Graduate Institute of Cancer Biology and Drug Discovery, Taipei Medical University, Taipei, Republic of China
Corresponding author: Huei Lee, PhD, Professor, Graduate Institute of Cancer Biology and Drug Discovery, Taipei Medical University, Room 5, 12F, No. 3, Park Street, Nankang District, Taipei, Taiwan 115, ROC; Fax: (011) 886-2-26558562; email@example.com
Lung cancers in women, in nonsmokers, and in patients with adenocarcinoma from Asia have more prevalent mutations in the epidermal growth factor receptor (EGFR) gene than their counterparts. However, the etiology of EGFR mutations in this population remains unclear. The authors hypothesized that the human papillomavirus (HPV) type 16/18 (HPV16/18) E6 oncoprotein may contribute to EGFR mutations in Taiwanese patients with lung cancer.
One hundred fifty-one tumors from patients with lung cancer were enrolled to determine HPV16/18 E6 and EGFR mutations using immunohistochemistry and direct sequencing, respectively. Levels of 8-oxo-7,8-dihydro-2′-deoxyguanosine (8-oxo-dG) in lung tumors and cells were evaluated using immunohistochemistry and liquid chromatography-mass spectrometry/mass spectrometry. An supF mutagenesis assay was used to determine H2O2-induced mutation rates of lung cancer cells with or without E6 expression.
Patients with E6-positive tumors had a greater frequency of EGFR mutations than those with E6-negative tumors (41% vs 20%; P = .006). Levels of 8-oxo-dG were correlated with EGFR mutations (36% vs 16%; P = .012). Two stable clones of E6-overexpressing H157 and CL-3 cells were established for the supF mutagenesis assay. The data indicated that the cells with high E6 overexpression had higher H2O2-induced SupF gene mutation rates compared with the cells that expressed lower levels of E6 and compared with vector control cells.
Lung cancer is the leading cause of cancer death and predominantly is induced by tobacco smoke. However, approximately 25% of all lung cancer worldwide is not attributed to smoking.[1, 2] In addition, nonsmokers with lung cancer are identified more often Asian women than among Caucasian women.[3, 4] Lung cancer in nonsmokers has recently been recognized as a disease entity distinct from tobacco smoke-associated lung cancer.[5, 6]
The activation of epidermal growth factor receptor (EGFR) by its gene mutations, including exon 19 in-frame deletions and exon 21 L858R substitution, are currently recognized as the most potent biologic predictors of EGFR tyrosine kinase inhibitor (TKI) sensitivity. These mutations are observed in approximately 40% to 60% of lung cancers in never-smokers, compared with approximately 10% to 20% of tobacco smoke-related lung cancers.[8, 9] Therefore, an exploration of the etiologic factor(s) in EGFR mutations may be useful in the prevention of lung cancer in nonsmokers, particularly among women.
A high frequency of EGFR mutations are common in nonsmoking women with lung adenocarcinoma in Asian populations, including Taiwanese.[3, 4] Nonsmokers with lung cancer more prevalently present with an adenocarcinoma subtype. The etiology of EGFR mutation in this population with lung cancer remains unknown. EGFR mutations are uncommon in tobacco smoke-related lung adenocarcinoma, occurring in only 15% of tumors from former smokers and in 6% of current smokers, compared with 52% of tumors in nonsmokers. In addition, environmental tobacco smoke (ETS) has been negatively associated with EGFR mutations in patients with lung cancer. Therefore, etiologic factor(s) other than tobacco smokes may be associated with EGFR mutations in lung cancer.
Inflammatory-induced oxidative stress also has been linked with the development of lung adenocarcinoma. A recent noteworthy report indicated that the level of 8-hydroxy-2′deoxyguanosine (8-OH-dG), an oxidative stress biomarker, was closely associated with EGFR mutation in lung cancer. The 8-OH-dG levels also are correlated with grade of dysplasia in human papillomavirus (HPV)-associated cervical carcinogenesis. Our previous report indicated that HPV type 16/18 (HPV16/18) infection was associated with the development of lung cancer in Taiwanese women who were never-smokers. The HPV16/18 E6 oncoprotein is expressed in lung tumors and is related to tumor suppressor protein 53 (p53) inactivation. A recent report indicated that the presence of HPV16/18 DNA is associated with EGFR mutations in Japanese lung cancer patients. Therefore, we hypothesized that HPV16/18 E6 expression may contribute to the occurrence of EGFR mutations in Taiwanese patients with non–small cell lung cancer (NSCLC) because of an increase in 8-OH-dG levels induced by HPV16/18 infection.
