Cancer biology

The weakest link?

Article metrics

Cellular lineages are defined by master regulatory proteins that dictate their fate and ensure their survival. The dependence on such factors of tumours that are resistant to treatment may prove to be their Achilles' heel.

The pigment-producing cells in the skin — melanocytes — have a master regulator called MITF (for ‘microphthalmia-associated transcription factor’). This factor is required for committing immature cells to the melanocyte lineage during development and is intimately involved in decisions regarding cell survival, growth and specialization (differentiation). Intuitively, one might expect that MITF would fiercely maintain melanocyte integrity, and discourage any deviation towards uncontrolled growth and malignancy. However, in this issue Garraway and colleagues (see Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma)1 report that melanoma cells tend to have extra, or ‘amplified’, copies of the gene that encodes MITF, and that under certain circumstances this gene can transform human melanocytes into cancerous cells. The melanoma cells still require MITF for survival, however, and for their characteristic resistance to drugs, presenting an unexpected target for the development of future therapies.

Garraway et al.1 began by looking for alterations in the genomes of cell lines that make up a standard sample set called the NCI60 panel, which contains eight melanomas. Remarkably, although there had been no previous evidence that MITF is mutated in human cancer, the authors found that the chromosomal region containing the MITF gene (designated 3p13–3p14) was amplified in most of the melanoma cell lines. Expanding their analysis to include human tissue samples revealed that the MITF gene was also amplified (ranging from 5 to 119 copies) in about 10% of primary melanomas and up to 20% of metastatic melanomas, but not in moles (melanocytic nevi), which are considered a pre-malignant stage of some melanomas. Moreover, the amplification of MITF was significantly associated with decreased five-year survival in patients with metastatic melanoma.

MITF is an intriguing candidate for an amplified oncogene (a cancer-promoting gene), as there is compelling evidence that, in addition to its role in differentiation, it represses cell proliferation by activating the expression of inhibitors of the cell cycle2,3. One of these, p16INK4a, is a well-known melanoma tumour suppressor. How can a gene whose normal product restricts cell proliferation be amplified in growing tumours? One possible mechanism is through ordered alterations that uncouple MITF from proliferation, and perhaps also from differentiation. For example, it is likely that loss of p16INK4a (or a mutation that produces an equivalent effect) is a crucial early step in melanoma progression, and a prerequisite to MITF amplification (Fig. 1).

Figure 1: A master regulator in melanoma.

The melanocyte-specific microphthalmia-associated transcription factor (MITF) is required for the commitment of embryonic cells to the melanocyte lineage during development and controls melanocyte survival, growth and specialization. Processes shown in bold type require MITF. Committed melanocyte precursors (melanoblasts) migrate to skin, where they can differentiate into pigment-producing melanocytes, or remain as melanocyte stem cells (green) in the hair follicle. Downstream targets of MITF include differentiation factors, survival factors and cell-cycle inhibitors such as p16INK4a. In melanoma, melanocytes progress through a series of pathological stages that can include pre-malignant nevi (moles) and invasive melanoma before becoming metastatic melanoma. These stages have been associated with early mutations of BRAF (asterisk), loss of p16INK4a, and other alterations. Garraway et al.1 show that extra copies of the MITF gene (MITF+) can occur in primary and metastatic lesions. The presence of amplified MITF seems to promote resistance to conventional melanoma chemotherapy and tumour-cell survival. Surviving residual tumour tissue could be heavily populated by melanoma stem cells (shown in green, currently hypothetical), arising from normal stem cells or from more highly differentiated melanocytic cells (red arrow).

In fact, the authors go on to show that in human melanocytes that have been genetically modified so that, among other things, p16INK4a activity is blocked, MITF can transform the cells. However, this transforming activity only occurs when MITF is overexpressed in the presence of a mutated form of the BRAF protein, a vital signalling factor in melanocytes. This finding is significant, because BRAF mutations occur early in melanoma and are found in most nevi and melanomas4,5.

What advantage, then, does enhanced MITF activity, whether through amplification of its gene or another mechanism, give the aspiring melanoma cell? The contributions of MITF the oncogene are undoubtedly as complex as those of MITF the master regulator. But clues may be gleaned from the actions of its targets, notably Bcl-2, a factor that promotes cell survival6. Because they must normally endure damaging ultraviolet radiation as well as the toxicity associated with biosynthesis of the melanin pigment, cells of the melanocyte lineage are primed for enhanced survival and depend heavily on factors that thwart cell-death pathways.

Lineage-specific survival mechanisms associated with MITF may account, at least in part, for the drug resistance that characterizes melanoma. Indeed, analysis of the available NCI60 pharmacological data revealed a significant correlation between MITF copy number and chemoresistance. Furthermore, Garraway et al. found that inhibiting MITF activity in melanoma cells harbouring extra copies of the MITF gene sensitized the cells to the growth-inhibitory effects of cisplatin and docetaxel — drugs currently used to treat melanoma, albeit relatively ineffectively. Agents that target MITF, or molecules further down the activation pathway that could be more suitable drug targets, may therefore enhance the therapeutic efficacy of conventional melanoma chemotherapy. It may also prove useful to screen for the presence of MITF amplifications before selecting treatment.

The discovery of MITF amplification in melanoma also backs up the theory of a link between cancer and stem cells — the immature cells that continuously divide to produce more highly specialized progeny. Melanocyte stem cells reside in the hair follicle7, where MITF has been implicated in their self-renewal and, through its target Bcl-2, in survival8. Cancer stem cells, possessing the proliferative potential and self-renewal capacity of normal stem cells, occur in malignancies of the blood and in some solid tumours9, although they have not been seen in melanoma. Melanoma stem cells retaining the properties of melanocyte stem cells, but carrying amplified MITF and/or other key mutations, might stand a better chance of evading conventional chemotherapy by surviving as a dormant residual disease and recurring as lethal metastatic melanoma (Fig. 1).

But is MITF a double-edged sword? The amplification of MITF suggests that melanocyte stem cells' dependence on this factor for self-renewal and survival might be maintained, and even amplified, in their malignant counterparts. Over-reliance on lineage survival factors such as MITF could be a weak link in an otherwise unbreakable chain, providing an opportunity well worth exploiting therapeutically.


  1. 1

    Garraway, L. A. et al. Nature 436, 117–122 (2005).

  2. 2

    Loercher, A. E., Tank, E. M., Delston, R. B. & Harbour, J. W. J. Cell Biol. 168, 35–40 (2005).

  3. 3

    Carreira, S. et al. Nature 433, 764–769 (2005).

  4. 4

    Davies, H. et al. Nature 417, 949–954 (2002).

  5. 5

    Pollock, P. M. et al. Nature Genet. 33, 19–20 (2003).

  6. 6

    McGill, G. G. et al. Cell 109, 707–718 (2002).

  7. 7

    Nishimura, E. K. et al. Nature 416, 854–860 (2002).

  8. 8

    Nishimura, E. K., Granter, S. R. & Fisher, D. E. Science 307, 720–724 (2005).

  9. 9

    Dean, M., Fojo, T. & Bates, S. Nature Rev. Cancer 5, 275–284 (2005).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Merlino, G. The weakest link?. Nature 436, 33–35 (2005) doi:10.1038/436033a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.