FOOD BIOLOGICAL CONTAMINANTS Validation of an Immunoassay for Detection and Quantitation of a Genetically Modified Soybean in Food and Food Fractions Using Reference Materials: Interlaboratory Study LIPP ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 83, NO. 4, 2000 MARKUS LIPP and ELKE ANKLAM 1 European Commission, Joint Research Center, Institute for Health and Consumer Protection, Food Products and Consumer Goods Unit, T.P. 260, I-21020 Ispra (Va), Italy JAMES W. STAVE Strategic Diagnostics, Inc., 111 Pencader Dr, Newark, DE 19702-3322 Collaborators: J. Bahrs-Windsberger, M.T. Barreto Crespo, E. Benvenuto, G. Berben, P. Brignon, U. Busch, C. Cordes, E. Eklund, M. El Jaziri, K.H. Engel, F. Eriksen, S. Eriksson, A. Felinger, C. Fogher, I. Folch, S. Garrett, C. Hischenhuber, H. H`rtner, K.D. Jany, R. Krska, L. Kruse, H.A. Kuiper, C. Lobre, M. Miraglia, W. Mhller, J.M. Nunes Costa, L. Oliveira, P. Patel, K. Pietsch, B. P`pping, P. Remler, J. Rentsch, M. Schulze, M. Sutton, G. van Duijn, F. van Kan, A. Wurz An immunoassay for detection of a specific geneti- cally modified soybean (Roundup-Ready ® ) was val- idated on dried soybean powder in an interlaboratory study. Different percentages of ge- netically modified soybeans in nonmodified soy- bean matrix were evaluated in a blind study. Thirty-eight laboratories from 13 countries partici- pated. The immunoassay was evaluated for 2 end- points: (1) To give a semiquantitative result, i.e., determination of a given sample above or below a given threshold, or (2) to compute a quantitative result, i.e., percentage of genetically modified soy- beans in the sample. Semiquantitative results showed that a given sample which contained <2% genetically modified soybeans was identified as below 2% with a 99% confidence level. Quantita- tive use of the assay resulted in a repeatability (r) and reproducibility (R) that were computed to be RSD r =7% and RSD R = 10%, respectively, for a sample containing 2% genetically modified soy- beans. Application of this method depends on availability of appropriate reference materials for a specific food matrix. Only matrix-matched refer- ence materials can be used for analysis of food or food fractions. T he requirements for labeling of food and food products derived from genetically modified organisms (GMO) within the European Union is set out by the Regulation 258/97/EEC on Novel Foods and Novel Food Ingredients (1). Because 2 GMO products (Roundup Ready ® soybean and BT176 corn) were on the market before the Novel Foods Reg- ulation came into force, their labeling requirements are dealt with separately by Regulation 1139/98/EEC (2). To ensure that consumers are fully informed of the con- tent of the food they purchase, adequate information is pro- vided on product labels. Labeling is mandatory if GMO products are no longer substantially equivalent to their con- ventional counterpart, e.g., if a new protein and/or the ge- netic modification of DNA can be detected. Analytical meth- ods for detection of both newly expressed protein or inserted DNA for GMOs are necessary to determine compliance with labeling requirements. Such methods should be validated at the European Union level to encourage application of har- monized procedures. DNA-based methods for specific de- tection of the economically most important GMOs are al- ready available, e.g., (3–7). Recently, a DNA-based screening method based on de- tection of 2 genetic elements, the 35S promoter and the nos terminator, by means of polymerase chain reaction (PCR) was published (8) and validated on a European scale (9). These 2 genetic elements are important for expression of genes that are present in nearly all genetically modified plants to-date (10, 11). The method seems well suited to screening for the presence of GMOs. However, it does not identify specific GMOs in the sample because these ele- ments, the 35S promoter and the nos terminator, are present in numerous GMOs. No internationally validated method based on analysis of proteins is currently available. This study describes a valida- tion study of a specific protein-based method for detecting RoundupReady soybeans in food-related matrixes. LIPP ET AL.: JOURNAL OF AOAC INTERNATIONAL VOL. 83, NO. 4, 2000 919 Received July 17, 1999. Accepted by AH January 18, 2000. 1 Author to whom correspondence should be addressed. Downloaded from https://academic.oup.com/jaoac/article/83/4/919/5656367 by guest on 01 March 2023