THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol . 267, No. 22, Issue of August 5, pp. 15620-15625.1992 Printed in U. S. A.

Solubilization and Partial Purification of Somatostatin-28 Preferring Receptors from Hamster Pancreatic ,8 Cells*

(Received for publication, April 8, 1992)

Maura Maletti$§, Mats AnderssonT, Jean-Claude Marie$, Gabriel RosselinS, and Viktor Mutt7 From the Slnstitut National de la Santk et de la Recherche Medicale U. 55, H6pital Saint-Antoine, 75571 Paris Ceder 12, France and the VDepartment of Biochemistry II, Karolinska Institute, S-104 01 Stockholm, Sweden

Somatostatin-28 (SRIF-28) preferring receptors were solubilized from hamster cell insulinoma using the zwitterionic detergent 3-[(3-~holamidopropyl) dimethylammonio]- 1-propanesulfonate. The binding of the iodinated [L~U~-D-TRP~~-T~~~']SRIF-Z~ analog (referred to as '261[LWY] SRIF-28) to the solubilized fraction was time-dependent, saturable, and reversi- ble. Scatchard analysis of equilibrium binding data indicated that the solubilized extract contained two classes of SRIF-28-binding sites: a high affinity site ( K d = 0.3 nM and B,,, = 1 pmol/mg protein) and a low affinity site (& = 13 nM and BmaX = 4.7 pmol/mg protein). The binding of '2SI[LWY]SRIF-28 to solubi- lized SRIF-28 receptors was sensitive to the GTP an- alog guanosine-5'-O-thiotriphosphate, suggesting that receptors are functionally linked to a G-protein. By anion-exchange chromatography of the solubilized ex- tract followed by chromatography on wheat germ ag- glutinin, a 46-fold purification of SRIF-28 receptors was obtained. At this stage of purification, only high affinity sites were found (& = 1 nM) and the GTP effect was not maintained. A specific protein of 37 kDa was identified by sodium dodecyl sulfate-polyacryl- amide gel electrophoresis after photoaffinity labeling. We suggest that this protein is the putative SRIF-28 receptor or a subunit thereof.

Somatostatin-28 (SRIF-28),' an N terminally extended, bioactive form of somatostatin-14 (SRIF-14), was isolated from pig gut and hypothalamus (1,2) and sheep hypothalamus (3), after the initial isolation of SRIF-14 from sheep hypotha- lamic extracts (4). Both peptides inhibit not only the release of growth hormone but also that of insulin and glucagon (5). The inhibitory actions of somatostatin are initiated after

*This work was supported by Grants 1010 and 8902 from the Swedish Medical Research Council, by the Wenner-Gren Foundation, and by the Institut National de la Sant6 et de la Recherche MBdicale. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

5 To whom correspondence should be addressed INSERM U.55, Hijpital Saint-Antoine, 75571 Paris Cedex 12, France. Tel.: 33-1- 49284637; Fax: 33-1-43433234.

The abbreviations used are: SRIF-28, somatostatin-28; SRIF-14, somatostatin-14; SRIF, somatotropin release-inhibiting factor; SMS 201-995, D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-Thr-ol; Gi, inhibitory GTP-binding protein; CHAPS, 3-[(3-~holamidopropyl) dimethyl- ammoniol-1-propanesulfonate; HEPES, 4-(2-hydroxyethyl)-l-piper- azineethanesulfonic acid GTP-yS, guanosine-5'-O-thiotriphosphate; WGA, wheat germ agglutinin; ANB-NOS, N-5-azido-2-nitrobenzoy- loxysuccinimide; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

binding to high affinity receptors, characterized in various tissues (reviewed in Ref. 6). It has been suggested that distinct SRIF-28 and SRIF-14 receptors exist in the pancreatic islets, with p cells predominantly under the control of SRIF-28 and a cells regulated by SRIF-14 (7). In hamster insulinoma, consisting mostly of @ cells (8), receptors with higher affinity for SRIF-28 than SRIF-14 have been detected (9, 10). These findings are supported by studies in vitro indicating that SRIF-28 is more potent than SRIF-14 in blocking insulin secretion from pancreatic p cells (11, 12). This specificity of the SRIF-28 seems to be characteristic of the p cells (13). In contrast, in pancreatic acini (14, X ) , in brain and in GH3 cells (16-18) somatostatin receptors display higher affinity for SRIF-14 than for SRIF-28. The observation that SRIF- 28 exhibits receptor binding activities distinct from those of SRIF-14, the differences in the biological actions of the two peptides, and the high concentration of SRIF-28 in certain tissues suggest that this protein may serve not only as a precursor of SRIF-14 but also possesses independent physio- logical functions (19). The solubilization of active SRIF recep- tors has been reported in rat exocrine pancreas (20), brain (21), and in a human gastric tumoral cell line HGT-1 (22). While this article was in preparation, the cloning of two different somatostatin receptors, both with higher affinity for somatostatin-14 than somatostatin-28, has been reported (23). This confirms the existence of SRIF receptor subtypes.

The hamster @ cell insulinoma is one of the richest sources of SRIF receptors (10) and also contains other regulatory peptide receptors, related to the p cell function (24, 25). In the present work, we describe the solubilization, in a ligand- free form, of SRIF-28 preferring receptors from p cell insulin- oma. Partial purification of the receptors by DEAE and WGA- Sepharose chromatography has also been achieved. The bind- ing properties of this solubilized receptor were characterized and its functionality was supported by its interaction with G- proteins.

