Molecular and Cellular Endocrinology, 89 (1992) 33-38 0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/92/$05.00
33
MOLCEL 02849
Binding of sex-hormone-binding globulin (SHBG) to testicular membranes and solubilized receptors C.S. Porto a, L.C. Abreu aj G.L. Gunsalus b and C.W. Bardin b a Department of Pharmacology, Endocrinology Section, Escola Paul&a de Medicina, Go Paul0 04044, Brazil, and ’ The Population Council, New York, NY 10021, USA
(Received 17 March 1992; accepted 10 July 1992)
Key words: Cell receptor; Sex-hormone-binding
globulin; Androgen-binding
protein; Membrane; (Rat testis)
Summary
Sex-hormone-binding globulin (SHBG) binds to a specific protein on the surface of prostate, epididymis, and a human breast cancer cell line (MCF-7), and is internalized by these cells. The present study demonstrated specific binding of SHBG to receptor on membranes prepared from rat testes. The binding was saturable, specific, and time and temperature dependent. Scatchard analysis of these binding studies suggested that SHBG binds to a single class of sites on testicular membranes with a K, at 37°C of 5 x 10e8 M and a binding capacity of 30 * 0.6 pmol/mg protein. These binding characteristics are similar to the SHBG receptor on human prostate and MCF-7 cells. Solubilization of the receptor resulted in a 5-fold increase in its binding capacity (158 f 0.3 pmol/mg protein) and a IO-fold decrease in binding affinity (Kd at 37°C = 6.5 X lo-’ MI. The apparent molecular weight of the testicular SHBG receptor, as estimated by gel filtration, was M, = 174,000. Conclusion: a specific binding site for SHBG was identified on testicular membranes. This binding site has been tentatively identified as a SHBG receptor based on its physical properties in testicular membrane preparations and following solubilization.
Introduction
The binding of steroid hormones to extracellular proteins plays an important role in their transport, distribution and biology. In the blood of many species steroid hormones are bound to specific high affinity steroid hormone-binding proteins, sex-hormone-binding globulin (SHBG) and corticosteroid-binding globulin (CBG) (Bardin et al., 1981; Siiteri et al., 1982a; Westphal, 1983). For years, it was believed that the only function of these proteins is to regulate the concentration of free steroids in plasma. However, a number of reports have provided evidence for the presence of intracellular SHBG and CBG in tissues not thought to produce these proteins @n-din and Petra, 1980; Sinnecker et al., 1988, 1990). Although intracellular
Correspondence to: C.W. Bardin, The Population Council, 1230 York Avenue, New York, NY 10021, USA. Abbreviations: SHBG, sex-hormone-binding globulin; ABP, androgen-binding protein; CBG, corticosteroid-binding globulin; TC, tyramine-cellobiose; CHAPS, 3-(3-cholamidopropyl)-dimethylammonioj-1-propanesulfonate.
immunoreactive SHBG has been found in monkey testis (Bordin and Petra, 19801, this immunoreactive product could not be distinguished from SHBG (ABP) produced by these cells. However, in vivo studies showing that rabbit [3H]SHBG infused into rats was rapidly cleared by testis and prostate, are consistent with cellular uptake (Sakiyama et al., 1988). The fact that SHBG (or ABP) binds specifically to the plasma membrane of decidual endometrium (Avvakumov et al., 19861, epididymis (De Besi et al., unpublished observations), human prostate (Hryb and Rosner, 1985; Hryb et al., 19891, a prostate tumor cell line, LNCaP (Nakhla et al., 19901, and a human breast cancer cell line, MCF-7 (Port0 et al., 19911, and that many of these cells internalize SHBG suggested that SHBG receptors might also be present on testicular membranes. The present study was designed to investigate this possibility. Materials
and methods
Ligand Rabbit SHBG was purified by affinity chromatography (Kotite et al., 19821, labeled with a radioiodinated
34
tyramine-cellobiose (TC) (Sigma Chemicals Co., St. Louis, MO, USA) adduct as described by Pittman et al. (1983). The labeled protein was purified by size exclusion on Bio-gel P-6 (200-400 mesh; Bio-Rad Laboratories, Richmond, CA, USA). The radiolabeled SHBG was greater than 97% precipitable by 10% trichloroacetic acid plus 2% phosphotungstic acid and bound quantitatively to an androgen affinity column (Port0 et al., 1991). The specific activities were 1.2-1.6 x 10” cpm/pg protein. Ligand was used without added testosterone. This ligand has been used previously to demonstrate receptors on MCF-7 cells (Port0 et al., 1991). Testicular membranes and solubilization of SHBG receptor
Testes were obtained from albino rats (Wistar, 2BAW colony) (Vale, 1949) at 60 days of age, and the plasma membranes were prepared by differential centrifugation. All subsequent procedures were carried out at 4°C in solutions supplemented with protease inhibitors (5 mM N-ethylmaleimide and 10 mM phenylmethylsulfonyl fluoride) (Sigma Chemical Co.). Testes were placed in buffer A (0.3 M sucrose in 25 mM Tris-HCI, pH 7.4) and homogenized with a Brinkmann Polytron homogenizer using two 20 s bursts. The homogenate was serially centrifuged at 1000 X g, 15,000 X g, and 100,000 X g. The final 100,000 X g pellet was ‘resuspended in buffer B (25 mM Tris-HCl, pH 7.4, 10 mM CaCl,) and either stored at -70°C for binding experiments or used immediately for preparation of solubilized SHBG receptor. In preliminary studies membranes were treated with various concentrations of Triton X-100, Nonidet P-40 (NP-40) and CHAPS (Sigma Chemical Co.). Both Triton X-100 and NP-40 released membrane proteins, but receptor binding activity was lost at -70°C. By contrast, CHAPS solubilized membrane protein and preserved binding activity. The optimal procedure for solubilizing the receptor used membranes (8-10 mg protein/ml) resuspended in buffer B containing enzyme inhibitor, 10% (v/v) glycerol (Fisher, Waltham, MA, USA) and 10 mM CHAPS. The suspensions were stirred at 4°C for 1 h and centrifuged at 100,000 x g for 1 h. The supernatants, containing the SHBG receptor, could be stored at - 70°C for 24 h without loss of activity. Membrane binding assay
