The NCI60 is a powerful tool for cancer gene discovery.

The NCI60 is a panel of 59 human cancer cell lines, established by the Developmental Therapeutics Program (DTP) of the NCI/NIH to screen compounds for anticancer activity. To date more than 100,000 compounds and natural extracts have been tested across this panel, and data on the compounds, their activity, and the cell lines themselves are organized into searchable databases. The Activity (A) Database contains data about the activity of each compound across the entire panel of cell lines. One use is to "fingerprint" the activity of a compound, then search for compounds with similar activity. The Structure (S) Database contains 2-D and 3-D descriptors of the structures of nearly 500,000 compounds. The Target (T) Database contains data on the cell lines. This database includes information about individual molecules as well "omic" style profiling at the DNA, mRNA, protein and functional levels. For example, array-based mRNA expression profiling data is available for each cell line. These databases make the NCI60 a powerful tool for cancer gene discovery, allowing scientists to test hypotheses and find correlations simultaneously across a multiple cell lines from a variety of cancer types. The study published by L.A. Garraway & al. in the July, 2005 issue of Nature used the information available on the NCI60 to best advantage to identify a novel cancer gene, MiTF. For more information on the NCI60 cell-based screen, see J.N. Weinstein.

MiTF is the master regulator of melanocyte development. 

MiTF, a bHLH-ZIP transcription factor, is considered to be the master regulator of melanocyte development. Cutaneous melanocytes are the pigment cells of the skin, synthesizing melanin in response to solar radiation. They must tolerate mutagenic UV radiation and the toxic process of melanin biosynthesis. Consequently, pro-survival, anti-cell-death factors play an important part in their normal function. For example, G.G. McGill & al. showed that Bcl-2 is a direct target of MiTF modulating melanocyte lineage survival and cell viability. Bcl-2 germ line deletion results in melanocyte loss in mice. J. Du & al. showed that CDK2 is specifically regulated by MITF in melanocytes, and that CDK2 depletion suppresses growth of melanoma cells. G.G. McGill & al. showed that c-Met, an oncogene involved in epithelial-mesenchymal transitions, is a direct target of MiTF, not surprising considering the neural crest origin of melanocytes.

A melanoma is a malignant melanocytic tumor. 

Figure 1. A nevus (left) and a melanoma (right).

Melanoma is characteristically drug resistant. The current standard of care is chemotherapy with cisplatin and docetaxel, two relatively nonspecific antineoplastic agents.

MiTF is selectively amplified in melanoma. 

L.A. Garraway & al. used a 100K SNP array for copy number analysis on the genomes of the NCI60 cell lines ( Figure 1a). The 100K SNP array by Affymetrix has high marker density and probe redundancy, making it ideal for identifying parts of the genome that are present in increased number. The genomes from 58 of the NCI60 were hybridized with the chip, along with a two copy (normal diploid) control. Data was processed by dChipSNP, software developed by the Cheng Li Group (Harvard School for Public Health/ Dana Farber Cancer Institute) and the Wing Wong Lab (Stanford University) for LOH and copy number analysis. The signals were summed from both alleles at each SNP locus. Hierarchical analysis clustered cell lines mainly by tissue of origin ( Figure 1b, Supplementary Figure 1). A segment of chromosome 3p was amplified in 6 of the 8 melanoma cell lines. Once the 3p amplicon was characterized, gene expression data on the NCI60 (from the T Database) allowed the identification of MiTF as the only amplified gene with increased expression.

The MiTF amplification was confirmed in samples from real human tumors. Quantitative PCR detected MiTF gene amplification in tissue samples from human melanocytic tumors. ( Figure 2a). A separate tissue array of human melanocytic tumors confirmed that MiTF is amplified in primary and metastatic melanomas, but not benign tumors. FISH detected MiTF gene copy number increase ( Figures 3a and 3b). AQUA protein analysis confirmed increased MiTF protein levels in samples with MiTF amplification ( Figures 3d, 3e and 3f). Five-year survival information was available for each sample on the tissue array, and MiTF amplification correlated with decreased 5-year survival ( Figure 3c).

