Search

To investigate the enterotrophic effects of bolus doses of long chain triglycerides, two groups of eight female Wistar rats were fed identical diets with 48.2% total calories as the essential fatty acid rich oil Efamol. To one group the oil was given in twice daily bolus doses by gavage, while for the other group the oil was mixed with the remainder of the feed and thus consumed over 24 hours. The animals were killed after 20 to 22 days. Bolus dosing significantly increased parameters of mucosal mass along the length of the small intestine in association with an increase in two hour accumulation of vincristine arrested metaphases in small intestinal crypts. In a second experiment, four replicate studies were carried out, each involving two groups of 12 rats respectively fed as described above. After 21 days one animal from each group was killed every two hours, providing regular plasma samples over 24 hours for measurement of gastrin, cholecystokinin, peptide tyrosine-tyrosine and enteroglucagon. Bolus dosing markedly enhanced release of peptide tyrosine-tyrosine and enteroglucagon, but not of gastrin or cholecystokinin. Thus, the enhanced enterotrophic effects of bolus doses of long chain triglycerides could be mediated by release of a distally located gut peptide, perhaps enteroglucagon.

The distribution of immunoreactivity for the potent vasoconstrictor endothelin-1 was studied in adult and developing human gut using antisera to endothelin-1 (1-21) and the C terminus of big endothelin-1. The coexistence of these peptides with other neuropeptides was investigated using comparative immunocytochemistry. Endothelin-1 like immunoreactivity was detected in extracts of adult (range 20-60 fmol/g wet weight) and fetal (33 fmol/g) gastrointestinal tract and was shown by chromatography to be the predominant isoform of endothelin present in both. It was localised by immunocytochemistry to ganglion cells in the submucous and myenteric plexuses and to scattered nerves, whereas big endothelin-1 like immunoreactivity was found in the submucous plexus only. Colocalisation studies showed immunoreactivity for both endothelin-1 and vasoactive intestinal peptide in the same ganglion cells of the submucous plexus. Although endothelin-1 immunoreactivity was not detected by immunocytochemistry in the fetal human gut until the 32nd week of gestation, big endothelin-1 was found as early as 11 weeks in the developing neural structures and epithelial cells. The latter were shown to be endocrine cells by their immunoreactivity for chromogranin. Our results indicate that endothelin is a neuropeptide found in adult human gut which shows transient expression in endocrine cells during development.

Results from a recent, new assay suggest that omeprazole, a potent inhibitor of gastric acid secretion, is genotoxic. The principle of this assay is that the non-proliferating zone of surface gastric epithelial cells can be selectively removed by controlled digestion so that any incorporation of tritiated thymidine into these cells represents unscheduled DNA synthesis. Parietal cells (which are located below the uppermost proliferating cells) and proliferating cells in semiconservative, regular DNA synthesis could always be shown in the digested fraction, and as regular DNA synthesis takes up a thousand fold more thymidine than unscheduled DNA synthesis, any signal from unscheduled synthesis would therefore be swamped. The digestion process was also uneven, as histological analysis showed denuded patches of mucosa, and gland like structures were seen in the digest. Quantification of the number of silver grains over the nuclei showed no increase in low level labelling after omeprazole administration, indicating that there was no unscheduled DNA synthesis. The labelling index of undigested gastric tissue from omeprazole treated rats was not significantly different from that of the control group, despite an increase in the plasma gastrin value.

Autoantibodies reacting with endocrine cells in the gastrointestinal mucosa were found by indirect immunofluorescence in 22 out of 268 sera (8.2%) obtained from patients with coeliac disease, Crohn's disease, ulcerative colitis, irritable bowel syndrome, and from subjects without bowel disease. A double immunofluorescence technique showed that the autoantibodies reacted with cells secreting gastric inhibitory polypeptide (glucose dependent insulinotropic polypeptide, GIP), secretin, somatostatin or enteroglucagon. Most sera contained antibodies against more than one cell type. Neither the presence of a particular antibody nor the pattern of antibody combinations appeared to be specific for any diagnostic category. The mean plasma GIP concentrations, however, both fasting and two hours after a test meal, were significantly lower in subjects with GIP cell autoantibodies. Thus gut hormone cell autoantibodies may be markers of impaired hormone secretion.

