If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The human androgen receptor (AR) gene contains a highly polymorphic CAG repeat sequence within exon 1. In-vitro studies have shown a relationship between CAG repeats in the AR gene and its transactivation potential. This variation in length may play a role in anovulatory infertility. The objective of this study was to investigate whether CAG polymorphism of the AR gene has a predictive value for ovarian reserve, response and cycle outcome in an egg donor programme. CAG length of the AR gene was determined in 147 oocyte donors. All donors underwent ovarian stimulation with a gonadotrophin-releasing hormone antagonist protocol (n = 355). No differences were reported in days of stimulation, gonadotrophin doses, and number of oocytes retrieved. Clinical outcomes were not affected by the CAG repeat length of the AR gene; the primary end-point, antral follicle count, was significantly affected (P < 0.05). In conclusion, in a population of fertile egg donors AR gene CAG polymorphism does not affect ovarian response to gonadotrophins. Antral follicle count was associated with the CAG polymorphism genotype. This suggests that genetic factors may increase susceptibility to poor ovarian reserve, and that AR gene genotype could play a role in the natural ovarian ageing process.
Recent experiments in androgen receptor (AR) knockout mice have shown the importance of androgens. Female knockout-AR mice, although fertile, display a diminished ovarian reserve (
Ovarian stimulation is a largely used strategy in assisted reproductive techniques. Ovarian response, however, varies widely in women undergoing ovarian stimulation (
). Recently, the Bologna criteria standardized the definition of poor ovarian response in a simple and reproducible manner. Poor ovarian response to ovarian stimulation usually indicates a reduction in follicular response, resulting in a reduced number of retrieved oocytes. It has been recognized that, in order to define poor response in IVF, at least two of the following three features must be present: advanced maternal age or any other risk factor for poor ovarian response; a previous poor ovarian response; and an abnormal ovarian reserve test (
ESHRE working group on Poor Ovarian Response Definition. ESHRE consensus on the definition of ‘poor response’ to ovarian stimulation for in vitro fertilization: the Bologna criteria.
About 9–24% of women undergoing IVF respond more poorly than expected to the ovarian stimulation protocol prescribed in accordance with their clinical characteristics (
). Experience in primates has shown an increase in the number of growing follicles after the stimulation of the AR with the administration of testosterone or 5-dihidrotestosterone (
). In recent years, some strategies have been developed to increase androgen concentrations in the ovarian milieu in order to increase ovarian response (
The biological effect of androgens is mediated through the activation of the AR and, therefore, their action does not depend solely on serum androgen levels. The human AR is encoded by a single copy gene on the X-chromosome at Xq12 (
Immunohistochemical distribution of progesterone, androgen and oestrogen receptors in the human ovary during the menstrual cycle: relationship to expression of steroidogenic enzymes.
Immunohistochemical localization of androgen receptor in the human ovary throughout the menstrual cycle in relation to oestrogen and progesterone receptor expression.
). Androgen activated AR exerts its biological effects by stimulating target genes via a sequence of processes, including ligand binding, homodimerization, nuclear translocation, DNA binding and a complex formation with co-regulators and general transcription factors (
The human androgen receptor gene contains a highly polymorphic CAG repeat sequence within exon 1. In-vitro studies showed a relationship between the number of CAG repeats in the AR and its transactivation potential (
). Recently, variations of the AR-CAG tract, while still within the normal polymorphic range (11–38), have been linked to an increase in the severity of different diseases associated with low or high androgenic activity (
Although within the normal polymorphic range, some studies have shown that relatively long tracts are associated with an increased risk of male infertility caused by impaired sperm production (
). Data on the functional importance of the CAG repeat in women is sparse, but conditions associated with androgen insensitivity, including hirsutism, androgenic alopecia and breast cancer have been studied (
). Because AR-CAG length can be correlated to androgenicity and androgen-regulated diseases, it is possible that variations in this length may play a role in the cause of anovulatory infertility (
To our knowledge nothing is known about the potential hazardous effect of CAG repeats in AR gene and ovarian stimulation. To investigate the possibility of a correlation between CAG polymorphism in the AR gene and ovarian stimulation, we evaluated ovarian stimulation in a non-confounding model in women participating in an egg donation programme. Egg donors are young and fertile women with normal ovulation; therefore, there is minimal variability in oocyte and embryo quality. Moreover, data have been published showing a relationship between CAG repeats and ovarian reserve in infertile patients, as cited above, but no results have been published that include fertile patients. The aim of this study was to investigate whether CAG polymorphism in the AR gene had a predictive value for ovarian response to stimulation and ovarian reserve in oocyte donors as a model of normoovulatory women.