MiTF promotes tumors in conjunction with BRAF mutation and p16 pathway inactivation. 

The six NCI60 melanoma lines with the MiTF amplification also have mutated BRAF and inactivation of the p16 pathway. To test whether MiTF can transform melanocytes in cooperation with oncogenic BRAF, MiTF overexpression studies were done in modified melanocytes that had both the p16 and p53 pathways inactivated. Transfection of MiTF or oncogenic BRAF alone had little or no effect on growth factor requirements, while co-transfection conferred robust growth factor independence ( Figure 4a). Only the cells co-transfected with oncogenic BRAF and MiTF were able to form anchorage-independent colonies in soft agar ( Figures 4c and 4d).

Next, the authors demonstrated the dependence of some melanoma lines on MITF. Dominant-negative mutants of MiTF (dnMiTF) have been identified in genetic screens of mice. Transfection with dnMiTF inhibited growth in three of the NCI60 melanoma cell lines, MALME-3, SKMEL-5 and UACC-257 ( Figure 4e).

MiTF knockdown sensitizes MALME-3 melanoma cells to chemotherapeutic drugs.

Finally, the authors investigated a possible role for MiTF in the drug resistance characteristic of melanoma. A supervised analysis of available pharmacological data on the entire NCI60 found that amplifications in 3p were associated with a significant decrease in sensitivity to 270 compounds, where 0-75 compounds would be expected at random ( Supplementary Figures 5a and 5b). This is not particularly informative, because it is already known that 3p is selectively amplified in melanoma cell lines, and that melanoma is characteristically drug resistant. Limiting the analysis to the eight melanoma lines did not produce a statistically significant result ( Supplementary Figure 5c). Introduction of dnMiTF into MALME-3 cells, an NCI60 melanoma line with 6 copies of the MiTF gene, resulted in a four- to five-fold sensitization to cisplatin and docetaxel Figure 4F, Supplementary Figure 5d).

MiTF, BRAF and the p16 pathway are intimately involved in melanocyte function and melanocytic tumor development.

Mutations in BRAF mutations are common, occurring in at least 50% of both nevi and melanomas. M.R. James & al. showed that certain polymorphisms in the BRAF gene predispose to melanoma. P.T. Wan & al. showed that the most common oncogenic form of BRAF, with a valine substituted for a glutamic acid at position 600, is also one of the most active mutants, its in vitro kinase activity being 500 fold greater than that of wild type BRAF. BRAF is a protein kinase component of the MAP kinase signaling pathway that functions downstream effector of Ras and upstream of MEK and ERK. The MAP kinases, and hence activated BRAF, promote advancement of the cell cycle through cyclin D1.

Another common genetic change in melanoma is p16(Ink4a) loss. p16 is a tumor suppressor gene; its activation forces the Rb tumor suppressor protein into a hypophosphorylated state, thereby exerting inhibitory effects on cell cycle progression through cyclin D1.

A simplified model of the progression of melanoma involving BRAF and p16 mutation was illustrated in a 2003 Oncogene review by D.C. Bennett.

Figure 2. Bennett's model of the progression of melanoma.

More recent work by C. Michaloglou & al. supported a model for the nevus growth pattern involving BRAF mutation and subsequent p16 activation.

A variety of studies link MiTF with the p16 or MAP kinase pathways. A.E. Loercher & al. showed that p16(Ink4a) is a direct target of MiTF that is required for cell cycle exit and efficient melanocyte differentiation. S. Carreira & al. showed that MiTF cooperates with p16 effector Rb1 to activate p21Cip1 expression, regulating cell cycle progression. M. Wu & al. and Xu & al. showed that dual MiTF phosphorylations by ERK and p90 Rsk-1 activate its transcriptional activity while simultaneously targeting it for proteasomal degradation. D.M. Molina & al. characterized an ERK binding domain on MiTF, and demonstrated that ERK1 and ERK2 form stable complexes with MiTF that promote its phosphorylation. In this context, it is not surprising that L.A. Garraway & al. found that MiTF on its own has no ability to transform, but but it is transforming in conjunction with oncogenic BRAF and p16 and p53 pathway inactivation.

