338 COMPUTATIONAL STRUCTURAL AND FUNCTIONAL PROTEOMICS A MOLECULAR MECHANI SM FOR THESTRUCTURE-FUNCTIONAL ALTERATIONS IN MUTANT FORMS OF HUMAN P5 3 PROTEIN Pintus S.S.* 1 , Ivanisenko V.A. 2 1 Novosibirsk State University, Novosibirsk, Russia; 2 Institute of Cytology and Genetics SB RAS, Novosibirsk, Russia * Corresponding author: e-mail: pintus@bionet.nsc.ru Keywords: gain-of-function mutations, p53, structure alignment, molecular dynamics Summary Motivation: The exact molecular mechanisms of the effects of the gain-of-function mutations are not known as yet (Gualberto et al., 1998). Here, we performed a computer analysis of the possible molecular mechanisms of the gain-of-function effect of cancer-prone mutations in human p53. Results: In this study we identified new functional sites in mutant human p53 protein, using the structure alignment technique. Molecular dynamics simulations were used to demonstrate the efficiency of the found sites. We also suggested a molecular mechanism of the gain-of-function effect in the mutant p53 protein. Introduction The p53 protein is a key factor for cell cycle regulation and apoptosis in tumor cells. If DNA of a cell is damaged, the protein can stop the cell cycle in the G1 phase (Kastan et al., 1991); this allows the cell to undergo reparation before replication starts or apoptosis that would result in cell death (Lowe et al., 1993). The p53 gene is mutated in 45–50 % of human tumor cells (Greenblatt et al., 1994). The prevailing reason for cancer diseases is point substitutions in the p53 gene. The substitutions are of particular interest because they often do not inactivate the p53 product (Soussi et al., 2000). The substitutions include the so-called gain-of-function mutations that result in new oncogenic properties of a mutant protein. We proposed that some mutations in the p53 gene may result in appearance of new functional sites. These sites determine the gain-of-function effect. To test this assumption, we searched for new functional sites in the mutant forms of the p53 protein. Methods and Algorithms The previously developed program, PDBSiteScan, was used to search for new functional sites in the p53 protein. Its algorithm is based on step-by-step extension of a set of juxtaposed residues of a predictor site and a protein using restricted search (Ivanisenko et al., 2002). PDBSiteScan searches for active, binding and posttranslational modification protein sites; the program structurally aligns the 3d structure of a protein with every exemplary site stored in the PDBSite database to identify the site searched in the protein (Ivanisenko et al., 2002). To search for new sites in the mutant forms of the p53 protein, the data were processed as follows. The 3d structure of a normal protein was taken from the PDB database (Berman et al., 2000); then, a normal residue was assigned a mutant name in the resName field for every ATOM record of a mutant residue. Thus, a new entry in the PDB format was obtained. This entry contains the atom coordinates for the normal protein and a primary sequence for the mutant. PDBSiteScan aligns only atoms of the main protein chain known to be not highly altered due to missense mutations (Soussi et al., 2000). On the other hand, PDBSite demands identity of two primary sequences of the exemplary site and the one to be predicted. In this way, search for the functional sites in the tertiary structure of BGRS'2004