Biochemical interpretation of quantitative structure-activity relationships (QSAR) for biodegradation of N-heterocycles: a complementary approach to predict biodegradability.

Prediction of the biodegradability of organic compounds is an ecologically desirable and economically feasible tool for estimating the environmental fate of chemicals. We combined quantitative structure-activity relationships (QSAR) with the systematic collection of biochemical knowledge to establish rules for the prediction of aerobic biodegradation of N-heterocycles. Validated biodegradation data of 194 N-heterocyclic compounds were analyzed using the MULTICASE-method which delivered two QSAR models based on 17 activating (OSAR 1) and on 16 inactivating molecular fragments (GSAR 2), which were statistically significantly linked to efficient or poor biodegradability, respectively. The percentages of correct classifications were over 99% for both models, and cross-validation resulted in 67.9% (GSAR 1) and 70.4% (OSAR 2) correct predictions. Biochemical interpretation of the activating and inactivating characteristics of the molecular fragments delivered plausible mechanistic interpretations and enabled us to establish the following biodegradation rules: (1) Target sites for amidohydrolases and for cytochrome P450 monooxygenases enhance biodegradation of nonaromatic N-heterocycles. (2) Target sites for molybdenum hydroxylases enhance biodegradation of aromatic N-heterocycles. (3) Target sites for hydratation by an urocanase-like mechanism enhance biodegradation of imidazoles. Our complementary approach represents a feasible strategy for generating concrete rules for the prediction of biodegradability of organic compounds.