Article. Reference. Effect of insulin on glucose- and arginine-stimulated somatostatin secretion from the isolated perfused rat pancreas

Transcription

1 Article Effect of insulin on glucose- and arginine-stimulated somatostatin secretion from the isolated perfused rat pancreas GERBER, Pietro, et al. Abstract The effects of exogenous insulin on somatostatin secretion from the isolated perfused rat pancreas have been investigated in the presence of 5.6 mm glucose and when somatostatin secretion was stimulated by either glucose (16.7 mm) or arginine (20 mm). Insulin (15 mu/ml) significantly and rapidly suppressed glucose- and arginine-stimulated somatostatin release. However, at 5.6 mm glucose and, in the absence of other stimulators of somatostatin release, insulin had no effect on the somatostatin secretion rate. Reference GERBER, Pietro, et al. Effect of insulin on glucose- and arginine-stimulated somatostatin secretion from the isolated perfused rat pancreas. Endocrinology, 1981, vol. 109, no. 1, p DOI : /endo PMID : Available at: Disclaimer: layout of this document may differ from the published version.

2 /81/ $02.00/0 Endocrinology Copyright 1981 by The Endocrine Society Vol. 109, No. 1 Printed in U.S.A. I Effect of Insulin on Glucose- and Arginine-Stimulated Somatostatin Secretion from the Isolated Perfused Rat Pancreas* PIETRO P. G. GERBER,t ELISABETH R. TRIMBLE, CLAES B. WOLLHEIM, AND ALBERT E. RENOLD Institut de Biochimie Clinique, Universite de Geneve, Sentier de la Roseraie, 1211 Geneve 4, Switzerland ABSTRACT. The effects of exogenous insulin on somatostatin secretion from the isolated perfused rat pancreas have been investigated in the presence of 5.6 mm glucose and when somatostatin secretion was stimulated by either glucose (16.7 mm) or arginine (20 mm). Insulin (15 mu/ml) significantly and rapidly suppressed glucose- and arginine-stimulated somatostatin release. However, at 5.6 mm glucose and, in the absence of other stimulators of somatostatin release, insulin had no effect on the somatostatin secretion rate. (Endocrinology 109: 279, 1981) ; T HE PANCREATIC hormones, insulin, glucagon, and somatostatin, appear to play a role in controlling nutrient homeostasis (1). In addition to the effects of pancreatic hormones on target tissues, an intraislet action of the hormones is possible. Thus, glucagon is known to stimulate the release of insulin (2-4) and pancreatic somatostatin (5, 6), somatostatin to inhibit both insulin (7-9) and glucagon (10-13) secretion, and insulin to suppress glucagon secretion (14, 15). In these studies, higher concentrations of the hormones than those found in peripheral blood were employed. This may be necessary for the assessment of paracrine effects, since the concentrations of the hormones to which the pancreatic endocrine cells are exposed are not precisely known, but they are certainly higher than the circulating levels. The increased plasma concentrations of glucagon and somatostatin in dogs with experimental insulinopenic diabetes may suggest an inhibitory action of insulin on the pancreatic a-cells (16) similar to that exerted on the a cells (16, 17). However, all efforts to demonstrate a direct action of insulin on somatostatin secretion from mammalian pancreas preparations have so far given negative results (6, 18-20). In the present study, we investigated the effect of insulin on somatostatin release from the isolated perfused rat pancreas under basal conditions and during stimulation with glucose or arginine. Received May 6, This work was supported by Research Grants 3, SR and SR from the Swiss National Science Foundation. t To whom requests for reprints should be addressed. Materials and Methods The isolated pancreas was prepared from overnight fasted male Wistar rats, weighing g, by the method of Penhos et al. (21) with minor