4734.Full-Text.Pdf
Total Page:16
File Type:pdf, Size:1020Kb
4734 Vol. 10, 4734–4741, July 15, 2004 Clinical Cancer Research CpG Island Methylation Is Responsible for p14ARF Inactivation and Inversely Correlates with p53 Overexpression in Resected Non–Small Cell Lung Cancer Han-Shui Hsu,1 Yu-Chien Wang,2 occurred at 26%, 9%, and 0%, respectively, of microdis- Ruo-Chia Tseng,2 Jer-Wei Chang,2 sected NSCLCs. Jung-Ta Chen,3 Chuen-Ming Shih,5 Conclusions: Our data suggest that p14ARF 5 CpG 4 2 hypermethylation is the predominant mechanism involved Chih-Yi Chen, and Yi-Ching Wang in the aberrant expression of the p14ARF gene. In addition, 1 Division of Thoracic Surgery, Taipei Veterans General Hospital, p14ARF 5CpG hypermethylation occurs inversely to P53 National Yang-Ming University School of Medicine, Taipei; overexpression. 2Department of Life Sciences, National Taiwan Normal University, Taipei; 3Department of Pathology and 4Division of Thoracic Surgery, Taichung Veterans General Hospital, Taichung; and 5Division of INTRODUCTION Pulmonary and Critical Care Medicine, Department of Internal The INK4a/ARF locus on human chromosome region 9p21 Medicine, China Medical College Hospital and Institute of Medicine, Chung Shan Medical University, Taichung, Republic of China encodes two distinct proteins, which are translated in different reading frames from alternatively spliced transcripts (1). P16INK4a is specified by the ␣ transcript composed of exons ABSTRACT 1␣, 2, and 3. The other product, P14ARF, is encoded by the Purpose and Experimental Design: The molecular mech- smaller  transcript, which is composed of exons 1, 2, and 3 anisms by which the p14ARF gene is altered in non–small (2). It is known that P14ARF and P16INK4a are upstream cell lung cancer (NSCLC) are complex and unclear. Using regulators in the P53 and RB tumor suppressor pathways, re- genetic and epigenetic analyses, we examined various mo- spectively. P16INK4a inhibits cyclin-dependent kinases 4 and 6 lecular alterations including the loss of protein and mRNA to phosphorylate RB protein, resulting in the induction of G1 expression, and 5CpG hypermethylation, allelic imbalance, arrest (3). P14ARF prevents the MDM2-mediated degradation and mutation of the p14ARF gene in a series of 102 NSCLC of P53 protein (4, 5). Thus, P14ARF overexpression leads to the samples, in parallel with clinicopathological and prognostic stabilized P53, which then induces several biological responses analyses. To clarify the biological significance of p14ARF against DNA damage, such as G1 arrest, G2-M arrest, and alterations, its relationship with p16INK4a and p53 alter- apoptosis (5–7). It has also been reported that the P14ARF- ations was also examined. mediated G1 and G2 arrests are abolished in mouse embryonic Results: We found that 34% of NSCLC patients had fibroblasts lacking functional P53, indicating that P53 acts aberrant P14ARF protein expression, which was more fre- downstream of P14ARF in a cell cycle regulatory pathway (8). quent in adenocarcinomas (AD; 44%) than in squamous cell Inactivation of the p16INK4a gene results from intragenic -A high concordance was mutation, homozygous/hemizygous deletions, and promoter hy .(0.024 ؍ carcinomas (22%; P observed between alterations in protein and mRNA expres- permethylation, and these genetic and epigenetic alterations (sion and 5CpG hypermethylation (P < 0.001). The p14ARF have been detected frequently in a variety of human cancers (9 hypermethylation inversely correlated with P53 overexpres- including non–small cell lung cancer (NSCLC; Refs. 10–12). This mutually exclusive relationship for However, the pathogenic and biological significance of p14ARF .(0.001 ؍ sion (P alteration between p14ARF and p53 was also supported by a gene alterations in human cancers is still unclear. Although the worse prognosis of AD patients with positive P14ARF ex- p14ARF null mice (where only the exon 1 is lost) develop and of AD patients with P53 overexpres- spontaneous tumors at an early age (8), no germ-line mutations (0.01 ؍ pression (P ,Our data also indicated that hemizygous/ affecting the p14ARF-specific exon 1 have been identified (13 .(0.006 ؍ sion (P homozygous deletion and mutation in the p14ARF gene 14), and yet many INK4a/ARF exon 2 mutations alter both the P16INK4a and P14ARF amino acid sequences (15). Alteration of the P14ARF protein expression is found by using immuno- histochemistry in 25–41% of NSCLCs (16, 17). However, con- Received 12/10/03; revised 3/26/04; accepted 4/19/04. trary results have been reported when loss of P14ARF expres- Grant support: NHRI93A1-NSCLC06–5 and NSC92–2320-B-003– sion was correlated with abnormal  transcripts (16, 17). 