Biochemical properties of somatostatin receptors

Biochemical

Properties

of Somatostatin

Receptors

Terry Reisine, Hai-Tao He, Stephanie Rens-Domiano, Jean-Michel Martin, Karen Raynor, Steven Borisiow, and Kyriaki Thermos Somatostatin (SW) induces its biological actions by binding to and stimulating membrane-associated receptors. To investigate the molecular mechanisms by which SRIF induces its biological effects, we have characterized the biochemical properties of SRIF receptors. SRIF receptors can ba solubilized in an active form with the detergant CHAPS and can be detected with the highaffinity SRIF analog [1asl] MK 678. The pharmacological characteristics of solubilized SRIF receptors from brain are similar to the receptors in membranes, suggesting that the solubilized receptors retain their biological activity. Solubilired SRIF receptors appear to be tightly associated with GTP-binding proteins, since analogs of GTP can greatly reduce agonist labeling of the solubilized SRIF receptor. The solubilized SRIF receptor migrates as a mass of approximately 400 kd and is a glycoprotein since it can specifically interact with lectin columns. The solubilization of the SRIF receptor has allowed for its purification by affinity chromatography. The purified SRlF receptor migrates as a mass of 60 kd in denaturing gels. Using aftinity ch~a~~phy, the receptor can be purified to near homogeneity. Present studies are directed toward sequencing and cloning cDNA encoding the SRIF receptor in order to further characterize its physical properties and expression. 0 1990 by W. 8. Saunders Company.

OMATOSTATIN-14 (SRIF- 14) is a multipu~o~ peptide that can act as a neurotransmitter or hormone. SRIF- 14 is unevenly distributed in the central nervous system (CNS) and is present in some peripheral organs such as the pancreas and gut.’ In the brain, SRIF- 14 is known to regulate the firing activity of neurons, as well as the release of catecholamines and serotonin.1,2 A role for SRIF- 14 in regulating basal ganglia function has been suggested and the peptide may be involved in the expression of motor behaviors.3 The SRIF-14 present in limbic areas may also be involved in memory acquisition and learning. The most well-established neuroendocrine action of SRIF- 14 is its regulation of anterior pituitary function.4 SRIF- 14 is the most potent and effective natural inhibitor of growth hormone and thyroid-stimulating hormone secretion from the pituitary. In more peripheral endocrine organs such as the pancreatic islets, SRIF-14 produced in the 6 cells is an effective inhibitor of glucagon and insulin secretion from the cyand p cells, respectively. SRIF- 14 induces its biological actions by interacting with and stimulating membmne-bound receptors. Receptors for SRIF- I4 are coupled to different cellular effector systems via pertussis toxin-sensitive guanine nucleotide binding regulatory proteins (G proteins).‘.’ G proteins couple SRIF- 14 receptors to the catalytic subunit of adenylyl cyclase and SRIF14 has been shown in a number of tissues to inhibit adenylyl cyclase activity. In addition to blocking CAMP formation, SRIF- 14 also inhibits Ca++ influx into pituitary cells.8.9 The inhibition of Ca++ influx may be a critical event in SRIF14’s attenuation of hormone release from the pituitary. The effects of SRIF-14 on Ca+’ influx are inde~ndent of its actions on adenylyi cyclase activity and mediated by pertussis

S

From the Department ~~Pharmaeoit~gy, University ?~Penns.vi~~ania School ~~Medic~ne~ Pbi~ade~phia, P.4. Supported by National Insiitutes ofHealth Grants No. GM 34781 and MH 45533. and by the Ojice qf Naval Research (NOOOl4-88-K0048)‘). Address reprint requests to Terry Reisine. PhD, Department of Pharmacolog?~, University ~rpennsylvania School of Medicines 36th St and Hamiiton Walk, Philade~hia~ F% 19104. 0 1990 by W.B. Saunders Company. 00260495/90/3909-2020$03.00/O 70

