MITF pathway mutations in melanoma.

Microphthalmia-associated transcription factor (MITF), a key regulator of melanocyte development, belongs to the class of basic-helix ⁄ loop ⁄ helixleucine-zipper transcription factors. Microphthalmia-associated transcription factor regulates multiple genes related to melanin synthesis (e.g. TYR, DCT), control of apoptosis (e.g. BCL2) and cell cycle progression (e.g. p21, p16, CDK2, TBX2) as well as factors controlling additional important biological activities (Steingrimsson et al., 2004). In humans, genomic loss-of-function mutations of MITF are associated with 10–15% of Waardenburg Syndrome or Tietz Syndrome characterized by skin hypopigmentation, ocular pigmentation defects and deafness. Transcription factors PAX3, CREB (cAMP-reponsive element binding protein), and SOX10 play important roles in regulating MITF expression in melanocytes. Loss-offunction mutations of SOX10 and PAX3 also occur in Waardenburg Syndrome and phenocopy melanocytic phenotypes of MITF mutation ⁄ deficiency. Signal responsive and tissue restricted regulation of MITF appear to play central roles in the growth, development, differentiation, and survival of melanocytes. Recent studies have identified gain-of-function aberrations of MITF, which have led to its classification as a human oncogene. These include evidence of MITF amplification in 10–20% of melanomas (Garraway et al., 2005). Microphthalmia-associated transcription factor expression was also seen to be dysregulated by the product of the translocation EWS-ATF1, in human Clear Cell Sarcomas where enforced MITF expression could functionally rescue knockdown of EWSATF1. In addition, TFE3 is a very close relative of MITF’s, which has been shown to be genetically redundant with MITF in certain non-melanocytic settings and to participate in translocation breakpoints of several human malignancies (Renal Carcinomas and Alveolar Soft Parts Sarcoma). Translocation ⁄ fusion breakpoints represent less ambiguous oncogenic lesions than genomic amplifications, and it is also notable that TFE3, TFEB, or MITF have been seen capable of replacing one another in a variety of experimental ‘gene-replacement’ studies, suggesting all may confer oncogenic activity. Although MITF amplifications have been previously recognized, somatic mutations within MITF or genes regulating MITF expression have not been identified, until now. Cronin et al. (2009) from the laboratory of Yardena Samuels at NIH now present an important study demonstrating the presence of somatic mutations of MITF or SOX10 in human melanomas. These mutations cause amino acid substitutions in conserved residues in defined functional domains, exon skipping, or truncations. Interestingly, one of the identified MITF mutations, which was caused by a splice-site alteration, exhibits higher activation of the TYR and DCT promoters when cotransfected with SOX10. This MITF 4TD2B mutation is the first evidence of a gain-offunction mutation of MITF in human melanoma. The mutation may function to enhance protein levels by deleting exon 2B which was previously implicated in control of ubiquitin-dependent proteolysis of MITF. Cronin et al. also identify SOX10 mutations, which produce a truncated form of the protein or cause amino acid substitutions. The authors demonstrate that five somatic MITF mutations, including 4TD2B, did not activate the p21 promoter, in contrast to wildtype MITF. This finding raises the possibility that promoter-selective MITF loss-of-function somatic mutation may confer tumorigenic activity (such as diminished expression of a cell cycle negative regulator). Such activity might even be dominant-negative, if the less-active protein were to occupy the target promoters. Additionally, MITF mutant protein may increase the expression of other prooncogenic genes, like BCL2, TBX2, c-MET, or CDK2, or it may diminish prodifferentiation genes, like those involved in pigmentation, in a promoter-specific manner. A question that therefore arises is: how might MITF mutant proteins differentiate a particular promoter, for instance, DCT ⁄ TYR from p21 promoters? One mechanism could rely upon the participation of cofactor(s). Microphthalmia-associated transcription factor can interact with the histone acetyl transferases CBP ⁄ p300 to activate target genes (Price et al., 1998; Sato et al., 1997). Furthermore, it has been shown that BRG1, the ATPase subunit of the SWI ⁄SNF complex, can physically interact with MITF allowing MITF to transactivate target genes (de la Serna et al., 2006). Because appropriate regulation of MITF is important for differentiation, survival and cell cycle control, specific cofactors might confer regulated expression of different target genes which modulate the biological state of the cell. The identification of MITF transcriptional partners required for MITF activity may thus provide valuable information pertinent to understanding MITF’s activities. Microphthalmia-associated transcription factor genomic mutations occur frequently at the helix-loop-helix leucinezipper domain in the Waardenburg syndrome which participates in DNA binding and dimerization. Although only relatively few somatic mutations of MITF have thus far been identified within primary and metastatic melanomas, it may be notable that their locations more commonly affect the transactivation domain, suggesting that changes in the transcriptional activity of MITF may be sufficient to trigger secondary events required for tumorigenesis. Cronin et al. (2009) also identify somatic mutations within Sox10, an upstream transcriptional regulator of MITF. The precise modes of action of these mutations will remain to be dissected. However, the epistatic relationship between these factors in the lossof-function setting (Waardenburg Syndrome) implies that similar regulatory relationships might plausibly occur in the gain-of-function setting. These studies expand not only our amazement at the activities in which MITF engages, but also the repertoire of molecular aberrations that may contribute to malignant melanoma. Moreover, the mechanistic links to non-melanoma cancers which share genomic abnormalities in MITF-related oncogenes (TFEB and TFE3) suggest that a search for somatic Coverage on: Cronin, J.C., Wunderlich, J., Loftus, S.K. et al. (2009). Frequent mutations in the MITF pathway in melanoma. Pigment Cell Melanoma Res. 22, 435–444.