Modelling volatile fatty acid dynamics and rumen function in lactating cows

Research output: Thesisinternal PhD, WU

Abstract

Mathematical models are developed to quantity and integrate the various processes involved with rumen fermentation. Three extant mechanistic models of rumen fermentation were studied (Baldwin et al., Danfær and Dijkstra et al.), each with a truly dynamic representation but different conceptual approach. The models were compared on mathematical representation of individual processes and their prediction accuracy was evaluated. Although the models predicted similar rates of substrate degradation and rumen outflow of organic matter, total crude protein and microbial protein, they differed substantially in representation of the underlying microbial mechanisms. The model of Baldwin et al. performed best in prediction of the combination of rumen pool sizes and duodenal flows, whereas the model of Dijkstra et al. was evaluated to deliver the most realistic outflow of rapidly fermentable carbohydrates. Further, it was identified that all models needed improvement with respect to the prediction of amounts and type of volatile fatty acids (VFA) produced. In a subsequent evaluation is was investigated to what extent individual model elements, of a selection of five, could be responsible for inaccurate VFA predictions. The results suggested that inaccuracy of stoichiometric coefficients of VFA yield from fermented substrate (VFA coefficients) and incorrect representations of VFA absorption kinetics are the most likely causes. New values of VFA coefficients were derived by regression of a stoichiometric model of VFA yield against data of VFA molar proportions observed in vivo in rumen fluid of lactating cows. Inputs to the model were observed rates of rumen substrate degradation. Regression against simulated data sets including random error indicated that the accuracy of this method to estimate VFA coefficients is acceptable. Estimates from regressions against in vivo data delivered new sets of VFA coefficients for roughage-rich and concentrate-rich diets. In a follow-up study the representation of stoichiometry was made pH-dependent. With regression of this model against in vivo data a profound effect of rumen pH on the type of VFA formed from rapidly fermentable carbohydrates was established. Besides VFA production, the rumen concentrations and the amount and profile of VFA available for the cow are also affected by absorption and metabolism of VFA by epithelial tissues in the rumen wall. A mechanistic model was constructed that represents the dynamics of these processes, including the effects of changes in VFA concentration differences between different compartments, the effect of competitive inhibition between VFA and the effect of changes in surface area and epithelial mass. Although some essential characteristics of VFA transport and intra-epithelial metabolism could be reproduced by the model, it was concluded that there is a definite need for more experimental data. It is concluded that various levels of functioning need to be included when representing whole rumen function. Besides intrinsic degadation characteristics and passage of ingested substrates, environmental conditions in the rumen and the functionality of the rumen wall need to addressed.  
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Wageningen University
Supervisors/Advisors
  • Tamminga, Seerp, Promotor
  • France, J., Promotor, External person
  • Dijkstra, Jan, Co-promotor
Award date26 Nov 2007
Place of Publication[S.l.]
Print ISBNs9789085047858
Publication statusPublished - 2007

Keywords

  • cows
  • lactation
  • volatile fatty acids
  • dynamic models
  • rumen fermentation
  • absorption
  • rumen epithelium
  • nutrition physiology
  • cell metabolism

Fingerprint Dive into the research topics of 'Modelling volatile fatty acid dynamics and rumen function in lactating cows'. Together they form a unique fingerprint.

Cite this