These data may all be explained by modifying Bennett's model of the progression of melanoma.
1. In a melanocyte, MiTF is expressed, and its targets control cell growth in both a positive (Bcl-2, CDK2) and negative (p16) way.
2. Chronic exposure to mutagenic UV radiation inappropriately activates the MAP kinase pathway, most often by an activating mutation in BRAF.
3. The melanocyte begins dividing more quickly, forming a clonal nevus.
4. Tumor suppressors cause the characteristic growth arrest.
5. Over the course of years or decades, tumor suppressor gene function is lost.
6. Loss of tumor suppressor genes tips the balance of MiTF targets toward pro-proliferative signals, so MiTF gene amplification may be transforming, may be selectively favored, and may contribute to disease progression.

REFERENCES

Bennett, D.C., Human melanocyte senescence and melanoma susceptibility genes. Oncogene, 2003. 22(20):3063-3069.

Carreira S., et al., Mitf cooperates with Rb1 and activates p21Cip1 expression to regulate cell cycle progression. Nature, 2005. 433(7027):764-9.

Du, J., et al., Critical role of CDK2 for melanoma growth linked to its melanocyte-specific transcriptional regulation by MITF. Cancer Cell, 2004. 6(6):565-76.

Garraway, L.A., et al., Integrative genomic analyses identify MITF as a lineage survival oncogene amplified in malignant melanoma. Nature, 2005. 436(7047):117-22.

James, M.R., et al., BRAF polymorphisms and risk of melanocytic neoplasia. J Invest Dermatol. 2005. 125(6):1252-8.

Lin M., et al., dChipSNP: significance curve and clustering of SNP-array-based loss-of-heterozygosity data. Bioinformatics, 2004. 20(8):1233-40.

Loercher A.E., et al., MITF links differentiation with cell cycle arrest in melanocytes by transcriptional activation of INK4A. J Cell Biol, 2005. 168(1):35-40.

McGill, G.G., et al., Bcl2 regulation by the melanocyte master regulator Mitf modulates lineage survival and melanoma cell viability. Cell, 2002. 109(6):707-18.

McGill, G.G., et al., c-met expression is regulated by mitf in the melanocyte lineage. J Biol Chem, 2006. [Epub ahead of print]

Michaloglou C., et al., BRAFE600-associated senescence-like cell cycle arrest of human naevi. Nature, 2005. 436(7051):720-4.

Molina D.M., Grewal S., and Bardwell L., Characterization of an ERK-binding domain in microphthalmia-associated transcription factor and differential inhibition of ERK2-mediated substrate phosphorylation. J Biol Chem. 2005. 280(51):42051-60.

Wan, P.T., et al., Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell, 2004. 116(6):855-67.

Weinstein, J.N., Linking Drugs and Genes: Pharmacogenomics, Pharmacoproteomics, Bioinformatics, and the NCI-60, in Oncogenomics: Molecular Approaches to Cancer, C. Brenner and D. Duggan, Editors. 2004, John Wiley & Sons, Inc.: Hoboken, NJ. p. 117-137.

Wu M., et al., c-Kit triggers dual phosphorylations, which couple activation and degradation of the essential melanocyte factor Mi. Genes Dev, 2000. 14(3):301-12.

Xu, W., et al., Regulation of microphthalmia-associated transcription factor MITF protein levels by association with the ubiquitin-conjugating enzyme hUBC9. Exp Cell Res, 2000. 255(2):135-43.