Published ahead of print on July 5, 2007, doi:10.1164/rccm.200704-543OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200704-543OC
IL13 Promoter Polymorphism –1112C/T Modulates the Adverse Effect of Tobacco Smoking on Lung Function1 Center for Human Genomics and Department of Medicine and Pediatrics, Wake Forest University School of Medicine, Winston-Salem, North Carolina; 2 Department of Epidemiology and Biostatistics, University of South Carolina, Columbia, South Carolina; and 3 School of Public Health, Saint Louis University, St. Louis, Missouri Correspondence and requests for reprints should be addressed to Alireza Sadeghnejad, M.D., Ph.D., Center for Human Genomics and Departments of Medicine and Pediatrics, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157. E-mail: asadeghn{at}wfubmc.edu
Rationale: Although the duration and amount of cigarette smoking correlate with reduction in pulmonary function, there is still variation among individual responses. IL-13 is involved in pulmonary inflammation, remodeling, and susceptibility to chronic obstructive pulmonary disease (COPD). Objectives: We investigated whether the relationships between smoking and the lung function measures FEV1 and FEV1/FVC ratio are modulated by IL13 polymorphisms.
Methods: Smokers ( Measurements and Main Results: The number of pack-years smoked was associated with a lower value for both %predicted FEV1 and FEV1/FVC (P < 0.001).The three SNPs were not associated with lung function measures; however, there was a significant combined effect between smoking and the promoter polymorphism –1112C/T on %predicted FEV1 (P for interaction < 0.03 for mean %predicted FEV1 and < 0.0001 for 90th percentile %predicted FEV1). Every 20–pack-year increment in smoking was associated with a 2.4% reduction in mean %predicted FEV1 in the common homozygous (CC) or heterozygous (CT) promoter genotypes, and an 8.2% reduction in mean %predicted FEV1 in minor allele homozygotes (TT, recessive model). Conclusions: An IL13 polymorphism in the promoter region may modulate the adverse effects of cigarette smoking on pulmonary function in long-term cigarette smokers.
Key Words: interleukin 13 • polymorphism • tobacco smoke • gene–environment interaction
Chronic obstructive pulmonary disease (COPD) is the fourth leading cause of death in the United States and its prevalence continues to increase. Although the primary known etiologic factor for the development of COPD is cigarette smoking, about 25% of chronic cigarette smokers develop COPD (1). This suggests that other factors, such as genetic predisposition, may influence the incidence or severity of this disease. The best-known genetic risk factor involved in the development of COPD is  1-antitrypsin deficiency (2). However, the frequency of this rare variation (1–2%) does not adequately explain the genetic predisposition to COPD in the general population. Therefore, it is reasonable to consider that other genetic factors may be involved in the pathogenesis of COPD. Data from both population and functional studies suggest that the IL13 gene may affect lung function. Several case-control studies have shown that polymorphisms in IL13 were associated with asthma and its related phenotypes (3–5). In a Dutch study, the –1112C/T polymorphism of IL13 has been reported to be associated with COPD compared with healthy control subjects (6). This association was not confirmed in another case-control study with Japanese and Egyptian subjects (7). In Chinese Han residents of Beijing, He and colleagues reported that the TT genotype of –1112C/T was not an independent risk factor for COPD but increased the risk for smokers of developing COPD (8). Data from functional studies also support that IL-13 and its gene may influence lung function. Pulmonary expression of IL-13 was reported to produce inflammation, mucus hypersecretion, subepithelial fibrosis, and eotaxin production (9). Inducible targeting of IL-13 to the adult lung in a murine model has resulted in emphysema with increased lung volume and compliance, mucus overproduction, and inflammation comparable to human COPD (10). Vladich and coworkers directly compared the activity of wild-type (WT) IL-13 and IL-13 R130Q (+2044G/A, rs20541) on primary human cells that were involved in the effector mechanisms of allergic inflammation (11). Their results showed that IL-13 R130Q was significantly more active than WT IL-13 in inducing STAT6 phosphorylation and CD23 expression in monocytes and hydrocortisone-dependent IgE switching in B cells. In an experiment using stimulated human T cells, the rare variant (T) at –1112C/T in IL13 resulted in increased levels of IL-13 through reduced inhibition of IL-13 production (12). In both human and murine CD4+ Th2 lymphocytes, the IL13–1112T allele has been reported to be associated with an enhanced promoter activity (13). On the basis of the results of previous reports, we investigated the association of three polymorphisms in IL13 and their interaction with cigarette smoking on lung function in whites with long-term tobacco use. Some of the results of this study have been previously reported in the form of an abstract (14).
