O. W. Stephanie Yap, M.D.,* Yasmin A. Chandrasekher, Ph.D., and Linda C. Giudice, M.D., Ph.D.

Transcription

1 FERTILITY AND STERILITY VOL. 70, NO. 3, SEPTEMBER 1998 Copyright 1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. REPRODUCTIVE BIOLOGY Growth factor regulation of insulin-like growth factor binding protein secretion by cultured human granulosa-luteal cells O. W. Stephanie Yap, M.D.,* Yasmin A. Chandrasekher, Ph.D., and Linda C. Giudice, M.D., Ph.D. Department of Gynecology and Obstetrics, Stanford University Medical Center, Stanford, California Received October 6, 1997; revised and accepted May 8, Supported by Stanford Medical Scholars Program, Stanford, California, and by National Institutes of Health grant HD-31579, Bethesda, Maryland. Presented at the 3rd International Symposium on Insulin-Like Growth Factor Binding Proteins, Sydney, New South Wales, Australia, February 6 10, Reprint requests: Linda C. Giudice, M.D., Ph.D., Department of Gynecology and Obstetrics, East Pavilion HH 333, Stanford University Medical Center, Stanford, California (FAX: ; giudice@stanford.edu). * Present address: Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco /98/$19.00 PII S (98) Objective: To examine the effects of epidermal growth factor (EGF), transforming growth factor- (TGF- ), and fibroblast growth factor (FGF) on insulin-like growth factor binding protein (IGFBP) secretion by cultured human granulosa-luteal cells. Design: Granulosa-luteal cells obtained at the time of oocyte harvest for IVF were cultured in serum-free medium in the presence or absence of EGF, TGF-, or FGF. Conditioned medium then was analyzed by Western ligand blot and immunoradiometric assays. Setting: An academic medical center. Patient(s): Women undergoing IVF. Intervention(s): None. Main Outcome Measure(s): IGFBP-1 secretion. Result(s): By Western ligand blot analysis, IGFBP-1 levels were 1.4-fold to 7.4-fold higher in conditioned medium from cells cultured in the presence of EGF than in control medium. By immunoradiometric assay, IGFBP-1 levels increased from 1.6 to 9.9 times over control. The TGF- had no apparent effect, and FGF did not consistently stimulate IGFBP-1 secretion. Conclusion(s): The EGF may decrease intrafollicular bioavailable IGF levels by increasing inhibitory IGFBPs, thereby leading to arrest of follicular development. Interactions between the EGF and IGF systems may be involved in the processes governing human ovarian follicle maturation and atresia. (Fertil Steril 1998;70: by American Society for Reproductive Medicine.) Key Words: Insulin-like growth factors, IGFBPs, epidermal growth factor, human granulosa-luteal cells Insulin-like growth factors (IGFs) are believed to play a role in cyclic follicular development in the human ovary. The IGFs alone or in synergy with gonadotropins are thought to regulate granulosa and theca cell function and their mitogenic and steroidogenic effects in the human ovary demonstrated in vitro (1). IGF-II has been postulated to have a role in follicle selection and follicular growth in the human ovary because there is localization of abundant IGF-II messenger RNA in theca cells and granulosa cells (2, 3) and there are elevated IGF-II levels in follicular fluid from estrogen-dominant follicles (4). The IGF binding proteins (IGFBPs) bind circulating IGFs and modulate IGF action at the cellular level. Both enhancing and inhibitory IGFBP effects on IGF actions within the ovarian follicle have been postulated (1, 5, 6). Seven IGFBPs have been identified in humans (7, 8), and five of the seven known binding proteins have been found in the human ovary (1). Studies on the regulation of IGFBPs report hcg and IGF-II inhibition of IGFBP-2 production and release by human luteinizing granulosa cells (9). The IGF peptide interactions via nonreceptor-mediated mechanisms are thought to regulate IGFBP-3 release into conditioned medium by luteinizing granulosa cells (9). Polypeptide growth factors have been implicated in the processes of follicular growth and steroidogenesis. Only a few studies have examined the interactions of growth factors with IGFBPs in human granulosa cells. Epidermal growth factor (EGF), for example, has been shown to augment the secretion of IGFBP-1 in human granulosa cells (10) and IGFBP-3 in porcine granulosa cells (11). Other 535

