Production and Characterization of Antibodies to Gonadotropin-releasing Hormone Receptor*

Antibodies to the gonadotropin-releasing hormone (GnRH) receptor of bovine pituitary membranes have been raised in rabbits by immunization with affinity- purified receptor preparations. These antibodies did not compete with 1251-labeled GnRH analog (Buserelin) for binding to the receptors but did precipitate rat and bovine solubilized receptors labeled with lz5I-Busere- lin. Binding of the antibodies to the receptors was also demonstrated by immunoprecipitation of 1251-labeled purified receptors and photoaffinity-labeled receptors. The antibodies did not have a GnRH-like activity but rather inhibited, in a dose-dependent manner, GnRH-stimulated luteinizing hormone release from cultured rat pituitary cells. In addition, the antibodies did not inhibit luteinizing hormone release stimulated by high K' concentration. This suggests that the antibodies recognize domains of the receptor other than the binding site of the hormone and thereby inhibit the biological response. These GnRH receptor antibodies provide a useful tool for studying GnRH receptor structure, function, localization, and biosynthesis. The hypothalamic decapeptide gonadotropin-releasing hormone (GnRH)' is the primary regulator of the reproductive cycle in both sexes. Binding of this hormone to specific receptors on pituitary gonadotrope cell membranes initiates the processes that lead to gonadotropin release. Characterization of the GnRH receptor in the pituitary has indicated that the receptor is a glycoprotein which contains sialic acid residues (1,2) and that membrane phospholipids

The hypothalamic decapeptide gonadotropin-releasing hormone (GnRH)' is the primary regulator of the reproductive cycle in both sexes. Binding of this hormone to specific receptors on pituitary gonadotrope cell membranes initiates the processes that lead to gonadotropin release. Characterization of the GnRH receptor i n the pituitary has indicated t h a t the receptor is a glycoprotein which contains sialic acid residues (1,2) and that membrane phospholipids are involved in the interaction between the hormone and the receptor (3). Photoaffinity labeling of pituitary GnRH receptors with a photoreactive derivative of GnRH, results in the identification of a single specific protein with an apparent molecular mass of 60 kDa (4). This protein has been identified as the physiological pituitary receptor for GnRH (5).
Purification of the GnRH receptor as well as production of anti-receptor antibodies has been hampered by the low density of receptors in pituitary membranes. Although solubilization of bovine pituitary GnRH receptor has been reported * This work was supported by the United States-Israel Binational Science Foundation and by the Fund for Basic Research administered by the Israel Academy of Sciences and Humanities. 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. The abbreviations used are: GnRH, gonadotropin-releasing hormone; Buserelin, [D-Ser(t-Bu)', des-Gly", ethylamide1GnRH; LH, luteinizing hormone; CHAPS, 3-[(3-cholamidopropyI)dimethylammoniol-1-propane sulfonate; SDS, sodium dodecyl sulfate. previously (6-9), the purification of GnRH receptors from rat pituitary has been achieved only recently (10). A homogenous receptor preparation was obtained following two cycles of affinity chromatography on [biotinyl-~-Lys~]GnRH coupled to avidin-agarose (10,11). We report here the preparation and characterization of antibodies to the purified GnRH receptor from bovine pituitary membranes.

Iodination and Pituitary Membrane Preparations-Buserelin and
[azidobenzoyl-~-Lys~]GnRH were iodinated by the lactoperoxidase method (4,14). Specific activity of the labeled peptides was approximately 1.0 mCi/pg, as measured by self-displacement in the pituitary radioligand receptor assay. Bovine pituitary membranes were prepared as previously described (15). Briefly, bovine pituitaries taken from a nearby slaughter house were kept at 4 "C in assay buffer, 25 mM Tris phosphate, pH 7.4. The glands were homogenized for 90 s in an Ultra-turrax homogenizer in assay buffer at 4 "C and centrifuged for 15 min at 1, OOO X g. The supernatant solution was then centrifuged for 20 min at 20,000 X g. The bottom, white layer was discarded while the top, loose pellet was resuspended in assay buffer and centrifuged for 20 min at 20,000 X g. This procedure was followed twice until the pellet appeared homogeneous. Rat pituitary membranes were prepared as previously described (10).
Purification of GnRH Receptors and Production of Antibodies-Bovine and rat pituitary membrane preparations were solubilized as described (lo), with modifications. The pellet was suspended in 25 mM Tris phosphate buffer containing 5 mM CHAPS, shaken for 60 min at 4 "C, and centrifuged (90 min at 100,000 X g ) . This procedure was repeated, and the supernatants were combined and kept in 1 mM CHAPS, 25 mM Tris phosphate buffer containing 10% glycerol and 1 mM phenylmethylsulfonyl fluoride. GnRH receptors were purified using affinity chromatography (10) on [biotinyl-D-L~s'IG~RH immobilized on avidin-agarose. New Zealand White rabbits were injected subcutaneously with affinity-purified receptor preparations mixed with 2 volumes of Freund's adjuvant in a final volume of up to 3 ml. The first two injections were conducted with receptor preparations eluted from the affinity column (10) with 0.5 M NaCl (each injection contains approximately 7 pmol of receptor, purification fold >15). Three more injections were conducted with highly purified receptor preparations, i.e. the receptors were eluted from the affinity column with lo4 M Buserelin following purification on slab gels and slicing a segment around a M, of 60,000 (each injection contains approximately 6-8 pmol of receptor, purification fold >15,000). The rabbits were injected at 14-day intervals and were bled after at least three immunizations. Antibodies were detected after the fourth injection.
Binding Assays to Solubilized Receptors-Solubilized receptors (bovine, 200 to 300 pg of protein; rat, 50 pg of protein) were incubated with labeled Buserelin (50,000 cpm) in 0.5 ml of 25 mM Tris, 0.1% bovine serum albumin, 1 mM CHAPS containing 10% glycerol for 20 h at 4 "C. The reaction was stopped by the addition of 0.3 ml of icecold dextran-coated charcoal as described (10). When antibodies were used, at the end of the equilibrium binding reaction (2.5 h at 4 'C), immune or preimmune serum was added at a final dilution of 1:10 and the incubation continued at 4 "C. After 18 h, 100 pl of 0.2%

