Effect of Antibodies to Somatostatin on Acid Secretion and Gastrin Release by the Isolated Perfused Rat Stomach

Transcription

1 GASTROENTEROLOGY 1985;88:984-8 Effect of Antibodies to Somatostatin on Acid Secretion and Gastrin Release by the Isolated Perfused Rat Stomach GLENN M. SHORT, J. WILLIAM DOYLE, and M. MICHAEL WOLFE Department of Medicine, University of Florida College of Medicine, Gainesville, Florida The present studies were directed toward examining the effect of somatostatin on gastrin release and acid secretion by the isolated vascularly perfused rat stoma'ch. Rat stomachs were perfused in situ with Krebs-Ringer bicarbonate buffer containing 1% ovine erythrocytes and gassed with 9S% Oz-S% CO 2, Inclusion of pentagastrin in the perfusion buffer increased acid output from 2.2 ±.4 f-leq H+ Ih during control perfusion to 1 B.B ± 1.B f-leq H+ Ih (p <.1). In order to determine the effect of somatostatin on acid secretion and gastrin release, specific antibodies to somatostatin were included in the perfusate. Inclusion of antibodies to somatostatin in the buffer without pentagastrin did not significantly enhance acid output; however, gastrin concentration in the portal venous effluent increased from ls.l ± 2. to 2S.2 ± S.2 pg/ml at 4S min (p < O.OS). When antibodies to somatostatin were perfused in the presence of pentagastrin, acid output increased by 32% to 24.9 ± 1.2 f-leq H+ /h (p < O.OS); however, no change in gastrin concentration over basal was detected in the portal effluent. Results of these studies indicate that the capacity of the isolated rat stomach to secrete acid permits direct assessment of factors involved in the regulation of acid secretion. Under the conditions of these experiments, gastric somatostatin inhibits basal gastrin release and directly inhibits pentagastrin-stimulated acid secretion without affecting gastrin release. The interrelationships among gastrointestinal peptides involved in the regulation of gastric acid secretion are complex. The diversity of actions of these peptides and the vast array of chemical regulators that influence gastrin release and modulate gastric acid secretion in intact animals make it advantageous to investigate these interrelationships by means of direct in vitro techniques. Both antral culture of mucosal tissue (1,2) and the isolated perfused rat stomach (3-5) have been used to investigate gastrointestinal peptide interactions and their effects on gastrin release. However, neither method has provided a model for direct assessment of these peptides in the regulation of gastric acid secretion. The advantages of examining gastric acid secretion in an isolated canine stomach model have been previously demonstrated by Kowalewski and Scharf (5). The purpose of the present study was to examine the direct effects of the regulatory peptide somatostatin on gastrin rele&se and gastric acid secretion by the vascularly perfused rat stomach, using an isolated model with the capacity to secrete acid. Received June 5, Accepted October 29, Address requests for reprints to: M. Michael Wolfe, University of Florida College of Medicine, Box J-214, JHMHC, Gainesville, Florida Dr. Short is the recipient of National Research Service Award AM 7277 from the National Institutes of Health. Dr. Wolfe is a recipient of Clinical Investigator Award AM 147 from the National Institutes of Health and the AGA/Pharmacia Research Scholar Award. This work was presented, in part, at the Annual Meeting of the American Gastroenterological Association in New Orleans, Louisiana on May 23, The authors thank G. M. Reel for technical assistance and J. Wilhite for assistance in the preparation of this manuscript by the American Gastroenterological Association /8 5/$3.3 Methods and Materials Preparation of Antibodies to Somatostatin Cyclic tetradecapeptide somatostatin (15-28) (Beckman Instruments, Inc., Palo Alto, Calif.) was conjugated to bovine serum albumin (Sigma Chemical Co., St. Louis, Mo.), and rabbits were immunized with the somatostatin-bovine serum albumin conjugate, as previously described (7). Antibodies to somatostatin were identified in the sera of each of the immunized rabbits. Antibodies to Abbreviation used in this paper: KRB, Krebs-Ringer bicarbonate.

