Detection of trace aflatoxin M1 in human urine using a commercial ELISA followed by HPLC.

Aflatoxin is a liver carcinogen, and rapid, inexpensive methods to detect its urinary biomarkers are needed. We used a commercial enzyme-linked immuno-sorbent assay (ELISA) for aflatoxin M1 in urine (Helica Biosystems) to test 52 Haitian samples. Using this ELISA, we detected traces above the limit of detection (0.2 ng/ml urine) but below the limit of quantitation (0.4 ng/ml) in 14 samples. Liquid chromatography of all 52 Haitian urine samples revealed that only 11 had quantifiable AFM1 (mean: 29.5 pg/ml, standard error: 10.8, range: 2.94-96.5 pg/ml). The Helica ELISA may have detected forms of aflatoxin other than AFM1 in the Haitian samples, or matrix enhancement may have affected results at low AFM1 concentrations. This ELISA may serve as an initial, qualitative indicator of aflatoxin exposure for epidemiological purposes. But this method's utility as a precise and specific indicator of AFM1 concentrations will require additional refinement and validation.

Urinary aflatoxin M1 in Port-au-Prince and a rural community in north-east Haiti.

Aflatoxins (AFs) are hepatocarcinogenic mycotoxins that can contaminate grains and oil seeds in tropical and sub-tropical areas and have been detected in maize and peanut products of Haiti. The first objective was to assess human exposure to AFs among Haitians at an urban hospital (GHESKIO) and a rural health centre (HCBH). The second objective was to test the association between AF exposure and reported dietary intake of potentially contaminated foods, such as maize, peanut products and milk. Measurement of urinary AFM1 by HPLC revealed that among 367 participants 14% and 22% at GHESKIO and HCBH, respectively, had detectable AFM1. The maximum and median AFM1 concentrations for all detected samples were 700 pg AFM1 ml(-1) and 11.7 pg ml(-1), respectively. Detection of AFM1 was significantly associated with peanut consumption (p < 0.05). Controlling for diet and age group in a logit model, patients who reported peanut consumption the day of the survey and patients from HCBH had greater log odds of excreting detectable AFM1 (p < 0.001 and 0.002, respectively); females had lower log odds (p = 0.020). Recalled frequency of consuming non-dairy animal-sourced foods, an indicator of diet quality, approached significance (p = 0.056) as an inverse predictor of urinary AFM1 detection. The findings augur the need for interventions that will improve food safety in Haiti and limit exposure to AFs, particularly among rural communities.

A comparative study of the human urinary mycotoxin excretion patterns in Bangladesh, Germany, and Haiti using a rapid and sensitive LC-MS/MS approach.

An improved "dilute and shoot" LC-MS/MS multibiomarker approach was used to monitor urinary excretion of 23 mycotoxins and their metabolites in human populations from Asia (Bangladesh), Europe (Germany), and the Caribbean region (Haiti). Deoxynivalenol (DON), deoxynivalenol-3-glucuronide (DON-3-GlcA), T-2-toxin (T-2), HT-2-toxin (HT-2), HT-2-toxin-4-glucuronide (HT-2-4-GlcA), fumonisin B1 (FB1), aflatoxins (AFB1, AFB2, AFG1, AFG2, AFM1), zearalenone (ZEA), zearalanone (ZAN), their urinary metabolites α-zearalanol (α-ZEL) and ß-zearalanol (ß-ZEL), and corresponding 14-O-glucuronic acid conjugates (ZEA-14-GlcA, ZAN-14-GlcA, ß-ZEL, α/ß-ZEL-14-GlcA), ochratoxin A (OTA), and ochratoxin alpha (OTα) as well as enniatin B (EnB) and dihydrocitrinone (DH-CIT) were among these compounds. Eight urinary mycotoxin biomarkers were detected (AFM1, DH-CIT, DON, DON-GLcA, EnB, FB1, OTA, and α-ZEL). DON and DON-GlcA were exclusively detected in urines from Germany and Haiti whereas urinary OTA and DH-CIT concentrations were significantly higher in Bangladeshi samples. AFM1 was present in samples from Bangladesh and Haiti only. Exposure was estimated by the calculation of probable daily intakes (PDI), and estimates suggested occasional instances of toxin intakes that exceed established tolerable daily intakes (TDI). The detection of individual mycotoxin exposure by biomarker-based approaches is a meaningful addition to the classical monitoring of the mycotoxin content of the food supply.