modifications. The perfusate was modified Krebs-Ringer bicarbonate buffer containing 1 mm Ca ++, 40 mg/ ml dextran, 2 mg/ml human serum albumin, and 400 kiu/ml aprotinin. The buffer was continuously gassed with a mixture of 95% 02 and 5% C02; the partial pressure of oxygen in the buffer entering the pancreas was 450 mm Hg, and the perfusion rate was 4 ml/min. Previous studies (22) have shown that steady basal concentrations of somatostatin are achieved in the effluent after the first 20 min of perfusion. Baseline samples were accordingly obtained 28, 29, and 30 min after starting the perfusion (designated -2, -1, and 0 min). The experimental variables were introduced at 0 min (i.e. 30 min after the perfusion had begun) according to protocols that are explained in the text. The somatostatin RIA used has been described in detail previously (22); its sensitivity was pg/ml. Statistical analysis was made by Student's t test. The following substances were used: Dextran T-40 (Pharmacia Fine Chemicals AB, Uppsala, Sweden); human serum albumin (Swiss Red Cross, Bern, Switzerland); aprotinin (Trasylol) kindly provided by Prof. G. L. Haberland (Bayer A. G., Wuppertal, West Germany); cyclic somatostatin and N-tyrosylated somatostatin for standard and iodinated tracer, respectively (Serono, Freiburg, West Germany); somatostatin antiserum, a gift from Dr. J. Ardill (Queen's University, Belfast, United Kingdom); and pork insulin (Actrapid; Novo Industri A/S, Copenhagen, Denmark). Results Throughout the experiments shown in Fig. 1, the glucose concentration was kept constant at 5.6 IllM. After 0 279

3 280 GERBER ET AL. Endo 1981 n = 6, x :t SEM n = 7, x ± SEM Vol109 No I SRIF (pg/min) GLUCOSE Insulin 5.6mM SRIF (pg/min) GLUCOSE 5.6mM ARGININE 20 mm Insulin minutes FIG. 1. Effect of exogenous insulin (15 mu /ml) on somatostatin (SRIF) secretion from the isolated perfused rat pancreas in the presence of 5.6 mm glucose. The addition of insulin (0-30 min) was without effect on the SRIF secretion rate. min of equilibration and the recording of three baseline somatostatin secretion rates (-2, -1, and 0 min), insulin was infused for 30 min (0-30 min) at a concentration of 15 mu/ml (0.63 ILg/ml). Sampling continued for 20 min after insulin was withdrawn (30-50 min). The somatostatin secretion rate did not vary significantly from the initial basal rate, either during the period of insulin infusion or subsequently. At this nonstimulatory glucose concentration, therefore, somatostatin secretion was not affected by insulin. The next experiments were designed to evaluate further whether insulin affects somatostatin secretion by adding insulin after first inducing stimulation of somatostatin secretion by glucose or arginine. The basal perfusate contained 5.6 mm glucose (-30 to 0 min). Somatostatin secretion was then stimulated, either by increasing glucose to 16.7 mm or by adding 20 mm arginine (in the presence of 5.6 mm glucose) during the period from 0-40 min. After 20 min of stimulation with either glucose or arginine, insulin was infused into a side-arm to give a final concentration of 15 mu/ml insulin in the perfusate during the second half of the stimulatory period (20-40 min) without modifying the concentration of any other constituent. As shown in Fig. 2, 16.7 mm glucose caused minutes FIG. 2. Effect of exogenous insulin (15 mu/ml) on glucose-stimulated somatostatin (SRIF) secretion from the isolated perfused pancreas. The glucose concentration was 5.6 mm from -30 to 0 min and 16.7 mm from 0 min to the end of the experiment. Insulin was added from min. SRIF secretion was significantly stimulated by 16.7 mm glucose (P < 0.02 at all time points, compared with the basal secretion rate). The effect of insulin on stimulated SRIF release was significant at 22, 26, 28, and 40 min (P < 0.05) and at 23, 24, 27, and 30 min (P < 0.025). a significant increase in the somatostatin secretion rate at all time points at which it was measured. The subsequent addition of insulin resulted in an inhibition of this glucose-induced somatostatin release. At 8 of 12 time points during the 20 min of insulin infusion, the somatostatin secretion rate was significantly less than the mean somatostatin secretion rate for the previous 20 min when 16.7 mm glucose was infused in the absence of insulin. Furthermore, the somatostatin secretion rate during insulin infusion with 16.7 mm glucose never significantly exceeded that during the initial baseline period when 5.6 mm glucose was infused alone. Arginine (20 mm) added to 5.6 mm glucose resulted in an increase in somatostatin secretion (Fig. 3). The further

4 EFFECT OF INSULIN ON SOMATOSTATIN 281 addition of insulin caused a rapid inhibition of somatostatin secretion, so that the secretion rate during insulin infusion was significantly less than the mean secretion rate during the previous 20 min of arginine infusion at all time points measured. In control experiments where 16.7 mm glucose or 20 mm arginine plus 5.6 mm glucose was infused for the period from 0-40 min (without the addition of insulin at min), the somatostatin secretion rate during the second 20-min period of stimulation (20-40 min) was similar to that during the first 20-min period (0-20 min; Table 1). In addition, the results of all the experiments have been expressed as integrated incremental somatostatin secretion for the period from 0-40 min (Table 2). In control experiments, when glucose was infused for 40 min without insulin, the integrated incremental somatostatin secretion from 0-20 min was similar to that for the subsequent min (Table 2A). By contrast, in the experiments where insulin was added during the period from min, the integrated incremental somatostatin secretion was markedly reduced compared to that during the 0-20 min period of the same experiment (P < 0.05, paired statistics) and compared to G 5.6 mm SRIF (pg/min) n = 6, x :t SEM GLUCOSE 16.7 mm Insulin r so minutes FIG. 3. Effect of exogenous insulin (15 mu/ml) on arginine-stimulated somatostatin (SRIF) secretion from the isolated perfused pancreas. Arginine stimulated SRIF secretion (P < at 1 min and P < 0.05 at 3, 5, 6, 7, 10 and 20 min, compared with the basal secretion rate). The effect of insulin on stimulated SRIF release was significant at all time points (P < 0.02). G, Glucose.,...; ;:;., 0 +I tl ;;;., ;:;., +I tl 00 a:> ;:; a:> 0 ;;; 0 +I +I., +I +I ;:!; ::! ;;; 0 +I +I «>., tl +I «> ::! 0 tl tl 0 «> tl tl ;;;., ;:; +I +I ;:; :::! +I +I., c :::!: :g., a:> +I +I 0 E ;:!; :::!: E== +I tl ;:; 18.,., +I +I g., g ;:;., +I +I +I +I ;::: +I tl :::! 1!5 «> 2 +I +I., ;s; a:> +I +I - a:> ::! 0 +I +I «> 7 tl +I a:> tl I +I <0 - - ':t'o'o:t II II. 5 0 :10 c :10 BE- E 0 < '

5 282 GERBER ET AL. Endo 1981 Vol109 No I TABLE 2. Integrated incremental somatostatin secretion during stimulation with 16.7 mm glucose or 20 mm arginine in the presence or absence of insulin during the second part of the stimulatory period (picograms per 20 min; mean ± SEM) 0-20 min p min A. Glucose (16.7 mm) without 2293 ± 353 NS 3318 ± 454 insulin p NS <0.001 B. Glucose (16.7 mm) with in ± 479 < ± 661 sulin (20-40 min) C. Arginine (20 mm) without 3708 ± 776 NS 2854 ± 497 insulin p NS <0.005 D. Arginine (20 mm) with insu ± 678 < ± 413 lin (20-40 mm) A + C, n = 4; B, n = 6; D, n = 7. that during the 20- to 40-min period of the control experiments carried out in the absence of insulin (P < 0.001, unpaired statistics). Similarly, when insulin was introduced from min in the arginine-stimulated pancreas, there was a marked reduction in the integrated incremental somatostatin secretion for this period compared with that during 0-20 min in the same experiment (P < 0.01) or that