| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
Lung Cancer Program, Lovelace Respiratory Research Institute, Albuquerque, New Mexico 87108 [L. C. P., K. K. D., D. M. K., S. A. B.]; Keck School of Medicine, Department of Preventative Medicine and Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, California 90089-9021 [F. D. G., T. K.]; Karmanos Cancer Institute and Wayne State University Department of Oncology and Internal Medicine, Detroit, Michigan 48201 [A. G. S.]; and Departments of Family and Community Medicine and Pathology, University of New Mexico, Albuquerque, New Mexico [T. J. B.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
These premises are supported by studies in laboratory animals examining BaP and NNK. The studies showed differences in retention patterns for the two carcinogens. The BaP accumulated in the central lung, whereas NNK accumulated in the peripheral lung (5 , 6) , indicating that differences in lipid solubility can affect the distribution of carcinogens in the lung. Because the type of DNA damage produced after exposure to BaP and NNK is different (e.g., bulky versus small DNA adducts), the genes affected in developing central and peripheral ACs may also be different. Similarly, genes altered in resulting tumors because of these exposures may also differ because tobacco and radiation lead to quantitative and qualitative differences in the spectrum of DNA damage. Supporting this hypothesis, inactivation of the ER gene by promoter hypermethylation was more frequent in never-smokers versus smokers and in rodent tumors induced by radiation than by NNK (7) . In contrast, the p16 gene was inactivated by promoter methylation more frequently in rodent tumors induced by NNK than radiation (8) .
Two genes frequently altered in human lung AC that play an important role in its pathogenesis are K-ras and MGMT. Mutations in K-ras, predominantly in codon 12, have been observed in 30–40% of lung ACs from smokers and former uranium miners but in no more than 11% of ACs from never-smokers (9, 10, 11, 12) . Another gene frequently inactivated in NSCLC is MGMT (13) . It is a DNA repair enzyme that protects cells from the carcinogenic effects of alkylating agents by removing adducts from the O6 position of guanine. Failure to repair this DNA adduct can lead to transition mutations in genes such as K-ras. The predominant mechanism for MGMT gene inactivation is aberrant methylation of a CpG island within the promoter region that leads to transcriptional silencing (13 , 14) . Limited studies (<50 tumors) of ACs from smokers detected MGMT methylation in 21% of tumors (13 , 15) . Tumor types with silenced MGMT often contain mutations in the K-ras gene (13) , suggesting that loss of MGMT may cause the ensuing mutation. Recent observations in colorectal and gastric tumors substantiate this relationship, showing a clear association between silencing of MGMT by promoter hypermethylation and G to A transition mutations in the K-ras gene (14 , 16) . Thus, if the MGMT gene is inactivated early in tumor development, this could predispose the cancer cell to the acquisition of somatic mutations. This, in turn, could lead to the development of aggressive malignant tumors and shortened survival.
The goal of the current study was to determine the prevalence for promoter methylation of MGMT and K-ras mutation by exposure, tumor location, and stage in ACs from smokers, former uranium miners, and never-smokers. The effect of MGMT methylation and K-ras mutation on survival was also assessed. 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.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Never-smokers (<100 cigarettes in their lifetime) were identified retrospectively after their participation in two telephone interviews in which their smoking history was obtained and then verified. Additional data collected from never-smokers included basic demographics and health, occupational, and family histories. Medical records verified the smoking status of all smokers and miners. When available, more detailed smoking information such as pack-years and duration was collected. The tumor registries provided information on gender, ethnicity, survival, age of diagnosis, and age of death. Radon exposures ranged from 2 to 4196 working-level months in former uranium miners.
Tumors were obtained from 193 non-Hispanic whites and 44 Hispanics, Native Americans, or African Americans. For statistical comparisons evaluating effects of race/ethnicity, results from Hispanics, Native Americans, and African Americans were grouped into one category.
Histological Diagnosis.
