Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

A polymorphism in the thyroid hormone receptor gene is associated with bronchodilator response in asthmatics


A pro-asthmatic culture milieu and β2-agonist (isoproterenol) were previously shown to regulate the expression of select transcription factors (TFs) within human airway epithelial and smooth muscle cells. This study tests 1116 single-nucleotide polymorphisms (SNPs) across 98 of these TF genes for association with bronchodilator response (BDR) in asthma patients. Genotyping was conducted using the Illumina HumanHap550v3 Beadchip in 403 non-Hispanic White asthmatic children and their parents. SNPs were evaluated for association with BDR using family and population-based analyses. Forty-two SNPs providing P-values <0.1 in both analyses were then genotyped in three adult asthma trials. One SNP 5′ of the thyroid hormone receptor-β gene was associated with BDR in the childhood population and two adult populations (P-value=0.0012). This investigation identified a novel locus for inter-individual variability in BDR and represents a translation of a cellular drug–response study to potential personalization of clinical asthma management.


Asthma is a chronic disorder characterized by inflammation, hyper-responsiveness of the bronchial muscles and narrowing of the airways that affects 300 million individuals worldwide.1 The increasing prevalence of asthma in recent decades has resulted in high rates of morbidity, mortality and annual health-care costs estimated to be tens of billions of dollars within the United States.2, 3, 4 Despite the availability of several classes of asthma therapies, large inter-individual variability in drug response has been described, which may be attributed in part to genetic factors.5, 6 Pharmacogenetic studies of β2-agonists, the most common asthma therapy, have identified multiple genes associated with bronchodilator response (BDR).5 The loci described to date, however, explain only a fraction of the variability in drug response, suggesting that other factors modulate BDR.

We previously described the differential expression of transcription factors (TFs) in two types of human airway (epithelial and smooth muscle) cell lines that are regulated by a pro-asthmatic culture milieu and β2-agonist.6 Specifically, the expression of 307 TFs was quantified following incubation with pro-inflammatory cytokines (interleukins 4 and 13, transforming growth factor-β), mediator leukotriene D4 and β2-agonist isoproterenol. Under these pro-asthmatic conditions, isoproterenol evoked changes (50% difference) in TF gene expression. Given that the role of these two airway cell types in asthma pathophysiology (that is, inflammation, remodeling and bronchoconstriction), we hypothesized that genes regulated by in vitro exposure to isoproterenol and a pro-asthmatic culture milieu would be good candidates for modulating drug response to β2-agonists in asthmatics. The aim of this study is to test the association of single-nucleotide polymorphisms (SNPs) in these TF genes with BDR in asthma trial populations treated with a short-acting β2-agonist.

Materials and methods

Study populations

The Childhood Asthma Management Program (CAMP) was a clinical trial of 1041 asthmatic children over an average period of 4.3 years.7, 8 A total of 403 non-Hispanic white probands and their parents were successfully genotyped on the Illumina HumanHap550v3 BeadChip (Illumina, San Diego, CA, USA). Each of the three replication trials consisted primarily of white adults with mild-to-severe asthma but no other significant comorbid medical conditions: Sepracor asthma trial (n=435);9 Leukotriene Modifier or Corticosteroid or Corticosteroid Salmeterol (LOCCS) trial (n=159);10 Effectiveness of Low Dose (LODO) Theophylline as Add-on Treatment in Asthma trial (n=155).9 Sepracor participants were selected to have BDR 15%. LOCCS patients were treated with a low-dose inhaled corticosteroid during a 4- to 6-week run-in period prior to randomization, which improved lung function in the range of 85–92% predicted.11 In all four trial populations, BDR was measured as the percentage difference in forced expiratory volume in 1 s (FEV1) after administration of two inhalations of albuterol (180 μg total) via a metered dose inhaler (BDR=100 × (postFEV1–preFEV1)/preFEV1). All participants or their guardians provided written informed consent, and all protocols were approved by the Institutional Review Board.

