Protein kinases are key signaling molecules that have been implicated in the development of many diseases. As specific kinase inhibitors are developed for therapeutic purposes, it is important to identify kinase/substrate networks to anticipate the results of kinase inhibition. Current methods of identifying novel kinase substrates are based on peptide modeling, computational methods, and protein-protein interaction traps. All of these approaches have significant limitations. We propose a genetic screen for identification of kinase substrates, called SIMPKIN (Simultaneous Identification of Multiple Protein KINase substrates). This method will be based on the simultaneous expression in bacteria of a kinase and a cDNA library of potential substrates. Phosphorylated proteins will be recovered through affinity chromatography, and the identities will be determined by a combination of gel electrophoresis and mass spectroscopy. As a test of SIMPKIN we will study the pim-1 kinase. We propose three specific aims for the Phase I portion of this SBIR application. (1) We will construct a bicistronic bacterial expression system for co-expression of the human pim-1 cDNA and cDNAs encoding known pim-1 substrates. Details of expression, purification, and analysis of the resulting phosphorylated proteins will be adjusted to permit recovery of at least some valid pim-1 substrates. (2) We will then co-express pim-1 kinase with a cDNA library derived from human RWPE2 prostate cancer cells, which contain biologically active PIM1 protein. The library plasmid pool will be """"""""spiked"""""""" with bicistronic plasmids encoding pim-1 and known substrates. The recovered phosphorylated substrates will be analyzed to determine the efficiency of the SIMPKIN method, and to estimate the size of the pim-1 substrate pool. (3) Potential novel substrates identified under specific aim (2) will be confirmed by co-expression of the protein with pim-1 in mammalian cells. The cDNAs will be inserted into mammalian expression plasmids, and transiently transfected (with or without a human pim-1 cDNA) into HEK293 human kidney cells. Phosphorylation of the potential substrate will be determined by a combination of immunochemical techniques and mass spectroscopy. These studies should confirm the utility of SIMPKIN for genome-wide identification of kinase substrates, and should provide a useful tool to permit the development of databases of kinase/substrate interactions in normal and pathologic tissues. ? ?