2 growth factors, including transforming growth factor- (TGF- ), fibroblast growth factor (FGF), and platelet-derived growth factor (PDGF), investigated primarily in animal models, are believed to be involved in regulating the processes of follicular development via interaction with the IGF-IGFBP system or independently (1). In this study, we investigated the effects of EGF (a known inhibitor of aromatase activity and present in high levels in follicles with arrested development), TGF- (a growth inhibitor), and FGF (a known mitogen of granulosa cells) on IGFBP secretion and progesterone production by cultured human granulosaluteal cells. MATERIALS AND METHODS Experimental Subjects and Granulosa Cell Cultures Luteinizing granulosa cells were obtained from 11 women undergoing controlled ovarian hyperstimulation for IVF. Briefly, the women were treated with leuprolide acetate (1 mg/d SC; TAP Pharmaceuticals, North Chicago, IL) beginning 2 weeks before an anticipated menses. After days, hmg therapy (225 IU/d IM; Serono Laboratories Inc., Randolph, MA) was begun. Ovulation was induced with 10,000 U of hcg therapy (Serono Laboratories Inc.), and follicles were aspirated transvaginally 35 hours after hcg stimulation. After oocyte removal, the remaining follicular fluid and cells were centrifuged and washed and the granulosa cell button was subjected to 0.1% hyaluronidase treatment in Dulbecco s modified Eagle s medium (DMEM)-F12 (GIBCO, Grand Island, NY). Recovered cells were layered over 50% Percoll and centrifuged at 600 g for 30 minutes. Granulosa cells recovered from the interface were washed, and cell viability was determined by exclusion of 0.1% trypan blue (average viability, 82.4%). Cells were plated at cells per well in 10% fetal bovine serum in DMEM-F12 for 1 day to allow for attachment. Cell monolayers were washed the next day and incubated in a previously described serum-free defined medium (9) for up to 4 days. The defined medium was designated 6F (9) and consisted of DMEM-F12 with the following additions: transferrin, 5 g/ml; aprotinin, 10 g/ml; sodium selenite, 25 nmol/l; L-glutamine, 200 mmol/l; penicillin, 50 U/mL; streptomycin, 50 mg/ml; fungizone, 0.25 mg/ml. Treatments included additions of the following growth factors: EGF (1, 20, 100 ng/ml; Collaborative Research Inc., Bedford, MA); TGF- (1 ng/ml; Upstate Biotechnology, Inc., Lake Placid, NY); and FGF (1, 10 ng/ml; Upstate Biotechnology, Inc., Lake Placid, NY). Conditioned medium was harvested after the indicated time (2 4 days), clarified by centrifugation, and stored at 70 C until assay or Western ligand blot analysis. The protocol was approved by the Stanford University Panel on the Use of Human Subjects in Medical Research and adheres to guidelines of the Helsinki Accords. All patients gave informed consent. Western Ligand Blots and Densitometry Insulin-like growth factor binding proteins in follicular fluid were analyzed according to the method of Hossenlopp et al. (12), as described previously (5). Conditioned medium (0.5 ml) from granulosa cell cultures was first concentrated 6- to 10-fold with use of Centricon-10 microconcentrators (Amicon, Beverly, MA). Nonreduced samples of conditioned medium (30 50 L), nonpregnancy serum (5 L), amniotic fluid (1 L 1:10 dilution), and seminal plasma (5 L) were electrophoresed on 12% sodium dodecyl sulfate (SDS)-polyacrylamide gels and electroblotted onto nitrocellulose. The