The somatostatin receptor (SSTR) is a 7-transmembrane G protein–coupled receptor that acts via several cellular effector systems, including adenylate cyclase, K+ channels, and voltage-dependent Ca2+ channels. 1,2 The receptor is part of a 5-receptor family, and these receptors have been cloned, characterized, and designated SSTRs 1–5. 1 The 5 SSTR subtypes are encoded on separate chromosomes and genes. The gene location for the human SSTRs 1, 2, and 3 were located by fluorescence in situ hybridization to chromosomes 14, 17, and 22, respectively. 3 Human SSTRs 4 and 5 were mapped to chromosomes 20 4 and 28, 5 respectively.
The pancreatic B cell displays all 5 receptor subtypes; however, the actual role of each subtype in islet hormone secretion has not been clearly defined. With the expression of the 5 SSTR subtypes in stable cell lines, Rossowski and Coy 6,7 developed and characterized high-affinity SSTR analogues for SSTRs 2, 3, and 5. These receptor analogues have been used to examine the physiologic roles of SSTR subtypes in islet hormone secretion.
In an in vivo rat model, Rossowski and Coy 6 showed that the Coy SSTR 5 agonist was as effective as somatostatin in suppressing insulin secretion under high glucose conditions. In the same model, the Coy SSTR 2 agonist was much more specific for inhibiting glucagon secretion, when compared with somatostatin and other receptor agonists. 7 In another study using these same agonists in isolated mouse pancreatic islets, Fagan et al 8 demonstrated that glucose-stimulated insulin secretion was suppressed by the Coy SSTR 5 agonist, whereas the SSTR 2 agonist showed no effect on insulin secretion. In the isolated perfused human pancreas, the infusion of exogenous somatostatin has been shown to inhibit insulin secretion, under low or high glucose conditions, and this effect can be reversed by immunoneutralizing the exogenous somatostatin with a potent somatostatin monoclonal antibody. 9 In the isolated perfused human pancreas model, the Coy SSTR 2 agonist, at a single concentration, inhibited insulin secretion under low glucose conditions. 10 In the same study, the Coy SSTR 5 agonist demonstrated a trend toward inhibiting insulin secretion, but the differences were not significant.
In the current study, the isolated perfused human pancreas model was used to determine the effect of SSTR agonists on regulation of insulin secretion in the isolated perfused human pancreas. Three specific SSTR agonists and octreotide were infused into the pancreas and subsequent changes in insulin secretion were measured and analyzed. In contrast to previous studies, dose-response curves were generated for the SSTR 2 agonist and the SSTR 5 agonist to determine the maximal agonist effect. Additionally, the effects of the SSTR 2 and 5 agonists were examined with and without immunoneutralization of intrapancreatic somatostatin.
RESEARCH DESIGN AND METHODS
Pancreata were obtained from 23 cadaveric organs donors following brain death due to either subarachnoid hemorrhage or trauma. The organ donors ranged from 5 to 66 years in age. Seventeen were male and 6 were female. These donors had no history of pancreatic disease.
Perfusion Experimental Setting
The method and materials were similar to those of previous studies done on isolated perfused human pancreas. 9,11 During organ procurement, the pancreas was perfused with chilled Viaspan solution (University of Wisconsin organ preservation solution, Du Pont Pharmaceuticals, Wilmington, DE) and transported back to the laboratory on ice. The pancreas was attached to the perfusion apparatus, and the splenic artery was cannulated with a 14-gauge catheter and single-pass perfusion was performed. The pancreatic duct was cannulated with an 18-gauge catheter, and all the leaking vessels were ligated. The splenic vein was cannulated with silastic tubing, which had multiple side holes.
The perfusate was a Krebs bicarbonate buffer that contained 0.5% human albumin (Alpha Therapeutics, Torrance, CA), 4% T-70 Dextran (Sigma Chemicals, St. Louis, MO) and 3.9 mmol/L glucose. The pH of the buffer solution was 7.4, and it was oxygenated with a 95/5% oxygen/carbon dioxide mixture. The temperature was maintained at 37°C. The perfusion rate was kept constant at 4 mL/min, and the perfusion pressure was 30 ± 4 cm H2O.
