A direct assessment of mycotoxin biomarkers in human urine samples by liquid chromatography tandem mass spectrometry

Abstract

Detection of mycotoxin biomarkers in urine of humans and animals provides a direct approach for assessing exposure to these mycotoxins as opposed to the indirect approach of food analysis, which in most cases is affected by the heterogeneity of the toxin in the food samples. Seven (7) mycotoxins and their metabolites (total 18 analytes) were selected and an LC-MS/MS method for their determination in human urine was developed and validated. The method consisted of direct analysis of two mycotoxin conjugates, deoxynivalenol-glucuronide and zearalenone-glucuronide without beta glucuronidase digestion of the urine samples. Since high method sensitivity is of utmost importance in such study, critical factors which could improve the analyte recovery and method sensitivity were investigated by a D-optimal experimental design. Urine samples (10 mL) were first extracted with 15 mL ethyl acetate/formic acid (99/1, v/v) followed by SAX SPE clean-up of the acidified aqueous fraction. Both extracts were combined and analyzed using an LC-MS/MS system operated in the positive ionization mode. A total run time of 28 min was adopted with all the 18 analytes eluting within 15 min. The method was validated by taking into consideration the guidelines specified in Commission Decision 2002/657/EC and 401/2006/EC. Forty samples obtained from volunteers within the laboratory research group were analyzed as part of a pilot study. All results were expressed per mg creatinine. A total of 9 samples were found contaminated with one or more of the following analytes: DON, OTA, OTα, 4-OH OTA, ZEN, CIT and β-ZOL. One-eighth (5/40) of the samples were contaminated with DON in the range of 3.7-67 ng mg(-1) creatinine. Samples with detectable levels of DON did not show any co-occurrence of DON-3Glu. One sample was found to be contaminated with 4-OH OTA (<LOQ), co-occurring with only OTA (0.2 ng mg(-1) creatinine). OTα (up to 4.4 ng mg(-1) creatinine) was detected in three other samples co-occurring with low levels of OTA (up to 0.3 ng mg(-1) creatinine) and no 4-OH OTA detected. ZEN was detected in 10% (4/40) of the samples analyzed. Three samples were contaminated with β-ZOL (3.3-20 ng mg(-1) creatinine), co-occurring with ZEN (<LOQ-10.8 ng mg(-1) creatinine). The ratio of ZEN/β-ZOL varied for all the three samples. α-ZOL was not detected in any of the 40 samples. CIT was detected in one sample at 4.5 ng mg(-1) creatinine. This is the first study carried out with a small group of the Belgian population to assess exposure to mycotoxins using biomarkers.