immunofluorescent localization of enteroglucagon cells the ...immunofluorescent localization...

Gut, 1971, 12, 311-318

Immunofluorescent localization of enteroglucagoncells in the gastrointestinal tract of the dogJULIA M. POLAK, S. BLOOM, I. COULLING, AND A. G. E. PEARSE

From the Department of Histochemistry, Royal Postgraduate Medical School, and the Institute of ClinicalResearch, Middlesex Hospital, London

SUMMARY Localization of the endocrine polypeptide cells responsible for 'glucagon-like immuno-reactivity' in the gastrointestinal tract of the dog has been achieved with an immunofluorescenttechnique using antibodies raised against porcine pancreatic glucagon. The cells, for which we

prefer the term 'enteroglucagon', could only be demonstrated by this technique in tissues fixed incarbodiimide.The enteroglucagon cells possess cytological, cytochemical, and ultrastructural characteristics

in common with those of the pancreatic CX2 cell and they are equivalent in the stomach to the A celland in the intestine to the L cell of the Wiesbaden terminology. Their distribution, predominantlyin fundus and jejunum, correlates precisely with the distribution of glucagon-like immunoreactivityby radioimmunoassay and bioassay.The storage form of enteroglucagon differs in many respects from that of pancreatic glucagon

although there are some close resemblances between the two forms of specific hormone-containinggranule. Elucidation of the role of enteroglucagon should be assisted by the ability to demonstrateenteroglucagon cells.

Finding that a pancreatectomized dog did not die,but lived to develop symptoms of what came to berecognized, over two centuries later, as experimentaldiabetes, Brunner (1683) concluded that the pancreaswas not so vitally important as had been supposed.He suggested that its functions could be taken overby other glands (eiusdem cum aliarum glandularum).A similar conclusion was drawn by Martinotti(1882), namely that dysfunction of the pancreascould be compensated by an augmented action ofthe glands of Lieberkuhn.These observations were clearly supported when

Sutherland and de Duve (1948) identified a glyco-genolytic, hyperglycaemic substance in acid-ethanolextracts of dog stomach and duodenum. At a laterdate, extracts of human stomach, duodenum, smallbowel, and colon were reported by Makman andSutherland (1964) to possess the glucagon-likeproperty of stimulating adenyl cyclase. Glucagon-like immunoreactivity was first identified in acid-ethanol extracts of canine stomach and duodenumby Unger, Eisentraut, Sims, McCall, and Madison

Received for publication 9 February 1971.

(1961) and subsequently in the gastrointestinal tractof all 12 species examined by Assan, Rosselin, andTchobroutsky (1968).The existence of two fractions of glucagon-like

immunoreactivity (GLI), obtained by filtration ofacid-ethanol extracts of canine jejunum, was re-ported by Unger, Ohneda, Valverde, Eisentraut,and Exton (1968) and by Valverde, Rigopoulou,Exton, Ohneda, Eisentraut, and Unger (1968). Thelarger fraction had a molecular weight of approxi-mately 7000, double the weight of pancreatic glu-cagon, and it was devoid of glycogenolytic, hyper-glycaemic properties. The lesser fraction was equalin apparent molecular weight to pancreatic glucagonand it possessed the two properties missing from thelarger fraction. Both fractions caused marked insulinrelease in dogs during hyperglycaemia and both werefound to differ immunologically from pancreaticglucagon. According to Valverde, Rigopoulou,Marco, Faloona, and Unger (1970) simple enzy-matic and chemical procedures could not convertthe heavier fraction of intestinal GLI into thelighter one.

In man, rat, and cow, it has been reported that311

on May 21, 2021 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.12.4.311 on 1 A

pril 1971. Dow

nloaded from

http://gut.bmj.com/

312

the gastric and duodenal content of GLI is rela-tively low but substantial amounts are present injejunum and ileum, the more distal segments tendingto have the highest levels. In these species the totalGLI of the gastrointestinal tract is less than thetotal pancreatic glucagon. In the dog, however,there is much more GLI in the stomach and thetotal gastrointestinal GLI is equal to total pancreaticGLI. Using radioimmunoassay, Unger, Ketterer,and Eisentraut (1966) found that immunologicalglucagon reactivity wa, low in the canine gastricantrum and high in the fundus. It was associatedwith the mucosal layer in both stomach and intestineand not, as described by Makman and Sutherland(1964), with the muscular layer. The proportionspresent in the villi and crypts were different injejunum and ileum.

