Genetic variation in SPAG16 regions encoding the WD40 repeats is not associated with reduced sperm motility and axonemal defects in a population of infertile males

Infertility impacts approximately 9% of couples globally, with reports ranging from 3.5% - 16.7% of couples [1], and it is estimated that male factor infertility plays a role in as many as 55% of cases [2]. There are multiple causes of male infertility, including congenital factors and environmental exposures [3‐5], as well as gene mutations which cause defects in spermatogenesis and sperm flagellar dysfunction.

The most studied mutations associated with abnormal sperm motility and male infertility cause the primary ciliary dyskinesias (PCD) [6, 7]. Genes involved in the structure and function of the “9 + 2” axoneme, the principle scaffold and regulator of motile cilia and flagella, have been extensively studied and characterized in the biflagellate eukaryotic algae Chlamydomonas, wherein loss of axonemal components has been shown to cause immotility [8, 9]. Most of the mutations identified to date that cause PCD in humans encode proteins that are associated with the axoneme outer doublets. However, relatively little is known about the contribution of mutations or genetic variation in genes encoding the central apparatus of the axoneme to sperm motility defects. It is known that the “9 + 2” axoneme is strongly conserved in overall structure and function [10, 11].

We previously reported that mice lacking Spag6, which encodes the mammalian orthologue of the Chlamydomonas axonemal central apparatus protein PF16, exhibited a phenotype of both male infertility associated with severe sperm motility defects and axonemal structural abnormalities [12], demonstrating that loss of an axonemal central apparatus protein can have dramatic effects in mammals as well.

SPAG16, the mammalian orthologue of the essential Chlamydomonas central apparatus protein PF20 [13, 14], is also required for male fertility and sperm motility in mice [15]. The Spag16 gene encodes two major transcripts in mice, which generate SPAG16L, a central apparatus protein, and SPAG16S, a protein localized to the male germ cell nucleus and cytoplasm. Male mice homozygous for a mutation ablating SPAG16L production have a sperm motility defect and are infertile [15]. The axonemes of SPAG16L-deficient mice also show abnormal responses to calcium, indicating that loss of SPAG16L disrupts either the ability to process or the ability to respond to key molecular signaling [16]. When the transcripts encoding SPAG16L and SPAG16S are both disrupted in a transgenic male mouse, there is haploinsufficiency and abnormal spermatogenesis in the chimeric state, implicating SPAG16S in fundamental processes of sperm formation distinct from the structural role of SPAG16L [17]. Importantly, the Spag16 mutant allele affecting both isoforms was not transmitted to offspring by chimeric males, demonstrating that heterozygous/non-homozygous Spag16 mutation may cause a pronounced phenotype, and that the two murine SPAG16 isoforms each play different but essential roles.

We have also reported previously that sperm from human subjects heterozygous for a frame shift mutation in SPAG16 exhibited instability of important central apparatus components SPAG16L, SPAG17, and SPAG6 [18]. While both subjects in this report were fertile, observed biochemical instability of the sperm axoneme suggests that even heterozygous disruption of SPAG16 has phenotypic consequences that may reduce fecundity. Additionally, a preliminary study of oligozoospermic and oligoasthenozoospermic males has identified 29 previously unreported mutations in SPAG16 transcripts isolated from ejaculated sperm, further suggesting a critical role for SPAG16 in male fertility [19].

In the present study, 60 males of Western European (Italian) origin volunteered for semen and genomic analyses following diagnosis of male factor infertility. We evaluated the genomic region encoding the highly conserved WD domains of SPAG16 proteins (exons 10–16). WD regions are semi-conserved 20 amino acid sequences giving rise to a β-propeller tertiary structure. These domains are the only known functional domains found in SPAG16 proteins, both isoforms of which contain seven such domains. These β-propeller structures are classically thought to mediate protein-protein interactions (reviewed in [20]), playing crucial roles in macromolecular assembly relevant to a diverse array of cellular tasks, including cell growth and division, intracellular communication, apoptosis, and transcription regulation [21]. Recent reports have also demonstrated that some WD domains interact directly with specific RNA sequences, further expanding the range of potential roles played by members of this class of proteins [22].

To better understand potential links between SPAG16 gene variation and specific mechanisms of infertility, semen analysis was performed for each subject [23]. In addition, sperm ultrastructure was assessed by transmission electron microscopy (TEM) and described according to previously established standards [24, 25].

Based on our previous observations of Spag16 gene function in mice, and the influence of a heterozygous mutation affecting the WD repeat region in humans on central apparatus stability, we hypothesized that functional SNPs in the human SPAG16 gene could cause defects in sperm motility and ultrastructure manifest in both the heterozygous and the homozygous states. The present study analyzed genetic variation in a population of infertile males, and their association with parameters that are potentially regulated by SPAG16.

In the SPAG16 regions analyzed, both major and minor alleles were present in the study group for four known polymorphisms, and no previously unreported mutations were detected. At rs2042791 [29], the minor allele replacement of an adenosine by a cytosine residue causes the replacement of a glutamine at position 361 [GenBank: NP_078808.3] with a histidine. Previous genotyping studies have reported a minor allele frequency of 0.39 [26] in individuals of Caucasian ancestry for this SNP. Fisher’s exact test demonstrated no significant difference between the measured minor allele frequency of 0.408 in our sample population and that established in control populations (p > 0.05). While the present study lacks sufficient power to detect small effects of these alleles on variation in observed parameters of sperm characterization, among patients carrying this SNP no single factor was uniformly deficient within the distribution of sub-normal semen analysis of the study participants. Full genotype and semen analysis results are included for reference (Additional file 2: Table S2).