filter-immobilized proteins were incubated overnight with cpm each of [ 125 I]IGF-I and [ 125 I]IGF-II at 4 C, washed, air dried, and visualized by autoradiography with exposure to Kodak X-Omat AR film (Eastman Kodak Company, Rochester, NY) at 70 C for up to 5 days. The autoradiograms were scanned with a laser densitometer (LKB, Bromma, Sweden). The integrated areas under the absorbance curves for each band were used to determine the relative amounts of IGFBPs. Immunoradiometric Assays for IGFBP-1 The IGFBP-1 in conditioned medium collected from granulosa cell cultures was measured with use of an immunoradiometric assay kit (Diagnostic Systems Laboratories Inc., Webster, TX). The sensitivity of the assay was 0.15 ng/ml. The intraassay and interassay coefficients of variation were 3.7% and 3.5%, respectively. Conditioned medium samples were assayed in triplicate. Hormonal Assays Progesterone secretion was measured by a RIA kit (Diagnostic Products Corporation, Los Angeles, CA). The sensitivity of the assays were 0.1 ng/ml. The intra-assay and interassay coefficients of variation were 5.9% and 10%, respectively. Conditioned samples were assayed in triplicate. Statistical Analysis The IGFBP-1 and progesterone production by granulosa cells among the various treatment groups were analyzed by analysis of variance. Comparisons among the means were made with Scheffé s F-test. RESULTS Secretion of IGFBP by Cultured Granulosa- Luteal Cells Western ligand blot analysis showed the presence of IGFBPs in conditioned medium from cultured granulosaluteal cells (Fig. 1A and B). In accordance with previous studies (9), conditioned medium from luteinizing granulosa cells cultured in basal 6F medium showed a major band at 536 Yap et al. IGFPBs in human granulosa-luteal cells Vol. 70, No. 3, September 1998

3 FIGURE 1 Production of IGFBP by cultured luteinizing granulosa cells. Autoradiographs of Western ligand blots of conditioned medium from granulosa cells cultured in control medium (6F, lane a) and 6F EGF at the concentrations shown (lanes b d). The effect of EGF on IGFBP-1 (migrating as the 28-kDa band) secretion was evident after 2 days in culture. Cultures were established for 24 hours in 10% fetal bovine serum; they were washed, and then the medium was changed to the above treatments. (A and B), Data from cells obtained from two individuals. Conditioned medium from individual wells in culture was analyzed separately. Nonpregnancy serum (NPS) (lane e); seminal plasma (SP) (lane f); and amniotic fluid (AF) (lane g) are shown for reference. The bands corresponding to IGFBP-3, -2, -5, -4* (glycosylated 4), and -4 are also indicated. (C), The densitometric analysis of IGFBP-1 secretion by cultured granulosa cells. Patients 1 and 2 in (C) correspond to the patients in (A) and (B), respectively. Each bar represents the mean SEM of triplicate cultures. The integrated areas under the absorbance curves for each band are shown as arbitrary units per millimeter (AU mm) kd, which has been identified as IGFBP-3. A doublet band present at 31 kd was identified as IGFBP-2 (9). IGFBP-5 and glycosylated IGFBP-4 are also believed to be migrating in this region. Insulin-like growth factor binding FERTILITY & STERILITY 537

4 FIGURE 2 Levels (means SEM) of IGFBP-1 in conditioned medium of triplicate wells by immunoradiometric assay. The IGFBP-1 (ng/ml) in conditioned medium from granulosa cells cultured in the absence and presence of EGF (at the indicated concentrations). protein-1 is only observed in conditioned medium from control cultures as a faint 28-kD band after a prolonged autoradiographic exposure time (9). Meanwhile, a 24-kD species known to be IGFBP-4 (9) is variably detected. Regulation of IGFBP Release by Growth Factors Western ligand blot analysis of conditioned medium from luteinizing granulosa cells cultured in the presence of ng/ml