Gene selection and genotyping

We selected 98 candidate genes, which code for isoforms of 59 TFs, that were previously shown to be differentially expressed in lung cells (50% up- or downregulation) in response to isoproterenol and pro-asthmatic conditions.6 A total of 1116 SNPs across these candidate genes and 20 kb on either side were successfully genotyped in CAMP using the Illumina HumanHap550v3 BeadChip (Illumina). Data cleaning and quality control of this genotype data has been previously reported.12 Follow-up genotyping in the three replication populations used a Sequenom MassARRAY MALDI-TOF mass spectrometer (Sequenom, San Diego, CA, USA). Each SNP had a greater than 95% completion rate and a Hardy–Weinberg equilibrium P-value of >0.01.

Statistical analyses

The primary outcome measure of the association analyses was acute BDR to the inhaled β2-agonist albuterol, dichotomized by the median value in each population (as shown in Table 1) due to variability in BDR distribution across the four trial populations. CAMP was used for discovery analysis using the available genome-wide SNP data, and the top SNPs were then genotyped in the three adult asthma trial populations for replication analysis of BDR. Given that the sample size of the CAMP trial limits the power to detect a genetic association, we performed both a population-based association test as well as a family-based association test and selected the top loci identified by both analyses to carry forward for replication in additional asthma populations. Specifically, 42 SNPs that provided P-values <0.1 in both the family-based analysis (PBAT) of the trios13 and population-based analysis of the probands using PLINK v1.5 ( were carried forward for genotyping and replication analysis. Our rationale for using both tests was to confirm that the loci identified in the population-based test were not the result of population stratification.

Table 1 Baseline characteristics of four asthma trial populations

The Haplo.stats package in R was used to estimate the haplotype structure, which applies the expectation-maximization algorithm.14 Haplotype estimates with posterior probabilities of 96% were used to calculate its prevalence in each population. The haplotype effect was specified as an additive model and adjusted for nongenetic covariates including sex, height, pre-bonchodilator FEV1 and age. Haplotype associations were considered significant only if a global haplogroup test and a subsequent specific haplogroup test each provided P-values <0.05.

Replication analysis in the three adult asthma trial populations consisted of population-based tests using PLINK. In all analyses, the additive model was used and adjusted for nongenetic covariates including sex, age, height and prebronchodilator FEV1. Combined P-values were calculated from the one-sided P-values of the replication populations using Fisher's method.12 The allelic and summary odds ratios of the mutant allele were estimated using the DerSimonian-Laird random-effects meta-analysis approach as implemented in the rmeta package in R.15 The variation in drug–response phenotype attributed by the rs892940 genotype is estimated using a logistic regression model in the Design package within R. Linkage disequilibrium (LD) among SNPs was determined by correlation coefficient values (r2) as calculated using PLINK.


Baseline characteristics of the four asthma trial populations are detailed in Table 1. CAMP consisted of children ranging from ages 5–13 years, whereas the three replication populations were composed primarily of adult asthma cases. Other distinctions among the four clinical trials include the gender composition with LOCCS and LODO recruiting fewer males than CAMP and Sepracor, as well as differences in the mean and distribution of BDR across the four trials. For example, Sepracor participants were selected to have BDR 15%, reflected in a higher mean BDR of 40.3% (s.d.=21.6). In addition, LOCCS participants were previously treated with an inhaled corticosteroid, which may have improved their lung function and explain in part their lower mean and more normally distributed BDR (that is, skewness=0.038 and kurtosis=0.444). Given the phenotypic variability across our asthma trial populations, we did not apply conventional clinical thresholds of BDR for classifying patients as ‘responders’ (12% or greater BDR).16, 17 Instead, the median BDR value of each trial was used to distinguish responders within that trial.

