Abstract
Women with polycystic ovary syndrome (PCOS) undergoing IVF–embryo transfer based-assisted reproductive technology (ART) treatment show variable ovarian responses to exogenous FSH administration. For better understanding and control of PCOS ovarian responses in ART, the present study was carried out to compare the follicular hormones and the expression of granulosa cell genes between PCOS and non-PCOS women during ART treatment as well as their IVF outcomes. Overall, 138 PCOS and 78 non-PCOS women were recruited for the present study. Follicular fluid collected from PCOS women showed high levels of testosterone. The expression of aromatase was found significantly reduced in luteinized granulosa cells from PCOS women. In cultured luteinized granulosa cells isolated from non-PCOS women, their exposure to testosterone at a level that was observed in PCOS follicles could decrease both mRNA and protein levels of aromatase in vitro. The inhibitory effect of testosterone was abolished by androgen receptor antagonist, flutamide. These results suggest that the hyperandrogenic follicular environment may be a key hazardous factor leading to the down-regulation of aromatase in PCOS.
Introduction
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder found in 6–10% women of reproductive age and represents a leading cause of female infertility (Goodarzi et al. 2011, Jayasena & Franks 2014). Ovarian follicular development is altered in PCOS, where most ovarian follicles, after leaving the primordial follicle resting pool, are arrested at the small antral stage and fail to complete their maturation for ovulation. This results in the accumulation of a high number of small antrum follicles in the ovary, a morphology of polycystic ovary (Dumesic et al. 2008, Dumesic & Richards 2013).
The ovarian morphological abnormality is accompanied with hormonal disruption in PCOS, including significant increases in androgens and luteinizing hormone (LH), and reduction in follicle-stimulating hormone (FSH) (McCartney et al. 2002). LH is one of the pituitary-released gonadotropins, which, by activating its receptors in theca cells of small secondary follicles, induces the production of theca cell-derived androgens (Richards et al. 1987). In PCOS, LH is hypersecreted contributing to the excessive androgen level (Blank et al. 2006). FSH is another gonadotropin essential to both follicle maturation and ovarian steroidogenesis (Payne & Hales 2004). It activates FSH receptors (FSHRs) in granulosa cells of antral follicles leading to the transcription of aromatase (CYP19), a key enzyme that catalyzes the conversion of androgens to estrogens (Payne & Hales 2004, Stocco 2008), which is required for follicle maturation (Chen et al. 2012). In PCOS, the expression of FSHR in granulosa cells appears to be up-regulated (Catteau-Jonard et al. 2008), which is believed to be responsible for the observed hyperresponsiveness of PCOS granulosa cells to FSH both in vitro and in vivo (Erickson et al. 1992, Mason et al. 1994, Coffler et al. 2003). Although FSH hypersensitivity accelerates estrogen production, PCOS women were found unable to sustain their estrogen level (Coffler et al. 2003).
PCOS women often seek fertility treatment by assisted reproductive technology (ART), where pituitary suppression followed by exogenous FSH administration is a standard procedure to stimulate the ovary and maximize follicle maturation and ovulation for IVF. We have observed in the IVF clinic that PCOS patients undergoing ART treatment have various ovarian responses to exogenous FSH, from insufficient to over-stimulated, probably because of the complicated FSH-regulatory system in PCOS (Erickson et al. 1992, Mason et al. 1994, Coffler et al. 2003). The following procedures for ART often had to be cancelled either due to the lack of oocyte numbers or to avoid possible development of ovarian hyperstimulation syndrome, a life-threatening complication.
For better understanding and control of PCOS ovarian responses in ART, the present study was carried out to compare the follicular hormones and granulosa cell gene expression between PCOS and non-PCOS women during their ART treatment as well as their IVF–embryo transfer (ET) outcomes, including the number of oocytes retrieved, fertilization rate, percentage of high quality grade embryo, implantation, and clinical pregnancy rates. Follicular fluid collected from PCOS women, particularly small follicles, showed high levels of testosterone. Whereas, the expression of aromatase was found significantly reduced in luteinized granulosa cells from PCOS women. In isolated non-PCOS human luteinized granulosa cells, we also demonstrated that exposure of the cells to testosterone at a level that was observed in PCOS follicles down-regulated aromatase in vitro.
Materials and methods
Patients
A total of 216 female patients at the IVF center of Shanghai First Maternity and Infant Hospital from June 2012 to June 2013 were recruited for the present study. Among the patients, 138 were diagnosed with PCOS based on the presence of at least two of the flowing criteria (Ng et al. 2005): i) ovulatory disturbance (oligomenorrhea or amenorrhea), ii) hyperandrogenism as defined by hirsutism, seborrhea, and/or testosterone >0.7 ng/ml and/or androstenedione >2.2 ng/ml as measured on day 3 of the menstrual cycle, and iii) the presence of more than 12 follicles of 2–9 mm in diameter in each ovary under B-ultrasound and/or ovarian volume higher than 10 ml. The other 78 women met the following inclusion criteria and were grouped as non-PCOS: i) both ovaries were present, ii) menstrual cycle length was between 25 and 35 days, iii) there were no current or past diseases affecting the ovaries, gonadotropin or sex steroid secretion, iv) there were no clinical signs of hyperandrogenism, and v) FSH levels were ≤10 mIU/ml on day 3 of the cycle. Causes of infertility in the non-PCOS group include tubal blockage (n=46), pelvic adhesions (n=11), or male factor (n=21). All the procedures were approved by the Ethics Committee of Tongji University and all patients have given informed consents.
