mcmahon@cc.ucsf.edu
(415) 502-1317 (lab); (415) 502-3179 (fax)

1450 3rd St., MC 0128; PO Box 589001
San Francisco, CA 94158-9001

deliveries: 1450 3rd Street, HD-340; San Francisco, CA 94158

Full Biosketch

Roch-Philippe Charles, PhD; Eric Collisson, MD; Stephan Gysin, PhD; Takashi Hirano; Vicky Marsh, PhD; Anny Shai, PhD; Christy Trejo, BS

Work in my laboratory focuses on the role of RAS-activated signal transduction pathways in the aberrant physiology of melanoma, pancreas and lung cancer using in vivo mouse model systems and in vitro cell culture based systems. The RAS-family of GTP binding proteins transmit signals into the interior of the cell via activation of a number of cytosolic signal transduction pathways. Prominent among these is the RAF-MEK-ERK MAP kinase and the PI3'-kinase-PDK-AKT signaling pathways. Components of both of these pathways are linked to the initiation and progression of human cancer both by their association with human RAS oncogenes and by evidence of somatic mutations in BRAF, PIK3CA, AKT and PTEN that occur in a wide range of human malignancies.

RAS genes are mutated in 25% of all human malignancies with a striking prevalence (~85%) in pancreatic cancer. However more recently, somatic mutations of the BRAF gene were identified in a large range of human malignancies with a striking prevalence (~65%) in melanoma. Indeed, mutation of BRAF is the earliest and most frequent somatic mutation known to occur in the sporadic form of the melanoma. Moreover, mutation or epigenetic silencing of PTEN or mutational activation of PIK3CA or AKT have also been demonstrated in a large number of tumor types.

Although the molecular genetics of melanoma, pancreas and lung cancer have been explored in some detail, there is a large gulf in our understanding of how mutations in oncogenes and tumor suppressors influence the aberrant behavior of cancer cells. Furthermore, there is an urgent need for new diagnostic and therapeutic tools for all three of these diseases. Consequently we are conducting a series of projects to understand the role of oncogenes and tumor suppressors in the initiation and progression of melanoma, pancreas and lung cancer.

We are taking three main approaches to explore fundamental aspects of the cell and molecular biology of these diseases:

1. Mouse models of Ras and BRaf-induced tumorigenesis
To explore the initiation and progression of cancer by BRaf in a more authentic in vivo setting, we used homologous recombination in ES cells to derive mice in which the expression of activated BRaf(V600E) is elicited by the action of Cre recombinase. By this scheme BRaf(V600E) is expressed at normal physiological levels in a cell type specific and temporally regulated manner. To do this we generated ES cells carrying a targeted allele of mouse BRaf that expresses normal BRaf (BRafCA) prior to Cre-mediated somatic recombination. However, under the influence of Cre recombinase the normal exon 15 of the BRaf gene will be replaced by a mutated exon 15 that encodes the V600E activated form of the protein kinase.

Using BRaf(CA) mice we have generated mouse models of BRaf(V600E)-induced lung cancer and melanoma. The former is published in Genes & Development and the latter is under review at Nature Genetics. We are now using BRafCA and similarly designed KRas(LSL) mice from Tyler Jacks to conduct a head-to-head comparison between oncogenic KRas(G12D) and BRaf(V600E) in the transformation of bronchioalveolar stem cells (BASC) in vivo. In addition, we are exploring the use of both BRaf(CA) and KRAS(LSL) mice as models of pre-clinical therapeutics to test inhibitors of RAF-MEK-ERK signaling either as preventatives or therapeutics to treat KRas(G12D) or BRaf(V600E)-induced malignancy. To this end we have recently obtained substantial quantities of a BRaf inhibitor and a MEK inhibitor from colleagues in the private sector.

In addition to the experiments outlined above, BRaf(CA) mice are currently being bred to other transgenic/knock-in mice in which Cre recombinase is expressed in a tissue specific manner. We are especially interested in generating a mouse model of malignant melanoma. The mice are also being bred to strains of mice in which the expression of tumor suppressor genes such as Pten, Ink4a, Arf or Trp53 can be extinguished by Cre recombinase. Initially we will breed these animals onto mouse strains in which conditional Tyr::Cre:ER(T2) is expressed either in melanoblasts (stem cells) or in mature melanocytes to explore the role of BRaf in the initiation and progression of melanoma. However, we also anticipate that such mice will be of considerable utility in the generation of mouse models of ovarian, thyroid and colon cancer, malignancies in which the activation of BRaf is directly linked to initiation and progression of the disease. Indeed, we are already collaborating with Dr. Tyler Jacks and Dr. Cory Abate-Shen on models of BRaf(V600E)-induced colon cancer and prostate cancer progression respectively. Clearly, mouse models that accurately recapitulate the features of human cancer will be very useful tools in understanding the earliest stages of the initiation and progression of these diseases and in testing new therapies to target these diseases.

More recently, we have engineered new alleles of BRaf that can be regulated by Flp recombinase as well as a latent allele of mouse Pik3ca that we expect to be very interesting in the months and years ahead. We have also established more substantive links to the pharmaceutical and biotechnology industry to obtain new inhibitors of PI3'-kinase, BRAF, AKT and others.

2. Molecular determinants of pancreatic cancer cell metastasis
Using human pancreatic cancer cell lines, we have previously conducted high throughput profiling techniques (Array CGH and Affymetrix mRNA expression analysis) to explore the genetic alterations and gene expression changes that occur in the initiation and progression of pancreatic cancer. By these techniques we identified the KRAS gene as a potential mediator of pancreatic cancer metastasis. Using an orthotopic implantation model and pancreas cancer cells labeled with Luciferase to allow in vivo imaging we are exploring the ability of KRAS and its downstream effectors to promote pancreatic cancer metastasis. In addition, we are establishing the orthotopic model as a test platform to evaluate therapeutic agents targeted against various RAS effectors to determine if this system might be used to evaluate new approaches to treat pancreatic cancer in patients.

3. Regulation of the pro-apoptotic BH3 only proteins BIM and BMF downstream of NRAS-BRAF-MEK-ERK signaling
Mutationally activated BRAF or NRAS is expressed in 90% of all melanomas. Using conditionally active forms of BRAF, we have demonstrated that BRAF-MEK-ERK signaling can suppress apoptosis. Moreover, suppression of apoptosis is accompanied by suppression of expression of both BIM and BMF, two related pro-apoptotic members of the BCL-2 family that play an essential role in melanocyte apoptosis. In these experiments we are exploring the biochemical mechanisms by which BRAF activation regulates BIM and BMF expression using cultured melanoma cells derived from our BRaf(CA) melanoma mouse model (see Aim 1) mouse Melan-a cells expressing conditionally active BRAF(V600E):ER and bona fide human melanoma derived cell lines. Although there is strong genetic and biochemical evidence for an important role for BRAF signaling in the initiation and progression of human melanoma, and for BIM and BMF in the regulation of melanocyte survival, we will test the hypothesis that these proteins are linked in a direct biochemical pathway. We believe that the proposed experiments will provide a general system for the analysis of the earliest effects of BRAF in the conversion of normal melanocytes into metastatic melanoma cells. Moreover, these studies will lead to a better understanding of the role(s) of BRAF and BIM in controlling apoptosis in mammalian cells.