Published ahead of print on March 11, 2005, doi:10.1164/rccm.200410-1312OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200410-1312OC
G-Protein–coupled Receptor Polymorphisms Are Associated with Asthma in a Large German PopulationUniversity Children's Hospital, Ludwig Maximilian's University Munich, Munich; Institute of Epidemiology, GSF–Research Centre for Environment and Health, Neuherberg; University Children's Hospital Dresden, Dresden; University Children's Hospital Leipzig, Leipzig; and Institute of Epidemiology, University of Ulm, Ulm, Germany Correspondence and requests for reprints should be addressed to Michael Kabesch, M.D., University Children's Hospital, Ludwig Maximilian's University Munich, Lindwurmstrasse 4, D-80337 München, Germany. E-mail: michael.kabesch{at}med.uni-muenchen.de
Rationale: Recently, a new asthma susceptibility gene, GPRA (G-protein–related receptor for asthma), has been identified by positional cloning. Initial association studies in a Finnish and Canadian population suggested an association with asthma and elevated serum IgE levels. Objective: In a large, nested case-control study, associations between GPRA polymorphisms, asthma, and serum IgE levels were analyzed. Methods: Using matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) technology, 1,872 German children aged 9 to 11 years (including 624 children with asthma and/or bronchial hyperresponsiveness) were genotyped for seven polymorphisms in the GPRA gene. Measurements: Hardy-Weinberg equilibrium was assessed, and association studies with single nucleotide polymorphisms (SNPs) and haplotypes were performed. Main Results: SNP 546333 increased the risk for asthma (odds ratio [OR], 1.40; 95% confidence interval [CI], 1.04–1.88; p = 0.025) and concomitant asthma and bronchial hyperresponsiveness (BHR; OR, 2.38; 95% CI, 1.22–4.66; p = 0.009). Also, SNP 585883 was associated with asthma (OR, 1.34; 95% CI, 1.04–1.72; p = 0.022) and asthma in combination with BHR (OR, 2.71; 95% CI, 1.45–5.09; p = 0.001). Furthermore, SNP 585883 was associated with elevated serum IgE levels (OR, 1.63; 95% CI, 1.10–2.42; p = 0.015). Haplotype combinations of risk alleles increased the OR for asthma to 1.83 (95% CI, 1.08–3.08; p = 0.024) and for asthma and concomitant BHR to OR 3.51 (95% CI, 1.08–11.37; p = 0.036). Conclusions: These results indicate that GPRA polymorphisms increase the susceptibility for asthma and BHR, and to a lesser degree for the elevation of serum IgE, in a German population, confirming initial observations in other white populations.
Key Words: asthma • children • GPRA • IgE • polymorphism In the last 2 years, three asthma genes have been identified by positional cloning (1–3). Only recently, Laitinen and coworkers (4) suggested a fourth potential asthma gene discovered by the same basic approach on chromosome 7p. The gene has been named GPRA for G-protein–coupled receptor for asthma susceptibility. GPRA polymorphisms were reported to be associated with elevated serum IgE levels in a Finnish and Canadian population. Furthermore, two GPRA isoforms, differing in length between 371 (isoform A) and 377 (isoform B) amino acids, were identified and transcript expression of both isoforms were detected in the lung. Immunohistochemical staining indicated that GPRA isoform B is significantly overexpressed in smooth muscle cells of human patients with asthma compared with normal control subjects. Taken together, these initially reported results indicate that GPRA may play a role in the development of asthma susceptibility. Thus, we tested the hypothesis proposed by Laitinen and coworkers, that GPRA polymorphisms are associated with asthma and the elevation of serum IgE in a large asthma case-control study of German children (n = 1,872).
Population Description Between 1995 and 1996, cross-sectional studies were conducted in Munich (International Study of Asthma and Allergies in Childhood II [ISAAC II]), in Dresden (ISAAC II), and in Leipzig to assess the prevalence of asthma and allergies in 7,963 schoolchildren aged 9 to 11 years (5, 6). Informed, written consent was obtained from all parents of children included in the study. All study methods were approved by the local ethics committees. Because the populations and phenotyping methods have been described in detail before (5), only an overview of the methods pertaining to this analysis is given here. For this analysis, all children who had a doctor's diagnosis of asthma and/or showed bronchial hyperresponsiveness (BHR; n = 671; Munich, n = 264; Dresden, n = 276; Leipzig, n = 131) were selected from the total study population. These children were matched at a 2:1 ratio with a random selection of healthy children without asthma or atopy and without a diagnosis of BHR (n = 1,342; Munich, n = 528; Dresden, n = 552; Leipzig, n = 262), and finally, only children of German origin who had both DNA and IgE data available were included (n = 1,872; Munich, n = 690; Dresden, n = 789; Leipzig, n = 393).
Subject Phenotyping
Spirometry and bronchial challenge.
Blood analysis.