Pectin has been shown to minimise the fall in blood glucose seen in patients who are troubled by hypoglycaemia attacks after gastric surgery. We therefore performed 50 g glucose tolerance tests with and without 14.5 g pectin on 11 post-gastric surgery patients. After pectin, the high postprandial levels of glucose, insulin, and enteroglucagon were significantly reduced as was the fall in blood glucose between 90 and 120 minutes. These effects of pectin may reflect slower uptake of glucose from the gastrointestinal tract and provide evidence to support the use of unabsorbable carbohydrate gelling agents in treating hypoglycaemia after gastric surgery.

Motilin is found in the upper intestine of pig and man and in man is released by acid. A similar release by acid has now been found in the pig and is associated with markedly reduced immunostainable motilin in the upper small intestine. Clamping the arteries and veins to the stimulated segment immediately reversed the rise of plasma motilin, indicating the motilin release to be an entirely local phenomenon. The apparent half life of endogenous motilin was 3.9 minutes. No release of motilin was seen after a meal and the possible physiological role of motilin thus remains speculative.

Seven patients with gut and pancreatic endocrine tumours have been treated with a long acting somatostatin analogue (SMS 201-995), given as a twice daily subcutaneous injection. This produced dramatic improvement in their endocrine related symptoms, in association with a fall in circulating tumour peptides. One of these patients has now been treated for seven months with this analogue which has controlled his previously life threatening diarrhoea caused by a malignant VIP secreting tumour. He gives his own injections twice daily, and has returned to a full and active life. This is a promising agent both for acute treatment of peptide hypersecretion, and for the long term management of some patients who are unresponsive to other available therapy.

Motilin is a hormonal peptide found in the duodenum and jejunum which potently influences gastrointestinal tract motility. Its role in human physiology is not yet established. After a standard hospital lunch the plasma concentration of motilin showed a small, transient, but significant rise in 28 healthy subjects. Individual food components either stimulated (oral fat) or suppressed release (oral glucose). Plasma motilin levels were, in addition, altered to an equal extent by intravenous nutrients, with glucose and amino acids suppressing release, and intravenous fat causing a significant rise in plasma concentration. These results demonstrate a consistent response to food stimuli, whether oral or intravenous. The release mechanism appears to be complicated and after a balanced meal, containing food components which both stimulate and suppress release, there is only a small net change.

The fasting output of bile into the bowel was investigated in nine healthy volunteers by hepatobiliary scanning. Subjects were studied twice, on each occasion for 2.5 hours. A significant flow of radioactivity from the gall bladder to the duodenum was observed on one or two occasions during each experiment. A 32 +/- 4% mean decrease in counts over the gall bladder was recorded, indicating partial emptying of this organ in the interdigestive state. The output of bile into the bowel was found to be related to fluctuations in fasting plasma motilin levels in that a significant motilin increment (18 +/- 4 pmol/l, p less than 0.005) paralleled the appearance of radioactivity in the duodenum. The onset of gall-bladder emptying regularly preceded the peak in plasma motilin (mean: 25 +/- 2 minutes). Atropine, intravenously, 0.6 mg followed by 0.3 mg, nearly abolished both fasting biliary output and plasma motilin fluctuations. Thus, bile output appears to occur frequently in fasting humans, but our data do not allow any conclusions as to the possible causal relationship between fasting gall-bladder emptying and release of motilin. Cholinergic influences appear to be of importance in the regulation of interdigestive biliary output in man.

A female patient is described with a single pancreatic tumour producing vasoactive intestinal polypeptide (VIP), insulin, and pancreatic polypeptide. The initial presentation was with diarrhoea and hypokalaemia and a raised plasma VIP was demonstrated. Her symptoms improved with metoclopramide administration and absolute concentrations of 28 aminoacid (peak IV) VIP were found to have fallen. She then developed hypoglycaemia with hyperinsulinism. All symptoms resolved after surgical excision. This case emphasises the potential of these tumours to contain more than one endocrine cell type synthesising different biologically active peptides.

During a control infusion noradrenaline and alcohol each provoked carcinoid flushing in four of five patients and pentagastrin in two of five patients. When tetradecapeptide somatostatin was infused on another day no patient flushed at any time, even when 16 microgram of either noradrenaline or pentagastrin were administered. Carcinoid flushing was not associated with release of gastrin or any of the other vasoactive or postprandially released gut regulatory peptides measured. In a sixth patient with severe prolonged carcinoid flushing, subcutaneous Des AA1, 2, 4, 5, 12D Trp8 somatostatin markedly reduced the incidence and severity of flushing for two days. Somatostatin is thus a potent inhibitor of carcinoid flushing, but no evidence has been found for the gut hormones measured to be mediators of flushing.

