Follicle-Stimulating Hormone Receptor Gene Haplotypes and Male Infertility in Estonian Population and Meta-Analysis

Abstract

Follicle-stimulating hormone (FSH) is crucial for male fertility and it exerts its effects via a gonad-specific receptor (FSHR). In the present study, the common G-29A, A919G, and A2039G polymorphisms in the FSHR gene were analysed in 150 (36 non-obstructive azoospermia and 114 individuals with oligozoospermia) patients and 208 normozoospermic men. The results showed that the FSHR polymorphisms were not associated with either azoo- or oligozoospermia as the distributions of alleles, genotypes, and haplotypes among patients and controls were similar. Amongst normozoospermic men, those carrying at least one minor A allele (GA and AA genotypes) of the G-29A polymorphism had a smaller mean testicular volume compared to men with GG homozygosity (25.8 ml vs. 27.4 ml, respectively; P=0.013). In a subsequent meta-analysis combining our data with previous studies, the G-29-A919-A2039 haplotype was shown to be more prevalent in normozoospermic men than in azoospermic patients (38.4% vs. 33.9%, respectively; χ²test, P=0.045), indicating that this haplotype may be a protective factor against male sterility. In conclusion, we suggest that FSHR haplotypes are not considerable risk factors for spermatogenic failure. The protective nature of G-29-A919-A2039 haplotype cannot be concluded without additional studies.

KEYWORDS:

• follicle-stimulating hormone receptor gene
• male infertility
• single nucleotide polymorphisms

Abbreviations
FSH:	=	follicle-stimulating hormone
FSHR:	=	follicle-stimulating hormone gonad-specific receptor
SNP:	=	single nucleotide polymorphism
RFLP:	=	restriction fragment length polymorphism
PCR:	=	polymerase chain reaction
T:	=	testosterone

INTRODUCTION

Male fertility is a sophisticated process that is regulated by a set of different hormones including follicle-stimulating hormone (FSH). FSH is critical for the production of mature spermatozoa. It has been proposed that in the adult testis, FSH is acting as a spermatogonial survival factor [Ruwanpura et al. 2008]. This gonadotropin is also a necessary signal for Sertoli cell proliferation, a process that occurs during fetal, neonatal, and prepubertal life [Cooke et al. 1992; Cortes et al. 1987]. Each Sertoli cell supports a fixed pool of germ cells during their development into spermatozoa, thus determining the final size of the testicles and, at least in part, sperm production capacity in adulthood [Arslan et al. 1993; Johnson et al. 1984; Orth et al. 1988].

In order to receive the stimulatory effect of FSH, Sertoli cells produce FSH receptor (FSHR) on their membrane surface. Given the significant role of FSH in fertility, genetic abnormalities of FSHR would be expected to affect normal reproduction. Mutations in the FSHR gene have been detected and shown to cause infertility in women [Aittomäki et al. 1995]. Though, so far, no mutation in men has been indicated as an unequivocal cause for their infertility [Simoni et al. 1999; Tapanainen et al. 1997; Tuerlings et al. 1998], homozygous Ala189Val mutation is shown to result in reduced testicular volume, increased serum FSH concentration, and suppression of spermatogenesis to a variable extent [Tapanainen et al. 1997].

While mutations affecting the FSHR gene are rare, polymorphisms seem to be a common phenomenon. Mutation screening of the FSHR gene has led to the discovery of various single nucleotide polymorphisms (SNP) in the core promoter [Wunsch et al. 2005] as well as in the non-coding and coding regions of the gene [Gromoll and Simoni 2005]. Currently, two common SNPs in exon 10 of FSHR are the most thoroughly studied. Both are tightly linked A to G non-synonymous substitutions at nucleotide position 919 (codon 307) and 2039 (codon 680) leading to Thr to Ala and Asn to Ser amino acid changes, respectively [Simoni et al. 1997]. In females, these SNPs have been shown to influence serum FSH levels during the follicular phase of the menstrual cycle [Greb et al. 2005] and the sensitivity of FSHR to FSH stimulation in assisted reproduction [Behre et al. 2005; Perez Mayorga et al. 2000].

