The relationship between sperm nuclear DNA fragmentation, mitochondrial DNA fragmentation, and copy number in normal and abnormal human ejaculates

While it is common to clinically evaluate sperm nuclear DNA fragmentation, less attention has been given to sperm mitochondrial DNA. Recently, a digital PCR assay has allowed accurate estimation of the proportion of fragmented mtDNA molecules and relative copy number. To determine the correlation of classical sperm parameters, average mtDNA copies per spermatozoon and the level of mtDNA fragmentation (SDF-mtDNA) to that of nuclear DNA fragmentation (SDF-nDNA), measured as the proportion of global, single-strand DNA (SDF-SSBs) and double-strand DNA breaks (SDF-DSBs). To determine whether the level of nuclear and mitochondrial DNA fragmentation and/or copy number can differentiate normozoospermic from non-normozoospermic samples. Ejaculates from 29 normozoospermic and 43 non-normozoospermic were evaluated. SDF was determined using the sperm chromatin dispersion assay. mtDNA copy number and SDF-mtDNA were analyzed using digital PCR assays. Relative mtDNA copy increased as sperm concentration or motility decreased, or abnormal morphology increased. Unlike SDF-mtDNA, mtDNA copy number was not correlated with SDF-nDNA. SDF-mtDNA increased as the concentration or proportion of non-vital sperm increased; the higher the mtDNA copy number, the lower the level of fragmentation. Non-normozoospermic samples showed double the level of SDF-nDNA compared to normozoospermic (median 25.00 vs. 13.67). mtDNA copy number per spermatozoon was 3× higher in non-normozoospermic ejaculates (median 16.06 vs. 4.99). Although logistic regression revealed SDF-Global and mtDNA copy number as independent risk factors for non-normozoospermia, when SDF-Global and mtDNA copy number were combined, ROC curve analysis resulted in an even stronger discriminatory ability for predicting the probability of non-normozoospermia (AUC = 0.85, 95% CI 0.76-0.94, p