Efforts to demonstrate the localization of GLI inthe gastrointestinal tract by immunofluorescence,carried out in several laboratories, have hithertobeen unsuccessful, with a single reported exception.This is the demonstration by Baxter-Grillo (1970)of fluorescence in 'all the cells of the epithelium' ofthe duodenal villi in a 16-day-old chick embryo,using an antiserum to porcine pancreatic glucagon.

Structural evidence for 'glucagon-producing cells'in the intestinal mucosa of the rat was produced byOrci, Pictet, Forssmann, Renold, and Rouiller(1968). Specifically, they claimed that one of the twotypes of non-digestive epithelial cell, as identifiedby electron microscopy, was responsible for theproduction of enteroglucagon. The cell responsiblecontained secretion granules closely resemblingthose of the pancreatic insular a2 cell and it was

therefore described as the 'intestinal A cell' byForssmann, Orci, Pictet, Renold, and Rouiller(1969). In the first paper of the series (Orci et al,1968) the average diameter of the invariably roundgranules of the A cell was recorded as 250 nm. In thesecond paper (Forssmann et al, 1969) it was de-scribed as varying between 500 and 700 nm. At thehigher end of this scale granules should be visibleby optical microscopy.

In the fundus of cat, dog, and rabbit stomachVassallo, Solcia, and Capella (1969) found a smallnumber of endocrine cells with an ultrastructuralpattern similar to the pancreatic a2 cell, with numer-ous dense granules with an average diameter of200 nm. Using dark-field microscopy they foundcells whose granules gave a silver-white luminiscence,a property exhibited by the granules of the pan-creatic a2 cells, in the fundic mucosa of dog and catbut not in the pylorus or duodenum.With these indications in mind, the present work

was undertaken in order to ascertain (1) whetherany of the endocrine polypeptide cells in the gastro-

Julia Polak, S. Bloom, L Coulling, and A. G. E. Pearse

intestinal mucosa of the dog contain a hormoneimmunologically similar to pancreatic glucagon;(2) whether there are cytological, cytochemical, orultrastructural resemblances between any endocrinepolypeptide cell and the pancreatic islet a2 cell; and(3) whether cells reacting with anti-glucagon sera,if present, can be shown to possess a2 cell charac-teristics.We were able to detect immunofluorescent cells

in both stomach and intestine, which reacted withantibodies to pancreatic glucagon, and we decidedto call the substance responsible by its hormonalappellation 'enteroglucagon' rather than to employthe perhaps more accurate term 'glucagon-likeimmunoreactivity'.

Materials and Methods

Four puppies (two labrador crosses and twomongrels), between 8 and 12 weeks old wereused. Following intraperitoneal injection ofnembutal, samples of mucosa were taken from thegastric fundus, cardia and pylorus, duodenum,high, mid- and low jejunum, high, mid- and lowileum, colon, and pancreas.

Small pieces of mucosa from each of the aboveregions, and similar small pieces of pancreas, wereprocessed in a variety of different ways, for thevarious procedures outlined below.

IMMUNOFLUORESCENCESamples were fixed at 40 in formol-calcium,methanol-free paraformaldehyde (Graham andKarnovsky, 1966) or 2% carbodiimide (Kendall,Polak, and Pearse, 1971). To produce this last fixa-tive l-ethyl-3-(3-dimethylaminopropyl)-carbodiimidehydrochloride (Sigma), or l-cyclohyexyl-3-(2-mor-pholinoethyl)-carbodiimide metho-p-toluene sulph-onate (Aldrich) were dissolved in OO1M phosphatebuffer saline at pH 7-1 immediately before use.Paraffin sections were prepared from tissues fixed informol-calcium. Other tissues from this fixative werewashed and stored in gum sucrose. Tissues fixed inparaformaldehyde and in carbodiimide were washedfor 24 hours in 30% sucrose phosphate buffer salineand stored, if necessary, in the same buffer.

