CDKN2A Mutation and Deletion Status in Thin and Thick Primary Melanoma1

Vol. 6, 3511–3515, September 2000 Clinical Cancer Research 3511

Adrian R. Cachia,2,3 James O. Indsto,2 involved in the progression rather than initiation of sporadic Kathryn M. McLaren, Graham J. Mann, and malignant melanoma. Mark J. Arends INTRODUCTION Department of Tissue and Cell Pathology, Institute of Clinical Pathology and Medical Research, Westmead Hospital, Westmead, The incidence of malignant melanoma is rising in devel- New South Wales 2145, Australia [A. R. C.]; Westmead Institute for oped countries. Investigations of potential tumor suppressor Cancer Research, University of Sydney at Westmead Hospital, genes involved in melanoma formation have shown either hem- Westmead, New South Wales 2145, Australia [J. O. I., G. J. M.]; izygosity or LOH4 for DNA markers within chromosome 9p21 Department of Pathology, University Medical School, Teviot Place, in 34–86% of melanoma cell lines and tumors examined by Edinburgh EH8 9AG, Scotland [K. M. M.]; and Molecular several authors (1–7). Histopathology, University of Cambridge Department of Pathology, Addenbrooke’s Hospital, Cambridge CB2 2QQ, United Kingdom One gene within this deleted region is CDKN2A, which can [M. J. A.] produce two independent transcripts through the alternate splicing of a separate exon 1. One transcript encodes the cdk inhibitor protein, p16INK4A (p16). The p16 protein binds to and ABSTRACT blocks the catalytic activity of two cyclin D-bound, cdks, cdk4 Human melanoma cell lines and tumor tissue from and cdk6. cdk4 is able to phosphorylate the retinoblastoma familial and sporadic melanomas have frequent, nonrandom (pRb) protein, leading to the release of associated proteins that chromosomal breaks and deletions on chromosome 9p21, a activate genes necessary for progression from the G1 to the region that includes the tumor suppressor gene CDKN2A/ S-phase of the cell cycle. It has been directly demonstrated that p16INK4A. Germ-line mutations within this gene have been melanoma-associated mutant p16 proteins are defective in their observed in some familial melanoma kindreds, but somatic ability to inhibit the catalytic activity of the cyclin D1/cdk4 mutation in sporadic primary melanoma is infrequent. Thir- complex (8–12). Loss of functional p16 protein may permit ty-nine archival, paraffin-embedded, sporadic, primary excessive cdk4 activity and therefore represents loss of a regu- cutaneous malignant melanomas (20 >3-mm-thick and 19 latory constraint on progression through the cell cycle. In addi- <0.75-mm-thick cases) were examined for mutations of the tion, experiments have shown that wild-type p16 can prevent CDKN2A gene using single-strand conformational polymor- cellular transformation by H-ras and myc oncogenes (13). Thus, phism analysis and direct sequencing. No mutations were by these mechanisms, CDKN2A can function as a tumor suppressor gene (14–18). detected. Loss of heterozygosity for the 9p21 microsatellite ARF marker D9S942 was detected in 6 of 17 informative thick The mRNA for p14 is generated by transcriptional splicing of exons 2 and 3 shared by CDKN2A with an alternative These .(0.006 ؍ lesions (35%) but 0 of 18 thin lesions (P exon 1 (E1␤; Refs. 19–21). The protein p14ARF is translated in results support other studies indicating that intragenic mu- a different reading frame from p16 and acts on a completely tation is an infrequent mechanism of CDKN2A inactivation different pathway, the regulation of the p53 tumor suppressor. in primary melanoma. The finding of loss of heterozygosity Binding of p53 to hdm2 results in its translocation to the for the 9p21 microsatellite D9S942 in thick but not thin cytoplasm and degradation; p14ARF binds to hdm2, sequesters it primary melanoma suggests that deletion or inactivation of in the nucleolus, and thereby enables p53 levels to stabilize and CDKN2A or other tumor suppressor gene(s) at this locus is rise in appropriate conditions (22). We have performed mutational analysis of the CDKN2A gene by SSCP analysis on 39 cases of sporadic primary cutaneous malignant melanoma. Of the 11 studies to date reporting Received 2/25/00; revised 6/15/00; accepted 6/21/00. somatic CDKN2A mutations in melanoma, only five involved The costs of publication of this article were defrayed in part by the primary sporadic melanomas and in all of these the lesions payment of page charges. This article must therefore be hereby marked examined were either Ͼ0.75 mm in Breslow thickness (23–25) advertisement in accordance with 18 U.S.C. Section 1734 solely to or of unspecified thickness (26, 27). Breslow thickness is the indicate this fact. 1 A. R. C. was supported by a grant from the United Kingdom Cancer most powerful histopathological predictor of melanoma pro- Research Council. M. J. A. is supported by grants from the Scottish gression currently known (28). None of the studies in which Hospitals Endowment Research Trust, Association for International CDKN2A mutations were detected in primary melanoma exam- Cancer Research, and the Cancer Research Campaign. J. O. was sup- ined tumors Ͻ0.75 mm thick (23, 24). Most have been on ported by the New South Wales Cancer Council and the Melanoma Foundation of the University of Sydney. 2 The first two authors contributed equally to this work. 3 To whom requests for reprints should be addressed, at Department of Tissue and Cell Pathology, ICPMR, Westmead Hospital, Westmead 4 The abbreviations used are: LOH, loss of heterozygosity; cdk, cyclin- NSW 2145, Australia. dependent kinase; SSCP, single-strand conformational polymorphism.

