|Ph.D Student||Senitzki Alon|
|Subject||DNA Structural Flexibility and the Mechanism of P53|
Binding to its Target Sites
|Department||Department of Biology||Supervisor||Professor Tali Haran|
|Full Thesis text|
The tumor suppressor p53 operates as a transcription factor (TF) by binding DNA response elements (REs) at gene regulatory regions. Upon binding to its REs, p53 has been shown to regulate the expression of more than 200 target genes, thus leading to diverse cellular outcomes. Cell-cycle arrest, DNA repair, senescence and apoptosis, constitute only a partial representation of p53 transcriptional repertoire. The p53 RE is composed of two repeats of a decameric half-site with the degenerate consensus of RRRCWWGYYY (R = A, G; W = A, T; Y = C, T). The two half-sites within the p53 RE may be separated by a random spacer of up to 18 bases. However, it has been shown that in REs of highly inducible target genes, the spacer length is bellow 3-bp, and in about half of all validated p53 REs, the two half-sites are contiguous.
p53 binds its REs as a dimer of dimers, through its DNA binding domain (DBD) to form DNA-bound tetramer, which is the protein functional unit. In each dimer, two DBD's interact with two pentameric repeats within a half-site. The binding of p53 onto its REs is highly cooperative, which is supported by contacts within the tetramerization domain, and also by protein-protein interactions between the DBD's. The loose consensus definition of the p53 target site implies that discrimination between different REs cannot be based on direct readout alone, and it was suggested that sequence-dependent structural features of the p53 REs also contribute to p53 differential binding through indirect-readout mechanism.
Here I show that p53 REs display structural asymmetry, which is represented by difference in the sequence-dependent structural flexibility of the half-sites composing each full-site. This was determined experimentally, in term of torsional flexibility, which is measured by the cyclization kinetics assay, for six half-sites belonging to three well characterized p53 REs (p21-5', p21-3' and BaxB). Moreover, structural asymmetry of p53 REs is shown here to be prevalent also at a genomic scale, represented as the overall difference in flexibility of two half-sites within each RE, from calculations of structural deformability of 155 validated REs without spacer.
By using the p21-5' RE as a model p53 binding site, I show here that changing the directionality of its two composing half-sites and/or their flanking sequences environment, effects mainly the DNA binding affinity of dimeric p53 complexes and its binding cooperativity. Specifically, the presence of rigid flanking sequence, and in particular when adjacent to a flexible half-site, results in higher dimer binding affinity and lower binding cooperativity. Moreover, by collaborating with the group of Prof. Alberto Inga (CIBO, University of Trento, Italy), we show that these changes in binding affinity are related to transactivation capacity under basal cellular levels of p53.