Somatostatin Receptor Subtype 2 Mediates Inhibition of Gastrin and Histamine Secretion From Human, Dog, and Rat Antrum

Transcription

1 GASTROENTEROLOGY 1996;111: Somatostatin Receptor Subtype 2 Mediates Inhibition of Gastrin and Histamine Secretion From Human, Dog, and Rat Antrum MUHAMMAD ZAKI,* LESLIE HARRINGTON,* ROBERT MCCUEN,* DAVID H. COY, AKIRA ARIMURA, and MITCHELL L. SCHUBERT* *Department of Medicine, Medical College of Virginia and McGuire Veterans Affairs Medical Center, Richmond, Virginia; Department of Medicine, Tulane University Medical Center, New Orleans, Louisiana; and Department of Medicine, Tulane University, Belle Chase, Louisiana Background & Aims: In gastric antrum, somatostatin exerts a tonic inhibitory influence on gastrin and histamine secretion. Five different subtypes of somatostatin receptors, designated sst 1 5, have been identified. sst 2,sst 3, and sst 5 subtypes have been localized to the stomach by molecular biological techniques. The aim of this study was to identify the sst subtype regulating gastrin and histamine secretion in human, dog, and rat stomach. Methods: Mucosal segments from human, dog, and rat antrum were superfused with various concentrations of somatostatin 14 and somatostatin analogues selective for sst 2,sst 3, and sst 5. Gastrin and histamine were mea- sured by radioimmunoassay. Results: Somatostatin 14 and the sst 2 agonist EC 5 20 inhibited gastrin and hista- mine secretion from all three species in a concentrationdependent manner, whereas the sst 3 and sst 5 agonists had no significant effect. Neutralization of endogenous somatostatin with the somatostatin antibody increased gastrin and histamine secretion. The increases were not affected by the gastrin antagonist but were abolished by the sst 2 agonist, implying that the inhibitory influence of ambient somatostatin is exerted directly on the hista- mine cell rather than indirectly via changes in gastrin secretion. Conclusions: Somatostatin inhibits gastrin and histamine secretion in human, dog, and rat antrum by activating sst 2 receptors on gastrin and histamine cells. I n the stomach, somatostatin cells are located in proximity to their target cells, which are parietal and entero- chromaffin-like (ECL) cells in the fundus and gastrin and histamine-containing cells in the antrum. 1 3 The func- tional correlate of this anatomic coupling is a restraint exerted by somatostatin on the secretion of histamine, gastrin, and acid Neutralization of somatostatin with somatostatin antibody augments histamine secretion from enriched canine fundic ECL cells in culture 9 as well as from superfused rat fundic mucosal segments, 11 indicating that endogenous somatostatin exerts a continuous inhibitory influence on histamine secretion in the fundus of the stomach. More recently, we have shown that superfusion of rat and human antral mucosal segments with somatostatin antibody augments histamine secretion, 3 indicating that endogenous somatostatin also exerts an inhibitory influence on histamine secretion in the antrum of the stomach. The exact cell type responsible for histamine secretion in the antrum has not been identified. In contrast to ECL cells of the fundus, histamine cells in the antrum seem to be insensitive to gastrin. 3 In rat, dog, and human antral muco- sal segments 3,10 as well as in isolated vascularly perfused rat 4,5 and pig stomach, 8 addition of somatostatin antibody augments gastrin secretion; this indicates that endogenous somatostatin exerts a continuous inhibitory influence on gastrin secretion. Five different subtypes of somatostatin receptors, designated sst 1 5, belonging to the superfamily of G-protein coupled receptors, have been identified by molecular cloning techniques. 12,13 sst 2, sst 3, and sst 5 (formerly 4) mes- senger RNA have been localized to the stomach by subtype- specific polymerase chain reaction and solution hybridiza- tion/nuclease protection analysis using sequence-specific complementary RNA probes Selective peptide analogues have been developed that display a high degree of selectivity for binding to sst 2, sst 3, and sst 5 ; no selective ligands are available to activate sst 1 and sst 4. 13,17,18 In the present study, we have used well-characterized selective so- matostatin analogues (EC 5 20 [sst 2 ], DC [sst 3 ], and DC [sst 5 ]) to identify the somatostatin subtype regulating gastrin and histamine secretion in the antrum of human, dog, and rat stomach. The results indicate that somatostatin inhibits gastrin and histamine secretion in all three species by activating sst 2 receptors present on these cells. Materials and Methods Materials Somatostatin 14 was obtained from Bachem Inc. (Tor- rance, CA). The selective sst agonists EC 5 20 (sst 2 selective; Abbreviations used in this paper: ECL, enterochromaffin-like; IC 50, median inhibitory concentration; sst, somatostatin receptor. This is a U.S. government work. There are no restrictions on its use /96/$0.00

