A Potential Relationship Between Estrogen Receptors Polymorphisms, Sperm Function and in vitro Fertilization Success: A Preliminary Study^*

Male infertility affects 20% to 25% of couples and in the half cases, conception failures etiology remains idiopathic [26]. It is estimated that approximately 15% of spermatogenic failure and/or sperm dysfunction is the result of gene mutations and chromosomal aberrations [26]. Some genetic disturbances can lead to androgen-estrogen imbalances, resulting in disrupted sperm function and conception problems. Studies on the polymorphic variations of estrogen receptors are becoming more common, but they are insufficient in addressing the genetic changes in estrogens and their receptors concerning the physiology and pathophysiology of the male reproductive system.

Impaired sperm production results from a multitude of factors: oxidative stress, environmental factors, hormonal imbalance, congenital diseases, genetic causes, and a host of other issues. Still, in approximately 50% of male infertility cases, the etiology remains unknown [20, 31].

Estradiol is a key factor for germ cell survival [25], and when production is disturbed, impaired sperm production occurs. Two main estrogen receptor subtypes mediate estrogen-related actions and are present in different stages of human germ cells [7, 8]. In recent years, there has been increasing interest in estrogen signal transduction disorders that may result from genetic polymorphisms in genes encoding estrogen receptors. There have been few studies investigating the direct relationship of these polymorphisms with semen parameters and sperm fertilizing ability. These studies have produced conflicting results. The results may vary due to ethnic differences in the population surveyed, genetic background of the participants, population sample size, or environmental factors or habits [5, 28]. The studies focusing on the ethnic diversity of single nucleotide polymorphisms (SNPs) showed a dependence on the relationship between the genes encoding estrogen receptor SNPs and osteoporosis [22, 38].

Estrogens have come to the forefront for their role in the pathophysiology of male infertility. These steroid hormones are produced in the testis from testosterone through the actions of cytochrome P450, and they act via specific receptors known as estrogen receptor 1 (ESR1) and 2 (ESR2) [11, 18]. Separate genes encode these receptors: the ESR1 gene is localized on chromosome 6q25.1-q25.2 [14, 27], and the ESR2 gene is found on chromosome 14q23.2-q23.3 [13, 15]. Estrogen receptors are expressed in almost all tissues of the male reproductive tract, and their expression varies along different stages of development [9, 17]. This suggests an influence of estrogens on the maturation process and function of the male reproductive system. Estrogens, acting through their receptors, regulate Sertoli and Leydig cell numbers and impact spermatogenesis [4, 6]. In animal models, impaired function or knock-out of ESR1 leads to infertility by reducing seminal fluid resorption and atrophy of the testes. In ESR1 knock-out mice, lowered mating frequency, inferior sperm parameters, and decreased fertilization potential have been observed [1, 12, 29]. ESR1 and ESR2, as well as aromatase P450, are expressed in male gametes [3, 25, 30], further strengthening a suggested relationship between estrogens and male fertility.

The development and increasing availability of in vitro fertilization techniques encourages exploration of factors that may influence their outcome. The influence of estrogens and their receptors on fertilization mechanisms can be observed in conventional in vitro fertilization (IVF) vs. the intracytoplasmic sperm injection (ICSI) model. Comparing the presence of ESR genetic polymorphisms with IVF outcomes allows an examination of estrogen action and the fertilization process. According to the literature, two ESR1 polymorphisms (rs9340799 and rs2234693) were described to have an association with azoospermia and severe oligozoospermia [21, 33]. ESR2 rs1256049 polymorphism was found to be linked with male infertility and it is probably associated with its influence on luteinizing hormone secretion [2]. The remaining two polymorphisms were suspected to be associated with fertility, but the results were inconclusive.

