Optimized combination of multiple biomarkers to improve diagnostic accuracy in male fertility

Artificial insemination is the general method of breeding for genetic improvement in offspring. However, almost half of the insemination cases fail to achieve full-term pregnancy, due to male infertility or subfertility. To maximize the success of insemination, accurate semen quality testing is required prior to insemination. Even though basic semen analyses have been used to provide preliminary information, it cannot fully identify the superior or inferior fertility bulls. Therefore, more powerful and easy to use methods for the prediction of male fertility are required, such as proteomic or microarray chips. During past decades, omics approaches have been developed and suggested the numerous fertility-related potential biomarkers. Our previous study identified the fertility related protein markers, enolase1 (ENO1), ATP synthase, H⁺ transporting, mitochondrial F1 complex, beta subunit (ATP5B), voltage-dependent anion channel 2 (VDAC2), phospholipid hydroperoxide glutathione peroxide (GPx4), and ubiquinol-cytochrome-c reductase complex core protein 2 (UQCRC2) in bovine spermatozoa. In the present study, we perform a marker combination assay using the western blot data of ENO1, ATP5B, VDAC2, GPx4, and UQCRC2 from 20 individual bull semen samples. And then, we identified the predictive ability of these markers for normal (non-return rate (NRR) ≥ 70%) and normal fertility (NRR<70%) in bulls. ENO1, a single protein marker, achieved an area under the curve (AUC) of 0.86 and 90% discriminatory power between normal and below-normal fertility bulls, with 90% sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Although no meaningful changes existed in overall accuracy (70-85%) to discriminate the normal and below-normal fertility between ENO1 single marker and combined marker panels, multiple marker combination methods using ENO1, VDAC2, GPx4, and UQCRC2 provided absolute sensitivity and NPV, with higher specificity (70%) and PPV (77%). ENO1 can be used as a fertility marker candidate, but there were limitations for providing absolute information about normal and below-normal fertility. Although the combined use of fertility-related markers cannot provide absolute accuracy, it can help in indicating below-normal fertility in bulls. These results may contribute to the maintenance cost in the animal industry, via selection of bulls with inferior fertility.