All cases were reviewed histologically in concert with review of the pathology reports. Tumor location was evaluated based on the tumor’s relationship to the pleura and large bronchi. On the basis of the tumor position, a central or peripheral location was conferred. The tumor was then classified broadly as AC or mixed AC/other tumor (such as AC with squamous cell carcinoma). The WHO International Histological Classification of Tumors (17)
was used to further subclassify the ACs. This detailed subclassification was condensed into four categories for statistical analyses: (a) bronchoalveolar carcinoma; (b) acinar, papillary, and solid with mucin; (c) mixed AC that comprised adenosquamous carcinoma, mixed small cell/AC, and spindle cell; and (d) large cell undifferentiated.
Microdissection and Nucleic Acid Isolation.
DNA was obtained by microdissection. Some lesions were contaminated extensively with normal tissue or were small in size. Therefore, for the small lesions it was essential to include normal appearing cells to ensure that after bisulfite modification and column clean up of the DNA template enough sample remained to conduct the MSP assay as described below. DNA was isolated by digestion of the tissue with Pronase (1%) followed by phenol-chloroform extraction and ethanol precipitation.
MSP.
The methylation status of the MGMT promoter was determined using our nested, two-stage MSP assay (18)
. Briefly, after amplification by stage I primers, two different stage II amplifications were performed. In one stage II amplification, the primers used recognize a sequence in which CpG sites are unmethylated. These primers allowed us to verify the ability to amplify the target region because stage I products are often not detectable with amplification of DNA from formalin-fixed tissue. Depending on the band intensity obtained using the unmethylated stage II primers, 2–7 µl of the stage I PCR product were used in the MSP with stage II primers that recognize only sequence in which CpG sites are methylated. Normal human lung tissue and cell lines positive for MGMT methylation (SkuLu1) served as negative and positive controls, respectively.
The use of relatively large amounts of DNA from the stage-I PCR could lead to false priming, i.e., amplification of unmethylated alleles. Therefore, methylation-positive tumor samples were also analyzed a second time where the resulting PCR product was subjected to methylation-specific restriction enzyme digestion. MSP was performed in triplicate for each sample. One sample was incubated with the restriction enzyme TaqI, which cuts at one CpG site (TCGA) within the amplified region of the methylated MGMT promoter. A second sample was digested with BstUI, which cuts at two CpG sites (CGCG) within the amplified sequence. Thus, the methylation states of three CpG sites were assessed. If either restriction site showed digestion, the sample was scored as positive for methylation.
RFLP Analysis of the K-ras Gene.
All AC samples were analyzed for mutation in codon 12 of the K-ras gene using the BstN1 mutant allele enrichment method where the resulting PCR product was digested with the BstNI restriction enzyme (New England BioLabs) to generate fragments of 27 and 89 bp for wild-type codon 12 (19)
. Point mutations at either of the first two positions of codon 12 did not digest.
Sequencing of MGMT and K-ras.
DNA from five methylation-positive lung tumors was amplified using methylation-specific primers. Two of the tumors cut at both restriction sites, whereas the other three cut at only one site. The 123-bp PCR product was ligated into the PCR II vector using the TA cloning kit (Invitrogen, San Diego, CA). Five or six clones from each AC sample were commercially sequenced in both directions (UNM Center for Genetics in Medicine, Albuquerque, NM). Three to six clones from two methylation negative tumors were also sequenced.
After identification of tumors containing mutant K-ras alleles, the PCR products were directly sequenced using the dideoxy chain termination method with Sequenase DNA polymerase (USB, Cleveland, OH). Sequencing primers (5'-TGATTCTGAATTAGC-3'), end labeled with 32P by T4 polynucleotide kinase (USB), were annealed to 500 ng of heat-denatured, amplified DNA and extended for 1.5 min. The reaction products were separated on an 8% acrylamide denaturing gel and visualized by exposure to Kodak BMR film at -80°C.
IHC.