Clinical data collection
A blood test was performed on the 2nd–5th day of the menstrual cycle before the treatment to determine basal levels of estradiol (E2), testosterone, progesterone, LH, and FSH in the patients. During treatment, patient data were documented, including age and BMI. IVF results include follicle number, numbers of collected oocytes, fertilization rate, cleavage rate, number of available embryos and transferred embryos, good embryo quality rate (grades I and II), cryopreserved embryo number, and implantation rate.
Ovarian hyperstimulation
Diphereline (Ipsen Pharma Biotech, Cambridge, UK; 1.25 or 1.88 mg), a gonadotropin-releasing hormone agonist, was intramuscularly injected at the mid-luteal phase of the menstrual cycle in each patient, to suppress pituitary secretion of gonadotropin hormones and prevent premature ovulation. After pituitary suppression was achieved as evidenced by plasma E2 levels of ≤50 pg/ml, the absence of ovarian follicles and endometrial thickness ≤6 mm by transvaginal ultrasound examination (Barash et al. 1998), patients were daily injected with Gonal-F (Merck–Serono), a recombinant human FSH (rhFSH), starting from the 5th day of the menstrual cycle. The initial rhFSH dose was determined by a variety of factors, including age, number of antral follicles, basal FSH level, and history of ovarian response. Trans-vaginal ultrasound using a 5-MHz vaginal transducer attached to a sector scanner (Model SSD-620, Aloka, Tokyo, Japan) was performed and serum sex hormones were assessed to monitor follicular development and adjust the dose of rhFSH. When the lead follicle achieved 18 mm in diameter, the lead two were 17 mm or the lead three were 16 mm, patients were subcutaneously injected with recombinant human chorionic gonadotropin (hCG, Ovidrel, Merck–Serono, 250 μg) to trigger oocyte maturation. Blood was collected right before the injection of first rhFSH and hCG respectively.
Oocyte retrieval, follicle fluid, and luteinized granolosa cell collection
Thirty-four to 36 h after the administration of hCG, ovarian follicles were aspirated using a single-lumen, 17-gauge needle (Cook Medical, Bloomington, IN, USA) guided by trans-vaginal ultrasonography. Follicles were classified into two groups, the large follicle group with the diameter >18 mm and the small follicle group with diameter <10 mm as measured by ultrasonography. All the samples were collected by the same operator to ensure the size of the follicles. Large follicles were used to isolate luteinized granulosa cells. The oocytes were dissected from the collected follicles under a dissecting microscope. After collecting the oocytes, the remaining granulosa cells with fluid were transferred into sterile tubes (Axygen Scientific, Union City, CA, USA) and centrifuged at 200 g for 10 min. Afterwards, the supernatant was aspirated and collected as follicular fluid and cell pellets were washed with PBS (Beyotime, Shanghai, China) and subsequently a red blood cell lysis buffer (Beyotime) to eliminate red blood cells. The cells were stored at −80 °C or re-suspended for culture.
IVF and ET
The oocytes isolated were used for IVF or ICSI based on the condition of semen. No more than three embryos were transferred into the uterine cavity 2–3 days after IVF or ICSI. Patients were intramuscularly injected with progesterone (60 mg/day, Tongyong Pharmaceutical Co., Shanghai, China) from the day oocyte retrieval was performed till 14 days after ET. Serum β-hCG was measured 14 days after ET and clinical pregnancy was defined by the presence of an intrauterine gestational sac with fetal heartbeats 3–4 weeks after ET.
Luteinized granulosa cell culture
The isolated luteinized granulosa cells from non-PCOS group were seeded at 0.75×106 cells/ml into six-well plates and cultured in DMEM/F12 (Gibco) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin and streptomycin sulfate (Life Technologies, Inc.) at 37 °C in a 5% CO2 incubator. On the next day after seeding the cells, testosterone (5, 15, or 60 ng/ml, Sigma) or the solvent was added into the culture medium. Forty-eight hours after the treatment with testosterone, the cells were collected for subsequent analysis.
RNA isolation and real-time PCR analysis
RNAs were extracted from cells using the TRIzol reagent (Invitrogen) according to the manufacturer's instruction. The extracted RNAs that show distinct 18s and 28s bands in RNA electrophoresis were used in subsequent experiments. The quantity of RNA was measured by the NanoDrop 2000c U.v.–Vis Spectrophotometer (Thermo Scientific, Waltham, MA, USA). RNAs were reversely transcribed into cDNA with random primers (Takara, Shiga, Japan) in a 20 μl reaction buffer at 37 °C for 15 min, and terminated by heating at 85 °C for 5 s followed by cooling at 4 °C. The sequences of primer pairs used in the present study are shown in Table 1. The primers were verified with their efficiency close to 100%. Quantitative real-time PCR was performed on the PCR system (Roche Diagnostics Ltd) with the SYBR reagent (Takara Bio, Inc., Otsu, Japan). The reaction was performed at 95 °C for 10 s and followed by 40 cycles of heating at 95 °C for 5 s and subsequently 60 °C for 34 s. The dissociation stage was initiated at 95 °C for 15 s, followed by one cycle of 60 °C for 1 min and 95 °C for 15 s. PCRs were carried out in triplicate for each sample. Gene expression was analyzed using housekeeping gene β-actin and cycle threshold (Ct) method. Fold-induction values (x) were calculated using the following formula, x=2−ΔΔCt, where Ct is the mean value of all replicates of a given gene, ΔCt is the difference between the Ct value of the gene in target and that of β-actin, and ΔΔCt is the difference between ΔCt values of the samples for each target and the ΔCt of a control sample.