A total of 1116 SNPs across the 98 candidate genes were tested for association with BDR in the CAMP trial using both family-based and population-based methods. SNPs providing P-values <0.1 using both analytical methods were considered to be the most robustly associated polymorphisms. Table 2 lists 42 such markers that were not correlated (linkage disequilibrium, LD) with each other, indicated by correlation coefficients (r2<0.8). These were subsequently genotyped in the three adult asthma trials (listed in Table 2). SNP association analyses in these follow-up populations identified five SNPs that provided P-values <0.1 in one or more of the replication populations (Table 3). SNP rs892940 in the thyroid hormone receptor B (THRB) locus is associated with BDR in CAMP, and replicated in LODO and Sepracor with a Fisher's combined P-value of 0.0012, which meets Bonferroni's significance threshold (0.0012). This is a common SNP (minor allele frequency of 41.7% in CAMP) that is located 2.5 kb 5′ of the THRB gene. Figure 1 shows that individuals with the minor allele are 33% more likely to respond to β2-agonists compared with those with the major allele, with a summary odds ratio of 1.33 (95% confidence interval 1.11–1.58; Table 4). However, the percentage of phenotypic variation attributed to the rs892940 is small with estimates of 0.75%, 0.30%, 1.04% and 0.25% in CAMP, LOCCS, LODO and Sepracor, respectively. Using SNP genotype data from the hapmap CEU population (, this SNP was determined to be in LD (r2>0.8) with another SNP (rs4858119), located 2.8 kb 5′ of THRB. In addition, genotype data from the 1000 Genomes Project confirms the LD between these 2 SNPs and identifies 14 other SNPs in the LD block, located within 50 kb of the THRB gene. However, none of these SNPs are coding. It remains to be determined whether any of these SNPs regulate the expression of THRB.

Table 2 Differentially expressed TF genes associated with BDR in CAMP
Table 3 SNPs associated with BDR in the four asthma trials
Figure 1

Odds ratios (ORs) indicate greater likelihood of a high β2-agonist bronchodilator response for the mutant allele of rs892940 in CAMP, LODO and Sepracor (SEP) trials. Boxes represent the point estimate for each study, the width of which is proportional to the standard error. The summary OR is represented as a diamond, the width of which is proportional to the standard error. Horizontal lines represent 95% confidence intervals.

PowerPoint slide

Table 4 Odds ratios (95% CI) of the minor allele relative to major allele of rs892940 for each asthma trial population, summary OR and test for heterogeneity

As multiple markers providing modest associations were identified across genes (Table 2), haplotype analysis was conducted to determine whether the haplotypic effects were stronger than single marker associations. Although significant haplotypic effects (P-values <0.05) were found for several candidate genes (RUNX1, TCF12, PARP1 and AP3) in CAMP, none of these haplotype associations replicated in the additional asthma populations.

Two SNPs in the vitamin D receptor, with the lowest P-values in CAMP, were nominally associated with BDR in LODO but did not meet significance criteria when the P-values were combined using Fisher's method. Similarly, rs3858444 in the Wilms tumor 1 isoform B was moderately associated with BDR in CAMP and replicated in LODO only, yielding a high combined P-value. Finally, rs2249650 in the runt-related TF 1 (RUNX1) gene replicated in Sepracor but was nonsignificant when the P-values were combined across the replication trials.


In this manuscript, we identified a non-coding SNP (rs892940) located 5′ of the THRB gene that is associated with response to β2-agonists in the childhood asthma trial (family-based association test P-value=0.001, population-based P-value=0.09) and replicated this association in two adult asthma populations (combined P-value of 0.0012 in three replication populations and 0.0007 in all populations). Previous work by our group demonstrated that the expression of this gene is altered by exposure to a β2-agonist in human airway epithelial and smooth muscle cells, co-treated with pro-inflammatory cytokines and leukotriene D4, which are known to be elevated in asthmatic patients.6 Taken together, this thyroid hormone receptor gene is a novel candidate for regulation of variable response to a common asthma therapy. Further studies are necessary to determine whether the associated SNP or any variant in LD with it regulates the expression or activity of the THRB gene in response to bronchodilators. Genetic variants associated with BDR may facilitate genetic tests for predicting individual asthma therapy outcomes.