Sections of paraffin-embedded tissue were deparaffinized and antigen retrieved by microwaving. Sections were incubated overnight at 4°C with the primary antibody (1:10,000), mouse anti-MGMT monoclonal antibody (Abcam, Cambridge, United Kingdom). Slides were then rinsed and incubated with the secondary antibody, biotinylated antimouse IgG, at a 1:200 dilution for 1 h. Avidin-biotin complex was formed using the Vectastain ABC kit (Vector Labs, Inc., Burlingame, CA). Sections were then developed using 3,3'-diaminobenzidine (DAB Peroxidase Substrate kit; Vector Labs) and counterstained with hematoxylin. Negative controls were conducted by replacing the primary antibody with mouse IgG. A section was considered to be immunohistochemically positive for MGMT if tumor nuclei were stained.
Data Analysis.
Stratified analyses were used to assess associations between MGMT promoter methylation status and the presence and type of K-ras mutations in tumors with respect to the characteristics of the individual and the tumor (e.g., stage and histological subtype). Differences in distribution of determinants (e.g., smoking) in methylated and unmethylated tumors and in tumors with and without K-ras mutations were tested using 
2 statistics for categorical variables or t test for continuous variables. Logistic regression models were used to assess joint effects. Stratified Kaplan-Meier survival estimates and fitted proportional hazard models were used to estimate the relative risk of death associated with methylation and mutation status, adjusted for age, race/ethnicity, gender, stage at diagnosis, and smoking. Data were also analyzed using a Cox regression analysis model with survival data censored at 60 months past diagnosis to meet the assumptions of proportional hazards. All analyses used SAS software (version 8e; SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
presents selected demographic and clinicopathological variables from cases of AC separated by exposure. Age at diagnosis and race did not differ statistically between smokers, miners, and never-smokers, and smoking dose did not statistically differ between smokers or miners. Former miners were all men, and the majority of never-smokers were women. Tumor location (central versus peripheral), classification, or stage did not differ statistically across the three exposure groups.
|
, Table 2
). Five tumors shown to be methylated by MSP and positive by restriction digestion at either the BstUI and/or TaqI site were selected for sequencing to assay the methylation status of seven additional CpG sites (nos. 1, 2, 4, 5, 8–10; Fig. 2
). Methylation densities ranged from 60 to 90% within the 10 CpG sites evaluated in the MGMT promoter, a region that correlates with loss of MGMT expression (13)
. Sequencing of two tumors negative for methylation revealed no methylated CpG sites. Unmethylated alleles were also simultaneously detected in the majority of tumors because of contamination with stromal and inflammatory cells. No methylation was detected in DNA from our normal lung control.
|
|
|
. The increased frequency of MGMT methylation in never-smokers was independent of age at diagnosis, gender, race, and tumor stage (P = 0.04). No difference in methylation frequency was observed with respect to tumor site, gender, or race (Table 2)
.
The frequency of MGMT methylation increased significantly with tumor stage (P = 0.04) and tumor size (T1–T2 versus T3–T4; P = 0.05; Table 2
). Multivariate analysis of the association between methylation frequency and tumor stage also indicated a statistically significant relationship between methylation and tumor stage when stage I tumors were compared with stages II–IV (P = 0.01). This association was independent of age, gender, race, and smoking status.
IHC was used to examine the relationship between MGMT methylation status and MGMT protein expression. In randomly selected tumors, MGMT protein staining was concordant with MSP results in 24 of 29 tumors (83%). Seven of 10 tumors positive for methylation as shown by MSP and confirmed by RFLP showed no protein staining within the nucleus, whereas 17 of 19 tumors negative for methylation by MSP showed positive nuclear staining (see Fig. 3
for representative results).
|
. In most tumors with a mutation, a wild-type allele was also detected, consistent with the dominant nature of the K-ras mutation. No difference was seen in K-ras mutation frequency between tumor location or gender. Tumors from non-Hispanic whites showed a trend (P = 0.06) toward a lower frequency of K-ras mutation when compared with tumors from other races. Multivariate analysis supported this trend (P = 0.12) after adjustment for age at diagnosis, gender, smoking status, and tumor stage.
K-ras Mutation Profile and Relationship to MGMT Methylation.
