Malignant melanoma, the most deadly form of skin cancer, is becoming more common; the incidence rate of this tumour in the US, for example, has doubled since the 1970s. Thanks to improvements in diagnosis and treatment, mortality rates are increasing only slowly, and a number of genes linked to this condition have recently been identified. There is, however, much still to learn about the genetics of melanoma development, and treatment options remain poor when compared to some other common tumours. A group led by Yardena Samuels of the National Human Genome Research Institute, NIH, Bethesda, USA has now sequenced the complete protein-coding regions – the exome – from a sample of melanomas, thus obtaining the most complete map to date of the genetic changes involved in the development of this tumour.

Samuels and her colleagues used next-generation sequencing techniques to capture and sequence exonic DNA from metastatic tumour and matched normal cells from fourteen patients diagnosed with malignant melanoma but not yet treated. About 12 Gb of sequence per sample was used to generate models of the exonic sequences of about 20,000 genes. The tumour DNA sequences were compared with databases of known mutations and data from the 1000 Genomes Project, and cross-compared with sequences from the matched normal tissue to identify putative somatic (tumour-specific) mutations. An initial list of over 5,000 possible mutations was reduced to 4,222 after further analysis; 2,589 of these were non-synonymous. The ratio of non-synonymous to synonymous changes, however, indicated that most of these were likely to be "passenger" mutations rather than mutations that drive tumour development. Somatic mutations were observed at a higher than expected frequency in a total of sixty-eight genes.

The researchers then identified somatic mutations that occurred in more than one of the 14 tumour samples studied. Not unexpectedly, the Val600Glu mutation, which is known to be common in melanoma, was found in exactly 50% of the tumours. After screening potential mutation "hotspots" in a further 153 melanoma samples, genes DCC, ZNF831 and TRRAP were identified as of particular interest. Six samples, including one commercial melanoma cell line, contained the same mutation in the TRRAP gene, which encodes the transformation/transcription domain associated protein. Knockdown of this protein in melanoma cells showed that this protein is essential for melanoma cell survival, suggesting that it is a previously unknown oncogene.

Sixteen more frequently-mutated genes were analysed using Sanger sequencing in 38 more samples. Interestingly, only one of these genes, BRAF, had previously been associated with melanoma development. The second most frequently mutated gene after BRAF in this set was GRIN2A, which was mutated in 34 of the 135 samples tested (a 25.2% mutation rate). This gene encodes the subunit of a form of ionotropic glutamate receptor containing the ligand-binding site; mutations were localised to several domains of the protein, including that site. Another gene found to be frequently mutated in these melanoma samples, PLCB4, encodes a protein that also acts in this pathway, downstream of GRIN2A, and previous work, some unpublished, has implicated other genes encoding glutamate signalling proteins in melanoma. Taken together, these results highlight the involvement of glutamate signalling in melanoma development. Antagonists of glutamate signalling have already been shown to limit tumour growth, and this work suggests further that they may prove to be effective treatments for this disease.

Reference:

Wei, X., Walia, V., Lin, J.C. & 11 others, and the NISC Comparative Sequencing Program (2011). Exome sequencing identifies GRIN2A as frequently mutated in melanoma. Nature Genetics, published online ahead of print 17 April 2011. doi:10.1038/ng.810