The role of these aforementioned FSHR polymorphisms in the etiology of male infertility is unclear and remains controversial. Previous studies in different ethnic groups revealed that the allelic variants of those polymorphisms are similarly distributed among fertile and infertile men [Pengo et al. 2006; Simoni et al. 1999; Song et al. 2001]. Recently an additional polymorphism in FSHR gene core promoter at position −29 has been evaluated in combination with A919G (Thr307Ala) and A2039G (Asn680Ser) SNPs [Ahda et al. 2005]. Gene haplotypes have better predictive power to disclose the existing associations between the gene variants and disease conditions that would remain hidden if only individual SNPs are analysed. The previous studies have demonstrated discrepant findings on the putative risk and protective nature of A-Ala-Ser and G-Thr-Asn haplotypes in severe spermatogenic impairment, respectively [Ahda et al. 2005; Pengo et al. 2006], thus reserving this conflicting issue for forthcoming studies.

In the present case-control study we investigated FSHR (G-29A, A919G, and A2039G) allele, genotype, and three-SNP haplotype distributions in azoo- and oligozoospermic infertile patients and normozoospermic men in Estonian population. Additionally we combined our data with previous studies [Ahda et al. 2005; Pengo et al.2006] for a meta-analysis to look at the FSHR haplotype frequencies in a larger population of infertile and fertile European men.

RESULTS

Table 1 shows the mean values of the clinical parameters in patients and controls. Genotype analysis confirmed a link between the major A alleles (A919-A2039) and minor G alleles (G919-G2039) in 149 patients and 207 controls. Two individuals, a single individual in the oligozoospermia (G919-A2039) and a single individual in the normozoospermia (A919-G2039) group, had recombinant alleles.

TABLE 1  Comparisons of Clinical Parameters Between Patients and Controls.

	Patients		Controls
Clinical parameters	n	Mean±SD	n	Mean±SD	Mann-Whitney U-test (P value)
Age	150	31.8±5.6	208	18.7±1.6	<0.0001
Testicular volume (ml)	146	18.6±6.8	208	26.6±4.9	<0.0001
Abstinence (days)	146	4.0±1.6	208	5.2±2.4	<0.0001
Sperm concentration (10^6/ml)	150	2.0±2.4	208	149.7±88.4	<0.0001
Sperm progressive motility (%)	107	19.4±16.2	206	57.9±9.9	<0.0001
FSH (IU/l)	144	12.3±11.6	79	3.3±1.6	<0.0001
T (nmol/l)	141	18.1±6.8	79	22.0±7.3	<0.0001

FSH: follicle-stimulating hormone; T: testosterone; SD: standard deviation.

When the three SNPs were analysed individually, allele and genotype frequencies did not significantly differ between patients and controls. Core promoter G-29A variation GG, GA, and AA genotype distributions were as follows: 38.9, 55.6, and 5.5% in azoospermic patients, 56.1, 36.0, and 7.9% in oligozoospermic patients, and 52.9, 40.9, and 6.2% in normozoospermic controls, respectively. The AA, AG, and GG genotypes of the A919G polymorphism were found in: 27.8, 61.1, and 11.1% of azoospermic patients, 35.1, 43.9, and 21.0% of oligozoospermic patients, and 32.2, 51.0, and 16.8% of control men, respectively. In the case of the A2039G polymorphism, the incidence of AA, AG, and GG genotypes were: 27.8, 61.1, and 11.1% in patients with azoospermia, 35.1, 44.7, and 20.2% in patients with oligozoospermia, and 31.7, 51.5, and 16.8% in controls, respectively. The percentages of the most common haplotypes containing G-29A, A919G, and A2039G SNPs were comparable in azoo- and oligozoospermia patients, as well as in the combined group of all infertile men and among normozoospermic controls (Table 2). Although the numbers and percentages of undecided haplotypes that were inferred from the indistinguishable double heterozygotic (A-Thr-Asn/G-Ala-Ser or A-Ala-Ser/G-Thr-Asn) individuals were excluded from Table 2, their frequencies among all infertile patients (18.7%) were comparable to those of control men (19.8%). Furthermore, the frequencies of combinations formed by FSHR haplotype pairs were also similar across all study groups (data not shown).