Table 1 Characteristics of primary melanomas analyzed Breslow thickness Clark level Mean mitotic Tumor-infiltrating Age Male Nodular mean and count lymphocytes, Melanomas n (yr) (%) (%) (range) (mm) II III IV V (/mm2) positive (%) Thin (Ͻ0.75 mm) 19 47.2 35 0 0.7 (0.50–0.74) 6 13 0.1 75 Thick (Ͼ3 mm) 20 61.6 60 55 5.5 (3.0–11.0) 15 5 7.3 75

metastatic tumors (3, 17, 29–32). To contribute information on the relative role of CDKN2A at different stages of primary melanoma tumorigenesis, we stratified our cases into two groups according to Breslow thickness and Clark level. LOH at the CDKN2A locus was assessed to determine whether there were different rates of deletion in thin versus thick melanomas and whether these rates were in line with those reported previously for sporadic primary melanoma.

MATERIALS AND METHODS Melanoma Cases. Twenty sporadic malignant melanomas of Breslow thickness 3 mm/Clark’s level IV or thicker and 19 sporadic melanomas of Breslow thickness 0.75 mm/Clark’s level III or thinner were retrieved from the Scottish Melanoma Group archives over a 3-year period from 1993 to 1995. The main clinical and histopathological features of the two groups are summarized in Table 1. All cases had been routinely pro- cessed and paraffin embedded. Two 10-␮m sections of tumor and normal skin for control were cut from the paraffin blocks in Fig. 1 Melanoma microdissection. A, H&E-stained section of a thick each case. Using a fresh scalpel, tumor tissue in each 10-␮m melanoma. B, H&E-stained, coverslipped tissue remaining after tumor has been microdissected away from a paraffin section contiguous to the section was carefully dissected from normal tissue with the aid section in the left panel. Contamination with normal stromal cells has of a dissecting microscope. Comparison to directly adjacent been minimized by retrieving only tumor tissue from within the edge of H&E-stained sections was used to minimize cross contamina- the neoplasm, leaving the host-tumor interface behind. tion of tumor with normal tissue, which may mask mutation detection (example dissection in Fig. 1). It is estimated after light microscopic examination of the poststained, dissected slides that contamination of dissected tumor tissue by normal 1 min, annealing for 1 min, and extension at 72°C for 2 min, stromal cells was Ͻ25% for thin lesions and Ͻ10% for thick with further extension for 10 min at the termination of the lesions. reaction. Primer sequences were manually and computer- DNA Preparation and PCR. The sections were depar- matched against the CDKN2A gene sequence published by affinized by sequential exposure to 600 ␮l each of xylene/100% Okamoto et al. (33), which includes corrections of sequence ethanol/100% ethanol in 1.5-ml Eppendorf tubes with 10 min published previously by Kamb et al. (2) and Serrano et al. (14). microcentrifugation at 13,000 rpm between each stage. The Primers used were: CDKN2A exon 1 sense, 5Ј-GGGAGCAG- tissue was then incubated overnight at 55°C in 400 ␮l of lysis CATGGAGCCG- 3Ј (13); CDKN2A exon 1 antisense, 5Ј- buffer [10 mM Tris (pH 8.3), 50 mM KCl, 0.45% Tween 20, and AGTCGCCCGCCATCCCCT-3Ј (13), annealing temperature ϭ Ј 2.5 mM MgCl2] with 0.5 mg of proteinase K (Boehringer Mann- (Tann 63°C); CDKN2A exon 2a sense, 5 -CCTGGCTCT- heim, Basel, Switzerland), followed by boiling for 30 min to GACCATTCTGT- 3Ј (novel, intronic); CDKN2A exon 2a anti- ␮ Ј Ј ϭ destroy proteinase K activity. A 10–30- l DNA aliquot was sense, 5 -GGGCAGCGTCGTGCACGGGT-3 (novel; Tann then subjected to PCR in 10 ␮l of PCR buffer (Life Technolo- 54°C); CDKN2A exon 2b sense, 5Ј-AACTGCGCCGAC- gies, Inc., Gaithersburg, MD), 