responses in serum, intestinal fluid, and colostrum. Direct contact of thesecretory tissue with living bacteria gave rise to IgG and IgA antitolies inexternal secretions. This did not occur when dead organisms were used.Parenteral inoculation of living or dead organisms gave rise to high serumantibody titres but no secretory antibody. This immune response by theintestine appears to be localized to the site exposed to antigen and the intestinedoes not behave as a single unit. Ogra and Karzon93 instilled polio vaccine intothe distal limb of a double-barrelled colostomy. Secretory IgA antibodyappeared in this segment, but only low titres were found in the proximalcolon, and no antibody activity whatsoever in the nasopharynx. Followingoral challenge, the virus was able to colonize the nasopharynx but not thecolon.When considering the gut humoral immune response it is important to

distinguish between studies which demonstrate solely local effects of localstimulation, and work in which, because of the absorption of antigen, asystemic immune response may contribute to what is observed locally.Orally administered antigen can give rise to serum antibody as well asantibody in gut secretions68"92,94,s. This could be due to absorption ofantigen, or proliferation of organisms, allowing antigen to reach systemicantibody-producing sites. Alternatively, immunocompetent cells from thelocal site could migrate to sites of systemic antibody production.

Following parenteral administration of antigen antibody may be detectedin gut secretions63 96'97, although this is not confirmed in all studies92'95;however, there is no doubt that parenteral immunization against infectionslocalized to the gut does confer protection98"99. The gut antibody response toparenteral immunization could be due to the effect of parenterally adminis-tered antigen reaching gastrointestinal sites of antibody production. This issupported by the observations of Koshland.00, suggesting that the use ofadjuvants localized parenterally administered antigen so that it did not reachthe gut in sufficient amounts to lead to the production of antibody.

It has also been shown by passive immunization experiments96 that serumantibody may reach the gut, probably by passive leakage of antibody.

In conclusion these various studies support the concept of local immuno-logical responses in the gastrointestinal tract. However, in any one situationthere may be reasons why serum and gut antibodies show some degree ofcorrelation.

STUDIES ON HUMAN SECRETORY ANTIBODY, INCLUDING ORALIMMUNIZATION

Antibody to bacteriaAn antibody response in intestinal secretions has been demonstrated duringacute cholera infection101, and more recent work68 using radioimmuno-diffusion claims to show that the antibody in jejunal juice in acute choleraexists in the IgG, A, and M classes. Intestinal antibody has also been reportedto follow infection with other bacteria, eg, E. coli102'103 and Shigella57, and inthe latter study anti-shigella antibody also appeared to exist in all threemajor immunoglobulin classes. In view of these findings, the possibility oforal immunization against enteric bacterial infections is of great interest.

In human subjects, Freter63 showed that a killed oral vaccine provided amore reliable stimulus of coproantibody than parenteral vaccination. The

489



j.com/


ownloaded from

http://gut.bmj.com/


coproantibody response was of short duration but could be maintained byregular oral immunization: faecal and serum antibody appeared to beindependent'04. Oral vaccination with attenuated strains of Shigellae hasprovided protection against natural infection in monkeys'05 and in man'08.In typhoid fever killed vaccines given orally in man provide only slightprotection against infection107; living attenuated strains appear to be moresuccessful-at least in animals'08. As has been mentioned, animal worksuggests that high titres of bacterial antibody in the gut and serum followoral administration of living, but not dead, organisms92. It has been shownthat live attenuated cholera vaccine produces better levels of serum antibodyand coproantibody than are obtained using killed oral vaccine, and that thelive vaccine was safe for human use'09. Further studies on the effectiveness oflive oral vaccines in protecting against enteric infections are needed.

Antibody activity to virusesThe gut can mount a local response to poliovirus93. Infection of the intestinaltract with living attenuated poliovirus in infants bears little relation to serumantibody titres"10. Previous intestinal infection with virus resulted in completeor partial resistance to re-infection, even when no demonstrable serum anti-body followed the first infection"'. Antibody to poliovirus has been found inthe stools of children with poliomyelitis84 or following oral Sabin vaccina-tion"12. Ogra, Karzon, Righthand, and MacGillivray95 showed that oralimmunization of infants with live attenuated poliovirus led to the appearanceof antibody in secretions (duodenal fluid or nasal washings). Parenteralimmunization with killed virus did not lead to production of antibody in thesecretions. Immunization by either route led to the appearance of serumantibody, and, as discussed above, there are a number of reasons why anorally administe1ed antigen may lead to antibody in serum. Other studieshave confirmed that antibody against viruses in saliva70"'3, small intestinalfluid70, and faeces14 is of IgA class.

Antibody activity to dietary antigensAs well as bacterial and viral antigens, the gut mucosa comes into contactwith foreign proteins which are ingested as part of the normal diet. Thus, it isnot surprising that antibody to dietary proteins has been detected in intestinalsecretions. Early studies concentrated on disease states. In gastrointestinalmilk sensitivity it was documented that affected infants had precipitins tomilk in the stool6t9"5. However, antibody to milk proteins in the absence ofintestinal disease has been found in faeces85 and parotid saliva"6. In coeliacdisease intestinal secretions may contain antibody to dietary protein17,although here again a recent report suggests that the incidence is not signi-ficantly greater than in a control group86. Two attempts have been made todefine the class of antibody responsible for these reactions5"69. However,neither can be considered technically entirely satisfactory. In both, goatantisera to human immunoglobulin were used, and as outlined earlierenteric contents can give rise to non-immunological lines.

The Protective Role of Colostrum

In any review of gastrointestinal antibody it is essential to consider the role ofcolostrum and milk. In many animal species, eg, cows and pigs, there is no

490



j.com/


ownloaded from

http://gut.bmj.com/


transplacental passage of antibody and the animal absorbs antibody fromcolostrum in the first few hours of life. Failure to obtain colostrum rendersthe animals very susceptible to septicaemic illness"18 due to infection withE. coli and this can be prevented by feeding the IgM fraction of colostrum"19.In the human infant, transplacental transfer of IgG occurs, but IgA in serumis absent at birth'20. Neonatal saliva contains neither IgG nor IgA, the latterappearing by about the fourth week of life, and reaching adult levels byabout six weeks'20-'22. Normal meconium contains IgG, and in normalinfants no IgA or IgM (although a recent report indicates that the meconiumof infants with cytic fibrosis does contain IgA, which is at least partly ofsecretory type'23).Human colostrum is rich in immunoglobulin, particularly IgA28, and it

contains antibody activity against bacteria36'66"24"123 and viruses514"126"127.This antibody activity is usually of IgA type36"128-'30. These antibodies cansurvive passage through the neonate gastrointestinal tractl24" 25. SecretoryIgA antibodies from colostrum appear to be resistant to trypsin36'124*Although IgA is sensitive to pepsin and HC1, these are absent from- thestomach in the first weeks of life, or present in very low concentrations'31.

