Nicotine and its derivatives, by binding to nicotinic acetylcholine receptors (nAChR) on bronchial epithelial cells, can regulate cellular proliferation and apoptosis via activating the Akt pathway. Delineation of nAChR subtypes in non–small-cell lung cancers (NSCLC) may provide information for prevention or therapeutic targeting. Expression of nAChR subunit genes in 66 resected primary NSCLCs, 7 histologically non-involved lung tissues, 13 NSCLC cell lines, and 6 human bronchial epithelial cell lines (HBEC) was analyzed with quantitative PCR and microarray analysis. Five nonmalignant HBECs were exposed to nicotine in vitro to study the variation of nAChR subunit gene expression with nicotine exposure and removal. NSCLCs from nonsmokers showed higher expression of nAChR α6 (P < 0.001) and β3 (P = 0.007) subunit genes than those from smokers, adjusted for gender. In addition, nAChR α4 (P < 0.001) and β4 (P = 0.029) subunit gene expression showed significant difference between NSCLCs and normal lung. Using Affymetrix GeneChip U133 Sets, 65 differentially expressed genes associated with NSCLC nonsmoking nAChR α6β3 phenotype were identified, which gave high sensitivity and specificity of prediction. nAChR α1, α5, and α7 showed significant reversible changes in expression levels in HBECs upon nicotine exposure. We conclude that between NSCLCs from smokers and nonsmokers, different nAChR subunit gene expression patterns were found, and a 65-gene expression signature was associated with nonsmoking nAChR α6β3 expression. Finally, nicotine exposure in HBECs resulted in reversible differences in nAChR subunit gene expression. These results further implicate nicotine in bronchial carcinogenesis and suggest targeting nAChRs for prevention and therapy in lung cancer. [Cancer Res 2007;67(10):4638–47]
- nicotinic acetylcholine receptor
- lung cancer
- quantitative polymerase chain reaction
Tobacco smoking is the major cause of lung cancer, and nicotine in tobacco smoke leads to both addiction and further metabolism into potent carcinogen(s). In addition, recent discoveries of functional acetylcholine receptors (AChR) on lung epithelial cells and lung tumors raise the question of whether exposure to nicotine could also participate in lung cancer pathogenesis by activating signal transduction pathways such as the Akt pathway ( 1). One model could be that nicotine by stimulating nicotinic AChRs (nAChR) would activate Akt in lung epithelial cells and perhaps stimulate cell proliferation and/or overcome apoptotic responses engendered by carcinogen exposure ( 1). If this model is true, then one may ask whether lung tumors have different nAChR expression patterns compared with normal lung tissues and whether lung cancers arising in smokers have different patterns compared with never smokers. Such differences would provide additional information that nicotine is playing a role via the nAChRs in lung cancer pathogenesis. In recent decades, there have been an increasing proportion of female nonsmokers compared with male smokers in patients with lung cancer ( 2). It is possible that both gender and smoking, or an interaction of both factors, are playing roles in lung carcinogenesis. Differences between men and women may also make them respond to tobacco smoke in different ways, and lung tumors derived from male smokers and female nonsmokers may have adopted different carcinogenic pathways. Thus, it would be important to analyze the role of nAChR expression in the context of gender as well.
