The bone morphogenetic proteins (BMPs) are growth factors involved in the folliculogenesis. Alteration in their expression may compromise the reproductive process in disease such as the polycystic ovary syndrome (PCOS). This study investigated the expression and role of granulosa cell (GC) BMP from normal cycling and PCOS women.
Methods and results
This prospective study was performed in GCs obtained from 14 patients undergoing IVF: i) six women with normal ovulatory cycles and tubal or male infertility and ii) eight women with PCOS. BMP2, BMP4, BMP5, BMP6, BMP7, and BMP8A and their receptors BMPR1A, BMPR1B, and BMPR2 were identified by RT-PCR in GCs from normally cycling and PCOS women. BMP4, BMP6, and BMP7 expressions were confirmed by immunohistochemistry. Quantitative transcript analysis showed the predominant expression of BMP6. In GCs from PCOS women, an overexpression of BMP6 (P<0.01) and BMPR1A mRNA (P<0.05) was observed. GC culture experiments demonstrated that basal estradiol (E2) production was threefold higher but FSH-induced E2 increment was twofold lower in PCOS compared with controls. In PCOS, BMP6 and BMP7 exerted a stimulatory effect on basal E2 production while BMP4 and BMP6 inhibited FSH-induced E2 production. FSH receptor and aromatase expression were not different between both groups.
The BMP system is expressed in human GCs from normal cycling and PCOS women. The BMP may be involved in reproductive abnormalities found in PCOS.
Polycystic ovary syndrome (PCOS) is a common endocrine disease affecting 6–10% of premenopausal women of childbearing age. This clinically heterogeneous disorder appears as the main cause of female infertility (1). The pathophysiology of this ovulatory disorder affecting folliculogenesis is still puzzling. There is evidence for a primary ovarian dysfunction in PCOS reflected by the dramatic increase in the number of small growing follicles at the expense of the dominant follicle selection from the increased pool of selectable follicles (2, 3, 4). Bone morphogenetic proteins (BMPs), which are functionally involved at all stages of folliculogenesis, are multifunctional growth factors belonging to the transforming growth factor β (TGFβ) superfamily. Evidence from studies on rodents indicate an autocrine/paracrine role for granulosa-derived BMP6, theca-derived BMP4 and BMP7, and a paracrine role of oocyte-derived growth differentiation factor 9 (GDF9), BMP15, and BMP6 on granulosa cell (GC) proliferation and on FSH-dependent follicle functions. BMP actions are mediated by their receptors (BMPRs) that share a different degree of affinity for their ligands (5). Unbalanced exposure to BMPs may alter the follicle sensitivity to FSH and thus impair the process of dominant follicle selection. In rodent and mammalian species, GDF9, another member of the BMP family, and BMP15 play a key role on folliculogenesis and female fertility (6, 7, 8, 9, 10). Disruption of follicle development and subsequent infertility were demonstrated in sheep sharing a homozygous mutation of BMP15 gene (9). In rodent ovary, BMP7 contributes to primordial follicle activation and also to the primordial-to-primary follicle transition (11). Later in the follicle growth process, loss of BMP6 expression should play a key role for the dominant follicle selection (12). BMP contribution to follicle selection is driven by suppressive or stimulating effects on FSH action (11, 13, 14). In contrast to rodent, bovine, and ovine species, the role of BMPs in human folliculogenesis has been poorly investigated. BMP6 and BMP7 were shown to stimulate FSH receptor expression in normal human GCs (15, 16). In comparison with healthy ovulatory women, a decrease in GDF9 expression was demonstrated in women with PCOS (10). In light of these data, disturbances at different stages of folliculogenesis may be due to alterations in the expression pattern of BMP genes in PCOS.
In this study, we have investigated the expression pattern of BMPs including BMP4, BMP6, and BMP7 and their receptors BMPR1A, BMPR1B, and BMPR2 in granulosa luteinizing cells from women affected by PCOS, which were compared with healthy ovulatory women. Moreover, the effects of BMPs on estradiol (E2) production in cultured GCs from normally cycling and PCOS women were studied.