2 920 ZAKI ET AL. GASTROENTEROLOGY Vol. 111, No. 4 equivalent to BIM 23060), DC (sst 3 selective; equiva- When several clusters of segments were obtained from a lent to BIM 23058), and DC (sst 5 selective; equivalent dog or human, the clusters were tested with different agents. to BIM 23052) were made by one of us (D.H.C.) and have One-milliliter samples of the effluent were obtained at 2 5- been characterized previously The somatostatin antiserum minute intervals and stored in 0.7-mL aliquots at 020 C for was raised by one of us (A.A.). 3,6 The gastrin antagonist PD subsequent measurement of gastrin and histamine by radioimmunoassay was a gift from Dr. John Hughes of Parke Davis (Cambridge, England). 22 Radioimmunoassay Superfusion of Antral Mucosal Segments Gastrin was measured in duplicate by radioimmunoassay The studies were performed on superfused mucosal seg- as described previously. 24 Gastrin antibody 1611 (final dilution, ments of human, dog, and rat antrum as previously described. 23 1:100,000) was a gift from Drs. J. H. Walsh and K. C. K. Lloyd In rat antrum, the muscle layer was stripped off the distal portion and provided by CURE/UCLA/DDC Antibody/RIA Core (Los of the stomach and a piece of mucosa, approximately 1 cm 2,was Angeles, CA). [ 125 I]gastrin was purchased from New England obtained from tissues adjacent to the pyloric junction. The cut Nuclear (Boston, MA). The limit of detection was 0.5 pmol/l edge of the remaining stomach was placed in formaldehyde for gastrin and the median inhibitory concentration (IC 50 ) was 11 { subsequent histological processing to verify the absence of pariintra-assay 3 pmol/l of sample (mean { SD; n Å 6 assays). Interassay and etal cells. The piece of mucosa (average weight, 75 { 5mg) coefficients of variation were 1% and 3%, respectively. was cut in 6 8 segments that were superfused together in the Histamine was measured in duplicate using a commercial ra- same chamber. A similar procedure was used in dog and human dioimmunoassay kit (Immunotech, Westbrook, ME). The kit antrum, except that several pieces (1 cm 2 ) could be obtained includes tubes coated with monoclonal antibody against acylated within 1 cm of the pyloric junction; these were cut into small histamine, acylating agent, and [ 125 I]histamine as a tracer. The segments, thus enabling superfusion of separate clusters of seg- limit of detection was 0.1 nmol/l histamine, and the IC 50 was 9 ments (average weight of each cluster, 281 { 17 mg and 244 { 2 nmol/l of sample (mean { SD; n Å 6 assays). Interassay and { 16 mg in dog and human antrum, respectively). Each cluster intra-assay coefficients of variation were 6% and 5%, respectively. was used for a different experimental procedure. Human tissue was obtained after gastrectomy for adenocarcinoma of the cardia Data Analysis (1 patient) or antrectomy for adenocarcinoma of the antrum (2 Gastrin and histamine secretion were expressed as the patients) and bleeding peptic ulcer (3 patients). mean increase or decrease from the basal level obtained during Before superfusion, the segments were washed with saline and the 5 minutes immediately preceding the experimental period. placed on a porous grid separating the two halves of a mini- Changes in secretion were tested for significance using Stuchamber (Swinnex 25; 1.4 ml volume; Millipore Corp., Bedford, dent s t test for unpaired values. All values are given as mean MA). Krebs bicarbonate solution containing 0.2% bovine serum { SE of n experiments on different animals. Concentrations albumin, 4% dextran, and 4.5 mmol/l glucose was perfused in eliciting 50% of maximal response (IC 50 ) were calculated using the bottom of the chamber at the rate of 1 ml/min and the linear regression analysis. effluent collected via a catheter leading from a small aperture at the top of the chamber. The perfusate was gassed with 95% O 2 Results and5%co 2. Drugs were delivered at the rate of 0.1 ml/min Basal Secretion of Gastrin and Histamine via a side arm close to the inlet. The entire preparation was From Rat, Dog, and Human Antral contained within a chamber maintained at 37 C. Mucosal Segments Experimental Design Mean initial basal gastrin and histamine secretion A 30-minute equilibration period was followed by an in rat, dog, and