The infusate was added to the perfusion solution via a sidearm infusion. The infusate was added at 0.1 mL/min (Harvard Pump, Harvard Apparatus, Boston, MA). Following a 60-minute equilibration period, randomly sequenced 10-minute stimulation periods followed by a 10-minute basal period were performed over a perfusion time of 188 ± 17 minutes. Aliquots of venous effluent were collected over a 2-minute period from the splenic vein catheter. The venous effluent was mixed with 500 kIU/mL Trasylol (FBA Pharmaceuticals, West Haven, CT) and then frozen at −20°C for subsequent radioimmunoassay. Immunoreactive insulin (IRI) levels in the splenic vein effluent were measured using a standard single antibody technique, as described previously. 12 Radioimmunoassays of the Krebs buffer with known concentrations of somatostatin antibody and antikeyhole limpet hemocyanin antibody were performed and demonstrated no interference by the infusate antibody with the insulin radioimmunoassay.
Antibody and Fab Fragments
In the study, 4 different infusions were used: (a) somatostatin monoclonal antibody (SAb) (John Walsh, CURE, VAMC, West Los Angeles, CA) is a high-affinity antibody that binds to the biologically active somatostatin ring, (b) somatostatin monoclonal antibody Fab fragment (SFAb), (c) antikeyhole limpet hemocyanin monoclonal antibody (KLH Ab) (John Walsh, CURE, VAMC, and (d) antikeyhole limpet hemocyanin monoclonal antibody Fab fragment (KLH Fab). The SFAb was prepared by papain digestion of the SAb and subsequent isolation of the Fab fragment using an Immunopure Fab Preparation Kit (Pierce, Rockford, IL). The SFAb was prepared as done in a previous study. 11 The KLH Ab and KLH Fab are nonspecific antibodies that are nearly identical to the SAb and SFAb in size. However, they differ at the hypervariable regions and served as a control infusate. The SAb, SFAb, KLH Ab, and KLH Fab were all prepared in phosphate-buffered saline, and phosphate-buffered saline controls were run. The final concentration of the SAb and the SFAb was 50 μg/mL. Intraislet somatostatin has never been directly calculated, but it has been estimated to be as high as 102 to 103 pM. 13 The concentration of the SAb and the SFAb was calculated to neutralize 100-fold excess of somatostatin.
Each pancreas was perfused with 3 SSTR-specific agonists and octreotide, in sequential order. The agonists were given in a single dose. The 3 receptor-specific agonists used in the study were DC 32-87 (binds SSTR 2), DC 25-12 (binds SSTR 3), and DC 32-92 (binds SSTR 5). The receptor-specific somatostatin agonists were provided by D. H. Coy (Tulane University, New Orleans, LA), and the octreotide was purchased commercially (Sandiz, Hanover, NJ). The 3 receptor-specific agonists and octreotide were prepared in 1% acetic acid in water. Final concentrations of 5, 25, and 50 ng/mL were used. These concentrations were chosen to approximate the anticipated physiologic intraislet somatostatin concentrations and 2 higher doses. 13
Statistical Analysis of the Data
Hormone levels in the splenic vein effluent were measured to assess the effect of the SSTR agonists, SAb, SFAb, KLH Ab, and KLH Fab. The insulin secretion in response to the test infusion is represented as the delta change (ΔX) from basal in picomoles. The basal value is the last value before pancreatic stimulation. This insulin secretion response represents the integrated insulin response seen during an infusion period. It is calculated as the weighted mean increase (or decrease) in insulin secretion above (or below) basal levels using the trapezoidal rule (area under the curve/10 minutes = Δ change). Since the baseline secretory rates vary between and within the pancreata, the basal secretory rate for each infusion is calculated based on the average basal insulin value obtained immediately preceding each infusion period. Insulin secretion due to infusate is also represented as the percentage of change from basal, which is calculated by dividing the Δ change from basal by the mean basal insulin level before infusion. The data are expressed as the standard error of the means. The influence of each infusate on the Δ change of hormone secretion from basal was analyzed by performing a paired Student t test. P < 0.056 was considered to be significant in all cases.