AChRs are divided into nicotinic (nAChR) and muscarinic (mAChR) subtypes. nAChRs are further subdivided into neuronal or muscle subtypes, which could also be present in nonneuronal or non-muscle tissues. Neuronal nAChRs are composed of different subunits including α1, α2, α3, α4, α5, α6, α7, α9, α10, β1, β2, β3, or β4. In addition to these different α and β subunits, the muscle type nAChR may also contain δ, ε, and γ subunits. Genes encoding for individual nAChR subunit are named CHRNA1, CHRNA2, CHRNA3, CHRNA4, CHRNA5, CHRNA6, CHRNA7, CHRNA9, and CHRNA10 for the α subunits and CHRNB1, CHRNB2, CHRNB3, and CHRNB4 for the β subunits. nAChRs are found to be present throughout the central nervous system (CNS) and in nonneuronal tissues, such as α3, α5, and α7 in bronchial epithelium ( 3, 4); α4 in alveolar epithelial cells ( 4); and α3, α5, α7, β2, and β4 in pulmonary neuroendocrine cells and human small cell lung cancer (SCLC) cell lines ( 5– 8), skin keratinocytes ( 9), vascular tissues ( 10), and human lymphocytes ( 11). nAChR holoreceptor is a pentamer consisting of five homologous or different nAChR subunits surrounding a ligand-gated channel ( 12) that responds to binding by ligands such as acetylcholine, nicotine, or its highly carcinogenic derivative 4(methylnitrosamine)-1-(3-pyridyl)-1-butanone (NNK; ref. 13). Neuronal nAChR consists of only two types of subunits: either a combination of α and β subunits or five copies of the same α subunits ( 14). Both the α and β subunits are thought to contribute to the physiologic properties of nAChR, where α subunit contains the principal sites for agonist binding, such as acetylcholine, and β subunits are believed to regulate the rate of binding and dissociation by agonists ( 15). Functional receptors in the brain are composed mainly of α4β2, α4α5β2, α6β2β3, α6β3, α4α6β2β3, and α7 ( 16, 17). In fact, α4β2 in the brain is thought to be responsible for nicotine addiction ( 18, 19). α6 associated with β2 and β3, α3 or α4 is present in dopaminergic and adrenergic neurons in the brain ( 20); and α6β3, in particular, is a functional nAChR ( 21). Receptor affinity for nicotine varies with different composition of nAChR subunits ( 20). Transfection studies have shown that the ratio of α/β subunits in nAChRs depends on the ratio of expression of the encoding nAChR subunit genes ( 22). nAChRs were first implicated in the growth regulation of lung cancer when nicotine was found to stimulate DNA synthesis in human SCLC cell lines ( 23), and this was supported by subsequent identification that the receptor involved was nAChR α7 ( 7, 24). We have previously shown that lung cancer cells expressed nAChR, and that nicotine, at concentrations found in smokers, blocked the induction of apoptosis in lung cancer cells ( 25); whereas West et al. showed that activation of nAChR resulted in downstream activation of the Akt pathway ( 1), protein kinase C pathway, and the mitogen-activated protein kinase (MAPK) pathways, leading to inhibition of apoptosis and promotion of growth and proliferation in human bronchial epithelial cells (HBEC; refs. 1, 25).
In the brain, nAChR α7 showed paradoxical up-regulation in response to chronic exposure to nicotine, whereas other nAChR subunits were down-regulated. It has also been shown that the nAChR α7 in the lungs of monkeys is up-regulated by exposure to nicotine ( 26). Up-regulation of functional nAChR α7 subunits has been shown in normal human bronchial epithelial cells (NHBE) upon exposure to nicotine ( 10). In SCLC cell lines, up-regulation of nAChR α7 has been shown in response to NNK stimulation, with overexpression and phosphorylation of serine-threonine protein kinase Raf-1 and extracellular signal regulated kinases 1 and 2 and activation of c-myc ( 27), leading to increased proliferation ( 1, 28). There is also speculation that chronic tobacco smoking may induce a positive feedback loop that amplifies nicotine response in NHBE cells ( 10, 29). On the other hand, tolerance to the adverse effects of nicotine could reflect desensitization of nAChR. Thus, it would be important to know if other nAChR subunit genes respond to nicotine exposure in a similar way, and these nicotine stimulation responses may give insight into their potential roles in nicotine addiction or bronchial carcinogenesis.
Recent reports of clinical trials of the nAChR α4β2 antagonist, which target α4β2 receptors in the brain, have shown its clinical efficacy in smoking cessation ( 30). It is possible that similar nAChR antagonists could block the effect on lung epithelial cells or tumor cells.
