Human Reproduction vol.7 no.9 pp. 1205 -1209, 1992

Expression of functional growth hormone receptors in human granulosa cells

Bjorn Carlsson, Christina Bergh1, Jane Bentham2, Jan-Henrik Olsson1, Michael R.Norman2, Hakan Billig, Paul Roos3 and Torbjorn Hillensjo1'4 Department of Physiology and 'Department of Obstetrics and Gynaecology, University of Goteborg, Sweden, 2Department of Clinical Biochemistry, Kings College, School of Medicine and Dentistry, London, UK and 'Department of Biochemistry, Biomedical Centre, University of Uppsala, Sweden 4

To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Sahlgrenska Hospital, S-413 45 Goteborg, Sweden

Both clinical and experimental evidence suggest that growth hormone may be of importance for ovarian function. The present study investigated whether growth hormone receptors are expressed in human granulosa cells. Granulosa cells were isolated either from natural cycles or from stimulated cycles in the course of in-vitro fertilization. Total RNA hybridized with a ^P-labelled rat growth hormone receptor cRNA probe revealed one major transcript with an estimated size of 4.5 kb and one minor transcript with an estimated size of 1.3 kb. Biotinylated growth hormone was used to analyse growth hormone binding. Competitive growth hormone binding was detected in freshly isolated granulosa cells, as well as hi cultured cells. Growth hormone augmented basal and/or foDicle stimulating hormone-stimulated steroidogenesis in granulosa cells obtained from patients with natural cycles, but the response to growth hormone stimulation showed considerable variation. We conclude that functional growth hormone receptors are present in human granulosa cells and that growth hormone, therefore, may have an important role In ovarian function. Key words: granulosa cells/growth hormone/growth hormone receptor/human/steroidogenesis

Introduction Although the gonadotrophins have a central role in regulating the ovary, there are now indications that growth hormone (GH) can influence human ovarian function. For instance, GH deficiency delays the onset of puberty and this can be prevented by GH treatment (Sheikholislam and Stempfel, 1972). In animals, GH deficiency also results in a delay in the onset of puberty, a decrease in ovarian luteinizing hormone (LH) receptor content, and a decrease in the response to human chorionic gonadotrophin (HCG). Administration of GH reverses all these effects (Adavis et al., 1981). More recently, GH has been used to increase the © Oxford University Press

responsiveness to gonadotrophins in patients who are resistant to gonadotrophin therapy (Homburg et al., 1988; Volpe et al., 1989; Ibrahim et al., 1991). It is presently unclear whether GH acts directly on the ovary or whether its action is indirect through somatomedin C/insulin-like growth factor-I (IGF-I). Recently, direct stimulatory effects of GH on steroidogenesis in cultured rat (Jia et al., 1986; Hutchinson et al., 1988), porcine (Hsu and Hammond, 1987) and human (Mason et al., 1990) granulosa cells have been reported. However, no study has demonstrated expression of the GH receptor gene or GH binding in human granulosa cells. We report here that the GH receptor gene is active in human granulosa cells and that these cells possess GH binding sites. Furthermore, we demonstrate that GH may stimulate steroidogenesis in cultures of human granulosa cells. Materials and methods Case material Granulosa cells were obtained after informed consent from patients undergoing gynaecological laparotomy for reasons unrelated to ovarian pathology (natural cycles, n = 9), or undergoing treatment for in-vitro fertilization (IVF, n = 12) due to infertility, usually tubal obstruction (stimulated cycles). The study was approved by the local ethics committee. The first group of women was operated on cycle days 7—14 and only the leading follicle (10—30 mm) was excised. The second group of women underwent ultrasound-guided vaginal follicle aspiration 34—36 h after receiving 10 000 IU HCG (Profasi, Serono, Italy). Their cycles had been stimulated by a combination of clomiphene citrate (Clomivid, Draco Ltd, Sweden), human menopausal gonadotrophin (HMG) (Pergonal, Serono, Italy) or a combination of gonadotrophin-releasing hormone (GnRH) agonist (Suprefact, Hoechst, FRG) and HMG. Follicular growth was monitored by daily serum oestradiol measurements, frequent ultrasound scans and, when appropriate, daily serum LH analyses. RNA analysis For analysis of RNA, the tissue was rapidly frozen in liquid nitrogen in the operating room. Total RNA was prepared essentially according to Chomczynski and Sacchi (1987) with minor modifications (Nilsson et al., 1990). A sample of 20 /ig of total RNA was electrophoresed through a 1 % agarose/2.2 M formaldehyde gel and transferred to Hybond-N membranes (Amersham, Buckinghamshire, UK) with a vacuum transfer system (LKB, Stockholm, Sweden). The membranes were baked at 80°C for 3 h and pre-hybridized at 56°C in 50% formamide, 25 mM NaH2PO4, 25 mM N^HPC^, 5 X SSC (1 X SSC is 1205

