Polymorphisms in IL13 pathway genes in asthma and chronic obstructive pulmonary disease


  • Edited by: Marc Humbert

B. Beghé, Department of Oncology, Haematology and Respiratory Diseases, University of Modena and Reggio Emilia, Via del Pozzo 71, 41100 Modena, Italy.


To cite this article: Beghé B, Hall IP, Parker SG, Moffatt MF, Wardlaw A, Connolly MJ, Fabbri LM, Ruse C, Sayers I. Polymorphisms in IL13 pathway genes in asthma and chronic obstructive pulmonary disease. Allergy 2010; 65: 474–481.


Background:  Asthma and chronic obstructive pulmonary disease (COPD) are chronic respiratory diseases involving an interaction between genetic and environmental factors. Interleukin-13 (IL13) has been suggested to have a role in both asthma and COPD. We investigated whether single nucleotide polymorphisms (SNPs) in the IL13 pathway may contribute to the susceptibility and severity of asthma and COPD in adults.

Methods:  Twelve SNPs in IL13 pathway genes –IL4, IL13, IL4RA, IL13RA1, IL13RA2 and STAT6– were genotyped in subjects with asthma (n = 299) and in subjects with COPD or healthy smokers (n = 992). Genetic association was evaluated using genotype and allele models for asthma severity, atopy phenotypes and COPD susceptibility. Linear regression was used to determine the effects of polymorphism on baseline lung function (FEV1, FEV1/FVC).

Results:  In asthmatics, three IL13 SNPs – rs1881457(–1512), rs1800925(–1111) and rs20541(R130Q) – were associated with atopy risk. One SNP in IL4RA1 [rs1805010(I75V)] was associated with asthma severity, and several IL13 SNPs showed borderline significance. IL13 SNPs rs1881457(–1512) and rs1800925(–1111) were associated with better FEV1 and FEV1/FVC in asthmatics. IL13 SNPs rs2066960(intron 1), rs20541(R130Q) and rs1295685(exon 4) were associated with COPD risk and lower baseline lung function in the recessive model. In females, but not in males, rs2250747 of the IL13RA1 gene was associated with COPD and lower FEV1.

Conclusion:  These data suggest that IL13 SNPs (promoter and coding region) and, to a lesser extent, IL4RA SNPs may contribute to atopy and asthma. We also provide tentative evidence that IL13 SNPs in the coding region may be of significance in COPD susceptibility.

Asthma and chronic obstructive pulmonary disease (COPD) are chronic inflammatory respiratory diseases that are most likely due to an interaction between genetic and environmental factors. Atopy is the most important risk factor of asthma (1) and tobacco smoking is the most important risk factor of COPD (2).

Interleukin-13 (IL13) is a Th2 cytokine implicated in (i) recruitment of inflammatory cells from the blood to the lung, (ii) immunoglobulin (Ig) E production by B cells, (iii) regulation of matrix metalloproteinases and (iv) induction of mucin production and secretion (3, 4). IL13 is produced by a variety of cells, including Th2 CD4+, Th1 CD4+, CD8+ T cells, mast cells, basophils and eosinophils. IL13 mediates its effects by interacting with a complex receptor system comprising an IL4Rα/IL13Rα1 heterodimer and the IL13Rα2 receptor (3).

The involvement of IL13 in the pathogenesis of asthma has been extensively documented (3–5). The administration of recombinant IL13 in the lungs of mice increases airway mucus secretion and causes sub-epithelial fibrosis, bronchial hyperresponsiveness, and eosinophilic airway inflammation (4). Cells expressing IL13 mRNA are increased in the bronchial tissue of atopic and nonatopic asthmatic subjects (6). Several IL13 genetic polymorphisms have been linked with susceptibility to develop atopic disease, such as bronchial asthma (7–9). Also, genetic polymorphisms in genes encoding IL4Rα, IL13Rα1, or STAT6 (10–12) predispose toward development of atopy, supporting the role of the IL13 pathway in contributing to risk of developing atopic asthma.

