Skip CCR Main Navigation National Cancer Institute National Cancer Institute U.S. National Institutes of Health
CCR - For Our Staff| Home |

Our Science – Kelly Website

Kathleen Kelly, Ph.D.

Portait Photo of Kathleen Kelly
Cell and Cancer Biology Branch
Head, Signal Transduction Section
Branch Chief
Center for Cancer Research
National Cancer Institute
Building 37, Room 1068
Bethesda, MD 20892

Kelly's Video Cast

Produced and edited by Natalie Giannosa


Dr. Kelly received her Ph.D. from the University of California, Irvine. She completed her postdoctoral training in the laboratory of Philip Leder, Harvard Medical School, and she has maintained an independent research program at the NCI since 1984. Dr. Kelly's interests have focused on the genetic regulation of cell growth, cancer progression and metastasis.


Our laboratory focuses on understanding the mechanistic consequences of specific genetic alterations that lead to the development of prostate cancer (PC), especially as related to progression. Prostate cancer (PC) is the most frequently diagnosed non-cutaneous cancer in men, and although organ confined PC is highly treatable with surgery and/or radiation, metastatic disease is incurable and leads to significant morbidity and mortality. We use two complementary approaches, genetically engineered mouse models (GEMM's) and xenograft models, to address mechanistic questions concerning the origin of PC metastasis, metastatic colonization of secondary organs, and therapeutic responses. Many significant questions concerning PC are approachable with mouse models. Some advantages of GEMM's are the unlimited availability of genetically-defined tumor tissue, the ability to longitudinally investigate various stages of PC progression, and the ability to manipulate the hormone environment. A strength of the laboratory is our ability to employ a wide variety of in vivo models and imaging modalities.

We have focused upon two models, both of which represent highly frequent genetic aberrations observed in human prostate cancer. The first is an aggressive PC model that initiates from deletion in prostate epithelium of the tumor suppressors, Pten and TP53. The goals of this project are to characterize the cellular and molecular phenotypes of Pten/TP53 null, castration sensitive and resistant, tumor-propagating populations and to apply the knowledge gained toward analyzing mechanisms of human prostate cancer progression and castration resistance. Our characterization of the Pb-Cre4; Ptenfl/fl, TP53fl/fl model revealed that the amplification and plasticity of PC progenitor cells contributes to the lethal, aggressive nature of Pten/TP53 null PC. Recent findings have identified PTEN deletion and AR pathway crosstalk as significant in human and mouse castration-resistant PC (CRPC) development. We have combined castration of tumor-bearing mice with FACS-based fractionation to assay and molecularly characterize tumor propagating cell fractions. We currently are identifying and analyzing mechanisms of Pten-loss dependent survival.

The second project uses a unique model that we developed, which recapitulates the most common mutation in PC, translocation of the ERG transcription factor locus behind the TMPRSS2 promoter. TMPRSS2-ERG (T-E) translocations occur in about 50% of clinically identified prostate cancers. Although T-E translocations are believed to occur early in tumorigenesis, their role in transformation is not known. T-E translocations are not transforming alone but collaborate with additional mutations. In order to model the natural expression of T-E, we produced transgenic lines containing a recombinant bacterial artificial chromosome (BAC) with an extended TMPRSS2 promoter driving genomic ERG. The use of the natural TMPRSS2 promoter has allowed us to investigate novel aspects of T-E regulation and expression. We have found that T-E is expressed in heterogeneous prostate epithelial cell populations, including progenitor-like populations where T-E expression correlates with increased clonogenicity, and that T-E is expressed in a subpopulation of castration-resistant cells. Our goal is to determine the mechanistic, functional effects of combining mutations that are commonly found in T-E translocation positive prostate cancers. Our approach combines de novo tumorigenesis in GEM models with the use of pre-neoplastic epithelial cell cultures to analyze specific characteristics of transformation.

This page was last updated on 2/20/2013.