All in good time : dynamics of the bovine estrous cycle investigated with a mathematical model

Bovine fertility is subject of extensive research in animal sciences, especially since a decline in dairy cow fertility has been observed during the last decades. One factor is reduced expression of estrous behavior. Fertility is a complex process, regulated by interactions between brain and reproductive organs. The objective of this thesis was to improve insight in the regulation of dairy cow fertility by developing and using a mechanistic mathematical model of the bovine estrous cycle. The model that was developed describes the dynamics of the bovine estrous cycle on individual cow level. It simulates follicle and CL development and the periodic changes in hormone levels that control these processes by a set of linked differential equations. The model captures a number of key physiological processes of the bovine estrous cycle, and serves as a starting point for further simulation studies, model validation, and extended models. The model was used to find candidate mechanisms that regulate follicular development. A normal estrous cycle contains 2 or 3 waves of follicular development, but why some cycles consist of 3 and others of 2 waves is unknown. Results showed that variation of (combinations of) model parameters regulating follicle growth rate or time point of CL regression can change the model output from 3 to 2 waves of follicular growth in a cycle. Several factors may perturb the regular oscillatory behavior of a normal estrous cycle. Such perturbations are likely the effect of simultaneous changes in multiple parameters. It was investigated how multiple parameter perturbation changes the behavior of the estrous cycle model, so as to identify biological mechanisms that could play a role in the development of cystic ovaries, a common reason for reproductive failure in dairy cows. Simulation results indicated that CL functioning, luteolytic signals, and GnRH synthesis are likely involved in the development of cystic ovaries. Empirical data of individual cows was used to identify mechanisms that explain individual differences in cycle characteristics by fitting the model to the data. Finding specific parameter configurations for individual cows shows the capability of the model to simulate ‘real’ data. Certain combinations of estimated parameter values induced a clear qualitative shift in model behavior (e.g. a different number of follicular waves), suggesting possible routes how environmental or genetic influences could affect estrous cycle characteristics. Experimental data to verify simulation results are not always available, but hypotheses based on the model predictions could be investigated in future animal experimen