Selective Loss of Wild-Type P16ink4a Expression in Human Nevi

The p16INK4a cyclin-dependent kinase inhibitor acts as a negative regulator of cyclin D-dependent kinases and is a critical gatekeeper at the G1–S checkpoint (Serrano et al., 1996Serrano M. Lee H.-W. Chin L. et al.Role of the INK4a locus in tumor suppression and cell mortality.Cell. 1996; 85: 27-37Abstract Full Text Full Text PDF PubMed Scopus (1408) Google Scholar). Accordingly, p16INK4a is frequently inactivated in human tumors, and deletions involving this locus occur frequently in melanomas. Inherited mutations in the p16INK4a gene are also associated with melanoma susceptibility in 40% of multiple-case melanoma families (Goldstein et al., 2006Goldstein A.M. Chan M. Harland M. et al.High-risk melanoma susceptibility genes and pancreatic cancer, neural system tumors, and uveal melanoma across GenoMEL.Cancer Res. 2006; 66: 9818-9828Crossref PubMed Scopus (326) Google Scholar). It is generally acknowledged that the progressive and gradual loss of p16INK4a expression correlates with the advancing stages of melanocytic disease progression. Thus, benign nevi show minimal loss of the p16INK4a gene, whereas allelic loss of this locus is common in dysplastic nevi, and in primary and metastatic melanomas (Talve et al., 1997Talve L. Sauroja I. Collan Y. et al.Loss of expression of the p16INK4/CDKN2 gene in cutaneous malignant melanoma correlates with tumor cell proliferation and invasive stage.Int J Cancer. 1997; 74: 255-259Crossref PubMed Scopus (107) Google Scholar;Sini et al., 2008Sini M.C. Manca A. Cossu A. et al.Molecular alterations at chromosome 9p21 in melanocytic naevi and melanoma.Br J Dermatol. 2008; 158: 243-250PubMed Google Scholar). Similarly, there is a progressive decrease in the expression level of p16INK4a protein from melanoma in situ to invasive and metastatic melanomas (Keller-Melchior et al., 1998Keller-Melchior R. Schmidt R. Piepkorn M. Expression of the tumor suppressor gene product p16INK4 in benign and malignant melanocytic lesions.J Invest Dermatol. 1998; 110: 932-938Crossref PubMed Scopus (79) Google Scholar;Sini et al., 2008Sini M.C. Manca A. Cossu A. et al.Molecular alterations at chromosome 9p21 in melanocytic naevi and melanoma.Br J Dermatol. 2008; 158: 243-250PubMed Google Scholar). The repression of p16INK4a expression is likely to commence earlier than disease manifestation, occurring during melanocytic proliferation, in benign and atypical nevi. Initial reports described qualitatively uniform immunohistochemical labelling for p16INK4a in nevi (Keller-Melchior et al., 1998Keller-Melchior R. Schmidt R. Piepkorn M. Expression of the tumor suppressor gene product p16INK4 in benign and malignant melanocytic lesions.J Invest Dermatol. 1998; 110: 932-938Crossref PubMed Scopus (79) Google Scholar), but more recent studies have found that the percentage of p16INK4a-positive nevus cells and the intensity of staining were heterogeneous among nevi (Michaloglou et al., 2005Michaloglou C. Vredeveld L.C. Soengas M.S. et al.BRAFE600-associated senescence-like cell cycle arrest of human naevi.Nature. 2005; 436: 720-724Crossref PubMed Scopus (1680) Google Scholar;Gray-Schopfer et al., 2006Gray-Schopfer V.C. Cheong S.C. Chong H. et al.Cellular senescence in naevi and immortalisation in melanoma: a role for p16?.Br J Cancer. 2006; 95: 496-505Crossref PubMed Scopus (311) Google Scholar). Reduced p16INK4a expression in nevi indicates that p16INK4a may contribute to the clonal expansion of nevus cells. This is supported by the clinical observation that patients with heterozygous p16INK4a mutations often develop larger, more numerous, and dysplastic nevi (Bishop et al., 2000Bishop J.A. Wachsmuth R.C. Harland M. et al.Genotype/phenotype and penetrance studies in melanoma families with germline CDKN2A mutations.J Invest Dermatol. 2000; 114: 28-33Crossref PubMed Scopus (98) Google Scholar). To precisely examine the expression of p16INK4a protein in nevi, we initially determined the percentage of nevus cells expressing detectable p16INK4a in 20 excised human nevi, comprising 15 compound nevi and 5 dysplastic nevi. As shown in Figure 1, all nevi displayed a heterogenous pattern of p16INK4a immunopositivity, and although the sample size was small we achieved borderline statistical significance on comparing p16INK4a expression in compound versus dysplastic nevi (Mann–Whitney test;P=0.053;Figure 1b). Considering that normal melanocytes at the dermal–epidermal junction have undetectable levels of p16INK4a (data not shown;Michaloglou et al., 2005Michaloglou C. Vredeveld L.C. Soengas M.S. et al.BRAFE600-associated senescence-like cell cycle arrest of human naevi.Nature. 2005; 436: 720-724Crossref PubMed Scopus (1680) Google Scholar), we sought to determine whether expression of p16INK4a was actively lost or simply not induced in some nevus cells. During this investigation, we discovered that the p16INK4a JC8 mouse monoclonal antibody did not detect the melanoma-associated p16INK4a R24P variant in western and immunohistochemical analysis (Figure 2a and b). This antibody, which is raised against full-length recombinant p16INK4a protein, is frequently used for immunohistochemical detection of p16INK4a in human cancers (Redman et al., 2008Redman R. Rufforny I. Liu C. et al.The utility of p16(Ink4a) in discriminating between cervical intraepithelial neoplasia 1 and nonneoplastic equivocal lesions of the cervix.Arch Pathol Lab Med. 2008; 132: 795-799PubMed Google Scholar). Using melanocytic nevi from melanoma-prone individuals from a single family (ID 31220) carrying a germline p16INK4a mutation encoding the R24P variant (Holland et al., 1995Holland E.A. Beaton S.C. Becker T.M. et al.Analysis of the p16 gene, CDKN2, in 17 Australian melanoma kindreds.Oncogene. 