All of these observations led to our current study of the quantitative mRNA expression analysis of various nAChR subunit genes in lung cancers, normal lung, and lung epithelial cells and their variation with the smoking history of the patients. In fact, we found significant differences in nAChR receptor subunit expression patterns in comparisons of tumor and normal tissue and also differences between lung adenocarcinomas, depending on smoking (nicotine) exposure.
Materials and Methods
Clinical characteristics of tumor and cell line samples. Total RNA was extracted from frozen tissue of 66 surgically resected non-SCLC (NSCLC) specimens and 14 normal lungs tissues. NSCLC tumor specimens were collected from Hong Kong Chinese patients undergoing surgical resection. Normal lung tissues used in this study were collected from patients with lung cancer undergoing surgical resection, and specimens were reviewed to show no tumor involvement. Written informed consent for tumor and normal lung tissues collection were obtained from patients recruited before surgery, and ethics approval for study protocol was obtained from the local Institutional Review Board of the University of Hong Kong (HKU)/Hong Kong Hospital Authority Hong Kong West Cluster. The demographic characteristics of these lung cancer patients, including their age, gender, smoking habits, and tumor-node-metastasis pathologic staging information, are summarized in Table 1 . Nonsmokers were patients who have never smoked for their lifetime. Smokers included patients who have been current active chronic smokers before surgery and patients who have been daily smoking for more than 6 months in the past but have quit smoking at the time of surgery. All 66 NSCLC tumors collected were included in quantitative PCR, and 49 (all primary lung adenocarcinomas) of these 66 specimens were used for microarray studies; in addition, of the 14 normal lung tissue specimens, 9 were used for microarray analysis, whereas 7 of them were included in quantitative PCR analysis (two normal lung tissue specimens were used in both microarray analysis and quantitative PCR analysis).
A panel of 13 NSCLC cell lines (9 NCI-H lung cancer cell lines and 4 HKULC lung cancer cell lines) were used in this study, and total RNA was extracted for quantitative PCR ( Table 1). The NCI-H lung cancer cell lines (all were NSCLC lines: H1437, H1648, H1770, H1819, H1993, H2009, H2087, H2122, and H2347) were maintained at the Hamon Center for Therapeutic Oncology Research, University of Texas Southwestern Medical Center at Dallas ( 31) and have been deposited in the American Type Culture Collection (ATCC) repository ( 32). The HKULC cell lines (HKULC1–4) were newly established lung adenocarcinomas from Hong Kong Chinese patients ( 33) being maintained and stored at the University of Hong Kong, HKSAR, China. The demographic characteristics of patients from whom the cell lines were established, including their gender, smoking habits, and tumor cell types, were known. The normal bronchial epithelial cell lines included in this study were one cell line derived from peripheral alveolar space, the small airway epithelial cell (Clonetics), and other bronchial epithelial cell lines, including NHBE (Clonetics), BEAS-2B (ATCC CRL-9609), and five NHBE cell lines of the HBEC-KT series (Dr. John Minna's Laboratory; ref. 34).
Complementary DNA synthesis. Total RNA was extracted from tissue specimens and cell lines. RNA samples (1 μg) were reverse transcribed in 20 μL reaction mix [5× First-Strand Buffer (Invitrogen), 100 mmol/L DTT (Promega), 1 mmol/L deoxynucleotide triphosphate (Amersham Biosciences), oligo-dT12-18 primers (Invitrogen) and random hexamer (Promega), RNaseOUT Recombinant RNase Inhibitor (Invitrogen), and Superscript II Reverse Transcriptase (Invitrogen)] with 1-h reaction at 42°C.
Quantitative PCR reactions. Twelve pairs of primers spanning across intron-exon junctions were designed for quantitative PCR targeting nAChR subunit genes α1, α3, α4, α5, α6, α7, α9, α10, β2, β3, and β4 ( Table 2 ), with 18S as the reference gene. Reaction conditions were validated separately for each pair of primers, with single peak of dissociation curves produced in each run of reaction.