EXPERIMENTAL PROCEDURES

Materials-SRIF-28, [Leu', ~-Trp", Tyr25]somatostatin-28, and synthetic cholecystokinin octapeptide (CCK-8) were purchased from Peninsula Laboratories (Belmont, CA). SRIF-14 was a gift by Dr. J. Diaz (Sanofi Recherche, Montpellier, France) and SMS 201-995 was from Sandoz Pharmaceutical (Basel, Switzerland). [Leu', ~-Trp", TyrZ5]SRIF-28 was radioiodinated with carrier-free lZ5I-Na (referred to as '251[LWY]SRIF-28) and purified by high performance liquid chromatography, according to the method previously described for gastric inhibitory polypeptide (25), with some modifications (26). The specific activity of '251[LWY]SRIF-28 was approximately 2000 Ci/ mmol. Other products were from the following sources: CHAPS, phenylmethylsulfonyl fluoride, pepstatin, leupeptin, and soybean trypsin inhibitor, Boehringer (Mannheim, Germany); GTP-yS, bovine serum albumin, and bacitracin, Sigma; lectin-coupled agarose, RBac- tifs IBF Biotechnics (Villeneuve la Garenne, France); protein stand- ards for calibration of gel chromatography and SDS-PAGE, Bio-Rad;

15620

Solubilized and Active Endocrine Pancreatic SRIF-28 Receptors 15621

ANB-NOS, Pierce Chemical Co.; Sepharose 6B and DEAE-Sepharose CL-GB, Pharmacia LKB Biotechnology Inc. (Uppsala, Sweden).

Solubilization of SRIF Receptors-Hamster pancreatic 0 cell mem- branes were prepared from serially transplantable tumors of Syrian hamster, as previously reported (25), but in the presence of 0.1 mM phenylmethylsulfonyl fluoride and 0.1% bacitracin. SRIF receptors were solubilized by treatment of the membranes with CHAPS. Mem- branes (2-4 mg protein/ml) were incubated, for 15 min a t 4 'C, in 25 mM HEPES buffer, pH 7.5, 0.3% (w/v) CHAPS, 35% (v/v) glycerol, and the following protease inhibitors: 0.1 mM phenylmethylsulfonyl fluoride, 1 p~ pepstatin, 1 p~ leupeptin, 0.1% (w/v) bacitracin and 0.2% (w/v) soybean trypsin inhibitor. The suspension was then centrifuged at 100,000 X g for 30 min a t 4 "C. The supernatant was removed and either used immediately or stored at -80 'C.

Binding Experiments with CHAPS-solubilized Receptors-Binding of 1251[LWY]SRIF-28 to solubilized receptors was evaluated by incu- bating the solubilized proteins (0.2-0.4 mg protein/ml) with 10 pM 1251[LWY]SRIF-28 at 4 "C in the presence and absence of unlabeled SRIF-28. The final volume of incubation was 400 pl, and the assay buffer was 25 mM HEPES, pH 7.5, 0.2% bovine serum albumin, 1 p M pepstatin, and 1 PM leupeptin. After incubation for 10 h, the separa- tion of 1251[LWY]SRIF-28 bound to solubilized receptors from free ligand was carried out by vacuum filtration through Whatman GF/B filters pretreated with 0.3% (v/v) polyethylenimine in H20 (27). The filters were then washed twice with 4 ml of ice-cold 50 mM Tris-HCI, pH 7.5, and the radioactivity retained on filters was counted in a LKB 1282 y-counter. Nonspecific binding was measured in the pres- ence of an excess (1 p ~ ) of unlabeled SRIF-28 and subtracted from total binding to obtain specific binding. Values of nonspecific binding were always lower than 10% of total binding.

Gel Filtration of Solubilized Receptors-CHAPS-solubilized SRIF receptors (2 ml) were applied to a column of Sepharose 6B (1.5 X 83 cm), previously equilibrated with 25 mM HEPES, pH 7.5, 0.1% CHAPS, and 20% glycerol at 4 "C. The column was eluted with the same buffer at a flow rate of 11 ml/h. Fractions of 2.5 ml were collected and aliquots (300 pl) were assayed for binding activity as described above.

DEAE-Sephnrose Chromatography of Solubilized Receptors-The CHAPS-solubilized extract of insulinoma membranes (665 ml) was applied to a DEAE-Sepharose CL-GB column (5 X 20 cm), previously equilibrated at 4 "C with 25 mM HEPES, pH 7.5, 40 mM NaCI, 20% glycerol, and 0.1% CHAPS, at a flow rate of 150 ml/h. After loading the sample, adsorbed proteins were eluted in the equilibrating buffer with a stepwise gradient of NaCI. Aliquots (100 p l ) from each fraction (12 ml) of the eluate were assayed for binding activity. Eluted frac- tions containing the highest specific activities (termed A and B) were pooled and stored at -80 "C.