Selected tissues were quenched in cold (- 165°)Arcton (Freon) 22 and cryostat sections were pre-pared. These were used either unfixed or post-fixedin cold methanol, cold acetone, or cold formol-calcium.

Cryostat sections were prepared from tissues fixedin formol-calcium, paraformaldehyde, and carbo-diimide.

Tissue was quenched in cold Arcton 22, freeze-



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

Immunofluorescent localization of enteroglucagon cells in the gastrointestinal tract of the dog

dried for eight hours at - 40° in a thermoelectricdryer, and embedded in 450 wax.

All sections were cut at 5 ,m and processed byan indirect immunofluorescence technique (Coons,Leduc, and Connolly, 1955).

AntiseraAntibodies were raised in rabbits and guinea pigsagainst porcine pancreatic glucagon using thecarbodiimide-glucagon polymer technique of Hed-ding (1969). The globulin fraction was obtained bycold precipitation with ammonium sulphate (Nairn,1969). Fluorescein-labelled goat anti-rabbit and anti-guinea pig globulin was obtained commercially(Microbiological Associates Inc).

ControlsThe following controls were used. (1) Anti-glucagonserum with added excess porcine pancreatic glucagonfollowed by fluorescein-labelled antiserum. (2)Normal rabbit or guinea pig globulin, followed bythe second layer. (3) Rabbit anti-human calcitoninand anti-ACTH globulin, followed by the secondlayer. (4) Fluorescein-labelled goat anti-rabbit andanti-guinea pig globulin fraction alone.

OPTICAL MICROSCOPYCytochemical and other techniques characteristicallypositive with pancreatic a2 cells were applied. Theseincluded dark-field luminescence, argyrophilia, leadhaematoxylin, oxidized phosphotungstic acid-haematoxylin (PTAH) the xanthydrol method fortryptophan, and the o-phthalaldehyde method(Takaya, 1970).

Small blocks of mucosa were taken from each ofthe regions already described and treated in thefollowing ways. After fixation in 6% glutaraldehydein 0 1M phosphate buffer saline (pH 7.4), and paraffinembedding, 5 ,tm sections were processed by theMasson-Hamperl (argentaffin) method (Pearse,1960); the Lillie xanthydrol method (Solcia,Sampietro, and Capella, 1969); McConnaill's leadhaematoxylin (Solcia et al, 1969).

After fixation in Bouin's fluid, and paraffin em-bedding, 5gm sections were processed by a modifiedGrimelius technique (de Grandi, 1970) and by theoxidized PTAH method.

After carbodiimide fixation, and the immuno-fluorescence procedure, sections were viewed bydark-field microscopy or restained by lead haema-toxylin or the Masson-Hamperl silver impregnation.

Fresh frozen cryostat sections were examined bydark-field microscopy and, subsequently, they werestained by the o-phthalaldehyde method.

Photomicrographs were taken in Ilford FP4(immunofluorescence) or Pan F (other methods).

ELECTRON MICROSCOPYSmall pieces of tissue from fundic mucosa, mid-jejunum, and pancreas, were processed immediatelyon removal from the animal for electron microscopy.These areas have been shown, by present assaytechniques, to contain the highest levels of intestinalglucagon.

Samples were fixed in 3% glutaraldehyde in0 1M phosphate buffer (pH 7.6) for two hours at 4°.Excess fixative was removed from the blocks byrepeated washing in 0.1M phosphate buffer contain-ing 0.1M sucrose. Following fixation the blocks weredehydrated in an ascending ethanol series, takenthrough epoxy propane, and finally embedded inAraldite CY 212. Other samples, with the primaryfixation described above, were post-fixed in Millonig's(1962) osmium tetroxide at 40 for two hours.

Sections were stained by lead citrate and uranylacetate and viewed in an AEI EM6B electron micro-scope.