toxin-~nsitive G proteins. *,’ SRIF- 14 can inhibit Ca+’ influx through multiple mechanisms. Several investigators have shown that SRIF-14 can reduce Ca++ currents in endocrine cells.“.’ ’ This effect is mediated by pertussis toxin-sensitive G proteins. lo SRIF-14 can also enhance K+ currents in pituitary cells, as well as neurons.“-t4 The stimulation of K+ conductance can hyperpolarize the ceil membrane thereby decreasing voltage-dependent Ca++ currents. Thus, SRIF- 14 receptors have the potential to couple to multiple cellular effector systems via G proteins. The properties of SRIF- 14 receptors have been examined using several approaches. A number of investigators have characterized SRIF- 14 receptors in SRIF- I4 responsive tissues with ligand binding techniques.’ The results of several studies have suggested that subtypes of SRIF-14 receptors are expressed in the brain and pituita~.‘-‘5 These subtypes have differing affinities for SRIF- 14 and its precursor somatostatin28 (SRIF-28). The results of functional studies have supported the hypothesis that distinct SRIF- 14 and SRIF-28 receptors exist.‘6.‘7 In this regard, SRIF-14 is more potent than SRIF28 in blocking glucagon secretion from pancreatic ty cells while SRIF-28 is more potent than SRIF-14 in inhibiting insulin secretion from fi cells. Autoradiographic studies of Amherdt et ali8 further suggest that N and @cells predominantly express SRIF-14 receptors and SRIF-28 receptors, respectively. In addition to the islet cells, the tumor cell lines AtT-20 and GH3, derived from the anterior pituitary, express SRIF-28 and SRIF-14 receptors, respectively.” Subtypes of SRIF-14 receptors are also expressed in the brain. and have similar affinities for SRIF- 14 and SRIF-28. Both Reubi2’ and Tran et al” reported that the octapeptide SMS-201-995 was able to distinguish subtypes of SRIF- I4 receptors in the brain that express similar affinities for SRIF14 and SRIF-28. Recently, Raynor and Reisine2’ showed that the SRIF- 14 analogs, CGP 23996” and MK 1j78’~bind selectively to distinct populations of SRIF- 14 receptors in the brain. These subpopulations of receptors have the same affinities for SRIF-I4 and SRIF-28. Interestingly, the MK 678sensitive receptors in the corpus striatum of the rat brain do not appear to be coupled to adenylyl cyclase. since MK 678 did not affect adenylyl cyclase activity in this brain region. This suggests that some subtypes of SRIF- I4 receptors may be coupled to different cellular effector systems, in much the Metabolism,

Vol39.

No 9, SuppI

(September). 1990: pp X-73

SOMATOSTATIN RECEPTORS

same manner that subtypes of muscarinic cholinergic or adrenergic receptors couple to different cellular effector systems. In an effort to investigate the bi~hemical properties of SRIF- 14 receptors, several groups of investigators have covalently tagged SRIF-14 receptors with radiolabeled SRIF14 analogs using crosslinking techniques and analyzed the characteristics of the receptors by gel electrophoresis and autoradiography. Williams et al*‘-*’ covalently labeled SRIF14 receptors from pancreatic acinar and pituitary membranes with [‘251]Tyr’-SRIF-14 and reported that the size of the receptor ranged between 94 and 88 kd. The labeled material in the sodium dodecylsulfate (SDS)-gels appeared as a smear after autoradiography. This was proposed to be due to the presence of oligosaccharides associated with the receptor. This hypothesis was supported by the finding that the labeled receptor specifically bound to lectin columns.26 Srikant and Pate128identified three different [‘*‘I] Tyr”-SRIF-14 labeled proteins from pancreatic acinar membranes. They suggested that these results support the hy~thesis that molecularly distinct subtypes of SRIF- 14 receptors are expressed in the pancreas. Thermos and Reisine*’ reported the labeling of anterior pituitary, AtT-20 cell and GHJ cell SRIF-14 receptors with the SRIF- 14 analog [r2’1]CGP 23996. The size ofthe receptors labeled in these tissues differed from those of Williams’ group25-27or Srikant and Patel.” A sharp protein band of 55 kd was specifically labeled in the pituitary cells. Interestingly, while AtT-20 and GHr cells appear to express pha~a~o1o~c distinct subtypes of SRIF-28 and SRIF- 14 receptors, respectively, o&y minor physicat differences appeared to exist between these subtypes of SRIF receptors in the two cell lines. Thermos et alz9 recently characterized the SRIF- 14 receptors of rat brain using [“‘I] CGP 23996 and photocrosslinking techniques. A sharp protein band of 55 to 60 kd was labeled in the brain. The labeled brain SRIF- 14 receptor was slightly acidic in charge as assessed by two~mensional polyacrylamide gel electrophoresis (PAGE) and appears to be a glycoprotein, since it could specifically bind to wheat germ aggiutinin (WGA). The apparent heterogeneity in size of the labeled SRIF-14 receptors in the various studies described above could be the result of differences in the techniques used, ligands used, or due to the expression of SRIF receptors of different size. Presently, however, no consensus exists concerning the size of SRIF- 14 receptors. In an attempt to further elucidate the biochemical properties of SRIF- 14 receptors, we30*3* have solubilized and purified the SRIF- I4 receptor from brain. A description of some of our findings is presented below. SOLUBILIZATION OF SRIF-14 RECEPTORS

