Effect of microbial isolates on microbial yield estimation in RUSITEC system

Abstract

The estimation of rumen microbial protein synthesis is one of the main points in the nitrogen (N)-rationing systems for ruminants, as microbial protein provides proportionately 0.4 to 0.9 of amino acids entering the small intestine in ruminants receiving conventional diets (Russell et al., 1992). Methods of estimating microbial protein synthesis rely on marker techniques in which a particular microbial constituent is related to the microbial N content. Marker : N values have generally been established in mixed bacteria isolated from the liquid fraction of rumen digesta and it has been assumed that the same relationship holds in the total population leaving the rumen (Merry and McAllan, 1983). However, several studies have demonstrated differences in composition between solid-associated (SAB) and fluid-associated bacteria in vivo (Legay-Carmier and Bauchart, 1989) and in vitro (Molina Alcaide et al, 1996), as well in marker : N values (Pérez et al., 1996). This problem could be more pronounced in the in vitro semi-continuous culture system RUSITEC, in which there are three well defined components (a free liquid phase, a liquid phase associated with the solid phase and a solid phase), each one having associated microbial populations.The objective of this experiment was to investigate the effect of using different bacterial isolates (BI) on the estimation of microbial production of four different diets in RUSITEC (Czerkawski and Breckenridge, 1977), using (15NH4)2 SO4 as microbial marker, and to assess what effects any differences would have on the comparison of microbial protein synthesis between diets.This study was conducted in conjunction with an in vitro experiment described by Carro and Miller (1997). Two 14-day incubation trials were carried out with the rumen simulation technique RUSITEC (Czerkawski and Breckenridge, 1977). The general incubation procedure was the one described by Czerkawski and Breckenridge (1977) and more details about the procedures of this experiment are given elsewhere (Carro and Miller, 1997).