Track: Manufacturing and Analytical Characterization - Chemical - Analytical - Impurities and Degradation - Other
Category: Late Breaking Poster Abstract
Mycotoxin Screening in Nicotine-Containing Products with LC-MS/MS
Purpose: Mycotoxins are naturally produced by certain types of fungi and can be hazardous to humans and animals, even at low concentrations. Mycotoxin contamination is most commonly associated with food products, such as grain. However, the same fungi can infect tobacco and the presence of mycotoxins in tobacco products should be assessed to reduce the risk to the consumer. To facilitate the quantitation of mycotoxins in nicotine-containing products, we developed and validated LC-MS/MS methods for 21 different mycotoxins in E-Liquid and other smokeless tobacco products. Methods: Mycotoxin standards were prepared in methanol and analyzed with HPLC-MS/MS using an AB Sciex API365 triple quadrupole mass spectrometer and an Agilent Zorbax SB-C18 column. Mobile phases A and B were 0.5% acetic acid and 5 mM ammonium acetate in DI water and in HPLC-grade methanol, respectively. Extracts of nicotine-containing E-Liquid, pouch, tablet, gum, and lozenge products were prepared in methanol and were filtered prior to analysis. Results: To analyze all 21 mycotoxins with a high degree of accuracy, precision, and linearity, three different LC-MS/MS methods were developed. One method used negative ionization to measure 3-acetyldeoxynivalenol (3-AcDON), alternariol (ALT), deoxynivalenol (DON), fusarenon-X (FUSX), nivalenol (NIV), zearalanol (ZEA; as sum of α-zearalanol and β-zearalanol, and zearalenone (ZON). Another method used positive ionization to measure aflatoxin B1 (AFB1), aflatoxin B2 (AFB2), aflatoxin G1 (AFG1), aflatoxin G2 (AFG2), diacetoxyscirpenol (DAS), HT-2 toxin (HT2), neosolaniol (NEO), ochratoxin A (OTA), and T-2 toxin (T2), as well as the antifungal treatment thiabendazole (THIA). A separate method was developed for fumonisins B1-B3 (FB1-3), also using positive ionization. All three methods used the same mobile phase and column, enabling the sample analysis in a single run of 28 minutes per sample. The three methods were then validated for accuracy, precision, linearity, specificity, limit of detection (LOD), limit of quantitation (LOQ), robustness, and specificity, according to ICH guidelines. All acceptance criteria were met in the validations. Method LODs were analyte-dependent. For OTA, THIA, ZEA, and ZON, the LOD was established at 1 ng/mL and the linear range was 2-100 ng/mL. AFB1 and AFB2 had an LOD at 5 ng/mL and a range of 10-500 ng/mL, whereas AFG1 and AFG2, T2, ALT, and DON, had an LOD of 10 ng/mL and range of 20-1000 ng/mL. The LOD was 25 ng/mL and the range was 50-2500 ng/mL for 3-AcDON, FUSX, NIV, DAS, HT2, and NEO. FB1-3 had an LOD of 100 ng/mL and a range of 200-5000 ng/mL. All of these LODs were below the limits established by regulatory authorities. Quantitation of mycotoxins in methanolic extracts of nicotine-containing products was verified via generating standard addition curves for each mycotoxin at four different spike levels, including a control at spike level 0 ng/mL. The standard addition curves all showed a linear correlation coefficient of 1.00 and good precision, with a method noise of less than 20 for each analyte. Conclusion: LC-MS/MS methods for over 20 mycotoxins were validated and verified in five different smokeless tobacco products, enabling the screening of a wide variety of mycotoxins without extensive sample preparation.