Lectin Affinity Chromatography-Two ml of wheat germ agglutinin (WGA)-agarose, in chromatographic tubes (1.1 X 10 cm), was equili- brated with 25 mM HEPES, pH 7.5,0.1% CHAPS, and 20% glycerol. The solubilized receptors (from the crude extract or the DEAE fraction A) were incubated with the gel for 5 h at 4 "C. Then the gel was washed with 12 ml of buffer before the elution of glycoproteins bound to WGA by 0.3 M N-acetylglucosamine (GlcNAc). Aliquots of 100 or 300 p1 from the collected fractions (2 ml) were assayed for binding activity as described above.

Cross-linking of Iz5I[L WYISRIF-28 to Soluble Receptors-Crude CHAPS-solubilized extract (64 pg of protein) as well as aliquots of fractions containing receptor activity that had been eluted from WGA-agarose or from DEAE-Sepharose (fractions A and B) or chromatographed on Sepharose 6B were incubated with 1251[LWY] SRIF-28 (60 pM) for 10 h at 4 "C in the presence or absence of 1 p~ unlabeled SRIF-28. The final volume of incubation was 800 pl and the assay buffer was 25 mM HEPES, pH 7.5, to which a mixture of protease inhibitors had been added: 1 GM pepstatin, 1 p~ leupeptin, 0.1% bacitracin, and 0.2% soybean trypsin inhibitor. After the incu- bation, a freshly prepared solution of ANB-NOS (20 mM in dimethyl sulfoxide) was added to a final concentration of 0.1 mM. After 5 min stirring in the dark on ice, the samples were exposed to UV light for 10 min under a 275 watt mercury-vapor lamp at a distance of 16 cm. The reaction was stopped by addition of 50 mM Tris-HC1, 2 mM EDTA (final concentration), pH 7.5, as a quenching buffer. The samples were then concentrated using Centricon-10 microconcentra- tors (Amicon) and analyzed by SDS-PAGE.

SDS-Polyacrylamide Gel Electrophoresis-The concentrated ma- terial was supplemented with sample buffer to a final concentration of 200 mM Tris-HCI, pH 6.8, 2% (w/v) SDS, 12% (v/v) glycerol, 0.001% (w/v) bromphenol blue, in either the presence or absence of

3% (v/v) 2-mercaptoethanol. The samples were heated for 30 min at 60 "C before application to polyacrylamide slab gels for electropho- resis according to Laemmli (28), using a 5% acrylamide stacking gel and a linear gradient resolving gel (10-15%). After staining with Coomassie Blue and destaining, gels were dried and exposed for 2-8 days at -80 "C with a Kodak X-Omat AR film and an enhancing screen.

Protein Determination-Protein concentration was determined by bicinchoninic acid assay (29) using a BCA kit (Pierce Chemical Co.) with bovine serum albumin as standard.

RESULTS

Conditions for Solubilization-Under our experimental con- ditions CHAPS was an appropriate detergent for solubilizing SRIF receptors, whereas with Triton X-100 no measurable binding activity was detected in the solubilized extract. To determine the optimal CHAPS concentration for solubiliza- tion of active somatostatin receptors, membranes of f l cell insulinoma (2-4 mg protein/ml) were incubated at 4 "C with various detergent concentrations. After centrifugation, the specific binding of 1251[LWY]SRIF-28 was measured in the solubilized fraction. Binding activity rapidly increased be- tween 0.1% and 0.4% CHAPS, with a maximum at 0.3-0.4%, where approximately 50% of membrane protein was solubi- lized. A drastic decrease of SRIF binding activity was observed between 0.5 and 1% CHAPS, whereas the concentration of solubilized protein still increased. In order to optimize extrac- tion conditions of active receptors, we have incubated the membranes with 0.3% CHAPS for different times. The con- centration of SRIF-binding sites in the extract reached a maximum at 15 min and rapidly decreased after 30 min. Addition of glycerol during solubilization resulted in a con- centration-dependent increase of the binding activity in the solubilized extract with a &fold improvement at 35% glycerol. CHAPS-glycerol-solubilized extracts frozen at -80 "C could be stored for several months without loss of binding activity. Therefore, routine solubilizations were performed for 15 min with 0.3% CHAPS and 35% glycerol.

Binding Characteristics of 12'Z[L WYISRIF-28 to Solubilized Receptors-The 1251[LWY]SRIF-28 binding of the solubilized extract was a time-dependent process. Fig. 1A shows that maximum binding at 4 "C was reached after 10 h of incuba- tion. The binding was stable for a t least 27 h. As shown in Fig. 1B, the binding of '251[LWY]SRIF-28 to solubilized recep- tors was reversible. When 1 PM unlabeled SRIF-28 or 100 PM of the non-hydrolyzable GTP analog GTPyS was added at equilibrium, about 40 and 35% of the bound radioactivity, respectively, was dissociated after 30 min of incubation at 30 "C. GTPyS, in combination with SRIF-28, accelerated the dissociation of 1251[LWY]SRIF-28 from solubilized receptors, reaching about 70% after 15 min of incubation.

GTPyS inhibited specific 12sI[LWY]SRIF-28 binding to solubilized receptors in a dose-dependent manner (Fig. 2). An inhibitory effect was observed already at 1 PM GTP+, and 100 ,LLM GTPyS (the highest concentration tested) caused a 70% decrease of specific 1251[LWY]SRIF-28 binding.