Somatostatin (cyclic growth-hormone release-inhibiting hormone--GH-RIH) was infused into dogs with gastric fistulae. Somatostatin inhibited gastric acid response to four gastric stimulants--insulin, food, histamine, and pentagastrin. Histamine- and pentagastrin-stimulated pepsins were inhibited similarly to inhibition of acid. Somatostatin inhibited the gastrin response to insulin and food.

Graded doses of 0.6, 1.3, and 3.3 pmol/kg/min of vasoactive intestinal peptide (VIP) were intravenously infused over 30 minute periods in four healthy volunteers and plasma VIP levels were measured by radioimmunoassay. Even with the smallest dose of VIP, plasma concentrations rose markedly above normal values. Infusion of higher VIP doses resulted in mean plateau levels of circulating VIP which were in the range of VIP values found in the Verner-Morrison syndrome. After cessation of the VIP infusions, plasma VIP levels fell strikingly by first order kinetics with an average disappearance half-time of one minute. The apparent metabolic clearance rate was about 9 ml/kg/min and the apparent volume of distribution for VIP was approximately 14 ml/kg. During infusion of the highest VIP dose, previously shown to induce one-fifth maximum pancreatic juice secretion, plasma concentrations of glucose, free fatty acids, and calcium were slightly but significantly raised, the pulse rate and the amplitude of blood pressure were increased, and cutaneous flushing occurred. The spectrum of effects accords well with some abnormalities seen in the Verner-Morrison syndrome. The present data, however, do not support a role for VIP as a circulating hormone, at least under physiological conditions.

Analytical subcellular fractionation techniques in combination with specific radioimmunoassays have been used to investigate the properties of the hormone-containing granules in human jejunal biopsy specimens. After gentle homogenisation, for all hormones except VIP, over 90% of the immunoreactivity was released in intact granules into the post-nuclear supernatant. Approximately 50% of the VIP immunoreactivity sedimented with the low speed pellet, probably reflecting the presence of this peptide in nerve fibres. The following hormone granules (equilibrium densities between parentheses) were separated by isopycnic centrifugation on the sucrose density gradients: VIP (1 x 17), motilin (1 x 20), and secretin (1 x 24). Gastrin and GIP granules were not resolved, both having the same modal density of 1 x 22 g/cm3. This combination of procedures thus provides a new quantitative technique for studying the properties of the gastrointestinal hormone-containing granules, and should therefore be of value in investigating pathophysiological problems in a variety of gastrointestinal diseases.

Peptide histidine isoleucine (PHI) is a newly discovered peptide from porcine intestine, which has sequence homologies with VIP, an established intestinal secretagogue. To study the effects of PHI in human jejunum, natural porcine PHI was infused intravenously at 10.7 +/- 1.7 pmol/kg/min (mean +/- SEM) in normal volunteers during steady state perfusion of the jejunum with an isotonic bicarbonate-electrolyte solution. Plasma PHI concentrations rose to 279 +/- 26 pmol/l (mean +/- SEM) in the first 20 minutes of the infusion reaching 417 +/- 45 at 40 minutes. At these concentrations PHI induced a net secretion of chloride and sodium and either decreased net absorption or increased net secretion of fluid and potassium, while bicarbonate transport remained unaffected. Peptide histidine isoleucine is a potent jejunal secretagogue in man.

Using the methods of immunocytochemistry and radioimmunoassay, five peptides (vasoactive intestinal polypeptide (VIP), substance P, somatostatin, met-enkephalin, and bombesin) have been found in the gall bladder and the biliary tracts of guinea pig and each of them possesses a characteristic distribution pattern. Networks of nerves containing each peptide were found in the smooth muscle, around blood vessels and, occasionally, in the lamina propria. The distribution of the peptide immunoreactive nerves in the gall bladder and biliary tract is similar to those found in the gut. Vasoactive intestinal polypeptide (11 +/- 1.5 pmol/g in the sphincters, mean +/- SEM) and substance P (21.5 +/- 1.8 pmol/g in the common bile duct) were found to be the most abundant peptides and a few VIP and substance P immunoreactive neurones were localised in the ganglionated plexus. Bombesin immunoreactive nerves were mainly seen in the sphincter of Oddi, where the mean concentration of extractable bombesin was 14.6 +/- 2 pmol/g. Somatostatin immunoreactive mucosal endocrine cells were identified in the epithelium of the common bile duct and the sphincter. The extractable somatostatin in these regions were 76 +/- 19 pmol/g and 162 +/- 30 pmol/g respectively.

