Network analysis and follow up studies were performed in yeast and in mammalian cancer cell lines using biochemical approaches and high throughput high content image analysis. Collectively, the results indicate that the ultimate peptidomimetic FTI targets are proteins in the nucleus acting at the kinetochore having as hubs the Aurora A and MAD2 proteins. We show that Aurora A is indeed mislocalised in HeLa cells upon treatement with the FTI peptidomimetic FTI 277. Furthermore, we show that proteins acting at the cross roads of the Ras/PKA and TORC/S6K pathways are down regulated in yeast genetically impaired in FTase activity or treated with an FTI peptidomimetic. Consis tent with this, HeLa and MCF 7 cells show poor phos phorylation of the ribosomal S6 protein, a target of TORC/S6K during starvation.

Moreover, we show that a FTI peptidomimetic up regulated the MDR gene response. Results and Discussion Basic assumptions and set up of the assay conditions It is well established that the basic components of the prenylation, transcriptional, cell cycle and trafficking machineries are conserved between yeast and humans. Moreover, S. cerevisiae has proven to be a powerful genetic tool to elucidate the mode of action of clinically relevant compounds at the genomic level. We reasoned that if there is a basic common mechanism by which FTIs act as antitumor agents in eukaryotic cells then this mechanism could be identified in yeast cells based solely on FTase activity inhibition and the results reciprocated in mammalian cells, irrespective of the FTI compound used in the two organisms.

Human and yeast FTases are conserved in structure and function. They are constituted by two subunits, called RAM1 and RAM2 in yeast cells. The b subunit RAM2 is shared with GGTase I, while the alpha subunit is specific for each enzyme, RAM1 for FTase and CDC43 for GGTase I. Importantly, for the aims of this study, cells deleted for the RAM1 Drug_discovery gene are viable as essential proteins, that cannot be farnesylated in the absence of FTase activity, are geranylgeranylated by GGTase I. Thus, budding yeast provides a unique genetic tool to decipher the global response to FTIs and to prove that FTI antiproliferative action relies solely on the inhibition of FTase activity. To define the expression signature of FTI or GGTI treated yeast cells we chose well studied and commercially available compounds. All of them had been already tested for their antiproliferative action in cancer cell lines. To be suitable for our study, we applied the fol lowing constraints. The compound should 1 be cell permeable . 2 be already demonstrated to possess a higher specifi city either for FTase or GGTase I in a mammalian cel lular system.