531
Cowan I strain of s. aureus in 1 mM CHAPS, 25 mM Tris phosphate was added (1 h at 4 OC), and the '251-Buserelin-receptor-antibody complex W-precipitated by centrifugation (20 min at 2,000 X g) and counted in a y-counter. Specific binding represents the radioactivity bound in the presence of lo-' M unlabeled Buserelin subtracted from the total bound radioactivity, Each point is the mean of triplicate incubations from at least three separate experiments, which varied by less than 10%.
Immunoprecipitation of Photoaffinity-&led and Iodine-labeled Receptors-Photoaffinity-labeled receptors were prepared by incubation of pituitary membranes with '261-[azidobenzoyl-~-Lys6]GnRH ( lo6 cpm) in the absence and presence of M Buserelin in a final volume of 1.0 ml of assay buffer. After 90 min at 4 "C, in the dark, the membranes were photolyzed (4) with a mercury lamp (7 min at 4 "C) and solubilized with 5 mM CHAPS. Purified receptor was iodinated with Na'"I and chloramine T (10). Photoaffinity-labeled and iodine-labeled receptors were incubated with immune or preimmune serum in 25 mM Tris, 1 mM CHAPS, 10% glycerol. After 18 h at 4 'C, Cowan I strain of Staphylococcus aureus was added (1 h at 4 "C), and the receptor-antibody complex was precipitated by centrifugation (20 min at 2000 X g) and prepared for SDS-polyacrylamide gel electrophoresis.
Cell Culture and Assay of Biological Activity-Rat pituitary cell cultures were prepared as previously described (2). The biological activity was assessed by the quantitation of LH released from the cells (2). After 48 h in culture, the medium was discarded and the cells were washed twice with medium 199. The cells were incubated with immune or preimmune serum alone (6 h at 37 "C) or preincubated with serum (2 h at 37°C) followed by additional incubation with 0.5 nM GnRH (4 h at 37 "C).
SDS-polyacrylamide Gel Electrophoresis-Gel electrophoresis was performed according to the method of Laemmli (16) using 9% SDSpolyacrylamide slab gels. After staining with Coomassie Blue (R-250) and destaining, the gels were dried for autoradiography. Autoradiography was performed using Kodak X-Omat AR film with a Dupont Cronex screen. High molecular weight standard mixture for gel electrophoresis was purchased from Sigma. The protein markers were: myosin, 205,000; @-galactosidase, 116,000; phosphorylase b, 97,400; bovine serum albumin, 66,000; egg albumin, 45,000; and carbonic anhydrase, 29,000.
Protein Determination-Protein concentrations were determined according to Bradford (17) with bovine serum albumin as a standard.

