Characterization of covalently cross-linked somatostatin receptors in hamster beta cell insulinoma

Functional characterization of somatostatin receptors expressed on hamster glucagonoma cells

We characterized somatostatin receptors expressed in hamster glucagonoma INR1G9 cells and the effects of somatostatin on glucagon secretion, proglucagon gene expression, and the adenosine 3',5'-cyclic monophosphate (cAMP)-dependent signal-transduction cascade. 125I-labeled somatostatin was displaced by somatostatin-14 and somatostatin-28 with a dissociation constant of 2 nmol/l. Stable GTP analogues decreased binding of 125I-somatostatin to its receptors, suggesting an interaction of somatostatin receptors with G proteins. Chemical cross-linking of 125I-somatostatin to its receptor revealed a molecular mass of the ligand-receptor complex of 47 kDa. Somatostatin inhibited forskolin-stimulated activation of adenylate cyclase [2.5 microM forskolin (161%) + 1 microM somatostatin (128%); P < 0.05] and protein kinase A [10 microM forskolin (143%) + 1 microM somatostatin (114%); P < 0.05] but did not influence basal activities of these enzymes. Forskolin-induced stimulation of cAMP generation was reduced by somatostatin [2.5 microM forskolin (306%) + 1 microM somatostatin (145%); P < 0.05]. Somatostatin inhibited forskolin, theophylline, and arginine stimulation of glucagon secretion. Basal as well as forskolin-, theophylline-, and isobutyl methylxanthine-induced proglucagon gene expression was significantly reduced by somatostatin. Our data show that, in INR1G9 cells, somatostatin receptors are at least in part coupled to the adenylate cyclase system. Somatostatin is a potent negative regulator of both basal and forskolin-stimulated proglucagon gene expression. The interaction with forskolin occurs at the level of adenylate cyclase. The effect of somatostatin on basal proglucagon gene transcription is most probably mediated by an unrelated second messenger system. Somatostatin may influence several functions of the pancreatic A cell.

Tissue-specific distribution of cross-linked somatostatin receptor proteins in the rat

Pharmacological studies have suggested that the somatostatin (SS) receptor is heterogeneous and exhibits SS-14-and SS-28-selective subtypes. Whether such subtypes arise from molecular heterogeneity of the receptor protein has not been definitively established. Previous reports characterizing the molecular properties of the SS receptor by the cross-linking approach have yielded divergent size estimates ranging from 27 kDa to 200 kDa. In order to resolve this discrepancy, as well as to determine whether SS-14 and SS-28 interact with specific receptor proteins, we have cross-linked radioiodinated derivatives of [125I-Tyr11]SS-14 (T*-SS-14) and [Leu8,D-Trp22,125I-Tyr25]SS-28 (LTT*-SS-28) to membrane SS receptors in rat brain, pituitary, exocrine pancreas and adrenal cortex using a number of chemical and photoaffinity cross-linking agents. The labelled cross-linked receptor proteins were analysed by SDS/PAGE under reducing conditions followed by autoradiography. Our findings indicate that the pattern of specifically labelled cross-linked SS receptor proteins is sensitive to the concentration of chemical cross-linking agents such as disuccinimidyl suberate and dithiobis-(succinimidyl propionate). Labelled high-molecular-mass complexes of cross-linked receptor-ligand proteins were observed only when high concentrations of these cross-linkers were employed. Using optimized low concentrations of cross-linkers, however, two major labelled bands of 58 +/- 3 kDa and 27 +/- 2 kDa were detected. These two bands were identified as specifically labelled SS receptor proteins subsequent to cross-linking with a number of photoaffinity cross-linking agents as well. We demonstrate here that the 58 kDa protein is the major SS receptor protein in the rat pituitary, adrenal and exocrine pancreas, whereas the 27 kDa moiety represents the principal form in the brain. Additionally, the presence of a minor specifically labelled band of 32 kDa was detected uniquely in the brain, and a minor labelled protein of 42 kDa was observed in the pancreas. The labelling pattern obtained with LTT*-SS-28 was identical to that observed with T*-SS-14. Labelling of the 27 kDa band by either ligand was inhibited by SS-14 and SS-28 in a dose-dependent manner. Densitometric quantification showed that SS-14 exhibited greater than 2-fold greater potency than SS-28 for inhibiting the labelling of the 27 kDa species. These findings emphasize the need for careful interpretation of cross-linking data obtained for SS receptors, and provide evidence for molecular heterogeneity and for a tissue-specific distribution of the two principal SS receptor proteins.