Materials and methods
Study population
Egg donation is the best model to evaluate the determinants of implantation for several reasons. First, oocyte and embryo quality vary minimally, as donors are young women with normal ovulation. Second, the preparation of the endometrium is similar, as all recipients receive the same hormone replacement protocol.
The selection and recruitment of donors was carried out in our clinic following strict quality criteria, including an extensive chromosomal and genetic evaluation. All donors studied were Mediterranean and met the legal requirements in Spain (Spanish Law 14/2006). All women included in the Instituto Bernabeu's egg donation programme have to pass a psychological evaluation. They were aged between 18 and 33 years, healthy, with no family history of hereditary diseases and with an antral follicle count (AFC) more than 8. Preferably, they had previous proven fertility (previous ongoing pregnancies). The donors underwent a complete gynaecological examination, karyotyping, and screening for infectious diseases, such as human immunodeficiency virus, hepatitis B and C, gonoccocia, and syphilis. In addition to the legal requirements, genetic screening was carried out for cystic fibrosis, fragile X syndrome and alpha and beta thalasaemia. Furthermore, guidelines from both the American Society for Reproductive Medicine and the European Society of Human Reproduction and Embryology for oocyte donors were followed (
In this study, the results of CAG polymorphism in the AR gene from 147 oocyte donors were included. These donors performed 355 COH cycles, and the results from stimulation and cycle outcome were included in the present research. The average number of COH cycles per donor is 2.6 ± 2.1.
All the participants included in the study gave their informed consent to collect peripheral blood samples suitable for molecular analysis. This study involved only retrospective analysis of anonymous medical records, and was approved by the Instituto Bernabeu Institutional Review Board (reference number 08/2012; 8 Febraury 2012).
Genotyping
DNA was isolated from peripheral blood lymphocytes according to manufacturer instructions (Wizard® Genomic DNA Purification Kit, Promega, USA) and stored at 4°C. CAG repeat of exon 1 in the AR gene was amplified from the genomic DNA using the TaKaRa LA Taq kit (Takara Bio Inc, Shiga, Japan) and primers flanking the CAG repeat region (
). The forward primer was 5′-CAGAATCTGTTCCAGAGCGTGC-3′ and labeled at 5′ with the dye 6FAM and the reserve primer was 5′-AAGGTTGCTGTTCCTCATCCAG-3′. Polymerase chain reaction was carried out under the following conditions: initial denaturation 95°C for 5 min, followed by five cycles of denaturation at 95°C, anneling at 55°C and polymerization at 72°C for 45 s with a final extension at 72°C for 5 min. For genotyping 1.0 µl of polymerase chain reaction product was mixed with 0.3 µl of LIZ 500 and 9.7 µl of Hi-Di formamide and analysed on ABI 310 DNA analyzer (Applied Biosystems, Madrid, Spain) using Gene Mapper software (Applied Biosystems) to ascertain the size of AR alleles and the number of repeats. The number of CAG repeats was calculated in relation to a series of standards obtained by direct sequencing.
Ovarian stimulation and oocyte retrieval
In accordance with Spanish Fertility Act requirements, all patients underwent an ovarian stimulation protocol with tailored doses of urinary FSH (Fostipur; Angelini, Barcelona. Spain). Gonadotrophin stimulation started from day 2 of the menstrual cycle, with doses varying between 150 and 300 IU/day depending on the age of the donor, body mass index and antral follicle count. The gonadotrophin-releasing hormone antagonist, cetrorelix 0.25 mg/day (Cetrotide; Merck-Serono, Geneva) was introduced according to a multiple-dose, flexible protocol. In all cases, triggering was exclusively performed with 0.4 mg of subcutaneous triptorelin (Decapeptyl; Ipsen Pharma, Paris). Ovarian response was monitored by transvaginal ultrasound and plasma oestradiol concentrations. Oocytes were aspirated 36 h after analogue administration by transvaginal, ultrasound-guided needle aspiration under sedation. Sperm and oocyte preparation, fertilization, embryo culture and transfer were carried out according to IVF laboratory guidelines.