Sequences of primers.
| Gene | Forward sequence | Reverse sequence | Product size (bp) |
|---|---|---|---|
| FSHR | TCTGTCACTGCTCTAACAGGG | TGCACCTTTTTGGATGACTCG | 131 |
| AR | CCTGGCTTCCGCAACTTACAC | GGACTTGTGCATGCGGTACTCA | 168 |
| LHCGR | ATTTGTCAATCTCCTGGAGGC | CACTCAGTTCACTCTCAGCA | 191 |
| FLK | GGCCCAATAATCAGAGTGGCA | TGTCATTTCCGATCACTTTTGGA | 105 |
| IGF1R | TGGTGGAGAACGACCATATCC | CGATTAACTGAGAAGAGGAGTTCGA | 123 |
| CYP19 | TGGAAATGCTGAACCCGATAC | AATTCCCATGCAGTAGCCAGG | 161 |
| CYP17 | GCTGCTTACCCTAGCTTATTTGT | ACCGAATAGATGGGGCCATATTT | 174 |
| PPARG | GGGATCAGCTCCGTGGATCT | TGCACTTTGGTACTCTTGAAGTT | 186 |
| β-actin | TCATGAAGTGTGACGTGGACATC | CAGGAGGAGCAATGATCTTGATCT | 156 |
Western blot
Cells were lysed in RIPA buffer (Beyotime) with proteases inhibitor and phenylmethylsulphonyl fluoride for 30 min on ice before centrifuged at 15 000 g for 20 min at 4 °C. Afterwards, the supernatant was collected and the protein concentration was determined by a BCA assay (Pierce Biotechnology, Rockford, IL, USA) and using a microplate reader (Multiskan MK3, Thermo Scientific). Proteins were mixed with the SDS sample buffer and boiled for 10 min before being separated by SDS–PAGE with 5% stacking gel and 10% separating gel (Beyotime) at 100 V for around 2 h and subsequently transferred to PVDF membranes (Millipore, Bedford, MA, USA). The membranes were blocked in 5% non-fat milk in TBST (0.01 M Tris–HCl, 0.15 M NaCl, 0.1% Tween 20, pH 7.4, Beyotime) for 1 h at room temperature and subsequently the antibody against aromatase (Cell Signaling Technology, Danvers, MA, USA, 1:1000) at 4 °C overnight. The membranes were then washed with TBST three times before being incubated with the HRP-conjugated secondary antibody (Proteintech, Chicago, IL, USA, 1:2000) for 1 h at room temperature. The blots were visualized by ECL (Millipore, Watford, UK) and the images were taken by FluorChem E (Protein Simple, San Francisco, CA, USA). Quantification of the blots was performed by the freely available ImageJ Software.
Steroid hormone assay
E2, testosterone, progesterone, LH, and FSH were measured by an automated chemiluminescence immunoassay analyzer (Siemens, Tarrytown, NY, USA). Intra- and inter-assay coefficients of variations were 6.1–8.7, 5–6.7, 1.4–2.9, 1.6–3.3, and 2.2–5.7% respectively. All samples were operated at least three times and the mean values were used for statistical analysis. The plates were read by an ELISA reader (Toyo-Sokuki Inc., Tokyo, Japan).
Statistical analysis
Data are mean±s.e.m. Student's t-test was used to compare PCOS and non-PCOS groups. One-way ANOVA followed by Tukey's post hoc tests was used to compare more than two groups. P<0.05 was considered statistically significant.
Results
Clinical results
Basic clinical characteristics of the patients for the present study are shown in Table 2. The PCOS patients were about 3 years younger than the non-PCOS patients on average. No significant difference in years of infertility was observed between the two groups. Although obesity is often seen in PCOS cases, the PCOS patients recruited in the present study showed normal BMI with no significant difference with the non-PCOS group. The PCOS patients were treated with the rhFSH at a lower initial dose for a longer administration duration in comparison to the non-PCOS patients, although similar total amounts of rhFSH was administrated to patients of both groups.
Basic characteristics of PCOS and non-PCOS women.
| Variables | Non-PCOS (n=78) | PCOS (n=138) | P value |
|---|---|---|---|
| Age (years) | 31.93±0.46 | 29.57±0.54 | <0.05 |
| BMI (kg/m2) | 21.33±0.28 | 22.62±0.64 | >0.05 |
| Years of infertility | 5.08±0.29 | 4.29±0.29 | >0.05 |
| Initial dose of rhFSH (IU) | 181.7±5.46 | 149.1±4.64 | <0.001 |
| Total dose of rhFSH (IU) | 2177.83±88.04 | 2058.9±140.40 | >0.05 |
| Duration of rhFSH (day) | 11.82±0.30 | 13.2±0.65 | <0.05 |
rhFSH, recombinant human FSH. Data other than P values are mean±s.e.m., n=number of patients.