during min in control experiments (P < 0.005; see Table 2, C and D). Discussion The results presented here demonstrate that somatostatin secretion stimulated by either glucose or arginine can be inhibited by an increase in the insulin concentration surrounding the pancreatic A-cells. Basal somatostatin release (i.e. 5.6 mm glucose without added stimuli) was unaffected by insulin. This finding under basal conditions confirms previous work in the rat (18, 19) and dog (6). With respect to the inhibition by insulin of glucosestimulated somatostatin release, our results appear at variance with those of Hermansen et al. (20), who failed to observe any effect of 25 mu/ml insulin on somatostatin release. The reason for this difference is unclear; apart from the species difference, the composition of the perfusate was also different (20). In the isolated perfused chicken pancreas, Honey and Weir (23) observed a stimulation of somatostatin secretion by 20 mu/ml insulin. While this finding contrasts markedly with the results reported here, the interaction between hormones in the pancreas of birds may be different from that occurring in mammals. Indeed, in the duck, the infusion of somatostatin in vivo resulted in a stimulation of glucagon secretion rather than the expected inhibition (24). The use of high concentrations of pancreatic hormones is necessary when investigating their possible paracrine effects. When insulin secretion was stimulated by 16.7 mm glucose, the insulin concentration measured in the mixed effluent of the isolated perfused pancreas preparation was 1 mu/ml (22). This concentration can be increased by a factor of 10 under maximal stimulatory conditions. It appears reasonable, therefore, to use a concentration of insulin large enough to permit the A cells to recognize an increase. Suppression of pancreatic somatostatin secretion by insulin under physiological conditions may explain the findings of increased somatostatin secretion in experimental insulinopenic diabetes in dogs (16) and rats (25). In addition, insulin treatment of alloxan-diabetic dogs partially normalized the increased somatostatin and glucagon levels measured in peripheral blood (16). In acute in vivo experiments, insulin infused iv inhibited mealstimulated pancreatic somatostatin secretion; it also inhibited stimulated somatostatin release from the gastric antrum but not from the fundus (26). However, it is not possible to determine from in vivo studies whether insulin has a direct effect on the pancreatic A-cell. In in vitro experiments using the perfused rat stomach, 5.6 mm glucose, after a delay, caused a stimulation of somatostatin secretion, which could be prevented by simultaneous insulin infusion (27). In conclusion, the results presented here suggest that insulin can exert a direct inhibitory influence on somatostatin secretion under certain conditions. It would seem that each of the three principal hormones of the pancreatic islet, insulin, glucagon, and somatostatin, may affect the secretion of the other two within the pancreatic islet. The concentration of one of the hormones in the pancreatic effluent may thus depend on the secretory activity of the other hormone-secreting cells in the islets. With regard to pancreatic polypeptide, knowledge is as yet insufficient to show whether it may be similarly involved in reciprocal paracrine regulation of pancreatic hormone secretion. Acknowledgment The authors acknowledge with gratitude the expert technical assistance of Jarmila Slesinger. References 1. Unger RH, Dobbs RE, Orci L 1978 Insulin, glucagon and somatostatin secretion in the regulation of metabolism. Annu Rev Physiol 40: Grodsky GM, Bennett LL, Smith DF, Schmid FG 1967 Effect of pulse administration of glucose or glucagon on insulin secretion in vivo. Metabolism 16: Samols E, Harris G, Marks V 1965 Promotion of insulin secretion by glucagon. Lancet 2: Ryan WG, Nibbe AF, Schwartz TB 1967 Beta-cytotrophic effects of glucose, glucagon and tolbutamide in man. Lancet 1: Patton GS, Ipp E, Dobbs RE, Orci L, Vale W, Unger RH 1977 Pancreatic immunoreactive somatostatin release. Proc Natl Acad Sci USA 74: Weir GC, Samols E, Loo S, Patel YC, Gab bay KH 1979 Somatostatin and pancreatic polypeptide secretion. Effects of glucagon, in-