2 April 1985 EFFECT OF SOMATOSTATIN ON ACID SECRETION 985 somatostatin in one antiserum preparation (117-3) obtained after three immunizations were used in the studies described in this report. Antiserum is immunoreactive with somatostatin and somatostatin 28, the latter inhibiting antibody binding of radiolabeled somatostatin by -75% (at 4% inhibition of antibody binding) relative to the potency of somatostatin Scatchard analysis (8) of binding to somatostatin by antibodies to somatostatin demonstrated that the total binding capacity of undiluted antiserum was 2.86 J.Lg/ml; the Ka of the antibody preparation was 5.2 x 1 1 M- l. Control rabbit serum used in this study was obtained from two healthy fasting New Zealand white rabbits. Both control serum and somatostatin antiserum were filtered through.2-j.lm plain membrane Nalgene filter units (Nalge Co., Nalgene Labware Division, Rochester, N.Y.) and were stored in sterile containers at 4 C until used in the studies described below. Preparation of Animals Male Sprague-Dawley rats weighing g were fasted overnight with free access to water. They were then anesthetized with 6 mglkg i.p. sodium pentobarbital (Butler Co., Columbus, Ohio) and submitted to laparotomy; 1 U of heparin sodium (Elkins Sinn, Inc., Cherry Hill, N.J.) was injected subcutaneously 15 min before surgery. Isolation of the rat stomach was performed by the method of Lefebvre and Luyckx (9) and adapted for use in the rat, as previously described (1). The superior and inferior mesenteric arteries and veins and the suprarenal and renal arteries and veins were ligated. The hepatic and pancreaticoduodenal arteries were then ligated and the spleen removed. Care was taken to preserve the right gastroepiploic artery. Although most of the pancreas was excluded from perfusion by ligation of pancreatic vessels, a small portion was perfused due to the intimate association of the right gastroepiploic artery with the pancreas. Pancreatic hormone effects, therefore, cannot be entirely eliminated using this model. The abdominal aorta was dissected carefully from the inferior vena cava and ligated above the celiac axis. An 18-gauge Teflon catheter was inserted into the abdominal aorta, and its tip was positioned and secured above the level of the superior mesenteric artery. No more than 3-6 s were permitted to elapse between ligation of the aorta and initiation of gastric perfusion. The portal vein was then isolated and a PE-16 catheter was inserted to collect portal venous effluent. An anchor tie was then secured around the porta hepatis. A drainage tube was inserted into the rat stomach via a duodenal incision immediately distal to the pylorus and secured with a ligature. A 14-gauge Teflon catheter was inserted through the mouth and positioned at the gastroesophageal junction; care was taken to preserve the left gastric artery. The gastric lumen was perfused with distilled water, ph 7., maintained at 37 C, at a rate of 1 mllmin with a Harvard peristaltic pump (Harvard Apparatus, Millis, Mass.). Gastric perfusate was collected continuously through the pyloric drainage tube immediately before and at 1-min intervals after initiation of vascular perfusion. Arterial perfusion was performed, both with and without recirculation, at a rate of 2 mllmin with a Harvard peristaltic pump using Krebs-Ringer bicarbonate (KRB) buffer, ph 7.4, containing 1% bovine serum albumin, 5 mm pyruvate (Fisher Scientific, Pittsburgh, Pa.). and 1% (vol/vol) fresh, washed ovine erythrocytes. Krebs- Ringer bicarbonate buffer Was gassed continuously with 95% Oz-5% CO2 and maintained at 37 C throughout the duration of the experiments. Recirculation was accomplished by gassing the portal venous effluent with 95% O 2-5% CO 2 and reinfusing into the abdominal aorta using the Harvard peristaltic pump. The perfusion rate was kept constant by replacing the sample volume (1 ml) of portal venous effluent with fresh perfusate. Portal venous effluent was