The prevalence of methylation of the p16, DAPK and RASSFIA genes was investigated in lung adenocarcinoma from smokers, former uranium miners and never smokers. The association between a common genetic alteration in adenocarcinoma, mutation of the K-ras gene and methylation of these genes, as well as survival was examined. Adenocarcinomas from 157 smokers, 46 never smokers and 34 former uranium miners were evaluated for methylation of the p16, DAPK and RASSFIA genes using the methylation-specific PCR assay. Comparisons were also made to prevalences of methylation of the MGMT gene and mutation of the K-ras gene previously examined in these tumors. The prevalence of methylation for all genes was similar between adenocarcinomas from smokers and never smokers, although the prevalence for methylation of the p16 gene tended to be higher in smokers compared to never smokers. A significantly higher prevalence for p16 methylation was seen in central vs. peripheral lung tumors. At least 1 gene was methylated in 35% of stage I tumors, whereas 2 and :3 genes were methylated in 40% and 16% of tumors, respectively. Methylation of all genes was independent of K-ras mutation, whereas methylation of the DAPK and RASSFIA genes was positively associated. Environmental tobacco smoke, the strongest lung cancer risk factor among never smokers, induces adenocarcinoma in part through inactivation of the p16, DAPK and RASSFIA genes. Adenocarcinomas may develop through 2 distinct processes: multiple gene inactivations through promoter hypermethylation and activation of the K-ras gene. (C) 2004 Wiley-Liss, Inc.

Loss of expression of the death-associated protein (DAP)-kinase gene by aberrant promoter methylation may play an important role in cancer development and progression. The purpose of this investigation was to determine the commonality for inactivation of the DAP-kinase gene in adenocarcinomas induced in mice by chronic exposure to mainstream cigarette smoke, the tobacco carcinogens 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and vinyl carbamate, and the occupational carcinogen methylene chloride. The timing for inactivation was also determined in alveolar hyperplasias that arise in lung cancer induced in the A/J mouse by NNK. The DAP-kinase gene was not expressed in three of five NNK-induced lung tumor-derived cell lines or in a spontaneously arising lung tumor-derived cell line. Treatment with 5-aza-2′-deoxycytidine restored expression; dense methylation throughout the DAP-kinase CpG island detected by bisulfite sequencing supported methylation as the inactivating event in these cell lines. Methylation-specific PCR detected inactivation of the DAP-kinase gene in 43% of tumors associated with cigarette smoke, a frequency similar to those reported in human non-small cell lung cancer. In addition, DAP-kinase methylation was detected in 52%, 60%, and 50% of tumors associated with NNK, vinyl carbamate, and methylene chloride, respectively. Methylation was observed at similar prevalence in both NNK-induced hyperplasias and adenocarcinomas (46% versus 52%), suggesting that inactivation of this gene is one pathway for tumor development in the mouse lung. Bisulfite sequencing of both premalignant and malignant lesions revealed dense methylation, substantiating that this gene is functionally inactivated at the earliest histological stages of adenocarcinoma development. This study is the first to use a murine model of cigarette smoke-induced lung cancer and demonstrate commonality for inactivation by promoter hypermethylation of a gene implicated in the development of this disease in humans.

The retinoic acid receptor beta (RAR-beta) gene encodes one of the primary receptors for retinoic acid, an important signaling molecule in lung growth, differentiation and carcinogenesis. RAR-beta has been shown to be down-regulated by methylation in human lung cancer. We have used previously lung tumors induced in mice to evaluate the timing and effect of specific carcinogen exposures on targeting genes altered in human lung cancer. These studies were extended to characterize the role of methylation of the RAR-beta gene in murine lung cancers. After treatment with the demethylating agent 5-aza-2'-deoxycytidine (DAC), RAR-beta was re-expressed in silenced cell lines or expressed at a higher rate than without DAC, supporting methylation as the inactivating mechanism. Bisulfite sequencing detected dense methylation in the area of the CpG island that contained the 5' untranslated region and the first translated exon in non-expressing cell lines, compared with minimal and heterogeneous methylation in normal mouse lung. Methylation-specific PCR revealed that this gene is targeted differentially by carcinogen exposures with the detection of methylated alleles in virtually all primary tumors associated with cigarette smoke or 4-methylnitrosamino-1-(3-pyridyl)-butanone (NNK) in contrast to half of tumors induced by methylene chloride or vinyl carbamate. RAR-beta methylation was also detected in 54% of preneoplastic hyperplasias induced by treatment with NNK. Bisulfite sequencing of both premalignant and malignant lesions detected dense methylation in the same area observed in cell lines, substantiating that this gene is functionally inactivated at the earliest histologic stage of adenocarcinoma development. These studies demonstrate that aberrant methylation of RAR-beta is an early and common alteration in murine lung tumors induced by several environmentally relevant exposures.