Serum levels of the hormones measured are shown in Table 3. The PCOS group showed significantly higher basal LH and lower basal FSH levels as compared to the non-PCOS group (P<0.01), although basal E2 levels were found to be similar between the two groups. After pituitary suppression (initial day), all the hormones measured including LH, FSH, and E2 dropped to a low level in both PCOS and non-PCOS groups and no significant difference in the residual hormones was found between them. On the day when hCG was injected (hCG day), no difference in progesterone and E2 between the two groups was found. The LH level remained low on the hCG day in both groups with the non-PCOS group showing relatively higher LH levels.
Hormonal levels in PCOS and non-PCOS women.
| Variables | Non-PCOS (n=78) | PCOS (n=138) | P value |
|---|---|---|---|
| Basal | |||
| FSH (IU/l) | 7.53±0.18 | 6.39±0.25 | <0.01 |
| LH (IU/l) | 4.53±0.25 | 8.67±1.57 | <0.01 |
| LH/FSH | 0.57±0.05 | 1.3±0.18 | <0.001 |
| E2 (ng/ml) | 61.53±6.10 | 63.09±11.50 | >0.05 |
| Initial day | |||
| FSH (IU/l) | 3.21±0.21 | 3.05±0.22 | >0.05 |
| LH (IU/l) | 1.99±0.12 | 1.44±0.19 | >0.05 |
| E2 (ng/ml) | 22.15±1.07 | 22.53±2.22 | >0.05 |
| hCG day | |||
| Progesterone (ng/ml) | 1.17±0.29 | 0.81±0.16 | >0.05 |
| LH (IU/l) | 1.81±0.18 | 1.18±0.19 | <0.05 |
| E2 (ng/ml) | 2125.23±113.9 | 2744.96±311.05 | >0.05 |
Data other than P values are mean±s.e.m., n=number of patients.
ART related parameters are shown in Table 4. A larger number of follicles were collected from the PCOS than non-PCOS group, although similar numbers of oocytes were obtained from the collected follicles of the two groups. Thus, the oocyte obtaining rate (oocyte number per follicle number) was lower in the PCOS than non-PCOS group. The oocytes obtained from the PCOS group showed no difference in fertilization rate and cleavage rate as compared with non-PCOS ones. The embryos derived from PCOS oocytes showed similar good quality rate, implantation rate, and pregnancy rate to those in non-PCOS ones.
IVF outcomes of PCOS and non-PCOS women.
| Variables | Non-PCOS (n=78) | PCOS (n=138) | P value |
|---|---|---|---|
| Follicle number | 11.95±0.62 | 18.0±1.23 | <0.001 |
| Oocyte number | 9.00±0.85 | 10.26±0.96 | >0.05 |
| Oocyte obtaining rate (oocyte number per follicle) | 0.89±0.06 | 0.60±0.05 | <0.01 |
| Fertilization rate | 0.70±0.02 | 0.77±0.04 | >0.05 |
| Cleavage rate | 0.84±0.03 | 0.92±0.02 | >0.05 |
| Available embryo number | 3.99±3.03 | 5.6±4.16 | <0.05 |
| Good quality embryo rate | 0.69±0.06 | 0.61±0.05 | >0.05 |
| Cryopreserved embryo number | 2.39±3.06 | 4.64±4.31 | <0.05 |
| Embryo transfer number | 1.6±1.08 | 0.95±1.01 | <0.01 |
Data other than P values are mean±s.e.m., n=number of patients.
Analysis of follicular fluid
We were able to collect a sufficient amount of follicular fluid from 25 patients in each group, PCOS or non-PCOS, of the cohort for the analysis of follicular hormones. Both PCOS and non-PCOS groups showed low follicular LH levels (below 0.5 IU/l, Fig. 1A) with no difference between the two groups probably due to pituitary suppression. The FSH levels in both small and large follicles were found significantly reduced in PCOS compared to that in non-PCOS patients (Fig. 1B). The testosterone levels in both large and small follicles from PCOS patients were found to be respectively higher than those in non-PCOS patients, although only in small follicles, the difference between PCOS and non-PCOS is statistically significant (Fig. 1C). The E2 level in follicles showed no significant difference between PCOS and non-PCOS (Fig. 1D). In both small and large follicles, the progesterone levels were found significantly lower in PCOS than those in non-PCOS (Fig. 1E).