of EGF revealed a specific increase in the levels of the 28-kD IGFBP-1 band compared with control cultures (Fig. 1A and B). Laser densitometry showed that IGFBP-1 levels increased 1.4- to 7.4-fold in conditioned medium of cells cultured in the presence of EGF compared with conditioned medium of control cultures (Fig. 1C). No detectable difference in the levels of the other IGFBPs was noted. The effect of EGF on IGFBP-1 secretion was detectable after 2 days in culture. A minimum of 1 ng/ml of EGF was sufficient to stimulate IGFBP-1 secretion. Quantitative analysis of conditioned medium collected from these granulosa cells after 2 days in culture by immunoradiometric assay confirmed the increase in IGFBP-1 secretion in response to EGF (Fig. 2). By immunoradiometric assay, IGFBP-1 secretion was increased 1.6- to 9.9-fold over controls. The values for IGFBP-1 secreted by granulosa cells cultured in control conditioned medium ranged from 0.5 to ng/ml, whereas the values for IGFBP-1 secreted by granulosa cells cultured in the presence of EGF ranged from to ng/ml (P ). Unlike EGF, TGF- had no apparent effect on IGFBP-1 secretion (data not shown). Transforming growth factor- (1 ng/ml) was likewise without effect on the secretion of the other IGFBPs on Western ligand blot analysis. Treatment with FGF (1 10 ng/ml) did not stimulate IGFBP-1 secretion consistently (data not shown). Similarly, FGF did not affect the release of other binding proteins by luteinizing granulosa cells. Effects of EGF on Progesterone Secretion Progesterone values after 2 days of cell culture varied between and 1, ng/ml for control conditioned medium and varied between and 2, ng/ml for conditioned medium from EGFtreated granulosa-luteal cells. This 1.3- to 1.9-fold stimulatory effect of EGF on progesterone secretion by luteinizing granulosa cells was not significant (P 0.117). DISCUSSION The IGFBPs are produced in the human ovary (3, 9, 13, 14), and studies suggest a role for them in follicular development. It is thought that a complex system of inter- 538 Yap et al. IGFPBs in human granulosa-luteal cells Vol. 70, No. 3, September 1998

5 actions involving growth factors, hormones, and IGFBP proteases regulate IGFBP levels, and hence IGF effects on cells. Epidermal growth factor is present in human follicular fluid (15), and EGF and EGF receptors are present in human granulosa and theca cells (16). Cultured granulosa-luteal cells express IGFBP-1 messenger RNA (9, 17) and secrete IGFBP-1 (17 19). De novo synthesis of IGFBP-1 by granulosa-luteal cells in culture has been documented (18, 20). Cultured human granulosa-luteal cells secrete IGFBP-1, -2, and -3, as well as variable amounts of IGFBP-4. Incubation of luteinizing granulosa cells in serum-free medium with EGF resulted in a specific increase in IGFBP-1 secretion. According to Western ligand blot analysis, EGF concentrations of ng/ml stimulated IGFBP-1 levels 1.4- to 7.4-fold compared with controls. Immunoradiometric assays confirmed the increase in IGFBP-1 production (1.6- to 9.9-fold) from cells cultured in the presence of EGF. The data presented in this study are consistent with previous immunofluorometric studies that had demonstrated a 1.2- to 4.3-fold stimulation of IGFBP-1 secretion by cultured luteinizing granulosa cells (10). Results reported in that series of experiments noted that the stimulatory effect of EGF was observed at 10 ng/ml. Our data indicated that 1 ng/ml of EGF significantly and specifically stimulated IGFBP-1 secretion by day 2 in culture. The profiles of the other binding proteins remained unchanged in control versus EGF-containing medium. In a neonatal rat model, a similar stimulatory effect of EGF on serum IGFBP-1 levels has been reported (21). A specific increase in hepatic IGFBP-1 messenger RNA preceded the rise in serum IGFBP-1 on infusion of EGF, although the exact mechanism