AbstractsBiology & Animal Science

Mechanistic and Statistical Models to Understand CXCL12/CXCR4/CXCR7 in Breast Cancer.

by Sei-Won Laura Chang




Institution: University of Michigan
Department: Chemical Engineering
Degree: PhD
Year: 2015
Keywords: modeling; cancer; multi-scale modeling; chemokine; Chemical Engineering; Microbiology and Immunology; Molecular, Cellular and Developmental Biology; Science (General); Engineering; Science
Record ID: 2061490
Full text PDF: http://hdl.handle.net/2027.42/111458


Abstract

Signaling via the CXCL12/CXCR4 axis is instrumental to the metastasis of more than 20 cancers, yet blocking the pathway alone has not been effective as cancer therapy. Since cancer progression results from a complex network of interdependent biological events, preventing metastasis cannot be understood by studying only one gene or protein at a time. In this thesis, we employed mathematical and statistical models to examine complexity in the CXCL12/CXCR4/CXCR7 signaling axis. First, we performed a comprehensive analysis of CXCL12 isoform expression in breast cancer. This is the first study to correlate the expression levels of all six CXCL12 isoforms to cancer survival outcomes. Second, to understand mechanisms of physiological gradient formation, we built a hybrid agent-based model of cancer cell chemotaxis that links molecular scale events to chemokine gradient shaping and sensing. Third, to understand how co-expression of CXCR7 may alter CXCR4 signaling, we constructed a mechanistic model of CXCR4/CXCR7 receptor dynamics and signaling with an emphasis on shared signaling components. Themes arising from this work include the importance of non-specific binding of ligand to surfaces, receptor desensitization, gradient sensing, and compensatory effects resulting from the competition of shared signaling components.