| The gamma isoform of peroxisome-proliferator activated receptor (PPARgamma) is a member of the super family of nuclear hormone receptors and shows much promise as a chemopreventative and therapeutic target for colorectal cancer. Activation of PPARgamma by thiazolidinediones (TZDs) inhibits proliferation and induces differentiation in human colon cancer cells. RS5444, a novel TZD, is a high affinity and high specificity ligand for PPARgamma. We have shown that RS5444 markedly reduced the proliferation of MOSER S human colorectal cancer cells under anchorage dependent and independent conditions. The inhibitory effect of RS5444 was irreversible. RS5444 also significantly repressed the invasive phenotype, but not motility, of these tumor cells.;Towards elucidating the activated PPARgamma controlled genomic program responsible for these observed phenotypes, functional genomic analysis was performed using a two-class longitudinal microarray data set in the presence and absence of RS5444. Differential expression of genes was calculated using an empirical Bayesian modification to the multivariate HotellingT2 score. We have demonstrated this statistical machine learning technique to be superior in controlling type II error in our dataset than more commonly used algorithms for two-class analysis. Likewise, through the use of several bioinformatics techniques, including frequency-based pathway and ontology analysis, we found a yet unappreciated tumor-suppressing network involving a feedback mechanism between PPARgamma, DSCR1 and calcineurin-mediated signaling of NFATc in colorectal cancer cells. To this end, we have demonstrated a direct connection between NFATc and DSCR1 in MOSER S colorectal cancer cells. Likewise, we have demonstrated a correlation between the sensitivity of PPARgamma in other colorectal cancer cells, and the messenger abundance of DSCR1. Finally, we have demonstrated that knockdown of DSCR1 messenger obviates the phenotypic effects of activated PPARgamma in vitro.;To our knowledge these data represent, for the first time, a network between PPARgamma, DSCR1, and NFATc signaling in the context of tumor-suppressor activity. This preliminary evidence is consistent with our working hypothesis that an oncology patient's receptiveness to TZD treatment may be largely dependent on the specific tumor's endogenous abundance of DSCR1. We believe without a critical endogenous level of DSCR1, activated PPARgamma may revert to a tumor-activator instead of a tumor-suppressor. |