There is some debate as to whether colostral antibodies are derived fromserum or synthesized in the breast. IgA antibodies may be present in colos-trum, which are not demonstrable in serum'30,132. Tissue culture studies onhuman and monkey mammary gland would also seem to indicate that IgAantibody is synthesized locally'33. Leucocytes derived from colostrum andcultured in vitro appear to be able to synthesize IgA only'34, whereas peri-pheral blood leucocytes can synthesize all three immunoglobulin classes135'136.

In human subjects it is now established that breast-fed infants have a lowerincidence of enteric infection with E. coli 137-139 and less septicaemic illness"4than bottle-fed infants. It has been suggested that the low pH of faeces frombreast-fed infants may inhibit E. coil'4' and that this effect is due to the poorbuffering capacity of breast milk'42.

It is tempting to try to relate the beneficial effects of milk to its content ofspecific antibody, but the evidence for this is limited. A recent report showsthat milk has a bacteriostatic effect on growth of E. coli related to a combinedeffect of specific antibody and lactoferrin'43.

The Function of Secretory IgA in the Gastrointestinal Tract

COMPLEMENT FIXATIONThe mode of action of IgA remains obscure. It is certainly able to agglutinatebacteria36, but it is very doubtful if it can be bactericidal-this requirescomplement for the lysis of bacteria, and IgA antibodies are unable to fixcomplement23"132"144. One report suggests that IgA can accomplish bacterio-lysis in the presence of complement and lysozyme130; however, it is possiblethat the IgA fraction used in this study was not pure, and this work has notbeen confirmed92.

OPSONIZATIONIt has been suggested that IgA may act as an opsonin, rendering bacteriamore available for phagocytosisl465"46, although the evidence is conflicting92.It is difficult to envisage the value of such a system to the intestinal tract,

491



j.com/


ownloaded from

http://gut.bmj.com/


unless phagocytes in the intestinal wall are attracted to and then clear IgA-coated bacteria.

ANTITOXIN ACTIVITYIgA could act as an antitoxin. Antitoxin has been detected in intestinalsecretions of rabbits97l147 and monkeys'48 after oral administration ofcholera toxin. Diphtheria antitoxin of IgA class has been demonstrated innasal fluid after local antigenic stimulation'49. No intestinal antitoxin againstE. coli enterotoxin has yet been found.

INTERFERENCE WITH BACTERIAL GROWTHBrandtzaeg65 has shown in vivo coating of streptococci with IgA from saliva.Coated organisms exhibited interference with growth-the 'long chainphenomenon' which occurs when bacteria are grown in media which containantibody'5015'.

AGGLUTINATION AND THE PREVENTION OF ADHERENCE

It may be that agglutination of bacteria in itself is beneficial. As alreadymentioned, secretory IgA may be more efficient in agglutinating bacteria thanserum IgA because of its dimeric structure. Bacteria are thought to be re-moved from the bowel by peristalsis'52, and it is possible that aggregation ofbacteria hastens this process. Potentially pathogenic organisms appear to bemore adherent to mucosal surfaces than avirulent strains'53"54 and selectiveattachment of bacteria to mucosa may be an important factor in the controlof local bacterial populations155. Freter has shown that one effect of anti-body is to prevent adherence of pathogens (V. cholerae) to intestinal mucosal56or to inhibit the growth of bacteria which adhere to the mucosa157, but theseexperiments were done using serum antibody.

ANTIVIRAL ACTIONAntiviral activity can be demonstrated by the prevention of viral replicationin tissue culture"14"126 and by the prevention of viral colonization of locallyimmunized segments of gut93. The exact mechanism of the antiviral action isnot known.

MASKING OF ANTIGENIC SITESIgA may mask antigenic sites. Bacteria coated with secretory IgA appear to benon-antigenic in rabbits. Furthermore, if bacteria are exposed to secretoryIgA antibody, the bactericidal effect of IgG and IgM antibody is blocked92.One hypothesis suggests that IgA is directed only against commensal bacteria,leaving potential pathogens uncoated so that these can then stimulate theproduction of bactericidal IgG and IgM antibodies'58. It has been suggestedalso that IgA may play a similar role in relation to dietary antigen'59, pre-venting the formation of IgG and IgM class antibodies. It has been men-tioned that antibody to dietary antigens is present in intestinal secretions evenin normal subjects85'86"'l6. In patients with congenital deficiency of IgA, thereis a greatly increased incidence of serum antibody to milk proteins'60-162 andruminant IgMl6l-163, this antibody being of IgG class. An increased incidenceof serum precipitins to dietary antigens has also been found in coeliacdisease86"164"165 and inborn absence of IgA seems to be associated with thedevelopment of some cases of coeliac disease'66"167. Coeliac disease could

--492



j.com/


ownloaded from

http://gut.bmj.com/


represent a functional failure of the masking action of IgA antibodies againstdietary proteins'59.

This attractive hypothesis is far from proven. Neonates rapidly developserum IgG and IgM antibody to ingested bovine serum albumin94. There maybe increased absorption of dietary protein in this period'68"169. It is possiblethat in the absence of local IgA in the gut in the first few weeks of lifel22these ingested proteins may be able to produce a systemic antibody response.However, this does not explain the high incidence of anti-BSA antibodiesreported in older subjectsl70" 71. There is little evidence for quantitative IgAdeficiency in intestinal secretions in the majority of coeliacs, and indeed thelevels may be raised52'53. Furthermore, antibodies to commensal bacteria arenot exclusively IgA, and IgA antibody can appear in response to pathogenicbacteria57'68. Additional work is clearly necessary, and in studies on intestinalantibody the class of immunoglobulin involved should be determined.

Gut Immunoglobulin and Antibodies in Pathological States

Lehner172 has shown that the salivary IgA levels of individuals prone to cariesare significantly lower than those of caries-free individuals, despite the factthat the group prone to caries had higher serum IgA levels. By contrast,patients with recurrent oral ulceration showed no difference in salivary IgAlevels compared with controls.

Brandtzaeg has shown high levels of IgA in saliva of patients with macro-globulinaemia and presents evidence that these are probably due to selectivetransfer of IgM from the serum. In contrast, elevated saliva IgG levels inpatients with IgG myeloma were probably dependent on the raised serumlevels with no evidence of selective transport27.

Aspirates from the normal acid-secreting stomach have been reported tocontain no immunoglobulin49 but appreciable amounts can be detected usingsuitable sampling methods41. In achlorhydric patients, the gastric juicecontains high levels of immunoglobulin41. These levels cannot at present berelated to studies on the immunoglobulin-containing cells in the gastricmucosa, which show that in the normal gastric mucosa there is a majority ofIgA-producing plasma cells49"173. In atrophic gastritis, the proportion of IgA-staining cells has been reported to be much less than normal49 and in anotherstudy'73 the proportion of IgA-staining cells was shown to be much in-creased.