To further evaluate the effect of solubilization on the spec- ificity of SRIF-binding sites, the ability of SRIF-28, SMS 201-995, and SRIF-14 for inhibiting 1251[LWY]SRIF-28 bind- ing to solubilized receptors was tested (Fig. 3). The concen- trations of peptides required to produce half-maximal inhibi- tion (1cb0) of 1251[LWY]SRIF-28 binding to solubilized recep- tors were in the following order: SRIF-28 = 0.76 nM) < SMS 201-995 (ICs, = 9 nM) < SRIF-14 (ICE, = 32 nM). These results clearly indicated the presence of SRIF-28 preferring binding sites in the solubilized f l cell insulinoma. CCK-8, up to 1 HM, was unable to compete with '251[LWY]SRIF-28. Scatchard analysis of the data, using the LIGAND program

15622 Solubilized and Active Endocrine Pancreatic SRIF-28 Receptors

0 IO 20 30 Time (hours)

B

40

20

04 . , . , . , . . I 0 20 40 60 80

Time (min)

FIG. 1. Time course of association and dissociation of spe- cific '261[LWY]SRIF-28 binding to the solubilized extract. Panel A , kinetics of association of '251[LWY]SRIF-28 (10 PM) to solubilized membranes (0.2-0.4 mg protein/ml) were followed at 4 "C. Panel B, after a plateau value has been reached in the association kinetics described in panel A , dissociation of bound '251[LWY]SRIF- 28 was initiated at 30 "C by addition of 1 p~ unlabeled SRIF-28 (.) or 100 p~ GTPyS (0) or 1 p~ SRIF-28 + 100 p~ GTPyS (0) compared to control (A). Results are expressed as percentage of specific binding (in panels A and B ) . Each point represents (in panels A and B ) the mean of three experiments performed in duplicate.

-9 -8 -7 -6 -5 -4 G T P 6 (1% M)

FIG. 2. GTPrS inhibition of specific 1261[LWY]SRIF-28 binding to solubilized receptors. The effect of varying concentra- tions of GTPyS on specific '251[LWY]SRIF-28 binding to solubilized receptors in the crude extract (O), and in the fractions A (0) and B (0) eluted after DEAE chromatography (see Fig. 5) was tested. Values are expressed as percent of specific binding obtained in the absence of GTP-yS. An experiment representative of three separate studies is presented.

(30), indicated the presence in the solubilized extract of two classes of SRIF-28-binding sites (Table I): a high affinity binding site with a Kd of 0.32 f 0.04 nM and a B,,, of 0.94 f 0.17 pmol/mg of protein and a low affinity binding site with a Kd of 13.3 f 5 nM and a B,,, of 4.65 f 1.07 pmol/mg of protein.

Gel Filtration of Solubilized SRIF Receptors-To estimate the apparent molecular weight of the solubilized SRIF recep- tor complex, the CHAPS-solubilized extract was subjected to

-11 -10 -9 -8 -7 -6 -5

Peptide (log M) FIG. 3. Competitive inhibition of binding of '2KI[LWY]

SRIF-28 to solubilized receptors by somatostatin analogs. Solubilized receptors (150 pg of protein) were incubated at 4 "C for 10 h with 10 PM '251[LWY]SRIF-28 and increasing concentrations of unlabeled SRIF-28 (O), SMS 201-995 (A) or SRIF-14 (m). Data are expressed as percentage of the specific binding measured in the absence of peptides. Each point represents the mean of four separate experiments performed in duplicate. Inset, Scatchard analysis of data from competitive displacement by SRIF-28. The data are represent- ative of four separate experiments.

gel filtration on Sepharose 6B at 4 "C. Aliquots of collected fractions were incubated with '251[LWY]SRIF-28 to deter- mine the elution profile of binding activity. Fig. 4 shows that the solubilized SRIF receptor complex was eluted as a single peak corresponding to an apparent molecular mass of 550 kDa.

DEAE-Sepharose Anion-Exchange Chromatography of the Solubilized SRIF-28 Receptor-In order to partially purify the SRIF-28 receptor, solubilized extract of insulinoma mem- branes was applied to DEAE-Sepharose and the eluted frac- tions were assayed for '251[LWY]SRIF-28 binding activity. As shown in Fig. 5 (left), the solubilized binding sites were found to distribute into two main peaks, referred to as fraction A and fraction B and eluted, respectively, at 40 and 150 mM NaC1. SRIF-28 strongly inhibited '251[LWY]SRIF-28 binding to receptors in fractions A and B (Fig. 5, right). Binding parameters were: Kd = 0.12 f 0.03 nM for the fraction A and 0.55 f 0.16 nM for the fraction B, as calculated by Scatchard analyses (Table I). This indicates that both fractions con- tained high affinity receptors for SRIF-28. Table I1 shows the results obtained after the partial purification of the crude extract on DEAE-Sepharose. The purification was 2.7-fold for the fraction A. The inhibition by GTP-yS of the specific '251[LWY]SRIF-28 binding in fractions A and B was less efficient in comparison to that in the crude extract (Fig. 2). At 100 PM GTP-yS, the inhibition represented 45 and 20% decrease of specific binding in A and B, respectively.