TABLE 2 The Numbers and Percentages of Haplotypes Consisting of FSHR Gene G-29A, A919G, and A2039G SNPs in Patients and Controls.

	Azoospermia patients		Oligozoospermia patients		All infertile patients		Controls
Haplotype	n	%	n	%	n	%	n	%
A-Thr-Asn	9	12.5	25	11.0	34	11.3	44	10.6
A-Ala-Ser	1	1.4	20	8.8	21	7.0	26	6.3
G-Thr-Asn	19	26.4	91	39.9	110	36.7	154	37.0
G-Ala-Ser	15	20.8	63	27.6	78	26.0	109	26.2

Comparing the clinical parameters among normozoospermic men according to the genotype of each of the SNPs revealed that men that carried at least one minor A allele (GA and AA genotypes) of the G-29A polymorphism had smaller mean testicular volumes compared to men with GG homozygosity (25.8 ml vs. 27.4 ml, respectively; multivariate linear regression model adjusted by age and abstinence, r=−1.69, P=0.013). However, no other relationships between sperm characteristics (sperm concentration and progressive motility), hormone (FSH and testosterone, T) levels, and FSHR genotypes were discerned (data not shown).

In a subsequent meta-analysis, we did not observe a correlation between A-Ala-Ser haplotype and male infertility. However, the G-Thr-Asn haplotype was more prevalent in normozoospermic men than in azoospermia patients (38.4% vs. 33.9%, respectively; χ²test, P=0.045; Fig. 1) and may therefore be classified as a protective factor against male sterility.

FIGURE 1  Distribution of the G-Thr-Asn haplotype among azoo-, oligozoo-, and normozoospermic individuals in the present study and those of Ahda et al. [2005] and Pengo et al. [2006], as well as in the meta-analysis. Pearson's Chi-squared (χ²) test with Yates' continuity correction was used for statistical analysis.

DISCUSSION

In the present study azoo-, oligozoo-, and normozoospermic men were genotyped for the most common FSHR allelic variants in the core promoter and exon 10. We analysed the SNPs individually and as haplotypes with respect to male fertility status. The distributions of FSHR allele, genotype, and haplotype frequencies among azoo-, oligozoo-, and normozoospermic men were similar. We identified that the presence of at least one minor A allele at position −29 was associated with markedly reduced testicular volume among men with sperm counts within the normal range. However, FSHR variants did not noticeably impact the other clinical fertility parameters studied.

Previous studies examining potential associations between FSHR polymorphisms and male fertility parameters have produced contradictory results [Ahda et al. 2005; Pengo et al. 2006; Simoni et al. 1999; Song et al.2001]. A recent meta-analysis suggests that there is a lack of a causal association between the A2039G (Asn680Ser) polymorphism and impaired spermatogenesis [Tüttelmann et al. 2007]. Yet, studies have not investigated the G-29A polymorphism in relation to other clinical parameters besides sperm count and serum FSH [Ahda et al. 2005]. We compared the clinical parameters of fertile men according to their G-29A genotypes. Men with the less prevalent GA and AA genotypes had smaller testicles and also tended to have a relatively, though not statistically relevantly, lower sperm count (data not shown). Based on databases, G-29A at the core promoter was thought to be a part of the potential binding domain for the ETS transcription factors with likely interruption of the binding motif in the case of G to A exchange [Simoni et al. 2002]. It is possible that this alteration could have a negative effect on the expression of the FSHR gene and thus might decrease the cells' responsiveness to FSH. As FSH is necessary for Sertoli cell proliferation in fetal to prepubertal development, the diminished FSH sensitivity could finally lead to reduced testicular size and reduced sperm production in adulthood [Cortes et al. 1987; Orth et al. 1988]. Nevertheless, Wunsch et al. [2005] reported no major changes in the transcriptional activity induced by this polymorphism in vitro. Also, no significant effect of G-29A polymorphism was observed on serum FSH levels and ovarian response in women undergoing controlled ovarian stimulation for IVF treatment [Wunsch et al. 2005]. Furthermore, as in the current study, the G-29A alleles and genotypes were equally distributed in sterile and fertile men in accordance with Ahda et al. [2005], this genetic change, particularly when considered alone, probably does not have a profound influence on the overall male reproductive success.