The involvement of IL13 and its genetic variants in COPD is less clear. In experimental studies the overexpression of IL13 in the adult murine lung caused emphysema (13), and the number of IL13+ cells was elevated in the bronchial submucosa of adult symptomatic smokers with symptoms of chronic bronchitis compared to asymptomatic smokers (14). These data are still controversial: Saha et al. (15) reported that IL13 expression was not increased in sputum and bronchial mucosa of COPD patients compared to nonsmoking controls. Finally, IL13 mRNA levels in lung homogenates from subjects with emphysema were correlated with the lung function parameter forced expiratory volume in 1 s (FEV1) (16). Taken together, these data suggest that IL13 may be involved in the pathogenesis of asthma and COPD.

Considering the biological interaction between IL13 and the IL4 receptor, we sought to examine the major component of the IL13 pathway by genotyping 12 known single nucleotide polymorphisms (SNPs) in IL13, IL4, IL13RA1, IL13RA2, IL4RA1 and STAT6 in asthmatics, subjects with COPD and healthy smokers. Thus, the aim of our study was to investigate whether SNPs in the IL13 pathway may contribute to the susceptibility and severity of asthma and COPD in adults.

Subjects and methods


Three cohorts were included in this study. Adult asthmatics were recruited from Nottingham, UK (n = 176) and Padova, Italy (n = 123). The following inclusion criteria were used: physician-diagnosed asthma, Caucasian, age 16–60 years, <10 pack-year cigarette smoking history, documented asthma for >1 year, and no other respiratory diseases. Comprehensive phenotype data were collected, including details of current medication, co-existing atopic disorders such as hay fever and/or eczema, and family history of asthma and atopy. Each subject underwent a lung function test. Skin prick tests for common inhaled allergens (Dermatophagoides, grass, tree, cat, dog) were determined in the Italian population. Asthma severity was evaluated according to British Thoracic Society guidelines (1). Blood samples (10 ml) were obtained from all subjects, and genomic DNA was isolated using the Puregene DNA extraction kit (Gentra, QIAGEN, West Sussex, UK). Approval was obtained from the Nottingham and Padova local research ethics committees.

Subjects with COPD and healthy smokers (controls) were recruited from five UK centres. Smokers (n = 537) were recruited in Nottingham; criteria were Caucasian, age >40 years, and >10 pack-year smoking history. Criteria for subjects with spirometry-defined COPD (n = 455) were Caucasian, age >45 years, and >10 pack-year smoking history. The combined population (n = 992) recruited for smoking history or COPD diagnosis was stratified into (i) healthy smokers (n = 176) – postbronchodilator (postBD, salbutamol) FEV1 >80% and postBD FEV1/FVC >0.7 – and (ii) COPD subjects (n = 599) – postBD FEV1 <80% and postBD FEV1/FVC <0.7. Subjects missing data or those not meeting these criteria were excluded from the case–control analyses.

SNP selection, genotyping and haplotypes

SNPs were chosen for putative function, previous association with asthma phenotypes or ability to tag the haplotypes spanning the regions. SNPs were genotyped using a KASPar reagent kit (KBiosciences, Hertfordshire, UK). Hardy–Weinberg equilibrium was assessed and IL13 and IL4RA haplotypes were identified using haploview software (17).

Statistical analyses

For the asthma cohort contingency table, analyses were completed with genotype or allele counts in the different groups using GraphPad Prism 5.0. For genotype analyses, 2 × 3 tables were generated and differences in groups were determined by chi-square test. For allele analyses, 2 × 2 tables were generated and differences were evaluated by Fisher exact test and odds ratio. In the asthma cohort, dichotomous traits tested were atopy, determined by positive skin test; rhinitis questionnaire; and asthma severity, evaluated according to British Thoracic Society guidelines (1) [stage 1 (mild asthma) vs stage >3 (severe asthma)]. Using spss v15 (SPSS Inc., Chicago, IL, USA), logistic regression analyses were completed for COPD susceptibility using additive (e.g. AA vs AC vs CC), recessive (AA/AC vs CC) or dominant (AA vs AC/CC) models, including age, sex and smoking pack-years as covariates. By linear regression we determined the contribution of each SNP to baseline FEV1 (liters) or FEV1/FVC using additive, recessive or dominant models, including age, sex, height and smoking pack-years as covariates. We did not correct for multiple testing, as linkage disequilibrium (LD) suggested these were not independent tests. A P-value of <0.05 was considered significant for all analyses.


