Print

Damsky, William

Cell and Molecular Biology Program

2012

Ph. D.

Melanoma is a major health problem both in the U.S. and worldwide. It is estimated that in the U.S. alone approximately 70,000 new cases of melanoma are diagnosed annually with nearly 9,000 deaths. Recurrent genetic alterations have been identified in melanoma over the past several years; however, the relationship of these changes to basic aspects of melanoma biology such as progression of benign nevi to melanoma and metastasis of melanoma to distant sites remain largely uncharacterized. Here, several novel mouse models of melanoma have been generated to explore the effects of such recurrent genetic changes. By using the Tyr::CreERT2 transgenic mouse strain, specific activating or inactivating mutations can be made somatically to the melanocytes of mice. In these models, mice develop melanoma precursors (nevi) and melanomas from their own endogenous melanocytes, which accurately recapitulate complex physiological microenvironmental conditions induced in response to melanocyte transformation.
By combining alterations to MAPK signaling (BrafV600E activating mutation), PI3.K/Akt signaling (Pten inactivation), Cdkn2a (the familial melanoma tumor suppressor locus), mTORC1 signaling (Lkb1 inactivation), Wnt signaling ([Beta]-catenin stabilization or inactivation), p53 pathway (p53 inactivation), and the adherens junction (E-cadherin), specific roles of individual proteins and signaling pathways in melanoma formation and progression were dissected. Further, characterization of pathway interactions was also possible. First, in a mouse model based on melanocyte-specific Cdkn2a inactivation combined with BrafV600E mutation, mice form growth arrested melanocytic nevi, which progress at low rates to melanoma as mice age. Through characterization of this model, inhibition of mTORC1 signaling was identified as key functional mediator of BrafV600E-induced melanocytic nevus growth arrest. mTORCI reactivation was found to allow subsequent progression of nevi to melanoma.
Using this mouse model, we propose a new paradigm for understanding BrafV600E-induced growth arrest in vivo and term this process oncogene-induced metabolic growth arrest. Second, using a mouse model based on melanocyte specific BrafV600E mutation combined with Pten inactivation, we identify Wnt signaling though [Beta]-catenin as a key mediator of melanoma metastasis in vivo. While inactivation of [Beta]-catenin in melanomas results in a vast reduction in metastasis, stabilization of [Beta]-catenin results in a dramatic enhancement of this process. The most metastatic variant of the model Pten/Braf/Bcat-STA, underwent extensive characterization and comparison to human melanomas with the same genetic changes. Overall, these novel mouse models have allowed identification of fundamental pathways that underlie melanoma formation, progression, and metastasis. Further, these models represent a suite of human-relevant, experimentally-tractable melanoma models that are likely to be central to future studies of melanoma biology.