solubilize SRIF- 14 receptors, rat brains were used as a source of receptors since SRIF-14 receptors are expressed in relatively high density in this tissue and the pharmacological characteristics of the receptors have been extensively examined.’ Solubilized SRIF-14 receptors were labeled with the cyclohexapeptide SRIF-14 agonist, [‘2JI] MK 678. [‘251]MK 678 binding to brain membrane-associated SRIF- k4 receptors is of high affinity, saturable, and selective. The characteristics of [ “‘11 MK 678 binding to rat brain SRIF-14 receptors have been described eBewhere.22 SRIF-14 receptors were solubiTo

71

lized with the detergent CHAPS. Twenty percent to 30% of membrane bound SRIF- 14 receptors were solubilized in the presence of 10 mmol/L CHAPS. Specific, D-Trp’-SRIF-14 displaceable [‘*‘I] MK 678 binding to the solubiiized SRIF14 receptors represented approximately 70% of total [‘25I] MK 678 binding. [1251]MK 678 binding to the solubilized SRIF-14 receptor reached equilibrium by 90 minutes. The half-life of dissociation of [12sI] MK 678 binding to the solubilized receptor was approximately 60 minutes. [ ‘251]MK 678 binding to the soiubilized SRIF-14 receptor was of high affinity and saturable. Analysis of MK 678 displacement of [“‘I] MK 678 binding to the solubilized SRIF-$4 receptors revealed two [r251]MK 678 binding sites. The high-affinity binding sites had Ku and B,, values of 1.1 nmol/L and 452 fmol/mg protein, respectively, while the low-affinity sites had values of 35 nmol/L and 1,875 fmollmg prot, respectively. [125I] MK 678 binding to brain membranes could also be resolved into two sites with similar characteristics as the binding to the ~lubiliz~ SRIF-14 receptor. (**‘I] MK 678 binding to the solubilized SRIF-14 receptor was potently inhibited by D-Trp8-SRIF- 14, SRIF- 14, and SRIF-28. The inactive peptide SRIF-28’T’4did not affect [‘2sI] MK 678 binding to the solubilized SRIF- 14 receptor nor did a number of biologically active peptides unrelated to SRIF-14. The pharmacological characteristics of [‘251]MK 678 binding to the solubilized and membrane-Lund SRIF-14 receptors were similar, su~esting that SRIF-14 receptors were maint~ned in an active state following solubilization. [r2sI] MK 678 binding to the solubilized and membrane bound SRIF-14 receptors could be affected by the presence of monovalent cations. Na’ ions reduced [‘*‘I] MK 678 binding to the SRIF-14 receptor. Similarly, Na+ ions have been shown to diminish the affinity of other receptors negatively coupled to adenylyl cyclase for agonists3’ This has been suggested to be due to Na” ions intemcting directly with inhibitory receptors to uncouple the receptors from G proteins. SRIF-14 receptors are coupled to G proteins and dissociation of the receptor from G proteins is known to reduce the affinity of the SRIF-14 receptors for agonists. The solubilized SRIF- 14 receptor appears to retain its coupling to G proteins. This is suggested by the ability of the GTP analog GTPyS to abolish [‘251]MK 678 binding to the solubilized SRIF- 14 receptors. To establish whether pert&s toxin-sensitive G proteins are coupled to the solubilized SRIF- 14 receptor, studies were carried out in the tumor cell line AtT20, which expresses a high density of SRIF receptors that are linked to pertussis toxin-sensitive G proteins.33,34 SRIF receptors could be solubilized from AtT-20 cell membrane and those receptors could be specifically labeled with [r2sI] MK 678. Pretreatment of AtT-20 cells with pertussis toxin for 24 hours uncouples membmne bound SRIF- 14 receptors from G proteins and eliminates specific [12?] MK 678 binding to membrane-bound SRIF-14 receptors.33 No specific [‘?I MK 678 binding to solubilized SRIF- 14 receptors from pertussis toxin-treated AtT-20 cells could be detected. This finding suggests that solubilized SRIF-14 receptors are coupled to pertussis toxin-sensitive G proteins. The coupling of SRIF14 receptors to G proteins most likely allows the receptors to be detected by [‘*‘I] MK 678, otherwise the afBnity of the