Autoantibodies to intrinsic factor in gastric juice have been shown tooccur in both IgG and IgA classesl74-'76, and in the case of intrinsic factortype II antibodies, IgA antibodies predominated. Parietal cell antibodies ingastric juice were seen to be predominantly IgA, or IgG and IgA combined.There appears to be no correlation between the immunoglobulin class of theautoantibodies in gastric juice and in serum'75.

In adult coeliac disease there are marked alterations in the immuno-globulin-containing cells in the lamina propria and small intestine, theproportion of IgM-containing cells being increased52"177, and these changespersist after treatment with a gluten-free diet. In one study'77 the number ofIgG-containing cells was also elevated.The jejunal juice in untreated coeliac disease has been reported to contain

increased amounts of IgM52 and increased amounts of IgG, A, and M53.

493



j.com/


ownloaded from

http://gut.bmj.com/

494 D. J. C. Shearman, D. M. Parkin, and D. B. L. McClelland

In one study52 IgM levels returned to normal on a gluten-free diet whereas inthe other report there was no change following treatment.

In ulcerative colitis, the density of IgA-containing cells in rectal mucosa islower than in normal tissue, although still greater than that of IgG- and IgM-containing cells178. More extracellular IgA was seen. There is also a report ofthe presence of large numbers of IgD-containing cells in rectal mucosa inulcerative colitis'79. A recent report suggests the presence of faecal antibodiesin colitic patients which react with the patient's own anaerobic flora67.

D. J. C. SHEARMAND. M. PARKIND. B. L. MCCLELLAND

Gastrointestinal Section of theUniversity Department of

Therapeutics and the RoyalInfirmary of Endinburgh

References

"Davies, A. (1922). An investigation into the serological properties of dysentery stools. Lancet, 2, 1009-1012.'Thompson, G. R., and Trexler, P. C. (1971). Gastrointestinal structure and function in germ-free or gnoto-

biotic animals. Gut, 12, 230-235.'Crabbe, P. A., and Heremans, J. F. (1969). The significance of local IgA in the physiology of the intestinal

mucosa. Folia med. neerl., 12, 100-106.'Tomasi, T. B., Jr., and Bienenstock, J. (1968). Secretory immunoglobulins. Advanc. Immunol., 9, 1-96.'Tomasi, T. B., Jr., and DeCoteau, E. (1970). Mucosal antibodies in respiratory and gastrointestinal disease.

Advanc. intern. Med., 16,401-425.'Hferemans, J. F. (1968). Immunoglobulin formation and function in different tissues. Curr. Top. Microbiol.

Immunol., 45, 131-203.'Brandtzaeg, P., Fjellanger, I., and Gjeruldsen, S. T. (1970). Human secretory immunoglobulins. 1. Salivary

secretions from individuals with normal or low levels of serum immunoglobulins. Scand. J. Haematoi.,Suppl., 12.

"Kraft, S. C., and Kirsner, J. B. (1971). Immunological apparatus of the gut and inflammatory bowel disease.Gastroenterology, 60, 922-951.

'Bull, D. M., and Tomasi, T. B., Jr. (1968). Deficiency of Immunoglobulin A in intestinal disease. Gastroen-terology, 54, 313-320.

°Andersen, S. B., Glenert, J., and Wallevik, K. (1963). Intestinal degradation of gamma-globulin in the dog.Nature (Lond.), 199, 1096-1097.

"Andersen, S. B., Glenert, J., and Wallevik, K. (1963). Gammaglobulmn turnover and intestinal degradation ofgammaglobulin in the dog. J. clin. Invest., 42, 1873-1881.

"Schultze, H. E., and Heremans, J. F. (1966). Molecular Biology of Human Proteins, Vol. 1 Elsevier, Amster-dam and New York.

'9Hurlimann, J., and Zuber, C. (1968). In vitro protein synthesis by human salivary glands. 1. Synthesis ofsalivary IgA and serum proteins. Immunology, 14, 809-817.

"Van Furth, R., and Aiuti, F. (1968). Immunoglobulin synthesis by tissues of the gastrointestinal and res-piratory tracts. Protides biol. Fluids, 16, 479-484.

5Crabb6, P. A., Carbonara, A. O., and Heremans, J. F. (1965). The normal human intestinal mucosa as amajor source of plasma cells containing yA-immunoglobulin. Lab. Invest., 14, 235-248.

"Crabb6, P. A., and Heremans, J. F. (1966). The distribution of immunoglobulin-containing cells along thehuman gastrointestinal tract. Gastroenterology, 51, 305-316.

7Gelzayd, E. A., Kraft, S. C., and Fitch, F. W. (1967). Immunoglobulin A: localization in rectal mucosalepithelial cells. Science, 157, 930-931.

"Tomasi, T. B., Jr., Tan, E. M., Solomon, A., and Prendergast, R. A. (1965). Characteristics of an immunesystem common to certain external secretions. J. exp. Med., 121, 101-124.

'Rubin, W., Fauci, A. S., Sleiseinger, M. H., and Jeffries, G. H. (1965). Immunofluorescent studies in adultceliac disease. J. clin. Invest., 44, 475-485.

2'Tourville, D. R., Adler, R. H., Bienenstock, J., and Tomasi, T. B., Jr. (1969). The human secretory immuno-globulin system: immunohistological localization of yA, secretory 'piece' and lactoferrin in normalhuman tissues. J. exp. Med., 129, 411-429.

"'Rowe, D. S., Crabb6, P. A., and Turner, M. W. (1968). Immunoglobulin D in serum body fluids and lymphoidtissues. Clin. exp. Immunol., 3, 477-490.

2"Tada, T., and Ishizaka, K. (1970). Distribution of yE-forming cells in lymphoid tissues of the human andmonkey. J. Immunol., 104, 377-387.

2"South, M. A., Cooper, M. D., Wollheim, F. A., Hong, R., and Good, R. A. (1966). The IgA system. 1. Studiesof the transport and immunochemistry of IgA in the saliva. J. exp. Med., 123, 615-627.

24Strober, W., Blaese, R. M., and Waldmann, T. A. (1970). The origin of salivary IgA. J. Lab. clin. Med., 75,856-862.

"6Stiehm, E. R., Vaerman, J. P., and Fudenberg, H. H. (1966). P.asma infusions in immunologic deficiency states:Metabolic and therapeutic studies. Blood, 28, 918-937.



j.com/


ownloaded from

http://gut.bmj.com/

The demonstration andfunction of antibodies in the gastrointestinal tract 495

"Selner, J. C., Merrill, D. A., and Claman, H. N. (1968). Salivary immunoglobulin and albumin: developmentduring the newborn period. J. Pediat., 72, 685-689.

"7Brandtzaeg, P. (1971). Human secretory immunoglobulins. II. Salivary secretions from individuals withselectively excessive or defective synthesis of serum immunoglobulins. Clin. exp. Immunol., 8, 69-85.