The study demonstrated that activation of the SSTR 2 causes suppression of insulin secretion. Furthermore, even during the immunoneutralization of endogenous intrapancreatic somatostatin, the SSTR 2 agonist reversed the effect of somatostatin immunoneutralization by suppressing insulin secretion, while the SSTR 5 agonist had no significant effect on insulin secretion. These results support a role for the SSTR 2 in the suppression of insulin secretion in the isolated perfused human pancreas.
Octreotide and the SSTR 2 Agonist Inhibit Insulin Secretion
To investigate the role of SSTR agonists on insulin secretion, 3 receptor specific agonists and octreotide were infused into isolated perfused human pancreata. The effects of the 3 SSTR agonists and octreotide are demonstrated in Figure 1. Octreotide infusion resulted in significant suppression of insulin secretion. The average suppression from basal secretion was 2855 ± 1093 pM (n = 18; P < 0.05). The effect of the SSTR 2 agonist (DC 32-87) was a decrease in insulin secretion of 1468 ± 480 pM (n = 17; P < 0.05). The average decrease of insulin secretion from basal during the SSTR 3 (DC 25-12) infusion was −174 ± 570 pM (n = 15; P = NS). The SSTR 5 agonist (DC 32-92) showed no significant suppression of insulin secretion. The average insulin suppression from basal was 1415 ± 1212 pM (n = 15; P = NS).
SSTR 2 Agonist Shows Significant Insulin Suppression at All Doses
To determine the maximal effect of the SSTR 2 and SSTR 5 agonist, dose-response curves were generated using the 2 agonists, and the data are presented as a percentage of inhibition of insulin secretion from baseline. The dose-response curves of the SSTR 2 and 5 agonists are shown in Figure 2. Infusion of the SSTR 2 agonist at increasing doses resulted in significant suppression of insulin secretion. The percentage of insulin inhibition from basal secretion was 23.2 ± 3.2% (5 ng/mL), 43.9 ± 9.3% (25 ng/mL), and 46.2 ± 3.8% (50 ng/mL) (P < 0.05 at all concentrations). The percentage of inhibition of insulin secretion with the SSTR 5 agonist was 8.2 ± 4.3% (5 ng/mL), 12.4 ± 11.4% (25 ng/mL), and 40.7 ± 9.6 (50 ng/mL). Significant insulin inhibition was only achieved at a dose of 50 ng/mL (P < 0.05).
SAb and Fab Fragment of SAb Increases Insulin Secretion
The SAb and the Fab fragment of the antibody were sequentially infused into the same pancreata to determine which infusate was more effective in immunoneutralizing intrapancreatic somatostatin. The KLH Ab and the Fab fragment of the KLH antibody were used as control infusates. The effect of the SAb and the SFAb infusates on insulin secretion is demonstrated in Figure 3. SAb infusion showed a significant increase in insulin secretion. Insulin secretion increased with the SAb infusion from 1449.0 to 2730.7 pM. This increase represented an 88% increase from baseline (n = 10; P < 0.001). The SFAb infusion caused an increase of insulin secretion from a baseline of 1644.2 pM to a mean of 2979.1 pM. The results represented an 81% increase in mean insulin secretion from basal levels (n = 12; P < 0.001). The KLH Ab and the KLH Fab were used to rule out the possibility of a nonspecific antibody effect on the B-cell secretion of insulin. The KLH Ab showed an increase in insulin secretion from 1202.4 to 1252.1 pM, which was a 4% increase (n = 7; P = NS). Infusion of the KLH Fab showed an increase in insulin secretion from 1598.3 to 1682.5 pM, which was a 5% increase in insulin secretion (n = 8; P = NS). Neither the KLH Ab nor the KLH Fab achieved a significant change in insulin secretion.