Clinical characteristics and allele frequencies

Characteristics of the study subjects are reported in Table 1. As anticipated, FEV1 and FEV1/FVC of the smoking control subjects (n = 176) and COPD patients (n = 599) were significantly different (P < 0.0001). Comparison of other baseline features identified significant differences in age, sex and smoking pack-years; therefore, in subsequent COPD susceptibility analyses we adjusted for these variables. Allele frequencies were comparable across the cohorts and showed good concordance with data in the SNP database (dbSNP) (Table 2). All SNPs were tested for deviation from the Hardy–Weinberg equilibrium in asthma (n = 299) and in the smoking/COPD cohort (n = 992), and there was no deviation (P > 0.05, data not shown).

Table 1.   Characteristics of the study subjects
 Asthmatics (n = 299)All smokers (n = 992)Healthy smokers (controls) (n = 176)Smokers with COPD (n = 599)
  1. Values are mean ± SD. FEV1, forced expiratory volume in 1 s

  2. FVC, forced vital capacity; BD, bronchodilator; ND, not determined.

  3. Values in parentheses are the number of individuals with data available. Control subjects were defined as having postBD (salbutamol) FEV1 >80% and postBD FEV1/FVC > 0.7. Subjects with COPD were defined as having postBD FEV1 <80% and postBD FEV1/FVC <0.7. Individuals not meeting these criteria were excluded from the case–control analyses. Variables between control and COPD groups were compared by independent t-test ([continuous] variables or Pearson chi-square test [sex]). P < 0.0001 for all variables.

Age (years)39.24 ± 12.3163.33 ± 10.28 (992)54.38 ± 9.5265.96 ± 9.01
Female (%)65.643.8 (984)56.340.4
FEV1 (% predicted)92.37 ± 20.5756.01 ± 28.17 (975)96.03 ± 12.1540.31 ± 15.63
FEV1/FVC 75.94 ± 11.5855.3 ± 17.4 (971)77.3 ± 5.946.3 ± 12.6
PostBD FEV1 (% predicted)ND59.08 ± 27.14 (885)99.48 ± 11.7244.65 ± 15.52
PostBD FEV1/FVC ND55.6 ± 17.7 (873)79.1 ± 5.546.2 ± 12.0
Smoking pack-years2.98 ± 6.3243.52 ± 26.05 (969)32.74 ± 20.0447.61 ± 27.01
Table 2.   Gene location and minor allele frequencies of genotyped IL13 pathway SNPs
SNPGeneLocationLocationAllelesSelected for:Asthmatics (n = 299)All smokers (n = 992)Smoking controls (n = 176)COPD patients (n = 599)
rs2066960IL135:132022334IL13 Intron 1C/AhtSNP0.
rs20541IL135:132023863IL13 Exon 4C/TR/Q 1300.
rs1295685IL135:132024344IL13 Exon 4C/ThtSNP0.
rs1805010IL4RA16:27263704IL4RA Exon 5A/GI/V 750.470.450.460.46
rs1805015IL4RA16:27281681IL4RA Exon 11T/CP/S 5030.
rs1801275IL4RA16:27281901IL4RA Exon 11A/GR/Q 5760.
rs2250747 (males)IL13RA1X:117747776IL13RA1 Intron 1A/GTagging SNP0.
rs2250747 (females)IL13RA1X:117747776IL13RA1 Intron 1A/GTagging SNP0.
rs5946040 (males)IL13RA2X:114154022IL13RA2 Intron 5G/TTagging SNP0.
rs5946040 (females)IL13RA2X:114154022IL13RA2 Intron 5G/TTagging SNP0.
rs324011STAT612: 55788449STAT6C/TIntron0.400.420.420.43

Haplotype structure

Haplotype analyses of SNPs genotyped in the 5q locus (IL4/IL13) and the 16p locus (IL4RA) are shown for the different populations in Fig. 1. There is some redundancy in the IL4/IL13 genotyping in asthmatics (Fig. 1A, B); i.e., rs1881457-rs1800925 are in high LD (0.97), as are rs20541–rs1295685. In the IL4RA1 locus, rs1805015 and rs1801275 are in high LD. Similar findings were observed in the smoking cohort (Fig. 1C, D).