human antral mucosal segments are listed minute sampling period. The sampling period consisted in Table 1. Basal secretion was reproducible between of the following sequence: (1) a 30-minute control basal animals and reverted to initial control levels at the end period, (2) a 20-minute period during which somatostatin 14 of each experimental period. or various selective sst agonists were superfused, and (3) a final 30-minute control period. Effect of Somatostatin and Selective sst For experiments involving the somatostatin antibody, the se- Agonists on Gastrin Secretion From Rat, quence consisted of (1) an initial 30-minute control period; (2) Dog, and Human Antral Mucosal a 30-minute period during which somatostatin antibody (final Segments dilution, 1:200) was superfused; (3) a 20-minute period during which the somatostatin antibody in combination with the gastrin Superfusion of rat, dog, and human antral mucosal antagonist PD (1 mmol/l) was superfused; (4) a 20- segments for 20 minutes with somatostatin 14 or the minute period during which the somatostatin antibody, gastrin selective sst 2 agonist EC 5 20 in the range of 0.1 pmol/ antagonist, and the sst 2 agonist EC 5 20 (0.1 mmol/l) were L to 0.1 mmol/l caused a prompt, reversible, and concentra- superfused; and (5) a final 30-minute control period. tion-dependent decrease in gastrin secretion in all three

3 October 1996 SOMATOSTATIN RECEPTOR SUBTYPES IN ANTRUM 921 Table 1. Basal Secretion of Gastrin and Histamine From Rat, Dog, and Human Antrum Gastrin (fmol/min) Histamine (nmol/min) Initial Final Initial Final Rat 110 { { { 3 9 { 3 Dog 28 { 3 34 { 4 22 { 4 20 { 4 Human 18 { 4 20 { 4 7 { 1 8 { 1 NOTE. Basal secretion reverted to initial control levels at the end of the experimental period. species (Figures 1 and 2). The IC 50 values for somatostatin 14 and the sst 2 agonist, respectively, were as follows: rat, and ; dog, and ; and human, and Maximal inhibition of gastrin secretion, expressed as the integrated 20-minute response, elicited by somatostatin 14 and the sst 2 agonist, respectively, were as follows: rat, 52% { 11% and 60% { 5%; dog, 32% { 3% and 27% { 5%; and human, 37% { 5% and 39% { 2% below basal level (P õ 0.01; n Å 4 8). Neither the sst 3 nor the sst 5 agonist had any significant effect on gastrin secretion (Figure 1). The equipotent effect of somatostatin 14 and the sst 2 agonist suggests that the sst mediating inhibition of gastrin release in all three species is of the sst 2 subtype. Effect of Somatostatin and Selective sst Agonists on Histamine Secretion From Rat, Dog, and Human Antral Mucosal Segments Superfusion of rat, dog, and human antral mucosal segments for 20 minutes with somatostatin 14 or the selective sst 2 agonist EC 5 20 in the range of 0.1 pmol/l to 0.1 mmol/l caused a prompt, reversible, and concentrationdependent decrease in histamine secretion in all three species (Figures 2 and 3). The IC Figure 1. Effect of somatostatin 14 (SS-14), the sst 2 agonist EC 5 50 values for somatostatin 14 20, the sst 3 agonist DC 25 12, and the sst 5 agonist DC on and the sst 2 agonist, respectively, were as follows: rat, 2 gastrin secretion from mucosal segments of (A) rat, (B) dog, and and ;dog, and ; (C) human antrum. The agents were superfused for 20 minutes and and human, and Maximal inhibition samples were obtained at 5-minute intervals. The results are ex- Effect of Somatostatin Antibody on Histamine and Gastrin Secretion From Rat, Dog, and Human Antral Mucosal Segments To determine whether a regulatory pathway linked the secretion of histamine and gastrin, experi- ments were performed under conditions in which the effect of endogenous somatostatin was eliminated by su- perfusion with somatostatin antibody. Superfusion with somatostatin antibody (final dilution, 1:200) for 30 min- 48% { 4% and 44% { 5% below basal level (P õ 0.005; n Å 4 8). Neither the sst 3 nor sst 5 agonist had any significant effect on histamine secretion. The equipotent effect of somatostatin 14 and the sst 2 agonist suggests that the sst mediating inhibition of histamine release in all three species is of the sst 2 subtype. The decrease in histamine could represent a direct inhibitory effect of somatostatin or an indirect effect resulting from the decrease in gastrin secretion. of histamine secretion, expressed as the integrated 20-min- ute response, elicited by somatostatin 14 and the sst 2 agonist, respectively, were as follows: rat, 47% { 6% and 42% { 5%; dog, 56% { 4% and 43% { 4%; and human, pressed as the integrated response during the 20-minute period. Data are mean { SE of 4 8 experiments.