MATERIALS AND METHODS
Study Patients and Cell Lines
Lung tumor specimens were collected between 1998 and 2004 from 151 patients with primary lung cancer in the Department of Thoracic Surgery, Taichung Veterans General Hospital, Taiwan. Among these, there were 98 men (65%) and 53 women (35%); 66 smokers (44%) and 85 nonsmokers (56%); 58 squamous cell carcinomas (38%) and 93 adenocarcinomas (62%); 50 stage I (33%), 28 stage II (19%), and 73 stage III (48%) tumors; 115 T1/T2 (76%) and 36 T3/T4 (24%) tumors; 90 lymph node-negative (N0) metastasis (60%), and 61 lymph node-positive (N1, N3, and N3) metastasis (40%). Patients were asked to submit a written informed consent for that was approved by the Institutional Review Board. The tumor type and stage of each collected specimen were determined histologically according to the World Health Organization classification. CL-3 cells were a gift from Professor P.-C. Yang (College of Medicine, National Taiwan University, Taipei, Taiwan). TL-1, and TL-4 cells were primarily established from patients' plural effusions, as described previously. H157 cells were kindly provided by Professor J. Y. Chen (Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan). The 4 cell lines were maintained in RPMI-1640 medium, which contained 10% fetal bovine serum supplemented with penicillin (100 U/mL) and streptomycin (100 mg/mL).
Plasmid Construction and Transfection Reaction
A small-hairpin RNA (shRNA) template was constructed using 2 complementary oligos, which, when partially annealed, created a loop region with a sequence complementary to E6 messenger RNA (mRNA). The oligos contained 19 nucleotides from the HPV16 E6 sequence (sh16E6) and a nonspecific control (NC) as follows: forward sh16E6, 5′-GATCCCGTTGTGTGATTTGTTAATTCAAGAGAT-3′; reverse sh16E6, 5′-AGCTAAAAACCGTTGTGTGATTTGTTAATCTTGAAA-3′; forward NC, 5′-GATCAACTACCGTTGTTATAGGTTTCAAGAGAT-3′; and reverse NC, 5′-AGCTAAAAAAACTACCGTTGTTATAGGTTCTTGAAA-3′. The shRNA template was cloned into the vector pCDNA-HU6 as described previously. HPV16 E6 was cloned, and E6 expression in H157 and CL-3 stable clones were established using a previous method. The HPV16 E6-specific shRNA plasmid and the HPV16 E6 expression plasmid (1 or 5 μg) were then mixed with TransFast transfection reagent (Promega, Madison, Wis) and added to TL1 and TL4 cells (1 × 106 cells). After 24 hours, interference with HPV16 E6 expression was confirmed by Western blot analysis and reverse transcriptase-polymerase chain reaction analyses.
Protein Extraction and Western Blot Analysis
Total protein was extracted from cells with a lysis buffer (100 mM Tris, pH 8.0; 1% sodium dodecyl sulfate [SDS]) followed by separation using SDS-polyacrylamide gel electrophoresis (PAGE) (12.5% gels) and Western blot procedures according to a previous report. Total protein was extracted from cells with a lysis buffer (100 mM Tris, pH 8.0; 1% SDS), and recovered protein concentrations were determined using the Bio-Rad protein assay kit (Bio-Rad, Hercules, Calif) followed by separation with SDS-PAGE (12.5% gels). After electrophoretic transfer to a polyvinyl dichloride membrane, nonspecific binding sites were blocked with 5% nonfat milk in Tris-buffered saline that contained 0.1% Tween-20. Signals were detected by incubating the membrane with antibodies against HPV16 E6, β-actin (Santa Cruz Biotechnology, Santa Cruz, Calif), mutL homolog 1 (MLH1), mutS homolog 2 (MSH2) (GeneTex Inc., Irvine, Calif), and p53 (Dako, Carpinteria, Calif) for 16 hours at 4°C, followed by a subsequent incubation with a peroxidase-conjugated secondary antibody (1:5000 dilution). Extensive washing with Tris-buffered saline/Tween-20 was performed after each step of antibody incubation to eliminate nonspecific binding. The protein bands were observed using enhanced chemiluminescence (Millipore, Billerica, Mass).
