Conjugation of deoxynivalenol (DON) with sulfur compounds is recognized as a significant reaction pathway, and putative DON-glutathione (DON-GSH) conjugates have been reported in planta. To understand and control the reaction of trichothecenes with biologically important thiols, we studied the reaction of DON, T-2 tetraol, and de-epoxy-DON with a range of model thiols. Reaction conditions were optimized for DON with 2-mercaptoethanol. Major reaction products were identified using HRMS and NMR spectroscopy. The results indicate that thiols react reversibly with the double bond (Michael addition) and irreversibly with the epoxide group in trichothecenes. These reactions occurred at different rates, and multiple isomers were produced including diconjugated forms. LC-MS analyses indicated that glutathione and cysteine reacted with DON in a similar manner to the model thiols. In contrast to DON, none of the tested mercaptoethanol adducts displayed toxicity in human monocytes or induced pro-inflammatory cytokines in human macrophages.

Conjugation with the biologically relevant thiol glutathione is one of the metabolic pathways for the mycotoxin deoxynivalenol (DON) in wheat. The occurrence of putative DON-cysteine conjugates has also been shown in wheat, likely in part as a result of degradation of the glutathione conjugates. It was reported that thiols react in vitro with DON at two positions: reversibly at C-10 of the α,β-unsaturated ketone and irreversibly at C-13 of the epoxy group. We synthesized pure DON-cysteine adducts and made analytical standards using quantitative NMR experiments. Compounds were characterized using NMR and LC-HRMS/MS and tested in vitro for toxicity. Cysteine conjugates were much less toxic than DON at the same concentration, and LC-HRMS analysis demonstrated that there was no detectable metabolism of the conjugates in human monocytes or human macrophages.

Glutathione (GSH) conjugates of the mycotoxin 4-deoxynivalenol (DON), 1, have been detected in plants by LC-MS, but their identities were not confirmed due to a lack of standards. We have synthesized DON-GSH conjugates in alkaline solution. The major products 2 and 5 were isolated and their structures determined by mass spectrometry and NMR spectroscopy as GSH adducts at C-13 and C-10 (via epoxide and Michael addition, respectively) of 1. Other Michael addition products were also tentatively identified by LC-MS. Concentrations of 2 and 5 were determined by quantitative NMR and are suitable for use as quantitative standards for LC-MS studies of plant and animal metabolism of 1. LC-MS showed that in the presence of human glutathione S-transferases of the alpha and mu classes, the reaction of DON and GSH proceeded with a half-life of 17 h, identical with the rate of the uncatalyzed reaction rate, indicating an absence of catalysis.

A glutathione (GSH) adduct of the mycotoxin 4-deoxynivalenol (DON), together with a range of related conjugates, has recently been tentatively identified by LC-MS of DON-treated wheat spikelets. In this study, we prepared samples of DON conjugated at the 10- and 13-positions with GSH, Cys, CysGly, γ-GluCys and N-acetylcysteine (NAC). The mixtures of conjugates were used as standards for LC-HRMS analysis of one of the DON-treated wheat spikelet samples, as well as 19 Norwegian grain samples of spring wheat and 16 grain samples of oats that were naturally-contaminated with DON at concentrations higher than 1 mg/kg. The artificially-contaminated wheat spikelets contained conjugates of GSH, CysGly and Cys coupled at the olefinic 10-position of DON, whereas the naturally-contaminated harvest-ripe grain samples contained GSH, CysGly, Cys, and NAC coupled mainly at the 13-position on the epoxy group. The identities of the conjugates were confirmed by LC-HRMS comparison with authentic standards, oxidation to the sulfoxides with hydrogen peroxide, and examination of product-ion spectra from LC-HRMS/MS analysis. No γ-GluCys adducts of DON were detected in any of the samples. The presence of 15-O-acetyl-DON was demonstrated for the first time in Norwegian grain. The results indicate that a small but significant proportion of DON is metabolized via the GSH-conjugation pathway in plants. To our knowledge, this is the first report of in vivo conjugation of trichothecenes via their epoxy group, which has generally been viewed as unreactive. Because conjugation at the 13-position of DON and other trichothecenes has been shown to be irreversible, this type of conjugate may prove useful as a biomarker of exposure to DON and other 12,13-epoxytrichothecenes.