Although the male gamete participates in a significant proportion of infertility cases, there are currently no proven molecular markers of sperm quality. The search for significant gene expression markers is partially hindered by the lack of a recognized set of reference genes (RGs) to normalize reverse transcription quantitative PCR (RT-qPCR) data across studies. The aim of this study is to define a set of RGs in assisted reproduction patients undergoing different sample collection and RNA isolation methods. Twenty-two normozoospermic men were included in the study. From each man, semen was either cryopreserved by slow freezing or analyzed fresh, and, for each, RNA was extracted with either phenol-free or phenol-based methods. In two cases, both methods were used to isolate RNA. Twenty putative RGs were analyzed and their mRNA abundance across samples was estimated by RT-qPCR. To determine the genes whose steady-state mRNA abundance remains unchanged, three different algorithms (geNorm, BestKeeper and NormFinder) were applied to the qPCR data. We found that RGs such as GAPDH or ACTB, useful in other biological contexts, cannot be used as reference for human spermatozoa. It is possible to compare gene expression from fresh and cryopreserved sperm samples using the same isolation method, while the mRNA abundance of expressed genes becomes different depending on the RNA isolation technique employed. In our conditions, the most appropriate RGs for RT-qPCR analysis were RPLP1, RPL13A, and RPLP2. Published discrepancies in gene expression studies in human spermatozoa may be due in part to inappropriate RGs selection, suggesting a possible different interpretation of PCR data in several reports, which were normalized using unstable RGs.
© 2015 American Society of Andrology and European Academy of Andrology.