Figure 1.

 Haplotype block structure of IL13 and IL4RA SNPs in asthmatics (Panels A and B, n = 299) and smokers (Panels C and D, n = 992). The location of each tested SNP is indicated. The intensity of shading represents D’ (a measure of LD), and numerical values were generated using haploview software (17).

IL13 pathway SNPs and atopy/asthma associations

In asthmatics, IL13 SNPs rs1881457(–1512) (odds ratio [OR] 2.1, confidence interval [CI] 1.09–4.4, P = 0.017), rs1800925(–1111) (OR 2.2, CI 1.13–4.40, P = 0.013) and rs20541(R130Q) (OR 2.3, CI 1.02–5.0, P = 0.029) were significantly associated with atopy (defined as a positive skin test), while rs2250747 (IL13RA1, tagging SNP) had a borderline association in females, but not in males (Table 3). While no single IL13 pathway SNP showed a significant association with hay fever, all the borderline associations were within the IL13 gene and included risk alleles rs1881457C(–1512), rs1800925T(–1111) and rs20541T(R130Q), with a strong concordance with the skin prick test analyses. The IL4RA SNP rs1805010(I75V) was associated with more severe asthma (OR 1.8, CI 1.1–3.3, P = 0.032). Multiple borderline associations (P ∼ 0.06) were found between IL13 SNPs and asthma severity, including risk alleles rs1881457C(–1512), rs1800925T(–1111), rs20541T(R130Q) and rs1295685T(exon 4) (data not shown).

Table 3.   Skin prick test and IL13 pathway SNPs in asthmatics
SNPNegative (n = 39)Positive (n = 70)GenotypeAlleleAllele95% CI
  1. 0, 1, and 2 represent the number of major, heterozygote and minor genotypes, respectively. For minor allele frequencies in controls vs COPD patients, see Table 2. MAF, minor allele frequency; OR, odds ratio; CI, confidence interval.

  2. *We did not correct for multiple testing as tested genes that have been previously reported to be associated with respiratory disease/phenotypes.

rs2250747 (males)100 0.00303 0.050.44210.44712.260.1119–45.64
rs2250747 (females)27300.05281100.140.05670.06783.1190.8292–11.73
rs5946040 (males)82 0.102112 0.180.28730.31332.000.4080–9.803
rs5946040 (females)141240.33161670.380.83730.32971.250.6179–2.529

IL13 pathway SNPs and COPD

Three SNPs in IL13– rs2066960A(intron 1), rs20541T(R130Q), and rs1295685T(exon 4) – were associated with COPD risk in smokers (OR ∼4.49–9.21), but only in the recessive model (Table 4). rs2250747G(intron 1) in IL13RA1 was associated with COPD risk in females, but not in males (OR 10.5, CI 1.21–91.4, P = 0.033) (Table 4).

Table 4.   Risk of COPD in smokers and IL13 pathway SNPs
SNPSmoking controls (n = 176)COPD patients (n = 599)AdditiveRecessiveDominant
012012P-valueOR95% CIP-valueOR95% CIP-valueOR95% CI
  1. OR, odds ratio; CI, confidence interval; ND, not determined (minor allele frequency too low); NA, not applicable (i.e., recessive model in males for X-linked gene).