Quantitative PCR reactions were carried out in triplicates. SYBR Green I (SYBR Green JumpStart Taq ReadyMix, Sigma) was used as the detection dye, and ribosomal 18S was used as the reference gene. Final reaction volume was 10 μL, with 1.25 μL of one-tenth TE-diluted cDNA from reverse transcription reaction, 0.1 μmol/L of each specific primer, and 5 μL SYBR Green Jumpstart Taq ReadyMix. Quantitative PCR cycles were set at 10-min denaturation followed by 40 cycles of 95°C for 15 s, 60°C for 5 s, and 72°C for 20 s and 72°C for 10 min as the final extension step. Dissociation curves were inspected for each pair of primers, and only one dissociation peak must be present for each run of reaction before the results were considered to be valid. A 5-fold serial dilution of a reference sample was used for construction of standard curves with respect to each pair of primers. Quantitative PCR was run for each tumor or cell line cDNA sample, and the Ct for a particular sample was obtained from the standard curves for a specific pair of primers. The Ct for unknown samples was compared with the Ct of reference samples to obtain the normalized relative amount for unknown samples. The normalized relative amount was then log 2 transformed, and approximation to normal distribution was estimated.
Excel RGB (red-green-blue color coding macro) coding was done with a black-blue-white scale representing highest to lowest level of nAChR gene expression detected in this study for expression pattern inspection. For quantitative PCR data, the normalized relative amount for all samples was multiplied by a common factor of 1,024 and then log 2 transformed to give values in the same range as the microarray data so that the same color scale could be used for both microarray and transformed quantitative PCR data for direct visual inspection and comparison.
Microarray GeneChip expression analysis. Total RNA (5 μg) was extracted with RNeasy Miniprep (Qiagen) protocol. The quality of the total RNA was checked with denaturing formamide gel electrophoresis, which showed two sharp and distinct bands of 18S and 28S. Quality check was also done by the Agilent Bioanalyzer with graphical analysis showing two distinct peaks of 18S and 28S without additional peaks of degradation. The total RNA was then hybridized onto Affymetrix GeneChip HG-U133 A and B sets according to standard protocols ( 35).
Absolute signals from an individual chip were captured and processed with the MicroArray Suite 5.0 software (Affymetrix). Captured signals were further scaled and normalized to median expression level with an in-house Visual Basic software “MATRIX” (Microarray Transformation in Excel) version 1.31 written by Luc Girard at the University of Texas Southwestern Medical Center at Dallas. The MATRIX program allowed input of multiple CHP files from MicroArray Suite 5.0 into an Excel spreadsheet where normalization, t tests, and color display were done and further statistical analysis could be done with raw, normalized, and transformed probeset signal intensities for each sample in Excel spreadsheet format.
The BRB ArrayTool 3.4.0 program (developed by Amy Peng Lam and Richard Simon at the Biometric Research Branch, National Cancer Institute) was used for significance analysis of microarray (SAM) to analyze for differentially expressed genes between different group phenotypes and for class prediction to identify signature genes that predict group phenotypes by computer algorithm such as support vector machine (SVM). All lung cancer samples were designated as either having a higher than the mean level of expression or a lower than the mean level of expression for individual subunit gene. Comparisons using SAM were made between samples above and below their respective mean levels of expression for individual nAChR subunit genes α1, α3, α4, α5, α6, α7, α9, α10, β2, β3, and β4 or different combinations of the expression levels of these nAChR subunit genes.