"Chodirker, W. B., and Tomasi, T. B., Jr. (1963). Gamma-globulins: quantitative relationships in human serumand nonvascular fluids. Science, 142, 1080-108 1.

29Tomasi, T. B., Jr., and Zigelbaum, S. (1963). The selective occurrence of VA globulins in certain body fluids.J. clin. Invest., 42, 1552-1560.

'°Bloth, B., and Svehag, S. E. (1971). Further studies on the ultrastructure of dimesic IgA of human origin. J.exp. Med., 133,1035-1042.

3'Grey, H. M., Abel, C. A., Yount, W. J., and Kunkel, H. G. (1968). A subclass of human v-A-globulins(v-A2) which lacks the disulfide bonds linking heavy and light chains. J. exp. Med., 128, 1223-1236.

"2Jerry, L. M., Kunkel, H. G., and Grey, H. M. (1970). Absence of disulfide bonds linking the heavy and lightchains: a property of a genetic variant of yA2 globulins. Proc. nat. Acad. Sci. (Wash.), 65, 557-563.

33Cederblad, G., Johansson, B. G., and Rymo, L. (1966). Reduction and proteolytic degradation of immuno-globulin A from human colostrum. Acta chem. scar;d., 20, 2349-2357.

3"Brown, W. R., Newcomb, R. W., and Ishizaka, K. (1970). Proteolytic degradation of exocrir.e and serumimmunoglobulins. J. clin. Invest., 49, 1374-1380.

35Steward, M. W. (1971). Resistance of rabbit secretory IgA to proteolysis. Biochim. biophys. Acta (Amst.), 236,440-449.

3"McClelland, D. B. L., et al (Unpublished).37Newcomb, R. W., and DeVald, B. L. (1969). Antibody activities of human exocrine yA diphtheria antitoxin.

Fed. Proc., 28, 765.3"Ishizaka, K., Ishizaka, T., Lee, E. H., and Fudenberg, H. (1965). Immunochemical properties of human

yA Isohemagglutinin. l. Comparisons with VG- and yM- globulin antibodies. J. Immunol., 95, 197-208.3"Shuster, J. (1971). Pepsin hydrolysis of Iga:Adelineation of two populations of molecules. Immunochemistry, 8,

405-411."'Wilson, I. D., and Williams, R. C.. Jr., (1969). Two distinct groups of immunoglobulin A (IgA) revealed by

peptic digestion. J. clin. Invest., 48, 2409-2416."McClelland, D. B. L., Finlayson, N. D. C., Samson, R. R., Nairn, I. M., and Shearman, D. J. C. (1971).

Quantitation of immunoglobulins in gastric juice by electroimmunodiffusion. Gastroenterology, 60,509-514.

"Bull, D. M., Bienenstock, J., and Tomasi, T. B., Jr. (1971). Studies on human intestinal immunoglobulinA. Gastroenterology, 60, 370-380.

"3Vaerman, J. P., and Heremans, J. F. (1970). Origin and molecular size ofimmunoglobulin-A in the mesentericlymph of the dog. Immunology, 18, 27-38.

"Thompson, R. A., Asquith, P., and Cooke, W. T. (1969). Secretory IgA in the serum. Lancet, 2, 517-519.""Waldman, R. H., Mach, J. P., Stella, M. M., and Rowe, D. S. (1970). Secretory IgA in human serum. J.

Immunol., 105,43-47."Thompson, R. A., and Asquith, P. (1970). Quantitation ofexocrine IgA in human serum in health and disease.

Clin. exp. Immunol., 7, 491-500."7Brandtzaeg, P. (1971). Human secretory immunoglobulins. V. Occurrence of secretory piece in human serum.

J. Immunol., 106, 318-323."Butler, W. T., Rossen, R. D., and Waldmann, T. A. (1967). The mechanism ofappearance ofimmunoglobulin

A in nasal secretions in man. J. clin. Invest., 46, 1883-1893.'"Odgers, R. J., and Wangel, A. G. (1968). Abnormalities in IgA-containing mononuclear cells in the gastric

lesion of pernicious anaemia. Lancet, 2, 846-849."Plaut, A. G., and Keonil, P. (1969). Immunoglobulins in human small intestinal fluid. Gastroenterology, 56,

522-530."'Kriebel, G. W., Jr., Kraft, S. C., and Rothberg, R. M. (1969). Locally produced antibody in human gastro-

intestinal secretions. J. Immuncl., 103, 1268-1275."'Douglas, A. P., Crabbe, P. A., and Hobbs, J. R. (1970). Immunochemical studies of the serum, intestinal

secretions and intestinal mucosa in patients with adult celiac disease and other forms of the celiacsyndrome. Gastroenterology, 59, 414-425.

"3Thompson, R. A., Asquith, P., and Cooke, W. T. (1970). Immunoglobulin content of jejunal fluid in coeliacdisease and other gastrointestinal diseases. In Abstracts of the IVth World Congress of Gastroenterology,Copenhagen, 1970, p. 8.

"'Kraft, S. C., Rothberg, R. M., and Kriebel, G. W., Jr. (1970). Nonimmunologic precipitin lines between serumand enteric contents giving false positive evidence of local antibody production. J. Immunol., 104, 528-529.

"Brown, W. R. (1971). Nonimmunoglobulin precipitin lines between intestinal fluids and antisera in immuno-diffusion studies: an explanation for their occurrence and how they can be avoided. J. Lab. clin. Med., 77,326-334.

'"McClelland, D. B. L., Parkin, D. M., Samson, R. R., and Shearman, D. J. C. (1972). Gastroenterology, inpress.

'Reed, W. P., and Williams, R. C., Jnr. (1971). Intestinal immunoglobulins in Shigellosis. Gastroenterology, 61,35-45.

5"Waldman, R. H., Bencic, Z., Sakazaki, R., Sinha, R., Ganguly, R., Deb, B. C., and Mukerjee, S. (1971).Cholera immunology. I. Immunoglobulin levels in serum, fluid from the small intestine, and feces frompatients with cholera and noncholeraic diarrhea during illness and convalescence. J. infect. Dis., 123,579-586.

59Brandtzaeg, P. (1971). Human secretory immunoglobulins. III. Immunochemical and physiochemical studiesof secretory IgA and free secretory piece. Acta path. microbiol. scand., Sect. B, 79, 165-188.

"'Rowe, D. S. (1969). Radioactive single radial diffusion: a method for increasing the sensitivity of immuno-chemical quantification of proteins in agar gel. Bull. Wld Hlth Org., 40, 613-616.

"Freter, R. (1962). Influence of various antibody characteristics on the titration of 0 antibodies by differentmethods. J. infect. Dis., 111, 25-36.

"Minden, P., and Farr, R. S. (1971). Antigen-antibody interactions and their measurement. In ImmunologicalDiseases, edited by M. Samter, Vol. 1, pp. 179-201. Little, Brown, Boston.



j.com/


ownloaded from

http://gut.bmj.com/


63Freter, R. (1962). Detection of coproantibody and its formation after parenteral and oral immunization ofhuman volunteers. J. infect. Dis., 111, 37-48.