sampled 5 min before perfusion, immediately prior to initiation of perfusion, and 15, 3, 45, and 6 min after initiation of perfusion. Samples were collected in icechilled tubes for subsequent measurement of gastrin levels. Five rats were used in each of the studies described below. Experimental Design Control experiments Were performed using KRB containing normal rabbit serum at a final dilution of 1: 2 in the presence or absence of pentagastrin,.6 J.Lg/kg h (Ayerest Laboratories, Inc., New York, N.Y.). a dose previously examined and found to maximally stimulate gastric acid secretion using this in vitro model (1). Gastric acid secretion by the isolated perfused rat stomach in response to pentagastrin stimulation proceeds over the entire 6-min collection period, although slightly smaller amounts of H+ are secreted during the last 2 min of perfusion (1). The effects of somatostatin on gastric acid secretion and antral gastrin release were examined by inclusion of somatostatin antiserum (1: 2 dilution) in the perfusion buffer. Gastric perfusate was collected, as indicated above, and gastric acid output was determined by titration of samples with.1 N NaOH to ph 7., using a model 61A Fisher accumet ph meter (Fisher Scientific, Pittsburgh, Pa.). Radioimmunoassay Determination of Gastrin Concentrations in the Portal Venous Effluent Portal venous effluent samples were assayed in triplicate with antibodies to gastrin (56-2), which measure all known bioactive gastrin species, using a previously described method (11). Dilutions of assay reagents were prepared in.2 M sodium barbital buffer, ph 8.4, containing 2% normal human serum. All standards and samples were included as 1-J.LI aliquots in a total incubation volume of 2 ml in 12 x 75-mm glass tubes. To each standard and sample were added 1 J.LI of gastrin antiserum at a final dilution of 1 : 3,, and 1 J.Ll of synthetic 1251_human gastrin I (G-17) (- 4 cpm) prepared by modification of the chloramine-t method of Hunter and Greenwood (12,13). Serum gastrin concentrations were calculated from the calibration curves using

3 986 SHORT ET AL. GASTROENTEROLOGY Vol. 88, No. 4 synthetic human gastrin I (G-17) standard concentrations of 5-4 pg/ml. Sensitivity of the assay was 4 pg/ml (4 fg gastrin per incubation tube). Statistical Analysis Results are expressed as the mean ± SEM of data obtained from 5 rats examined tinder each condition. The Student's t-test for unpaired values was used to compare gastric acid secretion under the various conditions. The Student's t-test for paired values was used to compare gastrin concentrations in the portal venous effluent before and after initiation of arterial perfusion under the different conditions. Statistical significance was assigned if p <.5. Results Gastric Acid Secretion in Response to Pentagastrin and Antibodies to Somatostatin During arterial perfusion with KRB buffer in the absence of pentagastrin (control), the 6-min gastric acid output was 2.2 ±.4 jleq H+ Ih (Figure 1). When perfusion was performed with buffer containing somatostatin antibodies alone, a small, but significant, increase in gastric acid output over control was detected. However, inclusion of pentagastrin alone in the perfusate increased gastric acid output to 18.8 ± 1.8 p,eq H+ Ih (p <.1) (Figure 1). When both pentagastrin and antibodies to somatostatin were perfused simultaneously with recirculation, there was a 32.3% ± 2.2% increase in gastric acid output from 18.8 ± 1.8 p,eq H+ Ih (during :c +- :J: C' IaJ -=- I- ::J a.. I- ::J (3 <t Co.NDITlo.N Figure 1. Total gastric acid output (6 min) under control conditions (n = 5) and with perfusate containing somatostatin antibodies (anti-s, 1 : 2 dilution) (n = 5) or pentagastrin (.6 J.Lg/kg. h) (n = 5). Results expressed as microequivalents of H+ per hour, mean ± SEM. Pentagastrin.61-'9/kg/h 3 p<.5 _ 25 -i-.c... + :I: 2 <T "" -=- 15 I- ::l a.. I- ::l 1 Q U <! 