25 pmol of each primer, 50–75 CCCGCCACTCTCA-3Ј (novel); CDKN2A exon 2b antisense, Ј Ј ϭ nmol of MgCl2 and 3 units of Taq DNA polymerase to a final 5 -CCAGCTCCTCAGCCAGGTCCACG-3 (novel; Tann volume of 100 ␮l. Because no disease-related mutations have 62°C); CDKN2A exon 2c sense, 5Ј-CGATGCCTGGGGC- been recorded in the 11-bp coding region of exon 3, this very CGTCTGC- 3Ј (13); CDKN2A exon 2c antisense, 5Ј-GTA- Ј ϭ short sequence was not examined. Four overlapping sets of CAAATTCTCAGATCATC-3 (novel, intronic; Tann 55°C). primers were chosen to span all of exons 1 and 2, yielding PCR Products were also submitted to specific restriction enzyme products of 203 bp or less for each set, to maximize the sensi- analysis to confirm they contained only CDKN2A sequence and tivity of mutation detection by subsequent SSCP. PCR reactions not CDKN2B. were performed on an Omnigene (Hybaid, Middlesex, United Mutation Controls. DNA was prepared from peripheral Kingdom) automated thermal cycler. A preliminary 15-min hot blood leukocytes derived from familial melanoma kindreds with start was followed by 30–40 cycles of denaturation at 94°C for known germ-line mutations (34). The DNA was amplified under

identical PCR conditions to the test samples, and the products were used as positive controls for subsequent SSCP screening. SSCP Analysis. Five to 10 ␮l of each PCR product were mixed with an equal volume of denaturing solution, heat dena- tured at 95°C for 3 min, and snap chilled in ice. Samples were loaded onto nondenaturing 10% 37.5:1 MDE polyacrylamide gels (Amersham Pharmacia Biotech, Little Chalfont, Bucks, United Kingdom), containing 5% glycerol and 0.7ϫ 0.09 M Tris-Borate, 0.002 M EDTA (pH 8.0). Duplicate gel runs were performed at 25°C. Electrophoresis was in 1ϫ TBE at 25 W for 90–120 min. DNA bands were visualized using standard com- mercial silver staining protocol and reagents (Bio-Rad Labora- tories, Hemel Hempstead, Herts, United Kingdom) according to the method of Holland et al. (34). DNA Sequencing. All CDKN2A PCR products with atypical band migration on SSCP were subjected to direct fluorescent dideoxy sequencing and analyzed in a Perkin-Elmer ABI 373 sequence analyzer (Merck, Poole, United Kingdom). In Fig. 2 Comparative electrophoresis profiles of the microsatellite marker D9S942, showing LOH of a thick primary melanoma. One ␮lof addition, two randomly selected products from each of the four PCR product was analyzed on the Corbett Research Gel-Scan 2000. sets of primers that showed normal migration on SSCP were Vertical axis, signal voltage; horizontal axis, gel retention time. Arrows, sequenced as controls. allele peaks. Other peaks are PCR slippage artifacts at 2-bp intervals. A, LOH Analysis. D9S942 is a highly polymorphic micro- normal profile. B, tumor sample showing reduction in the amount of the satellite locus located within 20 kb of the CDKN2A gene (35). larger allele by 66%, indicating LOH. PCR was performed using a 5ЈHEX-labeled forward primer in conditions carefully optimized to ensure that alleles of all sizes were amplified consistently. Twenty to 100 ng of DNA from in Fig. 2), whereas no informative thin lesions showed LOH (␹2 tumor were added to 10 ␮l of a reaction mixture of 10 mM ϭ 8.1; P ϭ 0.005). Thick lesions exhibiting LOH differed from Tris-HCl (pH 10.0), 2 mM MgCl ,50mM KCl, 5 pmol of 2 thick lesions without LOH in that Clark level 5 invasion (2 of 6 primers, and 0.2 unit of Taq DNA polymerase. The PCR pro- and 2 of 11, respectively), nodularity (5 of 6 versus 6 of 11) and tocol used was 95°C for 5 min; 5 cycles of 95°C for 30 s, 48°C males (4 of 6 versus 6 of 11) appeared