Mechanism of acid-induced release of secretin in rats. Presence of a secretin-releasing peptide.

In fasting rats, intraduodenal infusion of dilute hydrochloric acid results in significant increases in both pancreatic exocrine secretion and plasma concentration of secretin. To test the hypothesis that acid-induced release of secretin is mediated by a secretin-releasing factor (S-RF), anesthetized rats were prepared with pyloric ligation, duodenal and jejunal cannulas, and pancreatic duct cannulas. Donor rats were infused intraduodenally with 0.01 N HCl, 0.15 M NaCl, or a combination of 0.01 N HCl and 0.05 N NaHCO3 at 0.3 ml/min for 1.5 h, and the perfusates were collected via jejunal cannulas. The perfusates with pH adjusted to 6.0 were concentrated threefold and infused into the duodena of recipient rats. The concentrate of acid perfusate (CAP) significantly increased both pancreatic volume flow and bicarbonate output and plasma concentration of secretin, whereas concentrates of the saline perfusate (CSP) or the perfusate of a combination of 0.01 N HCl and 0.05 N NaHCO3 (CABP) did not influence pancreatic secretion or plasma concentration of secretin. The increased pancreatic secretion by CAP was attributed to increased circulating secretin because when secretin was immunoneutralized by a rabbit antisecretin serum, CAP-stimulated pancreatic secretion was abolished. The bioactivity of CAP was trypsin-sensitive and heat stable. The active substance in CAP had a molecular weight of less than 5,000 and greater than 1,000, as determined by ultrafiltration and bioassay. In conclusion, dilute HCl releases an S-RF into the upper small intestinal lumen to stimulate release of secretin. This substance, with molecular weight of less than 5,000, is heat stable and trypsin sensitive. Thus, the acid-stimulated release of secretin is mediated by a secretin-releasing peptide in the upper small intestinal lumen.

Identification of a transcriptional enhancer important for enteroendocrine and pancreatic islet cell-specific expression of the secretin gene.

It is well established that the gene encoding the hormone secretin is expressed in a specific enteroendocrine cell, the S cell. We now show that the secretin gene is transiently expressed in insulin-producing B cells of the developing pancreatic islets in addition to the intestine. Furthermore, secretin is produced by most established islet cell lines. In order to identify and characterize the regulatory elements within the secretin gene that control tissue-specific expression, we have introduced secretin reporter gene constructions into the secretin-producing HIT and STC-1 cell lines as well as the nonexpressing INR1-G9 glucagonoma line. Analysis of deletion mutants revealed that sequences between 174 and 53 bp upstream from the transcriptional start site are required for maximal expression in secretin-producing cells. This positive element functioned independently of position and orientation. Further deletions into the enhancer resulted in a stepwise loss of transcriptional activity, suggesting the presence of several discrete control elements. The sequence CAGCTG within the secretin enhancer closely resembles that of the core of the B-cell-specific enhancer in the insulin gene. Point mutations introduced into this putative element led to greater than 85% reduction in transcriptional activity. Gel mobility shift assays suggested that a factor in B cells closely related or identical to proteins that bind to the insulin enhancer interacts with the CAGCTG motif in the secretin gene.

Postprandial levels of prolactin and gut hormones in breast cancer patients: association with stage of disease, but not dietary fat.

Previous studies of the relationship between dietary fat and breast cancer have produced conflicting results and have provided no definitive evidence of a mechanistic link between fat and breast tumorigenesis. We conducted a study to compare postprandial levels of prolactin (Prl), a hormone suspected of promoting the growth of some human breast cancer, and several gut hormones, i.e., gastrin (Gs), vasoactive intestinal polypeptide (VIP), neurotensin (Nt), and cholecystokinin (CCK), following high- and low-fat isocaloric test meals. Data were obtained in the posttreatment period from 13 patients with breast cancer (nine stage I and four stage II), who were disease free clinically, and nine healthy controls. Subjects admitted to the research unit on 2 days were given the high-fat meal on day 1 and the low-fat meal on day 2. Blood samples were drawn before (i.e., fasting) and after test meal consumption. All hormone analyses were performed by radioimmunoassay. Results indicated a significant rise in postprandial Prl levels for stage II patients, but not for stage I patients or the controls. Postprandial Gs levels were also elevated, whereas VIP levels were markedly reduced in patients versus controls; these differences were most marked in stage II patients. No significant intergroup differences were noted in postprandial levels of Nt and CCK. Hormone levels of patients and controls did not differ between the test meal situations, which indicated that some other component of the test meals might have been responsible for altered Prl and Gs levels. The differences observed between the stage I and II patients indicated that diet may influence the aggressiveness of tumor behavior and development through alterations in postprandial hormone release.