Lectin Affinity Chromatography-The solubilized SRIF-28 receptor in cell insulinoma was found to be a glycoprotein by its interaction with wheat germ lectin. Crude CHAPS extract was incubated with WGA-agarose and, after a wash, 0.3 M GlcNAc was used to elute adsorbed glycosylated pro- teins. Fig, 6 shows that a large part of the total soluble binding activity could be retained on WGA-agarose and specifically eluted with 0.3 M GlcNAc. Scatchard analysis of the lectin eluted binding activity demonstrated the presence of both high and low affinity binding sites, as found in the crude extract (Table I): &H = 0.55 f 0.12 nM and K ~ L = 6.4 f 2.3 nM. The overall purification was 12.5-fold. Therefore, as a next step in the purification, high affinity receptors in fraction A from DEAE-Sepharose were applied to WGA-Sepharose. The receptors retained and eluted in this step maintained a high affinity binding for SRIF-28. According to Scatchard analysis, the binding was consistent with the presence of a

Solubilized and Active Endocrine Pancreatic SRIF-28 Receptors 15623

TABLE I Binding parameters of solubilized receptors

The parameters were determined from Scatchard plots of the equilibrium specific binding data. Values are the mean f S.E. of three senarate exueriments (four for the crude extract).

Crude extract WGA-Sepharose DEAE-Sepharose

Fraction A Fraction B

WGA of DEAE fraction A

nM pmollmg nM p m o l l w 0.32 f 0.04 0.94 ? 0.17 13.3 2 5 4.65 f 1.07 0.55 f 0.12 11.8 f 2.7 6.4 f 2.3 52.3 f 20

0.12 f 0.03 2.5 f 0.4 0.55 f 0.16" 1.18 f 0.22 0.99 f 0.26* 43 f 7.3

Significantly different from the values obtained in fraction A ( p < 0.01, n = 3). Significantly different from the value obtained in fraction A before WGA ( p < 0.004, n = 3).

p 5 2

f 5.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6

Ka"

0 10 20 30 40 50 60

Fraction number

FIG. 4. Gel filtration of CHAPS-solubilized SRIF-28 recep- tors. Two ml of crude CHAPS-solubilized extract was applied at 4 "C to a Sepharose 6B column (1.5 X 83 cm) equilibrated with HEPES 25 mM, pH 7.5, 0.1% CHAPS, and 20% glycerol. Aliquots of the eluted fractions were assayed for specific binding of '251[LWY]SRIF- 28, as described under "Experimental Procedures." The column has been previously calibrated with blue dextran (void volume = Vo), thyroglobulin (Mr, 669,000), ferritin (Mr, 440,000), catalase (Mr, 232,000) and aldolase (Mr, 158,000). Fraction volume, 2.5 ml; flow rate, 11 ml/h. Inset, the logarithmic molecular weight of the marker proteins (0) is plotted versus Kav. The apparent molecular weight of the eluting binding site (0) was obtained from this plot.

0 20 40 60 80 IW Fraruon number

FIG. 5. DEAE-Sepharose chromatography of CHAPS-soh- bilized SRIF-28 receptors. Left, crude CHAPS-solubilized extract (665 ml) was applied at 4 "C to a DEAE-Sepharose column (5 X 20 cm) equilibrated with 25 mM HEPES, pH 7.5, 40 mM NaC1, 20% glycerol, and 0.1% CHAPS. Fraction volume, 12 ml; flow rate, 150 ml/h. Aliquots of the eluted fractions were assayed for specific binding of '251[LWY]SRIF-28 (O), as described under "Experimental Proce- dures." Absorbance at 280 nm is shown by the broken line (- - -). The stepwise gradient of NaCl is indicated by the dotted line (. . . .). Fractions used for further studies were named A and E , as indicated by the bars. Right, competitive inhibition of '251[LWY] SRIF-28 binding to solubilized receptors in DEAE fractions A (0) and B (0) by unlabeled SRIF-28. Results are expressed as percentage of specific binding, determined as described under "Experimental Procedures." Each point represents the mean of four separate exper- iments performed in duplicate.

single class of sites with a Kd of 0.99 k 0.26 nM and a B,,, of 43 f 7.3 pmol/mg of protein (Table I). These values corre- spond to a 46-fold purification (Table 11). At this stage of

purification, the GTP effect was not maintained. Affinity Labeling of Soluble SRIF Receptors-Crude solubi-

lized insulinoma extract was covalently labeled with '251[LWY]SRIF-28 by the heterobifunctional cross-linking reagent ANB-NOS, and then analyzed by SDS-PAGE. A radiolabeled band of 37 kDa was observed in the absence (Fig. 7, lune 1 ) but not in the presence (Fig. 7, lane 2 ) of 1 pM SRIF-28, indicating the specificity of the labeling.

The solubilized SRIF receptor in the -550 kDa complex observed by gel filtration chromatography (fraction 19 in Fig. 4) was also cross-linked and analyzed by SDS-PAGE. Again, a specific 37-kDa band was photolabeled (data not shown).

A further demonstration that the 37-kDa protein is the hormone-binding subunit of the putative SRIF receptor is provided by covalent labeling of the proteins in the two fractions (A and B) separated by DEAE chromatography. For both fractions a major band corresponding to an apparent molecular mass of 37 kDa was observed in SDS-PAGE analy- sis (Fig. 7, lunes 3 and 5 ) , specifically abolished by 1 p~ SRIF- 28 (Fig. 7, lanes 4 and 6).