"Tourville, D., Bienenstock, J., and Tomasi, T. B., Jr. (1968). Natural antibodies of human serum, saliva, andurine reactive with Escherichia coli. Proc. Soc. exp. Biol. (N. Y.), 128, 722-727.

"6Brandtzaeg, P., Fjellanger, I., and Gjernldsen, S. T. (1968). Adsorption ofimmunoglobulin A onto oral bacteriain vivo. J. Bact., 96, 242-249.

6"Mouton, R. P., Stoop, J. W., Ballieux, R. E., and Mul, N. A. J. (1970). Pneumococcal antibodies in IgA ofserum and external secretions. Clin. exp. Immunol., 7, 201-210.

'7Monteiro, E., Fossey, J., Shiner, M., Drasar, B. S., and Allison, A. C. (1971). Antibacterial antibodies inrectal and colonic mucosa in ulcerative colitis. Lancet, 1, 249-251.

6"Northrup, R. S., and Hossain, S. A. (1970). Immunoglobulins and antibody activity in the intestine and serumin cholera. II. Measurement of antibody activity in jejunal aspirates and sera of cholera patients byradioimmunodiffusion. J. infect. Dis., 121, Suppl., 142-146.

"Kletter, B., Freier, S., Davies, A. M., and Gery, I. (1969). Coproantibodies to milk: identification by radio-immunoelectrophoresis. J. Immunol., 103, 857-858.

7"Berger, R., Ainbender, E., Hodes, H. L., Zepp, H. D., and Hevizy, M. M. (1967). Demonstration of IgApolioantibody in saliva, duodenal fluid and urine. Nature (Lond.), 214, 420422.

"Springer, G. F., and Horton, R. E. (1969). Blood group isoantibody stimulation in man by feeding bloodgroup-active bacteria. J. clin. Invest., 48, 1280-1291.

7'Kraus, F. W., and Konno, J. (1963). Antibodies in saliva. Ann. N.Y. Acad. Sci., 106, 311-329."Evans, R. T., and Mergenhagen, S. E. (1965). Occurrence of natural antibacterial antibody in human parotid

fluid. Proc. Soc. exp. Biol. (N. Y.), 119, 815-819.7"Green, G. E. (1966). Inherent defense mechanisms in saliva. J. dent. Res., 45, Suppl. 3, 624-629.75Sirisinha, S. (1970). Reactions ofhuman salivary immunoglobulins with indigenous bacteria. Arch. oral Biol.,

15, 551-554."Balazs, V. (1962). Studies on the antibody content of anacid gastric juices. Experientia (Basel), 18, 72-73.7Girard, J. P., and Kalbermatten, A. de. (1970). Les immunoglobulines dans les secretions duodeno-pan-

creatiques et la bile. Schweiz. med. Wschr., 100, 336-338."'Dobson, C. (1966). Precipitating antibodies in extracts from the mucosa and tunica muscularis of the ali-

mentary tract of sheep infected with Oesophagostomum columbianum. J. Parasit., 52, 1037-1038."'Dobson, C. (1966). The demonstration ofantibodies in the intestinal mucus of sheep infected with Oesophago-

stomum columbianum by means of the percutaneous anaphylaxis test. Aust. J. biol. Sci., 19, 339-340.50Crandall, R. B., Cebra, J. J., and Crandall, C. A. (1967). The relative proportions of IgG-, IgA and IgM-

containing cells in rabbit tissues during experimental trichinosis. Immunology, 12, 147-158."'Hermans, P. E., Huizenga, K. A., Hoffman, H. N., Brown, A. L. Jr., and Markowitz, H. (1966). Dysgamma-

globulinemia associated with nodular lymphoid hyperplasia of the small intestine. Amer. J. Med., 40,78-89.

'2Parkin, D. M., McClelland, D. B. L., O'Moore, R. R., Percy-Robb, I. W., Grant, I. W. B., and Shearman,D. J. C. (1972). Intestinal bacteria flora and bile salt studies in hypogammaglobulinaemia. Gut, 13, 182-188.

"3Balizs, V., and Szollosy, E. (1962). Studies on the virus neutralising capacity of anacid gastric juices. Ex-perientia (Basel), 18, 515-516.

"4Lipton, M. M., and Steigman, A. J. (1963). Human coproantibody against polioviruses. J. infect. Dis., 112,57-66.

'"Davis, S. D., Bierman, C. W., Pierson, W. E., Maas, C. W., and lanneta, A. (1970). Clinical nonspecificity ofmilk coproantibodies in diarrheal stools. New Engl. J. Med., 282, 612-613.

8"Ferguson, A., and Carswell, F. (1972). Precipitins to dietary proteins in serum and upper intestinal secretionsof coeliac children. Brit. med. J., 1, 75-77.

'7Crabb6, P. A., Nash, D. R., Bazin, H., Eyssen, H., and Heremans, J. F. (1969). Antibodies of the IgA type inintestinal plasma cells of germfree mice after oral or parenteral immunization with ferritin. J. exp. Med.,130, 723-744.

"8Bazin, H., Levi, G., and Doria, G. (1970). Predominant contribution of IgA antibody-forming cells to animmune response detected in extraintestina! lymphoid tissues of germ-free mice exposed to antigen by theoral route. J. Immunol., 105, 1049-1051.

"IDolezel, J., and Bienenstock, J. (1971). yA and non-vA immune response after oral and parenteral immuniza-tion of the hamster. Cell Immunol., 2, 458468.

"9Burrows, W., Elliot, M. E., and Havens, I. (1947). Studies on immunity to Asiatic cholera. IV. The excretion ofcoproantibody in experimental enteric cholera in the guinea pig. J. infect. Dis., 81, 261-281.

"Burrows, W., and Havens, I. (1948). Studies on immunity to Asiatic cholera. V. The absorption of immuneglobulin from the bowel and its excretion in the urine and feces of experimental animals and humanvolunteers. J. in.fect. Dis., 82, 231-250.

9'Eddie, D. S., Schulkind, M.L., and Robbins, J. B. (1971). The isolation and biologic activities of purifiedsecretory IgA and IgG anti-Salmonella typhimurium "O" antibodies from rabbit intestinal fluid andcolostrum. J. Immunol., 106, 181-190.

9"0gra, P. L., and Karzon, D. T. (1969). Distribution of poliovirus antibody in serum, nasopharynx andalimentary tract following segmental immunization of lower alimentary tract with poliovaccine. J.Immunol., 102, 1423-1430.

"4Rothberg, R. M. (1969). Immunoglobulin and specific antibody synthesis during the first weeks of life ofpremature infants. J. Pediat., 75, 391-399.

