56 Biomarker-Based High Throughput Sperm Phenotyping: the Good, the Bad and the Ugly

Abstract

Abstract Research in our laboratory addresses the unmet need for developing high throughput phenotyping for collecting comprehensive phenotypes of production, reproduction and fitness traits, a major emphasis area of the USDA Blueprint for Animal Genome Research 2022-2027. Our short- and long-term goal within this emphasis area is to identify specific genomic determinants of aberrant sperm quality and integrity that impart unique morphometric and optical properties on functionally defective spermatozoa that can be captured in next generation flow cytometry at a high speed/throughput, and without the need for costly, time-consuming sample processing. We identified a large set of deleterious gene variants associated with specific sperm defects in the genomes of 85 genetically valuable bulls with suboptimal artificial insemination (AI) fertility. Ongoing efforts focus on developing a second generation, label-free sperm phenotyping pipeline, eliminating variance and costs associated with the conventional, probe-based approaches, based on the specific genomic determinants of aberrant sperm quality resulting in unique morphometric and optical properties of spermatozoa, captured by the next-generation, brightfield image-based flow cytometry (IBFC) and translatable to high-precision phenotyping of heritable production traits. First step is to phenotype and validate rare deleterious polymorphisms associated with aberrant sperm head and tail morphologies in aforementioned cohort of 85 genome-sequenced bulls grouped based on prevailing sperm defect phenotype. Focus is on sperm defects predicted to be heritable and non-compensable by increasing sperm number per AI dose, such as knobbed acrosomes (sperm head caps), nuclear vacuoles and aplastic defects of sperm mitochondrial sheath (the major source of energy to fuel sperm movement). Other potentially heritable defects of interest may not be manifested by aberrant sperm morphologies, and only detectable at subcellular/molecular level by applying biomarkers validated by past and ongoing genome-to-phenotype studies. In the second phase, a label free sperm phenotyping pipeline for bull fertility testing will be developed, translatable to somatic cell phenotyping for other heritable traits by using the dedicated FeatureFinder IFBC software integrated with our machine learning pipeline, followed by validation assessing sperm structural integrity, “gentle” live cell sorting and metabolomics. This approach literary puts artificial intelligence in artificial insemination (AI in AI approach). Altogether, such genome-to-phenome efforts align the livestock genomic selection with state-of-the-art animal andrology, veterinary diagnostics and precision breeding, supporting the improvement of livestock industry efficiency through global approaches integrating production, reproduction, nutrition, health, and welfare.