Characterization of sperm subpopulations and their relation with functional and metabolic profiles bulls with different fertility rates using artificial intelligence methods

• Grant number: 22/15848-4
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: June 01, 2023
• End date: May 31, 2026
• Field of knowledge: Agronomical Sciences - Veterinary Medicine - Animal Reproduction
• Principal Investigator: Fernando Ferreira
• Grantee: Roberta Ferreira Leite
• Host Institution: Faculdade de Medicina Veterinária e Zootecnia (FMVZ). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

The use of sperm from subfertile bulls implies high productivity losses and, therefore, has a great impact on production sustainability. Results from a previous study (Fapesp grant #2019/20114-7) associating functional and metabolomics analysis of sperm from bulls with different fertility rates showed important differences related to energy metabolism, oxidative status and correlations with sperm kinetics. Samples from low fertility bulls showed energy metabolism deficiencies and a higher percentage of cells with oxidative stress, as well kinetic patterns that can be related to a possible early hyperactivation and capacitation. Furthermore, samples from the high fertility groups presented an efficient energy metabolism accompanied by oxidative homeostasis and kinetic patterns characterized by high linearity and velocity. Considering these results and taking into account differences observed between sperm samples, even from the same bull, this study aims to characterize sperm subpopulations based on individual sperm kinetics of each sperm batch obtained by CASA (Computer-Assisted Sperm Analysis; Hamilton-Thorne®, IVOS II, US), correlating the subpopulations with their functional and metabolic profiles of sperm samples from Angus and Nellore bulls with high and low fertility rates with data obtained in the previous study. Reviews of subpopulations studies published to date highlight some limitations for the fact that these studies didn't present a comprehensive relationship with the sperm physiological status, or with different fertility patterns and, raise questions regarding the representativeness of the number of cells used in relation to the number of cells in the semen batch. Considering these raised limitations, this study will analyze individually a high number of cells with representativeness in relation to semen batches (about 50,000 cells from each batch) and use the relation of the subpopulations with the functional and metabolic profile of each batch for a better understanding of the physiological effects of sperm subpopulations on fertility. This analysis will allow the establishment of kinetic parameters values related to sperm cell viability related to energy metabolism, oxidative status, and patterns of hyperactivation and early capacitation. Moreover, the big data subpopulations analysis and its association with the sperm physiological profile will lead to the development of an algorithm for automatic reading of subpopulations from the files obtained in the CASA system, which will be validated and adjusted to the usual readings performed by semen collection centers, with the use of a database where this algorithm will be improved through the process of machine learning.