4 922 ZAKI ET AL. GASTROENTEROLOGY Vol. 111, No. 4 utes caused a significant increase in gastrin secretion above basal level from rat (32% { 4%; P õ 0.001; n Å 6), dog (63% { 8%; P õ 0.001; n Å 6), and human (64% { 5%; P õ 0.001; n Å 6) antral mucosal segments (Figure 4). The results support the notion that somatostatin exerts an inhibitory paracrine influence on gastrin secretion in various mammalian species. Somatostatin antibody also caused a significant increase in histamine secretion above basal level from rat (40% { 4%; P õ 0.001; n Å 6), dog (34% { 4%; P õ 0.001; n Å 6), and human (43% { 3%; P õ 0.001; n Å 6) antral segments (Figure 4). The increase in histamine secretion in all three species was not affected by concurrent superfusion with the gastrin antagonist PD (1 mmol/l) (Figure 4). In control studies, this concentration of antagonists abolished somatostatin secretion stimulated by gastrin 17 (1 mmol/l). The results imply that the increase in histamine secretion resulted from suppression of the inhibitory influence of somatostatin and was not affected by the concomitant increase in gastrin secretion. In further support of this notion, superfusion with the sst 2 agonist (0.1 mmol/l) inhibited histamine secretion in the presence of both the somatostatin antibody and the gastrin antagonist. Discussion This study shows that in the antrum of three mammalian species (human, dog, and rat), somatostatin Figure 2. Gastrin and histamine responses to superfusion of human antral mucosal segments for 20 minutes with somatostatin 14 (10 nmol/l) or the sst 2 agonist EC 5 20 (10 nmol/l). Dotted line indicates the level of basal secretion. Data are mean { SE of 6 8 experiments. Figure 3. Effect of somatostatin 14 (SS-14), the sst 2 agonist EC 5 20, the sst 3 agonist DC 25 12, and the sst 5 agonist DC on histamine secretion from mucosal segments of (A) rat, (B) dog, and (C) human antrum. The agents were superfused for 20 minutes and samples obtained at 5-minute intervals. The results are expressed as the integrated response during the 20-minute period. Data are mean { SE of 4 8 experiments.

