Genetics of Cutaneous Malignant Melanoma

Yan A. Su, MD, PhD, and Jeffrey M. Trent, PhD

Laboratory of Genetics at the National Center for Human Genome Research,
National Institutes of Health, Bethesda, Md.

Introduction

The genetic dissection of a complex disease such as malignant melanoma is a daunting task. An understanding of genetic changes associated with the familial occurrence of this disease is required, as well as the ability to distinguish those changes associated with disease progression. Melanoma is a model malignancy in which to study these aspects of tumorigenesis, because (1) a clear familial segregation occurs in 5% to 10% of cases, and (2) the histopathologically defined stages of melanocytic transformation lend themselves to genetic study. A global view of this (and other) complex diseases is problematic in that malignant transformation is a multistep process.[1] Accordingly, the ability to distinguish genetic changes that are critical in disease initiation and progression from non-tumor-specific background changes presents a significant challenge. However, progress has been made in understanding key events in the pathway of melanocytic transformation.[2-4] This review presents some of the recognized genetic changes known to contribute to the malignant potential of melanoma (Fig 1), focuses on those changes involving the germline of patients with familial melanoma, and discusses changes in the tumors of sporadic melanoma patients.

Fig 1. - Diagrammatic representation of chromosomes most significantly involved in the aspects of melanocytic transformation associated with disease predisposition, progression, and suppression (human chromosomes #1, #6, and #9). Overlapping fields indicate the proposed involvement of these chromosomes in multiple steps in the malignant pathway. Multiple genes from each chromosome (and other chromosomes not represented) are involved in each aspect of tumor development.

Hereditary Predisposition to Cutaneous Malignant Melanoma

For over 150 years, the medical literature has referred to the hereditary (or familial) occurrence of malignant melanoma. However, within the past year, investigators have begun to recognize the specific genetic changes that may be germinally associated with this important subset of patients with melanoma. Current estimates suggest that true hereditary cases of melanoma represent between 5% and 10% of all new patients.[2,3] Although numerous other etiologic factors have been shown (principally epidemiologically) to contribute to melanoma risk (eg, sun exposure, hair color, number of nevi, etc.), this review focuses on changes defined by genetic linkage analysis. As the pathways of genetic alterations are identified and studied, associations will be made between those epidemiologic variables and specific genetic changes.

Human melanoma susceptibility loci have been assigned to three different human chromosomes: 1, 6, and 9. The most compelling associations of a genetic change with familial occurrence have been the report of genetic linkage to chromosome 9p21 in hereditary cases,[5] followed by the identification of a gene termed p16 (or multiple tumor suppressor 1, MTS1) that frequently is mutated in the germline of patients with hereditary melanoma.[3,6] The identification of the p16 gene as the germinal change in hereditary melanoma has been controversial. The initial report suggested that this gene would represent a genetic alteration associated with a plethora of different malignancies, thereby playing a central unifying role in the tumorigenic process.[6] However, this was followed by suggestions that the genetic changes within the p16 gene may more likely represent events associated with in vitro propagation of cells and may not be as global in significance. Mutations of the p16 gene will be shown to represent major and possibly early contributory effects to melanocytic transformation (Fig 2).[3,7]

Fig 2. - Diagrammatic representation of the steps to analyze p16 mutations in melanoma kindreds.[7] The pedigree (upper left panel) shows a familial melanoma kindred with segregation of the mutant p16 allele (determined by linkage analysis). The shading pattern in the pedigree reflects the disease status of individuals: invasive melanoma (bottom left quadrant), melanoma in situ (upper left quadrant), dysplastic nevi (lower right quadrant), and unaffected (no shading). The upper right panel represents an autoradiograph indicating the presence of a mutated DNA fragment from affected family members which was detected by single-strand conformation polymorphism analysis (a technique for identifying base pair changes in human disease). The lower right panel presents an example of the graphic output of a nucleotide sequence analysis of the mutated DNA (after polymerase chain reaction amplification and cloning). This information can be used to unequivocally identify the specific mutation within a family, which can then be rapidly confirmed in other family members by hybridization (lower left panel) by allele-specific oligonucleotide (ASO) hybridization.[7] Photograph provided by Nic Dracopoli, MD, Director of Research, Sequana Therapeutics, Inc, San Diego, Calif.