Materials and methods
This prospective study concerned women undergoing IVF, women with anovulatory PCOS (n=8), and women with healthy ovulatory function undergoing IVF for tubal and/or male infertility (n=6). Their age ranged between 20 and 35 years. According to Rotterdam criteria 2003, the diagnosis of PCOS was based on the association of two out of three of the following criteria: i) ovulatory dysfunction, oligomenorrhea, or amenorrhea; ii) more than 12 follicles in the 2–9 mm range in each ovary at ultrasonography and/or an ovarian volume >10 ml; and iii) clinical and/or biological hyperandrogenism. Exclusion criteria for control women were: an age >35 years; a history of menstrual disturbance, hirsutism, or endometriosis; high prolactin or androgen serum levels; FSH >10 IU/l at day 3 after menstrual bleeding; and PCOS ovarian pattern at ultrasonography.
All patients gave informed consent before their inclusion in the study. This study was approved by the Institutional Review Board of the University Hospital of Caen. This work is supported by Program Hospitlalier de Recherche Clinique (PHRC), France.
Follicular growth was stimulated by a s.c. injection of recombinant FSH (rFSH; Gonal-F; Merck-Serono) at 150–225 IU/day on the second day of the menses induced by progesterone (Duphaston; Abbott). Follicular growth was monitored by serum E2 level and transvaginal ultrasonography. A GNRH antagonist (Cetrotide; Merck-Serono) was administered i.m. at dose 0.25 mg/day (multiple dose protocol) starting from the day when the dominant follicles reached a mean diameter 14 mm and/or when peripheral E2 level reached 700 pg/ml and was repeated until the day of i.m. administration of human chorionic gonadotropin (hCG). hCG (5000 IU i.m.) was administrated when at least three follicles reached a mean diameter of 18 mm. Oocyte retrieval was performed 36 h later under transvaginal ultrasound guidance.
Isolation of GCs
After oocyte retrieval, all follicular fluids from each patient were pooled and stored in a tube. Cells were collected by centrifugation for 10 min at 300 g. The cell pellet was resuspended in 1 ml minimal essential medium (PAN Biotech, Aidenbach, Germany), supplemented with 100 IU/ml penicillin, 100 μg/ml streptomycin, and 2 μg/ml amphotericin B. The suspension overlayed on 3 ml Ficoll (PAN Biotech) was centrifuged at 600 g for 5 min. Granulosa–luteinized cells were aspirated from the interface, washed with medium, and counted on a hemacytometer. One part of the cells was used for culture and the rest were stored at −80 °C until mRNA and proteins analysis.
RNA extraction and RT
Total mRNA was extracted from GCs using the kit Tri Reagent (Sigma–Aldrich). For each sample, RT was performed in a total volume of 40 μl with 1 μg RNA, 0.2 μg oligo-dT as primers in the presence of dNTP (500 μg; Promega), RNAsin (20 IU; Promega), and Moloney murine leukemia virus reverse transcriptase (200 IU; Promega) for 1 h at 42 °C.
The cDNA obtained was used to evaluate the BMPs' mRNA profile expression in GCs. Amplification was performed at 94 °C for 45 s for denaturing, at 60 °C for 45 s for annealing, and at 72 °C for 1 min for extinction with 1.5 IU Taq DNA polymerase (Promega) in a buffer containing 200 mmol/l dNTP, 1.5 mmol/l MgCl2, and 25 pmol of each primer (Eurogentec, Liege, Belgium) in a total volume of 50 μl. Thirty-three cycles of PCR for FSH receptor and actin were performed. Primers used in supplemental data are summarized in Table 1.
Clinical parameters in normo-ovulatory controls and patients with PCOS. Values are median with 10th–90th percentiles (in parentheses).
|Controls||PCOS patients||P value|
|Age (years)||29.5 (26–33)||29.5 (23–32)||0.85|
|BMI (kg/m2)||21.7 (18–23)||25.4 (19–41)||0.1|
|rFSH totals (IU)||1275 (1065–3690)||1350 (930–1470)||0.57|
|Number of follicles >14 mm||9 (4–12)||10 (3–19)||0.4|
|Number of follicles <14 mm||12 (1–16)||18 (3–30)||0.08|
Real-time RT-PCR analysis
Evaluation of gene expression levels was achieved by real-time quantitative PCR using SYBR Green. Real-time PCR was performed with 3 μl appropriate cDNA using GoTaq qPCR Master Mix (Promega) according to the manufacturer's instruction; forward- and reverse-specific primers for human BMP4, BMP6, BMP7, BMPR1A, BMPR1B, and BMPR2 (Supplementary Table 1, see section on supplementary data given at the end of this article) in a Stratagene Mx 3005P (Agilent Technologies, Santa Clara, CA, USA). The PCR protocol used an initial denaturing step at 95 °C for 3 min followed by 40 cycles at 95 °C for 15 s and annealing at 61 °C for 1 min. Results were analyzed using Mxpro v4.10 Software (Stratagene). Relative gene expression was calculated as a ratio of each target concentration to housekeeping gene actin concentration. The identity of all PCR amplicons was confirmed by sequencing.