The material retained and eluted from WGA by 0.3 M GlcNAc (fractions 7-8 in Fig. 6) has also been cross-linked using the same procedure and analyzed by SDS-PAGE. A protein band of 37 kDa was photolabeled (Fig. 7, lane 7). The labeling of this band was specifically extinguished when 1 p~ unlabeled SRIF-28 was included during binding (Fig. 7, lune 8).

DISCUSSION

This paper reports the solubilization and partial purifica- tion of SRIF-28 preferring receptors from the hamster endo- crine pancreas in an active and unoccupied form. The highest specific binding activity could be recovered when 0.3% CHAPS and 35% glycerol were used. The presence of glycerol was essential for obtaining high amounts of active solubilized receptors and for their stabilization. CHAPS has been effi- ciently employed for the solubilization of somatostatin recep- tors from the rat exocrine pancreas (20) and brain (21), although at higher concentrations (3 and 0.676, respectively) and using a longer extraction time (60 min for the brain).

Scatchard analyses of equilibrium competitive binding data indicated the presence in the solubilized extract of two classes of binding sites. The & for the high affinity site (0.32 f 0.04 nM) was similar to that reported for membrane-bound recep- tors (10). Solubilized p cell somatostatin receptors maintain the preference for SRIF-28 over SRIF-14 observed in mem- brane-bound receptors (9, 10). Also, the binding of '251[LWY] SRIF-28 to solubilized receptors is selective since CCK-8, which in pancreatic acinar cell membranes interacts with the somatostatin-14 receptor (31), did not interfere with SRIF-28

15624 Solubilized and Active Endocrine Pancreatic SRIF-28 Receptors TABLE I1

Purification of high affinity somatostatin-28 receptors from hamster insulinoma The data in this table are from a single preparation, which is representative of four similar experiments.

Total Total binding Specific protein activity activity

Purification Yield

mRn pmol pmollmg -fold % Crude extract 1077 1012 0.94 1 100 DEAE fraction A 120 300 2.5 2.7 30 WGA of DEAE fraction Ab 2.4 103 43 46 10

’ Protein was determined as described under “Experimental Procedures.” Values extrapolated from an experiment performed on a 3% aliquot of DEAE fraction A.

0 2 4 6 X 1 0 1 2

Fraction number

FIG. 6. WGA affinity chromatography of solubilized SRIF- 28 receptors. Three ml of the crude CHAPS-solubilized receptors was applied to WGA-agarose (2 ml). After washing the gel with 25 mM HEPES, pH 7.5, 0.1% CHAPS, and 20% glycerol, the adsorbed proteins were eluted with 0.3 M GlcNAc in the same buffer as shown by the arrow. Aliquots of the collected 2-ml fractions were assayed for specific binding of ’251[LWY]SRIF-28, as described under “Ex- perimental Procedures.”

CRUDE EXTRACT A

DEAE WGA B vr -66

-45

3 7 w * 9 I::

SRIF-28 - + - + - + 1 2 3 4 5 6 7 8

FIG. 7. SDS-PAGE analysis of ’261[LWY]SRIF-28 cross- linked to solubilized receptors. ’251[LWY]SRIF-28 was cross- linked to crude CHAPS-solubilized extract and to aliquots of receptor eluted from DEAE-Sepharose (fractions A and B ) and WGA-agarose. Cross-linking was performed in the absence (lanes 1,3, 5, and 7) or presence (lanes 2, 4, 6, and 8) of 1 p~ SRIF-28. The methods of cross-linking and autoradiography by SDS-PAGE are described under “Experimental Procedures.” The position to which protein markers (kDa) had migrated is indicated on the right of the WGA autoradi- ograph, obtained after a separate cross-linking experiment. The ar- rows on the left indicate the molecular mass (kDa) of the specific labeled protein.

receptors. However, there was an important shift in the affinity of the analog SMS 201-995, which bound efficiently to solubilized receptors, in contrast with its low affinity for the membrane-bound receptors (10). Changes in ligand affin- ities between membrane-bound and solubilized receptors have been previously reported (20, 32).

After purification on DEAE-Sepharose, only high affinity sites were found distributed in two eluted fractions (A and B). The fractions A and B showed different Kd values, al- though close to that found in the crude extract (Table I). The reason for this is unclear, but one possible explanation could be differences in interactions with G-proteins, indicated by

the different sensitivities of the two fractions to GTPyS. The low affinity sites, found in the crude extract, are no longer detected after DEAE-Sepharose chromatography. (The low affinity binding proteins may have been degraded during the procedure or alternatively may require >0.4 M NaCl for their elution.)