Results

IMMUNOFLUORESCENCEImmunofluorescent demonstration of entero-glucagon cells was obtained solely with tissues fixedin carbodiimide (CDI), and cut in the cryostat afterwashing. Other methods of preparation gave negativeor, occasionally, weak and uncertain results. Theglucagon-containing a2 cells of the pancreatic islets,on the contrary, could be demonstrated by immuno-fluorescence after most of the preparative proceduresdescribed.

All regions of stomach and intestine, with theexception of the pylorus and duodenum, containedenteroglucagon (EG) cells. These cells were mostnumerous in the fundus, and in mid- and terminaljejunum. The fundic EG cells were situated mostly inthe middle and deeper portions of the mucosa (Fig. 1)while in the intestine they tended to predominate inthe middle zone. Many of the EG cells appeared tohave their long axes along the basement membrane,and there was no evidence of their reaching thelumen of the gland. Immunofluorescence wasstrongest in the basal portions of the cell giving, attimes, a half-moon appearance. The degree oflocalization in the CDI preparations was precise,and background fluorescence was very low (Fig. 1).All the control sections were free from specificimmunofluorescence.

COMPARATIVE CYTOCHEMISTRYThe EG cells showed the following characteristics:(1) dark-field luminosity with both fresh cryostatand CDI-fixed cryostat sections; (2) fluorescencewith the o-phthalaldehyde method (Fig. 2); (3)

313



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

Julia Polak, S. Bloom, L. Coulling, and A. G. E. Pearse

Fig. 2

Fig. 1

........ t sectFig. 1 Fundic mucosa. Carbodiimide-fixed, cryostatsection. Indirect immnunofluorescence technique. ShowsGLI in three large basigranular EG cells situated in the

t, ~.. deeper portion of the glands. (x 950.)

IES|tj

l.l ............... .. ''

Fig. 2 Fundic mucosa. Fresh frozen cryostat section.The o-phthalaldehyde method shows fluorescence, in four

-large basigranular cells, in the shape ofa half-moon.' ~~~~~~~~~~~~~~~~~~~~~~~~~~~... ........ . .!Fig. 3 Fundic mucosa. Bouin-fixed. Argyrophiliademonstrated by Grimelius silver technique. Both ECand non-EC cells are situated in the lower and middlethird of the glands. Comparison with Figs. 1 and 2suggests that only a proportion of the non-EC cells areEG cells. ( x 950.)

Fig. 3

314



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

Immunofluorescent localization ofenteroglucagon cells in the gastrointestinal tract of the dog

Fig. 4 Jejunal mucosa. a. Specificimmunofuorescence in an EG cell.b. Dark-field photomicrograph ofthesame cell shows refractile granuleslike those ofpancreatic % cells.( x 1,400.)

K .! ]hI .41-

:.

KEC,,,.!ws .,

.: :.

:: :,.

*:t. + +:vj ... K:^wr w =w .. \\D

Fig. 5a Fig. 5bFig. 5 Fundic mucosa. a. Specific immunofluorescence in three EG cells (and part of a fourth). b. Leadhaematoxylin stain on same preparation shows moderately positive granulation in the EG cells with much strongerstaining in two groups ofEC cells. ( x 950.)

moderate tryptophan content, much weaker thanthe (normal) strong reaction given by the EC cells;(4) positive lead haematoxylin staining; (5) argyro-philia blt not argentaffinity (Fig. 3); (6) a weakpositive reaction with oxidized PTAH.To identify the cells showing specific immuno-

fluorescence pairs of reactions or tests were appliedto single sections. A section of the mid-jejunalmucosa shows specific immunofluorescence (Fig. 4a)and dark-field luminosity in the same cell (Fig. 4b).Immunofluorescence sections, post-fixed in 6%glutaraldehyde for 30 minutes, were washed andstained by the lead haematoxylin method. Theresults are shown in Figures 5a and b. The EG cells

gave a weak to moderate positive reaction by con-trast with the strongly reacting EC cells. Othersections, after immunofluorescence, were stained bythe Masson-Hamperl procedure. This demonstratedthat the EG cells were non-argentaffin; that is to saythey were not EC cells. Cryostat sections viewed bydark-field microscopy, and subsequently stainedwith o-phthalaldehyde, showed that the cells withthese two characteristics were identical.