5 + Penta9strlh Pento9astrin.6 }L9/ k9/h.6/49/k9/h Anti-S CONDITION Figure 2. Total gastric acid output (6 min) obtained with perfusate containing pentagastrin (.6 J.Lg/kg. h) (n = 5) in the absence or presence of somatostatin antibodies (anti-s, 1: 2 dilution) (n = 5). Results expressed as microequivalents of H+ per hour, mean ± SEM. perfusion with pentagastrin alone) to 24.9 ± 1.2 p,eq H+ Ih (p <.5) (Figure 2). The increment in gastric acid secretion in 'response to pentagastrin and somatostatin antisera perfusion without recirculation was nearly identical, increasing to 24.6 ± 1.6 p,eq H+ Ih (p <.5). Gastrin Release in Response to Pentagastrin and Antibodies to Somatostatin The mean basal gastrin level in the portal venous effluent before initiation of control perfusion was 1.1 ±.8 pg/ml. During arterial perfusion with KRB buffer alone (control), there was no significant change in the gastrin concentrations over basal levels (Figures 3 and 4). Inclusion of somatostatin antisera in the perfusate, however, increased gastrin concentration in the collected portal venous effluent by 66.8% ± 6.7% from 15.1 ± 2. to 25.2 ± 5.2 pg/ml at 45 min (p <.5) (Figure 4). When both pentagastrin and antibodies to somatostatin were perfused simultaneously, there was no significant increase in gastrin concentration in the portal venous effluent over basal levels (Figure 3). Discussion Somatostatin has been localized in several digestive organs, including the stomach, small bowel, biliary tree, and pancreas, where it possesses the capacity to modify basic physiologic functions, such as secretion, motility, and absorption (14-16). In addition to being found in several digestive organs, somatostatinlike immunoreactivity has also been demonstrated in the hypothalamus and retina (17,18), in close proximity to both pancreatic a-islet

4 April 1985 EFFECT OF SOMATOSTATIN ON ACID SECRETION 987 ~ 2 ~ i l:l 15 co:...j ~ 1 it: l V> ~ 5 ><: co: a. o -f- Control t If- p<.5 t Antl-S P'entoQcstrm P'entogostnn.6/L9/k9/h.6/L9/k9/h onti-s CONDITION Figure 3. Peak gastrin concentration in portal venous effluent under control conditions (n = 5) and in response to perfusion with somatostatin antibodies (anti-s, 1 : 2 dilution) alone (n = 5). pentagastrin (.6 JLg/kg h) alone (n = 5). or both pentagastrin (.6 JLg/kg' h) and somatostatin antibodies (anti-s, 1: 2 dilution) (n = 5). Results expressed as percent of basal concentration before initiation of arterial perfusion, mean ± SEM. and J3-islet cells (19). in the cardiac vagus (2). and in the glomerulus of the kidney (21). Due to its widespread location, somatostatin is a likely candidate for a local paracrine role in regulation of various endocrine and exocrine cells. Recent experiments performed in this (2) and other (14,22) laboratories have supported the hypothesis that antral somatostatin exerts inhibitory effects on gastrin release through paracrine pathways, accomplished by release of the peptide from somatostatin-containing cells into the immediate interstitial environment of the gastrin cell. Previous studies have suggested that somatostatin inhibits gastric acid secretion indirectly through inhibition of gastrin release and possibly through direct inhibition of parietal cells (23-27). In these studies, intravenous infusion of pharmacologic doses of somatostatin into humans (23), dogs (24,25), and rats (26) was shown to inhibit pentagastrinstimulated gastric acid secretion, suggesting a direct action on the parietal cell. Recently, Seal et al. (27) and Colturi et al. (28) demonstrated increases in venous plasma somatostatinlike immunoreactivity concentration in response to meals in dogs and humans, respectively. They also found that after intravenous infusion of somatostatin in doses sufficient to produce venous plasma somatostatinlike immunoreactivity levels within physiologic range, the acid response to the meals was significantly inhibited. The authors concluded that circulating somatostatin may play a physiologic role in suppressing gastric secretion