a Laboratory of Physiology, School of Medicine, Democritus University of Thrace, Alexandroupolis, Greece; b Department of

Transcription

1 Gonadotropin-releasing hormone antagonists do not influence the secretion of steroid hormones but affect the secretion of vascular endothelial growth factor from human granulosa luteinized cell cultures Byron Asimakopoulos, Ph.D., a Nikos Nikolettos, M.D., Ph.D., a Barbara Nehls, b Klaus Diedrich, M.D., Ph.D., b Safaa Al-Hasani, M.V.D., Ph.D., b and Eric Metzen, M.D., Ph.D. c a Laboratory of Physiology, School of Medicine, Democritus University of Thrace, Alexandroupolis, Greece; b Department of Obstetrics and Gynecology, University of Schleswig-Holstein, Campus Lübeck, Lübeck, Germany; c Institute of Physiology, Medical University of Lübeck, Lübeck, Germany Objective: To evaluate the secretion of E 2, P, and vascular endothelial growth factor (VEGF) in human granulosa luteinized cell cultures with the presence of a gonadotropin-releasing hormone (GnRH) agonist or antagonist. Design: In vitro cell culture study. Setting: Research laboratory of a university hospital. Patients: Granulosa luteinized cells were obtained from 24 patients undergoing ovarian stimulation for IVF treatment. Interventions: Granulosa cells were cultured for 48 hours with 1 nm of cetrorelix or leuprorelide. For a further 48 hours, granulosa cells were cultured with or without the combination of cetrorelix plus leuprorelide. Main Outcomes: At the end of each culturing period, the concentrations of E 2, P, and VEGF were measured in culture supernatants by immunoassays. Results: Estradiol and P concentrations were similar between the culture supernatants from controls and treatment groups. The VEGF concentrations in supernatants from cultures with cetrorelix (2, ,565.5 pg/ml) were moderately, but significantly, lower than in controls (2, ,907.1 pg/ml) or cultures with leuprorelide (2, ,403.1 pg/ml). Conclusions: The GnRH analogues do not affect steroidogenesis in human granulosa luteinized cell cultures. The GnRH antagonists moderately affect the secretion of VEGF from human granulosa luteinized cells. (Fertil Steril 2006;86: by American Society for Reproductive Medicine.) Key Words: GnRH analogues, estradiol, progesterone, VEGF, granulosa cells Granulosa cells support and regulate the development of the oocyte by the production not only of steroid hormones but also a plethora of growth factors. The function of granulosa cells is mainly under the influence of gonadotropins that sustain the growth, proliferation, and secretion of this type of cells. The ovulation-inducing LH surge, discharged by the pituitary gland, reprograms granulosa cell function, leading to their terminal differentiation. The LH surge orchestrates profound changes in the production and secretion of sex steroid hormones and cytokines from granulosa cells. Among the growth factors secreted from granulosa cells, vascular endothelial growth factor (VEGF) is of particular interest. The VEGF is responsible for the vascularization of Received August 29, 2005; revised and accepted January 20, Supported in part by the Deutscher Akademischer Austausch Dienst (German Akademic Exchange Service), award A/04/33529 (B.A.). Presented at the 84th Annual Meeting of the Deutsche Physiologische Gesellschaft (German Physiological Society), March 6 9, 2005, Goettingen, Germany. Reprint requests: Byron Asimakopoulos, Ph.D., Laboratory of Physiology, School of Medicine, Democritus University of Thrace, Dragana, Alexandroupolis, Greece (FAX: ; basima@ med.duth.gr). the follicle and the regulation of vascular permeability, thus participating in the supply of the oocyte with oxygen, nutrients, and regulatory molecules. The role of VEGF is essential for follicular selection and development, as well as for the formation of corpus luteum and its function (1 4). Particularly after LH surge, VEGF production is augmented to enhance the microvascular network of the follicle, and consequently, it contributes to the formation of a functional corpus luteum (5 7). The VEGF is also implicated in various pathophysiological conditions of the ovary, including the pathogenesis of ovarian hyperstimulation syndrome (OHSS), a potentially lethal complication of ovarian stimulation (8). During controlled ovarian hyperstimulation (COH), which is used for the preparation of women participating in IVF treatment, granulosa cells function under a different environment than in natural cycles. Apart from the supraphysiological doses of exogenously administered gonadotropins, during COH there is a continuous presence of molecules used to suppress the secretion of endogenous gonadotropins from the pituitary gland: gonadotropin-releasing hormone (GnRH) analogues (agonists or antagonists). 