Cells were mounted and fixed in 4% paraformaldehyde on slides precoated with polylysine. All the primary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA) and included exclusively those proven suitable for immunohistochemistry by the manufacturer. The primary antibodies included goat polyclonal antibodies against BMP4, BMP6, BMP7, BMPR1A, BMPR2, and aromatase (Supplementary Table 2, see section on supplementary data given at the end of this article). The secondary antibodies used were coupled with FITC for BMP7 and with Texas red for aromatase. Control staining was performed on adjacent serial sections and consisted in replacing the primary antibody with PBS–BSA 3%. Slides were mounted with mounting medium with DAPI (Santa Cruz Biotechnology). Images were captured using a confocal microscope (Olympus Fluo View FV 1000).
Primary cell culture
GCs were cultured in a 24-well plate with 0.5 ml medium (Milieu Essential Minimum de Eagle, Dutcher, Brumath, France) supplemented with SVF at 37 °C in an atmosphere of 5% CO2 in air. After 72-h culture, 10−7 M androstenedione, a substrate for P450 aromatase, was added to the media. GCs were cultured in the presence or absence of 500 mUI/ml FSH and in the presence or absence of BMP4, BMP6, or BMP7. BMP4, BMP6, and BMP7 were purchased from Sigma–Aldrich Co. Ltd. The dose of BMP used was 50 ng/ml as described previously in rat GCs (17). After 48 h of culture, the supernatant of culture media was collected and stored at −20 °C until assay for E2.
Assessment of E2 production
The level of E2 was measured in the 10 μl culture media by a RIA (P.A.R.I.S antibodies). The within-assay coefficient of variation (CV) was 5%, the between-assay CV was 10%, and sensitivity of estrogen assay was 60 pg/ml.
Because values from any parameter were not normally distributed, all comparisons between PCOS and control were performed using the nonparametric Mann–Whitney U test. Univariate analysis of correlation between expressions of the different parameters was performed with the nonparametric Spearman's test (Sigma Stat. V3.5, SYSTAT, Chicago, IL, USA).
Clinical details: Patients
The main clinical parameters from healthy ovulatory women and from patients with PCOS are summarized in Table 1. No differences were found between both groups for age, BMI, and the total dose of rFSH received by each patient. Small follicles were more numerous in PCOS ovaries but this difference was not significant.
BMPs and BMP receptor gene expression in human GCs from normal cycling women
BMP2, BMP4, BMP5, BMP6, BMP7, and BMP8A mRNA expressions were detected in GCs using PCR techniques (Fig. 1). No signal was found for GDF9 and BMP15 (data not shown). These results were confirmed for BMP4, BMP6, and BMP7 by immunohistochemistry (Fig. 2). Co-localization of aromatase and BMP7 confirmed that BMP7 was actually synthesized by human GCs (Fig. 2).
The activin receptor-like kinases 2 (ALK2 (ACVR1)), ALK5, BMPR1A, BMPR1B, and BMPR2 receptors mRNA were also detected in GCs (Fig. 1) and the presence of proteins BMPR1 and BMPR2 was confirmed by immunohistochemistry (Fig. 2).
Differential expression level of BMP4, BMP6, and BMP7 and their receptors in GCs from normal cycling women
All BMP expression levels were expressed as the ratio of BMP mRNA to actin mRNA. We observed different expression levels of the different BMPs and their receptors in GCs from normal cycling women. BMP6 was predominantly expressed in comparison with BMP4 and BMP7 (233±11 vs 16±4 and 2±1 respectively, P<0.01). The BMPR2 receptor was predominantly expressed (1570±318) compared with the BMPR1A (460±110) (Fig. 3).
