The genetic evolution of acral melanoma.

Acral melanoma is an aggressive type of melanoma with unknown origins, arising on the sole, palm, or nail apparatus. It is the most common type of melanoma in individuals with dark skin and is notoriously challenging to treat. Our study examined exome sequencing data from 139 tissue samples, spanning different progression stages, collected from 37 patients. We found that 78.4% of the melanomas displayed one or more clustered copy number transitions with focal amplifications, recurring predominantly on chromosomes 5, 11, 12, and 22. These genomic "hailstorms" were typically shared across all progression stages within individual patients. Genetic alterations known to activate TERT also arose early. By contrast, mutations in the MAP-kinase pathway appeared later during progression, often leading to different tumor areas harboring non-overlapping driver mutations. We conclude that the evolutionary trajectories of acral melanomas substantially diverge from those of melanomas on sun-exposed skin, where MAP-kinase pathway activation initiates the neoplastic cascade followed by immortalization later. The punctuated formation of hailstorms, paired with early TERT activation, suggests a unique mutational mechanism underlying the origins of acral melanoma. Our findings highlight an essential role for telomerase, likely in re-stabilizing tumor genomes after hailstorms have initiated the tumors. The marked genetic heterogeneity, in particular of MAP-kinase pathway drivers, may partly explain the limited success of targeted and other therapies in treating this melanoma subtype.

P38 Mediates Tumor Suppression through Reduced Autophagy and Actin Cytoskeleton Changes in NRAS-Mutant Melanoma.

Hotspot mutations in the NRAS gene are causative genetic events associated with the development of melanoma. Currently, there are no FDA-approved drugs directly targeting NRAS mutations. Previously, we showed that p38 acts as a tumor suppressor in vitro and in vivo with respect to NRAS-mutant melanoma. We observed that because of p38 activation through treatment with the protein synthesis inhibitor, anisomycin leads to a transient upregulation of several targets of the cAMP pathway, representing a stressed cancer cell state that is often observed by therapeutic doses of MAPK inhibitors in melanoma patients. Meanwhile, genetically induced p38 or its stable transduction leads to a distinct cellular transcriptional state. Contrary to previous work showing an association of invasiveness with high p38 levels in BRAF-mutated melanoma, there was no correlation of p38 expression with NRAS-mutant melanoma invasion, highlighting the difference in BRAF and NRAS-driven melanomas. Although the role of p38 has been reported to be that of both tumor suppressor and oncogene, we show here that p38 specifically plays the role of a tumor suppressor in NRAS-mutant melanoma. Both the transient and stable activation of p38 elicits phosphorylation of mTOR, reported to be a master switch in regulating autophagy. Indeed, we observed a correlation between elevated levels of phosphorylated mTOR and a reduction in LC3 conversion (LCII/LCI), indicative of suppressed autophagy. Furthermore, a reduction in actin intensity in p38-high cells strongly suggests a role of mTOR in regulating actin and a remodeling in the NRAS-mutant melanoma cells. Therefore, p38 plays a tumor suppressive role in NRAS-mutant melanomas at least partially through the mechanism of mTOR upregulation, suppressed autophagy, and reduced actin polymerization. One or more combinations of MEK inhibitors with either anisomycin, rapamycin, chloroquine/bafilomycin, and cytochalasin modulate p38 activation, mTOR phosphorylation, autophagy, and actin polymerization, respectively, and they may provide an alternate route to targeting NRAS-mutant melanoma.

Integrated genomic analyses of acral and mucosal melanomas nominate novel driver genes.