B.Carison et al.

0.15 M NaCl, 15 mM sodium citrate), 0.1% sodium dodecyl sulphate (SDS), 1 mM EDTA, 0.05% bovine serum albumin (BSA), 0.05% Ficoll, 0.05% polyvinylpyrrolidone (PVP), 200 /ig/ml calf liver RNA, and 200 /ig/ml salmon sperm DNA, and hybridized in the pre-hybridization buffer with the addition of 32P-labelled RNA probe. Antisense [32P]UTP-labelled RNA was synthesized from the EcoRI-linearized plasmid pT7T3 18U which contained a 560 bp BamHI fragment of the rat GH receptor/GH binding protein cDNA (Mathews et al., 1989). The membranes were washed several times in 0.1 x SSC, 0.1 % SDS at 60°C. Autoradiography was performed at -70°C for 10 days with an intensifying screen using Kodak XAR-5 film. A 0.24-9.5 kb RNA ladder (BRL, Life Technologies Inc., MD, USA) was used to estimate transcript sizes. GH binding studies Preliminary GH binding studies using l25I-labelled GH were unsuccessful, showing variable results and low specific binding. Therefore, we applied the following semi-quantitative GH binding assay: sulphosuccinimidyl 6-(biotinamido) hexanoate (NHS-LC biotin; Pierce Eurochemie BV, Holland) was covalently linked to recombinant human growth hormone (hGH; Kabi Pharmacia AB, Sweden) (Bentham et al., 1990). Human growth hormone was incubated with NHS-LC biotin at a 1:5 molar ratio in a 50 mM bicarbonate buffer (pH 9) for 2 h on ice. Free biotin was removed by desalting to phosphate-buffered saline (PBS) on a desalting column in a fast liquid chromatography system (Pharmacia Ltd, Milton Keynes, UK). Freshly isolated human granulosa cells were washed in PBS and centrifuged onto microscope slides using a Cytospin or cultured on slides for 4 days before use. The cells were air-dried for 30 min and fixed in acetone:methanol (1:1) for 1 min. Fixed cells were incubated with 1 % hydrogen peroxide in methanol for 20 min at room temperature to block endogenous peroxidase activity. The cells were incubated in PBS wiui 1% BSA for 10 min and then with biotinylated hGH ( 1 - 5 /xM), with or without unlabelled hGH or human prolactin (hPrl) at a 50-fold molar excess, for 90 min at room temperature in a humid chamber. Control incubations without biotinylated hormone were also included. The cells were washed and incubated with Vectastain ABC complex (Vector Laboratories, Burlingame, CA, USA) for 30 min before being washed and incubated with 0.05 % (w/v) diaminobenzidine and 0.1 % (v/v) hydrogen peroxide in PBS for 10 min. Finally, the cells were washed in water to terminate the reaction. The slides were examined with a Nikon photomicroscope and brown coloured staining was taken as evidence of GH binding. Cell culture and steroid assay Cell culture was as previously described (Olsson et al., 1990). The granulosa cells were prepared as follows: the excised follicle was opened with microscissors under aseptic conditions. The granulosa cells were removed from the follicle wall by flushing and gently scraping with a Pasteur pipette, and then suspended in culture medium for culture experiments and in PBS before freezing for mRNA preparation. In the latter case after the cells had sedimented, excess PBS was removed and the cell pellet snap-frozen in liquid nitrogen. The cells were then stored at 1206

