Epigenetic landscapes restrict the developmental potential of differentiated cells, but their deregulation allows cancer cells to de-differentiate into transcriptionally heterogeneous cell states. This transcriptional heterogeneity and phenotypic plasticity define many cancers, including melanoma, and are thought to be the root cause of metastatic spread. Although this plasticity often correlates with cancer progression and poor prognosis in patients, it is still unclear how epigenetic changes erode developmental constraints to allow cells to acquire a spectrum of biological features as they interconvert between transcriptionally distinctive cell states. By analyzing the DNA methylation profiles of primary and metastatic cutaneous melanomas from human patients we discovered an alternative promoter of the nuclear receptor NR2F2 becomes de-methylated in metastatic melanomas. This epigenetic change allows NR2F2-isoform 2 – expressed in neural crest cells and silenced when they differentiate into melanocytes – to become re-expressed. Preliminary functional studies suggest NR2F2-isoform 2 expression is critical for metastasis but not primary melanoma growth and it affects neural crest and epithelial-to-mesenchymal transition associated gene sets. Based on these studies, we hypothesize that de-repression of NR2F2-isoform 2 elicits transcriptional changes that influence the phenotypic heterogeneity that defines melanoma cells, allowing a subset of these cells to disseminate and metastasize to distant organs. To test this hypothesis, we propose to: • inhibit or promote NR2F2-isoform 2 promoter methylation in mouse and patient-derived melanoma models to measure changes in melanoma heterogeneity and metastatic potential; • identify transcriptional networks that are influenced by NR2F2-isoform 2 and its partners; and • correlate the expression of NR2F2-isoform 2, its transcriptional co-factors and targets in primary patient samples with their metastatic recurrence after primary melanoma resection. Successful completion of this project promises to significantly impact both the melanoma and metastasis fields because it will: • provide a conceptual framework for how epigenetic de-regulation of an alternative transcription factor isoform influences the de-differentiation of melanoma cells into phenotypically heterogeneous cell states; • identify the cell state responsible for metastatic progression in early-stage tumors; and • identify prognostic markers and therapeutic targets. These will allow us to predict, prevent, or treat metastatic progression in early-stage melanoma patients, reducing morbidity and mortality in patients with recurrent melanoma, while sparing tumor-free patients from toxic side effects and cost of unnecessary interventions.

We do not understand how melanocytes de-differentiate into transcriptionally heterogeneous melanoma cell states that define poor outcome in patients. Screening for changes in DNA methylation, we discovered de- repression of NR2F2-isoform2 in patients where melanoma had spread from skin to other body parts, and now propose to determine whether and how NR2F2-isoform2 affects melanoma heterogeneity and metastatic potential. By understanding NR2F2-isoform2 function, we expect to identify the cell state responsible for metastatic progression, and factors we can harness to predict, prevent, or inhibit metastasis in melanoma patients.