Response to nicotine exposure in NHBE cell lines. Five NHBE cell lines (HBEC-KT 1–5; ref. 34) were cultured, and an equal number of passages for each cell line were randomized into a nicotine group and control group. The HBECs were seeded and incubated for 24 h before addition of nicotine. At time 0 h, nicotine (100 nmol/L) was added to the nicotine group, and the same volume of culture medium without nicotine was added to the control group. Both groups were incubated for 72 h, after which medium from both groups was removed and replaced with the same volume of fresh medium without nicotine. Both groups were further incubated for 72 h until time 144 h was reached. Cells were harvested at time 0, 72, and 144 h. Total RNA was extracted for reverse transcription and quantitative PCR as described above.
Statistical analysis. Microarray analysis was done by the in-house program MATRIX 1.31, an Excel-based visual basic program written for analysis of microarray data. Student's t tests, with assumptions of two tails and unequal variance, were used for comparison of expression level of different subunit genes between groups of samples with different gender and smoking history. χ2 tests were used for comparing the distribution of smokers and nonsmokers with high or low mean expression of individual or combinations of nAChR subunit genes. SAM and class prediction with SVM were done with the BRB ArrayTools program version 3.4.0; t tests and χ2 tests were carried out with SPSS for Windows version 11.5.
Quantitative PCR analysis of nAChR subunit gene expression. Many of the nAChR subunit genes were expressed in the primary lung cancer specimens and normal lung tissue ( Fig. 1A ). Likewise, some but not all of these nAChR subunit genes were expressed in the NSCLC lines (with the exception of α4; whereas α3, α6, α9, and β2 were only expressed in a few of the tumor lines). HBECs only expressed α1, α5, α10, and β4 ( Fig. 1B). Significant difference was found in the levels of different CHRN gene expression when NSCLC tumors, normal lung tissues, NSCLC cell lines, and NHBE cell lines were compared ( Fig. 1C). When resected NSCLC were compared with normal lung tissues, there were statistically significant differences in the expression levels of CHRNA4 (mean NSCLC/mean normal lung = 3.10/6.32, P < 0.001) and B4 (mean NSCLC/mean normal lung = 7.56/5.96, P = 0.029).
A subgroup analysis with separate comparisons between the squamous cell carcinoma and adenocarcinoma against normal lung tissue was done. For the comparison between squamous cell carcinoma (n = 6) and normal lung tissues (n = 7), no statistically significant difference in the expression levels of any nAChR subunit gene was observed. When adenocarcinomas (n = 54) were compared with normal lung tissues (n = 7), significant differences were found in the expression of nAChR α4 (mean adenocarcinoma/mean normal lung = 1.92/6.32, P < 0.001) and β4 (mean adenocarcinoma/mean normal lung = 7.56/5.96, P = 0.029). However, CHRNA4 and CHRNB4 did not show statistically significant difference when NSCLC cell lines were compared with HBEC lines. Instead, the expression levels of CHRNA5 (mean NSCLC cell lines/mean HBECs = 6.83/5.00, P = 0.022), CHRNA7 (mean NSCLC cell lines/mean HBECs = 6.20/−0.05, P = 0.023), CHRNA9 (mean NSCLC cell lines/mean HBECs = 2.33/−1.46, P = 0.001), and CHRNB2 (mean NSCLC cell lines/mean HBECs = 1.84/−1.78, P = 0.037) were found to be significantly higher in NSCLC cell lines compared with NHBE cell lines.
Significant differences for the expression levels of the CHRNA6 (mean smokers/mean nonsmokers = 4.41/6.81, P < 0.001) and CHRNB3 (mean smokers/mean nonsmokers = 6.51/7.16, P = 0.007) subunit genes ( Fig. 2A and B ) were found when primary lung adenocarcinomas arising in smokers versus nonsmokers were compared. This difference remained after adjusting for gender (for men: CHRNA6, P = 0.003; CHRNB3, P = 0.031 and for women: CHRNA6, P = 0.001; CHRNB3, P = 0.027; Fig. 2C and D). Among the samples with low nAChR α6β3 expression level, 8 of 34 (23.5%) were smokers, and 3 of 24 (8.8%) were nonsmokers; and among the samples with high nAChR α6β3 expression levels, 3 of 24 (8.8%) were smokers, and 20 of 34 (58.8%) were nonsmokers. The proportion of smokers and nonsmokers among samples with low or high nAChR α6β3 gene expression was significantly different (χ2 distribution, P < 0.001).