Hormones in follicles from PCOS and non-PCOS patients with ART treatment. The concentration of LH (A), FSH (B), testosterone (C), E2 (D), and progesterone (E) was measured in small and large follicles collected from PCOS and non-PCOS patients during their ART treatment. Data are means±s.e.m., n=25. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Hormones in follicles from PCOS and non-PCOS patients with ART treatment. The concentration of LH (A), FSH (B), testosterone (C), E2 (D), and progesterone (E) was measured in small and large follicles collected from PCOS and non-PCOS patients during their ART treatment. Data are means±s.e.m., n=25. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Hormones in follicles from PCOS and non-PCOS patients with ART treatment. The concentration of LH (A), FSH (B), testosterone (C), E2 (D), and progesterone (E) was measured in small and large follicles collected from PCOS and non-PCOS patients during their ART treatment. Data are means±s.e.m., n=25. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Gene expression in luteinized granulosa cells and down-regulation of aromatase in PCOS
Genes that are associated with granulosa cell functions including FSHR, androgen receptor (AR), insulin-like growth factor 1 receptor (IGF1R), vascular endothelial growth factor receptor (FLK), LH receptor (LHCGR), CYP17, CYP19, and peroxisome proliferator-activated receptor gamma (PPARG) were compared between PCOS and non-PCOS. We were able to collect a sufficient number of luteinized large (>18 mm) follicle granulosa cells from 20 to 35 patients in each group, PCOS or non-PCOS, of the whole cohort for real-time PCR analysis of each gene. Among these genes, PCOS luteinized granulosa cells showed significantly increased transcripts of FSHR, and decreased LHCGR and CYP19 as compared to the non-PCOS group (Fig. 2). To confirm that aromatase (CYP19) expression is affected in PCOS, proteins were extracted from luteinized large follicle granulosa cells for western blot analysis, which showed that the protein level of aromatase was also down-regulated in PCOS group as compared to the non-PCOS group (Fig. 3).
Real-time PCR analysis of genes expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Data are means±s.e.m., n is indicated above each bar. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Real-time PCR analysis of genes expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Data are means±s.e.m., n is indicated above each bar. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Real-time PCR analysis of genes expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Data are means±s.e.m., n is indicated above each bar. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. *P<0.05, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Western blot analysis of aromatase expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Gapdh was used as a loading control. Data are means±s.e.m., n=12. ***P<0.001, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Western blot analysis of aromatase expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Gapdh was used as a loading control. Data are means±s.e.m., n=12. ***P<0.001, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Western blot analysis of aromatase expression in granulosa cells freshly isolated from PCOS and non-PCOS patients with ART treatment. Gapdh was used as a loading control. Data are means±s.e.m., n=12. ***P<0.001, t-test.
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Testosterone-induced downregulation of aromatase in human granulosa cells in vitro
The transcription of CYP19 is known to be promoted by FSH. Since that similar total amount of exogenous FSH (Table 2) was used in PCOS and non-PCOS patients, and that PCOS granulosa cells showed higher level of FSHR expression (Fig. 2), the downregulation of CYP19 in PCOS luteinized granulosa cells might be the result from factors other than FSH. Given the increased testosterone levels observed in PCOS follicles, we hypothesized that CYP19 expression in granulosa cells might be affected by testosterone. In order to test this, we isolated and cultured human luteinized granulosa cells from the non-PCOS patients. Since the testosterone level in small follicles of PCOS patients was determined to be around 15 ng/ml (Fig. 1C), we treated the cells with testosterone at a similar dose range. Testosterone (5, 15, and 60 ng/ml) was added to the culture medium to treat these cells. After 48 h, both mRNA (Fig. 4A) and protein (Fig. 4B) levels of aromatase in the cells were found to be significantly decreased by testosterone in a dose-dependent manner. Moreover, pretreatment of the cells with flutamide (2 μM, 24 h), an antagonist of AR, abolished the testosterone-induced reduction in CYP19 transcription (Fig. 4C), suggesting the role of AR in mediating the inhibitory effect of testosterone on CYP19. In addition, we also tested whether testosterone would influence the effect of FSH on aromatase in these cells. As shown in Fig. 4D, incubation with FSH (0.04–0.2 IU) for 48 h increased the mRNA level of CYP19 in these cells in a dose-dependent manner. Testosterone (15 ng/ml) was added in together with FSH (0.04 IU) to incubate these cells for 48 h. As shown in Fig. 4E, in the presence of testosterone (15 ng/ml), the FSH-induced CYP19 transcription was found to be slightly reduced (P=0.0528).
