Background
Fecundity is the wonderful biological ability to produce abundant healthy offspring and is affected by genetic and environmental factors [
1]. If pregnancy is planned, fertility may be expressed as time-to-pregnancy (TTP) [
2]. TTP is defined as the number of contraceptive-free cycles needed to conceive [
3]. A TTP greater than 12 months allocates the infertility status [
4,
5]. Usually trying to conceive in the first year succeeds for ~ 85% of cases [
6].
Impaired fertility is inherited and may be due to inactivating mutations in the gonadotropin and gonadotropin receptor genes [
7,
8]. Recent genetic studies have revealed that the pathogenesis of subfertility or infertility can be due to mutations in the follicle-stimulating hormone receptor (FSHR) gene [
9]. While mutations affecting FSHR are sporadic; polymorphism of the FSHR gene seems to be a common phenomenon [
9]. FSHR inactivating mutations may cause primary or secondary amenorrhea, infertility, and premature ovarian failure [
10]; whereas activating mutations can predispose to ovarian hyperstimulation syndrome, as a consequence of exogenous FSH administration, or to a spontaneous onset [
10‐
12].
In-vitro studies have shown that the A allele at the 29th position in the 5′ untranslated region of the FSHR gene is associated with impaired transcriptional activity [
13]. The polymorphism at position 29 in the promoter of the FSHR gene may contribute to the reduced receptor expression [
14]. The FSHR shows nucleotide polymorphisms in the promoter and in exon 10 [
15]. The single nucleotide polymorphisms in exon 10 results in four discrete allelic variants characterized by the amino acid combinations:
threonine (Thr)307-
asparagine (Asn)680, alanine (Ala)307-
Serine (Ser)680,
Ala307-Asn680 and
Thr307-Ser680 [
15]. The first two allelic variants are very frequent in the Caucasian population [
15]. At position 680, three FSH receptor variants are possible:
Asn/Asn, Asn/Ser, and
Ser/Ser; however,
Ser/Ser-680 predominates in the studied infertile population [
16].
The studies on FSHR polymorphism, performed on patients undergoing in-vitro fertilisation procedures show that women homozygous for the
Ser680 variant have higher follicular FSH levels and longer follicular phase length, which suggest a lower sensitivity to FSH. Thus the homozygous
Ala307-Ser680 variant is associated with a higher amount of FSH required for ovarian stimulation in women undergoing assisted reproduction [
15]. This suggests that the FSHR genotype can influence the ovarian response to FSH stimulation [
17,
18]. However, there are studies where this association was not confirmed [
19,
20].
FSH is responsible for follicular maturation and for the length and stability of the menstrual cycle [
21]. A longer cycle may be associated with more difficulties in conception; women who have the FSHR gene
Asn (Asparagine) exchanged for
Ser (Serine) at codon 680 have statistically proven longer menstrual cycles [
22].
Despite the numerous publications on the FSHR polymorphism impact on women’s reproductive function; an FSHR polymorphic relationship to TTP has not yet been studied. Here we aimed to assess the relationship of the FSHR polymorphism Serine/Serine, Asparagine/Asparagine and Asparagine/Serine variants on TTP in a sample of Lithuanian women.
Methods
Subjects
Klaipeda is the third largest city in Lithuania and has one obstetric department that performs approximately 3500 deliveries per year from all parts of the west region of Lithuania. Between March 2008 and May 2008, 291 consecutive selected 21–34-year-old (mediana [25–75%] – 27.0 [24.0–31.0] years) women who conceived naturally and delivered babies at the Klaipeda University Hospital were invited to participate in the study 0–3 days after delivery. This time interval was chosen because it allowed for the accurate recall of the time period preceding conception, and it was a time point at which the delivery outcome was already known. All selected women had planned pregnancies that were achieved without using assisted reproductive technologies. In 49.17% of cases women were nulliparous, the rest of them - multiparous. Prior to pregnancy planning, 67 (22.26%) women used hormonal contraception. Women were asked to complete a standardised questionnaire. It included questions concerning age, height, weight before pregnancy, the menstrual cycle, socioeconomic factors, lifestyle, sexual behaviour, and some other factors. The regular menstrual cycle was defined as 28 ± 7 days and this definition was explained to study participants. Women with irregular menstrual cycles were asked to report if the majority of their cycles are < 21 days or > 35 days. Only one participant reported irregular cycles < 21 days, so, she was excluded from the final analysis. The questions regarding gynaecological diseases were asked separately for non-infectious and infectious diseases. Women with unplanned pregnancies and those who received assisted reproductive technologies were not included in the study. However, women who get another infertility treatment, e.g. ovulation induction, treatment of infections - were enrolled into the study.