The p16 gene encodes a regulator of the cell cycle that binds to the cyclin-dependent kinase 4 (CDK4), thereby inactivating it.[6] Because cyclins are among the most important determinants of the initiation of DNA synthesis, a mutation in this gene (which serves to control cellular growth) would fail to exert its normal function as an inhibitor of cell division.

The defining feature of a gene involved in a hereditary disease is the recognition of its mutation in the germline of affected families. Defects within the p16 gene may be responsible for a significant subset of patients with hereditary melanoma.[7] However, a germline change in the p16 gene (or an apparently neutral mutation in the p16 gene) has not been shown in all patients with documented familial disease, which indicates that another gene within 9p21 or at another locus contributes to some familial cases.

Two additional chromosomal loci thought to encode melanoma susceptibility genes are 1p36 and 6p21. The suggestion from linkage analysis of the susceptibility locus on 1p36 remains controversial.[3,8] It is unclear whether factors related to clinical or histopathologic definition of patient populations have led to discrepancies between groups where opposite results of linkage analysis have been reported. Based on both population-based allele-association studies and linkage analysis, Walker et al[9] have reported a small positive lod score indicating the likely presence of a susceptibility locus on chromosome 6 in some melanoma kindreds. Whether this represents an actual gene-modulating melanoma susceptibility or some form of "HLA effect" (eg, contributing to the development of melanoma by modulating the immune system) is unknown. Others report no such association.[10]

Cytogenetic Changes in Sporadic Melanoma

We have contributed to the identification of genetic changes associated with sporadic melanoma in the identification of recurring sites of chromosome change in malignant melanoma. By analogy to the human hematopoietic malignancies, the identification of recurring sites of chromosome change in solid tumors (including melanoma) may provide important insights into the location of growth regulatory genes and may supply important diagnostic or prognostic data. While most cytogenetic analyses have been performed on advanced-stage metastatic melanomas and melanoma cell lines, a general pattern has emerged indicating a nonuniform distribution of chromosome rearrangements in this disorder.

Chromosomes most frequently altered in melanoma include chromosomes 1, 6, and 7, followed by chromosomes 9, 3, 2, 11, and 10.[11,12] Structural changes frequent this neoplasm, which is usually aneuploid and has a reported modal chromosome number ranging from 24 to more than 100 chromosomes per cell.[12] While no single chromosome abnormality defines melanoma, regions of recurrent alterations strongly implicate the location of growth regulatory genes that are important in melanoma biology (Fig 3). Among the chromosome band-regions identified by cytogenetic analysis as likely to harbor important genes are 6q16-23 and 9p21.

Fig 3. - Representative G-banded karyotype from a metastatic melanoma (auxiliary lymph node). The chromosome number is near tetraploid with multiple structural and numeric alterations. Clonal (ie, recurring and consistent) structural chromosome abnormalities in this patient's melanoma included t(3;?)(q11;?), t(7;?)(p22;?), del(9)(p22), t(10;?)(q24;?), t(1;11)(q11;p11), and del(12)(q 15) and several unidentified marker chromosomes. Photograph provided by Floyd Thompson, Director, Cytogenetics Core Service, Arizona Cancer Center, Tucson, Ariz.

Chromosome 6 alterations have been observed in the majority of metastatic melanomas, as well as in a subset of primary melanomas. Alterations involve nonreciprocal translocations or deletions with breakpoints clustering around the midproximal long arm, resulting in the apparent loss of all or part of the distal long arm of this chromosome.[13] This consistent finding (in more than 80% of cases) involving one or both homologues of 6q suggests that one or more tumor suppressor genes inhabit this location. Investigators from our laboratory[14] and others[15] have suggested that the tumorigenicity or metastasis of melanoma cell lines can be controlled by the introduction of a normal copy of human chromosome.[6] Evidence implicating chromosome 6 in the earliest stages of tumor progression has not yet been demonstrated adequately. Future studies combining the approaches of comparative genomic hybridization and fluorescence in situ hybridization should augment our knowledge in this area.

