NON STARCH POLYSACCHARIDE HYDROLYZING ENZYMES AS FEED ADDITIVES - DETECTION OF ENZYME-ACTIVITIES AND PROBLEMS ENCOUNTERED WITH QUANTITATIVE-DETERMINATION IN COMPLEX SAMPLES

Chromogenic substrates, an agar diffusion assay and viscosity reductio n were used to estimate D-glucanase and xylanase activities in water s oluble extracts of different feedstuffs and digesta supernatants. The dinitrosalicylic acid reducing sugar method was employed to calibrate results from different methods based on international units (IU, gluco se equivalents). The detection of dye release from chromogenic substra tes was a suitable method, allowing the detection of 0.05 IU of enzyme activity per mi of extract, although measurements in digesta supernat ants were limited in linearity (0.1-0.5 IU/ml supernatant). With the a gar diffusion assay the detection of enzyme activity was possible over a wider concentration range (extracts: 0.05-1 IU/ml, digesta supernat ants: 0.1-1 IU/ml), but visual evaluation led to inaccurate measuremen t. Accuracy can be improved by computer based evaluation of digital im ages. The use of viscosity reduction produced linear standard curves f rom 0.01 to 0.5 IU/ml in feed extracts, but reliability of measurement s depended on modification of substrates. Quantification of enzyme act ivities was influenced by matrix effects of complex samples. Cereal de pendant differences were found in various extracts of feed mixtures an d cereal extracts. Digesta supernatants partly inhibited enzyme activi ty, depending on the ori gin of the sample. Interaction of substrates with digesta components varied between methods. The sensitivity of the methods is comparable, however, all methods require specific calibrat ions to account for matrix-and enzyme specific effects.