to be more frequent in the for 20 s, and 72°C for 20 s; followed by 40 cycles of 95°C for former group; however, none of these differences reached sig- 20 s, 48°C for 15 s, and 72°C for 5 min; with a final extension nificance. Average Breslow thickness was identical in these two phase of 72°C for 5 min. An equal volume of denaturing sample LOH-defined groups, and there were no significant differences dye was added to the PCR products; the samples were heated to in age, sex, type, mitotic count, or presence of tumor-infiltrating 95°C for 3 min and snap chilled before loading onto an 8% 29:1 lymphocytes. acrylamide:bis gel containing 8 M urea and 0.6ϫ TBE and run in 0.6ϫ TBE buffer at 900 V and 40°C in a Gel Scan-2000 DNA Analyzer (Corbett Research Laboratories, Atlanta, GA). DNA DISCUSSION extracted from normal skin was subjected to the same protocol Our results are in agreement with the findings of most to determine the normal D9S942 profile for that individual. others in that CDKN2A is mutated very infrequently in sporadic Analysis for LOH was performed using the algorithm of Cawk- primary melanoma (25–27, 29, 37–39). However, along with a well et al. (36). A change in allelic ratios of Ͼ40% in at least recent study (39), which discriminated less sharply between thin duplicate samples was assessed as LOH. and thick primary melanomas, we have shown that there is a significantly higher frequency of LOH at CDKN2A in thick melanomas. RESULTS SSCP screening for mutation detection has a false-negative As shown in Table 1, the thick (average, 5.5 mm) mela- rate of 10–20%. The CDKN2A open reading frame is comprised nomas were more frequently from older males, nodular and with of nearly 75% G or C residues, a trait that further complicates higher mitotic indices than the thin melanomas (average, 0.7 mutation detection by SSCP. However, false-negative SSCP mm). Tumor-infiltrating lymphocytes were equally frequent in assays are unlikely to have eliminated detection of all mutations the two groups. No consistent SSCP anomalies were detected in or to have obscured a true difference between the two tumor 20 thick and 19 thin melanomas examined. Mutations were groups in this study. Mutations were detected by SSCP in all readily detected in all four positive control samples by SSCP, positive controls and wild-type sequence confirmed by direct and all of the PCR products checked by direct sequencing sequencing of the products of all four primer sets in a sample of showed wild-type sequence (data not shown). tumors. Another possible explanation for reports of low muta- Seventeen of 20 thick melanomas and 18 of 19 thin mel- tion frequency is that contamination by nontumor cells may anomas were informative at the D9S942 locus. Using the criteria mask mutation detection, especially in archival material (26). for assessment of LOH cited above, 6 of 17 (35%) of the Careful microdissection of tumor from normal tissue minimizes informative thick lesions showed LOH for this marker (example but cannot eliminate this risk (Fig. 1). Given that we did not

detect any CDKN2A mutations, we cannot comment on whether ified thickness, whereas the tumors studied here were specifi- CDKN2A mutation is an early or late event in the development cally stratified by Breslow thickness. The observed difference in of melanomas in which it is present. frequency of 9p21 LOH between thick and thin melanomas is Considering alternative mechanisms to mutation for p16 suggestive of a dosage effect from loss of one CDKN2A allele or inactivation, certain factors may influence the production of of another putative tumor suppressor gene in this