636 Fertility and Sterility Vol. 86, No. 3, September /06/$32.00 Copyright 2006 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 The GnRH analogues are broadly used in IVF to suppress the secretion of the pituitary gland, thus preventing premature LH surges. The binding of GnRH-agonists to GnRH receptors of gonadotroph pituitary cells initially results in a stimulatory (flare-up) effect, further leading to their desensitization and to a reduction in the number of GnRH receptors. On the other hand, GnRH-antagonists directly inhibit gonadotropin secretion through a competitive binding to GnRH receptors. The GnRH-antagonists in COH protocols have a number of advantages compared with GnRH-agonists: avoidance of the flare-up effect, reduction of the length of stimulation, and a lower incidence of the OHSS (9 14). However, GnRH-antagonist protocols appear to be associated with lower E 2 serum concentrations on the day of hcg administration and slightly lower pregnancy rates than GnRH-agonist protocols (14 15). This fact, along with the presence of GnRH receptors in the ovary, has fueled an ongoing debate on the ovarian actions of GnRH analogues. Several researchers who studied the direct effects of GnRH analogues on steroidogenesis in granulosa luteinized cell cultures have come to contradictory results. At the same time, it was hypothesized that GnRH-antagonists negatively affect the biosynthesis and/or the secretion of growth factors produced locally by normal ovarian cells (16). In this study, human granulosa luteinized cells (GLCs) were cultured with or without a GnRH-agonist (leuprorelide), a GnRH-antagonist (cetrorelix), or their combination. The aim was to investigate the secretion of E 2, P, and VEGF (namely VEGF 165, which is the predominant isoform) by measuring their levels in culture supernatants. In addition, the secretion of the soluble receptor 1 of VEGF (svegfr1), which is an important regulator of VEGF activity in vivo, was studied. MATERIALS AND METHODS The GLCs were obtained at the time of oocyte retrieval from 24 women (29 39 years old) following COH for IVF treatment because of male factor infertility in the Clinic of Obstetrics and Gynecology, University of Schleswig-Holstein, Campus Lübeck. The patients gave their verbal consent and did not receive any monetary compensation. Approval from the institutional review board was not required because GLCs are normally discarded after oocyte recovery. The COH followed the multidose antagonist protocol (Lübeck protocol) (9 11). Briefly, the ovarian stimulation was made with human menopausal gonadotropins (Menogon; Ferring Arzneimittel GmbH, Kiel, Germany) and recombinant FSH (Gonal-F; Serono International S.A., Geneva, Switzerland). Pituitary suppression was achieved with a daily dose of 0.25 mg of cetrorelix (Cetrotide; Serono International S.A., Geneva, Switzerland) from the sixth day of ovarian stimulation until the day of the induction of ovulation. The ovulation was induced with 10,000 IU of hcg (Choragon; Ferring Arzneimittel GmbH, Kiel, Germany). Following oocyte retrieval, intracytoplasmic sperm injection was performed, with the exception of two cases where conventional IVF was applied. Embryo transfers were made at day 2. Up to three cleaving embryos were transferred according to the German Embryo Protection Law. Clinical pregnancy was defined by the presence of positive fetal heartbeat. Cell Preparation After harvesting the cumulus-oocyte complex for IVF, granulosa cells were isolated from follicular fluids through repeated centrifugation and washing with phosphate-buffered saline or culture medium. They were plated into 24-well culture dishes at a density of 100,000 cells per well with the addition of 1 ml Dulbecco s Eagle Modified Medium F12 (Gibco, Grand Island, NY) supplemented with 10% Ultroser G (Ciphergen BioSystems Inc., Cergy-Saint-Christophe, France), 100 