-80°C until RNA extraction. Cells obtained in connection with IVF were initially collected in Earle's medium with 10% heat-inactivated human serum and then washed once in culture medium before use. In brief, cells were cultured in Medium 199 with Earle's salt solution (Gibco, Paisley, UK), 25 mM sodium bicarbonate, 50 /tg/ml gentamicin, 1% fetal bovine serum (FBS; Flow Laboratories, LabKemi AB, Goteborg, Sweden) and testosterone (0.1 /ig/ml; Sigma). In some experiments minimal essential medium (MEM) F12 (1:1) (Gibco) was employed. The cells were cultured ( 2 - 6 X 104 viable cells/well) in 0.5 ml medium in multiwell plates (Costar Ltd, Cambridge, MA, USA) under 5% CO2 in humidified air at 37 °C. Granulosa cells from natural cycles were cultured for 4 days with a change of medium after 2 days. Cells from stimulated cycles were pre-cultured for 2 days without hormone before hormone was added. The levels of progesterone and oestradiol in spent culture media were analysed by radioimmunoassay using well-defined antisera as outlined previously (Olsson and Hillensjo, 1986). The steroid levels were assayed directly after appropriate dilution with water. Hormones Highly purified human follicle stimulating hormone (FSH) (6200 IU/mg, 1.5% LH) and human LH (7300 IU/mg 0.5% FSH) were donated by Dr Peter Toriesen (Oslo, Norway). Human Prl was isolated from human pituitaries (Roos et al., 1979). Recombinant hGH (Genotrophin) was donated by Kabi Pharmacia AB (Stockholm, Sweden). Statistics Statistical differences between treatment groups were calculated by analysis of variance (ANOVA) followed by Student Neuman—Keul's test or a non-parametric sign test. A probability 0.05 was considered significant. Results Growth hormone receptor mRNA RNA was extracted from granulosa cells derived from a natural cycle and analysed by Northern blot. One major transcript was revealed with an estimated size of 4.5 kb and was detected with a 32P-labelled rat GH receptor RNA probe (Figure 1). In addition, a minor transcript of low abundance with an estimated size of 1.3 kb was detected. Binding of biotinylated hGH Biotin-labelled hGH (b-hGH) binding was studied in granulosa cells obtained from patients with cycles stimulated for IVF. Both freshly isolated and cultured granulosa cells exhibited GH binding and no staining was seen when b-hGH was competed with a 50-fold excess of unlabelled hGH (Figure 2A,B). Human prolactin in a 50-fold excess did not displace b-hGH (data not shown). Steroidogenesis in cultures The effect of hGH (0.1-1 mU/ml) on basal and FSH (100 ng/ml)-stimulated steroid production was investigated in granulosa cells from the largest follicles of nine patients with

Growth hormone receptors in granulosa cells

natural cycles (Table I). Overall there was a considerable variation in response. The median and range for GH-stimulated basal progesterone production were 1.85 (0.07-5.32) compared to 1.74 (0.084.42) for controls. Corresponding values for the effect of GH on the FSH-stimulated progesterone production were 9.22 (1.17-20.42) compared to 7.99 (0.95-18.9) for FSH alone. The median and range for GH-stimulated basal oestradiol production were 0.40 (0.10-2.28) compared to 0.40 (0.122.73) for controls. Corresponding values for the effect of GH on the FSH-stimulated oestradiol production were 1.85 (0.13 — 3.25) compared to 1.28 (0.10-3.22) for FSH alone. Human GH in combination with FSH was significantly stimulatory in granulosa cells from four of eight follicles in terms of oestradiol and in four of nine in terms of progesterone. In seven of nine experiments oestradiol and/or progesterone secretion in the presence of FSH was significantly stimulated by GH. In eight of nine follicles, GH increased basal progesterone

-28S

-I8S

Fig. 1. Analysis of growth hormone (GH) receptor/GH binding protein gene expression in isolated granulosa cells obtained from a natural cycle. Total RNA (20 /ig) was electrophoresed, transferred and hybridized with a 32P-labelled GH receptor RNA probe under conditions described in Materials and methods.