6 EFFECT OF INSULIN ON SOMATOSTATIN 283 sulin, and arginine. Diabetes 28:35 7. Curry DL, Bennett LL, Li CH 1974 Direct inhibition of insulin secretion by synthetic somatostatin. Biochem Biophys Res Commun 58: Turcot-Lemay L, Lemay A, Lacy PE 1975 Somatostatin inhibition of insulin release from freshly isolated and organ cultured rat islets of Langerhans in vitro. Biochem Biophys Res Commun 63: Wollheim CB, Kikuchi M, Renold AE, Sharp GWG 1977 Somatostatin and epinephrine-induced modifications of 45 Ca++ fluxes and insulin release in rat pancreatic islets maintained in tissue culture. J Clin Invest 60: Koerker DJ, Ruch W, Chideckel E, Palmer J, Goodner ChJ, Ensinck J, Gale ChC 1974 Somatostatin: hypothalamic inhibitor of the endocrine pancreas. Science 184: Gerich JE, Lovinger R, Grodsky GM 1975 Inhibition by somatostatin of glucagon and insulin release from the perfused rat pancreas in response to arginine, isoproterenol and theophylline: evidence for a preferential effect on glucagon secretion. Endocrinology 96: Samols E, Harrison J 1976 Remarkable potency of somatostatin as a glucagon suppressant. Metabolism [Suppl] 25: Wollheim CB, Blonde! B, Renold AE, Sharp GWG 1977 Somatostatin inhibition of pancreatic glucagon release from monolayer cultures and interactions with calcium. Endocrinology 101: Raskin P, Fujita Y, Unger RH 1975 Effect of insulin-glucose infusions on plasma glucagon levels in fasting diabetics and nondiabetics. J Clin Invest 56: Samols E, Harrison J 1976 Intra-islet negative insulin-glucagon feedback. Metabolism [Suppl] 25: Schusdziarra V, Dobbs RE, Harris V, Unger RH 1977 Immunoreactive somatostatin levels in plasma of normal and alloxan diabetic dogs. FEBS Lett 81: Muller WA, Faloona GR, Unger RH 1971 The effect of experimental insulin deficiency on glucagon secretion. J Clin Invest 50: Kadowaki S, Taminato T, Chiba T, Mori K, Abe H, Goto Y, Seino Y, Matsukura S, Nozawa M, Fujita T 1979 Somatostatin release from isolated perfused rat pancreas. Diabetes 28: Patel YC, Amherdt M, Orci L 1979 Somatostatin secretion from monolayer cultures of the neonatal rat pancreas. Endocrinology 104: Hermansen K, Orskov H, Christensen SE 1979 Streptozotocin diabetes: a glucoreceptor dysfunction affecting D-cells as well as B and A cells. Diabetologia 17: Penhos JC, Wu CH, Basabe JC, Lopez N, Wolff FW 1969 A rat pancreas-small gut preparation for the study of intestinal factor(s) and insulin release. Diabetes 18: Gerber PPG, Trimble ER, Wollheim CB, Renold AE, Miller RE 1981 Glucose and cyclic AMP as stimulators of somatostatin and insulin secretion from the isolated, perfused, rat pancreas: a quantitative study. Diabetes 30: Honey RN, Weir GC 1979 Insulin stimulates somatostatin and inhibits glucagon secretion from the perfused chicken pancreasduodenum. Life Sci 24: Strosser MT, Cohen L, Harvey S, Mialhe P 1980 Somatostatin stimulates glucagon secretion in ducks. Diabetologia 18: Hara M, Patton G, Gerich J 1979 Increased somatostatin release from the pancreases of alloxan diabetic rats perfused in vitro. Life ' 26. Rouiller D, Schusdziarra V, Unger RH 1979 Effect of insulin upon pancreatic and gastric release of somatostatin-like immunoreactivity. Diabetes (Abstract) 27. Chiba T, Taminato T, Kadowaki S, Abe H, Chihara K, Matsukura S, Goto Y, Seino Y, Fujita T 1980 Effects of insulin and pancreatic polypeptide on gastric somatostatin release. Diabetes 29:292 \ (