IU/mL penicillin, and streptomycin [Dulbecco s Eagle Modified Medium F12 (Gibco) 10% Ultroser G (Ciphergen BioSystems) penicillin streptomycin (DMEM/U/PS)]. An overnight incubation (37 C, humidified atmosphere, 5% CO 2 ) followed to allow attachment of the cells to the bottom of the plates and recovery from any effect of the in vivo exposure to GnRH analogues and gonadotropins during COH. Twentyfour hours later, the incubation medium was replaced. Experiment 1 The cultures derived from each patient were divided into controls (group 1) and treated cultures. Treated cultures were divided into those treated with cetrorelix (group 2) and those treated with leuprorelide (group 3). Each patient contributed at least two cultures to each group; consequently, the three groups were interdependent. With this experimental design, comparisons were made between cultures belonging to different groups but derived from the same patient. During the next 48 hours, group 1 control cultures were incubated in 1 ml of DMEM/U/PS, group 2 cultures in 1 ml of DMEM/U/PS plus 1 nm of cetrorelix, and group 3 cultures in 1 ml of DMEM/U/PS plus 1 nm of leuprorelide. The cultures were incubated at 37 C in a humidified atmosphere with 5% CO 2. After this period (day 3), the medium samples were collected and stored at 20 C. Experiment 2 After experiment 1, another 48-hour culturing period ensued, in which the group 1 controls were incubated under the same conditions (1 ml of DMEM/U/PS), while group 2 and group 3 were incubated in 1 ml of DMEM/U/PS supplemented with 1 nm of cetrorelix plus 1 nm of leuprorelide. At the end of this period (day 5), the medium samples were stored at 20 C. For technical reasons, experiment 2 was performed with GLC cultures only from nine of the patients. Measurements All measurements were made with commercial ELISA kits as follows: E 2 : DSL (DSL, Webster, Texas); sensitivity: 7 pg/ ml; intraassay coefficient of variation (CV): 3.3% 4.8%; interassay CV: 6.5% 8.2%. Fertility and Sterility 637

3 P: Quantikine DE2200 (R&D Systems Inc, Minneapolis, MN); sensitivity: 8.57 pg/ml; intraassay CV: 4.9% 7.6%; interassay CV: 2.7% 8.3%. VEGF: Quantikine DVE00 (R&D Systems Inc, Minneapolis, MN); sensitivity: 5 pg/ml; intraassay CV: 3.5% 6.5%; interassay CV: 5% 8.5%. svegfr1: Quantikine DVR100F (R&D Systems Inc, Minneapolis, MN); sensitivity 5.01 pg/ml; intra- and interassay CV: 2.6% 3.8% and 7% 8.1%, respectively. Data Analysis In both sets of experiments (experiment 1 and experiment 2), there were at least two cultures from every patient in every experimental group (group 1 control, group 2, and group 3). In this way, the experimental groups were interdependent, and the comparisons between them were made with Wilcoxonmatched pairs test. The possible influence of patient age on the ability of granulosa cells to secrete E 2, P, and VEGF was evaluated by examining the correlation between age and the measured concentrations of the hormones and VEGF with the Spearman rank test. Possible differences between the cultures that came from women who became pregnant and from those who did not were also investigated. Statistical analyses were performed with Statistica 6.0 (StatSoft Inc., Tulsa, OK). Differences were considered significant at values of P.05. RESULTS Experiment 1 The E 2 and P were secreted in all granulosa cell cultures with P in very high concentrations as expected (Table 1). No statistically significant differences were found among the experimental groups, although P concentrations in groups 2 and 3 were lower than in the control group. The VEGF was secreted in all granulosa cell cultures; svegfr1 was not detected in any supernatant. Granulosa cell cultures treated with a GnRH-antagonist secreted significantly lower levels of VEGF than controls (P.01525) and cultures treated with a GnRH-agonist (P.0077) (Table 2, Fig. 1). As expected, VEGF levels in granulosa cell cultures derived from individual patients were highly variable. TABLE 2 Vascular endothelial growth factor levels (pg/ ml) in granulosa cell culture supernatants on day 3 of experiments. Group n Mean SD Group 1 (control) 24 2, ,907.1 Group 2 (antagonist) 24 2, ,565.5 a,b Group 3 (agonist) 24 2, ,403.1 a P.01525, group 2 vs. controls. b P.0077, group 2 vs. group 3. Asimakopoulos. GnRH-antagonists and VEGF secretion. Fertil Steril