SupF Mutagenesis Assay
The mutagenesis assay procedure has been described previously. The shuttle vector pSP189 carries an supF gene (an aminoacylated amber suppressor transfer RNA) as the mutation target for studying mutagenesis in mammalian cells. Briefly, pSP189 (2 μg) was transfected into the control and E6-expressed H157 and CL3 cells by using Lipofectamine 2000 (Invitrogen, Carlsbad, Calif). The transfected cells were then cultured for another 72 hours in the absence or presence of H2O2 (0.1 or 1.0 μM). After a 72-hour period for repair and plasmid replication, the pSP189 plasmids were rescued from lung cells using the alkaline lysis extraction method. The purified plasmid DNA was digested with DpnI restriction endonuclease (New England Biolabs, Ipswich, Mass) to remove unreplicated plasmid DNA. The progeny pSP189 plasmids were then introduced into competent Escherichia coli MBM7070 using an Eppendorf Multiporator (2000 volts and 5 msec for the 1-mm gap cuvette; Eppendorf AG, Hamburg, Germany). E. coli MBM7070 carries an amber mutated lacZ gene as an indicator for supF compensation from the pSP189 plasmid. The transformed E. coli was then assayed for ampicillin resistance and mutations in the supF gene on agar plates containing ampicillin (50 μg/mL), 5-bromo-4-chloro-3-indoly-β-D-galactoside (X-gal) (120 μg/mL) and isopropyl β-D-1-thiogalactopyranoside (IPTG) (60 μg/mL). The E. coli that carried plasmids with supF mutations formed white or light blue colonies, whereas the E. coli that carrying plasmids with a functional supF gene formed blue colonies. The potential mutant colonies were selected and restreaked on agar plates to confirm their phenotypes.
Details of the immunohistochemistry were reported in a previous publication. Immunostaining results of HPV16/18 E6 expression in lung tumors were obtained from previous reports. Briefly, formalin-fixed, paraffin-embedded specimens were sectioned at a thickness of 3 μm. All sections were then deparaffinized in xylene; rehydrated through serial dilutions of alcohol; and washed in phosphate-buffered saline, pH 7.2, which was the buffer used for all subsequent washes. Sections were heated in a microwave oven twice for 5 minutes in citrate buffer, pH 6.0, and then incubated with anti-HPV16 E6 (1:100; Santa Cruz Biotechnology) and anti-8-oxo-7,8-dihydro-2′-deoxyguanosine (anti-8-oxo-dG) (1:100; Stressgen Biotechnologies, San Diego, Calif) for 2 hours at 25°C before incubating with 3% H2O2 for 10 minutes. The conventional streptavidin peroxidase method (LSAB Kit K675; Dako) was used to develop signals, and the cells were counterstained with hematoxylin. Negative controls were obtained by leaving out the primary antibody. For the scoring of HPV16/18 E6 immunostaining, “negative” lung tumor sections were those in which <10% of tumor cell nuclei in a paraffin section had positive immunostaining, and “positive” lung tumor sections were those in which >10% of tumor cell nuclei in a paraffin section had positive immunostaining. For the scoring of 8-oxo-dG immunostaining, “low” lung tumor sections were those in which <50% of tumor cell nuclei in a paraffin sections had positive immunostaining, and “high” lung tumor sections were those in which >50% of tumor cell nuclei in a he paraffin section had positive immunostaining.
