The past 30 years of cancer research has illuminated many signaling pathways that contribute to human cancer. However, the majority of this insight has been directed to a relatively small proportion of genes in the human genome, underscoring the fact that many genes regulating tumorigenesis have yet to be identified. Although high-throughput genomic approaches have begun to establish extensive catalogs of gene alterations in human tumors, the genetic lesions that functionally contribute to tumor genesis are often concealed by the complex chromosomal instability in cancer cell genomes. Thus, functional approaches are critical for identifying such cancer causing perturbations. My laboratory applies genome-wide RNA interference (RNAi) technologies to the unbiased discovery of cancer genes and networks. Specifically, we focus on 2 areas of cancer gene discovery.
Functional genetic screens for human tumor suppressors.
Experimental cell models of human cancer have recently been developed to examine the cooperating genetic events required to transform normal cells into malignant derivatives. In such models, primary cells from healthy human tissues are transformed into tumor cells by dysregulating known oncogenes and tumor suppressors in combination. Using these cell culture models as platforms, we are screening genome-wide RNAi libraries for candidate tumor suppressors. For instance, using a human mammary epithelial cell (HMEC) model of breast cancer, we discovered over 40 new tumor suppressor candidates that include well known tumor suppressors as well as genes with unappreciated roles in human cancer (Westbrook, et al., Cell 2005). We are applying genetic, cell biologic, and biochemical approaches to understand the functions of these genes in controlling malignant transformation.
We are also adapting new cellular models of transformation from other human cell types (ex. melanocyte) to perform parallel RNAi-based screens for tumor suppressors. These models will enable us to identify genes that suppress cellular transformation in common and distinct cell types. Such rapid, parallel screens will define tumor-type specific networks, and importantly, will uncover common pathways required for malignant conversion. Collectively, these types of approaches will provide a better understanding of the genetic landscape governing tumorigenesis and an entry point for therapeutic discovery.
The REST tumor suppressor network.
One of the novel tumor suppressors discovered in our genetic screens is the transcriptional repressor, REST (or RE1-Silencing Transcription Factor). REST is a master repressor of neuronal genes in non-neuronal tissues, thus serving as a regulator of proper cell identity. We have shown that REST is a potent suppressor of malignant transformation in human epithelial cells, and cell biologic, genomic, and signaling data supports a role for REST in human tumor suppression. More recently, we have discovered a ubiquitin ligase complex that ubiquitinates REST and controls its stability (Westbrook, et al., Nature, in press). Importantly, this ubiquitin ligase (SCFbTRCP) also antagonizes the functions of REST in neural differentiation and tumor suppression. Specifically, genetic experiments indicate that SCFbTRCP transforms human epithelial cells by triggering REST degradation. SCFbTRCP requires phosphorylation of REST to elicit its degradation, and we are currently using genetic approaches to identify the signal transduction that controls the physiologic and pathologic functions of REST. By discovering these modifiers of REST, we will establish new strategies to modulate REST tumor suppressor function in human cancers.
For more information, visit his lab webpage on BCM.edu
Cancer gene discovery
Functional genetic screens for human tumor suppressors
The REST tumor suppressor network