Taking advantage of our finding of a glycoprotein nature of SRIF-28 receptors, proteins in fraction A were further purified by chromatography on WGA-agarose. The specific activity increased after DEAE chromatography to 2.5 pmol/ mg of protein and was further increased to 43 pmol/mg of protein after lectin chromatography of the fraction A, with a recovery of 10% after these two purification steps. WGA affinity chromatography has been useful in the purification of several receptors (33, 34). SRIF-28 receptors interacted strongly with WGA-agarose, suggesting that the receptor car- ries a glycosylated moiety containing N-acetylglucosamine residues and/or N-acetylneuraminic acid. Interestingly, solu- bilized SRIF receptors from hamster endocrine pancreas were eluted from the WGA gel by GlcNAc, whereas those from rat exocrine pancreas (20) could not be eluted by GlcNAc but only by triacetylchitotriose, which possesses higher affinity for WGA (35). This suggests that these two mammalian receptors differ in the nature of their carbohydrate chains. Initially, SRIF-28 receptors were solubilized as a large com- plex, as determined by gel exclusion chromatography. GTP- binding proteins may be one component of the SRIF receptor complex because a potent ligand binding inhibitory and dis- sociating effect of GTPyS was found in p cell insulinoma- solubilized receptors. This GTP-binding unit might be the Gi protein, since an inhibition of adenylate cyclase activity by somatostatin has been found in pancreatic /3 cells (36). This inhibitory effect was progressively lost during purification, in parallel with the decrease in the affinity for the hormone noted after WGA chromatography of the fraction A (Table I). A similar observation has been made during the purifica- tion of SRIF receptors from the human gastric tumoral cell line HGT-1, where the loss of the GTP effect was found to be accompanied by a decrease in the affinity (22).

Covalent cross-linking of ‘2sI[LWY]SRIF-28 to crude sol- ubilized receptors from p cell insulinoma revealed the pres- ence of a specifically labeled band of an apparent molecular mass of 37 kDa, both in the absence or presence of 2-mercap- toethanol (data not shown), suggesting that the receptor does not contain disulfide-linked subunits. The same specific band of 37 kDa was also identified when ‘251[LWY]SRIF-28 was cross-linked to the receptor after gel filtration, DEAE-Seph- arose, or WGA-agarose. This provides evidence for a specific 37-kDa binding protein in solubilized extracts from hamster /3 cell insulinoma, co-purifying with the high affinity binding activity in the different chromatographic steps. Until now, there is no consensus in the molecular species identified, which have been found to range in size from 27 to 200 kDa (10,13, 15,37-42). Some of the observed discrepancies might

Solubilized a n d Active Endocrine Pancreatic SRIF-28 Receptors 15625

provide further support for the existence of different receptor subtypes distributed in different tissues (13, 43). Recently, the cloning of two dissimilar SRIF-14 receptors stresses the idea of a large SRIF receptor family (23). A difference can be noted between the mass of the solubilized putative SRIF receptor in our study (37 kDa) and the several higher molec- ular mass species (45, 132, and 196 kDa) previously reported in the same hamster fi cell insulinoma (10). This can be explained by the fact that, in the earlier report, during the cross-linking of membrane-bound receptors, other proteins may have become covalently attached to the hormone-binding receptor subunit. In the present experiment, such proteins are likely to have been separated from the receptor during solubilization.

In conclusion, we describe the solubilization and a 46-fold purification of endocrine pancreatic high affinity SRIF-28 receptors. Moreover, we provide evidence for a specific SRIF- binding protein of 37 kDa and suggest that this protein is the putative SRIF-28 preferring receptor or a ligand-binding sub- unit thereof. Although the small size of this receptor is intrigu- ing, it is not unique since the prolactin and the growth hormone-releasing factor receptors are reported to have a mass of only 42 and 26 kDa respectively (44, 45). Moreover, the recently cloned somatostatin-14 receptors have calculated molecular masses of roughly 41 and 43 kDa for their protein components (23).

Acknowledgments-We thank Drs. E. Forgue-Lafitte, J. Mester, and G. Skoglund for helpful suggestions and A. Barakat for the preparation of membranes.

REFERENCES 1. Pradayrol, L., Jornvall, H., Mutt, V., and Ribet, A. (1980) FEBSLett. 109 ,

2. Schally, A,, Huang, W.-Y., Chang, R., Arimura, A,, Reddin T Millar R Hunkapiller, M., and Hood, L. (1980) Pruc. Natl. Acad. &i. 6. S. A. '77:

3. Bohlen, P., Brazeau, P., Benoit, R., Ling, N., Esch, F., and Guillemin, R.

4. Brazeau, P., Vale, W., Burgus, R., Ling, N., Butcher, M., Rivier, J., and

5. Meyers, C., Murphy, W., Redding, T., Coy, D., and Schally, A. (1980) Pruc.

6. Lewin, M. J. M. (1986) Scand. J. Gastroenterol. 2 1 (Suppl. 119), 42-46 7. Amberdt, M., Patel, Y., and Orci, L. (1987) J. Clin. Invest. 80, 1455-1458 8. Shapiro, S., Eto, S., Fleischer, N., and Baum, S. (1975) Enducrrnolugy 9 7 ,

9. Reubi, J.-C., Rivier, J., Perrin, M., Brown, M., and Vale, W. (1982)

10. Cotroneo, P., Marie, J.-C., and Rosselin, G. (1988) Eur. J. Biuchem. 174,

55-58

4489-4493

(1980) Bwchem. Biuphys. Res. Cummun. 9 6 , 725-734

Guillemin, R. (1973) Science 179 , 77-79

Natl. Acad. Scr. U. S. A. 77, 6171-6174

442-447

Endocrinology 110 , 1049-1051

219-224

11. Mandarino, L., Stenner, D., Blanchard, W., Nissen, S., Gerich, J., Ling, N., Brazeau, P., Bohlen, P., Esch, F., and Guillemin, R. (1981) Nature 2 9 1 , 76-77

