Investigating microphthalmia-associated transcription factor (MITF) and growth factors as initiators of melanoma

ABSTRACT Melanoma is the deadliest type of skin cancer. It can be cured with surgical resection if detected early, so advancing methods of prevention and early detection for melanoma is critical. To model human melanoma, the Zon lab developed a zebrafish model with melanocytes expressing human BRAFV600E in a p53 null background

(BRAFV600E;p53-/-). The tumors formed with this zebrafish model genetically and histologically resemble human melanoma, and the transparency of zebrafish makes them optimal models for studying melanoma initiation. We can image melanocytes in zebrafish using Tg(mitfa:mCherry) to express mCherry fluorescent protein under the

control of the promoter for mitfa, the zebrafish ortholog of the human MITF gene. When imaging BRAFV600E;p53- /- zebrafish over time, we initially observe high areas of mitfa expression, which we call cancer precursor zones (CPZs). Some CPZs progress to form tumors, while others remain growth arrested. CPZs that form tumors

reactivate an embryonic neural crest progenitor state. Incredibly, imaging the fluorescent reporter Tg(crestin:EGFP) allows for the identification and tracking of the single cell initiating melanoma since 100% of melanocytes expressing crestin go on to form tumors. Because neural crest reactivation is the defining

transformative step for CPZs to become tumors, understanding the mechanism of this transition is crucial, and it may reveal targetable pathways to prevent melanoma. I am interested in the extracellular microenvironment’s role in the neural crest reactivation of CPZs. By transplanting CPZs, I observed that CPZ melanocytes with high

mitfa:mCherry expression appear to have lower mitfa expression following transplant. One possible cause for this is the absence of ligands known to regulate mitfa activity when melanocytes are removed from their CPZ microenvironment. Upstream pathways known to regulate mitfa gene transcription include c-Kit, Wnt, and cAMP.

I hypothesize that mitfa expression changes during neural crest reactivation due to the absence of c-Kit, Wnt, and cAMP regulatory ligands. In Aim 1, I will test this by looking at protein and RNA changes to mitfa after transplants, chemically providing CPZ melanocytes with ligands for c-Kit, Wnt, and cAMP pathways, and using

a temperature sensitive mitfa mutant. A second preliminary observation was CPZ melanocytes expressing crestin within seven days of transplant, faster than usually observed in zebrafish. Since melanocytes are transplanted in growth-factor rich Matrigel, I hypothesize that specific growth factors in Matrigel direct

melanocytes to initiate melanoma. In Aim 2, I will test this by performing proteomics and single cell RNA sequencing to identify proteins that contribute to crestin induction and their cellular sources. These proteins will be further investigated with chemical and genetic approaches to test their sufficiency for neural crest reactivation

from CPZs. Collectively, investigating the contribution of mitfa and growth factors to neural crest reactivation will advance understandings of the mechanisms underlying melanoma initiation and contribute to a broad effort to understand the drivers of this complex process in vivo.