B

Fig. 2. Binding of biotinylated human growth hormone (hGH) to cultured human granulosa cells obtained from a stimulated cycle. (A) Granulosa cells incubated with biotinylated hGH. (B) Granulosa cells incubated with biotinylated hGH and a 50-fold molar excess of unlabelled hGH. The cells were then stained with Vectastain ABC complex as described in Materials and methods.

secretion but within each experiment, the difference did not reach statistical significance. A non-parametric sign test, however, showed that the effect of GH on basal progesterone and FSHinduced oestradiol secretion was significant (F 0.05). Figure 3 shows the absolute values obtained in one of the experiments (Experiment 3). The effect of GH on progesterone production in cultured granulosa cells obtained from patients with stimulated cycles was also investigated. Both basal, LH- and GH-stimulated cultures showed variable levels of steroid production. GH stimulated basal and LH-induced progesterone secretion in six of 12 experiments (data not shown). Discussion The present study demonstrates that the GH receptor gene is active in the human ovary. Two transcripts were detected with a GH receptor cRNA probe corresponding to the extraceUular domain of the GH receptor. Granulosa cells obtained from patients undergoing IVF possessed hGH binding sites that could be competed for with an excess of hGH. This binding is compatible with the presence of functional receptors, since hGH was found to augment steroidogenesis. The major GH receptor transcript detected in the present study is likely to correspond to the full-length cloned hepatic GH receptor predicted to encode a trans-membrane protein of 620 amino acids (Leung et al., 1987). However, even though the cloned GH receptor cDNAs in several species demonstrate a high degree of homology (Mathews et al., 1989; Smith et al., 1989), it is not known if the GH receptor is identical in all tissues. Minor heterogeneity is indicated by reports on the differences in GH binding (McCarter et al., 1990), GH receptor immunoreactivity (Barnard et al., 1985), and alternative splicing of the GH receptor mRNA encoding the extracellular domain (for discussion, see Godowski etal., 1989). The probe used in the present study recognizes only part of

Table I. Effect of growth hormone (GH) on steroid formation in granulosa cells from natural cycles Exp. Follicle diameter Cycle day GH Oestradiol no. (mm) (mlU/ml) Basal FSH 1 4 5 6 8 2 3 7 9

15 ND 12 15 15 30 15 15 10

11 ND ND 12 15 13 7 10 8

1 1 1 1 1 0.1 0.1 0.1 01

+22 +41 +1 -15 +7 -13 -20 -18 ND

Progesterone Basal FSH

+76** +45 +61 + 14** + 12 -17 +49 +44** +6 0 +20 + 3 4 " +7 0 +36 ND +33

+36

+1 + 8* +1 -24 + 11 -10 +58** +23** +26"

Results shown are from individual experiments of granulosa cell cultures during culture days 2—4. The cells were cultured in the absence of hormone or in the presence of human follicle stimulating hormone (hFSH) (100 ng/ml), human GH (0.1 or 1 mlU/ml) or the combination of FSH and GH. The results shown are the percentage increase (+) or percentage decrease ( - ) caused by GH on basal or FSH-stimulated steroid formation, respectively. In each experiment, there were 3 - 4 replicates, the means of which were used for the calculations. For further details see Materials and methods. ND = not determined; *P 0.05; **P 0.01 with ANOVA.

1207

B.Carlsson et al.