Effect of testosterone on the expression of aromatase in cultured non-PCOS human granulosa cells. (A and B) Real-time PCR of CYP19 (A) and western blotting for aromatase (B) in non-PCOS granulosa cells treated with testosterone (5, 15, and 60 ng/ml). Gapdh was used as a loading control for western blot. (C) Real-time PCR of CYP19 in non-PCOS granulosa cells treated with or without flutamide (2 μm, 24 h pretreatment) or testosterone (15 ng/ml, 48 h). (D) Cells were treated with FSH (0.04–0.2 IU) for 48 h before CYP19 mRNA level was measured by real-time PCR. (E) Cells were treated with FSH (0.04 IU) in the presence or absence of testosterone (15 ng/ml) for 48 h before the CYP19 mRNA level was measured. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. Data are means±s.e.m., n=7–9 (A), 4 (B), 9 (C), and 12 (D and E). *P<0.05 and **P<0.01, one-way ANOVA (A, B, C and D), t-test (E).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Effect of testosterone on the expression of aromatase in cultured non-PCOS human granulosa cells. (A and B) Real-time PCR of CYP19 (A) and western blotting for aromatase (B) in non-PCOS granulosa cells treated with testosterone (5, 15, and 60 ng/ml). Gapdh was used as a loading control for western blot. (C) Real-time PCR of CYP19 in non-PCOS granulosa cells treated with or without flutamide (2 μm, 24 h pretreatment) or testosterone (15 ng/ml, 48 h). (D) Cells were treated with FSH (0.04–0.2 IU) for 48 h before CYP19 mRNA level was measured by real-time PCR. (E) Cells were treated with FSH (0.04 IU) in the presence or absence of testosterone (15 ng/ml) for 48 h before the CYP19 mRNA level was measured. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. Data are means±s.e.m., n=7–9 (A), 4 (B), 9 (C), and 12 (D and E). *P<0.05 and **P<0.01, one-way ANOVA (A, B, C and D), t-test (E).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Effect of testosterone on the expression of aromatase in cultured non-PCOS human granulosa cells. (A and B) Real-time PCR of CYP19 (A) and western blotting for aromatase (B) in non-PCOS granulosa cells treated with testosterone (5, 15, and 60 ng/ml). Gapdh was used as a loading control for western blot. (C) Real-time PCR of CYP19 in non-PCOS granulosa cells treated with or without flutamide (2 μm, 24 h pretreatment) or testosterone (15 ng/ml, 48 h). (D) Cells were treated with FSH (0.04–0.2 IU) for 48 h before CYP19 mRNA level was measured by real-time PCR. (E) Cells were treated with FSH (0.04 IU) in the presence or absence of testosterone (15 ng/ml) for 48 h before the CYP19 mRNA level was measured. β-actin was used as internal control for relative mRNA level measurement by real-time PCR. Data are means±s.e.m., n=7–9 (A), 4 (B), 9 (C), and 12 (D and E). *P<0.05 and **P<0.01, one-way ANOVA (A, B, C and D), t-test (E).
Citation: REPRODUCTION 150, 4; 10.1530/REP-15-0044
Discussion
The present study compared patients under ART treatment with and without PCOS, and the results show that, despite the controlled LH and FSH levels in the circulation system during ART treatment, PCOS patients have significantly higher follicular testosterone and reduced expression of aromatase in luteinized granulosa cells. Treating luteinized granulosa cells isolated from non-PCOS women with a high level of testosterone similar to that observed in PCOS small follicles decreased the aromatase expression in luteinized granulosa cells in vitro. These results suggest for the first time that the down-regulation of aromatase in PCOS may be a result of the hyperandrogenic follicular environment.
Consistent with previous studies (Dickerson et al. 2010, de Resende et al. 2010, Nejad et al. 2011, Zhong et al. 2012), the present study has shown that PCOS women have similar fertilization rates, good embryo quality rates and clinical pregnancy rates in ART as compared to non-PCOS women (Table 4). A possible reason for this might be that the pituitary suppression treatment necessary for ART have overcome the high LH level in PCOS (Table 3). In addition, IVF and in vitro pre-implantation embryo culture for ART have released oocytes/embryos from pathological in vivo environment in PCOS, such as the high level testosterone. Given these reasons, ART is suggested to be a suitable treatment strategy for PCOS-associated infertility.
Although not all the samples from the present cohort contained a sufficient amount of materials for the analysis of follicular fluid, the available samples showed that hormonal levels in PCOS follicles remained remarkably different from that in non-PCOS even after pituitary suppression and exogenous FSH treatment in ART. For instance, progesterone is lower in PCOS follicles (Fig. 1E) suggesting impaired granulosa cells function, which may account for the lower oocyte obtaining rate in PCOS (Table 4). In addition, the FSH level in PCOS follicles was found to be lower than that in non-PCOS ones (Fig. 1B), even though the two groups had been treated with a similar total amount of exogenous FSH, which may suggest that not only the production of FSH but also other regulatory factors for FSH, such as its bio-degeneration, might have been altered resulting in a reduced FSH level in PCOS. Further investigation is required to clarify the mechanisms underlying FSH disruption in PCOS. Moreover, the testosterone level in PCOS follicles is high (Fig. 1C) even with a reduced low level of circulating LH (Table 3), which may suggest that theca cells, where most androgens are derived from, may have been intrinsically changed in PCOS. Alternatively, the processing of testosterone, such as its conversion to estrogens by granulosa cells, may have been inhibited in PCOS. This seems to be the likely case given the down-regulation of aromatase in PCOS granulosa cells as observed in the present study (Figs 2 and 3).