The gut hormone response to a breakfast meal was studied in 12 subjects hospitalised for an episode of acute diarrhoea (presumed infective) who were otherwise well and in 13 healthy control subjects. Fasting blood glucose concentrations were low but basal insulin concentrations were raised. Basal concentrations of pancreatic polypeptide and both basal and postprandial responses of motilin, enteroglucagon, and vasoactive intestinal polypeptide (VIP) were also significantly greater than controls. No abnormalities in plasma concentrations of gastrin, gastric inhibitory polypeptide (GIP) or pancreatic glucagon were found. The suggested physiological actions of the raised hormones may be relevant to the pathophysiology of diarrhoea.

VIP-, substance P- and met-enkephalin-containing innervation of the human gastroduodenal mucosa and Brunner's glands was studied by immunocytochemistry on whole mount tissue preparations. A dense VIP-containing nerve supply was found around fundic and pyloric glands, while the few and scattered substance P-immunoreactive fibres tended to run across the full thickness of the gastric mucosa. In the duodenum, both VIP and substance P were present in a striking nerve network in the villi as well as in the muscularis mucosae and around blood vessels. Both peptides were also immunostained in nerve bundles and neuronal perikarya between the lobules of Brunner's glands, while only very few fibres reached the proximity of acinar cells. Met-enkephalin-immunoreactivity was detected in a small number of nerve fibres, virtually confined to the basal parts of the mucosa and to the duodenal submucous plexus.

The dose response of duodenal bicarbonate production during synthetic porcine secretin infusions was studied in six healthy volunteers and related to plasma secretin immunoreactivity. Secretin was infused in each individual at four different doses from 0-1 to 2-7 CU/kg/h, each infusion lasting for 60 minutes. Mean maximal bicarbonate secretion was 33 +/- 4 mEq/h. The secretin plasma level for half maximal bicarbonate response was estimated to be 22 pmol/l. As this level is reported to be achieved by intraduodenal acidification in man, it is concluded that secretin may well play a part in the control of duodenal pH.

To assess the association between the putative intestinal trophic hormone enteroglucagon and the development of intestinal tumours, four groups of 20 rats underwent either jejunal transection or 20%, 50%, or 80% proximal small bowel resection. Tumours were induced with azoxymethane 10 mg/kg weekly for 12 weeks. At 26 weeks there was a promotion of colonic neoplasia from a median of 0.5 (range 0-3) per rat in the transection group to 1.0 (0-3) in the 50% resected group (p less than 0.01) but no significant promotion in the 80% resection group. In the small bowel, increasing resection resulted in a progressive promotion of tumours from a median of 1.0 (range 0-3) per rat in the transection group to 2.0 (0-5) in the 50% resection group (p less than 0.001) and 3.0 (0-11) in the 80% group (p less than 0.01). Plasma enteroglucagon was measured at 2, 16, and 26 weeks and was raised seven-fold in the 80% resected group (p less than 0.001). There was a significant correlation between enteroglucagon concentrations and number of duodenal tumours but not colonic tumours. Crypt cell production rate in the duodenum increased from 11.5 +/- 1.9 to 29.2 +/- 1.4 cells/crypt/h in the 80% resected group (p less than 0.001) and showed a close correlation with both enteroglucagon levels and tumour promotion in the small bowel. There were no changes in crypt cell production rate in the colon with resection. This study shows a close association between enteroglucagon concentrations, promotion of tumours and crypt cell production rate in the duodenum but not in the colon.

The effects of porcine vasoactive intestinal peptide (VIP) and bovine pancreatic polypeptide (PP) on jejunal, ileal, and colonic fluid transport were studied in the rabbit. VIP produced secretion in the small intestine (jejunum greater than ileum) but did not affect absorption in the colon. PP had no secretory effects in jejunum, ileum, or colon. The small intestinal secretion induced by VIP was not associated with raised cAMP concentrations in the mucosa; this suggests that the secretory effects of VIP in vivo are mediated by a mechanism other than stimulation of adenylate cyclase.

Immunofluorescence (anti-secretin), cytochemical, and ultrastructural studies were carried out on jejunal biopsies from 16 children with coeliac disease and from 17 controls with suspected malabsorption but normal jejunal morphology. In 11 of the 16 coeliacs there was generalized hyperplasia of endocrine cells and, specifically, of the secretin (S) cells. Further studies, on adult coeliacs as well as on children, combined with serum secretin assays, may establish whether the S cells are abnormally storing hormone because of inability to release it or because of excess production.