Characterization of the release of cholecystokinin from a murine neuroendocrine tumor cell line, STC-1.

The murine neuroendocrine cell line, STC-1, was found to contain 296.8 +/- 1.8 fmol of cholecystokinin-like immunoreactivity (CCK-LI) per mg cell protein. Immunocytochemical stain of STC-1 cells maintained in monolayer culture indicated that CCK-LI activity was present in 93% of the cells. Analysis by reverse-phase high-performance liquid chromatography indicated that STC-1 cells contained CCK-8 and an unidentified form as the predominant storage form. form. However, only CCK-8 was released into the culture medium upon stimulation by various secretagogues. The release of CCK-LI from STC-1 cells was stimulated by dibutyryl cAMP, forskolin, KCl, A23187, 4 beta-phorbol 12-myristate 13-acetate and luminal stimulants, e.g., sodium oleate, L-tryptophan, camostat and plaunotol. The release of CCK-LI from STC-1 cells was also stimulated by a neuropeptide, bombesin. The stimulatory effects of most of these agents were dose dependent. The stimulatory effects of dibutyryl cAMP, forskolin, and plaunotol were potentiated by 3-isobutyl-1-methyl xanthine, while that of camostat was not. The results obtained in this study indicate that the release of CCK from STC-1 cells shares the same characteristics of CCK release as from the CCK-secreting cells of the intestinal mucosa observed both in the dog and the rat in vitro and in vivo. Thus, the cellular mechanism of CCK release which appears to be cAMP- and Ca(2+)-dependent may be modulated by cellular protein kinase C activity. The STC-1 cell appears to be a suitable model for studying the mechanism of CCK release.

Bethanechol or cimetidine in the treatment of symptomatic reflux esophagitis: a double-blind control study.

We conducted a double-blind study to compare the effectiveness of oral bethanechol chloride or cimetidine in treating reflux esophagitis to evaluate the drugs' effects on the symptoms of esophagitis and its verification by endoscopy. Forty-three patients were treated with either 300 mg of cimetidine or 25 mg of bethanechol chloride, each administered four times a day for six weeks. In addition to this drug treatment, the patients all received conventional medical therapy. Patients who were treated with either of the two drugs experienced a decrease in symptoms and less severe endoscopic lesions. While cimetidine treatment resulted in complete endoscopic healing in 15 of 22 patients, bethanechol treatment resulted in the same healing in 11 of 21 patients. During therapy, neither endoscopic lesions or symptoms worsened. Our study indicated that either cimetidine or bethanechol is an effective drug in treating reflux esophagitis. The effects of the two drugs can be favorably compared.

A randomized controlled clinical trial of the serotonin type 3 receptor antagonist alosetron in women with diarrhea-predominant irritable bowel syndrome.

BACKGROUND: Irritable bowel syndrome (IBS) is a common gastrointestinal disorder seen in primary care practice. The symptoms of IBS, including abdominal pain, discomfort, and abnormal bowel function, may be modulated by activity of the serotonin type 3 receptor (5-HT(3)). The efficacy and tolerability of the 5-HT(3) receptor antagonist alosetron hydrochloride in nonconstipated female patients with IBS were evaluated in a double-blind, randomized, placebo-controlled trial. METHODS: Patients received either 1 mg of alosetron hydrochloride (n = 309) or placebo (n = 317) twice daily for 12 weeks, followed by a 4-week posttreatment period. Adequate relief of IBS pain and discomfort was the primary end point. Secondary end points included improvements in urgency, stool frequency, stool consistency, incomplete evacuation, and bloating. RESULTS: Seventy-one percent of patients were classified as having diarrhea-predominant IBS. Forty-three percent of alosetron-treated patients with diarrhea-predominant IBS reported adequate relief for all 3 months compared with 26% of placebo-treated patients (P<.001; percentage point difference = 17; 95% confidence interval, 8.0-25.4). Improvement with alosetron compared with placebo was observed by the end of the fourth week of treatment and persisted throughout the remainder of treatment. Alosetron significantly decreased urgency and stool frequency and caused firmer stools within 1 week of starting treatment. Effects were sustained throughout treatment and symptoms returned following treatment cessation. No significant improvement in the percentage of days with sense of incomplete evacuation or bloating was observed compared with placebo during the first month of treatment. Constipation was the most commonly reported adverse event. CONCLUSION: Alosetron hydrochloride, 1 mg twice daily for 12 weeks, is effective in relieving pain and some bowel-related symptoms in diarrhea-predominant female patients with IBS.

