Insights into Structure/Function Relationships in the p53 Tumor Suppressor from Chemical Probes Studies and Atomic Force Microscopy Imaging
It is well known that the functional properties of the p53 tumor suppressor depend critically upon its ability to bind DNA, particularly at its many genomic response elements but possibly also through nonspecific binding to nearby regions of various promoters. There is also evidence that p53 function is allosterically regulated and that p53 functions frequently as an important component of larger transcriptional complexes. Thus, a comprehensive structural model for the full protein bound to its DNA response elements is a prerequisite to a more complete understanding of p53 function and control. Such a model can lead to critical new insights into the binding specificity and selectivity of this protein as governed not only by specific protein-DNA contacts but by specific protein-protein contacts and allosteric regulation as well. Unfortunately, it has not yet been possible to apply traditional structural methods such as high resolution crystallography and nmr to this system and alternative, less direct methods are required in order to make further progress toward this goal.
We have developed a limited model of the p53 DNA binding domain peptide (p53DBD) bound tetramerically to a single DNA response element, the Waf1/Cip1/p21 site. Although this model offers insights into possible protein-protein interactions that can explain the high cooperativity observed in the binding of this peptide to DNA, it cannot explain the roles of other nonbinding regions of the protein and the possible role of allosteric regulation. We have begun a program in which we combine biochemical and biophysical methods and advanced molecular modeling with the new technique of dynamic force microscopy (DFM) to address more sophisticated questions on the DNA binding properties of p53 that, for various reasons, have not yet been directly investigated by other methods. Our central assumption is that the DNA binding properties of p53, and hence its function, are strongly influenced by regions of the protein in addition to the p53DBD. This influence may derive both from interactions of various regions of the protein with each other and with DNA, through the effects of the flexible linker regions and through allosteric conformational changes in the entire protein. Other phenomena related to DNA binding which are also relevant to function include the propensity of p53 to loop DNA through octameric complexes and the relative binding of p53 to various response elements having slightly different sequences including those which exist in multiple copies in the same promoter such as the mdm2 site, two of which exist in tandem repeat near the TATA box in the mdm2 promoter.
To address these questions, we describe experiments in which we combine the use of chemical probes to establish specific protein-DNA contacts with both contact mode and high resolution magnetic mode (MacMode) atomic force microscopy (afm) to investigate the complexes of p53 deletion mutant peptides. The results of these experiments can be used to develop a more comprehensive molecular model and of direct importance in assessing p53 function. Related studies employ MacMode afm to image axially strained DNA microcircles containing strategically located p53 response elements. Analysis of these images demonstrates clearly that when these sites are forced to kink by axial strain induced by thermal cyclization, they bend and overtwist comparably to that predicted by the structural model for the p53 nucleoprotein complex. Hence, these experiments show that the propensity for DNA bending in p53 complexes is encoded into the response elements themselves and therefore represent an important source of sequence selectivity in the DNA binding of p53.
P. Balagurumoorthy, Luda S. Shlyakhtenko, 1Stuart M. Lindsay and Rodney E. Harrington
Department of Microbiology and 1Department of Physics and Astronomy,