DNA Extraction and Sequencing for EGFR Mutation
The DNA extraction procedures and DNA sequencing methods were as described previously. Genomic DNA was isolated from primary tumor samples by overnight digestion with SDS and proteinase K at 37°C, followed by standard phenol-chloroform extraction and ethanol precipitation. For exons 19 and 21 EGFR gene mutations, the following primers were used: exon 19: forward, 5′-CCAGATCACTGGGCAGCATGTGGCACC-3′ reverse, 5′-AGCAGGGTCTAGAGCAGAGCAGCTGCC-3′; exon 21: forward, 5′-TCAGAG CCTGGCATGAACATGACCCTG-3 ′ reverse, 5′-GGTCCCTGGTGT CAGGAAAATGC-TGG-3′. All polymerase chain reaction assays were carried out in a 25-μL volume that contained 100 ng genomic DNA and 1.25 units of HotStarTaq DNA polymerase (Qiagen Inc., Valencia, Calif). DNA was amplified for 33 cycles at 95°C for 30 seconds, at 65°C for 30 seconds, and at 72°C for 45 seconds, followed by a 7-minute extension at 72°C. All polymerase chain reaction products were incubated with exonuclease 1 and shrimp alkaline phosphatase and then sequenced directly using an automated sequencing system (3100 Avant Genetic Analyzer; Applied Biosystems, Hitachi, Japan).
Cell-Permeant 2′,7′-Dichlorodihydrofluorescein Diacetate Fluorescence Detection of Reactive Oxygen Species
Fluorescence detection of cell-permeant 2′,7′-dichlorodihydrofluorescein diacetate (H2-DCF-DA) for reactive oxygen species (ROS) was used according to a previously report method. The cells were incubated with phenol red-free medium containing 10 μM at 37°C for 30 minutes, because H2-DCF-DA can diffuse through the cell membrane, inside, the acetate groups were cleaved by cellular esterases, and the resulting H2-DCF could not leave the cells. Reaction with ROS, primarily hydrogen peroxide (H2O2), resulted in the fluorescent molecule DCF (maximum emission, approximately 530 nm).
Determination of 8-Oxo-dG Levels by Liquid Chromatography-Mass Spectrometry/Mass Spectrometry
Lung tumors and lung cancer cells were available for the analysis of 8-OH-dG by liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) with online solid-phase extraction (SPE), as described previously.
The chi-square test, the Fisher exact test (2-tailed), and multivariate analyses were used for statistical analyses. P values < .05 were regarded as statistically significant. All analyses were performed using the SPSS Version 15.0 statistical package (SPSS, Inc., Chicago, Ill).
Coexistence of EGFR Mutation With HPV16/18 E6-Positive and High 8-Oxo-dG Levels in Lung Cancer Patients
The potential association of EGFR mutations with HPV16/18 infection and oxidative DNA damage was explored in 151 surgically resected lung tumor specimens from patients with NSCLC to determine their EGFR mutation status, E6 oncoprotein expression, and 8-OH-dG levels by direct sequencing and immunohistochemistry, respectively. EGFR mutations, as expected, were more frequent in women, nonsmokers, and patients with adenocarcinoma compared with men, smokers, and patients with squamous cell carcinoma (sex: 45% vs 21%; P = .002; smoking status: 40% vs 17%; P = .002; histology: 42% vs 10%; P < .001) (Table 1). HPV16/18 E6 expression was more common in women and patients with adenocarcinoma than in men and patients with squamous cell carcinoma, respectively (sex: 60% vs 37%; P = .005; histology: 54% vs 31%; P = .006 for histology) (Table 1), as reported previously. Therefore, we hypothesized that HPV16/18 E6 expression may contribute to EGFR mutation in Taiwanese patients with lung cancer.
Table 1. The Prevalence of Epidermal Growth Factor Receptor Mutations, Human Papillomavirus Type 16/18 E6 Expression, and 8-Oxo-7,8-Dihydro-2′-Deoxyguanosine Levels in Lung Tumors From Patients With Lung Cancer According to Sex, Smoking Status, and Tumor Histologya
No. of Patients (%)
Abbreviations: 8-Oxo-dG, 8-oxo-7,8-dihydro-2′-deoxyguanosine; AD, adenocarcinoma; EGFR, epidermal growth factor receptor; HPV 16/18, human papillomavirus type 16/18; SQ, squamous cell carcinoma.
The chi-square test was used for statistic analysis.
