Rapid Analysis and Authentication of Meat Using the MasSpec Pen Technology

• Main Authors: Abigail N. Gatmaitan (10280960), John Q. Lin (5622686), Jialing Zhang (1317267), Livia S. Eberlin (461381)
• Format: Other Non-Article Part of Journal/Newspaper
• Language: unknown
• Published: 2021
• Subjects: Molecular Biology; Ecology; Cancer; Inorganic Chemistry; Science Policy; Environmental Sciences not elsewhere classified; Biological Sciences not elsewhere classified; Chemical Sciences not elsewhere classified; per-sample prediction accuracies; MasSpec Pen analysis; MasSpec Pen Technology Food authent.; method; food fraud cases; per-sample accuracies; MasSpec Pen technology; grain-fed beef; type; MasSpec Pen feasibility testing; multiclass fish model; meat analysis; meat fraud; venison; test sets; Sockeye; Atlantic salmon
• Online Access: https://doi.org/10.1021/acs.jafc.0c07830.s001

Food authenticity and safety are major public concerns due to the increasing number of food fraud cases. Meat fraud is an economically motivated practice of covertly replacing one type of meat with a cheaper alternative raising health, safety, and ethical concerns for consumers. In this study, we implement the MasSpec Pen technology for rapid and direct meat analysis and authentication. The MasSpec Pen is an easy-to-use handheld device connected to a mass spectrometer that employs a solvent droplet for gentle chemical analysis of samples. Here, MasSpec Pen analysis was performed directly on several meat and fish types including grain-fed beef, grass-fed beef, venison, cod, halibut, Atlantic salmon, sockeye salmon, and steelhead trout, with a total analysis time of 15 s per sample. Statistical models developed with the Lasso method using a training set of samples yielded per-sample accuracies of 95% for the beef model, 100% for the beef versus venison model, and 84% for the multiclass fish model. Predictors of meat type selected included several molecules previously reported in the skeletal muscles of animals, including carnosine, anserine, succinic acid, xanthine, and taurine. When testing the models on independent test sets of samples, per-sample accuracies of 100% were achieved for all models, demonstrating the robustness of our method for unadulterated meat authentication. MasSpec Pen feasibility testing for classifying venison and grass-fed beef samples adulterated with grain-fed beef achieved per-sample prediction accuracies of 100% for both classifiers using test sets of samples. Altogether, the results obtained in this study provide compelling evidence that the MasSpec Pen technology is a powerful alternative analytical method for meat analysis and investigation of meat fraud.