Association of PIWIL4 genetic variants with germ cell maturation arrest in infertile Spanish men.

Dear Editor,

The PIWI proteins (originally P-element-induced wimpy testis in Drosophila) are predominantly present in the germ-line in diverse organisms and are involved in the processing of a class of small RNAs known as piRNAs (see Refs.1,2 for review). The human PIWI protein family consists of four members: PIWIL1–4. Of these, PIWIL4 is known to have essential roles in the first phases of spermatogenesis: its expression is restricted to gonocytes and it is required for transposon silencing.3 The lack of this gene in mice causes meiotic arrest in spermatogenesis.4 The goal of our study was to evaluate the frequency of several PIWIL4 genetic variants in our population to better define the relationship between PIWIL4 single nucleotide polymorphisms (SNPs) and both defective spermatogenesis and specific spermatogenic disorders.

We have genotyped four PIWIL4 SNPs (rs7110167 C/T, exon 7-synonimous; rs57607909 G/C, exon 9-missense; rs593690 T/C, exon 13-synonimous, and rs508485 C/T, 3’UTR) (), with a reported minor allele frequency (MAF) of >0.05 in the general Caucasian population (HapMap CEU), in genomic DNA from 79 nonobstructive infertile men (patients) with azoospermia (n = 61) or severe oligozoospermia (sperm counts <5 × 106 ml−1; n = 18), and 56 men diagnosed with obstructive azoospermia (as a consequence of congenital absence of the vas deferens or a previous vasectomy), who showed conserved spermatogenesis (controls). Patients underwent testicular biopsy, which showed a histological pattern with a total absence of germ cells (Sertoli cell-only syndrome, SCO; n = 40), >90% of maturation arrest, either in spermatogonia or in primary spermatocytes (MA; n = 22) and hypospermatogenesis (HS; n = 17). Men with a chromosomal aberration or with a Y-chromosome microdeletion had been previously excluded from the study. All the men included in this study were recruited from the Andrology Service of Fundaciσ Puigvert and gave their informed consent for the study, which was approved by the Institutional Ethical Committee. All of them were Caucasians of Spanish origin.

The current method for SNP determination was allelic discrimination using a Real Time PCR System and SNP Genotyping Assays (Applied Biosystems, Foster City, CA, USA). The χ2 test using Fisher exact test to the 5% limit was employed to compare carrier and allele frequencies in the patients and controls. Hardy–Weinberg equilibrium was tested by the χ2 test. The significance level was established at P < 0.05. These statistical analyses were performed using the SPSS software version 12 (Lead Technologies, Chicago, USA). Pairwise linkage disequilibrium (LD) between SNPs, haplotype frequencies, and association analyses were performed with SNPassoc, genetics and haplo.stats in R package (http://www.creal.cat/jrgonzalez/software.htm).

Our results show that the SNPs rs7110167, rs57607909, and rs593690 were in LD () in our cohort of individuals (D’ score >0.85), whereas rs508485 is independent of the other three (D’ <0.12). No significant differences in frequencies of allele, allele carrier, and genotypes were observed between patients and controls at the rs7110167, rs57607909, and rs593690 loci. Interestingly, we found that the frequencies of allele T (61.4% vs. 40.2%; P = 0.021) and allele T carrier (CT + TT) (86.4% vs. 60.7%; P = 0.033) in MA patients were higher than they were in the controls at the rs508485 locus. When the genotype frequencies were compared between the groups, a difference in genotype distribution was found in MA versus controls, although it was not statistically significant (P = 0.06) (). This SNP could have potential consequences for mRNA stability by altering the PIWIL4 3’UTR binding to miRNAs.

Linkage disequilibrium (LD) map of the PIWIL4 analyzed single nucleotide polymorphisms (SNPs), according to Haploview analysis (http://www.broadinstitute.org/scientific-community/software) of our population genotype data. Red boxes indicate strong evidence of LD according to D’ (% values are indicated) while white ones indicate strong evidence of recombination.

To further investigate the relationship of the four SNP in PIWIL4 and defects in sperm production, we performed a haplotype analysis of these variants in patient and control groups. A total of six haplogroups were found in a frequency > 0.05 (). The frequency of haplotype TGCT was higher in patients (29.4% vs. 18.9%; P = 0.036) and in SCO compared with controls (32.5% vs. 18.9%; P = 0.005), whereas a similar frequency of the haplotype TGCT was observed in HS and controls (17.6% vs. 18.9%). MA patients also presented a higher frequency of haplotype TGCT (33.4% vs. 18.9%; P = 0.24) and CGTT (14.3% vs. 7.6%; P = 0.24) than controls although the difference was not statistically significant, which is probably attributable to the low number of MA samples for the haplotype frequency estimation. Haplotypes CCTC (15.3% vs. 9.4%; P = 0.018) and CGTC (15.3% vs. 10.3%; P = 0.027) may also be risk factors for SCO infertility. A higher frequency of haplotype TGCT was also found in patients classified as azoospermic (29.3% vs. 18.9%; P = 0.036) and severely oligozoospermic (33.2% vs. 18.9%; P = 0.14), although the difference was not statistically significant for the oligozoospermic group; probably due to the lower number of samples.

Our results bring additional information from different populations about the involvement of PIWIL4 genetic polymorphisms in spermatogenic failure. An association was previously observed in an infertile Chinese population: patients with at least one rs508485 variant allele had a statistically significant risk of oligozoospermia (sperm counts <20 × 106 ml−1), but not of azoospermia.5 Our results contradict this as we have found this SNP to be associated with severe MA, a specific spermatogenic disorder leading to azoospermia. The cause of this discrepancy may be related to the influence of unknown ethnicity differences, but it could also be due to the absence of histological characterization of the infertile Chinese men. In fact, most of the testicular histological patterns in azoospermia correspond to SCO, which was not found to be associated with rs508485 SNP in the Spanish population. Under the assumption that genetic predisposition for spermatogenic impairment is most likely to be polygenic in nature, we conclude that genetic variants in PIWIL4 may contribute to the risk of spermatogenic deficiency in the Spanish population. We suggest that SNP rs508485, in the PIWIL4 gene is very likely to be relevant in genetic susceptibility to male infertility and specifically to premeiotic and meiotic maturation arrest, supporting the role of PIWI proteins in spermatogenesis and self-renewal of germ stem cells.2 Our findings may help to increase the understanding of the PIWIL4 genetic contribution to male infertility by defining the testicular histological phenotype associated with the presence of specific SNPs in PIWIL4. Previous data have provided evidence that other members of the PIWI pathway are affected by epigenetic mechanisms in human spermatogenic failure.6 Furthermore, the expression profile of transposon silencing genes (i.e. PIWIL2 and PIWIL4) was altered in azoospermia in cryptorchid boys.7 These, together with our results suggest that genetic and epigenetic alterations that affect the PIWI pathway contribute to unsuccessful sperm production, which might explain a subset of male infertility disorders.

AUTHOR CONTRIBUTIONS

This study was conceived and designed by SL. Samples and clinical data were selected by AM and LB. Experiments were performed by MN and XM. Data were analyzed by XM, MN, LB, and SL. The manuscript was written by SL, and was critically revised by XM and LB. All authors read and approved the final manuscript.

COMPETING INTERESTS

The authors declare no competing interests.

Table 1

Genetic description of PIWIL4 variants evaluated in our study

Table 2

Allele and genotype distributions of the four SNPs of PIWIL4 in infertile patients and controls

Table 3

Haplotype frequencies of four SNPs of PIWIL4 and association with infertile phenotypes