This study aims to analyze the relationship between ESR1 and ESR2 polymorphisms with sperm parameters and fertilization success defined as percentage of fertilized oocytes in both conventional in vitro fertilization, and intracytoplasmic sperm injection procedures. Selected ESR1 polymorphisms: rs9340799 (NC_000006.12:g.151842246A>G) and rs2234693 (NC_000006.12:g.151842200T>C) were recognized by XbaI and PvuII restriction enzymes, respectively and ESR2 polymorphisms: rs1256120 (NC_000014.9:g.64338283T>C), rs1256049 (NC_000014.9:g.64257333C>T) and rs4986938 (NC_000014.9:g.64233098C>T) were recognized by AlwNI, RsaI, and AluI, respectively We wanted to analyze the geno-type and allele’s distribution in patient subgroups with and without normozoospermia, as confirmed by concentration, motility with progressive movement, sperm morphology, and vitality (assessed using hypoosmotic swelling test).

The average age of male patients was 34 years, and their partners mean age was 32. The mean seminal parameters of the entire study group were as follows: sperm concentration of 2.5E7/mL, 18% sperm showing progressive motility, 4% spermatozoa with normal morphology and 53% live sperm. As defined using the 2010 WHO manual, normozoospermia, in all basic parameters (motility, concentration, and morphology), was observed in 28 patients (24% of the study group) [36]. Sperm parameters in the subgroups which underwent conventional IVF and ICSI are presented in Table 1. The concentration, motility, morphology, and viability of spermatozoa were significantly better in the subgroups which underwent conventional IVF. There was no difference in the average fertilization rate between groups (67% vs. 73%; p>0.05).

Restriction analysis was fully conclusive for 102 patients in the case of ESR1 and 92 in the case of ESR2. Despite the fact that the patients were a selected population treated by in vitro fertilization, all of evaluated SNPs distributions were consistent with the Hardy-Weinberg equilibrium (HWE). ESR1 and ESR2 genotypes did not differ significantly from those expected under HWE (p>0.05 for all genotypes), although for AluI analysis, only eight patients had genotype AA. For RsaI, genotype AA was not observed in any patient, and genotype GA was seen in eight patients. Concerning AlwNI restriction site, genotype TT was present just in four males. Absence of AA genotype analyzed by RsaI is consistent with the literature data as this genotype was marked the least frequent, regardless of the studied male population [19, 24, 37]. Concerning the distribution of genotypes, it is worth noting this study included a selected group of men, most of whom displayed decreased sperm parameters.

Using the chi-squared test and the Fisher exact test, we evaluated the association between sperm parameters and ESR1 and ESR2 polymorphisms. For ESR1, no predominance of any of the analyzed genotypes was found for any of the sperm parameters. The lowest p-value obtained was 0.07, and thus no statistical significance was found. This means that in the study population, the correct semen parameters, either oligozoospermia, asthenozoospermia or teratozoospermia, were not related to rs2234693 and rs9340799 polymorphisms. No relationship was observed between the studied seminological parameters and the genotypes of the ESR2 restriction sites. Only in the case ofrs1256120, the TT genotype is predominant among patients with ≥4% normal sperm morphology; however, this result is not statistically significant (p = 0.06) (Table 2).

Similarly, we observed no differences in sperm parameters between the genotypes using the Kruskal-Wallis and Mann Whitney U tests (p>0.05 for all sperm parameters in every SNP) (Table 3). For ESR1 polymorphisms, as related to sperm motility, rs2234693 and rs9340799 did not differ significantly between genotypes. Analysis of AluI, RsaI, and AlwNI restriction sites, all associated with the ESR2 gene, no correlation of genotypes was observed with sperm parameters. In the AluI study, p-values of 0.94, 0.69, and 0.88 (sperm concentration, morphology, and motility, respectively) between individual genotypes indicated no differences in analyzed parameters depending on the genotype. The results of sperm parameters indicate lower p-values within the rs1256120 TT genotype. However, there were no statistical differences observed (p = 0.1, p = 0.12, and p = 0.31 for concentration, motility, and morphology, respectively). In the case of RsaI restriction analysis, improved sperm parameters for the GG genotype were observed, but no statistical difference was found for any parameter (p = 0.32, p = 0.71, and p = 0.21 for concentration, motility, and morphology, respectively). In the studied population, however, no relationship between ESR2 polymorphisms and the seminological parameters has been shown (Table 3). Within the PvuII restriction site, there was no difference between the effectiveness of in vitro fertilization (measured by percentage of fertilized oocytes), regardless of genotype, both in the case of conventional IVF or ICSI (p = 0.16 and p = 0.75, respectively). For the XbaI restriction site, studies comparing the efficacy of conventional IVF and ICSI showed no difference. The association of XbaI restriction site polymorphisms with the effectiveness of fertilization by any of the methods used in in vitro fertilization may be excluded (IVF p = 0.73, ICSI p = 0.97). Due to the ESR2 localization in the midpiece area of the sperm, as well as the role of estrogens in the fertilization process, we analyzed the relationship of ESR2 polymorphisms with conventional IVF and ICSI fertilization rate no statistically significant relationship between individual genotypes with fertilization effectiveness in any of the methods. In the case of conventional IVF, we can observe a tendency towards a slightly lower percentage of fertilized egg cells for the AA genotype (p = 0.25; rs4986938; 1730 G>A) (Table 3).

