Semen Parameters Are Associated with Alterations in the Sperm Epigenome of Infertile Men

For half a century, the best studied and least predictive measure of male fertility status in clinical use has been the routine semen analysis. More recently, sperm strict morphology and sperm DNA fragmentation properties have shown more predictive power for male infertility. Most recently, sperm epigenetic profiles have been shown to correlate with both natural conception rates and embryo quality at IVF. Here we extend our prior work to describe new associations between routine semen parameters and sperm DNA methylation disruptions among infertile men, and seek to identify individual genes related to them. Prospective, controlled, multi-institutional clinical trial of infertile couples. Using array technology, we profiled DNA methylation levels at more than 480,000 CpG sites in the sperm from semen samples in 210 patients. Samples were collected from 11 major fertility clinics throughout the United States and Canada, and were derived from couples with no identified female-factor infertility issues, and those with a minimum sperm concentration >2 mill/mL. Data was normalized to correct for potential batch effects. Differentially methylated positions (DMPs) were detected by performing an F-test for each locus to measure the relation between semen parameters and methylation, with a p-value cut-off of .05 for statistical significance. When comparing sperm methylation patterns in men (n=26) with high vs. low progressively motile sperm, DMPs were observed in 7962 positions. In 258 genes, at least 2 loci were significantly differentially methylated; 7 genes had more than 6 significant loci, including the ANKRD11 gene, known to be associated with male infertility, with 10 DMPs. In men (n=42) with oligospermia (<15 mill/mL) vs. high sperm counts (>100 mill/mL), 4164 DMPs were observed. Differential methylation was detected in 3 or more positions of 112 genes, and 10 genes showed 7 or more DMPs. Among those genes was HDAC4 which is associated with low sperm counts in knockout mice. In men (n=45) with 0% vs. >2% normal strict morphology, we observed >20,000 DMPs. At least 4 significant DMPs were found in 548 genes and 13 genes had 8 or more DMPs, among which was the gene SLC9A3 with 9 DMPs. Loss of expression of SLC9A3 in mice has been associated with abnormal sperm morphology. There appear to be emerging and robust relationships between sperm DNA methylation patterns and abnormal semen parameters among infertile couples. Many of the genes implicated by these data have known roles in male infertility in both humans and mice. In addition to its ability to correlate with natural conception rates and IVF embryo quality, we expect that sperm epigenetic profiles will be found to correlate closely to abnormal routine semen parameters.