In addition to the frequent alteration of chromosome 6, chromosome 9 (especially the short arm 9p) has been suggested as a site of recurring chromosome abnormalities in both premalignant nevi and metastatic melanomas.[2,3,12] This observation helped to focus the search for a melanoma susceptibility gene within 9p21 as previously described. In addition to chromosomes 6 and 9, chromosomes frequently rearranged include 10q24-26, which may be associated with the invasive potential of melanoma.[16] However, the number of cases of early- stage lesions that have been studied successfully by chromosome analysis (and that have 10q alterations) currently is inadequate to provide convincing data. Chromosome 1, the most frequently altered chromosome in melanoma, is most often involved in translocations and/or deletions clustering around the centromere (1p11-q22).[2,12] However, chromosome 1 is altered in many other human solid tumors, and the significance of this finding (in the absence of a single defined rearrangement) is unclear.

The applications of comparative genomic hybridization and fluorescence in situ hybridization should increase our understanding of the early stages of melanocytic development, since these technologies allow the visualization of interphase nuclei independent of capturing cells in the metaphase of mitosis. As these procedures are increasingly combined with recently developed techniques for microdissecting tissue fragments from histopathologic sections, it is likely that the molecular chronology of genetic events associated with early events in melanoma progression will yield to analysis. This information should enhance our understanding of the genetic progression of cutaneous malignant melanoma and focus efforts toward the positional cloning of genes that are dysregulated by chromosomal rearrangements.

Evidence that chromosome rearrangements may portend clinical outcome in patients with metastatic melanoma is limited and uncorroborated.[17] Current long-range follow-up studies should lend further significance to this type of analysis in the future (unpublished results of authors). At present, it is impossible to directly define a biologic (or clinical) effect for a specific chromosome rearrangement in melanoma, with the possible exception of the association of 9p21 alterations with melanoma susceptibility.

Proto-oncogenes and Oncogenes

The activation of proto-oncogenes and the inactivation of tumor suppressor genes are fundamental mechanisms of tumorigenesis.[1] In normal cells, proto-oncogenes play a physiologic role in the process of cell proliferation and differentiation. Functions of known proto-oncogenes include growth factors, growth factor receptors, protein kinases, mediators of signal transduction, and transcriptional activators. The aberrant activation of a proto-oncogene may act as a cellular dominant and may confer specific growth advantage, thereby facilitating cell proliferation. The following highlights the current status of proto-oncogenes and oncogenes in the development of human melanocytic neoplasia (Fig 4).

Fig 4. - Ideogram of all human chromosomes demonstrating the location of proto-oncogenes, oncogenes, and tumor suppressor genes implicated in human malignant melanoma. Chromosomal loci were obtained from the GDB Human Genome Data Base. The abbreviations and proposed role of the majority of these growth regulatory genes are described in the text. Additional genes include MYBL: myb-like sequence; MYCL(K): myc-like sequence; and N-RASL: N-ras-like sequence.

Growth Factors and Growth Factor Receptors

Growth factors (GFs) and growth factor receptors (GFRs) are intensively studied in melanocyte biology.[18-21] Briefly, the expression pattern of GFs varies with the progression of melanocytic transformation. Normal human melanocytes, when cultured in vitro, require several GFs as well as 12-O-tetradecanoylphorbol-13-acetate (TPA) for growth. These requirements are decreased for cells derived from nevi and are unnecessary for in vitro growth of melanoma cells.[19-21] Basic fibroblast growth factor (bFGF), transforming growth factor-beta 2 (TGF-beta2), and transforming growth factor-alpha (TGF-alpha) are expressed in almost all melanoma cell lines examined, but their expression is not detected in cultured normal melanocytes. [4,19] These findings suggest that TGF-beta2, TGF-alpha, and bFGF may play important but as yet undefined roles in melanocytic transformation.

