Critical evaluation of indirect methods for the determination of deoxynivalenol and its conjugated forms in cereals.

A critical assessment of three previously published indirect methods based on acidic hydrolysis using superacids for the determination of "free" and "total" deoxynivalenol (DON) was carried out. The modified mycotoxins DON-3-glucoside (D3G), 3-acetyl-DON (3ADON), and 15-acetyl-DON (15ADON) were chosen as model analytes. The initial experiments focused on the stability/degradation of DON under hydrolytic conditions and the ability to release DON from the modified forms. Acidic conditions that were capable of cleaving D3G, 3ADON, and 15ADON to DON were not found, raising doubts over the efficacy of previously published indirect methods for total DON determination. Validation of these indirect methods for wheat, maize, and barley using UHPLC-MS/MS was performed in order to test the accuracy of the generated results. Validation data for DON, D3G, 3ADON, and 15ADON in nonhydrolyzed and hydrolyzed matrices were obtained. Under the tested conditions, DON was not released from D3G, 3ADON, or 15ADON after hydrolysis and thus none of the published methods were able to cleave the modified forms of DON. In addition to acids, alkaline hydrolysis with KOH for an extended time and at elevated temperatures was also tested. 3ADON and 15ADON were cleaved under the alkaline pH caused by the addition of KOH or aqueous K2CO3 to "neutralize" the acidic sample extracts in the published studies. The published additional DON increase after hydrolysis may have been caused by huge differences in matrix effects and the recovery of DON in nonhydrolyzed and hydrolyzed matrices as well as by the alkaline cleavage of 3ADON or 15ADON after the neutralization of hydrolyzed extracts.