[CANCER RESEARCH 61, 5636 –5643, July 15, 2001] Alterations of p14 ARF , p53, and p73 Genes Involved in the E2F-1-mediated Apoptotic Pathways in Non-Small Cell Lung Carcinoma Siobhan A. Nicholson, 1 Nader T. Okby, 1 Mohammed A. Khan, Judith A. Welsh, Mary G. McMenamin, William D. Travis, James R. Jett, Henry D. Tazelaar, Victor Trastek, Peter C. Pairolero, Paul G. Corn, James G. Herman, Lance A. Liotta, Neil E. Caporaso, and Curtis C. Harris 2 Laboratory of Human Carcinogenesis, National Cancer Institute, Bethesda, Maryland 20892 [S. A. N., M. A. K., J. A. W., M. G. M., L. A. L., N. E. C., C. C. H.]; Orange Pathology Associates, Middleton, New York 10940 [N. T. O.]; Armed Forces Institute of Pathology, Washington, DC 20306 [S. A. N., W. D. T.]; Mayo Clinic, Rochester, Minnesota 55905 [J. R. J., H. D. T., V. T., P. C. P.]; and The Johns Hopkins Oncology Center, Baltimore, Maryland 21231 [P. G. C., J. G. H.] ABSTRACT Overexpression of E2F-1 induces apoptosis by both a p14 ARF -p53- and a p73-mediated pathway. p14 ARF is the alternate tumor suppressor prod- uct of the INK4a/ARF locus that is inactivated frequently in lung carci- nogenesis. Because p14 ARF stabilizes p53, it has been proposed that the loss of p14 ARF is functionally equivalent to a p53 mutation. We have tested this hypothesis by examining the genomic status of the unique exon 1 of p14 ARF in 53 human cell lines and 86 primary non-small cell lung carci- nomas and correlated this with previously characterized alterations of p53. Homozygous deletions of p14 ARF were detected in 12 of 53 (23%) cell lines and 16 of 86 (19%) primary tumors. A single cell line, but no primary tumors, harbored an intragenic mutation. The deletion of p14 ARF was inversely correlated with the loss of p53 in the majority of cell lines (P  0.02), but this relationship was not maintained among primary tumors (P  0.5). E2F-1 can also induce p73 via a p53-independent apoptotic pathway. Although we did not observe inactivation of p73 by either mutation or DNA methylation, haploinsufficiency of p73 correlated positively with either p14 ARF or p53 mutation or both (P  0.01) in primary non-small cell lung carcinomas. These data are consistent with the current model of p14 ARF and p53 interaction as a complex network rather than a simple linear pathway and indicate a possible role for an E2F-1-mediated failsafe, p53-independent, apoptotic pathway involving p73 in human lung carcinogenesis. INTRODUCTION The Rb 3 and p53-mediated checkpoints are central to the control of cell cycle progression (1, 2). As linchpins of tumor suppression, the disruption of their function is a hallmark of carcinogenesis. These pathways are linked, and their activities are influenced, in turn, by a single genetic locus, INK4a, located on chromosome 9p21 (3, 4). This locus encodes two tumor suppressors in a manner unique to eukary- otic cells. p16 INK4a is encoded by exons 1, 2, and 3, and inhibits the cyclin-dependent kinases 4 and 6. This then inhibits the phosphoryl- ation of Rb, resulting in the sequestration of E2F transcription factors and prevents the transition from G 1 to S-phase of the cell cycle (5). p16 INK4a was identified as a tumor suppressor as a result of its association with germ-line mutations in familial melanomas (6 – 8) and with frequent deletion, mutation, and promoter methylation in sporadic human tumors and cell lines (9 –12). The recently discovered murine p19 ARF and its human homologue p14 ARF are the alternate transcripts of this locus (13). p14 ARF is encoded by a separate exon 1 that lies 20 kb upstream of exon 1 and shares exons 2 and 3 as read in an ARF, giving rise to a protein completely unrelated to p16 INK4a (14). Despite its unrelated structure, p14 ARF also is capable of causing cell cycle arrest in G 1 and G 2 . p14 ARF binds to and antagonizes the actions