Premium
Determination of type B fumonisin mycotoxins in maize and maize‐based products by liquid chromatography/tandem mass spectrometry using a QqQ linear ion trap mass spectrometer
Author(s) -
Faberi Angelo,
Foglia Patrizia,
Pastorini Elisabetta,
Samperi Roberto,
Laganà Aldo
Publication year - 2004
Publication title -
rapid communications in mass spectrometry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.528
H-Index - 136
eISSN - 1097-0231
pISSN - 0951-4198
DOI - 10.1002/rcm.1778
Subject(s) - chemistry , chromatography , mass spectrometry , selected reaction monitoring , ion trap , tandem mass spectrometry , detection limit , calibration curve , fumonisin b1 , triple quadrupole mass spectrometer , high performance liquid chromatography , analytical chemistry (journal) , mycotoxin , food science
A sensitive liquid chromatography/tandem mass spectrometry (LC/MS/MS) method for determining the type B fumonisin mycotoxins in corn‐based foodstuffs is described. Fumonisins FB 1 and FB 2 were extracted from a 1 g sample by homogenization with acetonitrile/water (75:25, v/v, 50 mmol/L formic acid, 25 mL final volume) and the extract was defatted on C 18 phase. Volumes of 5 mL of crude extracts were cleaned up on Carbograph‐4 cartridges. The final solution was analyzed by HPLC with electrospray ionization mass spectrometry in positive ion mode using multiple reaction monitoring with a QqQ linear ion trap mass spectrometer. Recoveries for spiked corn‐based foodstuffs ranged from 91–105% (RSD% ≤8%), and method detection limits were ≤2 ng/g for FB 1 and ≤1 ng/g for FB 2 . Two different spiking levels were tested (5000 and 100 ng/g for FB 1 , 1000 and 20 ng/g for FB 2 ). Quantitation was achieved by an external calibration procedure using matrix‐matched standards, with diclofenac added post‐cleanup as internal standard for the LC/MS/MS analyses. Calibration curves showed linearity in the concentration range 0.005–5 ng/μL of final extract (0.992 ≤ R 2 ≤ 0.995). Two other fumonisins, FB 3 and FB 4 , were identified in naturally contaminated samples of corn meal using an information‐dependent acquisition protocol that looped three experiments, including neutral loss scan, enhanced resolution scan, and enhanced product ion scan. FB 3 and FB 4 quantitation was estimated as peak area ratios relative to the FB 2 response in view of the lack of both standards. This work also includes an application of the present LC/MS/MS method to some maize and maize‐based product samples (corn meal, cornflakes and popcorn) collected from Italian stores. FB 1 and FB 2 contamination levels exceeding the European Union recommendation were found in 8 out of 15 corn meal samples. Copyright © 2004 John Wiley & Sons, Ltd.