chromosomal functional p16 protein at a transcriptional or translational level. region in advanced primary, but not early primary, melanoma. A It has been proposed that regulatory sequences upstream of exon corollary of this conclusion is that although CDKN2A is impor- 1 may be mutated or deleted, rendering the gene inactive (18, tant in further melanoma progression, the earliest genetic in- 40). In fact, Liu et al. (41) have demonstrated a mutation in the sult(s) that permits initiation of the malignant phenotype occur CDKN2A 5Ј untranslated region that creates an absent initiation outside the 9p21 region altogether. Alternatively, because five codon and is associated with melanoma risk. Methylation of a of the six thick tumors with LOH at D9S942 were nodular but CpG island upstream of exon 1 has been shown to be associated none of the thin tumors were, we cannot exclude the possibility with transcriptional silencing of the CDKN2A gene (42). A that the different rates of LOH observed in thick and thin quantitative reduction in p16 protein due to loss of a single allele melanomas may reflect different genetic pathways involved in may also be influential in modifying cell cycle control (26, the pathogenesis of nodular versus superficial spreading or 37, 43). lentigo maligna melanoma. One reason for the discrepancy between the low rates of CDKN2A mutation and the high prevalence of loss of expression ACKNOWLEDGMENTS of the p16 protein in tissues of advanced melanoma is that the We thank Elizabeth Holland for assistance with SSCP analysis and gene may be inactivated by homozygous deletion (2, 4, 13, 16, selection of controls. 17, 25, 27, 31, 32, 39, 43, 44, 45). Although the significance of CDKN2A as a melanoma susceptibility and tumor suppressor REFERENCES locus is not in doubt, controversy regarding the importance of CDKN2A inactivation in melanoma remains. At the heart of this 1. Fountain, J. W., Karayiorgou, M., Ernstoff, M. S., Kirkwood, J. M., Vlock, D. R., Titus-Ernstoff, L., Bouchard, B., Vijayasaradhi, S., lies discordance between the high rate of homozygous deletion Houghton, A. N., Lahti, J., Kidd, V. J., Housman, D. E., and Dracopoli, in melanoma cell lines [57 and 63% in two studies (2, 44)] and N. C. Homozygous deletions within human chromosome band 9p21 in the low rate of homozygous deletion in melanoma tumors melanoma. Proc. Natl. Acad. Sci. USA, 89: 10557–10561, 1992. [0–20% in five studies (17, 27, 32, 39, 45)]. Just as mutation of 2. Kamb, A., Gruis, N. A., Weaver-Feldhaus, J., Liu, Q., Harshman, K., CDKN2A has been attributed to cell culture artifacts or the Tavtigian, S. V., Stockert, E., Day, R. S., Johnson, B. E., and Skolnick, M. H. A cell cycle regulator potentially involved in genesis of many failure to detect mutation in uncultured melanomas blamed on tumor types. Science (Washington DC), 264: 436–440, 1994. contamination by stromal cells, so have these issues been cited 3. Isshiki, K., Seng, B. A., Elder, D. E., Guerry, D., and Linnenbach, with regard to homozygous deletion. Although this issue awaits A. J. Chromosome 9 deletion in sporadic and familial melanomas in clarification, the importance of homozygous deletion as a vivo. Oncogene, 9: 1649–1653, 1994. method of CDKN2A gene inactivation may be that there is a 4. Liu, Q., Neuhausen, S., McClure, M., Frye, C., Weaver-Feldhaus, J., concomitant requirement for inactivation of another closely Gruis, N. A., Eddington, K., Allalunis-Turner, M. J., Skolnick, M. H., linked gene to permit melanoma development or progression. Fujimura, F. K., and Kamb, A. CDKN2 (MTS1) tumor suppressor gene mutations in human tumor cell lines [published erratum appears in This would favor a single multilocus deletional event over the Oncogene, 11: 2455, 1995]. Oncogene, 10: 1061–1067, 1995. simultaneous occurrence of inactivating point mutations occur- 5. Walker, G. J., Palmer, J. M., Walters, M. K., Nancarrow, D. J., ring in two or more genes (13). Our finding of no CDKN2A Parsons, P. G., and Hayward, N. K. Refined localization of the mela- mutations but LOH for D9S942 in 35% of thick melanomas noma (MLM) gene on chromosome 9p by analysis of allelic deletions. examined does not prove the necessity of contiguous deletion of Oncogene, 9: 819–824, 1994. CDKN2A and neighboring tumor suppressor gene(s) in mela- 6. Cannon-Albright, L. A., Goldgar, D. E., Meyer, L. J., Lewis, C. M., Anderson, D. E., Fountain, J. W., Hegi, M. E., Wiseman, R. W., Petty, noma progression but would certainly be consistent with it. E. M., Bale, A. E., Olopade, O. I., Diaz, M. O., Kwiatkowski, D. J., In our study, 35% of thick melanomas showed 9p21 dele- Piepkorn, M. W., Zone, J. J., and Skolnick, M. H. Assignment of a locus tions that may affect both p16 and p14ARF function, if there for familial melanoma, MLM, to chromosome 9p13–p22. Science were a dosage effect from loss of one allele, or concomitant (Washington DC), 258: 1148–1152, 1992. inactivation of the second allele by a second event such as 7. Holland, E. A., Beaton, S. C., Edwards, B. G., Kefford, R. F., and mutation, methylation, or deletion. This would be predicted to Mann, G. J. Loss of heterozygosity and homozygous deletions on 9p21–22 in melanoma. Oncogene, 9: 1361–1365, 1994. confer a greater selective advantage over a mutation affecting a 8. Ranade, K., Hussussian, C. J., Sikorski, R. S., Varmus, H. E., single gene, because both of the central cell cycle control Goldstein, A. M., Tucker, M. A., Serrano, M., Hannon, G. J., Beach, D., pathways, mediated by Rb and p53, would be affected by such and Dracopoli, N. C. Mutations associated with familial melanoma a deletion. impair p16INK4 function. Nat. Genet., 10: 114–116, 1995. In summary, our results suggest that CDKN2A mutation is 9. Marx, J. Link to hereditary melanoma brightens mood for p16 gene. rare in primary melanomas both thinner and thicker than 0.75 Science (Washington DC), 265: 1364–1365, 1994. mm, which supports the concept of a limited role, if any, for 10. Koh, J., Enders, G. H., Dynlacht, B. D., and Harlow, E. Tumour- derived p16 alleles encoding proteins defective in cell-cycle inhibition. mutation of this gene in the early development of sporadic Nature (Lond.), 375: 506–510, 1995. melanoma. Previous studies of sporadic primary melanoma that 11. Wick, S. T., Dubay, M. M., Imanil, I., and Brizuela, L. Biochemical have reported CDKN2A mutation have all been performed on and mutagenic analysis of the melanoma tumor suppressor gene prod- melanomas either exceeding 0.75 mm in thickness or of unspec- uct/p16. Oncogene, 11: 2013–2019, 1995.

12. Reymond, A., and Brent, R. p16 proteins from melanoma-prone 31. Flores, J. F, Walker, G. J., Glendening, J. M., Haluska, F. G., families are deficient in binding to Cdk4. Oncogene, 11: 1173–1178, Castresana, J. S., Rubio, M-P., Pastorfide, G. C., Boyer, L. A., Kao, 1995. W. H., Bulyk, M. L., Barnhill, R. L., Hayward, N. K., Housman, D. E., 13. Otsuki, T., Clark, H. M., Wellmann, A., Jaffe, E. S., and Raffeld, M. and Fountain, J. W. Loss of the p16INK4a and p15INK4b genes, as well as Involvement of CDKN2 (p16INK4A/MTS1) and p15INK4B/MTS2 in human neighboring 9p21 markers, in sporadic melanoma. Cancer Res., 56: leukemias and lymphomas. Cancer Res., 55: 1436–1440, 1995. 5023–5032, 1996. 