  2. 0, 1, and 2 represent the number of major, heterozygote and minor genotypes, respectively. For minor allele frequencies in controls vs COPD patients, see Table 2. Logistic regression was used to compare genotype frequencies using the additive, recessive, and dominant models, with the covariates age, sex and pack-years included in the model.

rs2250747 (males)651202856200.8371.080.51–2.33NANANA0.8371.080.51–2.33
rs2250747 (females)6532116858110.6371.130.68–1.880.03310.51.21–91.40.6360.870.48–1.57
rs5946040 (males)631402825400.8190.910.43–1.97NANANA0.8190.910.43–1.97
rs5946040 (females)672811616440.7620.920.52–−1.620.6311.930.13–28.10.6560.870.47–1.60

IL13 pathway SNPs and lung function

In the asthmatic cohort, IL13 SNPs rs1881457(–1512) and rs1800925(–1111) were associated with higher FEV1 and FEV1/FVC using regression analyses (Fig. 2A–D). In the smoking cohort, in the recessive model, rs2066960(intron 1), rs20541(R130Q) and rs1295685(exon 4) were associated with lower FEV1, while rs2250747 was associated with lower FEV1 only in females (Fig. 2E–I).

Figure 2.

IL13 SNPs are associated with baseline FEV1 and FEV1/FVC in asthmatics (Panels A–D) and smokers (Panels E–I). Regression analysis was used to investigate the association between SNPs and baseline FEV1 and FEV1/FVC using the additive (A), recessive (R) and dominant (D) models. Data represent mean ± SE. Covariates included in the model were age, sex, height, and pack-years. *P ≤ 0.05.


This study confirms that SNPs in the IL13 pathway genes, particularly those spanning the IL13 promoter/coding region, may influence atopy development and, to a lesser extent, asthma severity. Our data also suggest a possible role for IL13 intron/exon polymorphisms in COPD development in smokers. Finally, our results show that genetic variants in the IL13 pathway may interact to affect baseline lung function in both asthmatics and smokers.

To further validate our findings, we completed data mining of the general 1958 birth cohort online data set (available at http://www.b58cgene.sgul.ac.uk/). Analyses identified an association between IL13 SNPs rs1881457(–1512), rs1800925(–1111) and rs20541(R130Q) and increased total IgE at age 44–45. These data show excellent concordance with identification of these SNPs as atopy risk alleles in the current study. Similarly, rs1800925(–1111) and rs2066960(intron 1) showed an association with increased FEV1 (at age 44–45) in this general population, confirming the importance of these SNPs in determining lung function as observed in our analyses in smokers. It is interesting to note that, in the general vs smoking populations, the direction of effect was reversed.

Overall our results are consistent with those from previous studies reporting the association of IL13 SNPs and atopy phenotypes. We replicated the association of the IL13-1111T variant reported by Howard et al. (8) with a positive skin test and the association of IL13-1512C and IL13-1111T variants with atopy, defined as increased total serum IgE levels reported in three different allergic cohorts of more than 3000 children (18).

Howard et al. (8) found that the IL13-1111T variant was associated with asthma per se; we did not formally test this, but we observed a borderline association between four IL13 SNPs, including the IL13-1111CT variant, and asthma severity. In addition, it would be difficult to identify the IL13-1111CT variant as a true risk polymorphism because of the strong LD observed between the four SNPs (–1512, –1111, R130Q and exon 4). In asthmatics, two of the SNPs associated with asthma severity were also associated with better lung function (higher FEV1 and FEV1/FVC). However, severity was established according to the pattern of medication required to keep asthma under control (1), and thus lung function in treated patients may not reflect severity.

DeMeo et al. (19), who analyzed 685 asthmatic children and their parents enrolled in the Childhood Asthma Management Program, reported a positive association between the IL13 R130Q polymorphism and various atopic phenotypes, including eosinophils, serum IgE and positive skin test, but not airway responsiveness, asthma diagnosis or asthma severity. This is not completely unexpected, as IL13 genetic variants have been associated with airway responsiveness but not with asthma in Chinese children (12). Furthermore, in a Costa Rican study, IL13 Arg130Gln was associated with asthma exacerbations only in children receiving treatment with inhaled steroids (20).

Finally, we found a borderline association between IL13 rs1800925 and the risk to develop allergic rhinitis, confirming the data reported by Huebner et al. (21) in a cohort of 923 children and by Black et al. (22) in the longitudinal 1946 birth cohort study of the UK Medical Research Council’s National Survey of Health and Development.

