The most commonly used method for the measurement of the level of endo-xylanase in commercial enzyme preparations is the 3,5-dinitrosalicylic acid (DNS) reducing sugar method with birchwood xylan as substrate. It is well known that with the DNS method, much higher enzyme activity values are obtained than with the Nelson-Somogyi (NS) reducing sugar method. In this paper, we have compared the DNS and NS reducing sugar assays using a range of xylan-type substrates and accurately compared the molar response factors for xylose and a range of xylo-oligosaccharides. Purified beechwood xylan or wheat arabinoxylan is shown to be a suitable replacement for birchwood xylan which is no longer commercially available, and it is clearly demonstrated that the DNS method grossly overestimates endo-xylanase activity. Unlike the DNS assay, the NS assay gave the equivalent colour response with equimolar amounts of xylose, xylobiose, xylotriose and xylotetraose demonstrating that it accurately measures the quantity of glycosidic bonds cleaved by the endo-xylanase. The authors strongly recommend cessation of the use of the DNS assay for measurement of endo-xylanase due to the fact that the values obtained are grossly overestimated due to secondary reactions in colour development.
Keywords: 3,5-Dinitrosalicylic acid; Beechwood xylan; Nelson-Somogyi; Reducing sugars; Wheat arabinoxylan; Xylo-oligosaccharides; endo-xylanase.