Binding experiments were conducted in 1.5 ml Eppendorf polypropylene centrifuge tubes, that had been pretreated for 24 h at 37°C with 1 ml buffer C (25 mM Tris-HCI, pH 7.4, 10 mM CaCl,) containing 0.1% bovine serum albumin (Sigma Chemical Co.) to minimize tube blanks, plus enzyme inhibitors. Preliminary experiments were conducted to determine the effect of
increasing concentrations of labeled SHBG on binding by testicular membranes. These studies showed that saturable binding occurred with 2.3 nM SHBG by 4 h. This concentration was used in all subsequent experiments. [iz51]TC-SHBG (2.3 nM, 75,000-100,000 cpm/tube) was incubated with membranes (0.3 mg membrane protein/ml) in buffer C for periods of l-6 h at 37°C in the absence and presence (lOOO-fold excess) of radioinert SHBG in a final volume of 0.3 ml. All assays were performed in duplicate. Bound [‘2sI]TC-SHBG was separated from free ligand by centrifugation at 13,000 x g for 10 min at 4°C. The supernatant was aspirated and the pellet was washed with buffer B and recentrifuged for 10 additional minutes. The final pellet was counted in an Abbott y-scintillation counter. Nonspecific binding was 15-20% of total binding. Assay of solubilized receptor
Immediately before the assay, the soluble extracts were diluted to 0.5 mg protein/ml with buffer B (4°C) containing 10% (v/v) glycerol and 10 mM CHAPS supplemented with protease inhibitors. Binding experiments with the CHAPS-solubilized receptor were performed by incubating aliquots of the diluted extracts (0.1 ml) with assay mixtures identical to those used for the membrane binding assay. Bound [ ‘251]TC-SHBG was separated from free ligand by precipitating the receptor-SHBG complex by sequential addition of 0.5 ml of buffer B containing 1 mg/ml bovine y-globulin (Sigma Chemical Co.) and 0.5 ml of 12% (w/v) polyethylene glycol (PEG-8000, Fisher, Waltham, MA, USA). After mixing, the tubes were incubated for 10 min and centrifuged at 13,000 X g for 10 min at 4°C. Supernatants were aspirated and the pellets were suspended in 1 ml of buffer B containing 10% (v/v) glycerol and 10 mM CHAPS. Polyethylene glycol (0.5 ml) was added and the tubes vortexed. After 10 min the tubes were centrifuged, the supernatant aspirated and the radioactivity in the pellet was measured. Under these conditions the nonspecific binding was 2022% of total binding. Gel filtration of the CHAPS-solubilized receptor 10 ml of the CHAPS-solubilized testicular receptor
(3-4 mg protein/ml) was applied to a column of Sepharose CL-6B (Pharmacia, Philadelphia, PA, USA) (1.5 x 70 cm) that previously was equilibrated with 10 mM CHAPS buffer at 4°C and eluted with the same buffer at a flow rate of 20 ml/h. Alternate fractions of 6 ml each were assayed for protein concentration and specific binding of [‘251]TC-SHBG as described above. The column was calibrated with blue dextran (void volume) and protein standards (thyroglobulin, M, = 669,bOO; ferritin, M, = 440,000; catalase, M, = 232,000; aldolase, M, = 158,000). The molecular weight of the
35
SHBG receptor was determined by interpolation from a plot of K, versus log molecular weight (Laurent and Killander, 1964). Other analytical procedures
Binding analysis was performed by weighted nonlinear least squares curve fitting as described previously (Rolland et al., 1976; Munson and Rodbard, 1980). Protein concentrations in the membrane preparations and soluble extracts were determined according to Bradford (1976). Results Binding of [‘251]TC-SHBG membrane
to SHBG receptor in testis
The binding of SHBG to testicular membranes is shown in Fig. 1, specific binding was the difference between total binding, which was 2528% of added [‘251]TC-SHBG at 37°C and nonspecific binding determined in the presence of a lOOO-fold excess of unlabeled SHBG. The time-course study showed that the specific binding reached a plateau by 4 h at 37°C. The specific binding at 37°C was 5fold higher than at 4°C showing a distinct temperature dependence (Fig. 1). Analysis of SHBG binding to testicular membranes at 37°C (Fig. 2) yielded a dissociation constant (K,) of 5 x lo-* M and a binding capacity of 30 f 0.6 pmol/mg membrane protein (mean f SEMI. The specificity of binding was established by comparing the abilities of unlabeled SHBG, transferrin, and a,-acid glycoprotein to compete with labeled SHBG for binding to testicular membranes. The binding of the radioactive ligand
;
i
I I, 5
I
I
I
10
15
20
I
25
BOUND (nM)
Fig. 2. Displacement by unlabeled SHBG of [‘251JTC-SHBG bound to testis membranes. [‘251~C-SHBG (2.3 nM) was incubated with membranes (0.3 mg membrane protein/ml) in the presence of unlabeled TeBG (O-7 PM) for 4 h at 3PC. Data from a single experiment are plotted in Scatchard coordinates. Curved line is the computer-generated best fit of raw data (01 to a high affinity, low capacity binding species (specific binding, solid straight line) and a low affinity, high capacity species (nonspecific binding, dashed straight line) (Rolland et al., 1976). Five experiments were performed; total binding in the absence of unlabeled SHBG was - 20% of total counts; nonspecific binding was - 15% of total binding.
was reduced 70% by SHBG. The other proteins did not compete for testosterone estradiol binding globulin (TeBG) binding sites (Fig. 3). Since the binding proper-
A
TIME (h) Fig. 1. Time-course of specific binding of [rz51~C-SHBG to testicular membranes. [ 1251JTC-SHBG (2.3 nM) was incubated with membranes (0.3 mg membrane protein/ml) with or without a lOOO-fold excess of unlabeled SHBG at 37°C (0) and 4°C CO). Specific binding is the difference between total binding (25-28% of added [‘2513rCSHBG) and nonspecific binding determined in the presence of lOOO-fold excess of unlabeled SHBG. Values expressed as mean+ SEM (n = 5 experiments).