RESULTS AND DISCUSSION
Bovine pituitary GnRH receptor was solubilized with the zwitterionic detergent CHAPS and purified by affinity chromatography on a n immobilized avidin column coupled to [biotinyl-~-Lys']GnRH, essentially as described for rat pituitary GnRH receptor (10). Antibodies were prepared in rabbits by immunization with these receptor preparations. The antiserum to the receptor did not inhibit '251-B~~erelin binding to both rat and bovine membrane preparations, even when the membranes were preincubated with antiserum for up to 6 h or when high concentrations (1:5 dilution) of dialyzed antiserum were used (data not shown). The antibodies were dialyzed prior t o preincubation with the membranes, since it has been shown that mono and divalent cations reduce the binding of the hormone to the membrane-associated and solubilized receptors (10,18). This result suggests that the anti-receptor antibodies are not directed against the hormone binding site of the receptor. The interaction of the antibodies with bovine and rat GnRH receptors was demonstrated by immunoprecipitation of the 1251-labeled hormone-receptor complex (Fig. 1). Solubilized receptor from bovine (Fig. 1, A to D ) or rat (Fig. 1, E and F ) pituitary was preincubated (2.5 h a t 4 "C) with 1 0 " ' M of lZ5I-Buserelin (calculated occupancy of the receptors by the ligand is approximately 15%) and subsequently incubated (18 h at 4 "C) with or without serum. The binding was either terminated by addition of dextran-coated charcoal (Fig. 1, A and  E ) or the hormone-receptor-antibody complex was precipitated with Cowan I strain of S. aurew (Fig. 1, B and F ) . Under these experimental conditions (1:10 dilution of antibody) about 75% of the bovine GnRH receptor and 64% of the rat GnRH receptor can be precipitated with the antibodies as compared to the dextran-coated charcoal binding assay. The higher percentage obtained with the dextran-coated charcoal binding assay, which is based on absorption of the free '251-B~serelin, may suggest that either the antibodies recognize one population of the receptors (10) or that this technique is more efficient than immunoprecipitation of the hormonereceptor complex. Supportive evidence for the second possibility is that the charcoal technique is also more efficient when compared t o receptor precipitation by the polyethylene glycol method (lo).' When anti-GnRH receptor serum was replaced by preimmune serum or buffer alone (Fig. 1, C and D, respectively), almost no '251-Buserelin-labeled receptor was precipitated. In addition, the antibody does not bind to 1251-Buserelin in the absence of the GnRH receptor (data not shown). These results indicate a direct interaction between the antiserum to the receptor and GnRH receptors from both bovine and rat pituitaries.
A specific interaction between the antibodies and the receptors was also demonstrated by immunoprecipitation of lZ5I-labeled purified receptors and photoaffinity-labeled receptors from both rat and bovine pituitary preparations (Fig. 2). in the identification of a single specific protein that has an apparent molecular mass of 60,000 daltons. Thus, the biochemical (15) and immunological evidence suggests that the structure of the GnRH receptor is highly conserved.
The effect of antibody binding to the receptor on its biological activity was next investigated. The inhibition of GnRHstimulated LH release as a function of antibody dilution is shown in Fig. 3. While a 1:104 dilution of the anti-receptor serum did not inhibit GnRH-stimulated LH release from cultured rat pituitary cells, higher concentrations inhibited the release in a dose-dependent manner, to a maximal level of about 90% inhibition. This inhibitory effect was abolished by more than 99% when the anti-receptor serum was preincubated with goat antiserum to rabbit IgG and the resulting precipitate was removed (data not shown). The amount of LH released from pituitary cells incubated (6 h at 37 "C) with the same dilutions of anti-receptor antibodies alone was not significantly different from basal LH release (760 & 140 ng/ ml). Preimmune serum had no significant effect on either To further analyze whether the anti-receptor antibodies' effect on GnRH-stimulated LH release is receptor-mediated, we examined the effect of antibodies on LH release stimulated by high K' concentration (62 mM). As shown in Table I, K'stimulated LH release was about 40% of the release stimulated by GnRH. However, although the anti-receptor antibodies significantly inhibited GnRH-stimulated LH release, they were ineffective when the stimulation was induced by high K' concentration, confirming that the antibody effect on GnRH-stimulated LH release is mediated through the GnRH receptor.
Several studies have shown that anti-receptor antisera may act as antagonists of hormone action (19)(20)(21)(22). In some other hormonal systems, anti-receptor antisera have hormone-like activity, e.g. insulin (23, 24), due to receptor cross-linking. Recently, the conversion of a GnRH antagonist to an agonist, using specific anti-hormone antibodies, suggests that receptor cross-linking, as such, is sufficient to induce LH release (25)(26)(27). Our findings indicate that the anti-receptor antibodies do not have GnRH-like activity, suggesting that they do not cause sufficient cross-linking of the receptors to induce a signal for LH release. Similar to observations with the 8adrenergic receptor (28), these antibodies inhibit GnRH-stimulated LH release by interacting with domains of the receptor distal to the hormone binding site. This suggests that the antibodies interfere with the coupling between the receptor and the effector system in pituitary cells and thereby inhibit LH release.
In summary, we report here the production and characterization of antibodies to the GnRH receptor. These antibodies provide a useful tool for further studies on the structure, function, and biosynthesis of the GnRH receptor.