The principal growth factor with autocrine effect on melanoma cells is bFGF, which can inhibit growth by either internalization of anti-bFGF antibodies or antisense oligo-DNA targeted against bFGF mRNA.[19] The bFGF gene is located on chromosome band region 4q26-q27, a chromosomal region that is infrequently altered in melanoma.

Epidermal GFR (EGFR) may be a marker for advanced malignant melanoma.[4,18-21] Expression of EGFR associated with extra copies of chromosome 7 (encoding EGFR) has been observed in late-stage melanoma cell lines but less frequently in nevi and primary melanoma cell lines.[22] In addition, normal melanocytes may demonstrate a mitogenic response to EGFR only when they are newly expanded in culture,[21] which is believed to be the point where these cells are least differentiated. Therefore, expression of EGFR may be associated with differentiation of human melanocytes. A human EGFR analog (Xmrk) in the Xiphophorus fish has significant importance in the development of hereditary melanoma.[23] In this system, overexpression of Xmrk is necessary and sufficient for neoplastic melanocytic transformation.[23] Another EGFR-related tyrosine kinase receptor, HER-2/neu, plays a role (when overexpressed) in human mammary tumors and may be clinically relevant in human melanocytic oncogenesis.[24]

Insulin-like GF-1 (IGF-1) also has been shown to stimulate the growth of human melanocytes and melanoma cells,[21] and its transcription is detected in some melanomas. IGF-1R kinase activity was detected recently in four of five melanoma cell lines examined; in one case, high levels of activity of IGF-1R correlated with an approximately 30-fold amplification of the IGF-1R gene.[25] Other attempts to search for restriction fragment length polymorphisms on the IGF-1 and IGF-1R genes in 28 melanoma tissues found no gross alterations.[26] These data suggest that aberrant expression of IGF-1 and IGF-1R may be involved in a subset of melanomas, but specific information regarding the importance of this finding is unavailable.

The transmembrane tyrosine kinase receptor c-met is also expressed in normal melanocytes and is stimulated by an exogenous ligand hepatocyte growth factor/scatter factor (HGF/SF), which results in an increase in growth and mobility of these cells.[19] This enhancement occurs only in the presence of bFGF or mast cell GF, suggesting that synergetic effects of GFs are required for melanocyte growth. Since c-met is not constitutively expressed in melanoma cells, it may play a role in the transition of melanoma to a metastatic phenotype.[19]

Mast cell GF and its ligand c-kit, a transmembrane tyrosine kinase receptor, appear to be essential to the normal growth and distribution of melanocytes in both mice and humans.[27] Mice with a deficiency in the mast cell GF and c-kit genes have white spots and are anemic and sterile, due to a reduction in the proliferation and differentiation of melanoblastic, hematopoietic, and germ cells.[28] In normal human melanocytes, c-kit is one of the most abundant proteins. Mutations in the c-kit gene were identified as a developmental defect in the human piebald trait.[28] Expression of c-kit has not been detected in most melanoma cell lines.[27] Also, in contrast to normal melanocytes, the growth of melanoma cells that express c-kit is inhibited (rather than stimulated) by mast cell GF.[29,30] Thus, while c-kit activity is probably an important marker of progression, it is more likely to be involved in a differentiation pathway rather than a pathway leading to the genesis of melanoma.

Signal Transducers

The oncogene ras has been extensively studied in human melanomas. H-ras and K-ras mutations are rarely detected in normal or dysplastic nevi,[4] and N-ras mutations were detected in only 5% to 24% of primary and metastatic melanomas.[4] Taken together with other studies, these results suggest that ras mutations are late events in the progression of a malignant melanomas. The transfection of the Ha-ras oncogene into cultured melanocytes resulted in tumorigenicity[4] and also in the generation of an isochromosome for the short arm of chromosome 6 - or iso(6p) - as the only cytogenetic abnormality in cells examined.[4] The induction of iso(6p) is of particular interest since it provides an independent in vitro model system for study of chromosome 6 abnormalities, which occur nonrandomly in metastatic melanomas.[12]

Transcription Factors

Many cellular proto-oncogenes, including c-myb, c-myc, c-fos, and c-jun, have been studied to determine a casual relationship with melanocytic tumorigenesis.[18] Data derived from these efforts suggest that none of the currently known dominant oncogenes is frequently responsible for melanocytic transformation.