BACKGROUND: Acral and mucosal melanomas are aggressive subtypes of melanoma, which have a significantly lower burden of somatic mutations than cutaneous melanomas, but more frequent copy number variations, focused gene amplifications, and structural alterations. The landscapes of their genomic alterations remain to be fully characterized. METHODS: We compiled sequencing data of 240 human acral and mucosal melanoma samples from 11 previously published studies and applied a uniform pipeline to call tumor cell content, ploidy, somatic and germline mutations, as well as CNVs, LOH, and SVs. We identified genes that are significantly mutated or recurrently affected by CNVs and implicated in oncogenesis. We further examined the difference in the frequency of recurrent pathogenic alterations between the two melanoma subtypes, correlation between pathogenic alterations, and their association with clinical features. RESULTS: We nominated PTPRJ, mutated and homozygously deleted in 3.8% (9/240) and 0.8% (2/240) of samples, respectively, as a probable tumor suppressor gene, and FER and SKP2, amplified in 3.8% and 11.7% of samples, respectively, as probable oncogenes. We further identified a long tail of infrequent pathogenic alterations, involving genes such as CIC and LZTR1. Pathogenic germline mutations were observed on MITF, PTEN, ATM, and PRKN. We found BRAF V600E mutations in acral melanomas with fewer structural variations, suggesting that they are distinct and related to cutaneous melanomas. Amplifications of PAK1 and GAB2 were more commonly observed in acral melanomas, whereas SF3B1 R625 codon mutations were unique to mucosal melanomas (12.9%). Amplifications at 11q13-14 were frequently accompanied by fusion to a region on chromosome 6q12, revealing a recurrent novel structural rearrangement whose role remains to be elucidated. CONCLUSIONS: Our meta-analysis expands the catalog of driver mutations in acral and mucosal melanomas, sheds new light on their pathogenesis and broadens the catalog of therapeutic targets for these difficult-to-treat cancers.

NRASQ61K melanoma tumor formation is reduced by p38-MAPK14 activation in zebrafish models and NRAS-mutated human melanoma cells.

Oncogenic BRAF and NRAS mutations drive human melanoma initiation. We used transgenic zebrafish to model NRAS-mutant melanoma, and the rapid tumor onset allowed us to study candidate tumor suppressors. We identified P38α-MAPK14 as a potential tumor suppressor in The Cancer Genome Atlas melanoma cohort of NRAS-mutant melanomas, and overexpression significantly increased the time to tumor onset in transgenic zebrafish with NRAS-driven melanoma. Pharmacological activation of P38α-MAPK14 using anisomycin reduced in vitro viability of melanoma cultures, which we confirmed by stable overexpression of p38α. We observed that the viability of MEK inhibitor resistant melanoma cells could be reduced by combined treatment of anisomycin and MEK inhibition. Our study demonstrates that activating the p38α-MAPK14 pathway in the presence of oncogenic NRAS abrogates melanoma in vitro and in vivo. SIGNIFICANCE: The significance of our study is in the accountability of NRAS mutations in melanoma. We demonstrate here that activation of p38α-MAPK14 pathway can abrogate NRAS-mutant melanoma which is contrary to the previously published role of p38α-MAPK14 pathway in BRAF mutant melanoma. These results implicate that BRAF and NRAS-mutant melanoma may not be identical biologically. We also demonstrate the translational benefit of our study by using a small molecule compound-anisomycin (already in use for other diseases in clinical trials) to activate p38α-MAPK14 pathway.

MYSM1/2A-DUB is an epigenetic regulator in human melanoma and contributes to tumor cell growth.

Histone modifying enzymes, such as histone deacetylases (HDACs) and polycomb repressive complex (PRC) components, have been implicated in regulating tumor growth, epithelial-mesenchymal transition, tumor stem cell maintenance, or repression of tumor suppressor genes - and may be promising targets for combination therapies of melanoma and other cancers. According to recent findings, the histone H2A deubiquitinase 2A-DUB/Mysm1 interacts with the p53-axis in hematopoiesis and tissue differentiation in mice, in part by modulating DNA-damage responses in stem cell and progenitor compartments. Based on the identification of alterations in skin pigmentation and melanocyte specification in Mysm1-deficient mice, we hypothesized that MYSM1 may be involved in melanoma formation. In human melanoma samples, expression of MYSM1 was increased compared with normal skin melanocytes and nevi and co-localized with melanocyte markers such as Melan-A and c-KIT. Similarly, in melanoma cell lines A375 and SK-MEL-28 and in murine skin, expression of the deubiquitinase was detectable at the mRNA and protein level that was inducible by growth factor signals and UVB exposure, respectively. Upon stable silencing of MYSM1 in A375 and SK-MEL-28 melanoma cells by lentivirally-mediated shRNA expression, survival and proliferation were significantly reduced in five MYSM1 shRNA cell lines analyzed compared with control cells. In addition, MYSM1-silenced melanoma cells proliferated less well in softagar assays. In context with our finding that MYSM1 bound to the c-MET promoter region in close vicinity to PAX3 in melanoma cells, our data indicate that MYSM1 is an epigenetic regulator of melanoma growth and potentially promising new target for tumor therapy.