Modeling microbial growth based on time-dependent kinetic mechanisms of digestion and passage in the ruminoreticulum

Abstract

The original Cornell Net Carbohydrate and Protein System (CNCPS) model, which was developed in 1992 and revised in 2004, included a rumen submodel with a set of equations that were algebraically programmed in a spreadsheet to predict bacterial N flow to the caudal tract of cattle. In this study, we propose a modification to the original CNCPS rumen submodel based on mathematical concepts that simulate the flow paths of solids and liquid in the rumen, which are known to be essential factors affecting rumen fermentation and microbial growth. This modification allows the quantification of the impact of aging mechanisms on the availability of both soluble and insoluble substrates for microbial growth in the rumen. Additionally, a correction has been proposed on peptide uptake by rumen bacteria. Literature data were used to evaluate the model adequacy of CNCPS versions 1992 and 2004, along with the proposed model, to determine their ability to predict bacterial N flow. The proposed model more accurately predicted the variable of interest, but there was an overall underprediction bias. Despite the rightful critics of the ability of mechanistic models to predict microbial crude protein flow from the rumen, the 1992 and 2004 versions of the CNCPS and the proposed modifications implemented within the CNCPS framework led to predictions that agreed with observational data at a discordance rate of 0.05, tolerance probabilities <0.001 for versions 1992 and 2004, and a tolerance probability equal to 0.13 for the proposed model given a sample size $n=39$. Therefore, the proposed modifications might improve the ability of CNCPS-based models to predict the bacterial N flow from the rumen, thereby resulting in improved predictive performance compared to the original CNCPS model. Thus, it seems that models built on time-dependent kinetics of food particles and fluid are more fitting than time-independent ones to predict bacterial N flow to the caudal tract.