Dialkylquinonimines validated as in vivo metabolites of alachlor, acetochlor, and metolachlor herbicides in rats

The oncogenicity of chloroacetanilide herbicides (1-5) is proposed to involve bioactivation to 2,6-dialkylbenzoquinonimines (quinonimines, 9) based on two earlier observations: (1) in vitro conversion of the alachlor (1) metabolite diethylaniline (7Et(2)) to 2,6-diethylquinonimine (9Et(2)) which reacts readily with GSH and (2) induction of sister chromatid exchanges in human lymphocytes for the parent herbicides and their purported 9 metabolites. This hypothesis lacks in vivo evidence for 9 formation which might be provided by analysis of urine and tissue for thiol adducts of 9. Accordingly, two mercapturates (10Et(2) and 10Me(2)) and a cysteine conjugate (11Me(2)) were prepared by addition of N-acetylcysteine or cysteine to 9Et(2) and the 2,6-dimethyl homologue (9Me(2)). Mercapturate 10Et(2) was characterized by HPLC using the urine of rats treated ip with hydroxyaniline 8Et(2), and both mercapturate 10Me(2) and cysteine conjugate 11Me(2) were found in the urine of mice administered hydroxyaniline 8Me(2). The mercapturates were then converted to the N,N-dimethyl-2,6-dialkyl-4-methoxy-3-(methylthio)-anilines (12Et(2), 12EtMe, and 12Me(2)) by alkaline permethylation, thereby providing a method for analysis of 9-derived thiol adducts in urine and liver as the 12 derivatives by GC/MS with selected ion monitoring, The urine of rats 0-6 h after ip treatment with 1, butachlor (2), acetochlor (3), metolachlor (4), and dimethachlor (5) at 0.74 mmol/kg yields permethylated derivatives which are definitively diagnostic for the 9 intermediates from each of the herbicides in amounts of 3-24-fold above the minimum detectable limit, as well as 1 and 2 orders of magnitude higher from the corresponding anilines (7) and hydroxyanilines (8), respectively. Similar liver analyses reveal tissue thiol adducts of 9 6 h after treatment with 7 and 8 but not with the parent herbicides. The yields of urinary 9 derivatives from the parent herbicides are higher from the 2,6-diethyl series (1 and 2) and the 2-ethyl-6-methyl derivatives (3 and 4) than from the 2,6-dimethyl analogue (5). These findings provide direct evidence in vivo that quinonimines are metabolites of 1-5 in rats.