The THRB gene is located on chromosome 3p24.2, encoding for the β-subunit of the thyroid hormone receptor, which is one of two genes (α and β) that code for several isoforms.18 The thyroid hormone receptor is located in the nucleus, and upon binding to the thyroid hormone, regulates (both repress and activate) transcription through binding to T3 response elements either as a homodimer or heterodimer with retinoid X receptor beta (RXRB). The thyroid hormone, mediated through activation of its receptor, has been implicated in the growth and development of the lung as well as other organs in pre- and post-natal stages.19, 20 In a study of rats treated with this hormone, one group showed increased relaxation of the renal artery smooth muscle along with elevated cyclic AMP, nitric oxide synthase and nitric oxide, which is a potent vasodilator.21 Thus, genetic variants in THRB may affect the expression of this receptor and have wide-spread downstream effects on transcription regulation that may contribute to inflammation, constriction of the bronchial smooth muscle and obstruction of the airways. However, given the multiple protein isoforms, an earlier knockout mouse study demonstrated biological redundancy of the receptor activity.22 In addition, the biological effect of a potential regulatory mutation, which may alter the level of the wild-type protein in specific cells depending on the available transcription machinery, likely differs from a non-synonymous variant that alters the protein function in all cells expressing the gene. Thus, variable expression of the thyroid hormone receptor-β isoform may be cell-specific and may not have the detrimental effects of a coding variant or another gene without functional redundancy. The mechanism by which THRB modulates BDR is unknown and further investigations are necessary to determine its role in β2-agonist response.

A limitation of our study was the sample sizes of the asthma trials, especially for LOCCS (n=159) and LODO (n=155), which may have reduced the power to detect genetic associations. To compensate for the reduced power, we selected only those SNPs associated with BDR in both family-based and population-based analyses in CAMP to carry forward for replication. In addition, there were ascertainment biases of the replication populations, which may have contributed to heterogeneity across the cohorts. Specifically, participants in the LOCCS trial were previously treated with glucocorticoids and, consequently, had well-controlled asthma compared with the other trials. Glucocorticoid treatment has been shown to alter arginine metabolism by inhibiting the induction of nitric oxide synthase by cytokines, thereby reducing nitric oxide production, resulting in improved lung function.23 This may explain, in part, for the lower mean BDR and more normalized BDR distribution observed in the LOCCS trial compared with the other populations. Also, 60% of LODO participants were taking a controller medication, such as a long-acting β2-agonist, that could modify BDR. Finally, the Sepracor trial recruited only high responders to albuterol (BDR 15%). As a result of the heterogeneity in BDR distributions across these studies, we dichotomized the phenotype using the median value of each study to distinguish responders from nonresponders, which differ from the conventional thresholds for classifying responders from nonresponders.16, 17 The reproducibility of our association results across the three replication trials, given the population heterogeneity, makes our study more robust. Moreover, whereas the initial association analyses were conducted in a childhood asthma population, the replication trials were composed primarily of adults, but each included some childhood cases.

The fact that multiple SNPs across a number of genes were only modestly associated with BDR and no stronger haplotype effect within these genes were found suggests that the genetic associations identified in this manuscript are likely due to LD with the causative variant(s). Further studies are necessary to determine the functional role, if any, of the associated SNP in THRB on the expression of this gene or if it is in LD with other functional variants.