Currently, only two studies have focussed on FSHR haplotypes in the context of male infertility [Ahda et al. 2005; Pengo et al. 2006]. Ahda et al. [2005] studied azoospermic men and suggested that the A-Ala-Ser and G-Thr-Asn haplotypes serve as risk and protective factors for male sterility, respectively. However, in the subsequent study, FSHR haplotypes were deemed to be unrelated to the etiology of azoo- and oligozoospermia [Pengo et al. 2006]. Our results corroborate the absence of an association between FSHR haplotypes and severely impaired spermatogenesis. However, our study included fewer azoospermic men than Ahda et al. [2005] which could account for this discrepancy.

The size of the study and control populations is an important aspect when conducting association studies focusing on heterogeneous pathologies like male infertility. In studies with a smaller number of participants the false-positive associations between the genetic changes and the characteristics under examination can be drawn, while the true associations may remain hidden. The false-negative conclusions mostly stem from the insufficient power to detect the true associations. However, the increased study population in case-control studies would enable one to avoid presenting biased results. Accordingly, we combined our results with those from earlier studies [Ahda et al. 2005; Pengo et al. 2006] to perform a meta-analysis of more than a thousand European participants. The findings of themeta-analysis support our conclusion that the A-Ala-Ser haplotype probably does not play a critical role in the genesis of male infertility. However, theG-Thr-Asn haplotype was more frequent among normozoospermic men than in azoospermia patients emphasizing its possible protective character. The size of the control group was greater than each of the study groups in meta-analysis. This contrasts with the Ahda et al. [2005] study in which the control group was a half the sizeof azoospermic cases, thus limiting their interpretations. In conclusion, we suggest that the studied FSHR polymorphisms and haplotypes do notrepresent a considerable risk parameter for spermatogenic failure, but the protective nature ofthe G-Thr-Asn haplotype cannot be concluded without additional studies.

Given the important role of FSH in male reproductive processes, it is interesting that different FSHR alleles or genotypes studied so far do not clearly affect clinical fertility parameters (e.g. serum T and FSH levels, sperm concentration, etc). To date, the two SNPs in exon 10 and the SNP in the core promoter region of the FSHR gene are the most studied mainly due to their relatively high frequency in the population. If those polymorphic loci would have even a slight yet not infertility causing effect on FSHR function, the FSH and testosterone levels would be expected to be influenced. Song et al. [2001] observed a correlation of the G allele of the A919G polymorphism with heightened FSH levels and lower testicular volume. However, none of these findings have been confirmed in other studies [Ahda et al. 2005; Asatiani et al. 2002; Simoni et al. 1999].

Undoubtedly, many more polymorphisms either in coding or non-coding areas are present in the FSHR gene, although at a very low frequency. These variations could alone or in combination with other genetic changes contribute to the impaired spermatogenesis or alterations in other fertility parameters. The fraction of infertile patients affected by those genetic changes would probably be very small. Therefore, larger association studies with more defined study groups are necessary to determine the influence of these rare polymorphisms on the normal testicular function.

MATERIALS AND METHODS

Study Participants

This study was approved by the Tallinn Medical Research Ethics Committee. A total of 150 infertile men were recruited. Patients with non-obstructive idiopathic azoospermia (n=36) or oligozoospermia (sperm count <20×10⁶ spermatozoa per ml of ejaculate, n=114) and without any obvious cause of infertility were considered eligible for the study. Patients with chromosomal abnormalities, Y chromosome microdeletions, and genital tract pathologies were excluded. The control group consisted of 208 military conscripts selected based on their sperm count of ≥75×10⁶/ml. All study participants underwent a medical examination, which included determination of testicular size using Prader orchidometer, measurements of serum FSH and T levels by chemiluminescence immunoassay (Immulite 2000; Siemens Medical Solutions, Los Angeles, CA, USA), and semen analysis according to the guidelines of the World Health Organization [WHO 1999].