in both dogs (27) and humans (28). In the present study only a small, and insignificant, increase in gastric acid output over control was detected when rat stomachs were perfused with buffer containing antibodies to somatostatin only (Figure 1). However, when pentagastrin and somatostatin antisera were perfused simultaneously both with and without recirculation, gastric acid output increased by ~32% (Figure 2), and no concomitant changes in portal venous effluent gastrin levels were observed (Figure 3). The reason for the failure of gastrin concentrations in portal venous effluent to increase is not entirely clear, but may be due, in part, to inhibition of gastrin release caused by a decrease in intraluminal ph (29). These results are in accord with the previously mentioned studies (23-28) and support the hypothesis that somatostatin directly inhibits acid secretion by the gastric parietal cell. However, although earlier studies suggested a physiologic role for circulating somatostatin in suppressing gastric acid output, the results of these studies, using the isolated perfused rat stomach to bind gastric somatostatin, suggest that somatostatin may inhibit acid secretion by paracrine pathways, accomplished by release of the peptide into the interstitial environment of the parietal cell. In the present study, arterial perfusion with specific antibodies to somatostatin only resulted in significant release of gastrin into the portal venous effluent. These observations are consistent with the findings of Saffouri et al. (22,3), but do not support those of Chiba et al. (14). who examined the effects of antisomatostatin -y-globulin on basal gastrin release using anesthetized rats, the isolated perfused rat stomach, and incubated rat antral mucosa. Although addition of somatostatin antibodies to incubated antral mucosa increased gastrin release, basal gastrin concentrations were not affected in intact animals or the perfused rat stomach (14). The reason for the different results obtained by Chiba et al. (14) is not clear, 3...J C I- a: ~ p<.5 o--<l Conl,ol Anli-S ~----4a----1t----~----~ III I I I I I I Ba.al 1 2 ~O 4 eo 6 TIME (min) Figure 4. Portal venous effluent gastrin concentrations under control conditions (n = 5) and in response to arterial perfusion with somatostatin antibodies (anti-s, 1: 2 dilution) (n = 5). Results expressed as picograms per milliliter. mean ± SEM.

5 988 SHORT ET AL. GASTROENTEROLOGY Vol. 88, No. 4 but, nevertheless, due to the widespread distribution of somatostatin-producing cells, and in light of previous experiments performed in this and other laboratories (2,14,16), it would appear that somatostatin exerts a local regulatory role on gastrin release. The capacity of the isolated vascularly perfused rat stomach to secrete acid, as demonstrated in these studies, permitted direct assessment of factors involved in the regulation of gastric acid secretion. Although somatostatin appears to have no effect on basal acid secretion in the rat, the modulation of gastric acid secretion by somatostatin appears to be mediated, at least in part, by pathways independent of its effect on gastrin release. Further investigation is required to define the complex interrelationships between somatostatin and other gastrointestinal regulatory peptides involved in modulation of gastric acid secretion. References 1. Wolfe MM, Reel GM, McGuigan JE. Inhibition of gastrin release by secretin is mediated by somatostatin in cultured rat antral mucosa. J Clin Invest 1983;72: Wolfe MM, Jain DK, Reel GM, McGuigan JE. Effects of carbachol on gastrin and somatostatin release in rat antral tissue culture. Gastroenterology 1984;87: Chiba T, Taminato T, Kadowaki S, et al. Effects of glucagon, secretin, and vasoactive intestinal polypeptide on gastric somatostatin and gastrin release from isolated perfused rat stomach. Gastroenterology 198;79: McIntosh CHS, Pederson RA, Koop H, Brown JC. Gastric inhibitory polypeptide-stimulated