With over 200,000 US cases per year, prostate cancer is the most commonly diagnosed malignancy in men and the second most common cause of male cancer deaths. With the development of effective screening programs, an increasing number of men are diagnosed and treated for clinically localized or locally advanced disease. Yet, the overwhelming majority of these cases will not progress to life-threatening stages. However, a significant number of men will suffer disease recurrence and their early identification through predictive models constitutes a great challenge for therapy. We aim to define the molecular distinctions between local and metastatic disease.

The PTEN tumor suppressor pathway is among the most frequently targeted signaling cascades of human cancer. Several years ago, I discovered that the loss of a single copy of PTEN is sufficient to permit tumors to develop in animal models of cancer. We later found that complete loss of PTEN paradoxically triggers senescence, an arrested state that delays or blocks cancer development in affected cells. These findings collectively explain why many cancer biopsies only display partial loss of this tumor suppressor and offer us a novel framework for understanding how tumors are initiated and progress to lethal therapy-resistant stages.

My lab is expanding on the above findings in collaboration with clinicians at Memorial Sloan Kettering, with the aim of identifying patients who have developed tumors with metastasis-favoring mutations.

Our inter-disciplinary team is generating mouse models that accurately reflect the core genetic changes of human metastatic prostate cancer. These models are the basis for pre-clinical therapy using small molecule drugs or RNA-interference technology developed at Cold Spring Harbor. In addition, the analysis of such tumors across species allows us to filter the vast amount of human genomic tumor information for relevant changes.

The PTEN/ PI 3-Kinase enzyme system translates extracellular growth cues into intracellular signals. To achieve this, PTEN turns cell membrane phospholipids into an off-state. This prevents membrane recruitment and activation of growth promoting kinases. Receptor tyrosine kinases revert PTEN function by activating PI 3 Kinase, leading to membrane recruitment and activation of the oncogenic AKT kinase (see snapshots, left).

In spite of its plasma-membrane function, PTEN has been consistently observed in cell nuclei, but mechanism and relevance of this localization have remained unclear. We have recently resolved this paradox by demonstrating that mono-ubiquitination of PTEN is essential for import. Patients harboring a germ line mutation in a PTEN ubiquitination site suffer from inheritable Cowden’s Disease. They develop pre-cancerous lesions, which display nuclear exclusion of PTEN. Since the cytoplasmic mutant retains catalytic activity, we can conclude that PTEN nuclear import is essential for tumor suppression. These findings demonstrate how a comprehensive view of disease initiation can be attained from integration of biochemistry, cell biology and human genetics and give rise to new ideas for combatting cancer.