Genotyping

The following FSHR polymorphisms were genotyped: G-29A (rs1394205), A919G (Thr307Ala, rs6165), and A2039G (Asn680Ser, rs6166). Genomic DNA was extracted from peripheral blood using a salting-out method [Aljanabi and Martinez 1997]. FSHR gene polymorphisms were analysed using the restriction fragment length polymorphism (RFLP) technique. Polymerase chain reaction (PCR) primers were: 5′-TCATAAGGGCACTGTGTGGA-3′ (forward) and 5′-TTGGCAGAGAAAAA CCCTGT-3′ (reverse) for G-29A; 5′-ACCCTGCACAAAGACAGTGA-3′ (forward) and 5′-GCTGTAGCTGGTCTCATTGT-3′ (reverse) for A919G, and 5′-TCTACCTCACAGTGCGGAAC-3′ (forward) and 5′-TAATAGTTCCTGACCAAT TTACCTTA-3′ (reverse) for A2039G.

PCR was performed in a total volume of 10 μl, containing 1×reaction buffer (800 mM Tris-HCl (pH 9.4–9.5), 200 mM (NH₄)₂SO_4, and 0.2% w/v Tween 20; Solis BioDyne, Tartu, Estonia), 2.5 mM MgCl₂ (Solis BioDyne), 0.25 mM dNTP (dATP, dCTP, dGTP, and dTTP; Fermentas, Vilnius, Lithuania), 0.4 μM primers (Metabion GmbH, Martinsried, Germany), 1U HOT FIREPol DNA polymerase (Solis BioDyne), and genomic DNA template (∼50 ng). GeneAmp^® PCR System 9700 cycler (Applied Biosystems, Foster City, CA, USA) was used for amplification using the following program: 10 min at 95^oC, followed by 35 cycles of denaturation at 95^oC for 30 s, primer annealing at 58^oC for 30 s (except for A2039G, in which case 55^oC was used), extension at 72^oC for 30 s, and final extension at 72^oC for 7 min. The PCR fragments were subsequently digested with endonucleases (Fermentas) according to the manufacturer's recommendations and cleavage products were separated by 3% agarose gel electrophoresis. The G-29A polymorphism creates a cleavage site for MboII, while A919G and A2039G SNPs were digested with Eam1105I and BseNI restrictases, respectively.

In the Caucasian population, the four most prevalent haplotypes formed by the three SNPs of the FSHR are A-29-A919-A2039 (A-Thr-Asn), G-29-A919-A2039 (G-Thr-Asn), A-29-G919-G2039 (A-Ala-Ser), and G-29-G919-G2039 (G-Ala-Ser). These four haplotypes were paired into ten combinations (genotypes), from which two represent indistinguishable double heterozygotes (A-Thr-Asn/G-Ala-Ser or A-Ala-Ser/G-Thr-Asn) that are considered together.

Meta-Analysis and Statistical Testing

Azoo- and oligozoospermia patients and normozoospermic men from this study and others [Ahda et al. 2005; Pengo et al. 2006] were combined for meta-analysis. Distributions of risk (A-Ala-Ser) and protective (G-Thr-Asn) haplotypes for male infertility, as determined according to a previous study [Ahda et al. 2005], were evaluated among 415 azoospermic and 291 oligozoospermic patients, and 486 normozoospermic controls. The control group consisted of 208 healthy normozoospermic men with a sperm count of ≥75×10⁶/ml (present study) as well as 92 [Pengo et al. 2006], and 186 [Ahda et al. 2005] normozoospermic men with a sperm count of ≥20×10⁶/ml according to WHO [1999] criteria.

Statistical analyses were performed using R2.7.2 statistical software (Free Software Foundation, Boston, MA, USA). Clinical parameters were provided as mean±standard deviation (SD) and were compared among cases and controls using a Mann-Whitney U-test. Pearson's Chi-squared (χ²) test with Yates' continuity correction was used to reveal possible associations between FSHR polymorphisms and male infertility within our data and subsequently in a meta-analysis. Linear regression analysis was used to determine correlations between clinical parameters and the polymorphisms studied among control men. A P value of <0.05 was considered statistically significant.

ACKNOWLEDGMENTS

The authors acknowledge all voluntary participants of the study. The study was supported by the Estonian Science Foundation (grants nos. 6498 and 6585), the Estonian Ministry of Education and Science (core grants nos. SF0180044s09 and PBGMR07903), and the European Union through the European Regional Development Fund through the Centre of Excellence in Genomics, Estonian Biocentre and University of Tartu.

Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.