AOAC 133rd Annual Meeting - Final Program

TUESDAY, SEPTEMBER 10, 2019 SYMPOSIUM: Applying Non-Target Data Acquisition for Target Analysis (nDATA) of Organic Contaminants and Biomarkers in Environmental and Food Samples 8:15 AM – 9:45 AM Grand Ballroom 1 Chairs: Paul Yang, Calibration and Validation Group Mark Sumarah, Agriculture and Agri-Food Canada 8:15 AM nDATA (non-target Data Acquisition Target Analysis) Workflow for Multiresidue Pesticide Screening by Ultra-High-Performance Liquid Chromatography-Quadrupole Orbitrap Mass Spectrometry with a Compound Database Jon Wong, U.S. Food and Drug Administration, Jian Wang, Willis Chow, Canadian Food Inspection Agency (CFIA), Roland Carlson, California Department of Food and Agriculture, James Chang, Thermo Fisher Scientific An nDATA workflow using UHPLC/ESI Q-Orbitrap full MS-data independent acquisition (DIA) and a compound database was developed to screen pesticides in fresh produce. MS/MS spec- tra of samples were acquired using variable and multiplex DIA (vDIA, mDIA) and identification of pesticides in samples was achieved using retention time (±0.5 min) and mass accuracy (±5 ppm) of the precursor ion (RTP by full MS) and retention time (±0.5 min) and mass accuracy (±10 ppm) of fragment ions (RTFI by DIA). This work was validated using mDIA and vDIA analysis of QuEChERS prepared produce samples spiked with pesticides. Of the 845 pesticides studied, RTP identified up to 765 and 796 pesticides whereas RTFI identified up to 729 and 764 pesti- cides at fortification levels of 10 and 100 μg /kg, respectively. To test for nDATA workflow transferability, a study involving a 50-pesticide compound database and 10 produce samples was evaluated in more than 20 laboratories. Laboratories that have already submitted results successfully completed the qualification and validation phases and were able to identify all pesticides in the proficiency phase of the study. The database continues to be expanded (currently >1000 chemicals) and provides a method with potential future use in comprehensive chemical screening. 8:40 AM Improving Selectivity in Untargeted Contaminant Analysis with Scanning Quadrupole Data Independent Acquisition (sqDIA) William Cooke, Greg Mercer, U.S. Food and Drug Administration A challenge adopting high-resolution mass spectrometry (HRMS) methods for routine contaminant analysis is data review and interpretation. Most data independent acquisition (DIA) methods generate precursor and product ions that can only be associated

by retention time. These methods produce vast amounts of data that can be difficult to review efficiently in a high-throughput regulatory setting. Identification of analytes can be particularly challenging in food samples with large amounts of co-eluting matrix peaks. To address this an alternative acquisition method, scanning quadrupole data independent acquisition (sqDIA), was used on an LC-QToF. In sqDIA, a fixed mass window (i.e., 20 m/z ) is applied to the quadrupole and continuously scanned over the mass range of interest, (e.g., 100-600 m/z ). Ions exiting the quadrupole are assigned a “Quad Time.” Ion mobility spectrometry data processing algorithms filter ions with matching Quad Times, producing pronounced spectral clean-up. Characteristic product ion spectra can be produced in the presence of co-eluting compounds with sufficiently different precursor masses (~8 m/z with 20 m/z quad window). Cleaner spectra simplify data review for known analytes and streamlines unknown compound identification. Here we describe sqDIA, its application to residue level pesticide analysis in food, benefits to unknown identification, and future work. 9:00 AM Untargeted Mass Spectrometry Data Acquisition: An Emerging Approach to Detect Gelatin Adulteration Based on Targeted Proteotypic Peptides Francis Beaudry, University of Montreal Following the internationalization of food production and manu- facturing, a significant increase of food fraud was discovered, ranging from false label claims to the use of additives and fillers to increase profitability. The accidental or fraudulent mixing of animal products or by-products from different animal species is an important preoccupation for consumers with health or ethical concerns. Gelatin is widely used during food processing as well as in cosmetics, nutraceutics, and medical formulations. It contains mainly type I, II, and III collagen polypeptides. We present a detailed method to perform targeted analysis using two common untargeted data acquisition strategies. Collagen proteins were methodically analyzed in silico to generate tryptic peptide mass lists and specific fragments. We will be comparing data acquired in data-independent acquisition (DIA) mode using 12 DIA isolation windows with an equal width for MS/MS scans to top-10 data-dependent acquisition (DDA) mode to auto- matically trigger series of MS/MS spectra from the information

12 SEPTEMBER 6–12, 2019 SHERATON DENVER DOWNTOWN HOTEL