9"Ogra, P. L., Karzon, D. T., Righthand, F., and MacGillivray, M. (1968). Immunoglobulin response in serumand secretions after immunization with live and inactivated poliovaccine and natural infection. NewEngl. J. Med., 279, 893-900.

"McCleery, J. L., Kraft, S. C., and Rothberg, R. M. (1970). Demonstration of antibody in rabbit feces afteractive or passive parenteral immunization. Digestion, 3, 213-221.

"Kaur, J., Burrows, W., and Furlong, M. A. (1971). Immunity to cholera: antibody response in the lowerileum of the rabbit. J. infect. Dis., 124, 359-366.

"Mosely, W. H., Woodward, W. E., Aziz, K. M. A., Mizanur Rahman, A. S. M., Alauddin Chowdhury,A. K. M., Ansaruddin Ahmed, and Feeley, J. C. (1970). The 1968-1969 cholera-vaccine field trial inrural East Pakistan. Effectivencss ofmonovalent Ogawa and Inaba vaccines and a purified Inaba antigen,with comparative results of serological and animal protection tests. J. infect. Dis., 121, Suppl., 1-9.



j.com/


ownloaded from

http://gut.bmj.com/


"Woodward, W. E., Mosley, W. H., and McCormack, W. M. (1970). The spectrum of cholera in rural EastPakistan. 1. Correlation of bacteriologic and serologic results. J. infect. Dis., 121, Suppl., 10-16.

0"Koshland, M. E. (1953). The origin of fecal antibody and its relationship to immunization with adjuvant.J. Immunol, 70, 359-365.

0'Freter, R., De, S. P., Mondal, A., Shrivastava, D. L., and Sunderman, F. W., Jr. (1965). Coproantibodyand serum antibody in cholera patients. J. infect. Dis., 115, 83-87.

"0Girard, J. P. (1970). Immunoglobulins in human duodenal fluid: local immunological response to ingestedbacterial antigen. (Abstr.) In Report on 4th meeting of the European Society of Clinical Investigation,pp. 31-32.

3"Lodinova, R., and Wagner, V. (1970). Development of faecal immunoglobulins and coproantibodies ininfants after artificial oral colonization with E. coli 083. Experientia (Basel), 26, 188.

04Freter, R., and Gangarosa, E. J. (1963). Oral immunization and production of coproantibody in humanvolunteers. J. Immunol., 91, 724-729.

"Formal, S. B., Kent, T. H., May, H. C., Palmer, A., Falkow, S., and LaBrec, E. H. (1966). Protection ofmonkeys against experimental shigellosis with a living attenuated oral polyvalent dysentery vaccine. J.Bact., 92, 17-22.

'Mel, D. M., Arsic, B. L., Nikolic, B. D., and Radovanic, M. L. (1968). Studies on vaccination against bacillarydysentery. IV. Oral immunization with live monotypic and combined vaccines. Bull. Wld Hith Org., 39,375-380.

"'Dupont, H. L., Hornick, R. B., Snyder, M. J., Dawkins, A. T., Heiner, G. G., and Woodward, T. E. (1971).Studies of immunity in typhoid fever, protection induced by killed oral antigens or by primary infection.Bull Wld Hlth Org., 44, 667-672.

"°'Cvjetanovi6, B., Mel, D. M., and Felsenfeld, 0. (1970). Study of live typhoid vaccine in chimpanzees. Bull.Wld Hlth Org., 42, 499-507.

0'Sanyal, S. C., and Mukerjee, S. (1969). Live oral cholera vaccine: report of a trial on human volunteersubjects. Bull. Wld Hlth Org., 40, 503-51 1.

"'Sabin, A. B., Michaels, R. H., Krugman, S., Eiger, M. E., Berman, P. H., and Warren, J. (1963). Effect oforal poliovirus vaccine in newborn children. 1. Excretion of virus after ingestion oflarge doses oftype 1 orof a mixture ofall three types, in relation to level ofplacentally transmitted antibody. Pediatrics, 31, 623-640.

"'Sabin, A. B., Michaels, R. H., Ziring, P., Krugman, S., and Warren, J. (1963). Effect oforal poliovirus vaccinein newborn children. 2. Intestinal resistance and antibody response at 6 months in children fed type 1vaccine at birth. Pediatrics, 31, 641-650.

"'Kawakami, K., Tatsumi, H., Tatsumi, M., and Kono, R. (1966). Studies on poliovirus coproantibody. I.Neutralising antibodies in feces of children following Sabin oral poliovirus vaccination. Amer. J.Epidemiol., 83, 1-13.

"'Douglas, R. G., Jr., Rossen, R. D., Butler, W. T., and Couch, R. B. (1967). Rhinovirus neutralizing antibodyin tears, parotid saliva, nasal secretions and serum. J. Immunol., 99, 297-303.

"'Akao, Y., Sasagawa, A., Shiga, S., and Kono, R. (1971). Comparative studies on the mode of neutralizationreaction of poliovirus type 2 with serum IgG and secretory IgA from mother's milk and fecal extract.Jap. J. med. Sci., 24, 135-152.

"'Katz, J., Spiro, H. M., and Herskovic, T. (1968). Milk precipitating substance in the stool in gastrointestinalmilk sensitivity. New EngI. J. Med., 278,1191-1194.

"Wubbena, P., and Buckley, R. (1969) In Proceedings of the 25th Annual Meeting of the American Academy oJAllergy, Miami Beach, Florida, March 1969.

"'Katz, J., Kantor, F. S., and Herskovic, T. (1968). Intestinal antibodies to wheat fractions in celiac disease.Ann. intern. Med., 69, 1149-1153.

"Barnum, D. A., Glantz, P. J., and Moon, H. W. (1967). Colibacillosis. CIBA Vet. Monogr. Ser., 2, 1-44."Logan, E. F., and Penhale, W. J. (1971). Studies on the immunity of the calf to colibacillosis. III. The local

protective activity of colostrum within the gastrointestinal tract. Vet. Rec., 89, 628-63 1.I"Haworth, J. C., and Dilling, L. (1966). Concentration ofgamm-A-globulin in serum, saliva and nasopharyn-

geal secretions of infants and children. J. Lab. clin. Med., 67, 922-933."'South, M. A., Cooper, M. D., Wollheim, F. A., and Good, R. A. (1968). The IgA system II. The clinical

significance of IgA deficiency: studies in patients with agammaglobulinaemia and ataxin-telangiectasia.Amer. J. Mcd., 44, 168-178.

"2Selner, J. C., Merrill, D. A., and Claman, H. N. (1968). Salilvary immunoglobulin and albumin: developmentduring the newborn period. J. Pediat., 72, 685-689.

"'Rule, A. H., Lawrence, D., Hager, HI. J., Hyslop, N., Jr., and Shwachman, H. (1971). IgA: presence inmeconium obtained from patients with cystic fibrosis. Pediatrics, 48, 601-604.

