Academic Background

Brian Lewis received his B.S. in Biology from the University of California, Los Angeles in 1991, and his Ph.D. from Johns Hopkins University in 1997. He performed postdoctoral studies at the NIH and Memorial Sloan-Kettering Cancer Center, supported by a Helen Hay Whitney Foundation postdoctoral fellowship. He is currently supported by a Burroughs Wellcome Fund Career Development Award in the Biomedical Sciences. Dr. Lewis joined the Program in Gene Function and Expression at the University of Massachusetts Medical School as an Assistant Professor in the winter of 2003.

Molecular Genetics of Pancreatic and Liver Cancers

Human solid tumors are marked by the presence of multiple genetic and epigenetic alterations. Through the analysis of primary tumor specimens, many of the genetic alterations that occur within particular tumor types have been identified. However, the specific molecular events that occur downstream of these genetic changes, and the mechanisms by which they influence tumor initiation, progression, and metastasis are still poorly understood. The recent success of targeted cancer therapies, particularly the success of Imatinib (Gleevec), demonstrates than an understanding of the underlying molecular defect(s) can lead to the design of rational, effective therapies. Therefore, one of the overriding goals of our lab is to identify correlations between specific genetic changes, tumor behavior, and signal transduction pathways.

We utilize novel mouse models generated with the RCAS-TVA gene delivery system (Fisher et al., 1999). RCAS retroviral vectors are based on an avian retrovirus, ALV, whose entry into target cells is mediated by its receptor, TVA. TVA is an avian gene and is not present in the genomes of mammals, such as humans and mice. Therefore, specific tissues within mice can be rendered susceptible to infection by RCAS viruses through the tissue-specific (e.g. pancreas-specific) transgene expression of TVA. We have utilized this technology to generate models for pancreatic and hepatic carcinomas.

Pancreatic Cancer

The pancreas is composed of three major compartments: the acinar cells that produce digestive enzymes; the ducts that conduct these enzymes to the gut; and the endocrine cells that produce hormones involved in glucose homeostasis. In humans, tumors derived from each of these compartments have been identified. Furthermore, the genetic alterations identified in each of these tumor types differ. To study these tumor types we have generated transgenic animals that express TVA in each of these compartments.

Using mice expressing TVA in acinar cells we have demonstrated that the histologic type of the tumor induced is dependent on the inciting oncogene. Polyoma midde T antigen (PyMT) induces acinar and ductal tumors. In contrast, c-myc induces pancreatic endocrine tumors. Furthermore, the tumors express the protein Pdx1, a transcription factor that identifies early pancreatic progenitor cells. Our findings suggest that, at least in this model, the pancreatic tumor type is dependent on the intiating genetic lesions, and that a pancreatic progenitor cell may be the target for oncogenic transformation.

Ongoing studies in the lab are aimed at:

1. identifying the nature of the transformed cell;
2. identifying the underlying molecular changes that occur in these tumors;
3. utilizing transgenic animals that express TVA in the ductal or endocrine compartments of the pancrease to identify the effects of oncogene expression in the cell lineages.

Hepatocellular Carcinoma

Hepatocellular carcinoma afflicts over 250,000 people annually worldwide. Most cases are associated with chronic hepatitis virus infection or causes of chronic liver damage. Common genetic events associated with the disease include loss of the p53 tumor suppressor gene.

We have generated transgenic animals that express TVA specifically within the liver. Delivery of RCAS viruses encoding PyMT induces tumors with high penetrance. However, these tumors do not invade or metastasize to distant sites. However, if the tumors are induced in mice deficient for the p53 tumor suppressor, the tumors induced are metastatic to the lungs. Therefore, the absence of p53 does not appear to affect tumor initiation in the liver, but rather impacts tumor progression and metastasis. We have intiated gene expression microarray analyses to identify genes that mediate these effects. Preliminary data from our lab suggest that a similar effect is obtained through the deletion of the locus encoding the p16 and p19 tumor suppressor proteins.

Current work in the lab includes:

1. expansion of the gene expression profiling studies;
2. evaluation of the roles of some of the candidate genes identified by the expression arrays in tumor metastasis;
3. elucidation of the cooperating genetic alterations required for tumor induction by c-myc.

Please also see the UMass Pancreas Program website.