Identification of cancer mechanisms through computational systems modeling

• Type of Material: Article
• Authors: Zhen Qi, Georgia Institute of Technology; Eberhard Voit, Emory University
• Language: English
• Date: 2014-06-01
• Publisher: AME PUBL CO
• Publication Version: Final Published Version
• Copyright Statement: © 2011 - 2016 Translational Cancer Research. All rights reserved.
• Final Published Version (URL): http://dx.doi.org/10.3978/j.issn.2218-676X.2014.05.03
• Title of Journal or Parent Work: TRANSLATIONAL CANCER RESEARCH
• Volume: 3
• Issue: 3
• Start Page: 233
• End Page: 242
• Grant/Funding Information: This work was supported by a grant from the National Institutes of Health (P01-ES016731, Gary W. Miller, PI), a contract from NIH/NIAID (HHSN272201200031C; Mary Galinsky, PI), and an endowment from the Georgia Research Alliance (EOV, PI).
• Keywords: COLON; Monte Carlo simulation; VALIDATION; Science & Technology; optimization; reverse engineering; PURINE METABOLISM; colorectal cancer; Cancer mechanisms; systems biology; mathematical model; Life Sciences & Biomedicine; Oncology; purine metabolism; PROFILE
• Research Categories: Health Sciences, Oncology

Abstract

Background: Colorectal cancer is one of the most prevalent causes of cancer death. It has been studied extensively for a long time, and numerous genetic and epigenetic events have been associated with the disease. However, its molecular mechanisms are still unclear. High-throughput metabolomics data, combined with customized computational systems modeling, can assist our understanding of some of these mechanisms by revealing connections between alterations in enzymatic activities and their consequences for a person's metabolic profile. Of particular importance in this context is purine metabolism, as it provides the nucleotides needed for cell proliferation. Methods and findings: We employ a computational systems approach to infer molecular mechanisms associated with purine metabolism in colorectal carcinoma. The approach uses a dynamic model of purine metabolism as the simulation system and metabolomics data as input. The execution of large-scale Monte Carlo simulations and optimization with the model permits a step-wise reduction in possibly affected enzyme mechanisms, from which likely targets emerge. Conclusions: According to our results, some enzymes in the purine pathway system are very unlikely the targets of colorectal carcinoma. In fact, only three enzymatic steps emerge with statistical confidence as most likely being affected, namely: amidophosphoribosyltransferase (ATASE), 5'-nucleotidase (5NUC), and the xanthine oxidase/dehydrogenase (XD) reactions. The first of these enzymes catalyzes the first committed step of de novo purine biosynthesis, while the other two enzymes are associated with critical purine salvage pathways. The identification of these enzymes is statistically significant and robust. In addition, the results suggest potential secondary targets. The computational method cannot discern whether the inferred mechanisms constitute symptoms of colorectal carcinoma, or whether they might be causative and critical components of the uncontrolled cellular growth in cancer. The inferred molecular mechanisms present testable hypotheses that suggest targeted experiments for future studies of colorectal carcinoma and might eventually lead to improved diagnosis and treatment.