HPV16/18 E6 Combined With 8-Oxo-dG Contributes More to the Occurrence of EGFR Mutations in Patients With Lung Cancer
To explore this hypothesis, the correlations among EGFR mutation, HPV16/18 E6 oncoprotein expression, and 8-oxo-dG level were examined. A higher frequency of EGFR mutation occurred in patients who had E6-positive tumors (41% vs 20%; P = .006) (Table 2) and in patients who had tumors with high 8-oxo-dG levels (36% vs 16%; P = .012), as indicated in Table 2. In addition, a positive correlation of HPV16/18 E6 expression with 8-oxo-dG levels in lung tumors also was observed in this study population (P = 0.034). We further examined whether the combination of HPV16/18 E6 expression and 8-oxo-dG level could contribute more to the occurrence of EGFR mutations in patients with lung cancer. In the study population overall, patients who had E6-positive tumors with high 8-oxo-dG expression had the greatest frequency of EGFR mutation, followed by those who had E6-positive tumors with low 8-oxo-dG expression, E6-negative tumors with high 8-oxo-dG expression, and E6-negative tumors with low 8-oxo-dG expression (44% vs 31% vs 28% vs 9%, respectively; P = .007) (Table 3). A similar trend also was observed in the subsets of nonsmokers (60% vs 38% vs 32% vs 9%, respectively; P = .017), women (70% vs 22% vs 36% vs 14%, respectively; P = .014), and patients with adenocarcinoma (60% vs 33% vs 39% vs 13%, respectively; P = .016) (Table 3), but not in the subsets of smokers, men, and patients with squamous cell carcinoma. These results suggest that the combined effects of HPV16/18 E6 and 8-oxo-dG may contribute more to the occurrence of EGFR mutations in patients with lung cancer, especially in nonsmokers, women, and patients with adenocarcinoma.
Table 2. The Correlation of Human Papillomavirus Type 16/18 (HPV16/18) E6 and 8-Oxo-7,8-Dihydro-2′-Deoxyguanosine (8-Oxo-dG) Levels With Epidermal Growth Factor Receptor Mutations and the Association Between HPV16/18 E6 and 8-Oxo-dG Levels in Lung Tumors in Patients With Lung Cancera
The chi-square test was used for statistic analysis.
Table 3. The Correlations of Epidermal Growth Factor Receptor Mutations With Human Papillomavirus Type 16/18 E6 and 8-Oxo-7,8-Dihydro-2′-Deoxyguanosine Levels in Patients With Lung Cancer According to Smoking Status and Tumor Histologya
HPV16-18 E6/8-Oxo-dG Expression
EGFR: No. of Patients (%)
Abbreviations: 8-Oxo-dG, 8-oxo-7,8-dihydro-2′-deoxyguanosine; EGFR, epidermal growth factor receptor gene; HPV 16/18, human papillomavirus type 16/18.
The chi-square test and the Fisher exact test were used for statistic analyses.
Squamous cell carcinoma
8-OH-dG Levels in Lung Cancer Cells Are Elevated by E6 Through Increased ROS Levels
We tested the possibility that 8-OH-dG levels in lung cancer cells could be elevated by HPV infection through increased ROS production. The potential for elevation of ROS and 8-OH-dG levels by E6 was tested using E6-positive TL-1 cells and E6-negative TL-4 lung adenocarcinoma cells to knock down E6 (TL-1 cells) and overexpress E6 (TL-4 cells). The ROS and 8-OH-dG levels were then evaluated by 2′,7′-dichlorodihydrofluorescein diacetate fluorescence detection and LC-MS/MS analysis, respectively. ROS levels decreased markedly in E6-knockdown TL-1 cells, but increased in E6-overexpressing TL-4 cells, as illustrated in Figure 1a. Consistently, 8-OH-dG levels also were reduced by E6 knockdown and elevated by E6 overexpression in these cells (Fig. 1b). However, a significant change in 8-OH-dG levels was observed with low-dose E6 knockdown or E6 overexpression in both cell types. These results suggest that E6 may elevate 8-OH-dG formation in lung cancer cells through increased ROS production.
8-OH-dG Levels Elevated by E6 May Occur Through Reduced Human MutL Homolog 1 and Human MutS Homolog 2 Expression
Next, we explored whether DNA mismatch repair (MMR) could be modulated by E6 and, in turn, would increase 8-OH-dG levels in lung cancer cells. Western blot analysis demonstrated that expression levels of human mutL homolog 1 (hMLH1) and human mutS homolog 2 (hMSH2) were significantly lower in HPV16 E6-positive TL-1 cells than in HPV16 E6-negative TL-4 lung adenocarcinoma cells (Fig. 2a). It is noteworthy that hMLH1 and hMSH2 expression levels were elevated by E6 knockdown in TL-1 cells; conversely, both expression levels were reduced by E6 overexpression in TL-4 cells in a dose-dependent manner (Fig. 2b,c). These results suggest that 8-OH-dG levels elevated by E6 may be caused in part by decreased hMLH1 and hMSH2 expression in lung cancer cells.