"'Kenny, J. F., Boesman, M. I., and Michaels, R. H. (1967). Bacterial and viral coproantibodies in breast-fedinfants. Pediatrics, 39, 202-213.

"'Michael, J. G., Ringenback, R., and Hottenstein, S. (1971). The antimicrobial activity of human colostralantibody in the newborn. J. infect. Dis., 124, 445-448.

""'Hodes, H. L. (1964). Poliomyelitis antibodies in human colostrum and milk. J. Pediat., 65, 319-320."'Sabin, A. B., and Fieldsteel, A. H. (I 962). Antipoliomyelitic activity ofhuman and bovine colostrum and milk.

Pediatrics, 29, 105-115."'Hodes, H. L., Berger, R., Ainbender, E., Hevizy, M. M., Zepp, H. D., and Kochwa, S. (1964). Proof that

colostrum polio antibody is different from serum antibody. J. Pediat., 65, 1017-1018."'Lubin, B., Boesman, M. I., Michaels, R. H., Kenny, J. F., and Gitlin, D. (1964). A comparison of the

V-globulins of human breast milk and serum. J. Pediat., 65, 1103-1104.3'Adinolfi, M., Glynn, A. A., Lindsay, M., and Milne, C. M. (1966). Serological properties ofVA antibodies to

Escherichia coli present in human colostrum. Immunology, 10, 517-526."'Agunod, M., Yamaguchi, N., Lopez, R., Luhby, A. L., and Glass, G. B. J. (1969). Correlative study ofhydro-

chloric acid, pepsin, and intrinsic factor secretion in newborns and infants. Amer. J. dig. Dis., 14, 400-414.

"Adinolfi, M., Mollison, P. L., Polley, M. J., and Rose, J. M. (1966). VA-Blood group antibodies. J. exp. Med.,123, 951-967.

"3Hochwald, G. M., Jacobson, E. B., and Thorbecke, G. J. (1964). C"'-amino acid ancorporation into trans-ferrin and f, A-globulin by ectodermal glands in vitro. (Abstr.) Fed. Proc., 23, 557.



j.com/


ownloaded from

http://gut.bmj.com/


3'Murillo, G. J. (1971). Synthesis of secretory IgA by human colostral cells. 5th. med. J., Bgham. (Ala.) 64,1333-1337.

"'Furth, R. van, Schuit, H. R., and Hijmans, W. (1966). The formationofimmunoglobulins by human tissues invitro. IV. Circulating lymphocytes in normal and pathological conditions. Immunology, 11, 29-40.

'"Chessin, L. N., Glade, P. R., Asofsky, R., Baker, P. D., Reisfeld, R., and Terry, W. (1968). Studies on humanperipheral blood lymphocytes in vitro. V. Biosynthesis of immunoglobulins. J. Immunol., 101, 4 58468.

'"Levi, D. (1941). Surgical treatment of congenital pyloric stenosis. Brit. med. J., 1, 963-965."31Hinton, N. A., and MacGregor, R. R. (1958). A study of infections due to pathogenic serogroups of E.

coli. Canad. med. Ass. J., 79, 359-364."3Mata, L. J., and Urrutia, J. J. (1971). Intestinal colonization of breast-fed children in a rural area of low

socioeconomic level. Ann. N. Y. Acad. Sci., 176, 93-109.'°Winberg, J., and Wessner, G. (1971). Does breast milk protect against septicaemia in the newborn ?Lancet, 2,

1091-1094.4'Ross, C. A. C., and Dawes, E. A. (1954). Resistance of the breast-fed infant to gastro-enteritis. Lancet, 1,

994-998."'Bullen, C. L., and Willis, A. T. (1971). Resistance of the breast-fed infant to gastroenteritis. Brit. med. J., 3,

338-343."'Bullen, J. J., Rogers, H. J., and Leigh, L. (1972). Iron-binding proteins in milk and resistance to E. coli

infection in infants. Brit. med. J., 1, 69-75.4'Ishizaka, T., Ishizaka, K., Borsos, T., and Rapp, H. (1966). C'l fixation by human isoagglutinins: fixation of

C'l by v-G and y-M but not by y-A antibody. J. Immunol., 97, 716-726."'Shklair, I. L., Rovelstad, G. H., and Lamberts, B. L. (1969) A study ofsome factors influencing phagocytosis

of cariogenic streptococci by caries-free and caries-active individuals. J. dent. Res., Suppl. 5, 48, 842-845.

'"Wernet, P., Breu, H., Knop, J., and Rowley, D. (1971). Antibacterial action of specific IgA and transport ofIgM, IgA, and IgG from serum into the small intestine. J. infect. Dis., 124, 223-226.

"'Burrows, W., Kaur, J., and Cercavski, L. (1971). The cholera enterotoxin and local immunity. Ann. N. Y.Acad. Sci., 176, 323-329.

"'Felsenfeld, O., Greer, W. E., and Felsenfeld, A. D. (1967). Cholera toxin neutralization and some cellular sitesof immune globulin formation in Cercopithecus aethiops. Nature (Lond.), 213, 1249-1251.

"Newcomb, R. W., Ishizaka, K., and DeVald, B. L. (1969). Human IgG and IgA diptheria antitoxins in serum,nasal fluids and saliva. J. Immunol., 103, 215-224.

"'Stollerman, G. H., and Ekstedt, R. (1957). Long chain formation by strains of group A streptococci in thepresence of homologous antiserum: a type specific reaction. J. exp. Med., 106, 345-355.

"'5Hahn, J. J., and Cole, R. M. (1963). Streptococcal M antigen location and synthesis, studied by immuno-fluorescence. J. exp. Med., 118, 659-666.

"'Dixon, J. M. S. (1960). The fate of bacteria in the small intestine. J. Path. Bact., 79, 131-140.'3Arbuckle, J. B. R. (1970). The location of E. coli in the pig intestine. J. med. Microbiol., 3, 333-340."4Labrec, E. H., Schneider, H., Magnani, T. J., and Formal, S. B. (1964). Epithelial cell penetration as an

essential step in the pathogenesis of bacillary dysentery. J. Bact., 88, 1503-1518."Gibbons, R. J., and van Houte, J. (1971). Selective bacterial adherance to oral epithelial surface and its role

as an ecological determinant. Infect. and Immunity, 3, 567-573."6Freter, R. (1969). Studies of the mechanism ofaction of intestinal antibody in experimental cholera. Tex. Rep.

Biol. Med., 27, 299-316."'5Freter, R. (1970). Mechanism ofaction of intestinal antibody in experimental cholera. Infect. and Immunity, 2,

556-562.'"'South, M. A. (1971). Enteropathogenic E. coli disease: new developments and perspectives. J. Pediat., 79,

1-11."'Beale, A. J., Parish, W. E., Douglas, A. P., and Hobbs, J. R. (1971). Impaired IgA responses in coeliac disease.