of MDM2, a negative regulator of p53 (15). Thus, it interferes with the ability of MDM2 to block transcription, to ubiquinate, and to transport p53 to the cyto- plasm for degradation (16 –22). In response to DNA damage, the accumulation of p53 results in cell cycle arrest or apoptosis. The loss of p14 ARF increases p53 degradation, thereby diminishing the p53 response to genotoxic stress. The role of p14 ARF in carcinogenesis was first demonstrated in mice by showing that ARF-null mice are highly tumor prone (23) and develop sarcomas, lymphomas, carcinomas, and gliomas and thus, die early in life (24). Mice that are heterozygous for ARF also develop tumors but die later in life. Mouse embryonic fibroblasts that lack ARF do not undergo replicative senescence in culture and can be transformed by oncogenic ras alone (23). On the basis of this evidence, it was hypothesized that p14 ARF functions as a tumor suppressor in vivo. The modulation of the p53 response to genotoxic stress by p14 ARF also raises the possibility that patients, whose tumors show a loss of p14 ARF , have a poorer prognosis. This protective role has been dem- onstrated in mice, where ARF protects cells against Myc-induced tumors, and on an ARF-null background, E-Myc transgenic mice die of aggressive lympholeukemias at an early age (25). Although alter- ations of p16 INK4a have been demonstrated in 50% of NSCLCs (26 –31), the status of p14 ARF has not been as well characterized. Although p14 ARF deletions are well described, intragenic mutations have not been identified within the unique exon 1. There may be additional associations between p14 ARF and other cell cycle regulators, oncoproteins, and tumor suppressors. p14 ARF is not activated directly by DNA damage, but rather by a subset of abnormal hyperproliferative signals including oncoproteins myc, ras, and v-abl (32) and E2F-1 (33). As one of the targets of phosphorylated Rb, the induction of p14 ARF by E2F-1 establishes cross-talk between the Rb and p53 pathways (33). The activities of the E2F family of transcrip- tion factors influence both cell cycle progression and apoptosis, and the loss of regulation is seen in many human cancers (34). In addition to p14 ARF , p73, an evolutionary relative of p53 and putative tumor suppressor, also is a downstream target of E2F-1. The activation of p73 has been shown recently to provide a means for E2F-1 to induce cell death in the absence of p53 (35). Furthermore, the disruption of p73 function inhibits E2F-1-induced apoptosis in p53-null cells (36). Because E2F-1-induced apoptosis uses both a p53-independent path- way (mediated by p73) and a p53-dependent pathway (mediated by p14 ARF ; Ref. 37), the loss of this apoptotic signal through the disrup- tion of both pathways may enhance tumor development, raising the selection pressure for these genetic alterations in tumors. We tested these hypotheses by studying alterations of exon 1 of p14 ARF and p73 in a series of NSCLCs and sought correlations with other clinical, genetic, pathological, and epidemiological markers examined previ- Received 1/26/01; accepted 5/10/01. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 These authors contributed equally to this work. 2 To whom requests for reprints should be addressed, at Laboratory of Human Carcinogenesis, National Cancer Institute, NIH, Building 37, Room 2C05, 37 Convent Drive, Bethesda, MD 20892-4255. Phone: (301) 496-2048; Fax: (301) 496-0497; E-mail: Curtis_Harris@nih.gov. 3 The abbreviations used are: Rb, retinoblastoma; ARF, alternative reading frame; MDM, murine double minute; SCLC, small cell lung carcinoma; NSCLC, non-SCLC; LOH, loss of heterozygosity; SSCP, single strand conformation polymorphism; WT, wild type. 5636 Research. on December 11, 2021. © 2001 American Association for Cancer cancerres.aacrjournals.org Downloaded from