Recipient protocol
A total of 355 recipient women underwent a standard protocol as previously reported (
). The number of previous IVF cycles with donor eggs per recipient was on average 1.5 ± 0.8. Patients with ovarian activity received either birth control pills (Yasmin; Bayer Hispania, Spain) or analogue depot (Decapeptyl depot 3.75; Ipsen Pharma, Paris) in the luteal phase of the previous cycle. Menopausal patients were treated with a sequential regimen of oestrogen and progesterone the month before the transfer. Oral oestradiol valerate (Progynova, Schering) or oestradiol patches releasing 50 µg daily (Dermestril 50; Rottapharm-Madaus) were used in increasing doses for the endometrial preparation. From day 12 of treatment, patients received up to 6 mg oestradiol valerate per day or three patches every other day and the duration of the treatment varied in accordance with the availability of a phenotypically matched donor, ranging from 14 to 24 days. After 13 days of oestradiol valerate administration, endometrial thickness and pattern were tested. If a trilaminar pattern was observed in a ≥7 mm endometrium, the aforementioned dose of oestradiol therapy was continued at least until the pregnancy test was carried out 2 weeks later. If the endometrium was not seen to be sufficiently developed, doses of oestradiol valerate were increased to 8 mg/day or four patches. From the day of oocyte retrieval until the pregnancy test, 600 mg of micronized progesterone (Utrogestan; Seid, Paris) were administered vaginally daily.
Statistical analysis
Values are presented as mean ± SD, and medians and range for continuous data and percentages for categorical variables. Data were analysed with Statistical Package for the Social Sciences (SPSS) software (version 20.0, SPSS, IBM Corp, Armonk, NY, USA). For CAG repeat length, we considered the biallelic mean, as in previous reports (
). The primary end-point was AFC and the secondary end-points were gonadotrophin consumption, stimulation length and total number of oocytes retrieved in donors. Linear regression was applied for AFC adjusting for age, previous fertility and smoking status as possible confounding factors, as these have been reported to affect ovarian reserve. Linear regression was applied to evaluate donor ovarian stimulation parameters adjusting for age, AFC, previous fertility and smoking status as possible confounding factors, as these have been reported to affect ovarian response. Clinical pregnancy was diagnosed as a pregnancy with ultrasound visualization of a gestational sac with fetal heartbeat at 8 weeks gestation. The implantation rate was defined as the number of gestational sacs observed divided by the number of embryos transferred. Spontaneous abortion was defined as the spontaneous loss of a clinical pregnancy before 20 weeks gestation. Logistic regression was carried out to evaluate the association between number of CAG repeats and categorical variables (positive beta-HCG, clinical pregnancy, implantation and spontaneous abortion rates). P < 0.05 was considered significant.
Results
Clinical characteristics
The clinical characteristics of the egg donors are detailed in Table 1. Donor age varied significantly (P = 0.002) between CAG repeat length of the AR gene, suggesting that the population was not homogenously distributed. Age was therefore used as a confounding variable in later statistical analyses. No significant differences were observed in body mass index, proven fertility or the number of smokers. The mean antral follicle count was 15 ± 5 (range 8–26) (Table 2). Linear regression analysis indicated that AFC was significantly associated with CAG repeat length (uncorrected P = 0.013; age-adjusted P < 0.001).
For this study, we included the results of the AR-CAG repeat counts obtained from 147 oocyte donors. The AR-CAG repeat numbers were in the normal polymorphic range in all women. The frequency of the CAG repeat length in the two alleles and the biallelic mean is represented in Figure 1. In the 147 oocyte donors analysed, the biallelic mean of CAG repeats ranged from 17 to 29. The shortest repeat lengths ranged from 11 to 27. In the longer of the two alleles, repeat lengths ranged from 20 to 34. The 21 and 25 CAG repeat alleles were found to be at the highest frequency for allele 1 and 2, respectively. The median CAG repeat in exon 1 of AR gene was found to be 23.
Figure 1Distribution of CAG repeats numbers in exon 1 of androgen receptor gene in egg donor. (A) Allele 1 is the shortest CAG repeat of the allele pair; (B) allele 2 is the longest CAG repeat of the allele pair; (C) Biallelic mean.
The 147 patients included in this study underwent 355 ovarian stimulation cycles. The ovarian stimulation parameters in the 355 ovarian stimulation cycles are presented in Table 3. No significant differences were reported in the number of oocytes retrieved, the gonadotropin doses or the days of stimulation.
Table 3Egg donation cycle outcomes.