Genetic Analyses
Statistical Analyses Hardy-Weinberg equilibrium (HWE) was tested as a genotyping quality control procedure using the  2 statistic, with expected frequencies derived from allele frequencies. Univariate associations between single nucleotide polymorphisms (SNPs) and outcomes for asthma and elevated IgE (90th percentile) were investigated using odds ratios (11). Haplotype frequencies were estimated using the expectation-maximization algorithm (12). To evaluate associations with traits, haplotype trend regression models were estimated, where the estimated probabilities of the haplotypes are modeled in a logistic regression as independent variables (13, 14). A stepwise approach was implemented with haplotypes from combinations of two to six SNPs. All statistical analyses were performed using the SAS statistical software package (version 9.1; SAS, Inc., Cary, NC); the haplotype analyses were implemented using the SAS/Genetics module, whereas the graphical overview of linkage disequilibrium (GOLD) was created using the GOLD 1.0 package (http://www.well.ox.ac.uk/asthma/GOLD).
Seven SNPs representing all previously observed haplotypes were selected for genotyping (Table 1). Call rates for SNPs in GPRA ranged from 96.7 to 99.1%, and all SNPs except 555608 were in HWE (Table 2). Thus, SNP 555608 was excluded from further haplotype analyses. In our large German sample, all SNPs showed allele frequencies similar to those originally reported in the Finish and Canadian population (Table 1). Using a normalized correlation coefficient, the degree of linkage disequilibrium was assessed between single polymorphisms, and a GOLD plot of the average linkage disequilibrium across the GPRA SNPs is shown (Figure 1). Color denotes the degree of linkage disequilibrium between a given pair.
A summary of association results between single GPRA polymorphisms, asthma, BHR, concomitant asthma and BHR, and elevated serum IgE (90th percentile) is given in Table 3 as odds ratios (OR) and 95% confidence intervals (95% CI). A significant association was observed between the SNPs 546333, and 585883 and the risk to develop asthma, whereas the minor allele of SNP 563704 was protective against the development of asthma. Associations with BHR per se did not reach statistical significance. However, in a subgroup analysis of children with concomitant asthma and BHR, the addition of BHR amplified the association between the two SNPs related to asthma: SNP 546333 was highly significantly associated with asthma and BHR (OR, 2.38; 95% CI, 1.22–4.66; p = 0.009) and a similar modification of the risk for asthma and concomitant BHR was observed for SNP 585883 (OR, 2.71; 95% CI, 1.45–5.09; p = 0.001). In addition, homozygote carriers of the polymorphic forms of SNP 563704 were protected against the development of elevated serum IgE levels (OR, 0.33; 95% CI, 0.12–0.93; p = 0.028). In contrast, having at least one polymorphic allele of SNP 585883 increased the risk for very high serum IgE levels (OR, 1.63; 95% CI, 1.01–2.54; p = 0.015).
For haplotype analysis, 1,794 samples were available in which genotyping was successful for all six GPRA polymorphisms (SNP 555608 was excluded from further analysis because of HWE). When all available haplotype combinations with a frequency more than 3% were assessed, haplotype CCAACC, which combined the risk alleles of all observed single SNP associations, showed a significant association with asthma (OR, 1.83; 95% CI, 1.08–3.08; p = 0.024) and with the subphenotype combination of asthma and BHR (OR, 3.51; 95% CI, 1.08–11.37; p = 0.036; Table 4). A borderline association with elevated IgE levels was observed for the same high-risk haplotype (OR, 2.16; 95% CI, 0.97–4.83; p = 0.060). Because our population was selected as a case-control population for asthma, the effects of GPRA polymorphisms on other atopic diseases like hay fever or atopic dermatitis could not be assessed because of potential selection bias in a case-control setting.
