Assessment of seminal cell-free DNA as a potential contaminate in studies of human sperm DNA methylation.

Abstract

Human seminal cell-free deoxyribonucleic acid (cfDNA) methylation patterns have not yet been thoroughly explored; however, recent work in mouse has suggested that some cfDNA encountered in the epididymis may contaminate DNA methylation studies assessing the mature spermatozoa. Such contamination could clearly prove to be a significant confounder, for many reasons, in epigenetic studies of male factor infertility. To explore the nature of seminal cfDNA methylation and the likelihood that it would be retained following standard semen sample processing for epigenetic analysis. We assessed 12 semen samples collected at Utah Fertility Center. For each sample, seminal cfDNA was isolated from the sperm pellet. The spermatozoa was split into three aliquots including one exposed to DNase to remove any additional cfDNA (termed "pure spermatozoa"), one not exposed to DNase, and one exposed to DNase but reintroduced to seminal cfDNA. We additionally assessed blood DNA as our benchmark for somatic cell DNA methylation patterns. DNA methylation was measured via Illumina's 850k array and assessed for differential regional methylation. Forty-six thousand three hundred fifty-two differentially methylated regions (FDR>40) were identified between pure spermatozoa and seminal cfDNA. We found at these sites that the average DNA methylation in cfDNA always fell somewhere between the average methylation in spermatozoa and blood. We also assessed each sperm treatment groups at all 46,352 regions of interest and found no significant differences at any of these sites. Our data suggest that seminal cfDNA is a clear mixture of both somatic and germline DNA and that cfDNA is not a contaminating feature in sperm DNA methylation studies following standard protocols in human sperm DNA extraction.