Taken together, our findings and data from the literature (19, 23) suggest a significant role for IL13 polymorphism in atopy in asthmatic subjects. However, we cannot exclude the possibility that the weak association of IL13 SNPs with asthma severity in our population may be due to the inability to exclude differential LD with unevaluated variants in IL13 or nearby genes, or the small number of subjects in this study.

While our findings are globally consistent with those of previous association studies of single SNPs in the IL13 gene itself with atopy and, to a lesser extent, asthma, they are not consistent with the results of studies supporting the role of individual IL4/IL4RA1 SNPs in susceptibility to both atopy and asthma. Indeed, while IL4-589CT has been widely reported to be a risk factor for both atopy and asthma phenotypes in different populations (24–26), we did not find any association between this common SNP and any of the phenotypes of atopy (positive skin test, serum IgE), hay fever, or asthma. Similarly, no associations were found between single SNPs in the gene encoding IL4RA1, such as the common Gln551Arg and Ser478Pro polymorphisms, and any atopy and asthma phenotypes frequently reported to be associated with asthma and allergic disorders (27–30); only a weak association between the IL4RA1I50V polymorphism and asthma severity was found in our asthmatic population. The lack of replication could be due to the different SNPs tested, to different ethnicity-specific effects of particular SNPs, or to the relatively small number of subjects analyzed in this and previous studies.

In the present study, we also demonstrated that three IL13 polymorphisms in smokers – rs2066960(intron 1), rs20541(R130Q) and rs1295685(exon 4) – are associated with COPD susceptibility. Furthermore, in females, but not in males, the rs2250747 SNP in IL13Rα1 is associated with COPD. These findings are novel and highlight a potential role for the IL13 intron/coding region SNPs in COPD susceptibility. Importantly, these same SNPs were determinants of baseline lung function in smokers.

We did not replicate the association between the IL13-1111CT polymorphism and COPD that was previously reported in the Dutch population (31). However, in that study the positive association could be due to the relatively small sample size (151 COPD patients and 78 healthy control smokers). On the other hand, in a larger population of 1073 smokers and former smokers, IL13-1111CT was not associated with lung function measures, and there was a significant combined effect of the promoter SNP IL13-1111CT on percent predicted FEV1, suggesting that the genetic variant may modulate the adverse effects of cigarette smoking on pulmonary function (32). In fact, IL13-1112CT has been reported to enhance IL13 activity in vitro (33).

The lack of replication of this association in our study may be due to a bias in the selection of healthy smokers. Indeed, the mean age of COPD patients was higher than that of healthy smokers, although we corrected for this in the regression model. It can be anticipated that some of the healthy smokers may develop COPD later in life. Other limitations of our study are that we did not have independent replication of the genetic effects observed, and several of the analyses are based on relatively small numbers.

Interestingly, we found a positive association between IL13 R130Q and COPD and impaired lung function that was not found in the Dutch study (31). Vladich et al. (34) reported that an R → Q replacement may amplify IL13 activity (e.g., STAT6 phosphorylation and CD23 expression in monocytes), particularly when associated with other SNPs in the IL13 pathway, such as - 1111CT (often found in LD) or V50R551 of IL4Rα (35). It is important to note that, while both the asthma/atopy and COPD analyses highlighted a role for IL13 SNPs in disease mechanisms, the asthma/atopy associations involved SNPs in the promoter, intron and exon regions, whereas the COPD signal was localized to intronic and coding region SNPs. These findings may in part explain the lack of correlation between COPD and IL13 expression (15) clearly observed in asthma and atopy (3).

In conclusion, the data reported here confirm that genetic variants in the IL13 pathway may be involved in the development of atopy and, to a lesser extent, severe asthma. Also, the novel positive associations found with COPD suggest a role for IL13 in COPD development in smokers as well.


This study was funded by The University of Nottingham (New Lecturer Fund to IS). We thank the study subjects for their contributions in making this study possible. We thank Drs L. Mancino and O. Coletti and the nurses P. and M. Bortolami for their contribution to the recruitment of asthmatic patients from Italy and Drs A. Henry and C. Stewart for DNA extraction.