Secretion cells in the gastrointestinal tract.

The morphology and distribution of secretin (S) cells were investigated in the human and the dog. S cells were well-visualized by the indired immunofluorescence antibody technique, using a highly specific rabbit anti-secretin sera. The fluorescence reaction was not blocked by an excess amount of gastrin, cholecystokinin, glucagon, vasoactive intestinal polypeptide, or motilin, whereas secretin blocked the reaction. S cells were seen in the mucosa of the antrum and duodenum in both humans and dogs, and throughout the entire length of the canine small intestine. They were not found in the mucosa of the esophagus, fundus of the stomach, or rectum. These cells were either pyramidal in shape or pearshaped and were one-third of the size of gastrin cells. The possible significance of S-cell distribution in the antrum and small intestine is discussed.

Role of secretin in basal and fat-stimulated pancreatic secretion in conscious rats.

The role of endogenous secretin in basal and fat-stimulated pancreatic exocrine secretion was investigated in conscious rats. Rats were prepared with chronic fistulas draining bile and pancreatic juice, which was collected and returned to the duodenum at all times. Six days postoperative rats were fasted overnight, and pancreatic protein and fluid secretion were monitored for 3 h under basal conditions (0.15 M NaCl, intraduodenally) and during 2 h of intraduodenal infusion of a 20% triglyceride emulsion (Liposyn). Solutions were infused at 4.6 ml/h. Rats received a single bolus injection of 0.1 ml antisecretin serum or normal rabbit serum starting in the second hour of the basal period, and the effect on basal and fat-stimulated pancreatic protein and fluid secretion was determined. Antisecretin serum significantly inhibited basal interdigestive pancreatic protein and fluid secretion by 43% and 36%, respectively. Infusion of 20% fat emulsion stimulated a 2.1-fold increase in pancreatic protein and fluid secretion. The stimulation of both protein and fluid secretion was significantly inhibited by 60% by antisecretin serum. Plasma secretin after 2 h of fat infusion was 17.7 +/- 1.8 pM and was greatly reduced by the presence of secretin antiserum. The results support the hypothesis that secretin released by fatty acids is an important mediator of the pancreatic protein and fluid secretory response to dietary fat in the rat.

Secretin-like immunoreactivity and biological activity in the antral mucosa.

Using immunohistocytochemical techniques, secretin cells are again demonstrated in the antral mucosae of both dogs and rats. Secretin-like immunoreactivity was found in the crude extracts of antral mucosae in 15 dogs [1.18 +/- 0.48 (+/- SE) ng/g wet wt of mucosae], and a similar amount of SLI was also found in 82 rat antral mucosae. Upon ion exchange chromatography, the extracts of dog antral mucosae exhibited a predominant species eluted by the same salt concentration as porcine secretin. The rat antral mucosal extract also produced a chromatogram exhibiting the same predominant species on the ion exchanger. The main immunoreactive secretin peak, when gel filtrated on a Sephadex G-50 (superfine) column, produced an elution profile identical to that of standard natural porcine secretin. These results indicated that antral mucosae of both animal species contain an immunoreactive secretin-like material of the same charge and size as natural porcine secretin. Intravenous injection of a preparation of partially purified secretin from the extracts of canine antral mucosae resulted in a significant increase in the pancreatic flow in anesthetized rats. We conclude that a small number of secretin cells are, therefore, present in the antral mucosae of dog and rat, and this observation is supported by the presence of an immunologically and biologically active secretin-like molecule with charge and size similar to those of porcine secretin in the canine mucosal extracts.

Evidence for the existence of CCK-producing cells in rat pancreatic islets.

