The use of enzymes to predict the digestibility of animal feeds

Publisher Website
Google Scholar
Abstract
Six techniques were investigated to find a quick, reliable laboratory method for the prediction of the in‐vivo digestibility of animal feeds. Four of these techniques comprised two digestions, one with pepsin in hydrochloric acid followed by another with cellulase. In three cases the cellulase was derived from Trichoderma viride which was used at two concentrations; in addition, at one of these concentrations a step using amyloglucosidase was introduced. In another technique cellulase derived from Aspergillus niger was tested to determine how effectively it dealt with starchy materials. The fifth technique comprised the separation of the cell walls by digestion with neutral detergent solution followed by digestion with cellulase (derived from T. viride) with an additional step using amyloglucosidase for samples containing starch. The last method was an in‐vitro technique using buffered rumen liquor followed by pepsin in hydrochloric acid. Because this technique proved to estimate digestibility more accurately than any of the simple chemical methods it provided a guide to the efficiency of the enzyme techniques. The digestible organic matter in the dry matter (DOMD) was determined in each case and correlated against the DOMD value obtained using an in‐vivo technique. Two of the methods showed as good a prediction of DOMD as the in‐vitro method using rumen liquor, and the correlation coefficients were comparable. The repeatability of the methods calculated as the coefficient of variation varied between 2.5 and 3.7%. For roughages containing little or no starch the shorter techniques provided an adequate evaluation of DOMD. Starches in the cereals were adequately broken down only when the additional digestion using an amyloglucosidase (Agidex) was included; this is illustrated by the correlation values obtained.