Thirty-nine percent of all observed K-ras codon 12 mutations were transitions; 61% were transversions. The predominant mutations observed were a GGT’ TGT transversion mutation (35% of all mutations) and a GGT’ GAT transition mutation (24%). Other mutations included GGT’ AGT (15%), GGT’ GTT (13%), GGT’ GCT (9%), and GGT’ CGT (4%). No significant relationship was seen between K-ras mutation profile (transition versus transversion mutation) and tumor location or exposure type. Among smokers, 88% of the transversion mutations observed were present in persons with 
30 pack-year smoking histories (P = 0.02 for comparison with 
30 pack years). In contrast, the frequency for transition mutations did not differ as a function of smoking dose.
Twenty-six of 237 ACs (11%) examined in this study showed alterations in both the K-ras and MGMT genes. Within those tumors, the prevalence of K-ras transversion mutations exceeded that of transition mutations (17 versus 9 tumors, respectively). Conversely, 12 tumors contained a transition mutation without MGMT methylation.
Effect of MGMT Methylation and K-ras Mutation on Prognosis of Patients with AC.
Kaplan-Meier survival estimates did not reveal any difference between patient survival with or without methylation of the MGMT gene (Table 3)
. 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 (Table 3
, Fig. 4
). Univariate analysis of exposure and demographics showed that survival was reduced in miners versus nonminers (P = 0.003) and in males versus females (P = 0.002; Table 3
). Clinicopathological factors also indicated reduced survival for individuals with central versus peripheral lung ACs (P = 0.004; Table 3
). Survival was also reduced in persons with bronchoalveolar carcinoma compared with other tumor subtypes (P = 0.02) and as a function of tumor stage at the time of diagnosis (P < 0.0001). A multivariate Cox proportional hazard regression analysis was subsequently performed using K-ras mutation type, MGMT methylation, exposure, demographics, and clinicopathological factors. The only independent prognostic marker was tumor stage (P < 0.0001; hazard ratio = 13.02).
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gene silencing by aberrant promoter methylation is emerging as a critical process for both initiation and progression of AC. Our previous studies substantiated that the inactivation of the p16 gene is likely one of the important initiating events for NCSLC; p16 is inactivated in alveolar hyperplasias and premalignant lesions in the bronchial epithelium from cancer-free smokers (20 , 21) . The current study indicates that inactivation of MGMT is likely a later event occurring closer to the development of malignancy. This conclusion is based on our finding of a significant difference in prevalence for inactivation between stages I and II–IV tumors and tumor size. Furthermore, only 3 of 40 (8%) biopsy samples with histologies, including normal, hyperplasia, metaplasia, and dysplasia from heavy smokers showed methylation of the MGMT gene (S. A. Belinsky, unpublished results). The different timing for inactivation of the p16 and MGMT genes correlates strongly with their prevalence for detection in sputum from cancer-free, heavy smokers (18) . Assembling a temporal genetic map for development and progression of lung cancer is critical to our other studies that are establishing a panel of genes inactivated by methylation to evaluate their sensitivity and specificity for predicting lung cancer risk through sputum analysis. Previous studies predicting squamous cell carcinoma showed the utility of the MGMT gene as a biomarker in sputum, and the current studies elucidate both the timing and extent of inactivation of this gene in AC.
Functional inactivation of the MGMT gene in tumors was evident based on loss of MGMT protein. The 83% concordance seen between MSP results and IHC corroborates previous studies that showed loss of MGMT protein expression associated with methylation in diffuse large B-cell lymphoma (22) and colorectal and brain tumors (13) . Moreover, bisulfite sequencing of a subset of tumors revealed dense methylation in the region of the gene amplified by MSP and previously associated with transcriptional silencing. The discordant results observed for three tumors in which methylation was detected in the presence of gene expression could indicate the timing seen for inactivation in this tumor type and reflect methylation of only one allele of this gene.