Expression analysis with Affymetrix GeneChips. No significant differential gene expression from the Affymetrix arrays was found that identified tumor groups with high or low expression by quantitative PCR for individual nAChR subunit genes α1, α3, α4, α5, α6, α7, α9, α10, β2, β3, and β4 as well as for combinations of low α4 and high β4, or with low α4 and low β2. However, with combined nAChR α6β3 expression levels, 173 genes were found to be differentially expressed between 34 samples with nAChR α6β3 low (n = 11) compared with samples with nAChR α6β3 high (n = 23), when the false discovery rate was controlled at 0.05 (i.e., allowed 5% chance of false discovery among the list of significant differentially expressed genes identified). Class prediction with leave-one-out analysis and SVM was done using these 173 genes. Repetitive training and testing with division of samples into training set (random two third) and test set (random one third) resulted in 65 signature genes ( Table 3 ) predictive of the “nAChR α6β3 phenotype” with 100% sensitivity and 90% specificity. When these 65 genes were used for prediction of all the 49 samples, the prediction sensitivity and specificity were 87% and 61%, respectively, for smoking habits.
Exposure of NHBE cell lines to nicotine. We wished to know if brief exposure to nicotine would alter the expression level of any of the nAChR subunit genes in bronchial epithelial tissue. We took advantage of our new immortalized HBEC to study this. When HBEC lines were exposed to 100 nmol/L nicotine and then RNA was harvested at 72 and 144 h, a significant increase in the expression levels of CHRNA1, CHRNA5, and CHRNA7 were found at 72 h, with return to baseline levels of expression upon nicotine removal ( Fig. 3 ). No significant increase or decrease was found in the levels of expression of other nAChR subunit genes analyzed.
The distribution, function, and ligand-binding affinity of nAChR depends on the composition of nAChR subunits, although the exact function and physiologic roles of individual nAChR subtypes and its subunits are not completely understood. To our knowledge, the pattern of expression of these nAChR subunits has not been reported in NSCLC. The quantitative PCR analysis of nAChR subunit gene expression levels allowed for quantitative comparison between groups of samples with respect to gender and smoking habits. We have shown overall statistically significant differences in the expression levels of CHRNA4 and CHRNB4 between NSCLC tumor and normal lung tissues as well as CHRNA6 and CHRNB3 between NSCLC tumors from smokers and nonsmokers ( Fig. 2) and CHRNA1 and CHRNA7 with respect to gender. The expression levels of CHRNA6 and CHRNB3 were statistically different even after adjusting for the effect of gender.
CHRNA4 is the major nAChR subtype found throughout the CNS and was thought to be responsible for nicotine addiction ( 19). The expression level of CHRNA4 was found to be significantly lower in NSCLC tumor compared with normal lung tissues. A modest elevation of CHRNB4 subunit gene expression in NSCLC compared with normal lung tissues was detected in this study. Such differences in the levels of expression of CHRNA4 and CHRNB4 were not found when comparing NSCLC cell lines or NHBE cells. There have been previous reports of functional polymorphism of nAChR α4β4 leading to variable nicotine-induced transmembrane conductance in vitro ( 36). The relatively lower level of CHRNA4 expression in NSCLC tumor might imply that CHRNA4 expression was down-regulated in the process of lung carcinogenesis. One possible mechanism for this to happen could be desensitization of nAChR subunit gene expression on chronic exposure to stimulation by its agonists such as nicotine. The differences in the levels of expression of CHRNA4 and CHRNB4 subunit genes might imply that they were involved in the process of lung carcinogenesis, but the potential mechanism deserve further evaluation.