The possible enhancement of EGFR mutation by the HPV E6 oncoprotein was evaluated by conducting supF gene mutagenesis with the pSP189 shuttle vector in lung cancer cells. EGFR mutations were statistically significantly more frequent in patients of East Asian ethnicity versus other ethnicities, and p53 is important for antioxidation in response to oxidative DNA damage. Therefore, a Taiwanese lung cancer cell line (CL3 cells with the wild-type p53 gene) and an oxidative stress-sensitive cell line (H157, which has a p53 nonsense mutation and a missense mutation in the oxidative damage repair gene MYH1) were used for the mutagenesis assay. Both cell lines were transfected with the control plasmid and the HPV16 E6 combinational DNA plasmid to establish stable clones, and E6 expression in both clones was evaluated by Western blot analysis (Fig. 3a,b). Two stable clones of each cell type were selected for the supF gene mutagenesis experiment. E6 expression levels in these stable clones were evaluated by Western blot analysis, and significantly higher E6 expression levels were observed in CL3 stable clone 2 than in the other 3 stable clones of both cell types (Fig. 3). Each stable clone of both cell types was treated with or without hydrogen peroxide (0.1 μM and 1.0 μM). Table 4 indicates that the mutation rate in each stable clone of both cell types was markedly elevated by treatment with hydrogen peroxide in a dose-dependent manner. It is noteworthy that a significantly higher mutation rate was observed for CL3 stable clone 2 than for CL3 stable clone 1 or for H157 stable clones 1 and 2. Therefore, hydrogen peroxide-induced supF gene mutagenesis was enhanced significantly by E6 transfection in lung cancer cells. Consequently, we expected that the EGFR mutation rate may be higher in patients with E6-positive tumors than in those with E6-negative tumors.
Table 4. The Effect of Human Papillomavirus Type 16/18 E6 on H2O2-Induced supF Gene Mutation in the Shuttle Vector pSP189a
A link has been demonstrated between oxidative stress induced by HPV infection and increased DNA damage and mutation rates. For example, LKB1 mutation commonly occurs in cervical cancer and may promote tumor progression. HPV16 infection has been considered an etiologic factor in head and neck squamous cell carcinoma (HNSCC) and has been observed predominately in nonsmokers. Landscape genomic analysis also has indicated a high mutation frequency of Notch1 (approximately 22%) in patients with HNSCC. A recent study demonstrated that Notch1 mutation was more common in nonsmokers than in smokers with HNSCC. PIK3CA mutations (28%) in the kinase domain of uterine endometrial adenocarcinoma have been associated with adverse prognostic value. Therefore, gene mutations induced by oxidative DNA damage conceivably may play a role in HPV-associated carcinogenesis.
DNA MMR is the major pathway controlling genetic stability; it acts by removing mispairings caused by faulty replication and/or mismatches that contain oxidized bases. Therefore, inactivation of the MMR pathway, and especially of hMLH1 and hMSH2, is associated with the mutagenesis induced by oxidative DNA damage. The expression of hMSH2 was significantly lower in patients who had lung cancer with p53-mutated tumors than in those who had p53 wild-type tumors, presumably because of a p53 response element identified in the promoter region of hMSH2 that is required for its transcription.[28, 29] These results appear to support our observation in lung cancer cells that hMLH1 and hMSH2 expression levels were decreased significantly in E6-positive lung cancer cells (Fig. 2), and the reduction in gene expression of both by E6 may explain in part the increased 8-OH-dG levels in E6-positive lung cancer cells. In the current study, the immunostaining results for 8-OH-dG in lung tumors were confirmed by LC-MS/MS of extracts of a subset of lung tumors from the study population (n = 30). The levels of 8-OH-dG determined by LC-MS/MS were relatively consistent with the immunostaining results obtained for 8-OH-dG (P = .055), as shown in Figure 4. In addition, the levels of 8-OH-dG in these lung tumors were correlated significantly with the EGFR deletion mutation at exon 19 (P = .008) (Table 5), but not with the base-substituted mutation at exon 21. This observation was consistent with a previous report indicating that lung tumors with lower hMLH1 expression had a high frequency of p53 deletion mutations. Therefore, we suggest that 8-OH-dG levels may contribute to EGFR mutations, and especially to the exon 19 deletion mutation, through reduced hMLH1 and hMSH2 expression in HPV16/18-infected patients with lung cancer.