Experiment 2 The cultures treated with a combination of GnRH-agonist GnRH-antagonist secreted similar levels of E 2, P, and VEGF as controls (Table 3, Table 4). Again, the svegfr1 was not detectable. The control cultures at day 5 secreted significantly lower amounts of E 2 (P.005), P (P.001), and VEGF than at day 3 (P.025). The statistical analysis did not show any relationship between patient age and the measured concentrations of hormones and VEGF in the two experiments. Among the 24 patients, five clinical pregnancies occurred. No statistically significant differences were found in the secretions between the cultures derived from patients who became pregnant and those who did not. DISCUSSION Many investigators have attempted to define direct effects of GnRH-agonists on human granulosa cells (17 25). With respect to steroidogenesis of granulosa cells, the results reported to date are particularly contradictory. In addition, a limited number of studies have been conducted to examine the direct effects of GnRH-antagonists on granulosa cell function, and these are mainly focused on steroidogenesis (26 28). The present study demonstrated that the in vitro exposure of human GLCs to GnRH analogues (agonist or antagonist) did TABLE 1 Estradiol (pg/ml) and P concentrations (ng/ml) in granulosa cell culture supernatants on day 3 of experiments. Group n E 2 P Group 1 (control) , , Group 2 (antagonist) , , Group 3 (agonist) , , Note: Values are mean SD. The differences were not statistically significant. Asimakopoulos. GnRH-antagonists and VEGF secretion. Fertil Steril Asimakopoulos et al. GnRH-antagonists and VEGF secretion Vol. 86, No. 3, September 2006

4 FIGURE 1 Levels of vascular endothelial growth factor (pg/ml) secreted from granulosa cell cultures on day 3 of experiments. The open boxes represent standard errors. The small dark boxes represent mean values. The whiskers represent standard deviations. *: p.01525, Group 2 vs Controls; : p.0077, Group 2 vs Group 3. Asimakopoulos. GnRH-antagonists and VEGF secretion. Fertil Steril not influence the secretion of E 2 and P. The secretion of steroid hormones was studied without stimulation by hcg, LH, or gonadotropins. The results of the present study are in agreement with those of Weiss et al. (28), who found that the in vitro treatment of GLCs with a GnRH-agonist or two different GnRH-antagonists (cetrorelix or ganirelix) did not exert any significant effect on basal or hcg-stimulated steroidogenesis. The results also indicated that the in vivo treatment with GnRH analogues did not have any effect on steroidogenesis of GLC. Other researchers have also reported a lack of direct effects of GnRH-agonists on ovarian steroidogenesis (17, 19). On the other hand, several other studies suggested that GnRH-agonists do directly act on granulosa cells, influencing secretion of steroid hormones (18, 20 22, 24 25). However, the type of influence is another matter of controversy because some studies reported inhibitory, whereas others reported stimulatory influence on E 2 and P secretion separately, or on both of them. Furthermore, Bussenot et al. (23) indicated that certain GnRH-agonists exhibit stimulatory actions, whereas others do not have any effect on E 2 production. This discrepancy may, at least, be partly explained by differences in the experimental settings. It is worth noting that most of these studies were of an experimental design different from the current one. As stated before, only a few studies have evaluated the effect of GnRH-antagonists on steroidogenesis of GLC cultures. Apart from Weiss et al. (28), there are also two other studies in which GLC derived from patients treated with a GnRH-antagonist or a GnRH-agonist were cultured, and TABLE 3 Estradiol (pg/ml) and P levels (ng/ml) in granulosa cell culture supernatants on day 5 of experiments. Treatment n E 2 P Control , , Agonist antagonist , , Note: values are mean SD. The differences were not statistically significant. Asimakopoulos. GnRH-antagonists and VEGF secretion. Fertil Steril Fertility and Sterility 639

