2013 Scientific Report

Cindy K. Miranti, Ph.D.

Cindy K. Miranti, Ph.D. Laboratory of Integrin Signaling and Tumorigenesis Dr. Miranti received her M.S. in microbiology from Colorado State University and her Ph.D. in biochemistry from Harvard Medical School. She was a postdoctoral fellow in the laboratory of Dr. Joan Brugge at ARIAD Pharmaceuticals, Cambridge, Massachusetts and in the Department of Cell Biology at Harvard Medical School. Dr. Miranti joined VARI in January 2000, where she is currently an Associate Professor. She is also an Adjunct Professor in the Department of Physiology at Michigan State University. From left: Frank, Cooper, Zarif, Berger, Nollett, Hildebrandt, Schulz, Miranti, Park Staff Penny Berger, B.S. Elly Park, Ph.D. Veronique Schulz, B.S. Students Alexis Bergsma, B.S. Jason Cooper, B.S. Amanda Erwin Sander Frank, B.A. Erin Hildebrandt, B.S. Eric Nollet, B.S. Jelani Zarif, M.S. Teacher Intern Erin Combs, M.S. 41

Van Andel Research Institute | Scientific Report Research Interests Our objective is to understand how cell adhesion and the tumor microenvironment promote prostate cancer progression and metastasis. Our work focuses on three major questions: 1) how do the androgen receptor (AR) and integrin interactions with the tumor microenvironment cooperate to promote prostate cancer bone metastasis? 2) how do oncogenes disrupt integrin signaling and prostate epithelial differentiation to promote tumorigenesis? and 3) how does the metastasis suppressor gene CD82/KAI1 regulate the tumor microenvironment to suppress prostate cancer metastasis? Our strategy is to develop cell- and animal-based models that accurately reflect the in vivo biology of human prostate cancer as observed in the clinic and use them to develop therapeutic strategies for prostate cancer. The AR/a6b1 integrin axis The human prostate gland contains basal cells which express and use integrins to adhere to laminin matrix. Basal cells do not express AR, but they differentiate into AR-expressing secretory cells that detach from matrix and lose integrin expression. In prostate cancer, the AR-expressing tumor cells retain abnormal expression of integrin a6b1. We hypothesize that abnormal cross-talk between AR and integrin a6b1 is crucial for prostate cancer development and progression to castration-resistant disease. We found that AR binds directly to the integrin a6 promoter to stimulate its transcription, while simultaneously decreasing the expression of other integrins. Control of integrin a6 expression by AR requires the fusion gene, TMPRSS2-Erg, suggesting cross-talk between AR and Erg. We discovered that AR stimulation of a6b1 expression activates a laminin-dependent survival pathway involving NF-kB/RelA activation and subsequent increased transcription of Bcl-xL. To understand the mechanisms that promote the survival of castration-resistant cancer, we screened tumors cells for NF-kB target genes whose expression depends on AR and integrin a6b1. We identified and validated BNIP3 as such a gene. BNIP3 expression is higher in castration-resistant cells and correlates with disease progression and poor patient outcome. Furthermore, loss of BNIP3 induces cell death. BNIP3 promotes mitochondrial-specific degradation through autophagy, and we hypothesize that BNIP3 promotes the emergence and survival of castration-resistant tumors by enhancing such mitophagy. The loss of Pten, which leads to enhanced PI3K signaling, occurs in 60% of advanced prostate cancers; however, PI3K inhibitors are not effective in patients. When plated on laminin to engage integrin a6b1, tumor cells were resistant to PI3K inhibition. Blocking PI3K in combination with blocking AR, integrin a6b1, RelA, or Bcl-xL resensitized the cells to such inhibition. Thus, interactions with the tumor microenvironment through AR/a6b1 is an important mechanism by which prostate tumor cells escape their reliance on PI3K signaling, and disrupting this pathway will be necessary for effectively blocking prostate cancer in vivo. Differentiation and oncogenesis The prostate cancer field is hampered by the lack of cell models that reflect in vivo events. We developed an in vitro differentiation model in which basal epithelial cells are differentiated into secretory cells that behave similarly to those in vivo. As is seen in vivo, the secretory cells are marked by their loss of integrin expression and loss of adhesion to matrix. In fact, the competency to activate AR requires the loss of matrix adhesion. Differentiation is accompanied by a dramatic increase in E-cadherin expression and increased cell-cell adhesion. At the same time, there is a switch in the basal cells from dependence on integrins and MAPK for survival, to E-cadherin and PI3K in the secretory cells. 42