Graded doses of 0-1, 0-3, 0-9, and 2-7 clinical units/kg/h of pure synthetic secretin were infused over 60 minute periods in six healthy volunteers. Duodenal bicarbonate output and pH were recorded and plasma secretin and motilin levels were measured by radioimmunoassay. During the infusions plasma motilin fell in a dose dependent manner to a nadir of 35%. This fall was linearly correlated with pancreatic bicarbonate output, whereas a non-linear correlation was observed between plasma motilin and both plasma secretin level and duodenal pH. It is suggested that plasma motilin levels are decreased by secretin-induced pancreatic bicarbonate juice flow. This may be important for the control of motilin secretion initiated by duodenal acidification and the concomitant delay in gastric emptying.

Plasma hormonal responses to insulin hypoglycaemia and to oral and intravenous glucose were investigated in chagasic patients with severe bowel disease and compared with controls matched for age, sex, weight, and race. After intravenous insulin, plasma concentrations of pancreatic glucagon and pancreatic polypeptide (PP) were reduced in the patients with Chagas's disease. These subjects also showed a subnormal rise in plasma insulin after oral glucose. Other hormone responses did not differ significantly from those in the normal controls. These results are compatible with partial denervation of the pancreatic alpha, beta, and PP cells in patients with chronic gastrointestinal Chagas's disease.

The endocrine polypeptide (APUD) cells responsible for the production and storage of secretin have been demonstrated, in canine duodenum, by the application of an indirect immunofluorescence technique, using anti-pure porcine secretin, to carbodiimide-fixed cryostat sections. The cells are identified, by the use of parallel cytochemical and ultrastructural preparations, as the small granular S cells. These are essentially restricted to the transitional zone of the duodenal mucosa, with whose lumen their apical projections probably always make contact. There is no need to alter the title of the cells since S can now stand for secretin instead of for small.

Peptide histidine isoleucine, a 27 amino acid peptide with close amino acid sequence homology to vasoactive intestinal peptide and secretin, is distributed throughout the mammalian intestinal tract, where it has been localised to intramural neurones. An intestinal perfusion technique has been used to study the effect of intravenous peptide histidine isoleucine (44.5 pmol/kg/min) on water and electrolyte transport from a plasma like electrolyte solution in human jejunum in vivo. Peptide histidine isoleucine infusion produced peak plasma peptide histidine isoleucine concentrations in the range 2000-3000 pmol/l, flushing, tachycardia and a reduction in diastolic blood pressure. Peptide histidine isoleucine caused a significant inhibition of net absorption of water, sodium, potassium and bicarbonate and induced a net secretion of chloride, these changes being completely reversed during the post-peptide histidine isoleucine period. These findings suggest that endogenous peptide histidine isoleucine may participate in the neurohumoral regulation of water and electrolyte transport in the human jejunum.

Refeeding starved rats with a fibre free 'elemental' diet increased crypt cell production rate (CCPR) in the proximal small intestine but not in the distal regions of the gut. Little effect on CCPR was seen when inert bulk (kaolin) was added to the 'elemental' diet. Addition of a poorly fermentable dietary 'fibre' (purified wood cellulose) had little effect on intestinal epithelial cell proliferation except in the distal colon where it significantly increased CCPR. A more readily fermentable 'fibre' (purified wheat bran) caused a large proliferative response in the proximal, mid and distal colon and in the distal small intestine. A gel forming 'fibre' also stimulated proliferation in the distal colon. There was no significant correlation between CCPR and plasma gastrin concentrations, but plasma enteroglucagon concentrations were significantly correlated with CCPR in almost all the sites studied. Plasma PYY concentrations also showed some correlation with CCPR, especially in the colon. Thus, whilst inert bulk cannot stimulate colonic epithelial cell proliferation, fermentable 'fibre' is capable of stimulating proliferation in the colon, and especially in the distal colon: it can also stimulate proliferation in the distal small intestine and it is likely that plasma enteroglucagon may have a role to play in this process.