The NSCLC cell lines showed other nAChR subunit gene expression differences when compared with NHBE cell lines (i.e., CHRNA5, CHRNA7, CHRNA9, and CHRNB2). The nAChR α5 subunit has been reported to be present in the brain and usually co-assembled with α3, α4, β2, or β4 to form various nAChR subtypes such as α4α5β2 and α3α5β4 ( 16). The presence of α5 subunit in nAChR has been shown to alter the calcium permeability and nicotine sensitivity in vitro ( 37). The nAChR α7 subunit was found to be important in the control of nicotine-induced calcium influx in SCLC ( 38) and was thus thought to be important in growth signal transduction induced by nicotine binding to nAChR. nAChR α9 was present mainly in inner ear hair cells and has been found to form nAChR with α10 subunits ( 39), but its function has not been fully characterized. nAChR β2 formed the nAChR α4β2 subtype that was thought to mediate nicotine addiction in the brain ( 20, 22). The fact that CHRNA5, CHRNA7, CHRNA9, and CHRNB2 subunit genes showed elevated levels of expression in NSCLC cell lines could indicate their involvement in constitutive cellular processes in cancer cell lines.
CHRNA6 and CHRNB3 showed differences in expression level between NSCLCs from smokers (relatively lower expression levels) and nonsmokers (relatively higher expression levels), even after adjustment for the effect of gender. The pattern of CHRNA6 and CHRNB3 gene expression in our study of NSCLCs is similar to the observations that CHRNA6 and CHRNB3 are coexpressed in the CNS ( 16), and they form functional nAChR in the brain ( 21). In fact, CHRNA6 and CHRNB3 were located closely on chromosome 8 (8p11.21 and 8p11.2, respectively), and it was possible that they share a common regulatory mechanism. The fact that both CHRNA6 and CHRNB3 subunit genes showed lower expression in NSCLCs from smokers when compared with nonsmokers could imply desensitization with chronic exposure to tobacco smoke. In NSCLC cell lines, no significant difference in the expression of nAChR subunit genes was found between cell lines derived from smokers and cell lines derived from nonsmokers. This could be attributed to the small number of NSCLC cell lines (n = 13) being studied, or that the effects of tobacco smoking on nAChR gene expression had ceased with in vitro culture.
A list of 65 differentially expressed genes was found to be predictors of nAChR α6β3 phenotypes, whereas no significant differentially expressed genes were found between samples with different levels of expression of individual nAChR subunit genes or combinations, including α4β4, α4β2, α6β2, etc. Some of these genes are involved in the regulation of apoptosis (p8; ref. 40), potential tumor suppressor genes in lung cancer [CEBPA ( 41) and NME1 ( 42)], or in other cancers [NDRG2 ( 43) and MXI1 ( 44)] and genes involved in tumor progression and metastasis [DPP4 ( 45), PLUNC ( 46), and SFRP4 ( 47)]. Both EGFR and MYC were found to be on the list of differentially expressed genes with respect to the α6β3 phenotypes. Binding of nAChR by nicotine was thought to activate downstream signaling pathway via Akt phosphorylation ( 1). It has been reported that nAChR may trigger the MAPK pathway with which EGFR and MYC were involved (reference: KEGG pathway hsa04010: classic MAPK pathway), eventually leading to promotion of cell growth and proliferation. EGFR expression was higher in the high α6β3 tumors. Inhibitors of EGFR prevented nicotine-induced Akt phosphorylation in mouse pheochromocytoma cell line (PC12; ref. 48). Thus, cross-talk between signaling downstream of EGFR and nAChR activation via the AKT and MAPK pathways may together promote carcinogenesis in this group of tumors ( 48). On the other hand, the low α6β3 tumors showed high MYC and low CEBPA expression. MYC overexpression had been found to lead to reduction in CEPBA-mediated cytochrome activation in the presence of EGF stimulation in proliferating cells ( 49). Together, the data suggest involvement of distinctive pathways mediated by EGFR or MYC in tumors that express different combination patterns of A6 and B3 nAChR subunits. Further investigation into their relation with nicotine exposure and verification with functional studies would be warranted.
