MITF from control of pigmentation to melanoma risk

Massachusetts General Hospital, Boston MA

Investigators

Linked publications & trials

Abstract

Abstract Human skin phototypes are routinely described in a gradient of 1 (redhaired) to 6 (dark constitutive pigment) which correlates with UV sensitivity and risk of melanoma. Parallel to this gradient is the increase of nevi (moles) from darker to lighter phototypes, a gradient which abruptly drops at phototype 1 (redhaired) where visible nevi are unexpectedly rare. We hypothesized, and verified in murine models, that BRAFV600E- induced nevi in redhaired mice are “invisible” but actually significantly more abundant compared to genetically matched black mice. They are clinically inapparent due to lack of dark eumelanin, but are easily identified via fluorescence tagging, and associated with profoundly elevated spontaneous melanoma transformation and UVA-induced melanoma risk in redhaired mice. This melanoma risk was traced to red pigment because “albino-redhaired” mice lack all pigment and are protected from either spontaneous (previously published), UVA-induced, or peroxide-induced melanoma. Spontaneous and UVA-induced “invisible” nevi and melanoma-genesis are proposed for study in Aims 1 and 2 for NRASQ61R pigment models, the 2nd most common human nevus and melanoma oncogene, to complement our extensive unpublished data for BRAFV600E. Several small molecule approaches will be tested to potentially mitigate pheomelanin-induced melanoma risk including induction of dark eumelanin synthesis. One such example was already seen to ”reveal” visible lesions in the “invisible” nevus- bearing redhaired mice. In addition, the fluorescent nevi, engineered into isogenic pigment backgrounds, will be flow-sorted from skin and transcriptomically scrutinized to mechanistically dissect the pathways and genes (including redox) that mediate the elevated melanoma risk among fair skinned individuals. Only a minority of melanomas arise from pre-existing nevi, so we use our models to examine both nevi and overall melanoma risk. We also discovered that the Parkinson’s Disease therapy L-Dopa significantly elevates pheomelanin synthesis in redheads. Separate from being a dopamine precursor, L-Dopa is coincidentally a chemical intermediate in melanin biosynthesis. Parkinson’s Disease and melanoma have long been associated with one another, but lacking a mechanistic explanation. We observed that low dose L-Dopa significantly elevates melanoma risk in BRAFV600E redhaired, but not in albino-red mouse models (which cannot make pheomelanin or eumelanin). Aim 3 will test L-Dopa’s melanoma causality in NRASQ61R redhaired models and whether topical skin darkening may modify the enhanced melanoma risk already observed for BRAFV600E. Finally, Aim 4 will examine the hypothesis that MITF regulates expression of an anti-oxidant pathway which includes enzymes controlling regeneration of reduced NADPH and glutathione—key factors buffering melanocyte oxidative stress. One such enzyme, the mitochondrial enzyme NNT, was recently shown to modulate melanocyte redox and melanosome differentiation. We will test evidence that NNT is a transcriptional target of MITF and functionally scrutinize a newly identified sequence variant in NNT, nominated by NCI collaborators as a candidate familial melanoma gene in humans.