13. Thermos, K., Meglasson, M., Nelson, J., Lounsbury, K., and Reisine, T. 12. Brown, M., Rivier, J., and Vale, W. (1981) Endocrinology 108,2391-2393

14. Taparel, D., Estgve, J. P., Susini, C., Vaysse, N., Balas, D., Berthon, G., Wunsch, E., and Ribet, A. (1983) Biochem. Biophys. Res. Cummun. 115 , 827-833

16. Srikant, C., and Patel, Y. (1987) Somatostatin Rece tor Evidence for 15. Srikant, C., and Patel, Y. (1986) J. Biol. Chem. 2 6 1 , 7690-7696

Structural and Functional Heterogeneity (Reichlin, i., ed) pp. 89-102, Plenum Press, New Yurk

(1990) Am. J. PhySiOl. 259. E216-E224

17. Thermos, J., and Reisine, T. (1988) Mol. Pharmacol. 33,370-377 18. Schonbrunn, A,, Rorstad, O., Westendorf, J., and Martin, J. (1983) Endu-

19. Srikant, C., and Patel, C. (1981) Nature 2 9 4 , 259-260 20. Knuhtsen, S., EstBve, J.-P., Cambillau, C., Colas, N., Susini, C. , and Vaysse,

21. He, H.-T., Rens-Domiano, S., Martin, J.-M., Law, S., Borislow, S., Wool- N. (1990) J. Bid. Chem. 2 6 5 , 1129-1133

kalis, M., Manning, D., and Reisine, T. (1990) Mol. Pharmacol. 37,614- 62 1

crinology 113 , 1559-1567

22. ReylIDesmars, F., Le Roux, S., Linard, C., Benkouka, F., and Lewin, M. J.

23. Yamada, Y., Post, S. R., Wang, K., Tager, H. S., Bell, G. I., and Seino, S.

24. Goldfine, I., Roth, J., and Birnbaumer, L. (1972) J. B i d Chem. 247,1211-

M. (1989) J. Biol. Chem. 264 , 18789-18795

(1992) Pruc. Natl. Acad. Sci. U. S. A. 89,251-255

25. Maletti, M., Portha, B., Carlquist, M., Kergoat, M., Laburthe, M., Marie,

26. Antoniotti, H., Fagot-Revurat, P., Esteve, J. P., Fourmy, D., Pradayrol, L.,

27. Bruns, R. F., Lawson-Wendling, K., and Pugsley, T. A. (1983) Anal.

28. Laemmli, U. K. (1970) Nature 227,680-685 29. Smith, 0. K., Krohn, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F.

H., Provenzano, M. D., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Anal. Biochem. 150,76-85

1218

J.C., and Rosselin, F. (1984) Endocrinology 115 , 1324-1331

and Ribet, A. (1984) J. Chromatogr. 296,181-188

Biochem. 132 , 74-81

30. Munson, P., and Rodbard, D. (1980) Anal. Biochem. 107 , 220-239 31. Sakamoto, C., Goldfine, I., and Williams, J. (1984) J. Biol. Chem. 2 5 9 ,

9623-9627 32. Patthi, S., Simerson, S., and Velicelebi, G. (1988) J. Biol. Chem. 2 6 3 ,

19363-19369 33. Lomasney,. J., Leeb-Lundberg, L. M. F.. Cotecchia, S., Regan, J., De-

Bernards, J., Caron, M., and Lefkowltz, R. (1986) J. Biol. Chem. 261 , 7710-771fi

34. Fujita-Yamaguchi, Y., Choi, S., Sakamoto, Y., and Itakura, K. (1983) J.

35. Goldstein, I. J., and Hayes, C. E. (1978) Ado. Carbohydr. Chem. Biochem.

. ._" . ." Bid. Chem. 258,5045-5049

36. Schuit, F., and Pipeleers, D. (1985) Endocrinology 117,834-840 37. Knuhtsen, S., EstBve, J.-P., Bernadet, B., Vaysse, N., and Susini, C. (1988)

38. Sakamoto, C., Nagao, M., Matozaki, T., Nishizaki, H., Konda, Y., and

39. Thermos, K., He, H.-T., Wang, H.-L., Margolis, N., and Reisine, T. (1989)

40. Bruno, J., and Berelowitz, M. (1989) Endocrinology 124 , 831-837 41. Murthy, K., Srikant, C., and Patel, Y. (1989) Endocrinology 125,948-956 42. Srikant, C., and Patel, Y. (1986) Biuchem. Biophys. Res. Commun. 139 ,

43. Reubi, J. C. (1984) Neurusci. Lett. 4 9 , 259-265 44. Okamura, H., Raguet, S., Bell, A,, Gagnon, J., and Kelly, P. (1989) J. Biol.

45. Zysk, J., Cronin, M., Anderson, J., and Thorner, M. (1986) J. Bid. Chem.

35, 127-340

Biochem. J. 254,641-647

Baba, S. (1988) J. Biol. Chem. 263,14441-14445

Neuruscrence 31 , 131-141

757-762

Chem. 264,5904-5911

2 6 1 , 16781-16784