Day 0 - 2

Day 2 - 4

Day 0 - 2

Day 2 - 4

Fig. 3. Effect of human growth hormone (hGH) (0.1 mU/ml) on basal and follicle stimulating hormone (FSH)-stimulated (100 ng/ml) oestradiol (a) and progesterone (b) production in cultured human granulosa cells during culture days 0-2 and 2-4. The granulosa cells were obtained from one follicle from a natural cycle. Bars represent mean $EM from triplicate wells. *P = 0.05 versus control (open bar) Data from Experiment 3 in Table I. the extracellular domain, providing insufficient information regarding the sequence homology between the liver and ovarian GH receptor. Besides the 4.5 kb transcript present in the ovary, a short transcript with an estimated size of 1.3 kb was also detected by the GH receptor probe. It is not clear what this transcript encodes. Besides the cloned full-length GH receptor, two alternative GH receptor cDNAs have been found. Both mouse and rat express a GH receptor transcript encoding a protein identical to the extracellular domain of the GH receptor but with a hydrophilic tail replacing the hydrophobic trans-membrane and intracellular domains (Smith etal., 1989; Baumbach et al., 1989), thus producing a soluble GH binding protein. In the rabbit, another alternative cDNA was cloned which was predicted to encode a membrane bound GH receptor identical to the GH receptor but with a truncated intracellular domain consisting of only eight amino acids (Leung et al., 1987). To assess whether GH binding sites were present on cultured human granulosa cells, we used a non-isotopic GH biotinavidin-HRP (horseradish peroxidase) system. Labelled hGH binding to granulosa cells was displaced with an excess of unlabelled hGH whereas hPrl was not as effective. This indicates the presence of somatogenic receptors in granulosa cells. In the present study, we have demonstrated an effect of GH on steroidogenesis in cultured granulosa cells, confirming a recent study by Mason et al. (1990) who demonstrated a marked stimulatory effect of hGH on basal aromatase activity in granulosa cells obtained from small follicles ( 5 - 7 mm diameter). We have studied granulosa cells from more mature follicles and found that the most pronounced effect on oestrogen production was seen in combination with FSH. A possibility is that the difference in responsiveness could depend on the degree of differentiation of the granulosa cells used. The granulosa cells used by Mason et al. (1990) were obtained from smaller follicles ( 5 - 7 mm) than the granulosa cells used in our experiments. In the present study, granulosa cells obtained from natural cycles appeared to be more responsive to GH stimulation than granulosa cells from patients with stimulated cycles. Taken togedier, this indicates that less differentiated granulosa cells are more sensitive to GH stimulation. We cannot explain the inhibitory effect of GH + 1208

FSH observed in a few cases. In these particular cases, the cells might be endowed with abundant prolactin receptors and GH could interact with them. Prolactin is known to inhibit steroid production in human granulosa cells in vitro as previously shown by several groups including our own. The mechanism by which GH sensitizes the ovary to gonadotrophins is unknown. According to the somatomedin hypothesis of GH action, liver-derived IGF-I mediates the action of GH (Daughaday, 1989). More recently, the somatomedin hypothesis has been extended to include extra-hepatic production of IGF-I (Isaksson et al., 1988). The extra-hepatic action of GH is also consistent with the wide tissue distribution of GH receptor gene expression in the rat (Mathews et al., 1989; Carlsson et al., 1990). The effects of IGF-I on cultured rat and porcine granulosa cells have been extensively studied under in-vitro conditions. In these systems, IGF-I participates in the regulation of granulosa cell proliferation and differentiation (Adashi et al., 1985). More recent data indicate that the IGFs are of importance in the regulation of human granulosa cell function. Human granulosa cells produce IGF-II and possibly also IGF-I (Voutilainen and Miller, 1987), have IGF-I receptors (Gates et al., 1987), and produce IGF-binding protein-1 (Suikkari etal., 1989). Under in-vitro conditions, IGF-I has been shown to be equipotent with FSH in stimulating oestrogen production and more potent in stimulating thymidine incorporation (Erickson etal., 1989; Olsson etal., 1990). A major problem with studies of GH action arises from its similarities to prolactin and related hormones. Cloning of GH and Prl and their receptors has further established the close relationship between the hormone systems (Boutin et al., 1988). In this study, three methods were used to investigate the presence of GH receptor in the ovary. Firstly, the demonstration of transcriptsrecognizedby a rat GH receptor probe in RNA isolated from granulosa cells. Secondly, competitive GH binding was detected on both freshly isolated and cultured granulosa cells. Thirdly, GH was found to enhance gonadotrophin-induced steroid production, thus distinguishing it from the inhibitory action of Prl in the same in-vitro system (Cutie and Andino, 1988). Our study demonstrates the presence of functional GH receptors