DNA sampling
A venous blood sample was drawn for DNA extraction from all 291participants.
DNA extraction was performed in a certificated “SORPO” laboratory of Thermo Fisher Scientific Inc. in Vilnius, Lithuania. DNA samples froze at-20 °C; were sent to the University of Munster in Germany. There are two known polymorphisms of clinical relevance in the hormone (FSH) receptor exon 10: Ala or Thr at position 307 (dbSNP numbers 6165), and Asn or Ser at position 680 (dbSNP numbers 6166). These give rise to two discrete allelic variants: Thr307/Asn680 and Ala307/Ser680. The allelic variants at codon 307 and 680 are almost invariably associated, therefore codon 680 was assessed, and all women were classified as homozygous (Ser/Ser or Asn/Asn) or heterozygous (Asn/Ser).
Genomic DNA was extracted from peripheral blood using a FlexiGene DNA extraction kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instruction. All samples were screened for the single nucleotide polymorphism (SNP) at position 2039 (codon 680) of exon 10 by the TaqMan allelic discrimination assay while using the ABI Prism 7000 sequence detection system (Applied Biosystems, Darmstadt, Germany). The probes (SNP indicated in bold lower case letters) were 5′-AGAGTCACCAgTGGTT-3′ (6-carboxyfluorescein fluorescence) and 5′-AGTCACCAaTGGTTC-3′ (VIC fluorescence). The primers were 5′-AAGGAATGGCCACTGCTCTTC-3′ (forward) and 5′-GGGCTAAATGACTTAGAGGGACAA-3′ (reverse). Each polymerase chain reaction (PCR) (25 μl) contained: 2 μl DEPC-treated water, 12.5 μl Universal master mix, 0.25 μl of each probe, and 4.5 μl of each primer (5 pmol). Using the TaqManassay, PCR was performed in two steps: absolute quantification and allelic discrimination. For absolute quantification, the cycles were as follows: stage 1: Probe binding at 50 °C for 2 mins (1 cycle); stage 2: denaturation at 95 °C for 10 mins (1 cycle), followed by 35 cycles at 95 °C for 15 s; stage 3:60 °C for 1 min. Whereas the allelic discrimination assay took 1 min at 60 °C.
Statistical analysis
Analyses were performed using SPSS 17.0 software. Women who conceived after ≥12 months of trying were classified as the risk group. The normality of distribution was tested using the Kolmogorov-Smirnov test. Student’s (t) criterion was used for comparison of means for normal distributions, and the Mann-Whitney (U) test was used in skewed distributions. In order to determine the difference between more than two groups; parametric and nonparametric dispersive analysis with ANOVA and Kruskal-Wallis test was performed. Bonferroni test was performed by comparing multiple pairs. For evaluating dependence between qualitative features χ2 criterion was used. Results were presented as mean (M) ± standard deviation (SD) or n (%). Biological, social, demographic, economic, sexual behaviour, genetic, living, working and environmental confounders variables were retaining in models. Univariate analysis of the OR for each variable was taken initially. Multivariate logistic regression step-wise enter method model was used to estimate the most important relationship factors. Odds ratios (OR) and 95% confidence intervals (CI) for time-to-pregnancy was calculated. The limit of significance was defined as a two-sided p-value of< 0.05.
The study was approved by the Lithuanian Bioethics Committee (21/12/2006 No. 59/2). The aim of the survey protocol was carefully explained to each subject of study entry, and a written informed consent was obtained.