of action remains to be elucidated. The investigators suggested that elevated IGFBP-1 levels decreased bioavailable IGFs, hence the observed associated decrease in perinatal growth. A number of studies have reported the effect of EGF on progesterone production by cultured granulosa cells (10, 22, 23). We describe a nonsignificant increase 1.3- to 1.9-fold over controls in progesterone secretion by granulosa-luteal cells cultured in the presence of EGF. Angervo et al. (10) noted a slight stimulatory effect on progesterone secretion by EGF. Serta and Seibel (22) reported that early and continuous exposure to EGF (beginning on culture day 4) maximally stimulated progesterone production. They also found that progesterone secretion by human granulosa-luteal cells was not enhanced by treatment with EGF during the first 3 days in culture. Our data support reports that EGF has a luteinizing effect on human granulosa cells (23, 24). The present study found that TGF- treatment had no apparent effect on IGFBP-1 secretion. Contrary to data on cultured granulosa cells from polycystic ovary syndrome follicles that showed TGF- stimulation of IGFBP-3 production (21), our data on granulosa-luteal cells from hyperstimulated follicles from normo-ovulatory women did not show any changes in IGFBP-3 levels in conditioned medium from control versus TGF- -treated cells by Western ligand blot analysis. Fibroblast growth factor, on the other hand, did not augment IGFBP-1 secretion consistently at a dose of 10 ng/ml. Its role as an intraovarian modulator of human follicle development remains to be clarified. Furthermore, TGF- and FGF did not have a stimulatory effect on progesterone secretion by luteinizing granulosa cells (data not shown). In conclusion, it is likely that complex interactions exist between the EGF and IGF systems in human ovarian follicle maturation and development. The precise mechanism of this interaction remains to be identified. Epidermal growth factor modulation of IGFBPs is likely to be cell type specific. In human cervical epithelial cells, for example, EGF acts as a positive regulator of IGF-I action by suppressing IGFBP-3 production (25). In the human ovary, EGF, a known inhibitor of granulosa cell estradiol production, may subserve its role in the process of luteinization by stimulating IGFBP-1 production and/or secretion with the resultant increased IGFBP-1 levels reducing intrafollicular bioavailable IGFs. References 1. Giudice LC, Cataldo NA, van Dessel HJHM, Yap OWS, Chandrasekher YA. Growth factors in normal ovarian follicle development. Semin Reprod Endocrinol 1996;14: El-Roeiy A, Chen X, Roberts VJ, LeRoith D, Roberts CT, Yen SSC. Expression insulin-like growth factor-i (IGF-I) and IGF-II and the IGF-I, IGF-II, and insulin receptor genes and localization of the gene products in the human ovary. J Clin Endocrinol Metab 1993;77: Zhou J, Bondy C. Anatomy of the human ovarian insulin-like growth factor system. Biol Reprod 1993;48: van Dessel HJHM, Chandrasekher Y A, Yap OWS, Lee PDK, Hintz RL, Faessen GHJ, et al. Serum and follicular fluid levels of insulin-like growth factor (IGF)-I, IGF-II, and IGF binding proteins -1 and -3 during the normal menstrual cycle. J Clin Endocrinol Metab 1995;81: Cataldo NA, Giudice LC. IGFBP profiles in human ovarian follicular fluid correlate with follicular functional status. J Clin Endocrinol Metab 1992;74: San Roman GA, Magoffin DA. Insulin-like growth factor binding proteins in healthy and atretic follicles during natural menstrual cycles. J Clin Endocrinol Metab 1993;76: Shimasaki S, Ling N. Identification and molecular characterization insulin-like growth factor binding proteins (IGFBP-1, -2, -3, -4, -5, and -6). Prog Growth Factor Res 1992;3: Oh Y, Nagall SR, Yamanaka Y, Kim HS, Wilson E, Rosenfeld RG. Synthesis and characterization of insulin-like growth factor-binding protein (IGFBP)-7. Recombinant human mac25 protein specifically binds IGF-I and -II. J Biol Chem 1996;271: Cataldo NA, Woodruff TK, Giudice LC. Regulation of insulin-like growth factor binding protein production by human luteinizing granulosa cells cultured in defined medium. J Clin Endocrinol Metab 1992; 76: Angervo M, Koistinen R, Seppälä M. Epidermal growth factor stimulates production of insulin-like growth factor binding protein-1 in human granulosa-luteal cells. J Endocrinol 1992;134: Mondschein JS, Smith SA, Hammond JM. Production of insulin-like growth factor binding proteins (IGFBPs) by porcine granulosa cells: identification of IGFBP-2 and -3 and regulation by hormones and growth factors. Endocrinology 1990;127: Hossenlopp P, Seurin D, Segovia-Quinson B, Hardouin S, Binoux M. Analysis of serum insulin-like growth factor binding proteins using western blotting: use of the method for titration of the binding proteins and competitive binding studies. Anal Biochem 1986;154: San Roman GA, Magoffin DA. Insulin-like growth factor binding proteins in ovarian follicles from women with polycystic ovarian dis- FERTILITY & STERILITY 539

6 ease: cellular source and levels in follicular fluid. J Clin Endocrinol Metab 1992;75: El-Roeiy A, Chen X, Roberts VJ, Shimasaki S, Ling N, LeRoith D, et al. Expression of the genes encoding the insulin-like growth factors, the IGF and insulin receptors, and the IGF-binding proteins-1-6 and the localization of their gene products in normal and polycystic syndrome ovaries. J Clin Endocrinol Metab 1994;78: Hofmann GE, Scott RT Jr, Brzyski RG, Jones HW Jr. Immunoreactive epidermal growth factor concentrations in follicular fluid obtained from in vitro fertilization. Fertil Steril 1990;54: Maruo T, Ladines-Llave CA, Samoto T, Matsuo H, Manalo AS, Ito H, et al. Expression of epidermal growth factor and its receptor in the human ovary during follicular growth and regression. Endocrinology 1993;132: Giudice LC, Milki AA, Milkowski DA, Danasouri IE. Human granulosa contain messenger ribonucleic acids encoding insulin-like growth factor binding proteins (IGFBPs) and secrete IGFBPs in culture. Fertil Steril 1991;56: Jalkanen J, Suikkari A-M, Koistinen R, Butzow R, Seppälä M, Ranta T. Regulation of insulin-like growth factor binding protein-1 production in human granulosa luteal cells. J Clin Endocrinol Metab 1989;69: Hamori M, Blum WF, Török A, Stehle R, Waibel E, Cledon P, et al. Immunoreactive insulin-like growth factor binding protein-3 in the culture of human luteinized granulosa cells. Acta Endocrinol (Copenh) 1991;124: Suikkari A-M, Jalkanen J, Koistinen R, Butzow R, Ritvos O, Ranta T, et al. Human granulosa cells synthesize low molecular weight insulinlike growth factor binding protein. Endocrinology 1989;124: Murray MA, Dickson BA, Smith EP, Hoath SB, Chernausek SD. Epidermal growth factor stimulates insulin-like growth factor binding protein-1 expression in the neonatal rat. Endocrinology 1993;133: Serta RT, Seibel MM. The influence of epidermal growth factor on progesterone production by human granulosa-luteal cells in culture. Hum Reprod 1993;8: Richardson MC, Gadd SC, Masson GM. Augmentation by epidermal growth factor of basal and stimulated progesterone production by human luteinized granulosa cells. J Endocrinol 1989;121: Steinkampf M, Mendelson C, Simpson E. Effects of epidermal growth factor insulin-like growth factor I on the levels of mrna encoding aromatase cytochrome P-450 of human ovarian granulosa cells. Mol Cell Endocrinol 1988;59: Hembree JR, Agarwal C, Eckert RL. Epidermal growth factor suppresses insulin-like growth factor binding protein 3 levels in human papillomavirus type 16-immortalized cervical epithelial cells and thereby potentiates the effects of insulin-like growth factor I. Cancer Res 1994;54: Yap et al. IGFPBs in human granulosa-luteal cells Vol. 70, No. 3, September 1998