5 TABLE 4 Vascular endothelial growth factor levels (pg/ ml) in granulosa cell culture supernatants on day 5 of experiments. Treatment n Mean SD SE Control 9 2, , Agonist antagonist 9 2, , Note: The differences were not statistically significant. Asimakopoulos. GnRH-antagonists and VEGF secretion. Fertil Steril steroidogenesis was determined either under basal conditions or after stimulation. Minaretzis et al. (26) reported reduced aromatase activity in GLC cultures derived from patients treated with the GnRHantagonist Nal-Glu, whereas the basal and gonadotropinstimulated P secretion values were similar to those of patients treated with the GnRH-agonist leuprorelide. Lin et al. (27) found that GLC cultures from patients treated with cetrorelix had a tendency to respond earlier to the in vitro hormone stimulation, in terms of P secretion, as compared with cultures derived from patients treated with the GnRHagonist buserelin. However, these investigators did not find any differences in P or E 2 secretion before stimulation. Again, it is worth mentioning that the experimental design in both studies was different from the present one. The most important difference was that these two studies tested the steroidogenesis of GLC after in vivo treatment with GnRH analogues and in vitro hcg stimulation, whereas the current study tested the secretion of steroid hormones after in vitro exposure to GnRH analogues. Among the previously mentioned studies, only the study by Weiss et al. (28) had a similar experimental design to ours and, consequently, similar results. Whereas the effect of GnRH analogues, especially of GnRHagonists, on steroidogenesis has gained considerable attention, the possible effects of GnRH analogues (agonists or antagonists) on the secretion of growth factors from human GLC have remained obscure. Recently, Weiss et al. (29) published a study on the effects of two GnRH-antagonists and a GnRH-agonist on the IGF system of GLC, demonstrating that the IGF system responded similarly in all three of the GnRH analogues. The present study investigated the secretion of VEGF and svegfr1. The GLC cultures secreted large amounts of VEGF but did not secrete svegfr1. The amount of secretion declined over time. Time-dependent secretion of VEGF has also been reported by other researchers (5). The present study is the first to demonstrate that GnRHantagonists moderately affect the secretion of VEGF from human granulosa cell cultures. Although moderate, this effect was statistically significant. In contrast to this result, GnRH-agonists or the combination of an agonist with an antagonist did not appear to influence the secretion of VEGF. These results indicate an interaction between GnRH receptor signaling and VEGF release and/or production. The GnRH receptor interacts with multiple G-proteins in a cell context-dependent manner: in hypothalamic neurons with Gq, Gs, and Gi; in pituitary gonadotroph cells with Gq; and in tumor cells with Gi (31 35). The activation of G- proteins triggers various pathways leading to the activation of several members of the mitogen-activated protein kinase (MAPK) superfamily, which in turn activate a variety of transcription factors in the nucleus (31 35). Moreover, in extrapituitary cells, the GnRH receptor can activate MAPK via alternative pathways including activation of epidermal growth factor receptor (34 35). In granulosa cells, it is neither known which G-proteins interact with GnRH receptors, nor which particular signaling pathways are activated by GnRH receptors. Therefore, at the moment, any speculation on certain mechanisms linking GnRH signaling and VEGF production and secretion appears to be very risky. Recently, Cunha-Filho et al. (30) investigated the role of GnRH-antagonists on VEGF production during minimally stimulated IVF cycles. They compared the intrafollicular concentrations of VEGF between women minimally stimulated with GnRH-antagonist protocol and women following natural-cycle IVF. According to their results, intrafollicular VEGF concentrations were lower in the GnRH-antagonist group, but the difference was not statistically significant (medians: 776 vs. 1,187.5 pg/ml). Although the results of Cunha-Filho et al. (30) appear to be contradictory to the results of the present study, we should emphasize that a direct comparison between the two studies is not possible because the one is an in vivo and the other is an in vitro study. For example, in our study, we investigated the secretion of VEGF from GLCs, whereas Cunha-Filho et al. (30), by measuring the intrafollicular concentrations of