A
.i
$0
6&o
COMPETITOR(P~‘) Fig. 3. Specificity of [1251JTC-SHBG binding to testicular membranes, in absence and presence of increasing amounts of unlabeled SHBG (o), transfertin (A) and a,-acid glycoprotein (MI for 4 h at 37°C. B, = binding of [‘251]TC-SHBG in the absence of competitor; B = binding of [ lZSIJTC-SHBG in the presence of competitor. Values expressed as mean f SEM (n = 5 experiments).
tics of these sites on testicular membranes are similar to SHBG receptors on prostate membranes (Rosner, 19901 the former are also tentatively designated SHBG receptors. Solubilization of the SHBG receptor
The ability of a variety of detergents (Triton X-100, NP-40 and CHAPS) to solubilize SHBG receptors from testicular membranes was evaluated. CHAPS was found to be efficient in solubilizing of the receptor and did not interfere with binding. To determine the optimum concentrations of CHAPS for solubilization of SHBG receptors, testis membranes (8 mg protein/ml) were incubated at 4°C for 1 h with various concentrations of this detergent (data not shown). Based on these studies 10 mM CHAPS was used in all studies of the soluble receptor. Various concentrations of glycerol were tested in order to improve the stability of the solubilized receptors. 10% glycerol provided the best protection for receptor; during 24 h at - 70°C there was no loss of activity (data not shown). The binding of [‘251]TC-SHBG to solubilized receptor was studied. Analysis of the competition data with SHBG revealed one population of binding sites (Fig. 4). A K, of 6.5 x lo-’ M and a binding capacity of
8
1
6
4
10
30
50
FRACTION
70
90
NUMBER
Fig. 5. Distribution of total protein (01 and specific binding of [‘2sI]TC-SHBG (0) after fractionation of CHAPS-solubilized receptor on Sepharose CL-6B. Fractions containing material capable of binding to [“‘l]TC-SHBG eluted with an apparent M, = 174,000. For protein standards see Materials and methods.
158 + 0.3 pmol/mg mated.
protein (mean k SEM) were esti-
I
Determination of the molecular weight of the solubilized SHBG receptor
The apparent molecular weight of the testicular SHBG receptor was estimated by fractionation of solubilized membrane preparations on Sepharose CL-6B at 4°C. When the resultant fractions were assayed for their ability to bind [12511TC-SHBG the receptor binding activity eluted in a symmetrical peak which was separated from the major protein peak by about 12 ml (Fig. 5). Calibration of the column with protein standards allowed estimation of the molecular weight from experiments illustrated in Fig. 5. Six experiments resulted in a molecular weight of 174,000 -t 2400 (mean f SEMI. Discussion BOUND Wl Fig. 4. Displacement by unlabeled SHBG of [“‘I]TC-SHBG bound to soluble receptor from testis membranes. [‘2’I]TC-SHBG (2.3 nM) was incubated with CHAPS-solubilized receptor (0.3-0.4 mg protein/ml) in the presence of increasing amounts of unlabeled SHBG (O-7 PM) for 4 h at 37°C. Data from a single experiment are plotted in Scatchard coordinates. Curved line is the computer-generated best fit of raw data (01 to a high affinity, low capacity binding species (specific binding, solid straight line) and a low affinity, high capacity species (nonspecific binding, dashed straight line) (Rolland et al., 1976). Four experiments were performed; total binding in the absence of unlabeled SHBG was - 9% of total counts; nonspecific binding was - 20% of total binding.
The results demonstrate specific SHBG binding sites on testicular membranes. These binding sites are similar to SHBG receptors described in other tissues that are capable of internalizing the ligand and of stimulating CAMP accumulation (Rosner, 1990). As a consequence, the SHBG binding sites in testes have also been termed receptors. The presence of a SHBG in blood of adult rat has been difficult to measure using conventional binding assays. Therefore one may not have expected SHBG receptors in rat tissues. Following the demonstration of ABP production in rat testes
37
and its secretion into the male reproductive tract, it was shown that this protein is released into blood of male rats (Gunsalus et al., 1978). Subsequently it was shown that SHBG and ABP, which are traditionally thought of as distinct hepatic and testicular proteins, respectively, were in reality products of the same gene (Joseph et al., 1987; Bardin et al., 1988; Petra et al., 1988). There is roughly 70% identity between the amino acid sequences of the rat and human proteins. Specific membrane-binding sites for SHBG were demonstrated by Strel’chyonok and Awakumov (19841 using human decidual endometrium; the K, of SHBG membrane complex was 3 x lo-” M at 4°C. Hryb et al. (1985, 1989) identified a SHBG receptor in human prostate membranes. High and low affinity sites were identified with K, values of 1.5 x 10-s M and 8.1 X 10e6 M, respectively; the solubilization of these receptors decreased their K, values to 1.4 X 10m9 M and 1 X lo-’ M, respectively. Our laboratory has studied the binding of rabbit SHBG to MCF-7 cells and shown K, values of 3 X 10-s M and 2 X lo-’ M for membrane and solubilized receptors, respectively (Port0 et al., 1992). The present report demonstrates one class of specific sites for SHBG on testicular membranes. The binding sites are saturable, specific, and have high affinity for SHBG (Kd = 5 X lo-’ M). Solubilization lowers the affinity for SHBG receptors on testis membranes (K, = 6.5 X lo-’ Ml. The reason for the large variation in the K, for the SHBG receptors noted above could relate to tissue-specific