Tracing the cellular origin of BrafV600E-induced cutaneous melanoma

Identification of the cancer cell of origin is a critical step towards earlier detection of malignancies, better prediction of tumor behavior and development of preventive therapies. It is unresolved as to whether cutaneous melanoma, the deadliest form of all skin cancers, originates from a resident adult melanocyte stem cell (MSC) or from a more differentiated melanocyte. Here, we used lineage-tracing approaches and mouse genetics to identify the cell of origin of BRAFV600E-induced melanoma and the compartments in which they reside. Whereas differentiated melanocytes located in the lower, bulbar region of hair follicles (HFs) were highly tumorigenic, amelanotic bulge MSCs were not. Because the vast majority of human melanomas are not thought to arise in HFs, we also induced the BRafV600E-driven melanomagenic program in interfollicular melanocytes located in mouse tails. By doing so, we created a model that faithfully recapitulates key histopathological features of human melanomagenesis. Similarly to bulge MSCs, amelanotic epidermal interfollicular melanocytes, which we show reside in the interscale hinges, did not efficiently form tumors. In contrast, differentiated melanin-producing interfollicular melanocytes were fully competent to initiate melanoma development. Virtually all targeted mature melanocytes clonally expanded and functioned as an equipotent population of tumor initiating cells even when transiting from the epidermis to the dermis, a step in which tumor-initiating cells undergo profound transcriptome reprogramming. These studies identify varying tumor susceptibilities among the different melanocytic lineage subpopulations in the mouse skin, unexpectedly highlighting mature pigment-producing cells as the melanoma cell of origin. Critically, this work provides direct in vivo evidence that fully differentiated somatic cells can be reprogrammed into cancer initiating cells.status: publishe