Total (n = 355)
Average ± SD (range)
Correlation coefficient Pearson (r)
Stimulation length (days)
11.6 ± 1.56 (7–18)
0.161
Gonadotropin used (IU)
2090.2 ± 546.51 (600–3900)
0.371
Number of retrieved oocytes
19.6 ± 8.74 (6–29)
0.166
Recipient age (years)
40.8 ± 4.39 (24–51)
–0.039
Follicular phase length (days)
18.8 ± 4.07 (9–32)
0.001
Recipient endometrial thickness (mm)
8.7 ± 1.63 (9–13)
–0.140
Number of oocytes received
12.9 ± 3.50 (9–17)
–0.008
Two pronuclei conventional IVF
7.4 ± 3.83 (1–16)
–0.052
Two pronuclei ICSI
8.3 ± 3.12 (1–17)
0.044
Transferred embryos
1.8 ± 0.47 (1–3)
–0.018
Positive beta-HCG (%)
245 (69.0)
Clinical pregnancy rate (%)
192 (54.0)
Implantation rate (%)
252/639 (39.4)
Spontaneous abortion rate (%)
22 (11.5)
Tests performed for statistical analysis were logistic regression for categorical variables and linear regression for continuous variables. No statistically significant differences were found in the above parameters in relation to androgen receptor repeat numbers as indicated by linear regression analysis adjusting for age, antral follicle count, previous fertility and smoking status as possible confounding factors.
Egg donation treatment outcomes are presented in Table 3. No significant differences in cycle outcome were observed in relation to CAG repeat length of the androgen receptor. No differences were observed in recipient age, endometrial thickness, days of hormone replacement therapy, number of oocyte received, fertilization rates for conventional IVF and ICSI, fertilization technique or the day of embryo transfer (data not shown). Also, no significant associations were found in biochemical pregnancy, clinical pregnancy rate, spontaneous abortion rate and implantation rate.
Discussion
Our data suggest that AFC is associated with CAG polymorphism of the AR gene but ovarian stimulation is not. Moreover, the clinical outcome is not associated with this polymorphism. To the best of our knowledge, these data show for the first time the relationship between, CAG repeat length in the AR gene and ovarian reserve and stimulation using a non-confounding model of egg donation.
Oocyte donation is the best model to evaluate the determinants of stimulation and embryo implantation potential. Donors are young women of similar age with normal ovarian function and, in our egg donation programme, with up to 90% of previous proven fertility (Table 1). Overall, no differences were reported in length of CAG repeats in relation to ovarian response and cycle outcome in our population. This is not surprising, because androgen-mediated actions occur in the early stages of follicular development. Recent approaches, however, have been developed with the ultimate aim of increasing serum androgens to deliver high androgen concentrations to growing follicles (
) and increase ovarian response to ovarian stimulation in patients with poor ovarian response. Two studies whose common goal was to increase androgen levels, one using testosterone (
), had different outcomes according to embryo quality and pregnancy rates. The data in the current study indicate that these strategies could improve the ovarian response by increasing the number of small antral follicles. This increase would be mediated by androgens instead of via a direct action on ovarian stimulation, because the current study found an association between CAG repeat number in the androgen receptor gene and ovarian reserve but not with ovarian response.
Success of assisted reproductive technology depends on the selection of a correct protocol for ovarian stimulation. This can be decided mainly by a proper assessment of ovarian reserve before the ovarian stimulation. Ovarian reserve is a major factor in determining success rates after IVF. Ovarian reserve relates to secondary, pre-antral and antral ovarian follicle pool. Various methods have been proposed to assess ovarian reserve. The most common are serum FSH, anti-Müllerian hormone (AMH) and AFC, the latter assessed by ultrasound (
). AFC provides a direct measure of ovarian reserve, whereas AMH, inhibin B and oestradiol are released from growing follicles and so their levels reflect the size of the developing follicle cohort (
A prospective, comparative analysis of anti-mullerian hormone, inhibin-B, and three dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation.
). AFC alone may be sufficient for the estimation of ovarian reserve. Gonadotrophin doses for ovarian stimulation usually increase with decreasing ovarian reserve. A recent meta-analysis, however, has shown that markers of ovarian reserve have only modest value in predicting the response to gonadotrophins (
Effect of follicle-stimulating hormone receptor N680S polymorphism on the efficacy of follicle-stimulating hormone stimulation on donor ovarian response.
The present study has shown a positive direct association between CAG repeat length in the AR gene and AFC. The average length of the CAG repeat in the normal fertile population is 22.1 ± 2 (range 14–31) (
). Similar results (23 ± 1, range, 17–29) in our population. Previous studies have indicated a relationship between the CAG repeat number on AR gene and AR function (
), because these repeats are located in the N-terminal region, altering the interaction with co-regulator proteins. Alteration of the transactivation function caused by loss of a co-regulator protein has been associated with androgen insensitivity (
). The explanation of this apparent disagreement could be found in the different population characteristics in different studies and in the effect of CAG repeats on the AR function. The present work seems to be the first to study a fertile population of egg donors with a range of shorter repeat lengths than patients with premature ovarian failure who carry longer CAG repeats; therefore, the results compare populations with a different range of repeats. On the other hand, the AR action depends on the CAG repeat length. Recent work supports previous evidence published by
). The highest AR activity was confined to that with 22 CAG repeats, the most frequent in normal population. This may at least be able to explain the discrepancy in data aiming to link CAG repeat length and physiological conditions.