14. Serrano, M., Hannon, G. J., and Beach, D. A new regulatory motif 32. Platz, A., Ringborg, U., Lagerlof, B., Lundqvist, E., Sevigny, P., in cell-cycle control causing specific inhibition of cyclin D/CDK4. and Inganas, M. Mutational analysis of the CDKN2 gene in metastases Nature (Lond.), 366: 704–707, 1993. from patients with cutaneous malignant melanoma. Br. J. Cancer, 73: 15. Liu, L., Lassam, N. J., Slingerland, J. M., Bailey, D., Cole, D., 344–348, 1996. INK4A Jenkins, R., and Hogg, D. Germline p16 mutation and protein 33. Okamoto, A., Demetrick, D. J., Spillare, E. A., Hagiwara, K., dysfunction in a family with inherited melanoma. Oncogene, 11: 405– Hussain, S. P., Bennett, W. P., Forrester, K., Gerwin, B., Greenblatt, 412, 1995. M. S., Serrano, M., Beach, D. H., and Harris, C. C. p16INK4 mutations 16. Cordon-Cardo, C. Mutations of cell cycle regulators. Biological and and altered expression in human tumors and cell lines. Cold Spring clinical implications for human neoplasia. Am. J. Pathol., 147: 545–560, Harb. Symp. Quant. Biol., 59: 49–57, 1994. 1995. 34. Holland, E. A., Schmid, H., Kefford, R. F., and Mann, G. J. 17. Maelandsmo, G. M., Florenes, V. A., Hovig, E., Oyjord, T., Enge- CDKN2A (p16INK4A) and CDK4 mutation analysis in 131 Australian braaten, O., Holm, R., Borresen, A. L., and Fodstad, O. Involvement of melanoma probands: effect of family history and multiple primary the pRb/p16/cdk4/cyclin D1 pathway in the tumorigenesis of sporadic malignant melanomas. Br. J. Cancer, 73: 909–916, 1996. melanomas. Genes Chromosomes Cancer, 25: 339–348, 1999. 18. Kamb, A. Cell-cycle regulators and cancer. Trends Genet., 11: 35. Ohta, M., Berd, D., and Shimizu, M. Deletion mapping of chromo- 136–140, 1995. some region 9p21–p22 surrounding the CDKN2 locus in melanoma. Int. J. Cancer, 65: 762–767, 1996. 19. Mao, L., Merlo, A., Bedi, G., Shapiro, G. I., Edwards, C. D., Rollins, B. J., and Sidransky, D. A novel p16INK4A transcript. Cancer 36. Cawkwell, L., Bell, S. M., Lewis, F. A., Dixon, M. F., Taylor, G. R., Res., 55: 2995–2997, 1995. and Quirke, P. Rapid detection of allele loss in colorectal tumours using 20. Duro, D., Bernard, O., Della, V. V., Berger, R., and Larsen, C. J. A microsatellites and fluorescent DNA technology. Br. J. Cancer, 67: new type of p16INK4/MTS1 gene transcript expressed in B-cell malig- 1262–1267, 1993. nancies. Oncogene, 11: 21–29, 1995. 37. Reed, J. A., Loganzo, F., Jr., Shea, C. R., Walker, G. J., Flores, J. F., 21. Stone, S., Jiang, P., Dayananth, P., Tavtigian, S. V., Katcher, H., Glendening, J. M., Bogdany, J. K., Shiel, M. J., Haluska, F. G., Foun- Parry, D., Peters, G., and Kamb, A. Complex structure and regulation of tain, J. W., and Albino, A. P. Loss of expression of the p16/cyclin- the p16 (MTS1) locus. Cancer Res., 55: 2988–2994, 1995. dependent kinase inhibitor 2 tumor suppressor gene in melanocytic 22. Tao, W., and Levine, A. J. p19 (ARF) stabilizes p53 by blocking lesions correlates with invasive stage of tumor progression. Cancer Res., nucleo-cytoplasmic shuttling of Mdm2. Proc. Natl. Acad. Sci. USA, 96: 55: 2713–2718, 1995. 6937–6941, 1999. 38. Wagner, S. N., Wagner, C., Briedigkeit, L., and Goos, M. Homozy- 23. Kumar, R., Lundh, R. B., Louhelainen, J., and Hemminki, K. gous deletion of the p16INK4a and the p15INK4b tumour suppressor genes Mutations in the CDKN2A (p16INK4a) gene in microdissected sporadic in a subset of human sporadic cutaneous malignant melanoma. Br. J. primary melanomas. Int. J. Cancer, 75: 193–198, 1998. Dermatol., 138: 13–21, 1998. 