ULTRASTRUCTURAL FINDINGSCells closely resembling pancreatic a2 cells werefound in both fundic and mid-jejunal mucosa (Fig. 6)These cells contained round, densely osmiophilic

315

KI,



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

Julia Polak, S. Bloom, l. Coulling, and A. G. E. Pearse

l11tiFtSis ; , ,3~~~~~~~~~~~~~~~~~~~I

Fig. 6 Fundic mucosa. Electron micrograph shows one Fig. 7 Fundic mucosa. Electron micrograph shows theof the two predominant non-EC endocrine polypeptide secondpredominant non-EC endocrine polypeptide cellcells of this region. This is an A cell (Wiesbaden of the region. This is an EC-like cell (Wiesbadenterminology) ( x 9,000.) terminology). ( x 15,000.)

granules with a diameter of between 200 and 300 nm.In the fundus they were considered to represent theA cells of the Wiesbaden terminology (Pearse,Coulling, Weavers, and Friesen, 1970) while in thejejunum they were regarded as the large granulecells (L) of the same terminology.A second endocrine polypeptide cell was found in

the fundic mucosa. This was the EC-like cell (ECL)shown in Figure 7. Both this cell and the A cellwere clearly distinguishable from the argentaffincell (EC), with its dense polymorphic granules. Nosmall granular cells (S) were found in the mid-jejunal samples examined.

Discussion

We may presume that at least four polypeptidehormones (secretin, cholecystokinin-pancreozymin,gastrin, and enteroglucagon) are produced by theeight or more endocrine cells at present recog-nizable in the mammalian gastrointestinal tract.Gastrin was localized by McGuigan (1968) in theG cells of porcine and human stomach, so named bySolcia, Vassallo, and Sampietro (1967) on the basisof their distribution and staining characteristics.

His observations were confirmed by Bussolati andPearse (1970) for the porcine antrum and by Pearseand Bussolati (1970) for human stomach. Thereremained, therefore, three or four endocrine poly-peptide cells in the stomach, and three in theintestine, which had no established product. Wehave now shown that in dog stomach the fundic Acell is the source of enteroglucagon while in theintestine (jejunum) this hormone is localized in theL cell (large granule). In the human jfundus Pearseet al (1970) were unable to demonstrate A-type cellsand the enteroglucagon demonstrated therein byAssan et al (1968) must be secreted by either D orECL cells, presumably the former. All these cells(A, D, EC-like, and L) belong to the APUD seriesof endocrine polypeptide cells (Pearse, 1968).Following the precedent set by the G cell we proposeto call the enteroglucagon producing cell the EGcell, although in the case of the pancreatic islets theold Greek alphabetical notation has not yet beensuperseded.The successful demonstration of the EG cell was

wholly dependent on prior fixation with carbodi-imide (Kendall et al, 1971). Demonstration of glu-cagon in the pancreatic a. cell was possible, on the

316



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

Immunofluorescent localization of enteroglucagon cells in the gastrointestinal tract of the dog 317

contrary, after a variety of fixatives or even in un-fixed tissues. We postulate that carbodiimides reactwith side-chain carboxyl groups, in the polypeptideor in its precursor protein, to yield intermediateswhich then cross-react with adjacent nucleophiles(ie, protein or polypeptide NH2 groups) to formpeptide bonds. We do not know the structure of thestorage form either of pancreatic glucagon or ofenteroglucagon. The 29 amino acid sequence ofporcine glucagon is known, and it shares a commonC-terminal octapeptide with secretin. Valverde,Rigopoulou, Marco, Faloona, and Unger (1970)indicated that peak II of their gastrointestinal GLIwas a molecule of similar size, glycogenolytic activity,and electrophoretic behaviour to pancreatic gluca-gon. It differed, however, with respect to its im-munological behaviour with anti-porcine and anti-beef glucagon and its molecular structure waspresumed, therefore, to differ from that of pancreaticglucagon. Our observations point to a similar con-clusion. It is clear, moreover, that the conformationof carbodiimide-fixed enteroglucagon, but not ofthe formaldehyde orglutaraldehyde-fixed hormone, issuch that the state ofthe antibody-combining site per-mits the formation of the antigen-antibody complex.