differences in the factors that regulate the conformation and stability of receptor and to variations in methods used in different laboratories. Recently, the SHBG receptor of human prostate membranes was solubilized and found to have an apparent M, of 167,000 as estimated by gel filtration (Hryb et al., 1989). The solubilized SHBG receptor from testicular membrane has similar size 04, = 174,000). A sequence in SHBG that interacts with the receptor was identified following trypsinization of SHBG and isolation of the resulting peptides for competition assay and sequence determination (Khan et al., 1990). the decapeptide corresponding to Furthermore, residues 48-57 inhibited SHBG binding to the SHBG receptor solubilized from human prostatic membranes (Khan et al., 1990). That this sequence is identical in the human and rat proteins suggests the binding site of the receptor may also be conserved across species (Bardin et al., 1988). In an effort to demonstrate a function for SHBG receptors, Nakhla et al. (1990) studied LNCaP cells, which possess membrane-associated binding sites for SHBG. Membrane-bound SHBG alone elicited little if any increase in intracellular CAMP concentration, but when membrane-bound SHBP subsequently bound di-
hydrotestosterone (DHT), CAMP levels increased N 1.7-fold. Even though DHT activation of membrane-bound SHBG is not of primary importance for androgen receptor-mediated control of cell proliferation in LNCaP-FGC cells (Damassa et al., 19911, the generation of CAMP via the SHBG receptor provides a potential mechanism for the protein kinase A pathway to interact with transcriptional processes that are mediated via androgen receptor (Damassa et al., 1991). Binding of steroid to SHBG to form a SHBG-steroid complex inhibits subsequent SHBG binding to cell receptors, whereas unliganded SHBG binds receptor freely and SHBG in the receptor-SHBG complex can then bind steroid (Hryb et al., 1990). Since some SHBG is always unliganded at physiological testosterone concentrations, it is free to associate with its receptor. Free steroid from plasma can then bind to SHBGreceptor complex and stimulate CAMP accumulation (Rosner, 1990). In conclusion, our results provide evidence for the presence of SHBG receptors on testicular membranes. Since, in the rat, Sertoli cells are the major site of synthesis of this extracellular binding protein, the presence of its receptor in the testis may provide a means for autocrine and/or paracrine regulation of testicular cells. Acknowledgements We gratefully acknowledge the assistance of Linda McKeiver in the preparation of the manuscript. These studies were supported by The Rockefeller Foundation, New York, USA, Funda@o de Amaaro a Pesquisa do Estado de SCo Paulo (FAESP), SCo Paulo, Brazil; Conselho National Desenvolvimento Cientifico e Tecnologico (CNPq), DF, Brazil; and The National Institutes of Health, Grant HD-13541. References Awakumov, G.V., Zhuk, N.I. and Strel’chyonok, O.A. (1986) Biochim. Biophys. Acta 881, 489-498. Bardin, C.W., Musto, N., Gunsalus, G., Kotite, N., Cheng, S.-L., Larrea, F. and Becker, R. (1981) Annu. Rev. Physiol. 43, 1899198. Bardin, C.W., Gunsalus, G.L., Musto, N.A., Cheng, C.Y., Reventos, J., Smith, C., Underhill, D.A. and Hammond, G. (1988) J. Steroid Biochem. 30, 131-139. Bordin, S. and Petra, P.H. (1980) Proc. Nat]. Acad. Sci. USA 77, 5678-5682. Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. Damassa, D.A., Lin, T.-M., Sonnenschein, C. and Soto, A.M. (1991) Endocrinology 129, 75-84. Gunsalus, G.L., Musto, N.A. and Bardin, C.W. (1978) Science 200, 65-66. Hryb, D.J., Khan, MS. and Rosner, W. (1985) Biochem. Biophys. Res. Commun. 128, 432-440. Hryb, D.J., Khan, M.S., Romas, N.A. and Rosner, W. (1989) J. Biol. Chem. 264, 5378-5383.
3x Hryb, D.J., Khan, MS., Romas, N.A. and Rosner, W. (1990) J. Biol. Chem. 265, 6048-6054. Joseph, D.R., Hall, S.H. and French, F.S. (1987) Proc. Natl. Acad. Sci. USA 84, 339-343. Khan, MS., Hryb, D.J., Hashim, G.A., Romas, N.A. and Rosner, W. (1990) J. Biol. Chem. 265, 18362-18365. Kotite, N.J. and Musto, N.A. (1982) J. Biol. Chem. 257, 5118-5124. Laurent, T.C. and Killander, J. (1964) J. Chromatogr. 14, 3177330. Munson, P.J. and Rodbard, D. (1980) Anal. Biochem. 107, 220-239. Nakhla, A.M., Khan, MS. and Rosner, W. (1990) J. Clin. Endocrinol. Metab. 71, 398-404. Petra, P.H., Que, B.C., Namkung, P.C., Ross, J.B.A., Charbonneau, H., Walsh, K.A., Griffin, P.R., Shabanowitz, J. and Hunt, D.F. (1988) Ann. NY Acad. Sci. 538, 10-24. Pittman, R.C., Carew, T.E., Glass, C.K., Green, S.R., Taylor, C.A.J. and Attie, A.D. (1983) Biochem. J. 212, 791-800. Porto, C.S., Gunsalus, G.L., Bardin, C.W., Phillips, D.M. and Musto, N.A. (1991) Endocrinology 129, 436-445.
Porto, C.S., Musto, N.A.. Bardin, C.W. and Gunsalus, G.L. (1002) Endocrinology 130, 2931-2936. Rolland, R., Gunsalus, G.L. and Hammond, J.M. (1976) Endocrinology 98, 1083-1091. Rosner, W. (1990) Endocr. Rev. I I, 80-91. Sakiyama, R., Pardridge, W.M. and Musto, N.A. (1988) J. Clin. Endocrinol. Metab. 67, 98-103. Siiteri, P.K., Murai, J.T., Hammond, G.H., Nisker, J.A., Raymore, W.J. and Kuhn, R.W. (1982) Recent Prog. Horm. Res. 38, 457510. Sinnecker, G., Hiort, O., Mitze, M., Donn, F. and Neuman, S. (1988) Steroids 52, 335-336. Sinnecker, G., Hiort, Q., Kwan, P.W. and DeLellis, R.A. (1990) Horm. Metab. Res. 22. 47-50. Strel’chyonok, O.A., Awakumov, G.V. and Survilo, L.I. (1984) Biochim. Biophys. Acta 802, 4.59-466. Vale, J.R. (1949) Cienc. Cult. 1, 156. Westphal, U. (1983) J. Steroid Biochem. 19, l-15.