also showed that the method used for the measurement of transactivation efficiency may highly influence the final conclusion. Enzyme-linked immunosorbent assay experiments might be a more efficient way to obtain these data (
). A poliglutamine tract of about 22 glutamine residues would represent the baseline activation status of the AR. A higher or lower number of glutamine residues would increase the repression of the receptor, leading to a reduction in its transactivation function and, consequently to a lower activation of the androgren-regulated genes (
The AR gene is X-linked and is known to undergo X-inactivation. This process is essentially random in normal women. As great variations in the degree of X-inactivation between tissues have been identified (
), future epigenetic studies might provide more unequivocal answers regarding the relative expression of CAG alleles. It is important to investigate tissue specific X chromosome inactivation patterns in future studies. Although in this study we have not investigated the pattern of X-inactivation, it is plausible that effects of repeat length at the functional level may be further amplified in the event of the shorter allele being preferentially inactivated. In the present study, however, and in a previous study (
), the association between CAG repeats lengths and ovarian reserve is only reported when the results are analysed from the biallelic mean. This suggests a random AR gene inactivation. These data agree with previous work reporting random inactivation (
Finally, advance identification of patients who will be affected by premature ovarian failure would be of great clinical advantage for such patients. Various predictive markers have been proposed such as AMH and AFC, but these are age related and depend on when the tests are performed (
A prospective, comparative analysis of anti-mullerian hormone, inhibin-B, and three dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation.
). Besides these parameters, genetic variability also seems to be an important factor. In conclusion, this research reveals that, in a population of fertile egg donors, CAG repeat length on AR gene is associated with different ovarian reserve but not with different ovarian response to ovarian stimulation. AR gene genotype could be an important factor for determining the natural ovarian ageing process, because the genotype does not change throughout life and so is independent of the time at which it is determined. AR gene genotype may help to predict ovarian behaviour with age. Egg cryopreservation could be an important option to preserve the fertility in at risk patients. Genotyping CAG in AR gene together with some additional markers may therefore provide a tool for identifying patients who will suffer lower ovarian reserve in the near future.
Acknowledgements
B. Lledó, J. A. Ortiz and J. Llácer were responsible for study design; J. Guerrero was responsible for sample collection and egg donor recruitment. B. Lledó, A. Turienzo J. A. Ortiz and R. Morales were responsible for laboratory experiments. B. Lledó, A. Turienzo and J. A. Ortiz were responsible for analysis; B. Lledó., J. Guerrero and J. Ten were responsible for embryology laboratory data collection. B. Lledó drafted the manuscript, and B. Lledó, J. A. Ortiz, R. Morales, J. Llácer, J. Guerrero, J. Ten and R. Bernabeu were involved in critical discussion.
References
Adachi M.
Takayanagi R.
Tomura A.
Imasaki K.
Kato S.
Goto K.
Yanase T.
Ikuyama S.
Nawata H.
Androgen-insensitivity syndrome as a possible coactivator disease.
ESHRE working group on Poor Ovarian Response Definition. ESHRE consensus on the definition of ‘poor response’ to ovarian stimulation for in vitro fertilization: the Bologna criteria.
Immunohistochemical localization of androgen receptor in the human ovary throughout the menstrual cycle in relation to oestrogen and progesterone receptor expression.
A prospective, comparative analysis of anti-mullerian hormone, inhibin-B, and three dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation.
Effect of follicle-stimulating hormone receptor N680S polymorphism on the efficacy of follicle-stimulating hormone stimulation on donor ovarian response.
Immunohistochemical distribution of progesterone, androgen and oestrogen receptors in the human ovary during the menstrual cycle: relationship to expression of steroidogenic enzymes.
Belén Lledó studied Biology at the University of Alicante, Spain, between 1996 and 2000. She received her PhD degree in Molecular Biology from the University of Alicante. In 2004, she moved to the Instituto Bernabeu in Alicante, Spain, where she first worked in the IVF and the Molecular Biology laboratories. She is currently a director of a genetic company, which forms part of the Instituto Bernabeu Group. Her research focus is on the genetic diagnosis of infertility patients.
Article info
Publication history
Published online: July 09, 2014
Accepted:
June 23,
2014
Received in revised form:
June 19,
2014
Received:
November 8,
2013
Declaration: The authors report no financial or commercial conflicts of interest.
00361-7&id=rbmo1182-fig-0001.jpg){kind=link}