Mechanism of action of somatostatin: an overview of receptor function and studies of the molecular characterization and purification of somatostatin receptor proteins

To determine whether somatostatin receptor subtypes arise from molecular heterogeneity of the receptor protein, we have cross-linked the putative receptor in normal rat tissues and in AtT-20 and GH3 cells, both chemically with SS-14, SS-28 and Tyr3 SMS ligands, as well as by photoaffinity labeling with an azido derivative of Tyr3 SMS (EE 581). Three prominent somatostatin receptor proteins of 58-kDa, 32-kDa, and 27-kDa size have been identified. These proteins exhibit a tissue-specific distribution, ligand selectivity, and relative preference for SS-14 and SS-28 binding, and thus qualify as somatostatin receptor subtypes. Using EE 581 as a photoaffinity probe, the 58-kDa and 32-kDa proteins have been purified to homogeneity from brain and AtT-20 cells by successive SDS-PAGE. The 58-kDa form has been trypsinized and amino acid sequence data obtained from four tryptic fragments. With the help of synthetic oligonucleotides derived from these sequences, work is currently in progress to clone the 58-kDa protein to elucidate its complete sequence, its expression, and its functional relationship to the somatostatin receptor and its pharmacological subtypes.

Pancreatic beta-cell somatostatin receptors

The characteristics of somatostatin (SRIF) receptors in rat pancreatic beta-cells were investigated using rat islets and the beta-cell line HIT-T15 (HIT). The biochemical properties of the SRIF receptors were examined with 125I-labeled des-Ala-1,Gly-2-desamino-Cys-3-[Tyr-11]- dicarba3,14-somatostatin (CGP 23996). 125I-CGP 23996 bound to SRIF receptors in HIT cells with high affinity and in a saturable manner. The binding of 125I-CGP 23996 to SRIF receptors was blocked by SRIF analogues with a rank order of potency of somatostatin 28 (SRIF-28) greater than D-Trp-8-somatostatin greater than somatostatin 14 (SRIF-14). To investigate the physical properties of the HIT cell SRIF receptor, the receptor was covalently labeled with 125I-CGP 23996 using photo-cross-linking techniques. 125I-CGP 23996 specifically labeled a protein of 55 kDa in HIT cell membranes. The size of the SRIF receptor in HIT cells is similar to the size of the SRIF receptor labeled with 125I-CGP 23996 in membranes of freshly isolated islets, suggesting that the physical properties of SRIF receptors in HIT cells and rat islet cells are similar. The binding studies suggest that beta-cells predominantly express a SRIF-28-preferring receptor. In freshly isolated islets, glucose- and arginine-stimulated insulin release was effectively blocked by SRIF-28 but not by SRIF-14. SRIF-14 did inhibit arginine-stimulated glucagon secretion from freshly isolated islets. The dissociation of the inhibitory effects of SRIF-28 and SRIF-14 on insulin and glucagon release from freshly isolated islets suggests that the two peptides act through different receptors in islets to regulate hormone secretion.

Solubilization of somatostatin receptors in hamster pancreatic beta cells. Characterization as a glycoprotein interacting with a GTP-binding protein

Somatostatin receptors of plasma membranes from beta cells of hamster insulinoma were covalently labelled with 125I-[Leu8,D-Trp22,Tyr25]somatostatin-28 (125I-somatostatin-28) and solubilized with the non-denaturing detergent Triton X-100. Analysis by SDS/PAGE and autoradiography revealed three specific 125I-somatostatin-28 receptor complexes with similar molecular masses (228 kDa, 128 kDa and 45 kDa) to those previously identified [Cotroneo, P., Marie, J.-C. & Rosselin, G. (1988) Eur. J. Biochem. 174, 219-224]. The major labelled complex (128 kDa) was adsorbed to a wheat-germ-agglutinin agarose column and eluted by N-acetylglucosamine. Also, the binding of 125I-somatostatin-28 to plasma membranes was specifically inhibited by the GTP analog, guanosine-5'-O-(3-thiotriphosphate) (GTP[S]) in a dose-dependent manner. Furthermore, when somatostatin-28 receptors were solubilized by Triton X-100 as a reversible complex with 125I-somatostatin-28, GTP[S] specifically dissociated the bound ligand to a larger extent from the soluble receptors than from the plasma-membrane-embedded receptors, the radioactivity remaining bound after 15 min at 37 degrees C being 30% and 83% respectively. After pertussis-toxin-induced [32P]ADP-ribosylation of pancreatic membranes, a 41-kDa [32P]ADP-ribose-labelled inhibitory guanine nucleotide binding protein coeluted with the 128-kDa and 45-kDa receptor complexes. The labelling of both receptor proteins was sensitive to GTP[S]. The labelling of the 228-kDa band was inconsistent. These results support the conclusion that beta cell somatostatin receptors can be solubilized as proteins of 128 kDa and 45 kDa. The major labeled species corresponds to the 128-kDa band and is a glycoprotein. The pancreatic membrane contains a 41-kDa GTP-binding protein that can complex with somatostatin receptors.