This is the first study to examine the frequency for MGMT methylation in AC from never-smokers. The significantly higher incidence for methylation in tumors from never-smokers than smokers may implicate carcinogens present in environmental exposures for this population. ETS is the most widely studied lung cancer risk factor among never-smokers (23)
. Twice as much nicotine is emitted in ETS as in mainstream smoke, suggesting that the never-smoker would have ample opportunity for exposure to tobacco-specific nitrosamines and polyaromatic hydrocarbons as well as 4-aminobiphenyl, a carcinogen that is enriched 
30-fold in ETS (23)
. We did not attempt to evaluate MGMT methylation as a function of ETS exposure because of the documented difficulty in quantitating ETS exposure and the fact that ETS exposure data were lacking on >35% of our never-smokers. There is precedent for targeting genes for inactivation by methylation through environmental exposure. Our previous studies on ER methylation showed a higher prevalence for inactivation of this gene in never-smokers than smokers and in rodent lung tumors induced by radiation than the tobacco carcinogen NNK (7)
. Moreover, recent studies have also shown that ACs from never-smokers display a distinct pattern of allelic imbalance compared with tumors from smokers (24)
. The precise mechanisms that underlie targeting of genes such as MGMT and ER for inactivation in AC from the never-smoker remain elusive. However, it is likely that germ-line defects in pathways that modulate carcinogen activation, detoxification, repair, and most recently de novo cytosine methylation are likely to play a significant role in susceptibility for lung cancer in this population (23
, 25
, 26)
. Studies from our laboratory have recently demonstrated a link between germ-line polymorphisms in genes involved in the protection of the cell from heritable DNA damage from tobacco exposure and epigenetic-mediated silencing of cancer genes by promoter hypermethylation (27)
. Subjects at risk for lung cancer with loss of either a wild-type allele of the NADPH quinone reductase or glutathione S-transferase P1 gene had a significantly increased risk for methylation of MGMT and/or p16 in their sputum.
Although mutation of the K-ras gene has been studied extensively in AC from smokers, fewer studies, mostly with small sample sizes (generally <20 cases), have been conducted in tumors from never-smokers (28)
. Vahakangas et al. (12)
, who evaluated 117 cases of ACs in never-smokers, reported an 11% prevalence for K-ras mutations that were largely transversions. In our study, 22% of ACs from never-smokers harbored a mutation in the K-ras gene, a prevalence similar to that observed in tumors from smokers. The predominance of transversion mutations in tumors from both smokers and never-smokers parallels previous studies of smokers and never-smokers and implicates exposure to polyaromatic hydrocarbons such as BaP, which is found in mainstream smoke and ETS. In fact, among smokers in our study, 88% of transversion mutations were present in tumors from heavy versus light smokers. Support for BaP in the etiology of this mutation was demonstrated by the targeting and persistence of DNA adducts at codon 12 of the K-ras gene after exposure of normal human bronchial epithelial cells to BaP diol expoxide (29)
. The susceptibility of the never-smoker to K-ras transversion mutations may be additionally compounded by the fact that women with a polymorphism in cytochrome P4501A1 that metabolizes BaP show an elevated risk for lung cancer (30)
. In fact, women comprised 75% of the never-smokers with AC in our study. The difference in prevalence for K-ras mutation in never-smokers from our study and others could stem from many variables, including enrichment for tumor cells through microdissection and use of the BstNI selection assay before genomic sequencing. The majority of earlier studies did not select for K-ras mutations before sequencing or conduct microdissection. Thus, with contaminating stroma and inflammatory cells often accounting for >50% of the tumor mass and a sensitivity of direct sequencing for mutant allele detection of 20%, it is likely that some mutations went undetected.
