© 2003 American Thoracic Society DOI: 10.1165/rcmb.4885 Polymorphisms in the IL13, IL13RA1, and IL4RA Genes and Rate of Decline in Lung Function in SmokersUBC McDonald Research Laboratories/iCAPTURE Center, St. Paul's Hospital, Vancouver, British Columbia, Canada; Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, Minnesota; and Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada Address correspondence to: Dr. A. J. Sandford, UBC McDonald Research Laboratories/iCAPTURE Center, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, V6Z 1Y6 Canada. E-mail: asandford{at}mrl.ubc.ca
Targeted expression of interleukin (IL)-13 in the adult murine lung has been shown to cause emphysema. We hypothesized that variants in the IL13, IL13RA1, and IL4RA genes would be associated with an accelerated rate of decline of lung function among smokers. We determined the allele frequencies of five polymorphisms in the IL13, IL13RA1, and IL4RA genes in 588 continuing smokers chosen from the NHLBI Lung Health Study for having the fastest (n = 282) and slowest (n = 306) 5-yr rate of decline of lung function (mean change in FEV1 %predicted/yr = -4.1 and +1.1, respectively). The IL4RA 551RR genotype was associated with rapid decline of lung function (odds ratio, 2.24; P = 0.043). However, none of the other four polymorphisms was associated with rate of decline in lung function. The association of 551RR with rapid decline of lung function became more significant in subjects who also had either the IL13 130RR or 1112TT genotypes. However, because multiple comparisons were made and only a few individuals had the 551RR genotype, these associations may represent type 1 error. Haplotypes consisting of alleles from the IL13 polymorphisms or from the IL4RA polymorphisms were not associated with rate of decline in lung function in smokers.
Abbreviations: airway hyperresponsiveness, AHR chronic obstructive pulmonary disease, COPD interleukin, IL single nucleotide polymorphism, SNP
Cigarette smoke exposure is the major cause of chronic obstructive pulmonary disease (COPD). However, only 1020% of active smokers develop clinical COPD (1), and cumulative smoking exposure was estimated to account for only 15% of the variation in forced expiratory volume in 1 s (FEV1) (2). Family (3) and twin studies (4) have shown that COPD is a complex genetic disease. Pulmonary inflammation induced by cigarette smoke and/or infection is a characteristic feature of COPD, but the precise mechanisms that lead to this disease remain largely unclear. In a recent study, Zheng and coworkers demonstrated that overexpression of interleukin (IL)-13 in the adult murine lung caused emphysema, elevated mucus production, and inflammation reminiscent of human COPD (5). IL13 is a Th2 cytokine that is strongly implicated in the pathogenesis of asthma and causes airway hyperresponsiveness (AHR) and eosinophilia (68). AHR, a characteristic feature of asthma, is a known predictor of accelerated rate of decline of lung function in smokers (9). Campbell has suggested that the site of lung injury and inflammation may be the major determinant of which pulmonary disease develops (10). Inflammation in the conducting airways may lead to asthma and inflammation in the lung parenchyma may result in emphysema. This is a conclusion similar to the "Dutch hypothesis," which proposes that there is a common susceptibility to COPD and asthma and that similar mechanisms can contribute to the pathogenesis of both disorders (11, 12).
