Strategie computazionali per lo studio delle interazioni tra macromolecole e per la ricerca di nuovi inibitori di proteine chinasi

The concept of "druggable genome" limits the molecular targets for which commercially viable compounds can be developed. In humans, whose genome accounts for about 30,000 genes, it has been calculated that a subset of no more than 3,000 genes express proteins able to bind drug-like molecules. This limit derives from the fact that the ability of a protein to bind a small molecule with the appropriate chemical properties does not necessarily make it a potential drug target, this latter feature being only applied to proteins that are also linked to disease. Among these, it has become evident that the proteins participating in intra- or intercellular communication represent the largest family with about 20-25% of the druggable genome consisting of kinases followed by G-protein coupled receptors (15%) and cation channels (5%), three of the fundamental groups of proteins implicated in signal transduction. For this reason, protein kinases have turned out to be one of the richest target mines, since drugs inhibiting specific kinases are in constant development, and some of them are currently investigated in clinical trials such as Gleevec and Iressa. Most of the structural information currently available about protein kinases is resulting from the continuously growing number of crystallographic structures solved. These studies suggest that the overall architecture of the kinases is quite similar, however, since the ATP-binding site possesses some distinct subsites, different kinases can be targeted by quite selective drugs. Two main objectives will be pursued: a) a search for new potent and selective inhibitors and b) an analysis of the interactions between kinases and related substrates. a) Our laboratory has a long-lasting expertise in targeting the human kinome, especially the protein kinase CK1, CK2 and Aurora-A, all implicated in carcinogenesis; we performed a new virtual screening approach searching for new potent and selective scaffolds against the three kinases. Virtual screening involves the rapid assessment of large libraries of chemical structures in order to guide the selection of potential drug candidates. The results obtained was really encouraging; in fact new good inhibitors for all the three kinases where obtained from the virtual screening strategy. b) The interaction of protein kinases with their substrates and/or effectors represents a key event in the overall picture of cellular regulation. In the case of CK2 more than 300 substrates have been identified the majority of which are phosphorylated both in vitro and in vivo. Our purpose is to highlight the molecular features underlying the interaction between CK2 and its peptides or protein substrates by using our protein-protein docking approach. Protein-protein docking is a computational tool for the prediction of three-dimensional structure of protein complexes from the coordinates of the component's structure.