33
MOLCEL 02849
Binding of sex-hormone-binding globulin (SHBG) to testicular membranes and solubilized receptors C.S. Porto a, L.C. Abreu aj G.L. Gunsalus b and C.W. Bardin b a Department of Pharmacology, Endocrinology Section, Escola Paul&a de Medicina, Go Paul0 04044, Brazil, and ’ The Population Council, New York, NY 10021, USA
(Received 17 March 1992; accepted 10 July 1992)
Key words: Cell receptor; Sex-hormone-binding
globulin; Androgen-binding
protein; Membrane; (Rat testis)
Summary
Sex-hormone-binding globulin (SHBG) binds to a specific protein on the surface of prostate, epididymis, and a human breast cancer cell line (MCF-7), and is internalized by these cells. The present study demonstrated specific binding of SHBG to receptor on membranes prepared from rat testes. The binding was saturable, specific, and time and temperature dependent. Scatchard analysis of these binding studies suggested that SHBG binds to a single class of sites on testicular membranes with a K, at 37°C of 5 x 10e8 M and a binding capacity of 30 * 0.6 pmol/mg protein. These binding characteristics are similar to the SHBG receptor on human prostate and MCF-7 cells. Solubilization of the receptor resulted in a 5-fold increase in its binding capacity (158 f 0.3 pmol/mg protein) and a IO-fold decrease in binding affinity (Kd at 37°C = 6.5 X lo-’ MI. The apparent molecular weight of the testicular SHBG receptor, as estimated by gel filtration, was M, = 174,000. Conclusion: a specific binding site for SHBG was identified on testicular membranes. This binding site has been tentatively identified as a SHBG receptor based on its physical properties in testicular membrane preparations and following solubilization.
Introduction
The binding of steroid hormones to extracellular proteins plays an important role in their transport, distribution and biology. In the blood of many species steroid hormones are bound to specific high affinity steroid hormone-binding proteins, sex-hormone-binding globulin (SHBG) and corticosteroid-binding globulin (CBG) (Bardin et al., 1981; Siiteri et al., 1982a; Westphal, 1983). For years, it was believed that the only function of these proteins is to regulate the concentration of free steroids in plasma. However, a number of reports have provided evidence for the presence of intracellular SHBG and CBG in tissues not thought to produce these proteins @n-din and Petra, 1980; Sinnecker et al., 1988, 1990). Although intracellular
Correspondence to: C.W. Bardin, The Population Council, 1230 York Avenue, New York, NY 10021, USA. Abbreviations: SHBG, sex-hormone-binding globulin; ABP, androgen-binding protein; CBG, corticosteroid-binding globulin; TC, tyramine-cellobiose; CHAPS, 3-(3-cholamidopropyl)-dimethylammonioj-1-propanesulfonate.
immunoreactive SHBG has been found in monkey testis (Bordin and Petra, 19801, this immunoreactive product could not be distinguished from SHBG (ABP) produced by these cells. However, in vivo studies showing that rabbit [3H]SHBG infused into rats was rapidly cleared by testis and prostate, are consistent with cellular uptake (Sakiyama et al., 1988). The fact that SHBG (or ABP) binds specifically to the plasma membrane of decidual endometrium (Avvakumov et al., 19861, epididymis (De Besi et al., unpublished observations), human prostate (Hryb and Rosner, 1985; Hryb et al., 19891, a prostate tumor cell line, LNCaP (Nakhla et al., 19901, and a human breast cancer cell line, MCF-7 (Port0 et al., 19911, and that many of these cells internalize SHBG suggested that SHBG receptors might also be present on testicular membranes. The present study was designed to investigate this possibility. Materials
and methods
Ligand Rabbit SHBG was purified by affinity chromatography (Kotite et al., 19821, labeled with a radioiodinated
34
tyramine-cellobiose (TC) (Sigma Chemicals Co., St. Louis, MO, USA) adduct as described by Pittman et al. (1983). The labeled protein was purified by size exclusion on Bio-gel P-6 (200-400 mesh; Bio-Rad Laboratories, Richmond, CA, USA). The radiolabeled SHBG was greater than 97% precipitable by 10% trichloroacetic acid plus 2% phosphotungstic acid and bound quantitatively to an androgen affinity column (Port0 et al., 1991). The specific activities were 1.2-1.6 x 10” cpm/pg protein. Ligand was used without added testosterone. This ligand has been used previously to demonstrate receptors on MCF-7 cells (Port0 et al., 1991). Testicular membranes and solubilization of SHBG receptor
Testes were obtained from albino rats (Wistar, 2BAW colony) (Vale, 1949) at 60 days of age, and the plasma membranes were prepared by differential centrifugation. All subsequent procedures were carried out at 4°C in solutions supplemented with protease inhibitors (5 mM N-ethylmaleimide and 10 mM phenylmethylsulfonyl fluoride) (Sigma Chemical Co.). Testes were placed in buffer A (0.3 M sucrose in 25 mM Tris-HCI, pH 7.4) and homogenized with a Brinkmann Polytron homogenizer using two 20 s bursts. The homogenate was serially centrifuged at 1000 X g, 15,000 X g, and 100,000 X g. The final 100,000 X g pellet was ‘resuspended in buffer B (25 mM Tris-HCl, pH 7.4, 10 mM CaCl,) and either stored at -70°C for binding experiments or used immediately for preparation of solubilized SHBG receptor. In preliminary studies membranes were treated with various concentrations of Triton X-100, Nonidet P-40 (NP-40) and CHAPS (Sigma Chemical Co.). Both Triton X-100 and NP-40 released membrane proteins, but receptor binding activity was lost at -70°C. By contrast, CHAPS solubilized membrane protein and preserved binding activity. The optimal procedure for solubilizing the receptor used membranes (8-10 mg protein/ml) resuspended in buffer B containing enzyme inhibitor, 10% (v/v) glycerol (Fisher, Waltham, MA, USA) and 10 mM CHAPS. The suspensions were stirred at 4°C for 1 h and centrifuged at 100,000 x g for 1 h. The supernatants, containing the SHBG receptor, could be stored at - 70°C for 24 h without loss of activity. Membrane binding assay