Expression level of BMPs and their receptors in GCs from PCOS women
BMP6 expression determined by quantitative RT-PCR was higher in GCs from PCOS women compared with healthy women (P=0.01; Fig. 4B). BMPR1A expression was also higher in GCs from PCOS women compared with healthy women (P<0.05; Fig. 4D). Otherwise, no significant differences were observed between both groups for BMP4, BMP7, BMPR1B, and BMPR2 expression levels (Fig. 4A, C, E and F). The BMP6 mRNA and BMPR1A mRNA levels were strongly and positively correlated in healthy women and PCOS women (Spearman's coefficient R=0.752, P<0.01). Moreover, no significant difference between healthy women and PCOS women was found in the mRNA expression level for aromatase gene (7221±4664 vs 15 546±2734), FSH receptor (70.7±81.9 vs 50.6±66.5), and in the progesterone production by GCs (164.9±116.8 pg/50 000 cells vs 226.3±176.3 pg/50 000 cells).
Effect of BMPs on basal and FSH-stimulated estrogen production in cultured GCs from healthy ovulatory and PCOS women
Basal E2 production by cultured GCs was threefold higher in PCOS women compared with control women (108±22.4 vs 380±45.9 pg/50 000 cells respectively, P<0.01; Fig. 5). Although absolute E2 production levels after FSH stimulation were similar in both groups, the FSH-induced E2 increase from baseline was weaker in cells from PCOS compared with healthy women (1.5- vs 3-fold increase respectively; Fig. 5). In cultured cells obtained from normally cycling women, BMP6 and BMP7 but not BMP4 decreased basal E2 production by GCs from healthy women whereas BMPs did not affect basal E2 production in PCOS women (Fig. 6A). FSH-stimulated E2 production by GCs from control ovary was not affected by BMP, whereas in PCOS ovary a dramatic decrease in FSH-stimulated E2 production was observed in the presence of BMP4 and BMP6 (Fig. 6B).
This study is the first to explore the expression pattern of the BMP family in human GCs. For this purpose, we screened mRNA expression of physiologically relevant BMPs and their membrane receptors in human GCs from healthy ovulatory women and from anovulatory PCOS women.
BMPs are the largest subfamily from the TGFβ superfamily. BMPs were initially identified in bone and cartilage (18, 19, 20) and thereafter in other tissues including the ovary (12, 21). These proteins were shown to regulate several key biological processes in the ovary including cell proliferation, differentiation, apoptosis, and steroidogenesis. Alterations in the BMPs system may be involved in human folliculogenesis disorders such as PCOS that represents the more prevalent endocrine disturbance in women of reproductive age.
We describe for the first time the expression pattern of the BMP in human granulosa–luteinized cells using a RT-PCR approach. We demonstrate that BMP2, BMP4, BMP5, BMP6, BMP7, and BMP8A and their BMPR1A, BMPR1B, and BMPR2 receptors are expressed in ovarian GCs from healthy ovulatory women.
Several of our data may be highlighted: first, BMP4 and BMP7 are expressed by human GCs, in contrast to rodents that exclusively express these proteins in theca cells (21). BMP7 synthesis by GCs was confirmed by immunohistochemistry experiments showing the co-localization of BMP7 and aromatase in GCs (22). Such results illustrate the species specificity of the expression pattern of BMP proteins. We were not able to identify any expression of GDF9 and BMP15 in GCs by RT-PCR. Our data do not confirm previous reports showing an expression of GDF9 in human GCs by an approach using nonspecific immunohistochemistry techniques (23). However, these discrepancies may be explained by the effect of the stimulation protocol used.
The study was also designed to investigate the influence of PCOS on the BMP expression. BMP6 and BMPR1A expression levels were higher in GCs from PCOS compared with normal healthy women. The difference in the number of small follicles in PCOS compared with ovary from healthy women may not fully explain BMP6 and BMPR1A overexpression as a progesterone production of similar magnitude by GCs from PCOS and control follicles suggests the same degree of luteinization between both groups.
Our study is the first to report an overexpression of BMPs in PCOS. This result was independent of the IVF stimulation protocols as the same protocol was used in the two groups. There is evidence that PCOS involves a primary ovarian dysfunction (2) characterized by a two- to threefold increase in follicle number together with an arrest of growing follicles at the 2–5 mm stage (3), thus explaining the impairment of dominant follicle selection, a hallmark of PCOS pathogenesis. The mechanism of such follicular arrest in PCOS is not fully explained. Studies of the influence of growth factors such as insulin-like growth factor 1 (24), TGFβ, and anti-Müllerian hormone (AMH) (25) on the follicle growth disturbance in PCOS were realized. Moreover, early appearance of LH receptor in GCs may drive the premature luteinization of GCs observed in PCOS (26).