BACKGROUND: Although the existence of cholecystokinin-like immunoreactivity (CCK-LI) in rat pancreas had been reported previously, it was never clearly demonstrated whether CCK is produced in rat pancreatic islets. AIMS: The purpose of this study was to elucidate the source of the CCK-LI, the molecular properties of CCK, and the expression of the CCK gene in islet cells. METHODS: Immunohistochemical studies of rat pancreas were carried out with different rabbit antisera against CCK-8 and CCK-related peptide including N-terminal CCK-33 (1-22) and gastrin-17, and colocalization with known islet hormones including insulin, glucagon, somatostatin, and pancreatic polypeptide was investigated. The major molecular form of CCK in the islets was determined by HPLC. RT-PCR and in situ hybridization were performed to demonstrate the presence of the CCK transcript in the pancreas. RESULTS: CCK-LI was found in the center of the islets, colocalized with insulin in B cells. The major molecular form of CCK in the islets was CCK-8. A 350-nucleotide fragment of PCR-amplified CCK cDNA was detected in the islet as well as the duodenum by RT-PCR. In situ hybridization showed that CCK messenger RNA was located in a large portion of the islets, and this was consistent with the immunohistochemical findings. CONCLUSION: CCK messenger RNA and immunoreactivity are expressed in adult rat pancreatic islets, indicating that CCK-producing cells are present in adult rat islets.

Physiological doses of secretin do not stimulate acute insulin release.

To study the effect of physiological increments in plasma secretin concentrations on basal and glucose-stimulated insulin release, bolus iv glucose injections (5 g) were given to normal weight volunteers (less than 108% ideal BW) before, during, and 30 min after a secretin infusion at a rate of 0.125 U/kg.h, raising mean plasma immunoreactive secretin to 35.5 +/- 8.3 pg/ml. Acute insulin responses to glucose were unaffected during or after the secretin infusion. Furthermore, when plasma glucose was elevated to postprandial levels (128--165 mg/dl), a similar secretin infusion also failed to alter acute insulin responses. In addition, no changes in basal glucose or insulin levels were found when endogenous secretin concentrations were increased by intraduodenal acid infusion. Thus, increases in plasma secretin to concentrations seen in the postprandial state fail to alter acute insulin secretion. It is unlikely that secretin plays any role in the intestinal stimulation of insulin secretion.

Infliximab for patients with refractory ulcerative colitis.

Conventional treatment options for patients with severe corticosteroid-refractory ulcerative colitis (UC) include intravenous cyclosporine, which is frequently limited by toxicity, or colectomy. The efficacy of infliximab was investigated in the treatment of 16 patients with severely active UC refractory to conventional therapy; 7 of these patients were considered for colectomy pending medical failure. All patients received a single infusion of infliximab, 5 mg/kg; 6 of 16 patients (38%) received a second infusion approximately 5 months later. Efficacy was assessed by clinical response (defined as the lack of symptoms) as well as endoscopic and histologic outcomes. Clinical, endoscopic, and histologic improvement was observed in 14 of 16 patients (88%) after treatment with infliximab. Surgery was avoided in six of seven surgical candidates (86%). Clinical remission was maintained in 14 of 16 patients (88%) for > or = 4 months, and 4 of 16 patients (25%) for 7-10 months. Most of the treated patients were completely withdrawn from corticosteroid therapy. Treatment with infliximab induced endoscopic remission at 30 days and a significant improvement from baseline in mean histologic score (p < 0.001). In conclusion, infliximab improved clinical, endoscopic, and histologic outcomes in patients with severely active UC refractory to conventional therapy, allowing corticosteroid sparing and reducing the need for colectomy.

Effect of alcohol on plasma secretin concentration and pancreatic secretion in dogs.

We studied the effect of alcohol on plasma secretin concentration and pancreatic secretion in dogs with gastric cannulas and pancreatic fistulas. Intragastric administration of ethanol resulted in significant increases in pancreatic secretion of water and bicarbonate, which accompanied simultaneous increase in plasma secretin concentration. A more pronounced increase in pancreatic secretion occurred when ethanol was administered in the postprandial period. Integrated secretin release after a meal, 4.5 +/- 0.7 ng/ml-2 hr, had also increased significantly to 7.6 +/- 1.3 ng/ml-2 hr when ethanol was given in addition to a meal. Cimetidine pretreatment blocked the increase in plasma secretin concentration and pancreatic secretion in response to a meat meal and ethanol administration. This action of cimetidine paralleled its potent inhibitory effect on the exaggerated gastric acid secretion from Heidenhain pouches of dogs in response to ingestion of a meal and ethanol administration. Intraduodenal infusion of ethanol, however, failed to affect the plasma secretin concentration and pancreatic secretion. The present study indicates that alcohol stimulates release of endogenous secretin and pancreatic secretion by increasing duodenal acid load from the stomach stimulated by ethanol administration in dogs.

