Performance of Commercial Colorimetric Assays for Quantitation of Total Soluble Protein in Thermally Treated Milk Samples

Abstract

Commercial colorimetric protein assays have been widely used, but their performance for protein quantitation in thermally processed milk samples has not been investigated. This study evaluated four colorimetric assays for quantitation of soluble protein in nonfat milk samples that were subjected to different degrees of heat treatment. Copper ion-based Pierce™ bicinchoninic acid protein assay (BCA) and Pierce™ modified Lowry protein assay (modified Lowry) consistently registered higher protein levels than those determined by dye binding-based Pierce™ Coomassie Plus (Bradford) assay (Coomassie Plus) or Pierce™ 660-nm protein assay (660-nm). When milk samples were autoclaved or dry-heated at ≥ 121 °C for 10 min, reduction in soluble protein content was reflected in measurements of Coomassie Plus and 660-nm assays but was not indicated by BCA or modified Lowry assays. In samples dry-heated at 176 °C, while no protein was detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), Coomassie Plus or 660-nm, BCA, and modified Lowry indicated protein levels that were 215.7 and 79.4%, respectively, of those of unheated samples. Analysis of lactose samples showed that lactose may contribute to a 13.4 or 1.0% overestimation in BCA or modified Lowry measurement of protein content in unheated dry milk samples. This study demonstrated that quantitation of protein in thermally treated milk samples using copper ion-based BCA and modified Lowry assays could be significantly affected by lactose and, to a greater degree, other reducing substances formed during heating likely via the caramelization and Maillard reaction pathways. In contrast, dye binding-based Coomassie Plus and 660-nm assays were not affected and may be more accurate for measurement of protein content in thermally treated milk samples.