SSTR 2 Agonist Plus SAb Shows Suppression of Insulin Secretion
To confirm the effectiveness of the SSTR agonists in suppressing insulin secretion, the SSTR 2 and 5 agonists were infused into the isolated perfused human pancreas during immunoneutralization of intrapancreatic somatostatin, using the whole SAb (Fig. 4). Infusion of the SSTR 2 agonist with the SAb showed suppression of insulin secretion 1128 ± 457 pM (n = 7; P < 0.05), while infusion of the SSTR 5 agonist plus the SAb showed an increase in insulin secretion of 566 ± 852 pM (n = 5; P = NS). The SSTR 5 agonist plus SAb did not achieve statistical significance.
In the current study, octreotide, a synthetic peptide known to act on SSTRs 2, 3, and 5, was shown to inhibit insulin secretion in the isolated perfused human pancreas. An SSTR 2-specific agonist, but not an SSTR 3 or 5 agonist inhibited insulin secretion at physiologic doses. Dose-response curves demonstrated that at the maximal dose, the SSTR 2 agonist inhibited insulin secretion by 46%. The SSTR 2 agonist, but not the SSTR 5 agonist, overcame the effect of the immunoneutralization of intrapancreatic somatostatin. The results suggest that activation of SSTR 2 plays an important role in the suppression of insulin secretion in the isolated perfused human pancreas.
Somatostatin is a ubiquitous hormone, which acts as a physiologic inhibitor 14 through a 5-receptor family. Recent immunohistochemical studies have identified all 5 receptor subtypes within the human pancreatic islet cell. 15 The predominant receptor subtypes in the human pancreas are SSTRs 1, 2, and 5, while SSTRs 3 and 4 are poorly expressed. In the B cell, SSTR 1 is the predominant receptor; however, SSTR 2 is expressed in 47% of the B cells and SSTR 5 is expressed in 87%. The presence of 3 receptor types on the B cell leads to the question of what is the specific function of each receptor type.
Due to the development of receptor specific analogues for the SSTRs 2, 3, and 5, it has been possible to target these SSTRs with exogenous somatostatin analogues. This has allowed the determination that the biologic actions of somatostatin are receptor subtype specific. In this study, several lines of evidence support the role that activation of SSTR 2 plays a role in insulin secretion in the isolated perfused human pancreas. In the current study, 3 receptor-specific agonists and octreotide were infused at a set dose of 5 ng/mL, which was estimated to be the islet concentration of somatostatin. Octreotide is a synthetic octapeptide that is used clinically in the treatment of acromegaly, metastatic carcinoid, and vasoactive intestinal peptide–secreting tumors. 16 As a somatostatin analogue, it has high binding affinity for human SSTRs 2, 3, and 5. 17 In this study, octreotide significantly suppressed insulin secretion when infused into the isolated perfused human pancreas. This suggests that activation of 1 or more of these 3 receptors suppresses insulin secretion in the isolated perfused human pancreas.
When specific receptor agonists were used, the SSTR 2 agonist significantly suppressed insulin secretion from baseline. At the maximal SSTR 2 agonist dose (50 ng/mL), insulin secretion was suppressed 46% from baseline. The SSTR 3 agonist had no effect on insulin secretion, while the SSTR 5 agonist significantly suppressed insulin secretion only at pharmacologic doses (50 ng/mL). These results suggest an important role for SSTR 2 in the control of insulin secretion in the isolated perfused human pancreas.