Discussion
The FSHR polymorphism’s impact on women’s reproductive function has been demonstrated in several studies [
9,
11,
16,
17,
23]; particularly in some diseases, such as the polycystic ovary syndrome and amenorrhea [
24‐
26]. Some investigations provide contradictory data on the relationship between single nucleotide polymorphisms, and their link to polycystic ovary syndrome and amenorrhea [
27‐
29]. The main reported findings on changes of hormonal dynamics in women with homozygote mutated
Ser680 throughout the menstrual cycle were with lower serum levels of estradiol, progesterone and inhibin A [
22]. However, these women had significantly higher FSH levels, and longer menstrual cycles [
12,
14,
18]. Patients with the
Ser680/Ser680 genotype are more resistant to FSH action and thus require a stronger stimulus for the same biological response [
22]. This finding is important in infertility treatment; patients with the homozygous FSHR
Ser680/Ser680 polymorphism have double the chance of having a resistance to clomiphene citrate [
30]. They require higher FSH dosages in order to show the same estradiol response during controlled ovarian stimulation [
24]. Furthermore, it was also demonstrated that the frequency of
Ser680/Ser680 variation in the control population is lower than if compared to the infertile women’s group [
23]. Thus it may be hypothesised that the
Ser680/Ser680 genotype could be directly related to a women’s fertility. To our knowledge, the FSHR polymorphism was never investigated in direct relation to TTP in a fertile population. In this study, women were considered to be fertile if they achieved pregnancy without using assisted reproductive technology methods. A large number of factors possibly affecting TTP were investigated along with the FSHR polymorphism in order to detect independent factors predicting longer TTP and to establish the role of FSHR polymorphism between other determinants of women’s fertility.
Our data confirmed that higher age, previous gynaecological diseases, and/or fertility problems pose as risk factors for longer TTP. An association for the risk of longer TTP due to hormonal contraception use prior to conception is more questionable. However, an older age of women using contraception could be the reason for this finding.
The relationship of the FSHR polymorphism to the length of the menstrual cycle was demonstrated in our study, as well as in previous publications [
22]. Differences in menstrual cycle length between the
Ser680/Ser680 and the
Asn680/Asn680 groups result in 12.5 vs. 13.5 menstrual cycles per year, respectively [
22]. Assuming no difference in age at the time of menopause; women with the
Ser680/Ser680 genotype would experience 30–40 cycles fewer, than women with an
Asn680/Asn680 genotype during their reproductive life [
22]. Some authors conclude that women with the
Ser680/Ser680 genotype have a lower chance to achieve pregnancy during the same time period if compared to the other variants [
23,
30]. Therefore, menstrual periods are stressful events that have certain disadvantages; such as blood loss, menstrual discomfort, and the effects of hormone fluctuations on mood, breast and other oestrogen-dependent organs. This gives rise to some speculation that fewer menstrual cycles during the reproductive lifespan might represent an evolutionary advantage and might influence fertility positively [
22]. Our data provide direct evidence that women with the
Ser680/Ser680 genetic variant had a lower chance of conception than females with
Asn680/Asn680 and
Asn680/Ser680 genetic variants.
We have demonstrated that the FSH receptor gene Serine/Serine variant polymorphism is associated with a fivefold lower likelihood to become pregnant during the first 12 months of attempts to conceive.
The other independent factors predicting a TTP of 12 or more months in the study group were older age, gynaecological diseases, fertility problems in the past, and the use of contraception prior to conception.
Some limitations of our study, especially related to the retrospective design, should be discussed. A retrospective design of the study was used in order to achieve a higher participation rate. Only one polymorphism in this region was evaluated, and furthermore, no replication in an independent cohort was attempted. However, it was previously demonstrated that immediately after delivery women can recall the period before conception very well, so data reported here can be treated as reliable [
2]. Because of the selected study design, it was not possible to include women who had miscarriages, ectopic pregnancies, or an induced abortion; as well as, to collect information on other important factors that may affect TTP, such as basal FSH levels and semen quality. Moreover, the study was conducted in only one region of the country, which represents one quarter of the entire Lithuanian female population.
Strengths and limitations of this study
-
The FSH receptor gene polymorphism may affect human reproduction by causing menstrual cycle disorders.
-
The present study demonstrates the effect of FSH receptor gene polymorphism on time to pregnancy* that has not been investigated till now.
-
The relationship of FSHr Serine680/Serine680 variant polymorphism to lower fecundity can have clinical relevance; e.g. more conservative infertility management can be suggested for women with unexplained infertility having this genetic variant.
-
Further studies including prospective studies on the impact of genetic factors on women’s fertility are needed.