Role of secretin on postprandial gastrin release in the dog: a further study.

To determine the physiologic role of secretin on gastric secretory function, the effects of secretin in a physiologic dose on gastrin release and gastric acid secretion were studied in four dogs with vagally innervated fundic pouches. Three sets of experiments were performed in each dog: (1) meal alone, (2) meal after intravenous cimetidine to suppress acid secretion and release of secretin, and (3) meal after intravenous cimetidine with simultaneous intravenous secretin, 0.03 clinical unit (CU)/kg-hr. A significant increase in plasma secretin concentration occurred after ingestion of a meal. The postprandial increase in the secretin level was abolished by intravenous cimetidine 200 mg, whereas intravenous cimetidine resulted in a marked increase in the postprandial plasma gastrin concentration which was significantly greater than that after a meal alone. A plasma secretin level comparable to that of the postprandial period could be achieved in the same fasting dogs by intravenous secretin, 0.03 U/kg-hr in 2% dog albumin solution. The postprandial plasma gastrin concentration and acid secretion following intravenous cimetidine were significantly decreased by simultaneous intravenous secretin at a dose of 0.03 U/kg-hr. The observations indicate that secretin plays a significant role in the regulation of release of gastrin and gastric secretion of acid in the postprandial state in dogs.

Effects of SMS 201-995, a somatostatin analogue, on the exocrine pancreatic secretion and gut hormone release in dogs.

The effect of SMS 201-995, an analogue of somatostatin, on pancreatic exocrine secretion was investigated in both interdigestive and digestive states in dogs. In four dogs with gastric and Thomas duodenal cannulas, the pancreatic juice was collected by direct cannulation of the main pancreatic duct. SMS 201-995 was infused intravenously at doses of 0, 15, 30, 60, and 120 ng/kg/hr for 2 to 3 hours in the following experimental conditions: (1) interdigestive pancreatic secretion, (2) pancreatic secretion stimulated by the intravenous infusion of both secretin, 0.06 CU/kg/hr, and cholecystokinin octapeptide (CCK8), 0.03 microgram/kg/hr, and (3) pancreatic secretion after ingestion of a test meal. Pancreatic juice was analyzed for volume and outputs of bicarbonate and protein. Plasma levels of motilin, pancreatic polypeptide (PP), CCK, and secretin were determined by radioimmunoassay. SMS 201-995 inhibited significantly the pancreatic secretion and release of hormones, including secretin, CCK, PP, and motilin, in all three experimental conditions. The inhibitory action of SMS 201-995 on pancreatic secretion and hormone releases was dose dependent.

Pathophysiology and significance of malabsorption after Roux-en-Y reconstruction.

Extensive metabolic studies were conducted in five normal controls and in five study patients after total gastric resection with Roux-en-Y (RY) reconstruction to determine the nutritional consequences of this particular technique of restoring gastrointestinal continuity. Although malabsorption of fat (19.2 +/- 2.2%) and nitrogen (22 +/- 2.5%) demonstrated in the study patients was moderate, it was significantly greater than normal ( less than 0.01). In spite of the demonstrated malabsorption, however, positive nitrogen balances (+ 0.33 +/- 0.18 gm/day) were maintained in the RY patients throughout the investigative period. These observations suggest that malabsorption after RY is infrequently of clinical significance, even in this "worst-case" situation characterized by complete removal of gastric tissue. Malnutrition should occur in only those patients with more limited gastric resections and RY reconstruction who are unable to increase caloric intake to cover losses due to malabsorption. A significant decrease in both trypsin and lipase concentrations and a marked delay in secretion of these enzymes was noted in the RY patients in response to a test meal (p less than 0.01). Malabsorption of fat and nitrogen in RY patients improved after exogenous pancreatic enzymes, but not after administration of tetracycline. Bacterial overgrowth as a cause of postoperative malabsorption may be less important than previously thought. Malabsorption after RY is due primarily to maldigestion brought about by duodenal bypass which, in turn, results in either an absolute or a relative pancreatic enzyme insufficiency.

