ABSTRACT
We present a new computational method for identifying regulated pathway components in transcript profiling (TP) experiments by evaluating transcriptional activity in the context of known biological pathways. We construct a graph representing thousands of protein functional relationships by integrating knowledge from public databases and review articles. We use the notion of distance in a graph to define pathway neighborhoods. The pathways perturbed in an experiment are then identified as the subgraph induced by the genes, referred to as activity centers, having significant density of transcriptional activity in their functional neighborhoods. We illustrate the predictive power of this approach by performing and analyzing an experiment of TP53 overexpression in NCI-H125 cells. The detected activity centers are in agreement with the known TP53 activation effects and our independent experimental results. We also apply the method to a serum starvation experiment using HEY cells and investigate the predicted activity of the transcription factor MYC. Finally, we discuss interesting properties of the activity center approach and its possible applications beyond the comparison of two experiments.
Subject(s)
Gene Expression Profiling/statistics & numerical data , Signal Transduction/genetics , Algorithms , Apoptosis/genetics , Cell Cycle/genetics , Cell Line, Tumor , Culture Media, Serum-Free , DNA, Complementary/biosynthesis , DNA, Complementary/genetics , Databases as Topic , Genes, p53/genetics , HumansABSTRACT
MLN944 (XR5944) is a novel bis-phenazine that has demonstrated exceptional efficacy against a number of murine and human tumor models. The drug was reported originally as a dual topoisomerase I/II poison, but a precise mechanism of action for this compound remains to be determined. Several lines of evidence, including the marginal ability of MLN944 to stabilize topoisomerase-dependent cleavage, and the sustained potency of MLN944 in mammalian cells with reduced levels of both topoisomerases, suggest that other activities of the drug exist. In this study, we show that MLN944 intercalates into DNA, but has no effect on the catalytic activity of either topoisomerase I or II. MLN944 displays no significant ability to stimulate DNA scission mediated by either topoisomerase I or II compared with camptothecin or etoposide, respectively. In addition, yeast genetic models also point toward a topoisomerase-independent mechanism of action. To examine cell cycle effects, synchronized human HCT116 cells were treated with MLN944, doxorubicin, camptothecin, or a combination of the latter two to mimic a dual topoisomerase poison. MLN944 treatment was found to induce a G(1) and G(2) arrest in cells that is unlike the typical G(2)-M arrest noted with known topoisomerase poisons. Finally, transcriptional profiling analysis of xenograft tumors treated with MLN944 revealed clusters of regulated genes distinct from those observed in irinotecan hydrochloride (CPT-11)-treated tumors. Taken together, these findings suggest that the primary mechanism of action of MLN944 likely involves DNA binding and intercalation, but does not appear to involve topoisomerase inhibition.
Subject(s)
Camptothecin/analogs & derivatives , Intercalating Agents/pharmacology , Phenazines/pharmacology , Animals , Antigens, Neoplasm , Camptothecin/pharmacology , Catalysis/drug effects , Cell Cycle/drug effects , Cell Line, Tumor , Cluster Analysis , DNA/chemistry , DNA/metabolism , DNA Topoisomerases, Type I/genetics , DNA Topoisomerases, Type I/metabolism , DNA Topoisomerases, Type II/genetics , DNA Topoisomerases, Type II/metabolism , DNA-Binding Proteins , Dose-Response Relationship, Drug , G1 Phase/drug effects , G2 Phase/drug effects , Gene Expression Profiling , HCT116 Cells , Humans , Intercalating Agents/chemistry , Irinotecan , Male , Mice , Mice, Nude , Mitosis/drug effects , Mutation , Neoplasms, Experimental/drug therapy , Neoplasms, Experimental/genetics , Neoplasms, Experimental/pathology , Phenazines/chemistry , Transplantation, Heterologous , Xenograft Model Antitumor Assays , Yeasts/drug effects , Yeasts/enzymology , Yeasts/geneticsABSTRACT
Agmatinase, which hydrolyzes agmatine to putrescine and urea, not only represents a potentially important mechanism for regulating the biological effects of agmatine in mammalian cells but also represents an alternative to ornithine decarboxylase for polyamine biosynthesis. We have isolated a full-length cDNA encoding human agmatinase whose function was confirmed by complementation in yeast. The single-copy human agmatinase gene located on chromosome 1 encodes a 352-residue protein with a putative mitochondrial targeting sequence at the NH(3)-terminus. Human agmatinase has about 30% identity to bacterial agmatinases and <20% identity to mammalian arginases. Residues required for binding of Mn(2+) at the active site in bacterial agmatinase and other members of the arginase superfamily are fully conserved in human agmatinase. Agmatinase mRNA is most abundant in human liver and kidney but also is expressed in several other tissues, including skeletal muscle and brain. Its expression in human liver is induced during hepatitis B virus infection, suggesting that agmatinase may play a role in the pathophysiology of this disease.