5 October 1996 SOMATOSTATIN RECEPTOR SUBTYPES IN ANTRUM 923 Figure 4. Effect of somatostatin antibody (1:200 final dilution) alone and in combination with the gastrin antagonist PD (Gastrin Antag; 1 mmol/l) and the sst 2 agonist EC 5 20 (0.1 mmol/l) on gastrin and histamine secretion in mucosal segments of human antrum. Dotted line indicates the level of basal secretion. Data are mean { SE of six experiments. secretion because it was not affected by the gastrin antagonist PD The pattern implied that endogenous somatostatin exerts an inhibitory influence not only on the gastrin cell but also on the histamine cell. Consistent with this notion, the sst 2 agonist inhibited histamine secretion in the presence of the gastrin antagonist. In the antrum, somatostatin cells are located in proximity to gastrin cells. Immunoneutralization studies indicate that endogenous somatostatin exerts a paracrine restraint on gastrin secretion in rat, dog, pig, and human stomach. 3 5,8,10 The present study confirms those findings; superfusion with somatostatin antibody significantly increased basal gastrin secretion in rat, dog, and human antrum by 30% 64%. More recently, we have shown that immunoneutralization of somatostatin with somatostatin antibody also augments histamine secretion from rat and human antrum. 3 The present study confirms these findings and extends them to include dog antrum. The exact cell type responsible for histamine secretion in the antrum of these species has not been identified. Unlike fundic ECL cells, the histamine-containing cells in the antrum seem to be insensitive to gastrin. Five subtypes of ssts have been cloned; based on amino acid sequence and ligand binding characteristics, they can be classified into two general subfamilies: the sst 2, sst 3, and sst 5 comprising one family and the sst 1 and sst 4 comprising the other. 13 All five subtypes are coupled via G proteins to inhibition of adenylate cyclase. Using Northern blotting, sst 2 mrna has been localized to the fundus and antrum of the stomach. 25 Using a similar panel of sst subtype analogues as used in the present study, it has been shown, in the in vivo rat, that inhibition of pentagastrinstimulated acid secretion is mediated via activation of sst 2. 18,20 The IC 50 for inhibition of acid secretion by somatostatin and the sst 2 agonist were each about 1.5 nmol/ L. 18 Although the studies indicate that somatostatin inhibits acid secretion by binding to sst 2 receptors, the precise cellular location of the receptors was not examined. sst 2 receptors have been identified on rat fundic ECL cells 14 ; it is likely that the inhibitory effect of somatostatin on acid secretion is mediated in part via this cell type. inhibits gastrin and histamine secretion by activating sst 2 receptors on these cells. This conclusion is based on evidence that can be summarized as follows. (1) Superfusion of human, dog, and rat antral mucosal segments with somatostatin 14 and the sst 2 agonist EC 5 20 caused a concentration-dependent decrease in gas- In the antrum, three separate isoforms of the sst 2 receptor, trin and histamine secretion in all three species, whereas occurring through alternate messenger RNA splicing the sst 3 and sst 5 agonists had no significant effect. So- at the 3 end of the coding segment, have been identified: matostatin 14 and the sst 2 agonist were equally potent a 2.7-kilobase, 2.4-kilobase, and a newly recognized 1.6- and inhibited gastrin and histamine secretion in each of kilobase variant. 25 The 1.6-kilobase species predominates the species with IC 50 values in a range from in the antrum and is particularly enriched in the mucosal to mol/l. cell fraction containing endocrine cells. It is tempting (2) Neutralization of endogenous somatostatin by superfusion to speculate that this isoform may mediate the inhibitory with the somatostatin antibody increased gastrin effect of somatostatin on gastrin and histamine cells. The and histamine secretion in all three species; the increase sst 2 isoforms have similar ligand sensitivities and cannot in histamine was not attributable to the increase in gastrin be distinguished by pharmacological agents. 13