The IL13 gene is located on chromosome 5q31, in a region in which genome searches have identified some linkage with atopy and atopic diseases in different populations (1319). Two single nucleotide polymorphisms (SNPs) of IL13 have been investigated in relation to asthma. One is located in the promoter region, at position 1112 relative to the transcription start site (2022). Another is a G
In this study, we hypothesized that an accelerated rate of decline in lung function in smokers from the Lung Health Study cohort would be influenced by the following polymorphisms: IL13 R130Q and T-1112C, IL13RA1 A1398G, IL4RA I50V, and Q551R. The Lung Health Study, sponsored by the National Heart, Lung, and Blood Institute was a clinical trial of smoking intervention and bronchodilator on the progression of COPD (32). This dataset provides an excellent opportunity to explore the associations of gene polymorphisms and rate of decline of lung function. In this cohort, we have previously found that the rate of decline of lung function was associated with the MZ genotype of the
Study Participants We selected 295 (282 of European descent and 13 of African descent) subjects who had the fastest decline of FEV1 and 320 (306 of European descent and 14 of African descent) subjects who had no decline of FEV1 from 3,216 subjects who continued to smoke during the initial 5 yr of follow-up in the Lung Health Study. The Lung Health Study is a multicenter randomized clinical trial. A total of 5,877 participants were smokers aged 3560 with spirometric evidence of moderate lung function impairment recruited from 10 North American medical centers with each center having 500 to 669 participants (37). The objective of the trial was to determine whether a program of smoking intervention and use of an inhaled bronchodilator could slow the rate of decline in pulmonary function over a 5-yr follow-up period (38). The lung function of the subjects was assessed as FEV1 %predicted, that is, FEV1 adjusted for age, height, sex, and race (32). To attain 300 participants for each group, arbitrary cutoff points of FEV1 %predicted/yr decrease 3.0% and increase 0.4% were used for rapid decliners and nondecliners, respectively. On average, 10.7% (range 8.312.3%) of the participants were selected from each center. In the rapid decliner group and nondecliner group, there were approximately the same percentage of subjects chosen from each individual center (6.915.2% and 5.615.1% for rapid decliners and nondecliners, respectively). The rapid decliners had a decrease in FEV1 of 153.7 ± 2.62 ml/yr (FEV1 %predicted/yr, 4.1 ± 0.06) and the nondecliners had an increase in FEV1 of 14.9 ± 1.51 ml/yr (FEV1 %predicted/yr, 1.1 ± 0.04).
Genotyping The A1398G polymorphism in the IL13RA1 3'UTR region was also detected by a PCR-RFLP method by using the following primers flanking the polymorphic region: 5'-GAA AGC CTC TCA GTG ATG GAG-3' (sense) and 5' GAG CTG CCT GTT TTT AAA TGG-3' (antisense). Amplification products were digested by MseI. MseI produced 17-bp, 29-bp, and 94-bp fragments when 1398A was present, and 17-bp and 123-bp fragments when 1398G was present. Detection of the I50V polymorphism in IL4RA was performed by a PCR-RFLP method described by Noguchi and colleagues with the modification that we used the following primers: 5' GAA GAG TCT GAT GCG GTT CC-3' (sense) and 5' GCC TCC GTT GTT CTC AGG TA-3' (antisense) (40). The Q551R polymorphism in IL4RA was genotyped as previously described (41). Template-free controls and known genotype controls were included in each experiment. The digested PCR products were electrophoresed on 3% agarose gels to separate the fragments. Genotypes were scored without knowledge of the phenotypes by two independent observers.
Statistical Analysis
The allele frequencies in the two ethnic groups are given in Table 1. Because the allele frequencies were different between ethnic groups, we limited our association study to those individuals of European descent, of whom we had the largest number of subjects. The characteristics of this population are shown in Table 2. Because genotype data on the IL13 C-1112T and IL13RA1 A1398G polymorphisms were not available in three participants and genotype data on the IL4RA I50V and Q551R polymorphisms were not available in two participants, the analyses include 588 participants for the IL13 R130Q polymorphism, 585 participants for the IL13 C-1112T and IL13RA1 A1398G polymorphisms, and 586 for IL4RA I50V and Q551R polymorphisms. The allele frequency of 130Q in the IL13 gene was 19.7%, which is similar to that in a German population (21%) (23) and a British population (26.7%), but less than in a Japanese population (43%) (31). The allele frequency of -1112T in the IL13 gene was 18.9%, which is similar to that in a German population (23%) (20). The allele frequency of 1398A in the IL13RA1 gene was 84.4%, which may be higher than that in a British population (31), although there are insufficient data for a direct comparison. The allele frequencies of 50I and 551R in the IL4RA gene were 53.8 and 21.5%, which are similar to those in another white population (50 and 20%) (44). The IL13 R130Q, IL13 C-1112T polymorphisms and IL4RA I50V, Q551R polymorphisms were in Hardy-Weinberg equilibrium in the European population (P = 0.4610.955). The IL13RA1 A1398G polymorphism was in Hardy-Weinberg equilibrium in the female study participants of European descent (P = 0.931).