Our data suggest that the BMP system may be a new protagonist involved in the etiology of the PCOS. BMP6 overproduction by GCs from PCOS ovary may be involved in the antral follicle accumulation and the anovulatory pattern observed in PCOS. In rodents, BMP6 inhibits FSH action on GCs by suppressing adenylate cyclase activity and by reducing progesterone production (14), which is essential for ovulation (27, 28, 29). The drop in BMP6 expression in GCs may be permissive at the time of dominant follicle selection, and in PCOS women, BMP6 overexpression may impair the process of selection. Indeed, in rat GC, BMP6 was shown to inhibit FSH-dependent gene expression of LH receptors (14). Based on these findings, we hypothesized that the early appearance of LH receptor in preantral follicles of PCOS women (24) should be FSH-independent and therefore a BMP6-independent event. In opposition to rodent models, BMP6 increases E2, androstenedione, and inhibin production in sheep (30). Taken together, these data suggest that an increase in BMP6 expression may explain some features related to anovulation and the arrest of dominant follicle selection observed in PCOS women. An in vitro study by Shi et al. (15) found that FSH receptor, inhibin, activin subunits, and AMH were overexpressed by human GCs cultured in the presence of BMP6. In contrast with these results, we observed in this study no difference in FSH receptor expression between normal and PCOS GCs, which overexpressed BMP6. Other authors found that PCOS GCs originating from subsets of small and large follicles overexpressed FSH receptors in comparison with control cells, a finding of which significance is clouded by the larger FSH dose in the control group which may induce downregulation of FSH receptor (31). The difference observed in these studies may be explained by the stimulation protocol used.
BMP2, BMP4, and AMH bind BMPR1A with different degrees of affinity, which results in triggering the intracellular phosphorylation of Smad 1/5/8 and thus induces a signaling pathway responsible for BMP effects. BMPR1A mediates AMH effect on Smad 1 activation and P450 side-chain cleavage enzyme in Sertoli cell line (SMAT-1) (32). BMP4 modulates FSH signaling in a way that inhibits basal and FSH-stimulated progesterone synthesis in the sheep (33). The enhanced expression of BMPR1A in PCOS GCs may contribute to amplify the effect of BMP4 and AMH in GCs. The increased expression of BMP6 and BMPR1A observed in this study may act synergistically in the pathogenesis of anovulation in PCOS women, as suggested by the positive correlation observed between BMP6 and BMPR1A overexpression in PCOS GCs.
We also performed the first functional studies based on the effect of BMPs on steroidogenesis in human cultured GC. In contrast with GCs from normal cycling women, in PCOS GCs, BMP4 or BMP6 induced a dramatic decrease in FSH-stimulated estrogen production while estrogen basal production was not affected. The overproduction of BMP6 by PCOS GCs may not explain such an inhibitory effect, as BMP4 and BMP7 also exerted such an inhibitory effects despite lack of overexpression of these proteins in PCOS GCs.
These results suggest that BMP6 overexpression in PCOS GCs may exert an inhibitory effect on FSH-stimulated E2 production and so may induce some degree of resistance to FSH in PCOS women (34). This contrasts with the absence of such an effect in GCs from normal cycling women and in rodent studies (11, 14). This inhibitory effect of BMP6 could affect aromatase activity as the levels of FSH receptor and aromatase mRNA were not modified in PCOS cells.
In the presence of FSH and BMP in cell culture medium, E2 production level from PCOS GCs is similar to the unstimulated basal E2 production level, highlighting the inhibitory effects of BMP on FSH-dependent steroidogenesis. In our study conditions, the basal E2 level is threefold higher in PCOS compared with normal GCs as described previously by Mason et al. (35), suggesting that BMP effects are mediated mainly by FSH and that enhanced E2 basal production in PCOS GCs is driven by another trigger independently from BMP.