Growth hormone receptors in granulosa cells

in the ovary, and thereby supports a role for GH in ovarian function. During final follicular maturation, the role of GH may be to augment FSH-stimulated oestradiol production. This is compatible with recent clinical studies showing that GH augments the response to gonadotrophin in ovulation induction (Homburg et al., 1988; Volpe et al., 1989; Ibrahim et al., 1991). However, the precise structure of the GH receptor in the ovary, its regulation and mechanism of action remain to be investigated. Acknowledgements We thank Lawrence Mathews for the rat GH receptor cDNA. The study was supported by the Swedish Medical Research Council (27,9295,5978), Kabi Pharmacia AB, the Sdderbergska Foundation, the Goteborg Medical Society, Joint Research Committee of Kings College Hospital and the Swedish Institute.

References Adashi.E.Y., Resnick,C.E., D'Ercole.A.J., Svoboda.M.E. and Van WykJ.J. (1985) Insulin-like growth factors as intraovarian regulators of granulosa cell growth and function. Endocrine Rev., 6, 400-420. AdavisJ.P., White.S.S. and Ojeda.S.R. (1981) Activation of growth hormone short loop negative feedback delays puberty in the female rat. Endocrinology, 108, 1343-1352. Barnard,R., Bundesen.P.G., Brylatt.D.B. and Waters.M. (1985) Evidence from the use of monoclonal antibody probes for structural heterogeneity of growth hormone receptor. Biochem. J., 231, 459-468. Baumbach.W.R., Homer.D.L. and Logan.J.S. (1989) The growth hormone-binding protein in rat serum is an alternatively spliced form of the rat growth hormone receptor. Genes Dev., 3, 1199-1205. Bentham^., Aldis.P. and Norman.M.R. (1990) Identification of growth hormone receptors using biotinylated human growth hormone. J. Endocrinol., V2A, 122. BoutinJ.-M., Jolicoerur,C, Okamura.H., Gagnon.J., Edery.M., Shirota.M., Banville.D., Dusanter-Fourt.I., Djiane,J. and Kelly,P.A. (1988) Cloning and expression of the rat prolactin receptor, a member of the growth hormone/prolactin receptor gene family. Cell, 53, 69-77. Carisson.B., Billig.H., Rymo,L. and Isaksson.O.G.P. (1990) Expression of the growth hormone-binding protein messenger RNA in the liver and extrahepatic tissues in the rat: co-expression with the growth hormone receptor. Mol. Cell Endocrinol., 73, Rl—R6. Chomczynski,P. and Sacchi,N. (1987) Single-step method for isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162, 156-159. Cutie.E.R. and Andino,N.A. (1988) Prolactin inhibits the steroidogenesis in midfollicular phase human granulosa cells cultured in a chemically defined medium. Fenil. Sterii, 49, 632-637. Daughaday.W.H. (1989) A personal history of the origin of somatomedin hypothesis and recent challenges to its validity. Perspect. Med., 32, 194-211. Erickson.G.F., Garzo.V.G. and Magoffm.D.A. (1989) Insulin-like growth factor-I regulates aromatase activity in human granulosa and granulosa luteal cells. J. Clin. Endocrinol. Metab., 69, 716-724. Gates.G.S., Bayer,S., Seibel.M., Poretsky.L., Flier.J.S. and Moses,A.C. (1987) Characterization of insulin-like growth factor binding to human granulosa cells obtained during in vitro fertilization.

J. Recept. Res., 7, 885-902. Godowski.PJ., Leung.D.W., Meacham.L.R., GalganiJ.P., HeUmiss,R., Keret.R., Rotwein.P.S., ParksJ.S., Laron,Z. and Wood.W.I. (1989) Characterization of the human growth hormone receptor gene and demonstration of a partial gene deletion in two patients with Laron-type dwarfism. Proc. Natl. Acad. Sri. USA, 86, 8083-8087.