Tumor Suppressor Genes

The tumor suppressor gene p53 is critical to the control of cell growth serving as a negative regulator and is the most frequently mutated gene in human malignant disorders.[1] Analysis of p53 mutation studies in melanocytic lesions has revealed a broad and variable range of mutation rates ranging from 9% to 85%.[4,31,32] Regardless of these controversial findings, p53 protein has been detected in higher percentages in malignant melanomas than in dysplastic nevi.[4] This suggests that the involvement of p53 mutations, if any, occurs in the later stages of melanocytic progression.

As described previously, p16 (MTS1/CDKN2/INK4/CDK4I) has been associated with familial melanoma (MLM) predisposition.[6,7] The p16 protein acts as a cell-cycle inhibitor by binding and inhibiting CDKs,[33] which is a typical feature of a tumor suppressor gene. The p16 gene has been localized in human chromosome 9p21,[6] a locus responsible for MLM.[5] The homozygous deletion and intragenic mutation of the p16 gene have been observed in many tumors, including 75% of human melanoma cell lines,[6] and p16 is now believed to be a tumor suppressor gene with considerable effect on hereditary cutaneous malignant melanoma.

Conclusions

Our understanding of critical genes associated with cutaneous malignant melanoma has expanded along with our ability to identify melanoma susceptibility genes, and the potential significance for presymptomatic diagnosis is significant. An increasing awareness of the presence of suppression-related sequences, as well as the likelihood of further definition of the genetic chronology associated with disease progression, should yield important new information. The intensive and ongoing efforts to dissect the genetic changes in this disorder should enhance our understanding of fundamental changes associated with melanocytic transformation.

References

1. Vogelstein B, Kinzler KW. The multistep nature of cancer. Trends Genet. 1993;9:138- 141.

2. Kraehn GM, Schartl M, Peter RU. Human malignant melanoma: a genetic disease? Cancer. 1995;75:1228-1237.

3. Skolnick MH, Cannon-Albright LA, Kamb A. Genetic predisposition to melanoma. Eur J Cancer. 1994;30A:1991-1995.

4. Albino AP. The role of oncogenes and growth factors in progressive melanoma-genesis. Pigment Cell Res. 1992(Suppl 2):199-218.

5. Cannon-Albright LA, Goldgar DE, Meyer LJ, et al. Assignment of a locus for familial melanoma, MLM, to chromosome 9p13-p22. Science. 1992;258:1148-1152.

6. Kamb A, Gruis NA, Weaver-Feldman J, et al. A cell cycle regulator potentially involved in genesis of many tumor types. Science. 1994;264:436-440.

7. Hussussian CH, Struewing JP, Goldstein AM, et al. Germline p16 mutations in familial melanoma. Nat Genet. 1994;8:15-21.

8. Bale SJ, Dracopoli NC, Tucker MA, et al. Mapping the gene for hereditary cutaneous malignant melanoma-dysplastic nevus to chromosome 1p. N Engl J Med. 1989;320:1367- 1372.

9. Walker GJ, Nancarrow DJ, Walters MK, et al. Linkage analysis in familial melanoma kindreds to markers on chromosome 6p. Int J Cancer. 1994;59:771-775.

10. Bale SJ, Greene MH, Murray C, et al. Hereditary malignant melanoma is not linked to the HLA complex on chromosome 6. Int J Cancer. 1985;36:439-443.

11. Fountain JW, Bale SJ, Housman DE, et al. Genetics of melanoma. Cancer Surv. 1990;9:645-671.

12. Thompson FH, Emerson J, Olson S, et al. Cytogenetics in 158 patients with regional or disseminated melanoma: subset analysis of near diploid and simple karyotypes. Cancer Genet Cytogenet. 1995. In press.

13. Trent JM, Thompson FH, Meyskens FL Jr. Identification of a recurring translocation site involving chromosome 6 in human malignant melanoma. Cancer Res. 1989;49:420-423.