In conclusion, the present data raise the question on the involvement of the BMP system in the pathogenesis of PCOS. The different expression patterns of BMP proteins observed in PCOS compared with normal ovaries suggest that gonadotropin stimulation of PCOS ovaries did not fully reverse the folliculogenesis disturbance present in PCOS, i.e. alterations of expression patterns of BMP proteins that may be involved in follicle development disruption. Further studies are needed to explore the mechanism involved in BMP overexpression in PCOS ovary and to study the signaling pathways involved in the BMP actions on human GCs.
This is linked to the online version of the paper at http://dx.doi.org/10.1530/EJE-12-0891.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This study was supported by PHRC 2009 (Hospital Clinical Research Program), number 0605, France.
MacielGABaracatECBendaJAMarkhamSMHensingerKChangRJEricksonGF. Stockpiling of transitional and classic primary follicles in ovaries of women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism2004895321–5327. (doi:10.1210/jc.2004-0643).
AaltonenJLaitinenMPVuojolainenKJaatinenRHorelli-KuitunenNSeppäLLouhioHTuuriTSjöbergJBützowR. Human growth differentiation factor 9 (GDF-9) and its novel homolog GDF-9B are expressed in oocytes during early folliculogenesis. Journal of Clinical Endocrinology and Metabolism1999842744–2750. (doi:10.1210/jc.84.8.2744).
BodensteinerKJMcNattyKPClayCMMoellerCLSawyerHR. Expression of growth and differentiation factor-9 in the ovaries of fetal sheep homozygous or heterozygous for the inverdale prolificacy gene (FecX(I)). Biology of Reproduction2000621479–1485. (doi:10.1095/biolreprod62.6.1479).
GallowaySMMcNattyKPCambridgeLMLaitinenMPJuengelJLJokirantaTSMcLarenRJLuiroKDoddsKGMontgomeryGW. Mutations in an oocyte-derived growth factor gene (BMP15) cause increased ovulation rate and infertility in a dosage-sensitive manner. Nature Genetics200025279–283. (doi:10.1038/77033).
Teixeira FilhoFLBaracatECLeeTHSuhCSMatsuiMChangRJShimasakiSEricksonGF. Aberrant expression of growth differentiation factor-9 in oocytes of women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism2002871337–1344. (doi:10.1210/jc.87.3.1337).
ShiJYoshinoOOsugaYKogaKHirotaYHirataTYanoTNishiiOTaketaniY. Bone morphogenetic protein-6 stimulates gene expression of follicle-stimulating hormone receptor, inhibin/activin β subunits, and anti-Müllerian hormone in human granulosa cells. Fertility and Sterility2009921794–1798. (doi:10.1016/j.fertnstert.2009.05.004).
MiyoshiTOtsukaFInagakiKOtaniHTakedaMSuzukiJGotoJOguraTMakinoH. Differential regulation of steroidogenesis by bone morphogenetic proteins in granulosa cells: involvement of extracellularly regulated kinase signaling and oocyte actions in follicle-stimulating hormone-induced estrogen production. Endocrinology2007148337–345. (doi:10.1210/en.2006-0966).
ZhaoSYQiaoJChenYJLiuPLiJYanJ. Expression of growth differentiation factor-9 and bone morphogenetic protein-15 in oocytes and cumulus granulosa cells of patients with polycystic ovary syndrome. Fertility and Sterility201094261–267. (doi:10.1016/j.fertnstert.2009.03.014).
JakimiukAJWeitsmanSRNavabAMagoffinDA. Luteinizing hormone receptor, steroidogenesis acute regulatory protein, and steroidogenic enzyme messenger ribonucleic acids are overexpressed in thecal and granulosa cells from polycystic ovaries. Journal of Clinical Endocrinology and Metabolism2001861318–1323. (doi:10.1210/jc.86.3.1318).
RaoIMMaheshVB. Role of progesterone in the modulation of the preovulatory surge of gonadotropins and ovulation in the pregnant mare's serum gonadotropin-primed immature rat and the adult rat. Biology of Reproduction1986351154–1161. (doi:10.1095/biolreprod35.5.1154).
Catteau-JonardSJaminSPLeclercAGonzalèsJDewaillyDdi ClementeN. Anti-Müllerian hormone, its receptor, FSH receptor, and androgen receptor genes are overexpressed by granulosa cells from stimulated follicles in women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism2008934456–4461. (doi:10.1210/jc.2008-1231).