1995; 11: 2289-2294PubMed Google Scholar), we were able to examine whether wild-type p16INK4a expression was selectively lost in nevus cells. In all, 11 nevi (8 compound nevi and 3 dysplastic nevi) were derived from five R24P carriers and 19 nevi (15 compound nevi and 4 dysplastic nevi) were excised from age-matched melanoma-affected controls, with no known family history of melanoma. The nevi ranged in size from 2 to 22 mm for the controls and from 3 to 20 mm for the R24P carriers. The indication for excision was clinical change within a clinically dysplastic (atypical) nevus to exclude melanoma (with low clinical suspicion of malignancy) in the majority of cases. We used dual immunofluorescence with JC8 and C20 on paraffin-embedded sections of nevi; the C20 antibody detects both wild-type and mutant p16INK4a (Figure 2). R24P-positive nevus cells expressing any combination of R24P and wild-type p16INK4a stained positively with C20, while the JC8 antibody stained positively only in cells in which wild-type p16INK4a expression was retained (see Figure 2b). The distribution of p16INK4a was both nuclear and cytoplasmic in all p16INK4a-positive nevus samples. In R24P-positive-only cells (JC8-/C20+) the distribution of mutant p16INK4a varied and showed nuclear and cytoplasmic, predominantly cytoplasmic, or predominantly nuclear localization (data not shown). As expected, nearly all p16INK4a-positive control nevus cells stained positively with both the JC8 and C20 antibodies. In contrast, a substantial proportion of nevus cells from R24P carriers showed strong positivity with only the C20 antibody (i.e. no JC8 positivity was seen) (Figure 2c) and thus, these nevus cells were negative for wild-type p16INK4a expression. As shown in Figure 2d, there was a highly significant difference between the control nevi and R24P-positive nevi (Mann–Whitney test;P<0.001); in the latter, wild-type p16INK4a expression was selectively lost in a subset (up to 31%) of nevus cells. These data confirm that the loss of wild-type p16INK4a expression commences early in melanocyte proliferation, with common benign nevi frequently containing cells with no p16INK4a. The mosaic pattern of p16INK4a expression in acquired nevi is reminiscent of the heterogeneity of B-RAF mutations found in human nevi (Lin et al., 2009Lin J. Takata M. Murata H. et al.Polyclonality of BRAF mutations in acquired melanocytic nevi.J Natl Cancer Inst. 2009; 101: 1423-1427Crossref PubMed Scopus (55) Google Scholar). It remains to be determined whether the mutant B-RAF is coexpressed with p16INK4a or if activated B-RAF and loss of p16INK4a contribute separately to the clonal expansion of melanocytes. Alternatively, activated B-RAF and p16INK4a deficiency may co-exist in a small percentage of nevus cells to enhance melanocyte proliferation and drive neoplastic transformation. The absence of a mutant-specific B-RAF antibody has so far precluded the execution of such an analysis. It is well established that activated B-RAF can promote nevus formation in murine and fish melanoma models (Patton et al., 2005Patton E.E. Widlund H.R. Kutok J.L. et al.BRAF mutations are sufficient to promote nevi formation and cooperate with p53 in the genesis of melanoma.Curr Biol. 2005; 15: 249-254Abstract Full Text Full Text PDF PubMed Scopus (493) Google Scholar;Goel et al., 2009Goel V.K. Ibrahim N. Jiang G. et al.Melanocytic nevus-like hyperplasia and melanoma in transgenic BRAFV600E mice.Oncogene. 2009; 28: 2289-2298Crossref PubMed Scopus (117) Google Scholar), and the arrested state of nevi does not appear to require p16INK4a (Dhomen et al., 2009Dhomen N. Reis-Filho J.S. da Rocha Dias S. et al.Oncogenic Braf induces melanocyte senescence and melanoma in mice.Cancer Cell. 2009; 15: 294-303Abstract Full Text Full Text PDF PubMed Scopus (438) Google Scholar). Nevertheless, the influence of p16INK4a loss on the development of B-RAFV600E-induced nevi needs to be investigated in human melanocytic tumors. This is particularly relevant as p16INK4a loss regulates the penetrance and latency of B-RAF-induced murine melanomas (Dhomen et al., 2009Dhomen N. Reis-Filho J.S. da Rocha Dias S. et al.Oncogenic Braf induces melanocyte senescence and melanoma in mice.Cancer Cell. 2009; 15: 294-303Abstract Full Text Full Text PDF PubMed Scopus (438) Google Scholar). This work was supported by program grant 402761 of the National Health and Medical Research Council of Australia (NHMRC), the Cancer Institute NSW, Cancer Council NSW, and an infrastructure grant to the Westmead Millennium Institute by the Health Department of NSW through the Sydney West Area Health Service. The Westmead Institute for Cancer Research is the recipient of capital grant funding from the Australian Cancer Research Foundation. HR is a recipient of a Cancer Institute NSW Research Fellowship and an NHMRC Senior Research Fellowship. RAS is a Cancer Institute NSW Clinical Research Fellow.