Binding experiments were conducted in 1.5 ml Eppendorf polypropylene centrifuge tubes, that had been pretreated for 24 h at 37°C with 1 ml buffer C (25 mM Tris-HCI, pH 7.4, 10 mM CaCl,) containing 0.1% bovine serum albumin (Sigma Chemical Co.) to minimize tube blanks, plus enzyme inhibitors. Preliminary experiments were conducted to determine the effect of
increasing concentrations of labeled SHBG on binding by testicular membranes. These studies showed that saturable binding occurred with 2.3 nM SHBG by 4 h. This concentration was used in all subsequent experiments. [iz51]TC-SHBG (2.3 nM, 75,000-100,000 cpm/tube) was incubated with membranes (0.3 mg membrane protein/ml) in buffer C for periods of l-6 h at 37°C in the absence and presence (lOOO-fold excess) of radioinert SHBG in a final volume of 0.3 ml. All assays were performed in duplicate. Bound [‘2sI]TC-SHBG was separated from free ligand by centrifugation at 13,000 x g for 10 min at 4°C. The supernatant was aspirated and the pellet was washed with buffer B and recentrifuged for 10 additional minutes. The final pellet was counted in an Abbott y-scintillation counter. Nonspecific binding was 15-20% of total binding. Assay of solubilized receptor
Immediately before the assay, the soluble extracts were diluted to 0.5 mg protein/ml with buffer B (4°C) containing 10% (v/v) glycerol and 10 mM CHAPS supplemented with protease inhibitors. Binding experiments with the CHAPS-solubilized receptor were performed by incubating aliquots of the diluted extracts (0.1 ml) with assay mixtures identical to those used for the membrane binding assay. Bound [ ‘251]TC-SHBG was separated from free ligand by precipitating the receptor-SHBG complex by sequential addition of 0.5 ml of buffer B containing 1 mg/ml bovine y-globulin (Sigma Chemical Co.) and 0.5 ml of 12% (w/v) polyethylene glycol (PEG-8000, Fisher, Waltham, MA, USA). After mixing, the tubes were incubated for 10 min and centrifuged at 13,000 X g for 10 min at 4°C. Supernatants were aspirated and the pellets were suspended in 1 ml of buffer B containing 10% (v/v) glycerol and 10 mM CHAPS. Polyethylene glycol (0.5 ml) was added and the tubes vortexed. After 10 min the tubes were centrifuged, the supernatant aspirated and the radioactivity in the pellet was measured. Under these conditions the nonspecific binding was 2022% of total binding. Gel filtration of the CHAPS-solubilized receptor 10 ml of the CHAPS-solubilized testicular receptor
(3-4 mg protein/ml) was applied to a column of Sepharose CL-6B (Pharmacia, Philadelphia, PA, USA) (1.5 x 70 cm) that previously was equilibrated with 10 mM CHAPS buffer at 4°C and eluted with the same buffer at a flow rate of 20 ml/h. Alternate fractions of 6 ml each were assayed for protein concentration and specific binding of [‘251]TC-SHBG as described above. The column was calibrated with blue dextran (void volume) and protein standards (thyroglobulin, M, = 669,bOO; ferritin, M, = 440,000; catalase, M, = 232,000; aldolase, M, = 158,000). The molecular weight of the
35
SHBG receptor was determined by interpolation from a plot of K, versus log molecular weight (Laurent and Killander, 1964). Other analytical procedures
Binding analysis was performed by weighted nonlinear least squares curve fitting as described previously (Rolland et al., 1976; Munson and Rodbard, 1980). Protein concentrations in the membrane preparations and soluble extracts were determined according to Bradford (1976). Results Binding of [‘251]TC-SHBG membrane
to SHBG receptor in testis
The binding of SHBG to testicular membranes is shown in Fig. 1, specific binding was the difference between total binding, which was 2528% of added [‘251]TC-SHBG at 37°C and nonspecific binding determined in the presence of a lOOO-fold excess of unlabeled SHBG. The time-course study showed that the specific binding reached a plateau by 4 h at 37°C. The specific binding at 37°C was 5fold higher than at 4°C showing a distinct temperature dependence (Fig. 1). Analysis of SHBG binding to testicular membranes at 37°C (Fig. 2) yielded a dissociation constant (K,) of 5 x lo-* M and a binding capacity of 30 f 0.6 pmol/mg membrane protein (mean f SEMI. The specificity of binding was established by comparing the abilities of unlabeled SHBG, transferrin, and a,-acid glycoprotein to compete with labeled SHBG for binding to testicular membranes. The binding of the radioactive ligand
;
i
I I, 5
I
I
I
10
15
20
I
25
BOUND (nM)
Fig. 2. Displacement by unlabeled SHBG of [‘251JTC-SHBG bound to testis membranes. [‘251~C-SHBG (2.3 nM) was incubated with membranes (0.3 mg membrane protein/ml) in the presence of unlabeled TeBG (O-7 PM) for 4 h at 3PC. Data from a single experiment are plotted in Scatchard coordinates. Curved line is the computer-generated best fit of raw data (01 to a high affinity, low capacity binding species (specific binding, solid straight line) and a low affinity, high capacity species (nonspecific binding, dashed straight line) (Rolland et al., 1976). Five experiments were performed; total binding in the absence of unlabeled SHBG was - 20% of total counts; nonspecific binding was - 15% of total binding.
was reduced 70% by SHBG. The other proteins did not compete for testosterone estradiol binding globulin (TeBG) binding sites (Fig. 3). Since the binding proper-
A
TIME (h) Fig. 1. Time-course of specific binding of [rz51~C-SHBG to testicular membranes. [ 1251JTC-SHBG (2.3 nM) was incubated with membranes (0.3 mg membrane protein/ml) with or without a lOOO-fold excess of unlabeled SHBG at 37°C (0) and 4°C CO). Specific binding is the difference between total binding (25-28% of added [‘2513rCSHBG) and nonspecific binding determined in the presence of lOOO-fold excess of unlabeled SHBG. Values expressed as mean+ SEM (n = 5 experiments).