Age, sex, smoking history, initial level of lung function (prebronchodilator FEV1 %predicted), and responsiveness to methacholine of the rapid and nondecliner groups are shown in Table 2. Each variable was significantly different between the two groups, so the frequencies of the genotypes between groups were analyzed by binary logistic regression to adjust for these potentially confounding factors. The genotype frequencies of the IL13 genes in the study groups are summarized in Table 3. Within the study groups of smokers who had a rapid decline and those who had no decline of lung function, the genotypes of IL13 R130Q and IL13 C-1112T occurred at a similar frequency. Because male subjects are hemizygous for the IL13RA1 gene, we stratified the subjects by sex. Homozygotes for the IL13RA1 1398A allele were more frequent in the groups of female smokers who had a rapid decline (77.6%) than those who with no decline (66.7%), but this difference was not significant (unadjusted P = 0.083, adjusted P = 0.159). No genotype frequency difference was found between male study groups (Table 4).
There were no significant differences in genotype prevalence for any polymorphism in the IL13 and IL13RA1 genes even after adjusting for the significantly different baseline values (age, sex, pack-years, baseline FEV1 %predicted, and response to methacholine). For the IL4RA I50V and Q551R polymorphisms, there were no significant differences when we compared the three separate genotypes of each polymorphism between the rapid decliners and nondecliners (Table 3). However, the frequency of the 551RR genotype in the rapid decliners was significantly higher than that in the nondecliners (7.1 versus 3.6%, adjusted OR = 2.24, and P = 0.043). Risma and coworkers recently showed that a functional alteration in the IL4RA requires the coexistence of two naturally occurring SNPs: 50V (termed V75 in the report by Risma and colleagues) and 551R, and the association of V50/551R with atopic asthma was greater than either allele alone (44). We also compared the frequency of individuals with at least one VR combination (VV/QR, IV/RR, or VV/RR) and those without a VR combination (II/QR, IV/QQ, II/QR, or VV/QQ) between rapid decliners and nondecliners, but there was no significant difference (P = 0.132) (data not shown). Possible interactions between the SNPs in the receptor subunits and ligand were analyzed. Table 5 shows that there was a significant difference in the frequency of IL4RA genotypes between rapid decliners and nondecliners in subjects with the IL13 130RR genotype. The frequency of the 551RR genotype was significantly higher in rapid decliners than in nondecliners (7.7 versus 2.0%, adjusted OR = 4.18, P = 0.016). A similar result was observed for the IL4RA Q551R and IL13 C-1112T polymorphisms (Table 5). No interactions were detected between IL4RA I50V genotype and the IL13 SNPs or between IL13RA1 and IL13 SNPs (data not shown).
There was strong disequilibrium between the 130Q allele of the IL13 R130Q polymorphism and the T allele of the IL13 C-1112T polymorphism (D/Dmax = 0.56, P < 0.00001), which is similar to that reported in a German population (15). Haplotype analysis of the IL13 R130Q and the IL13 C-1112T polymorphisms did not show association with the rate of decline of lung function (Table 6). There was no linkage disequilibrium between the I50V and Q551R SNPs of the IL4RA gene (D/Dmax = 0.10, P > 0.05). Haplotype analysis of those two SNPs did not show association with rate of decline of lung function (Table 6).
Because the regression analysis may not have adequately adjusted for the differences in baseline variablese.g., age, sex, smoking history, etc.we analyzed the data to determine whether there are any significant differences of the baseline values among different genotype groups. There were no significant differences among these groups (data not shown).