Interesting correlations between the pancreatica2 cell and the EG cell were provided by our stainingand cytochemical studies although there is noexplanation, in terms of structure, for the dark-fieldluminescence and o-phthalaldehyde staining of thetwo granule types. Both are clearly argyrophil andboth contain reactive indole groups (tryptophan)although the EG granules react less strongly thanthe a2 granules.The ultrastructural similarities between a2 cells

and EG cells were precisely those pointed out for a2and intestinal A cells by Orci et al (1968) andVassallo et al (1969). The former authors' suggestion,based solely on electron microscopical observations,that intestinal A cells produce enteroglucagon, wasthus correct.The distribution of EG cells in the dog stomach

and intestine exactly parallels the GLI levels of thedifferent regions, as determined by radioimmuno-assay. We have clearly shown that the enterochro-maffin cells, which are distributed throughout thegastrointestinal tract, contain no enteroglucagon.The physiological function of enteroglucagon has

not been elucidated. The ability to demonstrate thehormone in its storage form in the EG cell shouldfacilitate studies on the response of the latter todifferent stimuli, and throw some light on its role inthe gastrointestinal tract in health and disease.It seems likely that the straightforward views ofBrunner and Martinotti are an oversimplificationof the true state of affairs.

This work was made possible by grants from theWellcome Trust and the Medical Research Council.We are grateful to Dr B. A. L. Hurn (WellcomeResearch Laboratories) for the gift of anti-ACTHserum and to Professor G. Wolf Heidegger for draw-ing our attention to the contributions of Brunner(1683) and Martinotti (1883).References

Assan, R., Rosselin, G., and Tchobroutsky, G. (1968). Glucagon inblood and immunological specificity of glucagon moleculeprotein and polypeptide hormones. Excerpta Med. (Amst.)Sect., pp. 87-90.

Baxter-Grillo, D. L. (1970). Enzyme histochemistry and hormones ofthe developing gastrointestinal tract of the chick embryo. III.Enterochromaffin cells-their possible products, glucagon, 5-hydroxytryptamine and the relation of monoamine oxidase.Histochemie, 21, 129-135.

Brunner, J. C. (1683). Experimenta circa Pancreas, accedit diatribe deLympha et Genuino Pancreatis Usu. Amsterdam.

Bussolati, G., and Pearse, A. G. E. (1970). Immunofluorescentlocalization of the gastrin-secreting G cells in the pyloricantrum of the pig. Histochemie, 21, 1-4.

Coons, A. H., Leduc, E. H., and Connolly, J. M. (1955). Studies onantibody production. I. A method for the histochemicaldemonstration of specific antibody and its application to astudy of the hyperimmune rabbit. J. exp. Med., 102, 49-60.

de Grandi, P. (1970). The routine demonstration of C cells in humanand animal thyroid glands. Virchows Arch. Zellpath., 6, 137-150.

Forssmann, W. G., Orci, L., Pictet, R., Renold, A. E., and Rouiller,C. (1969). The endocrine cells in the epithelium of the gastro-intestinal mucosa of the rat. J. Cell Biol., 40, 692-715.

Graham, R. C., and Karnovsky, M. J. (1966). The early stages ofabsorption of injected horseradish peroxidase in the proximaltubules of mouse kidney: ultrastructural cytochemistry by anew technique. J. Histochem. Cytochem., 14, 291-302.

Hedding, L. G. (1969). The production of glucagon antibodies inrabbits. Hormone and Metabolic Res., 1, 87-88.

Kendall, P. A., Polak, Julia M., and Pearse, A. G. E. (1971). Car-bodiimide fixation for immunohistochemistry: Observations onthe fixation of polypeptide hormones. Experientia (Basel), inthe press.

McGuigan, J. E. (1968). Gastric mucosal intracellular localization ofgastrin by immunofluorescence. Gastroenterology, 55, 315-327.