secretion of somatostatinlike immunoreactivity from the stomach: inhibition by acetylcholine or vagal stimulation. Can J Physiol Pharmacol 1981; 59: Martindale R, Kauffman GL, Levin S, Walsh JH, Yamada T. Differential regulation of gastrin and somatostatin secretion from isolated perfused rat stomachs. Gastroenterology 1982; 83: Kowalewski K, Scharf R. Influence of cholinergic stimulation on gastric secretory response to histamine in ex vivo isolated canine stomach. Am J Dig Dis 1972;17: McGuigan JE, Wolfe MM. Gastrin radioimmunoassay. Clin Chern 1982;28: Scatchard G. The attraction of proteins for small molecules and ions. Ann NY Acad Sci 1949;51: Lefebvre P, Luyckx A. Factors controlling gastric-glucagon. J Clin Invest 1977;59: Short GM, Wolfe MM, McGuigan JE. Pentagastrin-stimulated gastric acid secretion by the isolated perfused rat stomach. Life Sci 1984;34: Wolfe MM, McGuigan JE. Immunochemical characterization of gastrinlike and cholecystokininlike peptides released in dogs in response to a peptone meal. Gastroenterology 1984; 87: Hunter WM, Greenwood FC. Preparation of iodine-131 labelled human growth hormone of high specific activity. Nature 1962;194: Rehfeld JF, Stadil F, Rubin B. Production and evaluation of antibodies for the radioimmunoassay of gastrin. Scand J Clin Lab Invest 1972;3: Chiba T, Kadowaki S, Taminato T, et al. Effect of antisomatostatin y-globulin on gastrin release in rats. Gastroenterology 1981;81: Polak JM, Pearse AGE, Grimelius L, Bloom SR, Arimura A. Growth-hormone release-inhibiting hormone in gastrointestinal and pancreatic D cells. Lancet 1975;i: Dancygier H, Klein U, Leuschner U, Hubner K, Classen M. Somatostatin-containing cells in the extrahepatic biliary tract of humans. Gastroenterology 1984;86: Brazeau P, Vale W, Burgus R, et al. Hypothalamic peptide that inhibits the secretion of immunoreactive pituitary growth hormone. Science 1973;179: Yamada T, Marskak D, Basinger S, Walsh J, Morley J, Stell W. Somatostatin-like immunoreactivity in the retina. Proc Natl Acad Sci USA 198;77: Gerich JE, Charles MA, Grodsky GM. Regulation of pancreatic insulin and glucagon secretion. Annu Rev PhysioI1976: Campbell G, Gibbins IL, Morris JL, et al. Somatostatin is contained in and released from cholinergic nerves in the heart of the toad Bufo marinus. Neuroscience 1982;7: Kurokawa K, Aponte GW, Fujibayashi S, Yamada T. Somatostatin-like immunoreactivity in the glomerulus of rat kidney. Kidney Int 1983;24: Saffouri B, Weir G, Bitar K, Makhlouf G. Stimulation of gastrin secretion from the perfused rat stomach by somatostatin antiserum. Life Sci 1979;25 : Bloom SR, Mortimer CH, Thorner MO, et al. Inhibition of gastrin and gastric acid secretion by growth-hormone release inhibiting hormone. Lancet 1974;ii:ll Barros D'Sa AAJ, Bloom SR, Baron JH. Direct inhibition of gastric acid by growth-hormone release-inhibiting hormone in dogs. Lancet 1975;i: Konturek SJ, Tasler J, Cieszkowski M, Coy DH, Schally AV. Effect of growth hormone release-inhibiting hormone on gastric secretion, mucosal blood flow, and serum gastrin. Gastroenterology 1976;7: M6rz R, Prager-Petz J, Pointner H. Effect of luminal somatostatin on pentagastrin stimulated gastric acid secretion in the rat. Am J Physiol 1983;245:G Seal A, Yamada T, Debas H, et al. Somatostatin-14 and -28: clearance' and potency on gastric function in dogs. Am J Physiol 1982;243:G Colturi TJ, Unger RH, Feldman M. Role of circulating somatostatin in regulation of gastric acid secretion, gastrin release, and islet cell function. Studies in healthy subjects and duodenal ulcer patients. J Clin Invest 1984;74: Saffouri B, DuVal JW, Makhlouf GM. Stimulation of gastrin secretion by intraluminal chemicals: regulation by intraluminal cholinergic and noncholinergic neurons. Gastroenterology 1984;87: Saffouri B, Weir GC, Bitar KN, Makhlouf GM. Gastrin and somatostatin secretion by perfused rat stomach: functional linkage of antral peptides. Am J PhysioI198;238:G