Chlorsulfuron is the first commercialized sulfonylurea herbicide, which targets acetohydroxyacid synthase (AHAS). Mutations in AHAS have caused serious herbicide resistance to chlorsulfuron. Quantitative description of the herbicide resistance in molecular level will benefit the understanding of the resistance mechanism and aid the design of resistance‐evading herbicide. We have recently established a MB‐QSAR (Mutation‐dependent Biomacromolecular Quantitative Structure‐Activity Relationship) method to conduct the 3D‐QSAR study in biomacromolecules. Herein, based on the herbicide resistance data measured for a series of AHAS mutants against chlorsulfuron, we constructed MB‐QSAR models to quantitatively predict the herbicide resistance and interpret the structure resistance relationships for AHAS mutants against chlorsulfuron. Quite well correlations between the experimental and the predicted pKi values were achieved for MB‐QSAR/CoMFA (q2=0.705, r2=0.918, r2pred=0.635) and MB‐QSAR/CoMSIA (q2=0.558, r2=0.940, r2pred=0.527) models, and interpretation of the MB‐QSAR models gave chemical intuitive information to guide the resistance‐evading herbicide design.

No abstract

No abstract

No abstract

Conquering the mutational drug resistance is a great challenge in anti-HIV drug development and therapy. Quantitatively predicting the mutational drug resistance in molecular level and elucidating the three dimensional structure-resistance relationships for anti-HIV drug targets will help to improve the understanding of the drug resistance mechanism and aid the design of resistance evading inhibitors. Here the MB-QSAR (Mutation-dependent Biomacromolecular Quantitative Structure Activity Relationship) method was employed to predict the molecular drug resistance of HIV-1 protease mutants towards six drugs, and to depict the structure resistance relationships in HIV-1 protease mutants. MB-QSAR models were constructed based on a published data set of Ki values for HIV-1 protease mutants against drugs. Reliable MB-QSAR models were achieved and these models display both well internal and external prediction abilities. Interpreting the MB-QSAR models supplied structural information related to the drug resistance as well as the guidance for the design of resistance evading drugs. This work showed that MB-QSAR method can be employed to predict the resistance of HIV-1 protease caused by polymorphic mutations, which offer a fast and accurate method for the prediction of other drug target within the context of 3D structures.