72

REISINE ET AL

receptors for agonists would be too low for equilibrium [ ‘25I] MK 678 binding to be measured. Consistent with the findings of the photocrosslinking studies 26,29the solubilized SRIF-14 receptor is a glycoprotein. The solubilized SRIF- 14 receptor could bind to a WGA column and specific [“‘I] MK 678 binding activity could be eluted from the column with the sugar N,N’,N”-triactylchitotriose. To determine the size of the solubilized SRIF- 14 receptor from brain, the solubilized receptor was subjected to gel filtration chromatography. Most of the solubilized SRIF- 14 receptor binding activity migrated at approximately 400 kDa. The large size of the solubilized SRIF-14 receptor may be due to its coupling with other proteins. G proteins are one component of the 400-kDa complex. Specific [‘25I] MK 678 binding to the 400-kd material was abolished by GTPyS. Analysis of the 400-kd material by SDS-PAGE and the detection of a-subunits of G proteins by Western Blot, using an antibody that cross-reacts with all a-subunits of the G proteins, revealed the presence of a substantial amount of (Ysubunit corn&rating with specific [‘25I] MK 678 binding activity. This further indicates that the 400-kDa material represents SRIF-14 receptors coupled to G proteins. PURIFICATION

OF THE SRIF-14

RECEPTOR

The ability to solubilize the SRIF- 14 receptor in an active form offered us the possibility to purify the receptor. The SRIF-14 receptor was purified to near homogeneity using affinity chromatography.30 A SRIF- 14 affinity column was constructed by covalently coupling D-Trp8-SRIF- 14 to Affigel 10. Solubilized receptor, in the presence of protease inhibitors and phospholipid, was applied to the column at 4°C to minimize degradation of the receptor and peptides. The solubilized receptor remained in contact with the column overnight and then the column was extensively washed until protein was no longer detected in the eluate. The receptor was then eluted from the column with a buffer containing 50 mmol/L sodium acetate. This buffer was slightly acidic (pH 5.5). In previous studies, it was observed that the affinity of SRIF- 14 receptors for agonists was highly sensitive to pH and that at pH 5.5, agonist labeling of the receptor was abolished. Thus, the slight acidity of the elution buffer would be expected to promote the dissociation of the receptor from the column. Furthermore, Nat ions greatly decrease the affinity of the SRIF- 14 receptor for agonists and therefore would be expected to elute the receptor from the column.35 The material eluted from the affinity column with the sodium