In a previous study, primary cultured luteinized granulosa cells from PCOS patients showed increased aromatase activity in vitro (Andreani et al. 1997). The present study using freshly isolated granulosa cells from luteinized follicles in PCOS, however, showed significantly reduced mRNA and protein levels of aromatase. Another previous study has also demonstrated that freshly isolated follicles from PCOS women have low mRNA expression of aromatase (Jakimiuk et al. 1998). Since freshly isolated cells/follicles may be more representative of in vivo, it is thus suggested that aromatase may be suppressed by in vivo environmental factor(s) in PCOS. For instance, it could be a result of the reduced follicular FSH in PCOS. However, FSHR was found up-regulated in PCOS luteinized granulosa cells (Fig. 2) which might compensate for the effect of low FSH in PCOS suggesting down-regulation of aromatase may be caused by factor(s) other than a low FSH level. The present results support that testosterone is a key factor responsible for down-regulation of aromatase in PCOS. First, at the averaged level in small follicles in PCOS patients, testosterone down-regulated both mRNA and protein levels of aromatase in cultured non-PCOS luteinized granulosa cells. Second, AR antagonist, flutamide, could abolish the inhibitory effect of testosterone. Moreover, in the presence of FSH, testosterone also showed some inhibitory effect on aromatase transcription in these cells in vitro (Fig. 4E). The present study is the first to show the effect of a high level of testosterone on the expression level of aromatase in human luteinized granulosa cells, which may provide explanation to the previous observation that the activity of aromatase in human granulosa cells was inhibited by androgens at high doses (Kirilovas et al. 2003). Since these PCOS patients had been controlled to have a reduced LH level during ART treatment, the follicular testosterone level in untreated PCOS patients may have gone higher than 15 ng/ml resulting in further inhibition of aromatase expression. The key role of aromatase is to convert androgens to estrogens. With lower aromatase expression, androgens would be accumulated. Therefore, the inhibition of aromatase by high testosterone as observed in the present study may also suggest a vicious cycle leading to hyperandrogenism, a key feature in PCOS (McCartney et al. 2002).
Of note, aromatase inhibitors have been proposed to treat PCOS patients, since they may stimulate pituitary secretion of FSH by reducing circulating estrogen levels (Polyzos et al. 2008, Misso et al. 2012, Pavone & Bulun 2013, Legro et al. 2014). However, clinical trials of aromatase inhibitors for PCOS yield controversial results regarding pregnancy rates (Badawy et al. 2009, Kar 2012, Roy et al. 2012). The present study has shown that aromatase expression in luteinized granulosa cells is down-regulated by high testosterone in PCOS patients. Whether further inhibition of residual aromatases in PCOS by these aromatases inhibitors would be effective as proposed requires further investigation.
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.
Funding
This work was supported in part by the National Major Basic Research Program of China (2012CB944903) and the Natural Science Foundation of China (no. 81471460).
References
Andreani CL, Diotallevi L, Lazzarin N, Pierro E, Giannini P, Lanzone A, Capitanio P & Mancuso S 1997 The cellular activity of different sized follicles in cycles treated with gonadotrophin-releasing hormone analogue. Human Reproduction 12 89–94. (doi:10.1093/humrep/12.1.89)
Badawy A, Shokeir T, Allam AF & Abdelhady H 2009 Pregnancy outcome after ovulation induction with aromatase inhibitors or clomiphene citrate in unexplained infertility. Acta Obstetricia et Gynecologica Scandinavica 88 187–191. (doi:10.1080/00016340802638199)
Barash A, Weissman A, Manor M, Milman D, Ben-Arie A & Shoham Z 1998 Prospective evaluation of endometrial thickness as a predictor of pituitary down-regulation after gonadotropin-releasing hormone analogue administration in an in vitro fertilization program. Fertility and Sterility 69 496–499. (doi:10.1016/S0015-0282(97)00542-6)
Blank SK, McCartney CR & Marshall JC 2006 The origins and sequelae of abnormal neuroendocrine function in polycystic ovary syndrome. Human Reproduction Update 12 351–361. (doi:10.1093/humupd/dml017)
Catteau-Jonard S, Jamin SP, Leclerc A, Gonzales J, Dewailly D & di Clemente N 2008 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 Metabolism 93 4456–4461. (doi:10.1210/jc.2008-1231)
Chen H, Guo JH, Lu YC, Ding GL, Yu MK, Tsang LL, Fok KL, Liu XM, Zhang XH & Chung YW et al. 2012 Impaired CFTR-dependent amplification of FSH-stimulated estrogen production in cystic fibrosis and PCOS. Journal of Clinical Endocrinology and Metabolism 97 923–932. (doi:10.1210/jc.2011-1363)
Coffler MS, Patel K, Dahan MH, Malcom PJ, Kawashima T, Deutsch R & Chang RJ 2003 Evidence for abnormal granulosa cell responsiveness to follicle-stimulating hormone in women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism 88 1742–1747. (doi:10.1210/jc.2002-021280)
Dickerson EH, Cho LW, Maguiness SD, Killick SL, Robinson J & Atkin SL 2010 Insulin resistance and free androgen index correlate with the outcome of controlled ovarian hyperstimulation in non-PCOS women undergoing IVF. Human Reproduction 25 504–509. (doi:10.1093/humrep/dep393)
Dumesic DA & Richards JS 2013 Ontogeny of the ovary in polycystic ovary syndrome. Fertility and Sterility 100 23–38. (doi:10.1016/j.fertnstert.2013.02.011)
Dumesic DA, Padmanabhan V & Abbott DH 2008 Polycystic ovary syndrome and oocyte developmental competence. Obstetrical & Gynecological Survey 63 39–48. (doi:10.1097/OGX.0b013e31815e85fc)
Erickson GF, Magoffin DA, Garzo VG, Cheung AP & Chang RJ 1992 Granulosa cells of polycystic ovaries: are they normal or abnormal? Human Reproduction 7 293–299.