6 924 ZAKI ET AL. GASTROENTEROLOGY Vol. 111, No. 4 In the present study, somatostatin 14 and the selective Srikant CB. The somatostatin receptor family. Life Sci 1995;57: sst 2 analogue EC 5 20 equipotently inhibited gastrin 13. Reisine T. Somatostatin receptors. Am J Physiol 1995;269: and histamine secretion from the antral mucosa of hu- G813 G820. man, dog, and rat stomach. The inhibition was not afreceptor subtype on rat enterochromaffinlike cells. Gastroenterol- 14. Prinz C, Sachs G, Walsh JH, Coy DH, Wu SV. The somatostatin fected by concomitant administration of a gastrin antagoogy 1994;107: nist, implying that somatostatin activates sst 2 receptors 15. Raulf F, Pérez J, Hoyer D, Bruns C. Differential expression of present on gastrin and histamine cells. five somatostatin receptor subtypes, SSTR1-5, in the cns and peripheral tissue. Digestion 1994;55(suppl 3): Bruno JF, Xu Y, Song J, Berelowitz M. Tissue distribution of so- References matostatin receptor subtype messenger ribonucleic acid in the 1. Larsson L I, Goltermann N, DeMagistris L, Rehfeld JF, Schwartz rat. Endocrinology 1993;133: TW. Somatostatin cell processes as pathways for paracrine se- 17. Raynor K, Murphy WA, Coy DH, Taylor JE, Moreau J P, Yasuda cretion. Science 1979;205: K, Bell GI, Reisine T. Cloned somatostatin receptors: identifica- 2. Larsson L I, Houggaard DM. Evidence for paracrine somatostatfinity tion of subtype-selective peptides and demonstration of high afinergic regulation of gastrin gene expression by double-staining binding of linear peptides. Mol Pharmacol 1993;43:838 cytochemistry and quantitation. J Histochem Cytochem 1994; : Rossowski WJ, Gu Z F, Gu Z F, Akarca US, Jensen RT, Coy DH. 3. Vuyyuru L, Schubert ML, Harrington L, Arimura A, Makhlouf GM. Characterization of somatostatin receptor subtypes controlling Dual inhibitory pathways link antral somatostatin and histamine rat gastric acid and pancreatic amylase release. Peptides 1994; secretion in human, dog, and rat stomach. Gastroenterology 15: ;109: Rossowski WJ, Coy DH. Specific inhibition of rat pancreatic insu- 4. Saffouri B, Weir GC, Bitar KN, Makhlouf GM. Stimulation of gaslogs. lin or glucagon release by receptor-selective somatostatin anatrin secretion from the perfused rat stomach by somatostatin Biochem Biophys Res Commun 1994;205: antiserum. Life Sci 1979;25: Lloyd KCK, Wang J, Aurang K, Grönhed P, Coy DH, Walsh JH. 5. Short GM, Doyle W, Wolfe MM. Effect of antibodies to somatotion Activation of somatostatin receptor subtype 2 inhibits acid secrestatin on acid secretion and gastrin release by isolated perfused in rats. Am J Physiol 1995;268:G102 G106. rat stomach. Gastroenterology 1985;88: MartıBnez V, Coy DH, Lloyd KCK, Taché Y. Intracerebroventricular 6. Schubert ML, Edwards NF, Arimura A, Makhlouf GM. Paracrine injection of somatostatin sst 5 receptor agonist inhibits gastric regulation of gastric acid secretion by fundic somatostatin. Am acid secretion in rats. Eur J Pharmacol 1996;296: J Physiol 1987;252:G485 G Hayward NJ, Harding M, Lloyd SAC, McKnight AT, Hughes J, 7. Sandvik AK, Holst JJ, Waldum HL. The effect of gastrin-releasing Woodruff GN. The effect of CCK B /gastrin antagonists on stimupeptide on acid secretion and the release of gastrin, somatocol lated gastric acid secretion in the anaesthetized rat. Br J Pharma- statin, and histamine in the totally isolated, vascularly perfused 1991;104: rat stomach. Scand J Gastroenterol 1989;24: Schubert ML, Makhlouf GM. Identical patterns of somatostatin 8. Holst JJ, Jorgensen PN, Rasmussen TN, Schmidt P. Somatostatin secretion from isolated antrum and fundus of rat stomach. Am restraint of gastrin secretion in pigs revealed by monoclonal anti- J Physiol 1988;254:G20 G24. body immunoneutralization. Am J Physiol 1992;263:G Schubert ML, Bitar KN, Makhlouf GM. Regulation of gastrin and G912. somatostatin secretion by cholinergic and noncholinergic intra- 9. Chuang C N, Tanner M, Lloyd KCK, Wong H, Soll AH. Endoge- mural neurons. Am J Physiol 1982;243:G442 G447. nous somatostatin inhibits histamine release from canine gastric 25. Sandvik AK, Dimaline R, Brenna E, Waldum HL. Differential ex- mucosal cells in primary culture. Am J Physiol 1993;265:G521 pression and regulation of SSTR 2 messenger RNA in rat gastric G525. antrum and corpus. Am J Physiol 1995;269:G542 G Chiba T, Kinoshita Y. Interaction between D cells and G cells. In: Walsh JH, ed. Gastrin. New York: Raven, 1993: Received March 6, Accepted May 31, Schubert ML, Harrington L, Makhlouf GM. Reciprocal paracrine Address requests for reprints to: Mitchell L. Schubert, M.D., pathways link histamine and somatostatin secretion in the fundus of rat stomach (abstr). Gastroenterology 1993;104:A Patel YC, Greenwood MT, Panetta R, Demchyshyn L, Niznik H, McGuire Veterans Affairs Medical Center, Code 111N, Division of Gastroenterology, 1201 Broad Rock Boulevard, Richmond, Virginia Fax: (804)