Abstract
Eight laboratories participated in a collaborative study to evaluate an enzyme-linked immunosorbent assay (ELISA) to determine soy, pea, and wheat proteins in pasteurized or ultra-high temperature (UHT) milk powders. To perform this assay, polyclonal antibodies for soy, pea, and wheat proteins were obtained from rabbit sera. Collaborators received calibration standards composed of milk powder containing 0–8% (w/w) vegetal protein in total protein and blind test samples containing approximately 1, 2, and 5% (w/w) vegetal protein. An indirect competitive ELISA was performed with a kit prepared by a participating laboratory; the kit contained plates coated with soy, pea, or wheat proteins, the corresponding specific antisera, enzyme-labeled second antibody, and substrate solution. Test samples and calibrants were extracted with phosphate-buffered saline, pH 7.4, containing 0.05% Tween and assayed with the ELISA kits. The degree of adulteration was affected by the type of heat treatment applied to the samples. The estimated percentage of vegetal protein addition was close to the theoretical value for pasteurized samples but much lower for UHT samples. For pasteurized samples, intralaboratory relative standard deviations ranged from 5 to 22% and interlaboratory relative standard deviations ranged from 14 to 34%.
Use of Enzyme Linked Immunosorbent Assay for detection of aflatoxin M1 in milk powder