Study Population Subjects were recruited from a cohort of tradesmen referred for a work-related, independent medical evaluation (15). Referrals were drawn from trade unions and using television and newspaper advertisements. Participants gave informed consent for their involvement, and the research protocol was reviewed and approved by the institutional review boards at Wake Forest University and Saint Louis University. As part of the referral process, an extensive questionnaire, a chest radiograph, and pulmonary function tests were obtained. The questionnaire detailed information about prior employment, smoking history, and personal and family health histories. The questionnaire was self-administered before evaluation, and the physician-examiner reviewed answers at the time of examination. Subjects were asked to quantify their cigarette smoking as packs per day, and ages of initiation and cessation of tobacco use. Chest radiographs were obtained and interpreted by a certified B-reader. Chest radiograph abnormalities were quantified according to the International Labor Organization (ILO) scoring system (16). Pleural abnormalities were recorded as present or absent. Pulmonary function testing was performed according to American Thoracic Society published guidelines (17, 18). The FEV1 used for analysis was the best of at least three replicate (within 5% of each other) FVC maneuvers. For the current study, subjects over 40 years of age with a greater than or equal to 20 pack-year history of cigarette smoking were included in the analysis. We did not genotype any subject who was not a smoker or who had smoked fewer than 20 pack-years. Because of the apparent overlapping susceptibility among smokers to both COPD and lung cancer, a past medical history of lung cancer was not an exclusion criterion. The presence of significant occupational exposure–induced lung disease (ILO scores > 1/1, 89 subjects), mesothelioma, and an anticipated survival of less than 1 year secondary to active cancer or other chronic diseases (226 subjects) were exclusion criteria.
Genotyping
Statistical Analysis All polymorphisms were tested for Hardy-Weinberg equilibrium using Haploview 3.2 software (http://www.broad.mit.edu/mpg/haploview) (20). Estimates of LD between SNPs were calculated using D' and r2 (21). We applied linear models, quantile regression, and logistic regression to evaluate the associations among pack-years smoked, genotypes, and their interaction on %predicted FEV1 and FEV1/FVC. We controlled for age (continuous covariate), sex, and status of smoking (current vs. ex-smoker) as potential confounders. Quantile regression is a robust statistical method that makes no assumptions about the distribution of the outcome variable (phenotype) in the population (22). It provides an inference about the entire distribution of the outcome variable. We considered the 10th, 25th, 50th, 75th, and 90th percentiles of %predicted FEV1. Standard errors and confidence intervals for the regression coefficients were obtained by generating 500 bootstrap samples. Pack-years entered the models as a continuous independent variable in the models for the considered quantiles, implying a linear relationship between all lung function indices and pack-years. Departures from linearity were tested and were not significant.
Genotyping for the three SNPs in IL13 was completed for more than 99% of subjects. The genotypes were in Hardy-Weinberg equilibrium and minor allele frequencies were similar to those of the Single Nucleotide Polymorphism database (dbSNP; http://www.ncbi.nlm.nih.gov/SNP/). LD measures between –1112C/T and +2044G/A were D' = 0.51 and r2 = 0.21. Between +2044G/A and +2525G/A, these measures were D' = 0.99 and r2 = 0.94. This population was white and mainly composed of men (97.0%, n = 1,073; Table 1). Approximately two-thirds of the population had quit smoking at the time of evaluation (Table 1). Each 10-year increment in the duration of smoking cessation was associated with an approximate 4% improvement in %predicted FEV1 (P < 0.001). The number of ex-smokers and the duration of smoking cessation were not different among genotypes. Pack-years smoked was associated with a reduction in %predicted FEV1 and in FEV1/FVC ratio (P values < 0.001; Table 2). None of the SNPs was associated with %predicted FEV1 and FEV1/FVC; however, the reduction in %predicted FEV1 due to smoking was significantly different in TT homozygotes versus CC genotype at polymorphism –1112C/T (P value for the interaction < 0.03; Figure 1). From linear regression, on average, every 20 pack-years of smoking was associated with a 2.4% reduction in %predicted FEV1 in homozygotes for the common allele (CC) and also for heterozygotes (CT) at –1112C/T, and an 8.2% reduction in %predicted FEV1 in homozygotes for the minor allele (TT). Table 3 shows the estimated mean %predicted FEV1 for genotypes of the three SNPs by the categories of pack-years smoked. In subjects homozygous for the minor allele (TT) at –1112C/T, those who smoked more than 80 pack-years had an estimated mean %predicted FEV1 that was 23.4% (= 82.0 – 58.6; Table 3) lower than the group who smoked 20–40 pack-years. In heterozygotes, this difference was 6.7% (= 76.4 – 69.7), and in homozygotes for CC at –1112C/T the difference was 5.2% (= 76.5 – 71.3; Table 3). Logistic regression analysis modeling the interaction between the genotypes of –1112C/T and categories of smoking on %predicted FEV1 < 80% resulted in comparable findings (P for interaction < 0.02).
Figure 2 shows results for quantile regression analysis. The slopes for the effect of pack-years smoked on different percentiles (10th, 25th, 50th, 75th, and 90th) of %predicted FEV1 across three genotypes of –1112C/T were quantified. When compared with the CC genotype, TT homozygotes had steeper slopes for 90th, 75th, and 50th percentiles of %predicted FEV1. The interaction terms between pack-years smoked and the TT genotype (vs. CC) at –1112C/T were significant for the 90th (P < 0.0001), 75th (P < 0.005), and 50th (P < 0.002) percentiles of %predicted FEV1.
