Use of the rumen simulation technique (RUSITEC) to compare microbial digestion of various by-products

Abstract

A semi-continuous culture system, adapted from the rumen simulation technique (RUSITEC) was used to compare the rumen micro-organism fermentation of different by-products. The following by-products were studied: (1) concentrates: citrus pulp; beet pulp; corn gluten feed; wheat bran; (2) roughages: untreated straw samples; straw samples treated with either ammonia or sodium hydroxide. The latter were also compared to alfalfa hay. The undigested feed residues from the bag were analysed for organic matter and cell-wall constituents (van Soest detergent procedure); liquid effluent was analysed for volatile fatty acids (VFA) and gas samples for CH 4, CO 2 and H 2. The substrates, representing a wide range of fermentability, fell into 4 groups with decreasing values according to OM digestibility and total daily VFA production: pulp by-products (80%, 93 mmoles), cereal by-products (68%, 69 mmoles), NaOH-treated straw and hay (55%, 60 mmoles) and untreated and NH 3-treated straw (34%, 40 mmoles). However, cellulose digestibility showed intra-group variations; the largest difference observed was between corn gluten feed and wheat bran (40.1 and 15.6%, respectively). In comparison with the other groups, cereal by-products showed a particular VFA pattern with a high molar proportion of propionate (32 vs. 20–27%) and a low proportion of acetate (51 vs. 58–74%), probably because of their residual starch content (16.7% DM in bran). Methane production relative to hexose fermented (mol mol −1) was higher for roughage (0.45–0.58) than for concentrate by-products (0.31–0.35). It was concluded that the RUSITEC technique offers a convenient means of comparing the extent of microbial degradation and the fermentation pattern of various by-products but that, in the future, biomass synthesis should also be measured.