Goodarzi MO, Dumesic DA, Chazenbalk G & Azziz R 2011 Polycystic ovary syndrome: etiology, pathogenesis and diagnosis. Nature Reviews. Endocrinology 7 219–231. (doi:10.1038/nrendo.2010.217)
Jakimiuk AJ, Weitsman SR, Brzechffa PR & Magoffin DA 1998 Aromatase mRNA expression in individual follicles from polycystic ovaries. Molecular Human Reproduction 4 1–8. (doi:10.1093/molehr/4.1.1)
Jayasena CN & Franks S 2014 The management of patients with polycystic ovary syndrome. Nature Reviews. Endocrinology 10 624–636. (doi:10.1038/nrendo.2014.102)
Kar S 2012 Clomiphene citrate or letrozole as first-line ovulation induction drug in infertile PCOS women: a prospective randomized trial. Journal of Human Reproductive Sciences 5 262–265. (doi:10.4103/0974-1208.106338)
Kirilovas D, Naessen T, Bergstrom M, Bonasera TA, Bergstrom-Pettermann E, Holte J, Carlstrom K, Simberg N & Langstrom B 2003 Effects of androgens on aromatase activity and 11C-vorozole binding in granulosa cells in vitro. Acta Obstetricia et Gynecologica Scandinavica 82 209–215. (doi:10.1034/j.1600-0412.2003.00144.x)
Legro RS, Brzyski RG, Diamond MP, Coutifaris C, Schlaff WD, Casson P, Christman GM, Huang H, Yan Q & Alvero R et al. 2014 Letrozole versus clomiphene for infertility in the polycystic ovary syndrome. New England Journal of Medicine 371 119–129. (doi:10.1056/NEJMoa1313517)
Mason HD, Willis DS, Beard RW, Winston RM, Margara R & Franks S 1994 Estradiol production by granulosa cells of normal and polycystic ovaries: relationship to menstrual cycle history and concentrations of gonadotropins and sex steroids in follicular fluid. Journal of Clinical Endocrinology and Metabolism 79 1355–1360. (doi:10.1210/jcem.79.5.7962330)
McCartney CR, Eagleson CA & Marshall JC 2002 Regulation of gonadotropin secretion: implications for polycystic ovary syndrome. Seminars in Reproductive Medicine 20 317–326. (doi:10.1055/s-2002-36706)
Misso ML, Wong JL, Teede HJ, Hart R, Rombauts L, Melder AM, Norman RJ & Costello MF 2012 Aromatase inhibitors for PCOS: a systematic review and meta-analysis. Human Reproduction Update 18 301–312. (doi:10.1093/humupd/dms003)
Nejad ES, Saedi T, Saedi S, Rashidi BH, Nekoo ZA & Jahangiri N 2011 Comparison of in vitro fertilisation success in patients with polycystic ovary syndrome and tubal factor. Gynecological Endocrinology 27 117–120. (doi:10.3109/09513590.2010.501872)
Ng EH, Chan CC, Yeung WS & Ho PC 2005 Comparison of ovarian stromal blood flow between fertile women with normal ovaries and infertile women with polycystic ovary syndrome. Human Reproduction 20 1881–1886. (doi:10.1093/humrep/deh853)
Pavone ME & Bulun SE 2013 Clinical review: The use of aromatase inhibitors for ovulation induction and superovulation. Journal of Clinical Endocrinology and Metabolism 98 1838–1844. (doi:10.1210/jc.2013-1328)
Payne AH & Hales DB 2004 Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocrine Reviews 25 947–970. (doi:10.1210/er.2003-0030)
Polyzos NP, Tsappi M, Mauri D, Atay V, Cortinovis I & Casazza G 2008 Aromatase inhibitors for infertility in polycystic ovary syndrome. The beginning or the end of a new era? Fertility and Sterility 89 278–280. (doi:10.1016/j.fertnstert.2007.10.016)
de Resende LO, dos Reis RM, Ferriani RA, Vireque AA, Santana LF, de Sa Rosa e Silva AC & Martins Wde P 2010 Concentration of steroid hormones in the follicular fluid of mature and immature ovarian follicles of patients with polycystic ovary syndrome submitted to in vitro fertilization. Revista Brasileira de Ginecologia e Obstetrícia 32 447–453. (doi:10.1590/S0100-72032010000900006)
Richards JS, Jahnsen T, Hedin L, Lifka J, Ratoosh S, Durica JM & Goldring NB 1987 Ovarian follicular development: from physiology to molecular biology. Recent Progress in Hormone Research 43 231–276. (doi:10.1016/B978-0-12-571143-2.50012-5)
Roy KK, Baruah J, Singla S, Sharma JB, Singh N, Jain SK & Goyal M 2012 A prospective randomized trial comparing the efficacy of letrozole and clomiphene citrate in induction of ovulation in polycystic ovarian syndrome. Journal of Human Reproductive Sciences 5 20–25. (doi:10.4103/0974-1208.97789)
Stocco C 2008 Aromatase expression in the ovary: hormonal and molecular regulation. Steroids 73 473–487. (doi:10.1016/j.steroids.2008.01.017)
Zhong YP, Ying Y, Wu HT, Zhou CQ, Xu YW, Wang Q, Li J, Shen XT & Li J 2012 Comparison of endocrine profile and in vitro fertilization outcome in patients with PCOS, ovulatory PCO, or normal ovaries. International Journal of Endocrinology 2012 492803. (doi:10.1155/2012/492803)
(F Yang and Y-C Ruan contributed equally to this work)