Table 5. Correlations Between 8-Oxo-7,8-Dihydro-2′-Deoxyguanosine Levels and Epidermal Growth Factor Receptor Mutation in Exons 19 and 21 in Non–small Cell Lung Tumorsa
EGFR Exon 19: No. (%)
EGFR Exon 21: No. (%)
Abbreviations: 8-Oxo-dG, 8-oxo-7,8-dihydro-2′-deoxyguanosine; EGFR, epidermal growth factor receptor gene; HPV 16/18, human papillomavirus type 16/18.
The chi-square test was used for statistic analysis.
In this study, p53 null H157 and p53 wild-type CL-3 lung cancer cells were used to explore whether HPV16 E6 oncoprotein could enhance H2O2-induced supF gene mutagenesis. Both vector-transfected cell lines with E6 expression had significantly higher H2O2-induced supF gene mutation rates than vector control cells, as indicated in Table 4. In addition, the E6-mediated, H2O2-induced supF gene mutation rate in p53 null H157 cells appeared to be lower than the rate in p53 wild-type CL-3 cells at high concentration of H2O2 (1 μM) (Table 4). Therefore, we suggest that the elevation in the H2O2-induced supF gene mutation rate by E6 oncoprotein was not associated with p53 aberration.
The association of EGFR mutations with HPV16/18 E6/8-oxo-dG levels was observed in nonsmokers, women, and patients with adenocarcinoma (Table 3). Multivariate logistic regression analysis was used to assess the parameter(s) that may contribute more to the occurrence of EGFR mutations in NSCLC patients. Among these parameters, adenocarcinoma was the most influential factor on the risk of EGFR mutations in patients with NSCLC (odds ratio [OR], 4.723, 95% confidence interval [CI], 1.559-14.306; P = .006) (Table 6), followed by 8-oxo-dG levels (OR, 3.060, 95% CI, 1.216-7.699; P = .018). No influence was noted for the other parameters on the risk of EGFR mutation in our study population. Patients who had E6-positive tumors had an OR of 1.901 for EGFR mutation risk compared with those who had E6-negative tumors, but the difference did not reach statistical significance (95% CI, 0.864-4.184; P = .111). This may have been because the influence of the parameters tumor histology and 8-oxo-dG on the risk of EGFR mutation was greater than the influence of E6. Therefore, the statistical significance of the E6 parameter may have been lost in this small study population (n = 151). Consistent with previous reports,[3, 4, 14, 31-33] EGFR mutation and E6 oncoprotein expression were concomitantly more frequent in nonsmokers, patients with adenocarcinoma, and women than in smokers, patients with squamous cell carcinoma, and men (Table 1). In addition, a positive correlation of EGFR mutation with 8-oxo-dG was more common in nonsmokers, patients with adenocarcinoma, and women than in their counterparts (Table 3). Therefore, at least in this study population, 8-oxo-dG levels may be elevated by HPV16/18 E6 oncoprotein and, in turn, may contribute to EGFR mutations in lung cancer, especially among nonsmoking women with lung adenocarcinoma.
Table 6. Multivariate Analysis of Odds Ratios for Epidermal Growth Factor Receptor Mutation in Non–small Cell Lung Tumors
In summary, the etiology of EGFR mutations in women, in never-smokers, and Asians with adenocarcinomas remains unclear. In this study, we provided evidence from cells in vitro and human tumors in vivo that a high prevalence of HPV16/18 infection in women, nonsmokers, and patients with adenocarcinoma from Taiwan may be associated with a higher frequency of EGFR mutations through increases in mutagenesis induced by oxidative DNA damage.
This work was supported by grants from the National Science Council (NSC99-2628-B-038-016-MY3 and NSC100-2314-B-38-043-MY3),Taipei Medical University (TMU101-AE1-B10), and Tung' Taichung MetroHarbor Hospital (TTM-TMU-102-02), Taiwan, Republic of China.