Adenocarcinoma (AC) is the most common type of lung cancer diagnosed in the United States, comprising up to 40% of tumors in smokers and 50-80% of tumors in never-smokers. Exposures to cigarette smoke, direct or second-hand, and radiation in the form of radon progeny are the major risk factors for lung AC in both smokers and never-smokers. The goal of the current study was to determine the prevalence for O(6)-methylguanine-DNA methyltransferase (MGMT) promoter methylation in a large sample of central or peripheral ACs from smokers (n = 157), former uranium miners (n = 34), and never-smokers (n = 46). The mutation rate at codon 12 of the K-ras gene was determined to assess whether activation of this oncogene was associated with MGMT methylation. The overall prevalence for MGMT methylation was 51%. Significantly more tumors from never-smokers than smokers exhibited MGMT methylation (66 versus 47%, respectively). In contrast, exposure to radon through uranium mining did not affect the prevalence for methylation. The frequency of MGMT methylation was increased significantly in association with tumor stage. K-ras mutations were detected in 24% of all ACs and 22, 24, and 28% of tumors from never-smokers, smokers, and miners, respectively. Alterations in both the K-ras and MGMT genes were seen in only 11% of ACs. Kaplan-Meier survival estimates did not reveal any difference between patient survival with or without MGMT methylation. In contrast, survival was significantly reduced over the initial 60 months after diagnosis for patients with a transition mutation in the K-ras gene compared with those with a transversion mutation. This investigation demonstrates that MGMT promoter hypermethylation is a common event in the progression of early stage AC of the lung. We have shown that the incidence of MGMT methylation was significantly higher in never-smokers than smokers and have detected a higher frequency of mutations within the K-ras gene than previously reported in never-smokers. This study also suggests that K-ras activation is independent of MGMT methylation.

Inactivation of the p16INK4a (p16) tumor suppressor gene by promoter hypermethylation and mutation within exon 3 of beta-catenin represent two of the more common gene alterations in human hepatocellular carcinoma (HCC). One exposure implicated in the development of liver cancer is hepatitis B or C viral infection, which causes chronic destruction and regeneration of liver parenchyma. Treatment of rats with high doses of the tobacco-specific nitrosamine 4-methylnitrosamino-1-(3-pyridyl)-1-butanone (NNK) also causes liver toxicity and a high incidence of tumors. The purpose of the current investigation was to define the prevalence of genetic alterations in p16 and beta-catenin in NNK-induced rat liver cancer to determine if the molecular mechanisms seen in human tumors are the same in this animal model. DNA isolated from 15 adenomas and 14 carcinomas was examined for methylation of p16 by methylation-specific PCR. p16 methylation was detected in five of 15 adenomas and eight of 14 carcinomas (45% of all tumors). Methylation of p16 was extensive within the 5'-untranslated region and exon 1alpha, areas shown to correlate with loss of gene transcription. Liver tumors were also screened for mutations within exon 3 of beta-catenin. Single strand conformation polymorphism and DNA sequencing revealed five mutations in four of 29 tumors (14%). Mutations were present in three adenomas and one carcinoma and were located within codons 33, 36 or 37. All mutations resulted in amino acid substitutions; three of these mutations occurred at potential serine phosphorylation sites. Our results link two important regulatory pathways altered in human HCC to cancer induced in the rat NNK model. The fact that common genetic alterations are observed between rodent and human HCC suggests that the rat NNK model could be useful for identifying additional genetic alterations critical to the initiation of HCC.