There were gender differences detected in the expression level of CHRNA7 in lung cancer cell lines. CHRNA7 had been most well studied in patients with schizophrenia ( 50). There were suggestions of CHRNA7 correlating with gender and nicotine dependence in patients with schizophrenia ( 51). CHRNA7 has been reported to be expressed in SCLC cell lines ( 5, 6). There was previous reports of sensitization of the nAChR α7 subunit in human SCLC cell lines by elevated carbon dioxide level at the expense of oxygen level ( 52). It was thus proposed that chronic pulmonary condition such as chronic obstructive pulmonary disease (COPD) may promote the growth of SCLC. This was further supported by demonstration of up-regulation of the nAChR α7 subunit of endothelial cells by second-hand smoke and hypoxia ( 53). In addition, the growth of pulmonary neuroendocrine cells was shown to be under the control of the nAChR α7 subunit ( 27), with formation of multiple hyperplastic foci indicative of increased proliferative activity in the lungs of smokers with COPD and other chronic pulmonary diseases ( 54). It was possible that, in addition to smoking status or nicotine exposure, the presence of nonneoplastic pulmonary disease that impairs lung oxygenation may influence the expression of CHRNA7.
CHRNA1, CHRNA5, and CHRNA7 showed significant reversible induction of expression on in vitro exposure to nicotine, providing direct evidence that these subunit genes respond to acute nicotine exposure and could mediate the immediate or short term effects of nicotine. Little is known about the functional significance of the expression these specific CHRNA1, CHRNA5, and CHRNA7 in the CNS or in lung cancer, although CHRNA7 had been reported to be expressed in SCLC cell lines ( 5) and was thought to be related to smoking in schizophrenic patients ( 50), as discussed above. Up-regulation of functional CHRNA7 has also been shown in NHBE cells on exposure to nicotine ( 10). The return of the expression levels of those three subunits to baseline upon cessation of nicotine exposure may reflect that continued or chronic exposure to nicotine (usually taken to be more than 10 days of continuous exposure to nicotine; ref. 55) was required for nicotine addiction or other cellular effects of nicotine mediated by specific nAChR subunits. The effects of chronic exposure to nicotine on nAChR subunit gene expression in these NHBE cells and whether the same response is maintained, or other nAChR subunit genes would be involved, in chronic nicotine exposure warrant further evaluation. As the baseline expression level of CHRNA6 and CHRNB3 are low, we cannot determine whether they are down-regulated on nicotine stimulation.
In summary, we have determined the pattern of expression of CHRN subunit genes in NSCLCs, normal lung, normal bronchial epithelial cells, and tumor cell lines. The combination of CHRNA6 and CHRNB3 high expression correlated with NSCLCs in nonsmokers, whereas the combination of low expression correlated with NSCLCs from smokers. This leads to the identification of differential gene expression with class predictive significance. With short-term exposure to nicotine, there were elevated level of expression for CHRNA1, CHRNA5, and CHRNA7 subunit genes. These nAChR subunit genes could be playing a role in the pathogenesis of bronchogenic carcinoma and may mediate the effects of nicotine addiction in lung cancer patients of different sexes. Further evaluation of the functions and roles played by these nAChR subunit genes in nicotine addiction and lung carcinogenesis are warranted to allow for opportunities in development of chemoprevention strategy.
Grant support: Lung Cancer Specialized Programs of Research Excellence grant P50CA70907 and the Hong Kong Special Administrative Region RGC grant 7468/04M.
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.
- Received December 21, 2006.
- Revision received February 21, 2007.
- Accepted March 8, 2007.
- ©2007 American Association for Cancer Research.