A. Understanding the behavior of bioregulatory networks is critical for future progress in biology and therapeutics. The complexity and high interconnectivity of these networks however often makes it difficult to comprehend directly how the system functions. Understanding of function will often require simulation by computer. It is questionable however whether the present state of knowledge is sufficient for computational simulations of large systems to yield useful results. We took an alternative approach, based on the notion that essential behavior is already encoded in simple subsystems, and that further complexity serves to modulate this behavior. This possibility is attractive from considerations of evolution and system robustness. A second premise for the current investigation is that the response characteristics of the fundamental subsystems are often switch-like. This theoretical approach, to the degree that it may be successful, will facilitate linkage between theory and experiment. Subsystems characterized by a combination of theory and experiment could then be combined to model larger networks. The subject of our investigation was the network that controls the induction of a set of genes in response to hypoxia. This was an attractive subject for theoretical study, because extensive (albeit yet incomplete) information has accumulated about relevant molecular and biological behavior, and because there is a clearly defined dependence of an output on an input. The input is the concentration of molecular oxygen, and the output is the activation of promoters that are under the control of hypoxia-regulated elements (HRE). We showed how a model (""""""""core"""""""") subsystem can be selected from a molecular interaction map, how it can be encoded for computer survey of parameter space, how switch-like behavior can be found, and how the results may predict or guide experiments. A manuscript describing this work has been prepared. B. In a study of gene expression changes associated with a drug resistant cell line we encountered a pattern of seemingly paradoxical changes consistent with the hypothesis that multi-drug resistance could primarily involve the development of a defect in apoptosis (recently published: Reinhold et al. 2003 Cancer Res.). The hypothesis, which we call """"""""permissive apoptosis resistance,"""""""" also entails gene expression changes that promote cell proliferation but that would induce apoptosis were it not for a primary apoptosis defect. Our study compared a human prostate cancer cell line (DU145) with a subline that had been selected for resistance to a camptothecin. cDNA microarray analysis of 1,648 cancer-related genes disclosed expression differences in 181 of these genes. Among these 181 genes, there were a larger than expected number of the apoptosis-related genes, suggesting a general defect in the apoptosis machinery. Indeed, the resistant line displayed reduced apoptotic responses to several agents other than camptothecin, including cisplatin, staurosporine, UV, ionizing radiation, and serum deprivation. Closer examination of the gene expression changes however disclosed the surprising fact that many of the changes were in the wrong direction for apoptosis resistance. In order to make sense of this surprising pattern, we charted the altered genes in a molecular interaction map, using the symbols defined by Kohn (1999 Mol. Biol. Cell). What emerged is that almost all of the genes altered in the expected direction fell into one of two categories. Most were in the downstream core apoptosis network, including the genes BAD, caspase-6, and some associated genes, while some were regulators of the key survival kinase, Akt/PKB. On the other hand, nearly all of the genes altered in the contrary direction were in pathways that impact both apoptosis and cell proliferation, including c-Myc, E2F1, and players in the NFkB and TGFb pathways. The changes in the drug-resistant cells were nearly always such that they would be expected to enhance both apoptosis and cell proliferation. These findings suggested a model (""""""""permissive apoptosis resistance"""""""") as one route by which cells could acquire multi-drug resistance. According to this model, cells would first be selected for one or more defects in the core apoptosis network, i.e., in genes that function primarily to control or execute apoptosis. When this defect is in place, the cells would then be free to acquire other gene expression changes that would enhance proliferation, but that would also have induced apoptosis were it not for the downstream defect.
| Solier, Stephanie; Sordet, Olivier; Kohn, Kurt W et al. (2009) Death receptor-induced activation of the Chk2- and histone H2AX-associated DNA damage response pathways. Mol Cell Biol 29:68-82 |
| Kohn, Kurt W; Aladjem, Mirit I (2006) Circuit diagrams for biological networks. Mol Syst Biol 2:2006.0002 |
| Pommier, Yves; Weinstein, John N; Aladjem, Mirit I et al. (2006) Chk2 molecular interaction map and rationale for Chk2 inhibitors. Clin Cancer Res 12:2657-61 |
| Kohn, Kurt W; Aladjem, Mirit I; Weinstein, John N et al. (2006) Molecular interaction maps of bioregulatory networks: a general rubric for systems biology. Mol Biol Cell 17:1-13 |
| Pommier, Yves; Sordet, Olivier; Rao, V Ashutosh et al. (2005) Targeting chk2 kinase: molecular interaction maps and therapeutic rationale. Curr Pharm Des 11:2855-72 |
| Pommier, Yves; Sordet, Olivier; Antony, Smitha et al. (2004) Apoptosis defects and chemotherapy resistance: molecular interaction maps and networks. Oncogene 23:2934-49 |
| Kohn, Kurt W; Riss, Joseph; Aprelikova, Olga et al. (2004) Properties of switch-like bioregulatory networks studied by simulation of the hypoxia response control system. Mol Biol Cell 15:3042-52 |
| Aladjem, Mirit I; Pasa, Stefania; Parodi, Silvio et al. (2004) Molecular interaction maps--a diagrammatic graphical language for bioregulatory networks. Sci STKE 2004:pe8 |
