Based on the results of computer simulations and high-resolution crystal structures, we have elucidated the extents of DNA sequence-dependent bending, twisting and stretching deformability. With these knowledge-based elasticity functions for DNA, we built a stereochemically feasible model for the tetrameric p53-DNA complex. The predicted directionality and magnitude of the DNA bending and twisting were subsequently confirmed by gel electrophoresis experiments. Now we are entering the second phase of this study, when the precise localization of the DNA bends is to be determined, and the complexes are to be compared for various p53 mutants. For this aim, we are using an original method, iodine-125 radioprobing, recently developed at NIH. Preliminary results for the wild type p53 are consistent with the predicted increase in DNA twisting and major groove bending in the consensus CATG motifs in the p53 binding sites. As a next step, we will analyze the DNA distortions in the complexes between various p53 mutants and the DNA response elements regulating the p53-related genes. New areas are: (1) DNA looping and transcription regulation; analysis of the multimeric complex with GalR repressor and HU protein. We will apply our mechanistic """"""""rules"""""""" to DNA containing ~100 base pairs, in order to interpret the available data on the GalR-HU binding to DNA, and to suggest new experiments, to elucidate the complicated 3D structure of this transcription regulation complex. (2) DNA binding to mutant hSRY protein (human testis determining factor). When mutant hSRY binds to cognate DNA, the degree of DNA bending differs from the wild type case. Based on the knowledge of the sequence-dependent properties of DNA, we will analyze how the atomic interactions at the DNA-protein interface lead to an increase or decrease in the level of the DNA deformation in the complex, which, in turn, is involved in regulation of transcription. Z01 BC 08371-17