The lack of association between this SNP and sperm dysfunction is not surprising, given that the glutamine replacement by histidine is predicted to be tolerated in the protein structure [27]. Further, the amino acid is not conserved in mammals – Mus musculus [GenBank:AAI20669.1, NP_080004.1, NP_083436.2] and Rattus norvegicus [GenBank:NP_001128200.1, AAI58603.1] reference sequences for the orthologous proteins in fact report a histidine at this position, consistent with the human minor allele. The same amino acid replacement is predicted in Pan troglodytes [GenBank:XP_001148592.1, XP_526016.2, XP_001148393.1]. These data are consistent with the hypothesis that structural rather than sequence homology is essential to maintenance of axoneme function.

The rs2042792 minor allele was not found at a significantly different frequency than in the reference population (p > 0.05), and did not correlate with assessed sperm phenotypes.

Rs2042791 exists in LD with rs2042792 in our population, but no phenotypic consequences were associated with any of the observed haplotypes.

Minor allele replacement of an adenosine with a cytosine in rs12623569 causes an amino acid replacement at position 425 [GenBank:NP_078808.3], with a threonine taking the place of a lysine. Previous genotyping studies have shown a minor allele frequency for this SNP in individuals of Caucasian ancestry of 0.30 [26]. The present sample population exhibited a minor allele frequency of 0.333, which was not significantly different than that in control populations (p > 0.05). While the lysine at position 425 encoded by the major allele is present in other mammals (Mus musculus – GenBank:NP_080004.1, NP_083436.2; Rattus norvegicus – GenBank:NP_001128200.1, AAI58603.1), replacement by threonine is predicted to be tolerated by the protein structure. Interestingly, minor allele carriage was found to be nominally associated with a higher percentage of normal accessory fibers of the axoneme. Although the statistical significance of this association was lost upon correction for multiple testing, further studies on a larger study population may be warranted to identify contributions of the SNP to sperm flagellar structure.

At rs16851495, minor allele replacement of guanine with an adenine residue does not affect protein translation directly, as the position lies outside the translated exon region. However, minor allele carriage was nominally associated with the reduced presence of normal axoneme structure in the study population. The statistical significance of this association did not withstand correction for multiple testing. However, the limited power of our study may have contributed to the absence of a robust association.

For all SNPs discussed, both heterozygous and homozygous individuals were present in the sample population, and statistical analysis demonstrated no difference in genotype frequency between the sample and control populations (p > 0.05). While the study did not have sufficient power to analyze possible contributions to complex traits, the results negate the hypothesis that amino acid modification at the tested positions may result in a loss of protein function that would mimic the severe phenotypes observed in transgenic mouse studies.

Our results suggest that non-synonymous amino acid substitutions at residues 361 and 425 in the SPAG16 protein are not sufficient to explain a reduction of sperm motility and fertility index, the presence of axonemal/periaxonemal alterations, or an increased percentage of sperm pathologies in the assessed patient population. Based on the strong homology between members of the WD protein family, it is tempting to suspect that these resides may be non-essential in other WD repeat proteins as well.

The profound defects observed in studies of Spag16 gene effects in other species suggest that a functional mutation would significantly perturb sperm function, and would be observed even in a small sample population. Our observations do not, however, preclude the possibility that alternative variations in the SPAG16 gene cause an increase in sperm ultrastructural alterations, and a reduction in sperm motility. Indeed, our previous studies suggest that more significant SPAG16 mutations, such as a frame shift, can reduce sperm central apparatus stability. Further studies of SPAG16 gene variation are warranted to offer a comprehensive understanding of the gene’s contributions to male fertility. The sensitivity of the present study to detect variation with the sample is limited by the lack of similarly extensive sperm analysis data, in particular TEM ultrastructural analysis, from a control population. Future development of this area of research should include control samples from men with known fertility and/or sperm count, motility, and structure in the normal range.

Future studies in mammals are necessary as well to explore the nature of Spag16 gene evolution, and the possibility that male germ cell-specific isoforms exist in multiple species. The only identified rat SPAG16 protein is similar to SPAG16S in size, and is derived from mRNA similar to murine SPAG16S, with a 5^′UTR from the region upstream of the first exon suggestive of independent transcription, rather than splice variation. A putative SPAG16S promoter and 5′UTR have been identified in humans as well [GenBank: EF591776.1], raising the possibility that SPAG16S maybe be common to mammalian male germ cells. We have recently shown that murine SPAG16S is enriched within nuclear speckles of male germ cells [30]. This unique localization, combined with phenotypic results from various Spag16 transgenic mice, strongly suggests a unique processing role for the SPAG16S protein, and thus a distinct role for the Spag16 gene apart from its structural role in the axoneme.

In a sample population of 60 infertile males of Western European origin, mutations in SPAG16 were not significantly associated with a single phenotype of sperm alteration. The findings suggest that mutations and or genetic variation in the SPAG16 regions encoding the protein’s WD repeats are not likely to be major causes of sperm motility and sperm flagellar defects.

Research was carried out in full compliance with the Helsinki Declaration of ethical principles for medical research involving human subjects. All patients signed a declaration of informed consent to participate in the research. The work was approved of and performed in full accordance with policies governing human subject research at the by the Ethics Committee for the University of Siena (CEL-AOUS) for specimens collected by non-invasive methods for clinical analysis, and the Virginia Commonwealth University Institutional Review Board.