In this investigation, we saw no relationship between MGMT hypermethylation promoter and the occurrence of K-ras transition mutations. Tumors with a transition mutation were more likely to possess wild-type than methylated MGMT, suggesting that these two genetic alterations develop by independent pathways. This finding contrasts with studies of colorectal and gastric cancer in which MGMT methylation was strongly associated with the occurrence of transition mutations in K-ras. Transition mutations predominate in these tumors types, accounting for 58 and 100% of all observed K-ras codon 12 mutations in colorectal and gastric cancers, respectively (14 , 16) . However, other tumor types that show a high rate of K-ras transition mutations such as pancreatic carcinoma do not exhibit MGMT methylation (31 , 32) . This suggests that inactivation of MGMT is not obligatory for the development of K-ras mutations. The timing for MGMT methylation and K-ras mutation is another key difference between colon and lung that could account for the difference seen. In the colon, methylation precedes mutation in the small colonic adenomas, whereas mutation is first seen in the larger adenomas (14) . In contrast, the K-ras gene has been frequently mutated in atypical adenomatous hyperplasia, one of the proposed precursor lesions to AC (33) . The facts that MGMT methylation is rarely found in histological lesions in the bronchial epithelium and shows a higher prevalence for inactivation in more advanced AC support the premise that K-ras mutation precedes methylation of this gene in AC. Although transition mutations were not associated with methylation of the MGMT gene, they did predict for decreased survival in the initial 60 months after cancer diagnosis. The fact that this variable was not significant after adjustment for tumor stage suggests that the presence of transition mutations could predict for tumor aggressiveness.
Improving survival from AC will clearly depend on continued identification of the pathways in which their alterations are critical to the initiation and progression of this fatal disease. Such discoveries will lead to the development of therapies that rely on modulating altered pathways rather than direct cytotoxicity. The current study provides strong evidence that inactivation of the DNA repair gene, MGMT, is one alteration that may be linked to tumor progression. Its high prevalence in ACs from both smokers and never-smokers additionally reinforces the likelihood that inactivating this gene renders the cell more sensitive to the propagation of DNA damage associated with chronic exposure to mainstream smoke and ETS.
|
FOOTNOTES |
|---|
1 This investigation was supported by NIEHS ES08801, P20 ES09871, ES05836, CA60691, NO1-CN65064, and the University of New Mexico Cancer Center Sequencing Shared Facility. 
2 To whom requests for reprints should be addressed, at Lovelace Respiratory Research Institute, 2425 Ridgecrest Drive, Southeast, Albuquerque, NM 87108. Phone: (505) 348-9456; Fax: (505) 348-4990; E-mail: sbelinsk{at}LRRI.org
3 The abbreviations used are: AC, adenocarcinoma; NNK, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone; ER, estrogen receptor; MGMT, O6-methylguanine-DNA methyltransferase; IHC, immunohistochemistry; MSP, methylation-specific PCR; ETS, environmental tobacco smoke; BaP, benzo(a)pyrene.
Received 1/20/03. Revised 4/ 4/03. Accepted 4/29/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. E. Sandercock, J. N. Hahn, L. Li, H.A. Luchman, J. L. Giesbrecht, L. A. Peterson, and F. R. Jirik Mgmt deficiency alters the in vivo mutational spectrum of tissues exposed to the tobacco carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) Carcinogenesis, April 1, 2008; 29(4): 866 - 874. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Schwartz, G. M. Prysak, C. H. Bock, and M. L. Cote The molecular epidemiology of lung cancer Carcinogenesis, March 1, 2007; 28(3): 507 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J MARSIT, M. R KARAGAS, A. SCHNED, and K. T KELSEY Carcinogen exposure and epigenetic silencing in bladder cancer. Ann. N.Y. Acad. Sci., September 1, 2006; 1076: 810 - 821. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Shao, J. Berenguer, A. C. Borczuk, C. A. Powell, T. K. Hei, and Y. Zhao Epigenetic Inactivation of Betaig-h3 Gene in Human Cancer Cells. Cancer Res., May 1, 2006; 66(9): 4566 - 4573. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Vuillemenot, J. A. Hutt, and S. A. Belinsky Gene Promoter Hypermethylation in Mouse Lung Tumors Mol. Cancer Res., April 1, 2006; 4(4): 267 - 273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-Y. Zhong, Y.-M. Zhou, G. C. Douglas, H. Witschi, and K. E. Pinkerton MAPK/AP-1 signal pathway in tobacco smoke-induced cell proliferation and squamous metaplasia in the lungs of rats Carcinogenesis, December 1, 2005; 26(12): 2187 - 2195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Hutt, B. R. Vuillemenot, E. B. Barr, M. J. Grimes, F. F. Hahn, C. H. Hobbs, T. H. March, A. P. Gigliotti, S. K. Seilkop, G. L. Finch, et al. Life-span inhalation exposure to mainstream cigarette smoke induces lung cancer in B6C3F1 mice through genetic and epigenetic pathways Carcinogenesis, November 1, 2005; 26(11): 1999 - 2009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Vineis and K. Husgafvel-Pursiainen Air pollution and cancer: biomarker studies in human populations Carcinogenesis, November 1, 2005; 26(11): 1846 - 1855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Belinsky Silencing of genes by promoter hypermethylation: key event in rodent and human lung cancer Carcinogenesis, September 1, 2005; 26(9): 1481 - 1487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Russo, A. Thiagalingam, H. Pan, J. Califano, K.-h. Cheng, J. F. Ponte, D. Chinnappan, P. Nemani, D. Sidransky, and S. Thiagalingam Differential DNA Hypermethylation of Critical Genes Mediates the Stage-Specific Tobacco Smoke-Induced Neoplastic Progression of Lung Cancer Clin. Cancer Res., April 1, 2005; 11(7): 2466 - 2470. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lu, J.-C. Soria, X. Tang, X.-C. Xu, L. Wang, L. Mao, R. Lotan, B. Kemp, B. N. Bekele, L. Feng, et al. Prognostic Factors in Resected Stage I Non-Small-Cell Lung Cancer: A Multivariate Analysis of Six Molecular Markers J. Clin. Oncol., November 15, 2004; 22(22): 4575 - 4583. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Mauderly, A. P. Gigliotti, E. B. Barr, W. E. Bechtold, S. A. Belinsky, F. F. Hahn, C. A. Hobbs, T. H. March, S. K. Seilkop, and G. L. Finch Chronic Inhalation Exposure to Mainstream Cigarette Smoke Increases Lung and Nasal Tumor Incidence in Rats Toxicol. Sci., October 1, 2004; 81(2): 280 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Hansel, A. Rahman, M. House, R. Ashfaq, K. Berg, C. J. Yeo, and A. Maitra Met Proto-Oncogene and Insulin-Like Growth Factor Binding Protein 3 Overexpression Correlates with Metastatic Ability in Well-Differentiated Pancreatic Endocrine Neoplasms Clin. Cancer Res., September 15, 2004; 10(18): 6152 - 6158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Dulaimi, I. I. de Caceres, R. G. Uzzo, T. Al-Saleem, R. E. Greenberg, T. J. Polascik, J. S. Babb, W. E. Grizzle, and P. Cairns Promoter Hypermethylation Profile of Kidney Cancer Clin. Cancer Res., June 15, 2004; 10(12): 3972 - 3979. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. C. Pulling, B. R. Vuillemenot, J. A. Hutt, T. R. Devereux, and S. A. Belinsky Aberrant Promoter Hypermethylation of the Death-Associated Protein Kinase Gene Is Early and Frequent in Murine Lung Tumors Induced by Cigarette Smoke and Tobacco Carcinogens Cancer Res., June 1, 2004; 64(11): 3844 - 3848. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Belinsky, D. M. Klinge, K. C. Liechty, T. H. March, T. Kang, F. D. Gilliland, N. Sotnic, G. Adamova, G. Rusinova, and V. Telnov Plutonium targets the p16 gene for inactivation by promoter hypermethylation in human lung adenocarcinoma Carcinogenesis, June 1, 2004; 25(6): 1063 - 1067. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||
represents unmethylated CpGs; 
represents methylated CpGs. Each row represents one clone. The methylation status of CpG no. 3 was also assessed by the methylation-specific restriction enzyme TaqI, whereas that of CpGs nos. 6 and 7 were assessed by BstUI.