Furthermore, we analyzed the association between SNP genotypes and fertilization rates from two methods of in vitro fertilization. We divided the patients into conventional IVF and ICSI groups. The patients were then further subdivided by fertilization of at least one oocyte or no oocytes. We used the chi-squared test to analyze the genotype distribution according to fertilization success. In the group which underwent conventional IVF, the males who were heterozygote TC carriers for the SNP recognized by PvuII had an improved treatment outcome than patients who did not have this genotype (p = 0.02). This was not observed for any of the ESR1 and ESR2 SNPs (Table 4).

For AluI and AlwNI restriction sites, there were no differences between genotypes and fertilization success in any of the utilized methods. Rs1256049 polymorphism (recognized by RsaI) and IVF efficacy, due to lack of AA genotype and low GA group genotype, were analyzed using the chi-squared test. This was performed to compare patients who had fertilized at least one oocyte to those with none. No relationship was found from this comparison in either the IVF or the ICSI groups (Table 4).

To assess the relationship of sperm parameters with the ESR1 and ESR2 gene polymorphisms, we proposed a comparison of their mean values within genotypes to the examined restriction sites. The presence of receptors in the sperm mitochondria, depending on the gene variant, could be related to varying mobility. However, the results did not confirm a correlation. The lack of association of different genotypes and semen parameters may be related to a relatively small population of normozoospermia cases. Normozoospermia was seen in 23 patients subjected to the analyses using PvuII (rs2234693; 397 T>C) and XbaI (rs9340799; 351 A>G), 19 for AluI (rs4986938; 1730 G>A), and 18 for RsaI (rs1256049; 1082 G>A). In the case of AlwNI (rs1256120; -458 T>C), normozoospermia existed in 20 individuals. This may explain the differing observations from those presented by the Lazaros et al. [24]. These authors found the relationship of the rs1256120 (defined by AlwNI) to higher gamete mobility as compared to the AG and AA genotypes; however, in the studied population of 114 men, as many as 85 had normozoospermia. However, it cannot be ruled out that, due to the post-genomic effect of estrogen receptors on sperm, structural disturbances within the restriction sites are not related to mitochondrial function and gamete mobility.

We found no statistically significant differences for the ESR1 and ESR2 restriction sites, indicating none of the genotypes were associated with the occurrence of normozoospermia in patients. The results presented are partly consistent with those published by Lazaros et al. [24]. Their work analyzed genotype distribution in patient groups with normal concentration and sperm motility then assessed groups with oligozoospermia and asthenozoospermia. They found that rs2234693 and rs9340799 polymorphisms did not correlate with sperm motility in patients with normal sperm kinetics. They noticed, however, in the asthenozoospermia group, CC and TC genotypes were associated with improved gamete motility than was the cytosine-free genotype, i.e. TT. In the case of XbaI restriction analysis, better sperm motility was demonstrated in patients with genotypes AG and GG than in the case of AA [23].