Immunohistochemical studies using an antiserum to a pure porcine vasoactive intestinal peptide, possessing no cross reactivity against the related hormones glucagon, secretin, and gastrin-inhibitory peptide, revealed a wide distribution of vasoactive intestinal peptide cells throughout the entire length of the mammalian and avian gut. The highest numbers of cells were present in the small intestine and more particularly in the large intestine in all species investigated. Three types of endocrine cell in the mammalian gut are sufficiently widely distributed to be considered as the sites for production of vasoactive intestinal peptide. In the avian gut there are only two identifiable cell types. Sequential immunofluorescence and silver staining showed, in the bird, that the enterochromaffin (EC) cell was not responsible. This procedure could not be used in our mammalian gut samples but here serial section immunofluorescence for enteroglucagon and vasoactive intestinal peptide indicated that the two cells were not identical and that each was differently localized in the mucosa. These results leave the D cell of the Wiesbaden classification as the most likely site for the production of vasoactive intestinal peptide. The final identification must come from successful immune electron cytochemistry but this has not yet been achieved.

Estimates of the G cell population were made in 24 resected human pyloric antra from counts of cells in multiple samples and from measurements of antral size. Measurements had been made previously in 20 subjects of acid output (basal and after pentagastrin) and in 10 subjects of plasma gastrin (basal and after insulin + bicarbonate). G cells were most dense near the pylorus, but their circumferential distribution was even. The G cell populations ranged from 8 to 15 (mean 10) million in four control patients and from 3 to 43 (mean 18) million in 15 patients with duodenal ulcer. Those with recurrent ulcer after vagotomy had either a low G cell count and incomplete vagotomy, or a high G cell count and apparently complete denervation. Two patients with hypergastrinaemia and duodenal ulcer had moderate (29 X 10(6)) or marked (56 X 10(6)) excesses of G cells. 'G cell hyperplasia' may represent the extreme end of the normal range of G cell numbers in the antrum, and can be assessed by semi-quantitative grading of G cell hyperplasia in antral biopsies. There were significant direct correlations between antral area and G cell density, between peak acid output and G cell population, and between basal plasma gastrin and G cell density (but not population). We suggest that, in patients with duodenal ulcer, acid and gastrin secretion are interrelated and that both are related to the masses of parietal cells and of G cells.

Immunoperoxidase positive HPP-producing cells have been examined ultrastructurally using the serial semithin-thin section technique. The features of the HPP cells were found to be identical with those of the previously described D1-cell. The D1-cells are scattered throughout the exocrine parenchyma and occur to a minor extent in the islets.

Mouth-caecum transit time (M-CTT) of a lactulose labelled liquid test meal has been measured in 27 coeliac patients and 10 healthy controls using the breath hydrogen technique. Although all patients were urged to maintain a gluten free diet, not all did, and there was, therefore, a wide range in the severity of fat malabsorption within the patient group. Gastric emptying of a 113Indium DTPA-labelled liquid test meal was also assessed in separate studies on six healthy controls and 11 of the coeliac patients. Fasting breath hydrogen concentrations and the response to lactulose, as assessed both by the rate of rise, and the peak breath hydrogen concentration reached, showed no difference between coeliacs and controls, regardless of the presence or absence of steatorrhoea. Mouth-caecum transit time in the 16 coeliac patients with steatorrhea (faecal fat greater than 7 g/24 h) was, however, significantly prolonged being 158 +/- 18 minutes (mean +/- SEM), compared with 70 +/- 9 minutes for the controls (p less than 0.02), and 83 +/- 15 minutes for the 11 coeliacs without steatorrhoea (p less than 0.002). Mouth-caecum transit time in the coeliac patients was linearly related to the 24 hour faecal fat excretion, r = 0.55, n = 27, p less than 0.01. Slow mouth-caecum transit in the coeliacs with steatorrhoea was not caused by delayed gastric emptying as the t1/2 for coeliacs with steatorrhoea was within the normal range. Coeliacs with delayed mouth-caecum transit had impaired insulin release but the postprandial profiles of the other peptides measured (cholecystokinin, GIP, secretin, motilin, neurotensin, enteroglucagon, and peptide YY) were all within the normal range in this group of partially treated coeliac patients.

Indirect immunofluorescence studies using an antiserum to purified porcine gastric inhibitory polypeptide indicate, in the gastrointestinal tract of dog and man, that this polypeptide is present in cells situated predominantly in the mid-zone of the glands in the duodenum and, to a lesser extent, in the jejunum. Absolute correlation of the gastric inhibitory polypeptide cell with one or other of the known endocrine-like cells identified by electron microscopy awaits confirmation by electron immunocytochemistry. It is here identified as an endocrine polypeptide cell of the APUD series and, provisionally, as the D(1) cell. While the hormonal status of a given polypeptide depends ultimately on physiological experiments the present results strengthen the view that gastric inhibitory polypeptide is indeed a hormone.