Asthma genes detected by positional cloning based on linkage studies had been anticipated for years and when they finally started to appear, they were celebrated enthusiastically as a big step toward the disentanglement of the Gordian knot of the complex asthma genetics. However, it seems that reports on positionally cloned genes deserve the same amount of initial skepticism as the ever-increasing number of results stemming from association studies using the candidate gene approach. Learning from experience with ADAM33, DPP10, and PHF11, it seems clear that the dissection of the functional relevance of SNPs in these genes as well as independent replications of the original associations in large population samples are key for the assessment of their impact on asthma susceptibility. Although functional studies of polymorphism effects are necessary to understand disease mechanisms, they are time-consuming and tedious. On the other hand, using modern genotyping techniques, replication of associations in independent populations can be achieved quickly to prove or disprove initial findings. This study examined seven SNPs in the GPRA gene. Although all tested SNPs are located within a short physical distance in the region of intron 2, they represent all possible haplotypes in the previously described 77-kb risk haplotype block harboring the GPRA gene (4). SNP 555608 has been finally excluded from association analyses because it showed deviation from expected HWE genotype frequencies. A departure from HWE may be indicative of methodologic assay problems in genotyping a certain SNP. Thus, it has been proposed to exclude SNPs deviating from HWE as an approach to improve data quality (15). However, distortion from HWE may also be caused by biological effects or population characteristics leading to a nonrandom distribution of alleles. This can be the case in populations with strong founder effects or clear biological and reproductive advantages of certain SNPs. In the case of GPRA, both methodologic and biological effects could be responsible for the observed deviation from HWE. Because we cannot rule out a methodologic genotyping error or a selection bias, SNP 555608 was excluded from further analyses following common practice in genetic studies to avoid false association reports. The associations initially observed between GPRA haplotypes, elevated serum IgE, and asthma in cohorts from Finland and Canada could be replicated in our large German case-control sample. A statistically significant increased risk to develop asthma was observed for two single SNPs. When a subphenotype of concomitant asthma and BHR was analyzed, the observed risk for both SNPs and the strength of these associations further increased. These findings may indicate that GPRA is a gene involved in lung-specific allergic inflammation because GPRA polymorphisms are strongly associated with BHR in the presence of an asthmatic airway inflammation but no association was observed with BHR alone. Because of the consistent association between GPRA and the regulation of serum IgE in our population and in that of Laitinen and coworkers (4), it may be speculated that GPRA is not only a pulmonary-expressed gene important for asthma susceptibility but that it may also play a more general immunomodulatory role. On the basis of the detection of GPRA molecules in ciliated cells of the airways and airway smooth muscle cells after sensitization (4), it is possible that GPRA may play an important role in conferring bronchial asthma by translating atopic sensitization into bronchial inflammation and BHR. In this sense, GPRA may be a link between systemic atopic sensitization and the development of pulmonary symptoms. Thus, GPRA polymorphisms could modify this translation from atopy to asthma. One could also hypothesize that individuals with asthma with certain GPRA polymorphisms are more likely to present BHR as an expression of a more severe form of asthma and that GPRA may also be involved in airway remodeling. However, to further investigate this hypothesis, cohort studies may be needed. An even more profound effect was observed when GPRA haplotypes were analyzed. Associations for all haplotypes composed of the six investigated SNPs with a frequency larger than 3% were assessed. The combination CCAACC, present in 9% of the population, showed a higher risk for asthma than any single SNP. Furthermore, the risk for asthma and concomitant BHR increased substantially from an OR of 2.71 for the maximum effect for a single SNP to 3.51 for the risk haplotype CCAACC (Table 4). It is interesting that the haplotype CCAACC conferring the highest risk for asthma and asthma with concomitant BHR in our study population is equivalent to the originally reported haplotype H4, which was associated with elevated levels of serum IgE in the Finnish population (4). We also investigated haplotype H2 (CCAG[T]CG) reported to be associated with asthma in the Quebec population but could not confirm the previous findings. In addition, the exclusion of SNP 555608 because of HWE in our population made it impossible to discriminate haplotypes H3 and H6 originally described by Laitinen and coworkers as nonrisk. In a parallel study by Melén and coworkers (16), however, the rather infrequent haplotype H6 conferred a risk for atopic sensitization in two populations. Because of the missing information on SNP 555608, both H3 and H6 had to be analyzed together in our study and a potential effect of H6 may have been missed. In contrast to the original study by Laitinen and coworkers in adults, we here present data on the association between GPRA polymorphisms and asthma, BHR, and IgE regulation in children. There may be profound differences in clinical as well as immunologic aspects of childhood asthma compared with adult asthma. It has been proposed that different forms of asthma may overlap during childhood, potentially representing different etiologies. Early transient, continuous, and late-onset wheezing can be distinguished in longitudinal studies of childhood asthma (17). In our cross-sectional study, children were evaluated at the age of 9 to 11 years, and the definition of asthma was based on lifetime prevalence assessed through questionnaires. Therefore, in our population, these three subgroups of wheezers cannot be determined unambiguously. However, because of the inclusion criteria in the questionnaire, transient wheezers should not have been labeled asthmatic in our study population. Even though some heterogeneity exists between different forms of childhood asthma and between childhood asthma and adult asthma, our results in combination with those of Laitinen and colleagues (4) indicate that mechanisms common to both childhood and adult asthma may be influenced by GPRA polymorphisms. How GPRA exerts its function and how SNPs in GPRA alter these mechanisms is still not clear. However, it seems that the consistent effects of GPRA polymorphisms now documented in different independent studies may well justify a closer look into the functional properties of GPRA in the development of asthma, BHR, and atopy.
The authors thank Cecilia Lindgren and Juha Kere from the Karolinska Institute, Stockholm, Sweden, for their kind supply of primer and assay information for genotyping. Genotyping was performed in the Genome Analysis Center of the GSF (Gesellschaft für Strahlenforschung).
Supported by the German Ministry of Education and Research (BMBF)/National Genome Research Network (NGFN): research grants NGFN 01GS 0122 and NGFN IE-S08T03. Conflict of Interest Statement: M.S.D.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; D.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; N.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; T.I. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.K.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; E.v.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form October 5, 2004; accepted in final form March 2, 2005
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS