Journal of the National Cancer Institute Advance Access originally published online on May 27, 2008
JNCI Journal of the National Cancer Institute 2008 100(11):757-759; doi:10.1093/jnci/djn156
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2008. Published by Oxford University Press.
EDITORIALS |
Tumor Suppression for ARFicionados: The Relative Contributions of p16INK4a and p14ARF in Melanoma
Affiliation of author: Cancer Research UK London Research Institute, London, UK
Correspondence to: Gordon Peters, PhD, Cancer Research UK London Research Institute, 44 Lincoln’s Inn Fields, London WC2A 3PX, UK (e-mail: gordon.peters{at}cancer.org.uk).
Glancing at the title of the article by Freedberg et al. (1), readers might wonder why almost 15 years after the connection between p16INK4a and melanoma was first reported, people still seem to be questioning its importance. The confounding factor is that the gene is embedded in a locus that encodes not one but three potential tumor suppressors, INK4a, INK4b, and ARF, each vying for a share of the limelight. The two INK4 proteins, p15INK4b and p16INK4a, are cyclin-dependent kinase (CDK) inhibitors that activate the retinoblastoma (pRb) pathway, whereas p14ARF, which exploits the second exon of p16INK4a in an alternative reading frame, is an activator of p53. Elevated expression of any one of these proteins causes cell cycle arrest, but the main arena in which their effects are manifest is the implementation of cellular senescence, the state of irreversible growth arrest elicited by various forms of cellular stress (2). Originally viewed as a mechanism that limits the proliferative capacity of primary cells in tissue culture, senescence is now recognized as a critical facet of tumor suppression in vivo, and the products of the INK4b-ARF-INK4a locus are markers for oncogene-induced senescence in premalignant lesions (3,4). However, for reasons that are far from clear, their relative contributions to senescence appear to be context- and perhaps species-dependent. A simplistic interpretation of the literature would be that whereas p16INK4a takes the leading role in human cells, ARF assumes this mantle in mouse cells and the understudy p15INK4b deputizes in chicken cells, but there is obvious scope for harmony and unison.
Lest they get the wrong impression, readers should be in no doubt that p16INK4a is an important determinant of susceptibility to melanoma and other cancers. In melanoma-prone families, and patients presenting with multiple primary melanomas, germline mutations can be found throughout the coding region of INK4a, including parts (exon 1
) that are not shared with p14ARF. Importantly, the mutations generally result in demonstrable effects on the ability of p16INK4a to bind to CDK4 and CDK6 or to impair cell proliferation and are consistent with the need to maintain the ankyrin repeats that mold the structure of the protein. Germline alterations are also found in noncoding regions of the gene, and, although hard to evaluate functionally, they segregate with melanoma in afflicted families. Moreover, the other high-penetrance melanoma susceptibility gene identified thus far, CDK4, encodes the principal target of p16INK4a. Thus, some rare melanoma kindreds have a germline mutation in CDK4 (R24C) that renders it insensitive to inhibition by INK4 proteins.
As well as having the functional attributes required to block cell proliferation, p16INK4a is activated in oncogene-induced senescence downstream of the RAS-RAF-MEK signaling pathway, and cells from p16INK4a-deficient individuals are resistant to RAS-induced arrest (note that ARF remains functional in these patients) (5). Importantly, p16INK4a expression is a marker of oncogene-induced senescence in vivo, an apt example being melanocytic nevi expressing activated B-RAF (4,6). Simply explanting human cells into tissue culture results in the accumulation of p16INK4, and virtually all established human cell lines have defects in either p16INK4a function or its downstream target pRb. The defects can be genetic, ranging from single point mutations to extensive homozygous deletions, or epigenetic, such as CpG methylation of the promoter.
A wide variety of sporadic cancers and cell lines carry these stigmata of escape from senescence, and it is exactly this sort of evidence that forms the basis of the report by Freedberg et al. (1). Using relatively sophisticated techniques, they conducted a detailed survey of DNA copy number and methylation status across the INK4b-ARF-INK4a locus in a collection of melanoma metastases from 58 patients. Arguing that it generally requires two hits to completely incapacitate a tumor suppressor gene, they conclude that p14ARF is more frequently inactivated than p16INK4a and by inference that it may play a more substantive role in human melanoma. A concern is that in some tumors the estimates of DNA copy number, based on multiplex ligation-dependent probe amplification (MLPA), show surprising variability across the locus. It is difficult to see how and why multiple localized deletions might arise during the escape from senescence. This is an important issue because conclusions about the inactivation of ARF are based on finding both hypermethylation and reduced copy number, without supporting expression data at either the protein or RNA level. It is also clear that the alterations often extend to p15INK4b and beyond, and this does not appear to be factored into the interpretation.
So why the focus on ARF? The answer is quite simple: modelling the effects of INK4b-ARF-INK4a alterations in mice points unequivocally to a prominent role for p19Arf in senescence and tumor suppression. Mice specifically lacking exon 1β are highly tumor prone, and mouse embryo fibroblasts (MEFs) derived from these animals are immortal and resistant to RAS-induced senescence (7). In contrast, mice specifically lacking Ink4a are much less tumor prone, unless treated with carcinogens, and the corresponding MEFs undergo oncogene-induced senescence in response to RAS (8,9). That said, both p19Arf and p16Ink4a are induced by RAS and both appear to contribute to tumor suppression in different contexts. Indeed, the most persuasive mouse models for melanoma are those in which both copies of Ink4a are inactivated and Arf is heterozygous (8), the mirror image of the scenario favored by Freedberg et al. (1) in human melanoma. The appeal of building a case for Arf in melanoma is that it could explain why p53 is rarely mutated in this tumor. However, it is clear that Arf can function as a melanoma tumor suppressor in the absence of p53 (10).
While it would of course be very satisfying to find direct parallels between the mouse and human models, we might also be able to learn from the differences, and it is worth considering what we would have deduced about human p14ARF if we did not have access to knockout mice. With the caveat that endogenous p14ARF is notoriously difficult to detect, a problem encountered by Freedberg et al. (1), it has not been implicated in senescence in human cells; nor does it appear to be activated by oncogenic RAS (11). That said, knocking down p14ARF by RNA interference clearly confers a proliferative benefit (12). While this presumably reflects its influence on p53, ARF is also credited with binding to a diverse array of cellular proteins, including MYC, E2F1, TIP60, NPM, and other equally intriguing candidates (13). In many cases, the physiological relevance remains uncertain because much of the evidence for these interactions relies on overexpression. Where tested, the known functions of ARF are generally attributable to the amino acids encoded by the unique exon 1β, so that mutations in the exons shared with p16INK4a are unlikely to have much of an impact on p14ARF. Indeed, it has proved remarkably difficult to inactivate ARF by tinkering with its sequence, suggesting that the protein makes multiple contacts with its clients (14). This could potentially explain why mutations in exon 1β are so rare and why, in contrast to p16INK4a, evidence that they affect ARF function remains equivocal. There are nevertheless examples of nonsense, frameshift, and splice site mutations that specifically incapacitate ARF, and importantly germline deletions of ARF are associated with predisposition to melanoma and neural system tumors. However, the interpretation remains uncertain because there are also examples of melanoma and neural system tumors where the alteration appears specific for p16INK4a (15) or extends over a considerable distance (16).
In the context of deletions, it is worth recalling that most of the hemi- and homozygous deletions of chromosome 9p21 in human cancers encompass the entire INK4b-ARF-INK4a locus, including p15INK4b. Although p15INK4b gets much less attention than its illustrious neighbors, it qualifies as a marker of oncogene-induced senescence in vivo (3) and combinatorial knockouts in mice have confirmed that it has an appreciable influence on tumor susceptibility (17). Interestingly, p15Ink4b appears to be upregulated in the absence of p16Ink4a, suggesting that it serves as a backup for loss of p16Ink4a. Codeletion of p15INK4b and p16INK4a would prevent any compensatory response, but there are nevertheless tumors in which p15INK4b appears to be specifically silenced by localized promoter methylation (18). In passing, it is worth noting that in chicken cells, which lack an equivalent of p16INK4a, it is p15INK4b that is implicated in senescence (19).
In summary, while the role of INK4a in melanoma is irrefutable, it is much less clear how ARF and INK4b fit into the picture, based on the available evidence. Bear in mind that our understanding of how the INK4b-ARF-INK4a locus is regulated remains rudimentary. In addition to the complications posed by the presence of overlapping and differentially spliced transcripts from the three tandemly arranged promoters, there is now compelling evidence that the locus is traversed by antisense transcription initiating from within the locus or from the nearby MTAP gene (16,20,21). There is also evidence for a regulatory domain upstream of INK4b (22), and a number of previous loss of heterozygosity studies have pointed to foci of deletion on chromosome 9p21 that do not affect the INK4b-ARF-INK4a locus. Rather than raising the specter of yet more tumor suppressors in the region, perhaps these deletions are revealing elements of chromatin structure and organization that underpin the regulation of our three aspiring prima donnas. For now, we should be content to let them share the limelight.
REFERENCES
1. Freedberg DE, Rigas SH, Russak J, et al. Frequent p16-independent inactivation of p14ARF in human melanoma. J Natl Cancer Inst (2008) 100(11):784–795.
2. Drayton S, Peters G. Immortalisation and transformation revisited. Curr Opin Genet Dev. (2002) 12(1):98–104.[CrossRef][ISI][Medline]
3. Collado M, Gil J, Efeyan A, et al. Tumour biology: senescence in premalignant tumours. Nature (2005) 436(7051):642.[CrossRef][Medline]
4. Michaloglou C, Vredeveld LC, Soengas MS, et al. BRAFE600-associated senescence-like cell cycle arrest of human naevi. Nature (2005) 436(7051):720–724.[CrossRef][Medline]
5. Brookes S, Rowe J, Ruas M, et al. INK4a-deficient human diploid fibroblasts are resistant to RAS-induced senescence. Embo J (2002) 21(12):2936–2945.[CrossRef][ISI][Medline]
6. Gray-Schopfer VC, Cheong SC, Chong H, et al. Cellular senescence in naevi and immortalisation in melanoma: a role for p16? Br J Cancer (2006) 95(4):496–505.[CrossRef][ISI][Medline]
7. Kamijo T, Zindy F, Roussel MF, et al. Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell (1997) 91(5):649–659.[CrossRef][ISI][Medline]
8. Krimpenfort P, Quon KC, Mooi WJ, Loonstra A, Berns A. Loss of p16Ink4a confers susceptibility to metastatic melanoma in mice. Nature (2001) 413(6851):83–86.[CrossRef][Medline]
9. Sharpless NE, Bardeesy N, Lee KH, et al. Loss of p16Ink4a with retention of p19Arf predisposes mice to tumorigenesis. Nature (2001) 413(6851):86–91.[CrossRef][Medline]
10. Ha L, Ichikawa T, Anver M, et al. ARF functions as a melanoma tumor suppressor by inducing p53-independent senescence. Proc Natl Acad Sci U S A (2007) 104(26):10968–10973.
11. Wei W, Hemmer RM, Sedivy JM. Role of p14(ARF) in replicative and induced senescence of human fibroblasts. Mol Cell Biol. (2001) 21(20):6748–6757.
12. Voorhoeve PM, Agami R. The tumor-suppressive functions of the human INK4A locus. Cancer Cell (2003) 4(4):311–319.[CrossRef][ISI][Medline]
13. Sherr CJ. Divorcing ARF and p53: an unsettled case. Nat Rev Cancer (2006) 6(9):663–673.[CrossRef][ISI][Medline]
14. Clark PA, Llanos S, Peters G. Multiple interacting domains contribute to p14ARF mediated inhibition of MDM2. Oncogene (2002) 21(29):4498–4507.[CrossRef][ISI][Medline]
15. Ghiorzo P, Gargiulo S, Pastorino L, et al. Impact of E27X, a novel CDKN2A germ line mutation, on p16 and p14ARF expression in Italian melanoma families displaying pancreatic cancer and neuroblastoma. Hum Mol Genet. (2006) 15(18):2682–2689.
16. Pasmant E, Laurendeau I, Heron D, Vidaud M, Vidaud D, Bieche I. Characterization of a germ-line deletion, including the entire INK4/ARF locus, in a melanoma-neural system tumor family: identification of ANRIL, an antisense noncoding RNA whose expression coclusters with ARF. Cancer Res. (2007) 67(8):3963–3969.
17. Krimpenfort P, Ijpenberg A, Song JY, et al. p15Ink4b is a critical tumour suppressor in the absence of p16Ink4a. Nature (2007) 448(7156):943–946.[CrossRef][Medline]
18. Herman JG, Civin CI, Issa JP, Collector MI, Sharkis SJ, Baylin SB. Distinct patterns of inactivation of p15INK4B and p16INK4A characterize the major types of hematological malignancies. Cancer Res. (1997) 57(5):837–841.
19. Kim SH, Rowe J, Fujii H, et al. Upregulation of chicken p15INK4b at senescence and in the developing brain. J Cell Sci. (2006) 119(12):2435–2443.
20. Yu W, Gius D, Onyango P, et al. Epigenetic silencing of tumour suppressor gene p15 by its antisense RNA. Nature (2008) 451(7175):202–206.[CrossRef][Medline]
21. Schmid M, Sen M, Rosenbach MD, Carrera CJ, Friedman H, Carson DA. A methylthioadenosine phosphorylase (MTAP) fusion transcript identifies a new gene on chromosome 9p21 that is frequently deleted in cancer. Oncogene (2000) 19(50):5747–5754.[CrossRef][Medline]
22. Gonzalez S, Klatt P, Delgado S, et al. Oncogenic activity of Cdc6 through repression of the INK4/ARF locus. Nature (2006) 440(7084):702–706.[CrossRef][Medline]
Related Articles in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 784-795.
J Natl Cancer Inst 2008 100: 755.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||