A
.i
$0
6&o
COMPETITOR(P~‘) Fig. 3. Specificity of [1251JTC-SHBG binding to testicular membranes, in absence and presence of increasing amounts of unlabeled SHBG (o), transfertin (A) and a,-acid glycoprotein (MI for 4 h at 37°C. B, = binding of [‘251]TC-SHBG in the absence of competitor; B = binding of [ lZSIJTC-SHBG in the presence of competitor. Values expressed as mean f SEM (n = 5 experiments).
tics of these sites on testicular membranes are similar to SHBG receptors on prostate membranes (Rosner, 19901 the former are also tentatively designated SHBG receptors. Solubilization of the SHBG receptor
The ability of a variety of detergents (Triton X-100, NP-40 and CHAPS) to solubilize SHBG receptors from testicular membranes was evaluated. CHAPS was found to be efficient in solubilizing of the receptor and did not interfere with binding. To determine the optimum concentrations of CHAPS for solubilization of SHBG receptors, testis membranes (8 mg protein/ml) were incubated at 4°C for 1 h with various concentrations of this detergent (data not shown). Based on these studies 10 mM CHAPS was used in all studies of the soluble receptor. Various concentrations of glycerol were tested in order to improve the stability of the solubilized receptors. 10% glycerol provided the best protection for receptor; during 24 h at - 70°C there was no loss of activity (data not shown). The binding of [‘251]TC-SHBG to solubilized receptor was studied. Analysis of the competition data with SHBG revealed one population of binding sites (Fig. 4). A K, of 6.5 x lo-’ M and a binding capacity of
8
1
6
4
10
30
50
FRACTION
70
90
NUMBER
Fig. 5. Distribution of total protein (01 and specific binding of [‘2sI]TC-SHBG (0) after fractionation of CHAPS-solubilized receptor on Sepharose CL-6B. Fractions containing material capable of binding to [“‘l]TC-SHBG eluted with an apparent M, = 174,000. For protein standards see Materials and methods.
158 + 0.3 pmol/mg mated.
protein (mean k SEM) were esti-
I
Determination of the molecular weight of the solubilized SHBG receptor
The apparent molecular weight of the testicular SHBG receptor was estimated by fractionation of solubilized membrane preparations on Sepharose CL-6B at 4°C. When the resultant fractions were assayed for their ability to bind [12511TC-SHBG the receptor binding activity eluted in a symmetrical peak which was separated from the major protein peak by about 12 ml (Fig. 5). Calibration of the column with protein standards allowed estimation of the molecular weight from experiments illustrated in Fig. 5. Six experiments resulted in a molecular weight of 174,000 -t 2400 (mean f SEMI. Discussion BOUND Wl Fig. 4. Displacement by unlabeled SHBG of [“‘I]TC-SHBG bound to soluble receptor from testis membranes. [‘2’I]TC-SHBG (2.3 nM) was incubated with CHAPS-solubilized receptor (0.3-0.4 mg protein/ml) in the presence of increasing amounts of unlabeled SHBG (O-7 PM) for 4 h at 37°C. Data from a single experiment are plotted in Scatchard coordinates. Curved line is the computer-generated best fit of raw data (01 to a high affinity, low capacity binding species (specific binding, solid straight line) and a low affinity, high capacity species (nonspecific binding, dashed straight line) (Rolland et al., 1976). Four experiments were performed; total binding in the absence of unlabeled SHBG was - 9% of total counts; nonspecific binding was - 20% of total binding.
The results demonstrate specific SHBG binding sites on testicular membranes. These binding sites are similar to SHBG receptors described in other tissues that are capable of internalizing the ligand and of stimulating CAMP accumulation (Rosner, 1990). As a consequence, the SHBG binding sites in testes have also been termed receptors. The presence of a SHBG in blood of adult rat has been difficult to measure using conventional binding assays. Therefore one may not have expected SHBG receptors in rat tissues. Following the demonstration of ABP production in rat testes
37
and its secretion into the male reproductive tract, it was shown that this protein is released into blood of male rats (Gunsalus et al., 1978). Subsequently it was shown that SHBG and ABP, which are traditionally thought of as distinct hepatic and testicular proteins, respectively, were in reality products of the same gene (Joseph et al., 1987; Bardin et al., 1988; Petra et al., 1988). There is roughly 70% identity between the amino acid sequences of the rat and human proteins. Specific membrane-binding sites for SHBG were demonstrated by Strel’chyonok and Awakumov (19841 using human decidual endometrium; the K, of SHBG membrane complex was 3 x lo-” M at 4°C. Hryb et al. (1985, 1989) identified a SHBG receptor in human prostate membranes. High and low affinity sites were identified with K, values of 1.5 x 10-s M and 8.1 X 10e6 M, respectively; the solubilization of these receptors decreased their K, values to 1.4 X 10m9 M and 1 X lo-’ M, respectively. Our laboratory has studied the binding of rabbit SHBG to MCF-7 cells and shown K, values of 3 X 10-s M and 2 X lo-’ M for membrane and solubilized receptors, respectively (Port0 et al., 1992). The present report demonstrates one class of specific sites for SHBG on testicular membranes. The binding sites are saturable, specific, and have high affinity for SHBG (Kd = 5 X lo-’ M). Solubilization lowers the affinity for SHBG receptors on testis membranes (K, = 6.5 X lo-’ Ml. The reason for the large variation in the K, for the SHBG receptors noted above could relate to tissue-specific