Design of Inhibitors of Photosystem II using a Model of the D1 Protein

Several inhibitors belonging to structurally different chemical classes were used to analyze the predictive power of an initial model of the herbicide binding niche of the D1 protein belonging to photosystem II (PS II) from plants. In the case of small PS II inhibitors, the estimation of relative activities was hampered by uncertainty about the binding modes. To overcome this problem, a bulky substituent was introduced into the inhibitors to act as a hook, resulting in an unambiguous orientation in the model. The comparison of the modelling results and the experimentally determined IC₅₀ values of different triazines suggested that the previously assumed volume of the binding niche had to be reduced by 20%. After refinement of the model, it was possible to estimate qualitatively, the relative in-vitro activity for inhibitors belonging to different families, as long as an unambiguous binding mode could be deduced either from steric demands or from IC₅₀ values of mutant D1 proteins. The usefulness of the refined model is demonstrated by the successful de-novo design of a potent class of herbicides, the triazolopyrimidines.