The primary known etiologic factor for the development of COPD is cigarette smoking, yet about 25% of chronic cigarette smokers develop COPD (1). Therefore, genetic susceptibility may influence individual predisposition to the disease. COPD is considered a complex genetic disorder and it is expected that many genes play a role in the pathogenesis of this disease. IL13 polymorphisms genotyped in these 1,073 whites were the most commonly examined in previous studies (3–5). In concordance with the previous reports, our analysis suggests that the SNP +2525G/A provides redundant information because it is in strong LD with SNP +2044G/A. The genotypes studied in IL13 were not associated with reduced lung function measures; however, the interaction between the polymorphism in the IL13 promoter region (–1112C/T) and tobacco smoking was significant. The estimated effect of smoking on %predicted FEV1 was significantly greater in those with the rare variant (TT) of polymorphism –1112C/T (promoter region) in IL13 compared with the major allele homozygous genotype (CC). The estimated effect of the heterozygous genotype at –1112C/T (CT) was similar to that of the CC genotype (recessive genetic model). Quantile regression analysis revealed that, for a given level of pack-years smoked, the effect of the TT genotype had a larger influence on the higher percentiles of %predicted FEV1 distribution. This suggests that the combined effect of the TT genotype and pack-years smoked may be greater in individuals with higher FEV1 values compared with those with lower FEV1 values. The IL13 variant (–1112T) that we report enhances the adverse effect of cigarette smoking on lung function has been shown to be associated with asthma-related phenotypes and COPD in independent population studies (3, 5, 6). In a Dutch population, among several IL13 polymorphisms, the most significant associations were observed for –1112C/T on asthma and bronchial hyperresponsiveness (3). In a Dutch case-control study (151 cases), an increased frequency of the rare variant of the –1112C/T polymorphism was observed in patients with COPD compared with healthy control subjects (P = 0.002), and also, compared with smokers with normal lung function (P = 0.01). In this Dutch study, genotypes of the R130Q polymorphism (+2044G/A) had a similar distribution in cases and control subjects (6). In Chinese Han residents in Beijing, He and colleagues reported that the TT genotype of –1112C/T was not an independent factor for COPD (111 cases) but increased the risk of COPD in smokers (8). There is one report that failed to show an association between –1112C/T and COPD (7). This study was a case-control design with 88 Japanese and 106 Egyptian patients with COPD (7). The general problem with previous reports on the association between IL13 –1112C/T polymorphism and COPD has been small sample size (6–8). There is also evidence from functional experiments that IL-13 and its gene (specifically, –1112C/T in the gene) may have relevant biological ties to our findings. The –1112C/T SNP is located in a region containing a binding site of the nuclear factor of activated T cells (NFAT) transcription factor that regulates IL13 and IL4 gene expression. The T allele at position –1112 may increase binding of the NFAT protein to this region (12). There is some evidence that the T allele at the –1112C/T polymorphism results in increased levels of IL-13 (12, 13). In a report to assess the functionality of this polymorphism, T cells were stimulated to produce higher amounts of IL-13 (12). In this experiment, the overproduction of IL-13 could not be reduced in the presence of the TT genotype. Recently, Cameron and colleagues showed that the IL13 –1112T allele enhanced promoter activity in primary human and murine CD4+ Th2 lymphocytes (13). In a murine model, Zhu and colleagues reported that pulmonary overexpression of IL-13 was reported to produce inflammation, mucus hypersecretion, subepithelial fibrosis, and eotaxin production (9). In a follow-up, this group observed that, when IL-13 was inducibly overexpressed in the adult mice lungs, findings were consistent with the following: emphysema with enhanced lung volumes and compliance, mucus metaplasia, and inflammation. This study suggested that IL-13 overexpression may be associated with pulmonary histopathologic findings comparable to emphysema (10).
The current study is the first to report on the association between IL13 polymorphisms and quantitative pulmonary function measures. A strength of our study is the overall large sample size (n = 1,073). However, the TT genotype of interest has a frequency of about 6.0% in whites. Despite the low frequency of the TT genotype (55 subjects of 1,063 in this report), the regression coefficient for the TT genotype ( Functional studies are needed to assess the significance of the interaction between –1112C/T and environmental exposures or other genes. In summary, this report suggests that the IL13 –1112C/T polymorphism may modify the adverse effects of cigarette smoking in the pathogenesis of COPD.
Supported in part by the Selikoff Fund for Environmental and Occupational Cancer Research, Saint Louis University. Originally Published in Press as DOI: 10.1164/rccm.200704-543OC on July 5, 2007 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form April 5, 2007; accepted in final form July 5, 2007
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
20 pack-years), aged at least 40 years old (n = 1,073), were genotyped for three single nucleotide polymorphisms (SNPs; –1112C/T [rs1800925], +2044G/A [rs20541, R130Q], and +2525G/A [rs1295685]) in the IL13 gene. Linear, quantile, and logistic regression methods were used to assess the effect of cigarette smoking (pack-years), IL13 polymorphisms, and their interaction on %predicted FEV1 and FEV1/FVC ratio. Age, sex, and current smoking status were included as confounders. 
= 0.41) was statistically different (P < 0.03) from the slope for the CC genotype (