Our hypotheses were based on the observation that the targeted expression of IL-13 in the adult murine lung causes emphysema (5). Recent studies have identified that 130Q or -1112T homozygotes had significantly higher levels of total serum IL-13 than those homozygous for 130R or -1112C (20, 22, 31). Of the two types of human IL-13 receptor complexes identified, IL13RA2 is considered to be a "decoy" receptor, whereas the heterodimer consisting of IL13RA1 and IL4RA acts as the functional receptor for IL-13. The noncoding variant of IL13RA1, 1398A, was associated primarily with high IgE levels (OR = 3.39 in males, 1.10 in females) rather than asthma (31). IL4RA 50V, 478S, and 551R isoforms are associated with functional changes of the receptor, and have been associated with atopy, atopic asthma, and variation in IgE level (2527). In this study, we investigated five SNPs in the IL13, IL13RA1, and IL4RA genes as risk factors for smoking-related accelerated decline in lung function. We compared genotype frequencies among smokers with a rapid decline in lung function and smokers without a decline in lung function in study participants of European descent. The ethnic composition of the study subjects is important for studies of this type. The study subjects were drawn from ten centers across North America, and the rapid and nondecliner groups had similar percentage of subjects from each center. It is reasonable to assume that these represent a sample of mixed European descent. However, we have no further data concerning ethnicity. As we mentioned in the results, the frequencies of the five SNPs studied were similar to other white populations, with the possible exception of IL13RA A1398G. For the IL13RA polymorphism it is difficult to compare our results to those of Heinzmann and coworkers (31), because these authors only reported the percentage of individuals who were homozygous 1398AA. Because only female subjects have AA genotype, the proportion of females in the group will affect this genotype frequency. Heinzmann and coworkers did not report the sex ratio in their study group, and therefore no comparison can be made with our data (31). However, unless there was a high proportion of males in the study by Heinzmann and colleagues, our results show a higher proportion of the A allele. We did not find any association between individual SNPs in the IL13 and IL13RA1 genes and rapid decline in lung function. However, we found that the IL4RA 551RR genotype was associated with rapid decline of lung function (OR = 2.24, P = 0.043), which became more significant in subjects who also had the IL13 130RR genotype (OR = 4.18, P = 0.016) or 1112TT genotype (OR = 4.11, P = 0.015). This indicates a genegene interaction between IL13 and IL4RA; however, this interaction effect was observed in individuals with the non-risk genotype of IL13 and the risk genotype of IL4RA for asthma. This is not consistent with a recent report that showed an interaction of the risk genotype for asthma in IL13 C-1112T with the risk genotype in IL4RA S478P (which was in strong linkage disequilibrium with Q551R) (45). There are two possible explanations. First, there is a true interaction of IL13 and IL4RA on the rate of decline of lung function, and its mechanism is unknown. Second, due to the multiple comparisons that were made and the low numbers of individuals who had the 551RR genotype, these associations may represent type 1 error. Therefore, it is important that this association is replicated in an independent cohort before definite conclusions can be drawn. There were no associations of haplotypes consisting of IL13 R130Q and C-1112T SNPs, or haplotypes consisting of IL4RA I50V and Q551R SNPs, with decline of lung function. The lack of association of the polymorphisms of the IL13 and IL13RA1 genes in this study could be due to several reasons. First, the emphysema caused by targeted expression of IL-13 in the adult murine lung in transgenic model described in the report of Zheng and coworkers (5) may only be relevant to the development of COPD in a minority of patients. IL-13 has been strongly implicated in the pathogenesis of asthma, AHR, and tissue eosinophilia in animal models (6, 8). Therefore, IL-13 may only contribute to COPD in those individuals who have concurrent asthma and/or AHR. Second, the mouse models are not exactly analogous to the situation in humans. For example, unlike human B cells, no IL13RA1 is expressed on the surface of murine B cells. Third, IL13RA1 and IL4RA are mainly expressed in smooth muscle cells and epithelial cells of the human bronchus rather than the pulmonary parenchyma (31). Campbell and colleagues suggested that the site of lung injury and inflammation might be the major determinant of whether asthma or emphysema develops in a susceptible host (10). Inflammation in the conducting airways may lead to asthma, and inflammation in the lung parenchyma may result in emphysema. This may explain the apparent limited role of IL13, IL13RA1, and IL4RA polymorphisms in determining rate of decline of lung function. Fourth, we used rate of decline of lung function rather than emphysema as a phenotype in this study. The relationship between lung function and degree of emphysema is complex and measurements of FEV1 may not accurately reflect the degree of tissue destruction in the lung (46). In summary, this is the first study to address the association of most of the function-altering SNPs or SNPs in linkage disequilibrium with the function-altering SNPs within the IL13, IL13RA1, and IL4RA genes with the rate of decline of lung function in smokers. Our data suggest that an IL4RA Q551R interaction with IL13 R130Q or C-1112T may be associated with lung injury. However, this data should be taken as hypothesis generating rather than as identification of true risk factors for a rapid decline of lung function.
The authors would like to thank Melissa Skeans and Helen Voelker for help with the statistical analysis, and Peter Paré for critical analysis of the manuscript. This study was supported by a grant from the Canadian Institutes of Health Research. The Lung Health Study was supported by contract N01-HR-46002 from the Division of Lung Diseases of the National Heart, Lung, and Blood Institute. A.J.S. was supported by a Parker B. Francis Fellowship and is the recipient of a Canada Research Chair in genetics. Received in original form April 25, 2002 Received in final form October 14, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||