14. Trent JM, Stanbridge EJ, McBride HL, et al. Tumorigenicity in human melanoma cell lines controlled by introduction of human chromosome 6. Science. 1990;247:568-571.

15. You J, Miele ME, Dong C, et al. Suppression of human melanoma metastasis by introduction of chromosome 6 may be partially due to inhibition of motility, but not to inhibition of invasion. Biochem Biophys Res Commun. 1995;208:476-484.

16. Parmiter AH, Balaban G, Clark WH Jr, et al. Possible involvement of the chromosome region 10q24-q26 in early stages of melanocytic neoplasia. Cancer Genet Cytogenet. 1988;30:313-317.

17. Trent JM, Meyskens FL, Salmon SE, et al. Relation of cytogenetic abnormalities and clinical outcome in metastatic melanoma. N Engl J Med. 1990;322:1508-1511.

18. Bennett DC. Genetics, development, and malignancy of melanocytes. Int Rev Cytol. 1993;146:191-260.

19. Halaban R. Growth regulation in normal and malignant melanocytes. Recent Results Cancer Res. 1993;128:133-150.

20. Rodeck U. Growth factor independence and growth regulatory pathways in human melanoma development. Cancer Metastasis Rev. 1993;12:219-226.

21. Shih EM, Herlyn M. Role of growth factors and their receptors in the development and progression of melanoma. J Invest Dermatol. 1993;100(Suppl 2):196S-203S.

22. Koprowski H, Herlyn M, Balaban G, et al. Expression of the receptor for epidermal growth factor correlates with increased dosage of chromosome 7 in malignant melanoma. Somat Cell Mol Genet. 1985;11:297-302.

23. Willbrodt J, Adam D, Malitschek B, et al. Novel putative receptor tyrosine kinase encoded by the melanoma-inducing Tu locus in Xiphophorus. Nature. 1989;341:415-421.

24. Peris K, Cerroni L, Chimenti S, et al. Proto-oncogene expression in dermal naevi and melanomas. Arch Dermatol Res. 1991;283:500-505.

25. Zhang J, Trent JM, Meltzer PS. Rapid isolation and characterization of amplified DNA by chromosome microdissection: identification of IGF1R amplification in malignant melanoma. Oncogene. 1993;8:2827-2831.

26. Linnenbach AH, Huebner K, Reddy EP, et al. Structural alteration in the MYB protooncogene and deletion within the gene encoding alpha-type protein kinase C in human melanoma cell lines. Proc Natl Acad Sci U S A. 1988;85:74-78.

27. Natali PG, Nicotra MR, Winkler AB, et al. Progression of human cutaneous melanoma is associated with loss of expression of c-kit proto-oncogene receptor. Int J Cancer. 1992;52:197-201.

28. Fleischman RA, Saltman DL, Stastny V, et al. Deletion of the c-kit protooncogene in the human developmental defect piebald trait. Proc Natl Acad Sci U S A. 1991;88:10855- 10889.

29. Funasaka Y, Boulton T, Cobb M, et al. c-kit-kinase induces a cascade of protein tyrosine phosphorylation in normal human melanocytes in response to mast cell growth factor and stimulates mitogen-activated kinase but is down regulated in melanomas. Mol Biol Cell. 1992;3:197-209.

30. Zakut R, Perlis R, Eliyahu S, et al. KIT ligand (mast cell growth factor) inhibits the growth of KIT-expressing melanoma cells. Oncogene. 1993;8:2221-2229.

31. Weiss J, Schwechheimer K, Cavenee WK, et al. Mutation and expression of the p53 gene in malignant melanoma cell lines. Int J Cancer. 1993;54:693-699.

32. Akslen LA, Mrkve O. Expression of p53 protein in cutaneous melanoma. Int J Cancer. 1992;52:13-16.

33. Serrano M, Hannon GJ, Beach D. A new regulatory motif in cell-cycle control causing specific inhibition of cycline D/CDK4. Nature. 1993;366:704-707.

Back to Cancer Control Journal Volume 2 Number 5