The identification of TFs that modulate BDR provides a better understanding of the inter-individual variability in response to β2-agonists, the most common class of asthma medications, as well as novel therapeutic targets for better symptom control. For example, antagonists, inhibitors or small interfering RNAs may be used to alter the expression of a specific TF gene. However, to date, few general TFs have been associated with asthma and asthma pharmacogenetics (that is, vitamin D receptor) since overexpression or suppression of such proteins are expected to result in wide-spread adverse effects. Therapeutic interventions to regulate the expression of TFs (for example, antisense oligonucleotides, TF decoys) would have to be cell-specific such as via aerosol or intra-tracheal administration, which specifically targeting TF expression in human lung cells such as airway epithelial and smooth muscle cells only, without affecting gene expression in other cell types or organs. Further studies are necessary to improve the administration of such therapies in humans in order to minimize adverse effects and optimize therapeutic benefits.


  1. 1

    Masoli M, Fabian D, Holt S, Beasley R . The global burden of asthma: executive summary of the GINA Dissemination Committee report. Allergy 2004; 59: 469–478.

    Article  PubMed  Google Scholar 

  2. 2

    Mannino DM, Homa DM, Pertowski CA, Ashizawa A, Nixon LL, Johnson CA et al. Surveillance for Asthma—United States, 1960-1995. In: Morbidity and Mortality Weekly Report. CDC Surveillance Summaries, 1998; 47 (No. SS-1): 1–28.

    Google Scholar 

  3. 3

    Moorman JE, Rudd RA, Johnson CA, King M, Minor P, Bailey C et al. National surveillance for asthma--United States, 1980-2004. MMWR Surveill Summ 2007; 56: 1–54.

    PubMed  Google Scholar 

  4. 4

    Rudd RA, Moorman JE . Asthma incidence: data from the National Health Interview Survey, 1980-1996. J Asthma 2007; 44: 65–70.

    Article  PubMed  Google Scholar 

  5. 5

    Duan QL, Tantisira KG . Pharmacogenetics of asthma therapy. Curr Pharm Des 2009; 15: 3742–3753.

    CAS  Article  PubMed  Google Scholar 

  6. 6

    Panebra A, Schwarb MR, Glinka CB, Liggett SB . Heterogeneity of transcription factor expression and regulation in human airway epithelial and smooth muscle cells. Am J Physiol Lung Cell Mol Physiol 2007; 293: L453–L462.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Childhood Asthma Management Program Research Group. The Childhood Asthma Management Program (CAMP): design rationale methods. Control Clin Trials 1999; 20: 91–120.

    Article  Google Scholar 

  8. 8

    The Childhood Asthma Management Program Research Group. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med 2000; 343: 1054–1063.

    Article  Google Scholar 

  9. 9

    American Lung Association Asthma Clinical Research Centers. Clinical trial of low-dose theophylline and montelukast in patients with poorly controlled asthma. Am J Respir Crit Care Med 2007; 175: 235–242.

    Article  Google Scholar 

  10. 10

    Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007; 81: 559–575.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Peters SP, Anthonisen N, Castro M, Holbrook JT, Irvin CG, Smith LJ et al. Randomized comparison of strategies for reducing treatment in mild persistent asthma. N Engl J Med 2007; 356: 2027–2039.

    Article  PubMed  Google Scholar 

  12. 12

    Fisher RA . Statistical methods for research workers. Hafner: New York, 1950.

    Google Scholar 

  13. 13

    Lange C, DeMeo D, Silverman EK, Weiss ST, Laird NM . PBAT: tools for family-based association studies. Am J Hum Genet 2004; 74: 367–369.

    Article  PubMed  PubMed Central  Google Scholar 

  14. 14

    Schaid DJ, Rowland CM, Tines DE, Jacobson RM, Poland GA . Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J Hum Genet 2002; 70: 425–434.

    Article  PubMed  Google Scholar 

  15. 15

    Veenstra DL, Saint S, Saha S, Lumley T, Sullivan SD . Efficacy of antiseptic-impregnated central venous catheters in preventing catheter-related bloodstream infection: a meta-analysis. JAMA 1999; 281: 261–267.

    CAS  Article  Google Scholar 

  16. 16

    Lung function testing: selection of reference values interpretative strategies. American Thoracic Society. Am Rev Respir Dis 1991; 144: 1202–1218.