To confirm the effectiveness of the SSTR agonists, the SSTR 2 and 5 agonists were infused into the isolated perfused human pancreas during immunoneutralization of endogenous intrapancreatic somatostatin. The benefit of the isolated perfused human pancreas model is that the pancreas is completely isolated from exogenous neural, hormonal, and nutrient influences; however, endogenous somatostatin is still present, most likely arising from the delta cell. In the isolated perfused human pancreas model under high and low glucose conditiond, Kleinman et al 9 infused somatostatin 14, a high-affinity SAb, or somatostatin 14 plus the SAb. In those studies, infusion of somatostatin 14, which binds to all 5 SSTR subtypes, resulted in inhibition of insulin secretion, while immunoneutralization of intrapancreatic somatostatin with the SAb significantly increased insulin secretion. The combined infusion of the SAb plus somatostatin 14 showed a reversal of the inhibitory effect of the exogenous somatostatin, demonstrating the effectiveness and specificity of the antibody. In another study, 11 the Fab fragment of the same SAb, somatostatin 14, or a combined infusion of the Fab fragment and somatostatin 14 were infused into the isolated perfused human pancreas under low glucose conditions. Exogenous somatostatin inhibited insulin secretion and the Fab fragment significantly increased insulin secretion by passive immunoneutralization of intrapancreatic somatostatin. The combined infusion of Fab fragment and somatostatin reversed the inhibitory effect of exogenous somatostatin, thereby showing that the Fab fragment is an effective tool in intrapancreatic somatostatin immunoneutralization. Both of these studies demonstrated that a high-affinity SAb and the Fab fragment of the antibody effectively immunoneutralize intrapancreatic somatostatin.
In the current study, a high-affinity SAb and the SFAb were infused in equimolar amounts sequentially into the same pancreata to assess whether the SAb or the SFAb would be more effective in immunoneutralizing intrapancreatic somatostatin. To rule out a nonspecific effect of the SAb and the SFAb fragment on insulin secretion, the KLH Ab and the KLH Fab were also used. The SAb fragment increased insulin secretion 88% from baseline while the SFAb increased insulin secretion 81% from baseline. Neither the KLH Ab or the KLH Fab showed any significant effect on insulin secretion, thereby making a nonspecific effect of the SAb and SFAb unlikely. Since the SAb and the SFAb had equal effectiveness in immunoneutralizing intrapancreatic somatostatin, the whole SAb was selected to serve as the tool for immunoneutralization of intrapancreatic somatostatin.
With an effective tool to immunoneutralize intrapancreatic somatostatin, the SSTR 2 and 5 agonists were infused alone or with the SAb. The combined infusion of the SSTR 2 agonist and the SAb showed a reversal of the antibody effect by suppressing insulin secretion approximately 1128 ± 457 pM (n = 7; P < 0.05) from baseline. The infusion of the SSTR 5 agonist plus SAb showed an increase in insulin secretion of 566 ± 852 pM (n = 5; P = NS). Even though the SSTR 5 agonist plus SAb showed a small increase in insulin secretion, the results were not significant. These results suggest that even though the SSTR 5 agonist demonstrates activity at pharmacologic doses, it probably does not play an important role in insulin secretion at physiologic levels. The results from the SSTR 2 agonist plus somatostatin antibody demonstrate that the activation of SSTR 2 is potent enough to completely reverse the effect of intrapancreatic somatostatin immunoneutralization.
In conclusion, the current study demonstrates that activation of SSTR 2 suppresses insulin secretion in the isolated perfused human pancreas in a dose-dependent manner. With immunoneutralization of endogenous somatostatin, the SSTR 2 agonist was still specific for inhibiting insulin secretion, while the SSTR 5 agonist had no significant effect on insulin secretion. These results suggest that SSTR 2 plays a role in regulation of insulin secretion from the isolated perfused human pancreas.