differences in the factors that regulate the conformation and stability of receptor and to variations in methods used in different laboratories. Recently, the SHBG receptor of human prostate membranes was solubilized and found to have an apparent M, of 167,000 as estimated by gel filtration (Hryb et al., 1989). The solubilized SHBG receptor from testicular membrane has similar size 04, = 174,000). A sequence in SHBG that interacts with the receptor was identified following trypsinization of SHBG and isolation of the resulting peptides for competition assay and sequence determination (Khan et al., 1990). the decapeptide corresponding to Furthermore, residues 48-57 inhibited SHBG binding to the SHBG receptor solubilized from human prostatic membranes (Khan et al., 1990). That this sequence is identical in the human and rat proteins suggests the binding site of the receptor may also be conserved across species (Bardin et al., 1988). In an effort to demonstrate a function for SHBG receptors, Nakhla et al. (1990) studied LNCaP cells, which possess membrane-associated binding sites for SHBG. Membrane-bound SHBG alone elicited little if any increase in intracellular CAMP concentration, but when membrane-bound SHBP subsequently bound di-
hydrotestosterone (DHT), CAMP levels increased N 1.7-fold. Even though DHT activation of membrane-bound SHBG is not of primary importance for androgen receptor-mediated control of cell proliferation in LNCaP-FGC cells (Damassa et al., 19911, the generation of CAMP via the SHBG receptor provides a potential mechanism for the protein kinase A pathway to interact with transcriptional processes that are mediated via androgen receptor (Damassa et al., 1991). Binding of steroid to SHBG to form a SHBG-steroid complex inhibits subsequent SHBG binding to cell receptors, whereas unliganded SHBG binds receptor freely and SHBG in the receptor-SHBG complex can then bind steroid (Hryb et al., 1990). Since some SHBG is always unliganded at physiological testosterone concentrations, it is free to associate with its receptor. Free steroid from plasma can then bind to SHBGreceptor complex and stimulate CAMP accumulation (Rosner, 1990). In conclusion, our results provide evidence for the presence of SHBG receptors on testicular membranes. Since, in the rat, Sertoli cells are the major site of synthesis of this extracellular binding protein, the presence of its receptor in the testis may provide a means for autocrine and/or paracrine regulation of testicular cells. Acknowledgements We gratefully acknowledge the assistance of Linda McKeiver in the preparation of the manuscript. These studies were supported by The Rockefeller Foundation, New York, USA, Funda@o de Amaaro a Pesquisa do Estado de SCo Paulo (FAESP), SCo Paulo, Brazil; Conselho National Desenvolvimento Cientifico e Tecnologico (CNPq), DF, Brazil; and The National Institutes of Health, Grant HD-13541. References Awakumov, G.V., Zhuk, N.I. and Strel’chyonok, O.A. (1986) Biochim. Biophys. Acta 881, 489-498. Bardin, C.W., Musto, N., Gunsalus, G., Kotite, N., Cheng, S.-L., Larrea, F. and Becker, R. (1981) Annu. Rev. Physiol. 43, 1899198. Bardin, C.W., Gunsalus, G.L., Musto, N.A., Cheng, C.Y., Reventos, J., Smith, C., Underhill, D.A. and Hammond, G. (1988) J. Steroid Biochem. 30, 131-139. Bordin, S. and Petra, P.H. (1980) Proc. Nat]. Acad. Sci. USA 77, 5678-5682. Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. Damassa, D.A., Lin, T.-M., Sonnenschein, C. and Soto, A.M. (1991) Endocrinology 129, 75-84. Gunsalus, G.L., Musto, N.A. and Bardin, C.W. (1978) Science 200, 65-66. Hryb, D.J., Khan, MS. and Rosner, W. (1985) Biochem. Biophys. Res. Commun. 128, 432-440. Hryb, D.J., Khan, M.S., Romas, N.A. and Rosner, W. (1989) J. Biol. Chem. 264, 5378-5383.
3x Hryb, D.J., Khan, MS., Romas, N.A. and Rosner, W. (1990) J. Biol. Chem. 265, 6048-6054. Joseph, D.R., Hall, S.H. and French, F.S. (1987) Proc. Natl. Acad. Sci. USA 84, 339-343. Khan, MS., Hryb, D.J., Hashim, G.A., Romas, N.A. and Rosner, W. (1990) J. Biol. Chem. 265, 18362-18365. Kotite, N.J. and Musto, N.A. (1982) J. Biol. Chem. 257, 5118-5124. Laurent, T.C. and Killander, J. (1964) J. Chromatogr. 14, 3177330. Munson, P.J. and Rodbard, D. (1980) Anal. Biochem. 107, 220-239. Nakhla, A.M., Khan, MS. and Rosner, W. (1990) J. Clin. Endocrinol. Metab. 71, 398-404. Petra, P.H., Que, B.C., Namkung, P.C., Ross, J.B.A., Charbonneau, H., Walsh, K.A., Griffin, P.R., Shabanowitz, J. and Hunt, D.F. (1988) Ann. NY Acad. Sci. 538, 10-24. Pittman, R.C., Carew, T.E., Glass, C.K., Green, S.R., Taylor, C.A.J. and Attie, A.D. (1983) Biochem. J. 212, 791-800. Porto, C.S., Gunsalus, G.L., Bardin, C.W., Phillips, D.M. and Musto, N.A. (1991) Endocrinology 129, 436-445.
Porto, C.S., Musto, N.A.. Bardin, C.W. and Gunsalus, G.L. (1002) Endocrinology 130, 2931-2936. Rolland, R., Gunsalus, G.L. and Hammond, J.M. (1976) Endocrinology 98, 1083-1091. Rosner, W. (1990) Endocr. Rev. I I, 80-91. Sakiyama, R., Pardridge, W.M. and Musto, N.A. (1988) J. Clin. Endocrinol. Metab. 67, 98-103. Siiteri, P.K., Murai, J.T., Hammond, G.H., Nisker, J.A., Raymore, W.J. and Kuhn, R.W. (1982) Recent Prog. Horm. Res. 38, 457510. Sinnecker, G., Hiort, O., Mitze, M., Donn, F. and Neuman, S. (1988) Steroids 52, 335-336. Sinnecker, G., Hiort, Q., Kwan, P.W. and DeLellis, R.A. (1990) Horm. Metab. Res. 22. 47-50. Strel’chyonok, O.A., Awakumov, G.V. and Survilo, L.I. (1984) Biochim. Biophys. Acta 802, 4.59-466. Vale, J.R. (1949) Cienc. Cult. 1, 156. Westphal, U. (1983) J. Steroid Biochem. 19, l-15.