Release of distal gut peptide YY (PYY) by fat in proximal gut depends on CCK.

We tested the hypothesis that the release of PYY by fat confined to the proximal small intestine is dependent on CCK. Using a multi-fistulated model, plasma PYY levels were compared in 6 dogs after 60 mM oleate was perfused into the proximal one-half of the small intestine following i.v. administration of saline or devazepide, a CCK-A antagonist. Plasma PYY increased with fat (P < 0. 05), but plasma PYY level was lower following devazepide at 60 min and 90 min (P < 0.05). We conclude that CCK serves as a foregut signal linking fat in the proximal gut with the release of distal gut PYY.

Presence of biologically and immunologically active secretin-like substance in the mammalian brain.

The present study involves the isolation and characterization of secretin-like immunoreactivity from the brains of pigs, rats and dogs. Secretin-like immunoreactivity was extracted with 0.1 N HCl and subjected to SP-Sephadex ion exchange chromatography and gel filtration on a Sephadex G-50 superfine column. The average amounts of secretin-like immunoreactivity in the extracts of 2 pigs, 7 rats and 6 dog brains were 0.25 ng/g, 2.4 +/- 0.2 ng/g and 0.34 +/- 0.07 ng/g fresh tissue weight, respectively. The secretin-like immunoreactivities in the brain extracts exhibited the same retention coefficient as natural porcine secretin on gel filtration and were eluted in the same salt gradient from the SP-Sephadex column. A partially purified secretin-like immunoreactivity isolated from canine brain exhibited the same bioactivity as natural porcine secretin to stimulate pancreatic volume flow in anesthetized rats (n = 4). These results indicated that secretin-like immunoreactivities from brain extracts possess the same molecular size and charge as natural porcine secretin and the secretin-like immunoreactivity isolated from dog brain is active in stimulating pancreatic secretion in anesthetized rats.

Neurohormonal regulation of secretin secretion in canine duodenal mucosa in vitro.

We examined the effects of cholinergic, peptidergic and GABAergic agents on secretin secretion from canine duodenal mucosal explants incubated in organ culture media. Carbachol (10(-12) to 10(-4) M), atropine (10(-6) to 10(-4) M), hexamethonium (10(-6) to 10(-4) M), and somatostatin did not alter basal secretion of secretin. Somatostatin (10(-7) to 10(-8) M) inhibited secretin secretion stimulated by pH 4.5. Met, Leu and their D-ala2-analogs inhibited both basal and pH 4.5-stimulated secretin. Naloxone reversed the inhibition caused by met-enkephalin at pH 7.4. GABA (10(-9) to 10(-6) M) stimulated both basal and pH 4.5-stimulated secretin secretion. GABA-stimulated secretin secretion was neuronal in nature, bicuculline sensitive and was mediated via post ganglionic cholinergic neurons. GABA-stimulated secretin secretion was inhibited by both somatostatin and metenkephalin, suggesting that GABA-stimulated secretin secretion may be under the inhibitory control of peptidergic agents as well.

Motilin-antimotilin reaction reveals two antigenic determinants in motilin.

The reaction of motilin with four rabbit antimotilin sera raised by immunization with synthetic porcine motilin-bovine serum albumin conjugate was studied with respect to various binding parameters and specificity. All four antisera exhibited an extremely high degree of specificity and high affinity (K greater than 10(11) M-1) for porcine motilin. Studying the various synthetic motilin fragments. These antisera appeared to contain binding sites reacting strongly with the N-terminal sequence in which the first three amino acid residues are essential for high affinity binding. One of the antisera, R-3-6, appeared to contain approximately 20% of its binding sites with high affinity for C-terminus-containing fragments. These results suggest that motilin possesses two antigenic domains along its primary structure; one contains the N-terminal tripeptide and the other contains the C-terminal nanopeptide as essential parts. Thus, in addition to heterogeneity in affinity, a given antibody preparation may be heterogenous with respect to the specificity along the sequence of a peptide. Gel filtration studies of the methanol extracts of human and dog plasma indicated that an immunoreactive motilin-like material with a molecular size similar to natural porcine motilin was measured by our routine radioimmunoassay.