Further studies are still needed to improve our understanding of the mechanism by which somatostatin inhibits insulin secretion. Studies utilizing a SSTR 1–specific agonist are still needed as this is the predominant receptor subtype on the beta cell. In addition, further studies will be needed to determine whether the effect occurs directly at the beta cells or indirectly through the action of SSTR 2 stimulation on other endocrine cells. We did not assess the role of glucagon in the regulation of insulin secretion by the SSTR subtype selective agonists in the current study. One interesting finding in this study was that the maximal insulin suppression was 46% with the SSTR 2 agonist. This degree of insulin suppression correlates with the degree of expression of the SSTR 2 in the B-cell population (47%). This correlation between the degree of insulin suppression by the SSTR 2 agonist and the presence of SSTR 2 on the human B cell might suggest a direct effect of the SSTR 2 agonists on the beta cells.
1. Reisine T, Bell GI. Molecular properties of somatostatin
receptors. Neuroscience. 1995; 67:777–790.
2. Patel YC. Somatostatin
and its receptor family. Front Neuroendocrinol. 1999; 20:157–198.
3. Yamada Y, Stoffel M, Espinosa III, R et al. Human somatostatin receptor
genes: localization to human chromosomes 14, 17, and 22 and identification of simple tandem repeat polymorphisms. Genomics. 1993; 15:449–452.
4. Demchyshyn LL, Srikant CB, Sunahara RK, et al. Cloning and expression of a human somatostatin
-14-selective receptor variant (somatostatin receptor
4) located on chromosome 20. Mol Pharmacol. 1993; 43:894–901.
5. Panetta R, Greenwood MT, Warszynska A, et al. Molecular cloning, functional characterization, and chromosomal localization of a human somatostatin receptor
type 5) with preferential affinity for somatostatin
-28. Mol Pharmacol. 1994; 45:417–427.
6. Rossowski WJ, Coy DH. Potent inhibitory effects of a type four receptor-selective somatostatin
analog on rat insulin release. Biochem Biophys Res Commun. 1993; 197:366–371.
7. Rossowski WJ, Coy DH. Specific inhibition of rat pancreatic insulin or glucagon release by receptor-selective somatostatin
analogs. Biochem Biophys Res Commun. 1994; 205:341–346
8. Fagan SP, Azizzadeh A, Moldovan S, et al. Insulin secretion is inhibited by subtype five somatostatin receptor
in the mouse. Surgery. 1998; 124:254–259.
9. Kleinman R, Ohning G, Wong H, et al. Regulatory role of intraislet somatostatin
on insulin secretion in the isolated perfused human pancreas. Pancreas. 1994; 9:172–178.
10. Moldovan S, Atiya A, Adrian TE, et al. Somatostatin
inhibits B-cell secretion via a subtype-2 somatostatin receptor
in the isolated perfused human pancreas. J Surg Res. 1995; 59:85–90.
11. Kleinman R, Gingerich R, Wong H, et al. Use of the Fab fragment for immunoneutralization of somatostatin
in the isolated perfused human pancreas. Am J Surg. 1994; 167:114–119.
12. El Shami AS, Durham JI. More on vitamin B12 results as measured with “boil” and “no-boil” kits. Clin Chem. 1983; 29:2115–2116.
13. Samols E, Stagner JI. Intraislet and islet-acinar portal systems and their significance. In: Samols E, Stagner JI, eds. Comprehensive endocrinology: the endocrine pancreas. New York: Raven Press, 1991: 93–124.
14. Reichlin S. Somatostatin
. N Engl J Med. 1983; 309:1495–1501.
15. Kumar U, Ramakrishnan S, Suresh S, et al. Subtype-selective expression of the five somatostatin
receptors (hSSTR1-5) in human pancreatic islet cells: a quantitative double-label immunohistochemical analysis. Diabetes. 1999; 48:77–85.
16. Lamberts SWJ, Krenning EP, Reubi JC. The role of somatostatin
and its analogs in the diagnosis and treatment of tumors. Endocr Rev. 1991; 12:450–482.
17. Patel YC, Srikant CB. Subtype selectivity of peptide analogs for all five cloned human somatostatin
receptors (hsstr 1–5). Endocrinology. 1994; 135:2814–2817.