24. Piccinin, S., Doglioni, C., Maestro, R., Vukosavljevic, T., Gaspa- 39. Fujimoto, A., Morita, R., Hatta, N., Takehara, K., and Takata, M. rotto, D., D’Orazi, C., and Boiocchi, M. p16/CDKN2 and CDK4 gene p16INK4a inactivation is not frequent in uncultured sporadic primary mutations in sporadic melanoma development and progression. Int. J. cutaneous melanoma. Oncogene, 18: 2527–2532, 1999. Cancer, 74: 26–30, 1997. 40. Holland, E. A., Beaton, S. C., Becker, T. M., Becker, T. M., Grulet, INK4 25. Healy, E., Sikkink, S., and Rees, J. L. Infrequent mutation of p16 O. M., Peters, B. A., Rizos, H., Kefford, R. F., and Mann, G. J. Analysis in sporadic melanoma. J. Investig. Dermatol., 107: 318–321, 1996. of the p16 gene, CDKN2, in 17 Australian melanoma kindreds. Onco- 26. Gruis, N. A., Weaver-Feldhaus, J., Liu, Q., Frye, C., Eeles, R., gene, 11: 2289–2294, 1995. Orlow, I., Lacombe, L., Ponce-Castaneda, V., Lianes, P., Latres, E., 41. Liu, L., Dilworth, D., Gao, L., Monzon, J., Summers, A., Lassam, Skolnick, M., Cordon-Cardo, C., and Kamb, A. Genetic evidence in N., and Hogg, D. Mutation of the CDKN2A 5ЈUTR creates an aberrant melanoma and bladder cancers that p16 and p53 function in separate pathways of tumor suppression. Am. J. Pathol., 146: 1199–1206, 1995. initiation codon and predisposes to melanoma. Nat. Genet., 21: 128– 132, 1999. 27. Ruiz, A., Puig, S., Lynch, M., Castel, T., and Estivill, X. Retention of the CDKN2A locus and low frequency of point mutations in primary 42. Merlo, A., Herman, J. G., Mao, L., Mao, L., Lee, D. J., Gabrielson, Ј and metastatic cutaneous malignant melanoma. Int. J. Cancer, 76: 312– E., Burger, P. C., Baylin, S. B., Sidransky, D. 5 CpG island methylation 316, 1998. is associated with transcriptional silencing of the tumor suppressor p16/CDKN2/MTS1 in human cancers. Nat. Med., 1: 686–692, 1995. 28. Elder, D. E., and Murphy, G. F. Malignant tumors (melanomas and related lesions): tumor progression phases and classification. In: D. E. 43. Cairns, P., Mao, L., Merlo, A., Lee, D. J., Schwab, D., Eby, Y., Elder and G. F. Murphy (eds.), Atlas of Tumor Pathology-Melanocytic Tokino, K., van der Riet, P., Blaugrund, J. E., and Sidransky, D. Rates Tumors of the Skin, pp. 103–130. Washington DC: Armed Forces of p16 (MTS1) mutations in primary tumors with 9p loss. Science Institute of Pathology, 1991. (Washington DC), 265: 415–417, 1994. 29. Ohta, M., Nagai, H., Shimizu, M., Rasio, D., Berd, D., Mastrangelo, 44. Pollock, P. M., Yu, F., Qiu, L., Parsons, P. G., and Hayward, N. K. M., Singh, A. D., Shields, J. A., Shields, C. L., and Croce, C. M. Rarity of Evidence for U.V. induction of CDKN2 mutations in melanoma cell somatic and germline mutations of the cyclin-dependent kinase 4 inhibitor lines. Oncogene, 11: 663–668, 1995. gene, CDK4I, in melanoma. Cancer Res., 54: 5269–5272, 1994. 45. Kumar, R., Smeds, J., Lundh Rozell, B., and Hemminki, K. Loss of 30. Straume, O., and Akslen, L. A. Alterations and prognostic signifi- heterozygosity at chromosome 9p21 (INK4–p14ARF locus): homozy- cance of p16 and p53 protein expression in subgroups of cutaneous gous deletions and mutations in the p16 and p14ARF genes in sporadic melanoma. Int. J. Cancer, 74: 535–539, 1997. primary melanomas. Melanoma Res., 9: 138–147, 1999.

Adrian R. Cachia, James O. Indsto, Kathryn M. McLaren, et al.

Clin Cancer Res 2000;6:3511-3515.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/6/9/3511

Cited articles This article cites 41 articles, 12 of which you can access for free at: http://clincancerres.aacrjournals.org/content/6/9/3511.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/6/9/3511.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/6/9/3511. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.