Network Analysis of Gene Expression Kinetics in Human Breast Cancer Cells
May 05, 2006 04:00 PM to 05:00 PM
Speakers:
Ethier, Steve
|
 |
Abstract: |
Experiments were designed to examine the gene expression networks regulated by the
epidermal growth factor receptor (EGFR) in the normal human mammary epithelial cell line MCF-
10A, and in the human breast cancer cell line, SUM-149. These cell lines were chosen due to their
absolute dependence on EGFR signaling for growth in vitro. To generate gene expression
networks from the microarray data, we used a simple dynamic model to analyze the time response
of the system following the inhibition of EGFR activity using small molecule tyrosine kinase
inhibitors specific for EGFR. The model is based upon a linear finite difference equation in which
the estimation of a transition matrix led us to establish the topology of the underlying network. Local
measures of connectivity like weighted connectivity and clustering coefficient allowed us to identify
characteristic major hubs from the breast cancer cell line. By defining a particular threshold, we
were able to convert the original weighted matrix to an adjacency matrix in which global measures
of connectivity can be applied. One of these measures, the mutuality or reciprocity, suggested a
cooperation decrease between major hubs in the cancer cell line as compared to the normal cells.
In addition, by comparing network topology between normal mammary epithelial cells and breast
cancer cells that both rely on EGFR signaling for growth and survival, we were able to identify
cancer-cell specific hub genes within the EGFR-regulated networks. Whereas some of the hub
genes, such as DUSP6 and AREG, would have been predicted based on prior knowledge of EGFR
signaling pathways in normal and neoplastic cells, many of the cancer cell specific hubs would not
have been predicted to be regulated by EGFR. Among these hub genes are IL-1A, IL-1B and
genes related to cytokine signaling and NF-kappaB activation. We confirmed that expression of
these cytokines was regulated by EGFR in the breast cancer cells and not in the normal cells, and
further demonstrated that blockade of these pathways leads to growth arrest and loss of viability in
the cancer cells but not in normal cells. Thus, kinetic analysis of the regulation of gene expression
networks in cancer cells and normal cells has the potential to identify novel features of cell
signaling and gene expression in cancer cells, which in turn, can predict novel targets for
therapeutic intervention. |
|
|