VEGF, evaluated the secretion of VEGF not only from granulosa, but also from theca cells and possibly other sites. In our opinion, the existence of such a difference in intrafollicular VEGF concentrations (776 vs. 1,187.5 pg/ml), although not statistically significant, suggests further investigation. For a better investigation of VEGF status, future clinical studies should evaluate the balance between free VEGF, total VEGF, and svegfr1. The findings of the present study are interesting for two reasons. First, they indicate that GnRH-antagonists do have a direct influence on the function of granulosa cells and possibly on the function of other ovarian cell types. Second, the present findings may provide an additional explanation for the lower incidence of OHSS in IVF cycles where COH follows GnRHantagonist protocols. Up to now, the reduction in the incidence of OHSS in GnRH-antagonist cycles has been attributed to the decreased number of small-sized follicles (13). However, it has been reported that VEGF is an important mediator in OHSS, 640 Asimakopoulos et al. GnRH-antagonists and VEGF secretion Vol. 86, No. 3, September 2006

6 whereas high levels of VEGF have been associated with increased risk and severity of this syndrome (8). In conclusion, the results of the present study indicate that GnRH analogues do not affect steroidogenesis in human GLCs, thus confirming previous reports. The GnRH-antagonists appear to moderately affect the secretion of VEGF from GLCs. Therefore, the present results emphasize the necessity for future investigation of the probable influence of GnRH-antagonists on growth factor expression and release at the ovarian level. REFERENCES 1. Ferrara N, Chen H, Davis-Smyth T, Gerber HP, Nguyen TN, Peers D, et al. Vascular endothelial growth factor is essential for corpus luteum angiogenesis. Nat Med 1998;4: Geva E, Jaffe RB. Role of vascular endothelial growth factor in ovarian physiology and pathology. Fertil Steril 2000;74: Ferrara N. Vascular endothelial growth factor: basic science and clinical progress. Endocrine Rev 2004;25: Tamanini C, De Ambrogi M. Angiogenesis in developing follicle and corpus luteum. Reprod Dom Anim 2004;39: Neulen J, Raczek S, Pogorzelski M, Grunwald K, Yeo TK, Dvorak HF, et al. Secretion of vascular endothelial growth factor/vascular permeability factor from human luteinized granulosa cells is human chorionic gonadotrophin dependent. Mol Hum Reprod 1998;4: Hazzard TM, Molskness TA, Chaffin CL, Stouffer RL. Vascular endothelial growth factor (VEGF) and angiopoietin regulation by gonadotrophin and steroids in macaque granulosa cells during the periovulatory interval. Mol Hum Reprod 1999;5: Molskness TA, Stouffer RL, Burry KA, Gorrill MJ, Lee DM, Patton PE. Circulating levels of free and total vascular endothelial growth factor (VEGF)-A, soluble VEGF receptors-1 and -2, and angiogenin during ovarian stimulation in non-human primates and women. Hum Reprod 2004;19: Rizk B, Aboulghar M, Smitz J, Ron-El R. The role of vascular endothelial growth factor and interleukins in the pathogenesis of severe ovarian hyperstimulation syndrome. Hum Reprod Update 1997;3: Felberbaum R, Reissmann T, Küpker W, Al-Hasani S, Bauer O, Schill T, et al. Hormone profiles under ovarian stimulation with human menopausal gonadotropin (hmg) and concomitant administration of the gonadotropin releasing hormone (GnRH)-antagonist Cetrorelix at different dosages. J Assist Reprod Genet 1996;13: Diedrich K, Felberbaum R. New approaches to ovarian stimulation. Hum Reprod 1998;13:1s 13s. 11. Felberbaum RE, Albano C, Ludwig M, Riethmüller-Winzen H, Grigat M, Devroey P, et al. Controlled ovarian stimulation for assisted reproduction with hmg and concomitant midcycle administration of the LHRH-antagonist Cetrorelix (Cetrotide) according to the multiple dose protocol results of a prospective non controlled phase III study. Hum Reprod 2000;15: Ludwig M, Felberbaum RE, Diedrich K. Interactions of GnRH analogues and gonadotrophins. In: Adashi EY, Baird DT, Crosignani PG, eds. Gonadotrophins and fertility. Vol. 1. Rome: Christengraf, 1999: Ludwig M, Felberbaum RE, Devroy P, Albano C, Riethmüller-Winzen H, Schüler A, et al. Significant reduction of the incidence of ovarian hyperstimulation syndrome (OHSS) by using the LHRH antagonist Cetrorelix in controlled ovarian stimulation