    Article  Google Scholar 

  17. 17

    Pellegrino R, Viegi G, Brusasco V, Crapo RO, Burgos F, Casaburi R et al. Interpretative strategies for lung function tests. Eur Respir J 2005; 26: 948–968.

    CAS  Article  PubMed  Google Scholar 

  18. 18

    Lazar MA . Thyroid hormone receptors: multiple forms, multiple possibilities. Endocr Rev 1993; 14: 184–193.

    CAS  PubMed  Google Scholar 

  19. 19

    Perez-Castillo A, Bernal J, Ferreiro B, Pans T . The early ontogenesis of thyroid hormone receptor in the rat fetus. Endocrinology 1985; 117: 2457–2461.

    CAS  Article  PubMed  Google Scholar 

  20. 20

    Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DJ, Evans RM . The c-erb-A gene encodes a thyroid hormone receptor. Nature 1986; 324: 641–646.

    CAS  Article  PubMed  Google Scholar 

  21. 21

    Bussemaker E, Popp R, Fisslthaler B, Larson CM, Fleming I, Busse R et al. Hyperthyroidism enhances endothelium-dependent relaxation in the rat renal artery. Cardiovasc Res 2003; 59: 181–188.

    CAS  Article  PubMed  Google Scholar 

  22. 22

    Gauthier K, Chassande O, Plateroti M, Roux JP, Legrand C, Pain B et al. Different functions for the thyroid hormone receptors TRα and TRβ in the control of thyroid hormone production and post-natal development. EMBO J 1999; 18: 623–631.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  23. 23

    Saleh D, Ernst P, Lim S, Barnes PJ, Giaid A . Increased formation of the potent oxidant peroxynitrite in the airways of asthmatic patients is associated with induction of nitric oxide synthase: effect of inhaled glucocorticoid. FASEB J 1998; 12: 929–937.

    CAS  Article  PubMed  Google Scholar 

Download references


This work was supported by the grants U01 HL65899 and P01 HL083069: The Pharmacogenetics of Asthma Treatment from the National Heart, Lung and Blood Institute (NHLBI). We thank all families for their enthusiastic participation in the CAMP Genetics Ancillary Study, supported by the National Heart, Lung, and Blood Institute, N01-HR-16049. We acknowledge the CAMP investigators and research team for collection of CAMP Genetic Ancillary Study data. Additional support for this research came from Grants N01 HR16044, HR16045, HR16046, HR16047, HR16048, HR16049, HR16050, HR16051 and HR16052 from the National Heart, Lung and Blood Institute. All work on data collected from the CAMP Genetic Ancillary Study was conducted at the Channing Laboratory of the Brigham and Women's Hospital under appropriate CAMP policies and human subject's protections. The CAMP Genetics Ancillary Study is supported by U01 HL075419, U01 HL65899, P01 HL083069, R01 HL086601 and T32 HL07427 Grants from the NHLBI, National Institutes of Health. We acknowledge the American Lung Association (ALA) and the ALA's Asthma Clinical Research Centers investigators and research teams for use of LOCCS and LODO data, with additional funding from HL071394 and HL074755 from the NHLBI, and Nemours Children′s’ Clinic. GlaxoSmithKline supported the conduct of the LOCCS Trial by an unrestricted Grant to the ALA. We acknowledge Sepracor for use of the Sepracor data. QLD receives funding from the Canadian Institutes of Health Research.

Author information



Corresponding author

Correspondence to Q L Duan.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Duan, Q., Du, R., Lasky-Su, J. et al. A polymorphism in the thyroid hormone receptor gene is associated with bronchodilator response in asthmatics. Pharmacogenomics J 13, 130–136 (2013).

Download citation


  • bronchodilator response
  • transcription factor
  • association
  • thyroid hormone receptor-β
  • asthma
  • pharmacogenetics

Further reading


Quick links