acetate buffer was subjected to SDS-PAGE and the proteins detected by silver stain. A single major protein band of 60 kd was detected in the eluate. This material appeared to be specifically eluted from the SRIF affinity column since application of solubilized brain proteins to an Affi-gel 10 column treated in the same manner as the SRIF-14 affinity column except that no SRIF-14 was coupled to the Al&gel 10. did not yield a 60-kd protein following elution with the sodium acetate buffer. The 60-kd material could also be eluted from the SRIF-14 affinity column with buffer containing D-Trp*SRIF- 14. Furthermore, in preliminary studies, GTPyS could elute the 60-kd protein from the SRIF-14 affinity column consistent with the finding that the solubilized SRIF-14 receptor must be coupled to G proteins in order to be in a high enough affinity state to bind to the SRIF- 14 immobilized to the Affi-gel 10 in the affinity column. To further establish that the material eluted from the SRIF14 affinity column is the SRIF- 14 receptor, it was shown that SRIF- 14 and the SRIF- 14 analog SMS-20 l-995 could block the binding of the 60-kd material to the affinity column while the inactive SRIF analog SRIF-28 [l-14] did not prevent the receptor from interacting with the column. GTPyS also blocked the 60-kd material from binding to the affinity column. This is most likely due to its uncoupling the receptor from G proteins, which would result in a reduction in the affinity of the receptor for SRIF-14. Finally, it was possible to covalently label the purified 60-kd material with the SRIF14 analog [‘25I] CGP 23996 using photocross-linking techniques. Excess SRIF- 14 could completely block the binding of [“‘I] CGP 23996 to the 60 kd material while SRIF-28 [l141 did not affect the binding. Thus, the elution of the 60kd material from the affinity column with sodium acetate buffer or SRIF-14, the selective blockade of the binding of the 60-kd material to the affinity column by SRIF- 14 and its analogs, and the specific labeling of the 60-kd protein with the SRIF-14 analog [‘25I] CGP 23996 suggests that the material eluted from the affinity column is the SRIF- 14 receptor. Presently, we are further characterizing the properties of the purified SRIF- 14 receptor. To establish that the purified receptor is functionally active, attempts are underway to reconstitute the receptor and purified G proteins into phospholipid vesicles. Activity of the receptor will be measured by its ability to stimulate GTPase activity and the specific binding of [‘251]MK 678. In addition, we are presently sequencing the SRIF-14 receptor. With the sequence information, we intend to raise antibodies against the receptor and generate oligonucleotides that will lx used to probe cDNA libraries for clones expressing the receptor.

REFERENCES

I.

Epelbaum J: Somatostatin

in the central

nervous system: Phys-

iology and pathological modifications. Prog Neurobiol 27:63-100, 1986 2. Cbesselet MF, Reisine T: Somatostatin regulates dopamine release in rat striatal slices and cat caudate nucleus. J Neurosci 3:232236, 1983 3. Martin-lversen M, Radke J, Vincent S: The effects of cysteamine on dopamine-mediated behaviors: Evidence for dopamine-somatostatin interactions in the striatum. Pharrnacol Biochem Behav 24: 1707-1714, 1986 4. Brazeau P, Vale W, Burgus R. et al: Hypothalamic polypeptide

that inhibits the secretion of immunoreactive pituitary mone. Science 129~77-79, 1972 5. Jakobs K, Schultz G: Occurrence of a hormone

growth sensitive

horin-

hibitory coupling component of the adenylate cyclase in S49 lymphoma cyc variants. Proc Nat1 Acad Sci 80:3899-3902. 1983 6. Reisine T, Zhang Y, Sekura R: Pertussis toxin treatment blocks the inhibition of somatostatin and increases the forskolin stimulation of CAMP accumulation and ACTH secretion from mouse anterior pituitary tumor cells. J Pharrnacol Expt Ther 232:275-282, 1985 7. Schonbrunn A, Rorstad 0, Westerdorf J, et al: Somatostatin analogues: Correlation between receptor binding affinity and biolog-