003 from the National Science Council (The Executive Yuan, Republic Various loss of heterozygosity (LOH) assays using different of China). The costs of publication of this article were defrayed in part by the microsatellite markers have indicated allelic loss in the 9p21 at payment of page charges. This article must therefore be hereby marked various frequencies (12, 18, 19). In two studies of NSCLC, advertisement in accordance with 18 U.S.C. Section 1734 solely to homozygous deletion of p14ARF was indirectly inferred from indicate this fact. allelotype pattern at 9p21 (18) or from amplification of a frag- Requests for reprints: Yi-Ching Wang, Department of Life Sciences, National Taiwan Normal University, No. 88, Sec. 4, Tingchou Road, ment containing noncoding intronic sequence (20), but this was Taipei 116, Republic of China. Phone: 886-2-29336876, extension 373; not confirmed in other studies. Sequence analysis revealed that Fax: 886-2-29312904; E-mail: [email protected]. exon 1 mutation occurs rarely in the p14ARF gene (16, 18, 20). Downloaded from clincancerres.aacrjournals.org on September 25, 2021. © 2004 American Association for Cancer Research. Clinical Cancer Research 4735 In data published to date, only a few reports showed p14ARF waxed in xylene, genomic DNA was extracted according to the methylation in 5–9% of primary NSCLCs (21–23). However, standard methods described above. For the RNA expression the correlation of p14ARF methylation with protein expression assay, the total RNA were prepared from matched pairs of was not examined in these studies. primary tumors and nearby normal lung tissues using Trizol The frequency and mechanisms of p14ARF gene inactiva- reagent (Invitrogen, Carlsbad, CA). cDNA was synthesized tion and their correlation with loss of protein expression in using SuperScript reverse transcriptase (Invitrogen) with the NSCLCs vary between different studies. Furthermore, data con- protocols provided by the manufacturer. cerning the frequency and mechanisms of p14ARF gene inacti- Analysis of Protein Expression: Immunohistochemistry vation have rarely been documented in the same series of Assay. Paraffin blocks of tumors were cut into 5-m slices NSCLC. To elucidate the possible mechanisms involved in and then processed using standard deparaffinization and rehy- p14ARF changes in NSCLC tumorigenesis, we performed a dration techniques. Polyclonal antibody p14ARF/p16 Ab-4 comprehensive genetic and epigenetic study of the p14ARF (1:100; NeoMarkers, Union City, CA) was used as the primary status in a series of 102 NSCLC samples and compared the data antibodies to detect P14ARF protein expression. The primary with the clinicopathological parameters and prognosis of the antibody was detected using biotinylated secondary antibody patients. To clarify the biological significance of p14ARF alter- (DAKO, Carpinteria, CA) according to the manufacturer’s in- ations, we also analyzed the relationship of p14ARF alterations structions. The sections were then counterstained with hematox- with p53 and p16INK4a alterations. The results indicated that ylin. The stains were graded negative when there was complete p14ARF alteration is involved in NSCLC tumorigenesis and absence of staining in the tumor cell nuclei, with adequate mainly results from 5ЈCpG hypermethylation. In addition, nuclear staining in surrounding normal stromal and epithelial p14ARF hypermethylation inversely correlated with P53 over- cells. The samples were assayed in batches including both expression and occurred frequently in tumors with p16INK4a negative- and positive-expression patients. The analysis for the hypermethylation. P53 expression was described previously (24). Analysis of mRNA Expression: Multiplex Reverse Transcription-PCR (RT-PCR) Assay. The RNA was ex- MATERIALS AND METHODS tracted, and 96 samples had adequate RNA for additional anal- Sample Preparation and Clinical Characterization of ysis. p14ARF mRNA expression was assayed in a multiplex Patients. Tissues were collected after obtaining permission RT-PCR analysis using the -actin gene as an internal control. from the appropriate Institutional Review Board and informed The coding regions of exons 1–2oftheP14ARF gene and the consents from the recruited patients. Surgically resected tumor -actin gene were amplified using primers described by Gazzeri samples from 102 patients with NSCLC were collected between et al. (16). Reactions were carried out in a volume of 25 l with 1999 and 2001. Of these patients, 54 had squamous carcinoma 1 l cDNA and 0.25 pmol primers on a DNA thermal cycler. (SQ), 42 had adenocarcinoma (AD), 2 had adenosquamous cell PCR was performed for 35 cycles with an annealing temperature carcinoma (AS), and 4 had large-cell carcinoma (LC). The of 70°C. The number of cycles and the amount of primers and histology of the tumor types and their stages were determined