In the analysis of sperm concentration as related to rs2234693 and rs9340799 polymorphisms (defined by PvuII and XbaI, respectively), Lazaros et al. noted that the genotype TT analyzed by PvuII is accompanied by a larger amount of sperm in the group with normal sperm concentrations. For XbaI restriction analyses, a higher semen concentration was observed in the AA genotype as opposed to AG and GG. Among the patients with asthenozoospermia, these relationships were not observed. Concerning ESR2, the researchers found, contrary to the results presented here, higher values of sperm motility in progressive movement in patients with the rs4986938 GG genotype. At the same time, similar to the present study, no relationship was observed between sperm concentration and AluI restriction site, nor was one noted between sperm parameters and rs1256049 genotypes. Discrepancies in these results are quite difficult to interpret. The size of the study groups was comparable [24]. The authors referred to WHO standards in 1999, where the diagnosis of oligozoospermia was made when the concentration of sperm was <20 million/mL, and the motility of the gametes with progressive movement was estimated at <50% for asthenozoospermia. Additionally, the result discrepancies may be due to the different ethnic backgrounds of the groups. The origin of patients in the study group is an important factor that may affect polymorphism distribution.

Solakidi et al. reported ESR1 localized to the equatorial segment of the sperm head. They suggest the possible involvement of this receptor in combining the cell membranes of the male and female gametes [32]. It seems, therefore, that individual polymorphisms of the gene encoding ESR1 in men could be associated with differing percentages of fertilized egg cells in their partners. For this reason, data analysis was performed in the current study. Although no difference was shown between the effectiveness of classical in vitro fertilization and genotype in the PvuII analyses, the potency of this method seems to be slightly lower in the CC genotype than in TT and TC variants. Similar relationships were not observed in the case of ICSI. For the XbaI restriction site, the efficacy of conventional IVF and ICSI showed no differences, which was also observed for AluI. In the case of conventional IVF, one can observe a tendency towards a slightly lower percentage of fertilized egg cells for the AA genotype. These results do not coincide with the results of Aschim et al. [2]. These researchers noticed the rarest occurrence of the AA genotype in infertile men, although no significant differences between genotypes were observed. However, this work used a different criterion for male infertility (sperm concentration <5 million/mL and no partner study), so the results are difficult to compare. For patients treated with ICSI in our work, the dominant trend of one of the genotypes analyzed by AluI in the context of the effectiveness of ICSI was not observed. It can be presumed that, while ESR1 and ESR2 polymorphisms may be related to conventional IVF efficacy (and perhaps a natural conception) and in ICSI conditions, they do not affect its result.

In the AlwNI and RsaI assessment, there were no differences between genotypes and the effectiveness of fertilization in any of the methods assessed. Zhang et al. conducted a study of this polymorphism in the context of fertilization effectiveness in the Chinese population [37]. These authors observed the occurrence of the AA genotype in 32 of 865 couples and showed that, although there was no difference between the genotypes and effectiveness of fertilization, it was the AA genotype that was associated with poorer embryo quality. These results encourage similar research in Caucasian populations.

Significantly higher occurrences of the TC genotype of the PvuII restriction site were seen where at least one oocyte was fertilized (p = 0.02). This was not observed for other genotypes analyzed with PvuII, nor for the genotypes identified by XbaI or any of the SNP ESR2 genotypes. Therefore, it is the only statistically significant relationship between the studied estrogen receptor polymorphisms and fertilization effectiveness assessed in this work. Therefore, the conclusion regarding a favorable relationship between the rs2234693 TC genotype and the fertilizing capacity of semen should be treated with caution. With the present size of the group, the statistical power ranged from 0.05 to 0.99 and the continuation of research with an increase in the size of the group would be advisable, because major determinant of statistical power is allele frequency.

Because of the polygenic nature of spermatogenic disorders, additional loci could be involved in spermatozoa defects and conception ability. This may be related to estrogens, their receptors, or other core genes involved in estrogenic and estrogen-related pathways. In most cases, we did not discover a relationship between both estrogen receptors polymorphisms and sperm function, except forrs2234693 and fertilization rate. There was no association with ESRs polymorphisms and standard WHO semen parameters.