SOMATOSTATI~

RECEPTORS

ical activity in GH pituitary cells. Endocrinology I 13:1559-1567, 1983 8. Reisine T, Guild S: Pertussis toxin blocks SRIF inhibition of calcium mobilization and desensitizes SRIF receptors. J Pharmacol Expt Ther 235:551-557, 1985 9. Koch B, Dorflinger L, Schonbrunn A: Pertussis toxin blocks both CAMP-rn~iat~ and CAMP-inde~ndent actions of somatostatin. J Biol Chem 260:13138-13145, 1985 IO. Lewis D, Weight F. Luini A: A guanine nucleotide-binding protein mediating the inhibition of voltage-dependent calcium current by somatostatin in a pituitary cell line. Proc Natl Acad Sci USA 83: 9035-9039, 1986 I 1. Mallard P, Vacher P, Dufy B, et al: Somatostatin blocks Ca++ action potential activity in prolactin-secreting pituitary tumor cells through coordinate actions on K+ and Ca++ conductances. Endocrinology 123:721-732, 1988 12. Wang H, Bogen C, Reisine T, et al: Somatostatin-14 and somatostatin-28 induce opposite effects on potassium currents in rat neocortical neurons. Proc Nat1 Acad Sci USA 8696 16-9620, 1989 13. Yatani A, Codina J, Sekura R, et al: Reconstitution of somatostatin and muscarinic receptor mediated stimulation of K+ channels by isolated G,, protein in clonal rat anterior pituitary cell membranes. Mol Endocrinol 1:283-293, 1987 14. Mihara S, North R, Suprenant A: Somatostatin increases an inwardly rectifying potassium conductance in guinea pig submucous plexus neurones. J Physiol 390:335-355, 1987 15. Srikant C. Pate1 Y: Receptor binding of somatostatin 28 is tissue specific. Nature 294:259-260, 198 1 16. Mandarin0 L, Stenner D, Blanchard W, et al: Selective effects of ~matostatin-14, -25, and -28 on in vitro insulin and glucagon secretion. Nature 29 I :76-78, 198 1 17. Brown M, Rivier J, Vale W: Somatostatin 28: Selective actions on the pancreatic beta cells and brain. Endocrinology 108:2391-2393, 1981 18. Amherdt M, Pate1 Y, Orci L: Selective binding of somatostatin14 and somatostatin-28 to islet cells revealed by electron microscopic autoradiopraphy. J Clin Invest 80:1455-1458, 1987 19. Thermos K, Reisine T: Somatos~tin receptor subtypes in the clonal anterior pituitary cell lines AtT-20 and GH3. Mol Pharmacol 33:370-377, 1988 20. Reubi J: Evidence for two somatostatin-I4 receptor types in rat brain cortex. Neurosci Lett 49:259-263, 1984 2 1. Tran V, Beal M, Martin J: Two types of somatostatin receptors

73

differentiated by cyclic somatostatin analogs. Science 228:492-495, 1985 22. Raynor K, Reisine T: Analogues of somatostatin selectively label distinct subtypes of somatostatin receptors in rat brain. J Pharmacol Ex Ther 251:510-577, 1989 23. Czernik A, Petrack B: Somatostatin receptor binding in rat cerebral cortex. Characte~tion using a non-reducible ~matostatin analogue. J Biol Chem 258:5525-5530, 1983 24. Veber D, Saperstein R, Nu HR, et al: A potent cyclic hexapeptide analogue of somatostatin. Nature 29255-58, 198 I 25. Sakamoto C, Goldtine I, Williams J: The ~matos~tin receptor on isolated pancreatic acinar cell plasma membranes. J Biol Chem 259~9623-9627, 1984 26. Susini C, Bailey A, Szecozka J, et al: Characterization of covalently cross-linked pancreatic somatostatin receptors. J Biol Chem 261:16738-16743, 1986 27. Lewis L, Williams J: Structural characterization of the somatostatin receptor in rat anterior pituitary membranes. Endocrinology 12 1:486-492,1987 28. Srikant C, Pate1 Y: Somatostatin receptors on pancreatic acinar cells. J Biol Chem 261:7690-7696, 1986 29. Thermos K, He HT, Wang H, et al: Biochemical properties of brain somatostatin receptors. Neuroscience 3 1:13 I- 142, 1989 30. He HT, Johnson K, Thermos K, et al: Purification of a putative brain somatostatin receptor. Proc Nat1 Acad Sci USA 86: 1480-1484, 1989 31. He HT, Rens-Domiano S, Martin J-M, et al: Solubilization of active somatostatin receptors from rat brain. Mol Pharmacol 37: 614-621, 1990 32. Wong Y, ~moliou-Men C, Barnard E: ADP-~~sylation with pertussis toxin modulates the GTP-sensitive opioid ligand binding domain in digitonin-soluble extracts of rat brain membranes. J Neurochem 51:114-121, 1988 33. Mahy N, Woolkalis M, Thermos K, et al: Pertussis toxin modifies the characteristics of both the inhibitory GTP binding proteins and the somatostatin receptor in anterior pituitary tumor cells. J Pharmacol Exp Ther 246779-785, 1988 34. Mahy N, Woolkalis M, Manning D, et al: Characteristics of somatostatin desensitization in the pituitary tumor cell line AtT-20. J Pharmacol Exp Ther 247:779-785, 1988 35. Mahan L, Reisine T, Axelrod J: Biochemical characterization of ~matos~tin receptors in AtT-20 cells. Sot Neurosci I2:6 14, 1986 (abstr)