Pre-screening method for somatic cell contamination in human sperm epigenetic studies.

Sperm epigenetic profiles are frequently studied and are of great interest in many fields. One major technical concern when assessing these marks is the potential for somatic cell contamination. Because somatic cells have dramatically different epigenetic signatures, even small levels of contamination can result in significant problems in analysis and interpretation of data. In this study we evaluate an assay, which we designed to offer a reliable 'pre-screen' for somatic cell contamination that directly assesses the DNA being used in the study to determine tissue purity. In brief, we designed an inexpensive and simple assay that utilizes the strong differential methylation between sperm and somatic cells at four genomic loci to assess the general purity of samples prior to performing expensive and time intensive assays. The assay is able to reliably detect contamination qualitatively by running the sample on an agarose gel, or quantitatively with the use of a bioanalyzer. With this technique we have found that we can detect potentially contaminating signals in samples of many different types, including those from patients with poor sperm phenotypes (oligozoospermia, asthenozoospermia, and teratozoospermia). We also have found that the use of multiple sites to determine potential contamination is key, as some conditions (asthenozoospermia specifically) appear at one site to reflect a somatic-like profile, while at all other sites it appears to have very typical sperm DNA methylation signatures. Taken together, the use of the assay described herein was effective at identifying contamination and could be implemented in many labs to quickly and inexpensively pre-screen samples prior to performing far more expensive and labor intensive procedures. Additionally, the principles applied to the development of this assay could be easily adapted for the development of other assays to pre-screen different tissue/cell types or model organisms.