Makman, M. H., and Sutherland, E. W., Jr. (1964). Use of liver adenylcyclase for assay of glucagon in human gastro-intestinal tractand pancreas. Endocrinology, 75, 127-134.

Martinotti, G. (1888). Sulla estirpazione del pancreas. G. Accad. Med.Torino, 36, 348-351.

Millonig, G. (1962). Further observations on a phosphate buffer forosmium solutions in fixation. In Proceedings of the 5th Inter-national Congress for Electronmicroscopy, Philadelphia.Edited by Sydney S. Breese, Jr. Academic Press, New York.

Nairn, R. C. (1969). Fluorescent Protein Tracing. Livingstone, Edin-burgh and London.

Orci, L., Pictet, R., Forssmann, W. G., Renold, A. E., and Rouiller,C. (1968). Structural evidence for glucagon producing cells inthe intestinal mucosa of the rat. Diabetologia, 4, 56-57.

Pearse, A. G. E. (1960). Histochemistry, Theoretical and Applied.Churchill, London.

Pearse, A. G. E. (1968). Common cytochemical and ultrastructuralcharacteristics of cells producing polypeptide hormones (TheAPUD Series) and their relevance to thyroid and ultimo-branchial C cells and calcitonin. Proc. roy. Soc. B., 170, 71-80.

Pearse, A. G. E., and Bussolati, G. (1970). Immunofluorescence studiesof the distribution of gastrin cells in different clinical states.Gut, 11, 646-648.

Pearse, A. G. E., Coulling, I., Weavers, B., and Friesen, S. (1970).The endocrine polypeptide cells of the human stomach, duo-denum and jejunum. Gut, 11, 649-658.

Solcia, E., Sampietro, R., and Capella, C. (1969). Differential stainingof catecholamines, 5-hydroxytryptamine and related com-pounds in aldehyde fixed tissues. Histochemie, 17, 273-283.

Solcia, E., Vassallo, G., and Sampietro, R. (1967). Endocrine cells inthe antro-pyloric mucosa of the stomach. Z. Zellforsch., 81,474-486.



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

318 Julia Polak, S. Bloom, I. Coulling, and A. G. E. Pearse

Sutherland, E. W., and de Duve, C. (1948). Origin and distribution ofthe hyperglycemic-glycogenolitic factor of the pancreas. J.biol. Chem., 175, 663-674.

Takaya, K. (1970). A new fluorescent stain with o-phthalaldehyde forA cells of the pancreatic islets. J. Histochem. Cytochem., 18,178-186.

Unger, R. H., Eisentraut, A. M., Sims, K., McCall, M. S., andMadison, L. L. (1961). Sites of origin of glucagon in dogs andhumans. Clin. Res., 9, 53.

Unger, R. H., Ketterer, H., and Eisentraut, A. M. (1966). Distributionof immunoassayable glucagon in gastrointestinal tissues.Metabolism, 15, 865-867.

Unger, R. H. Ohneda, A., Valverde, I.. Eisentraut, A. M., and

Exton, J. (1968). Characterization of the responses of circulat-ing glucagon-like immunoreactivity to intraduodenal and jin-travenous administration of glucose. J. clin. Invest., 47, 48-65.

Valverde, I., Rigopoulou, D., Exton, J., Ohneda, A., Eisentraut, A.,and Unger, R. H. (1968). Demonstration and characterizationof a second fraction of glucagon-like immunoreactivity injejunal extracts. Amer. J. med. Sci., 255, 415-420.

Valverde, I., Rigopoulou, D., Marco, J., Faloona, G. R., and Unger,R. H. (1970). Characterization of glucagon-like immunore-activity (GLI). Diabetes, 19, 614-623.

Vassallo, G., Solcia, E., and Capella, C. (1969). Light and E.M.identification of several types of endocrine cells in the gastro-intestinal mucosa of the cat. Z. Zellforsch., 98, 333-356.



j.com/


pril 1971. Dow

nloaded from

http://gut.bmj.com/

immunofluorescent localization of enteroglucagon cells the ...immunofluorescent localization...

Documents