Assessment of the relationship of number of sertoli cells with number of germ cells, spermatogenic activity and germ cell apoptosis rates in bullsusing different regression models

Abstract

This study aimed to model the quantities of germ cells, Leydig cells, and spermatogenic activity in relation to Sertoli cell numbers in bulls. Histological and histometric methods were used to process left testes of 36 bulls. Prediction models were constructed using both linear and non-linear regression. Standard error (SE), coefficient of correlation (r), R-squared (R²), mean squared error (MSE) and root mean squared error (RMSE) statistics were used to identify the optimal model. Statistical analysis revealed that, except for Leydig cell counts, Sertoli cell numbers may predict the overall composition of testicular cells. Spermatogenic activity and Sertoli cell count are most accurately correlated using nonlinear models. Total Sertoli cell numbers contributed to modest variance in germ cell apoptosis rates during meiosis (13.6%) and post-meiotic germ cell loss rates (7.90%), but these associations were not statistically significant. Cubic models were the best fit for describing the relationship between testicular morphometric parameters and Sertoli cells numbers (X). However, the diameter of seminiferous tubules (Y) was more accurately described by a growth model, represented by the equation Y=e(5.428+(0.022*X). It can be recommended that the non-linear model using the Sertoli cells numbers as the predictive biomarkers can be used for predicting bull’s spermatogenetic activity.