Lancet, 1, 1198-1200."'Buckley, R. H., and Dees, S. C. (1969). Correlation ofmilk precipitins with IgA deficiency. New Engl. J. Med.,

281, 465-469."'Tomnasi, T. B., Jr., and Katz, L. (1971). Human antibodies against bovine immunoglobulin M in IgA deficient

sera. Clin. exp. Immunol., 9, 3-10.62Ammann, A. J., and Hong, R. (1971). Anti-antiserum antibody as a cause of double precipitin rings in

immunoglobu in quantitation and its relation to milk precipitins. J. Immunol., 106, 567-569."'Leikola, J., and Vyas, G. N. (1971). Human antibodies to ruminant IgM concealing the absence of IgA in

man. J. Lab. clin. Med., 77, 629-638.64Heiner, D. C., Lahey, M. E., Wilson, J. F., Gerrard, J. W., Shwachman, H., and Khaw, K.-T. (1962). Pre-

cipitins to antigens of wheat and cow's milk in celiac disease. J. Pediat., 61, 813-830."'Alarc6n-Segovia, D., Herskovic, T., Wakim, K. G., Green, P. A., and Scudamore, H. H. (1964). Presence of

circulating antibodies to gluten and milk fractions in patients with nontropical sprue. Amer. J. Med., 36,485499.

"'Crabb6, P. A., and Heremans, J. F. (1966). Lack of gamma-A-immunoglobulin in serum of patients withsteatorrhoea. Gut, 7, 119-127.

'7Asquith, P., Thompson, R. A., and Cooke, W. T. (1969). Serum immunoglobulins in adult coeliac disease.Lancet, 2, 129-131.

"'Gruskay, F. L., and Cooke, R. E. (1955). The gastrointestinal absorption of unaltered protein in normalinfants and in infants recovering from diarrhea. Pediatrics, 16, 763-769.

"9'Matthews, D. M., and Laster, L. (1965). Absorption of protein digestion products: a review. Gut, 6, 411-426."'7Peterson, R. D. A., and Good, R. A. (1963). Antibodies to cow's milk proteins-their presence and signifi-

cance. Pediatrics, 31, 209-221."'Rothberg, R. M., and Farr, R. S. (1965). Anti-bovine serum albumin and anti-alpha lactalbumin in the serum

of children and adults. Pediatrics, 35, 571-588.72Lehner, T., Cardwell, J. E., and Clarry, E. D. (1967). Immunoglobulins in saliva and serum in dental caries.

Lancet, 1, 1294-1296.'73irus, I., Siegel, H., Yamaguchi, N., and Glass, G. B. J. (1968). Immunoglobulins IgA and IgG in gastric

mucosa of patients with atrophic gastritis and pernicious anaemia. Scand. J. Gastroent., 3, 43-57."'7Goldberg, L. S., Shuster, J., Stuckey, M., and Fudenberg, H. H. (1968). Secretory immunoglobulin A:

autoantibody activity in gastric juice. Science, 160, 1240-1241.



j.com/


ownloaded from

http://gut.bmj.com/


"5Strickland, R. G., Baur, S., Ashworth, L. A. E., and Taylor, K. B. (1971). A correlative study of immuno-logical phenomena in pernicious anaemia. Clin. exp. Immunol., 8,25-36.

""Goldberg, L. S., and Bluestone, R. (1970). Hidden gastric autoantibodies to intrinsic factor in perniciousanemia. J. Lab. clin. Med., 75, 449-456.

"S0ltoft, J. (1970). Immunoglobulin-containing cells in non-tropical sprue. Clin. exp. Immunol., 6, 413-420.7'Gelzayd, E.A.,Kraft,S. C., Fitch, F. W.,and Kirsner,J. B. (1968). Distribution of immunoglobulins in human

rectal mucosa. II. Ulcerative colitis and abnormal mucosal control subjects. Gastroenterology, 54, 341-347.

1"Crabbe, P. A. and Heremans, J. F. (1966). Presence of large numbers of plasma cells containing IgD in therectal mucosa of a patient with ulcerative colitis. Acta clin. belg., 21, 73-83.

The May 1972 IssueTHE MAY 1972 ISSUE CONTAINS THE FOLLOWING PAPERS

Partial gastrectomy for haemorrhage J. R. COCKS,A. M. DESMOND, B. F. SWYNNERTON, AND N. C. TANNER

The role of the sympathetic nervous system inhypoglycaemia-stimulated gastric secretion A. J.HODGE, J. R. MASAREI, AND B. N. CATCHPOLE

Extragastric gastrin M. G. KORMAN, C. SOVENY, ANDJ. HANSKY

The functional 'G' cell mass in atrophic gastritisM. G. KORMAN, R. G. STRICKLAND, AND J. HANSKY

Normal deglutitive responses of the human loweroesophageal sphincter MICHAEL D. KAYE AND J.PHILIP SHOWALTER

Inhibition of pancreatic secretion in man by cigarettesmoking T. E. BYNUM, TRAVIS E. SOLOMON, LEONARDR. JOHNSON, AND EUGENE D. JACOBSON

Effect of rifampicin on liver function in manPHILIPPE CAPELLE, DANIEL DHUMEAUX, MICHEL MORA,GERARD FELDMANN, AND PIERRE BERTHELOT

Chronic Budd-Chiari syndrome due to obstruction ofthe intrahepatic portion of inferior vena cava D. V.DATTA, S. SAHA, SAMANTA A. K. SINGH, B. B. GUPTA,B. K. AIKAT, K. S. CHUGH, AND P. N. CHHUTTANI

An immunofluorescent study of the distribution ofimmunoglobulin-containing cells in the normal andthe inflamed human gall bladder F. H. Y. GREENAND H. FOX

5-Hydroxyindole-secreting rectal carcinoid tumourIAIN M. MURRAY-LYON, M. SANDLER, H. D. CHEETHAM,J. A. E. WATTS, AND ROGER WILLIAMS

The effect of lymphocytes from sufferers fromrecurrent aphthous ulceration upon colon cells intissue culture A. E. DOLBY

Radioimmunoassay of carcinoembryonic antigen inserum of normal subjects and patients with coloniccarcinoma MARTIN S. KLEINMAN AND MICHAEL D.TURNER

Interference by Gastrografin with a spectrophoto-metric trypsin assay A. E. COWEN. HEATHER M.MCGEARY, AND C. B. CAMPBELL

Progress report The interrelationships of the pan-creatic enzymes K. G. WORMSLEY AND D. M.GOLDBERG

Notes and activities

Copies are still available and may be obtained from the PUBLISHING MANAGER,BRIIISH MEDICAL ASSOCIATION, TAVISTOCK SQUARE, LONDON, WCIH 9JR, price 87ip



j.com/


ownloaded from

http://gut.bmj.com/

The demonstration and antibodies in the gastrointestinal tract · destruction of immunoglobulin by...

Documents

Transcript of The demonstration and antibodies in the gastrointestinal tract · destruction of immunoglobulin by...