Homburg.R., Eshel.A., Abdalla.H.I. and Jacobs.H.S. (1988) Growth hormone facilitates ovulation by gonadotrophins. Clin. Endocrinol, 29, 113-117. Hsu.C.J. and HammondJ.M. (1987) Concomitant effects of growth hormone on secretion of insulin-like growth factor I and progesterone by cultured porcine granulosa cells. Endocrinology, 121, 1343 — 1348. Hutchinson,L.A., Findley.J.K. and Herington,A.C. (1988) Growth hormone and insulin-like growth factor-I accelerate PMSG-induced differentiation of granulosa cells. Mol Cell Endocrinol., 55, 61—69. Ibrahim,Z.H.Z., Mason.P.L., Buck.P. and Lieberman.B.A. (1991) The use of biosynthetic human growth hormone to augment ovulation induction with buserelin acetate/human menopausal gonadotropin in women with a poor ovarian response. Fertil. Sterii., 55, 202—204. Isaksson.O.G.P., Lindahl.A. and IsgaardJ. (1988) Action of growth hormone: current views. Acta Pediatr. Scand., 343, 12-18. Jia,X.-C., Kalmijn,J. and Hsueh.A.J. (1986) Growth hormone enhances follicle-stimulating hormone-induced differentiation of cultured rat granulosa cells. Endocrinology, 118, 1401 -1438. Leung.D.W., Spencer,S.A., Cachianes.G., Hammonds,G.R., Collins.C, Henzel.W.J., Barnard^., Waters.M.J. and Wood.W.I. (1987) Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature, 330, 537—543. Mason.H.D., Martikainen.H., Beard.R.W., Anyaoku.V. and Franks.S. (1990) Direct gonadotrophic effect of growth hormone on oestradiol production by human granulosa cells in vitro. J. Endocrinol., 126, R1-R4. Mathews.L.S., Engberg.B. and Nordstedt.G. (1989) Regulation of rat GH receptor gene expression. J. Biol. Chenu, 264, 9905-9910. McCarter.J., Shaw.M.A., Winer.L.A. and Baumann.G. (1990) The 20,000 Da variant of growth hormone does not bind to growth hormone receptors in human liver. Mol. Cell. Endocrinol., 73, 11-14. Nilsson.A., Carlsson.B., Mathews.L. and Isaksson.O.G.P. (1990) Growth hormone regulation of the growth hormone receptor mRNA in cultured rat epiphyseal chondrocytes. Mol. Cell. Endocrinol., 70, 237-246. Olsson,J.-H. and Hillensjo.T. (1986) Inhibitory effects of danazol on steroidogenesis in cultured human granulosa cells. Fertil. Sterii., 46, 237-242. OlssonJ.-H., Carlsson.B. and Hillensj6,T. (1990) Effect of insulin-like growth factor 1 on deoxyribonucleic acid synthesis in cultured human granulosa cells. Fertil. Sterii., 54, 1052-1057. Roos.P., Nyberg.F. and Wide.L. (1979) Isolation of human pituitary prolactin. Biochim. Biophys. Acta, 588, 368-379. Sheikholislam,B.M. and Stempfel,R.S. (1972) Heredity of isolated somatotropin deficiency: effects of human GH administration. Pediatrics, 49, 362-374. Smith.W.C, Kuniyoshi,J. and Talamantes.F. (1989) Mouse serum growth hormone (GH) binding protein has GH receptor extracellular and substituted transmembrane domains. Mol. Endocrinol, 3, 984-990. Suikkari,A.-M., Jalkanen.J., Koistinen.R., Butzow.R., Ritvos.O., Ranta.T. and Seppala.M. (1989) Human granulosa cells synthesize low molecular weight insulin-like growth factor-binding protein.

Endocrinology, 124, 1088-1090. Volpe.A., Coucus.G., Barreca^A., Artini.P.G., Minuto.F., Giardano.G. and Genazzini.A.R. (1989) Ovarian response to combined growth hormone-gonadotropin treatment in patients resistant to induction of superovulation. Gynecol. Endocrinol., 3, 125 — 133. Voutilainen.R. and Miller.W.L. (1987) Coordinate tropic hormone regulation of mRNAs for insulin-like growth factor II and the cholesterol side-chain-cleavage enzyme, P45Ossc, in human steroidogenic tissues. Proc. Natl. Acad. Sri. USA, 84, 1590-1594. Received on March 5, 1992; accepted on June 23, 1992

1209