for assisted reproduction. Arch Gynecol Obstet 2000;264: Ludwig M, Katalinic A, Diedrich K. Use of GnRH antagonists in ovarian stimulation for assisted reproductive technologies compared to the long protocol. Meta-analysis. Arch Gynecol Obstet 2001;265: Al-Inany H, Aboulghar M. GnRH-antagonist in assisted reproduction: a Cochrane review. Hum Reprod 2002;17: Hernandez ER. Embryo implantation and GnRH antagonists: embryo implantation: the Rubicon for GnRH antagonists. Hum Reprod 2000; 15: Casper RF, Erickson GF, Rebar RW, Yen SS. The effect of luteinizing hormone-releasing factor and its agonist on cultured human granulosa cells. Fertil Steril 1982;37: Parinaud J, Beaur A, Bourreau E, Vieitez G, Pontonnier G. Effect of a luteinizing hormone-releasing hormone agonist (Buserelin) on steroidogenesis of cultured human preovulatory granulosa cells. Fertil Steril 1988;50: Lanzone A, Panetta V, Di Simone N, Arno E, Fulghesu AM, Caruso A, et al. Effect of gonadotrophin-releasing hormone and related analogue on human luteal cell function in vitro. Hum Reprod 1989;4: Pellicer A, Miro F. Steroidogenesis in vitro of human granulosa-luteal cells pretreated in vivo with gonadotropin-releasing hormone analogs. Fertil Steril 1990;54: Miro F, Sampaio MC, Tarin JJ, Pellicer A. Steroidogenesis in vitro of human granulosa-luteal cells pretreated in vivo with two gonadotropin releasing hormone analogs employing different protocols. Gynecol Endocrinol 1992;6: Guerrero HE, Stein P, Asch RH, de Fried EP, Tesone M. Effect of a gonadotropin-releasing hormone agonist on luteinizing hormone receptors and steroidogenesis in ovarian cells. Fertil Steril 1993;59: Bussenot I, Azoulay Barjonet C, Parinaud J. Modulation of the steroidogenesis of cultured human granulosa lutein cells by gonadotropinreleasing hormone analogs. J Clin Endocrinol Metab 1993;76: Gaetje R. Influence of gonadotropin releasing hormone (GnRH) and a GnRH-agonist on granulosa cell steroidogenesis. Clin Exp Obstet Gynecol 1994;21: Uemura T, Namiki T, Kimura A, Yanagisawa T, Minaguchi H. Direct effects of gonadotropin-releasing hormone on the ovary in rats and humans. Horm Res 1994;41:7s 13s. 26. Minaretzis D, Alper MM, Oskowitz SP, Lobel SM, Mortola JF, Pavlou SN. Gonadotropin-releasing hormone antagonist versus agonist administration in women undergoing controlled ovarian hyperstimulation: Cycle performance and in-vitro steroidogenesis of granulosa-lutein cells. Am J Obstet Gynecol 1995;172: Lin Y, Kahn JA, Hillenjo T. Is there a difference in the function of granulosa-luteal cells in patients undergoing in-vitro fertilization either with gonadotrophin-releasing hormone agonist or gonadotrophin-releasing hormone antagonist? Hum Reprod 1999;14: Weiss JM, Oltmanns K, Gürke EM, Polack S, Eick F, Felberbaum R, et al. Actions of gonadotropin-releasing hormone antagonists on steroidogenesis in human granulosa lutein cells. Eur J Endocrinol 2001;144: Weiss JM, Krautmacher B, Polack S, Diedrich K, Ortmann O. Actions of GnRH antagonists on IGF-II, IGF-binding protein-2 and pregnancyassociated plasma protein-a in human granulosa-lutein cells. Eur J Endocrinol 2003;149: Cunha-Filho JS, Lemos N, Stein N, Laranjeira A, Passo EP. Vascular endothelial growth factor and inhibin A in follicular fluid of infertile patients who underwent in vitro fertilization with a gonadotropinreleasing hormone antagonist. Fertil Steril 2005;83: Sealfon SC, Weinstein H, Millar RP. Molecular mechanisms of ligand interaction with the gonadotropin-releasing hormone receptor. Endocr Rev 1997;18: Millar RP, Lu Z-L, Pawson AJ, Flanagan CA, Morgan K, Maudsley SR. Gonadotropin-releasing hormone receptors. Endocr Rev 2004;25: Millar RP, Pawson AJ. Outside-in and inside-out signalling: the new concept that selectivity of ligand binding at the gonadotropin-releasing hormone receptor is modulated by the intracellular environment. Endocrinology 2004;145: Millar R, Pawson A, Lu Z-L, Morgan K, Davidson L, Lopez de Maturana R, et al. New concepts in GnRH receptor function. In: Filicori M, ed. Updates in infertility treatment Bologna, Italy: Medimont Srl, 2004: Gründker C, Naor Z, Millar RP, Seger R, Leung PC, Emons G. Mechanisms of action of GnRH. In: Lunenfeld B, ed. GnRH analogs in human reproduction. London and New York: Taylor & Francis Group, 2005:1 34. Fertility and Sterility 641