Submitted to: Journal of Agricultural and Food Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/25/2015
Publication Date: 8/25/2015
Citation: Geng, P., Harnly, J.M., Chen, P., Luthria, D.L., Zhang, M. 2015. Differentiation of whole grain and refined wheat (T. aestivum) flour using a fuzzy mass spectrometric fingerprinting and chemometric approaches. Journal of Agricultural and Food Chemistry. 407:7875-7888. Interpretive Summary: Whole grains have becoming popular due to its numerous health-benefiting effects compared to their refined counterparts. The Whole Grains Council has created an official packaging symbol called the “Whole Grain Stamp” that helps consumers find real whole grain products. However, an fast and easy method to diffrentiate whole grains from refined grains is lacking. Using wheat as model, this study developed a quick and simple method to classify germ, bran, whole grain wheat flour and refined wheat flour. The method also dtected the major characteristic chemical components responsible for the classification.
Technical Abstract: A fuzzy mass spectrometric (MS) fingerprinting method combined with chemometric analysis was established to provide rapid discrimination between whole grain and refined wheat flour. Twenty one samples, including thirteen samples from three cultivars and eight from local grocery store, were studied. The three milling fractions (bran, germ, and endosperm) could be well distinguished. Germ was found to be the most phytochemical rich fraction. Sixteen biomaker components were tentatively identified. Most of the biomarkers revealed in this study, such as di-hexoside, tri-hexoside, apigenin glycosides, and unsaturated fatty acids, showed higher abundance in germ than in bran/endosperm. This characteristic can be used for the differentiation of refined flour (RF) and whole wheat flour (WWF). The state of refinement (whole vs. refined) of wheat flour was differentiated successfully despite the potential confounding introduced by wheat class (red vs. white; hard vs. soft) based on the principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA). Samples of 100% RF and 100% WWF from 3 cultivars (Hard Red, Hard White, and soft white) were physically mixed individually to provide 20, 40, 60, and 80% WWF of each cultivar. SIMCA was able to discriminate between 100%, 80%, 60%, 40%, 20% WWF and 100% RF.