Haplotypes of G ProteineCcoupled Receptor 154 Are Associated with Childhood Allergy and Asthma
http://www.100md.com
美国呼吸和危急护理医学 2005年第5期
Institute of Environmental Medicine, Centre for Allergy Research
Department of Biosciences, Karolinska Institutet
Clinical Allergy Research Unit and Clinical Immunology and Allergy Unit, Department of Medicine, Karolinska Institutet and University Hospital
Department of Occupational and Environmental Health, Stockholm County Council, Stockholm
Clinical Research Centre, Karolinska University Hospital, Huddinge
AstraZeneca R&D Mlndal, Mlndal, Sweden
Institute for Risk Assessment Sciences, University of Utrecht, The Netherlands
University Children's Hospital, Ludwig Maximilian's University Munich, Munich, Germany
GeneOS Limited, Helsinki
Department of Medical Genetics, University of Helsinki, Helsinki, Finland
Division for Immunology, Zurich University Children's Hospital, Zurich, Switzerland
Children's Hospital, Schwarzach, Austria
ABSTRACT
Rationale: Allergic diseases are influenced by both genes and environment. A 70-kb haplotype block in the G proteineCcoupled receptor for asthma susceptibility gene (GPR154; alias GPRA) on chromosome 7p was recently identified to influence susceptibility to asthma and elevated total serum IgE levels in adults. Objectives: To assess the impact of GPR154 on childhood allergic disease, including allergic sensitization, asthma, and rhinoconjunctivitis, in study populations with diverse environmental backgrounds. Methods: We studied farm children, Steiner school children, and two reference groups from five Western European countries in the cross-sectional PARSIFAL (Prevention of Allergy Risk factors for Sensitization In children related to Farming and Anthroposophic Lifestyle) study and a sample of children from the Swedish birth cohort study BAMSE. DNA samples from 3,113 PARSIFAL and 800 BAMSE children were genotyped for 7 GPR154 polymorphisms and haplotypes were inferred. The proportions of alleles and haplotypes (H1eCH7) were compared in affected children with their healthy counterparts. Results: Data indicate a global association of the haplotype block to sensitization (allergen-specific serum IgE 0.35 kU/L, p = 0.022), with significant haplotype-specific associations for H1, H5, and H6. Haplotypes H1 and H5 were also significantly associated with childhood allergic asthma (p = 0.045 and p = 0.023, respectively), and H5 to asthma regardless of sensitization. A broader involvement of GPR154 in allergic diseases was further supported in allergic rhinoconjunctivitis (H3: p = 0.046). The associated haplotypes could be allocated into risk (H5/H6) and nonrisk (H1/H3) groups, a pattern supported by allelic association of single nucleotide polymorphisms (SNPs) rs324384 and rs324396. Conclusions: Our results indicate that polymorphisms and haplotypes in the haplotype block of GPR154 are associated with asthma, rhinoconjunctivitis, and sensitization in European children.
Key Words: asthma children genetic association GPRA gene IgE
Linkage analyses for asthma susceptibility and IgE regulation loci performed some years ago initially held promise in the search for a major gene, but strong candidate genes have been identified only recently (1eC4). The G proteineCcoupled receptor for asthma susceptibility gene (GPR154; alias GPRA), an orphan G proteineCcoupled receptor gene on chromosome 7p, which was identified through positional cloning, is the latest candidate gene suggested to influence the susceptibility to asthma and elevated total IgE levels (4). The genetic evidence was supported by single nucleotide polymorphism (SNP) and haplotype associations, a distinct distribution of protein isoforms between bronchial biopsies from healthy adults and adults with asthma, and increased expression of the GPR154 gene in experimentally induced lung inflammation in mice. The observed 77-kb haplotype block showed a similar risk and nonrisk pattern in all three populations studied. Significant relative risks were reported for elevated total IgE among carriers of particular haplotypes (H4 or H5) and for asthma among homozygous H2 carriers (1.4, 95% confidence interval [CI] 1.1eC1.9 and 2.5, 95% CI 2.0eC3.1, respectively) (4). Given the complex pathophysiology and hereditary patterns of asthma and allergy, these results suggest that the GPR154 gene or other genetic variants in linkage disequilibrium with the identified haplotype block may be associated more broadly with allergic disease, including asthma, allergic sensitization, and rhinoconjunctivitis.
Replication is important as original genetic findings often correlate only moderately well with subsequent research on the same association, with the strongest effect estimate typically occurring in the first study (5, 6). Large, well powered original studies are naturally desirable to avoid spurious associations. Systematic metaanalysis and pooling of data are also powerful tools for assessing effects across several studies and different datasets (7).
Here we present results from GPR154 analyses exploring association in the same haplotype region to childhood asthma, allergic sensitization, and allergic rhinoconjunctivitis using two large and well characterized datasets, the Western European cross-sectional PARSIFAL (Prevention of Allergy Risk factors for Sensitization In children related to Farming and Anthroposophic Lifestyle) study and the Swedish birth cohort study BAMSE. The study population represents a broad spectrum of European children of different ages and lifestyles: farmers, children with an anthroposophic lifestyle, and subsamples of the general population from small communities and large cities.
METHODS
Participants and Questionnaire
The study included 3,113 school children 5 to 13 years old from the cross-sectional PARSIFAL study and 800 children at about 4 years of age from the birth cohort study BAMSE. PARSIFAL children originally included a total of 14,893 from 5 Western European countries. It was designed to investigate the role of different lifestyles and environmental exposures in farm children, Steiner school children (mainly from anthroposophic families), and two corresponding reference groups to identify protective factors for development of asthma and allergic disorders. In Austria, Germany, the Netherlands, and Switzerland farm children were recruited from schools in rural areas known to have a high percentage of farmers; in Sweden through the Farming Registry at the National Bureau of Statistics. Children with an anthroposophic lifestyle were recruited from classes in Steiner schools. The respective reference groups were recruited with similar methods from the same geographical areas, to match for environmental exposures not related to the lifestyles of interest. Parents completed a detailed questionnaire on allergic diseases, infectious history, and environmental exposures, largely based on questions from the International Study of Asthma and Allergies in Childhood (ISAAC) (8), BAMSE (9), the ALEX study (10), and a Swedish study focused on the anthroposophic lifestyle (11). A subsample of children (n = 4,854) and their parents were asked for consent to collect venous blood for allergen specific IgE analyses and DNA extraction. The present study includes 3,113 children (1,579 boys, 1,534 girls) with complete questionnaire data, adequate DNA material, and consent to genetic analyses.
The BAMSE study is based on a well characterized, representative cohort of children followed prospectively from birth, as previously described (9, 12). In brief, 4,089 newborn infants were recruited between 1994 and 1996 and data on parental allergic diseases and residential characteristics were obtained by questionnaire. When the children were 1, 2, and 4 years old, additional mailed questionnaires included questions on symptoms related to wheezing and allergic disease (largely following ISAAC [8]). Blood samples drawn at 4 years of age (mean 4.3) were available from 2,298 children with questionnaire data and informed consent for genetics. In a caseeCcohort sampling design, a sequential random sample of 709 children was selected as a primary subcohort for genetic analyses (13), from whom 686 blood samples provided adequate DNA. Children with reported doctor's diagnosis of asthma up to age four were identified as asthma cases (n = 62), and the remaining children with no asthma reported constituted random controls. From among the 2,298 children not included in the subcohort, a further 114 children with reported asthma were also selected, resulting in a total of 176 asthma cases and 800 samples altogether. The randomly sampled subcohort also allows for unbiased analysis of other outcomes, e.g., sensitization, by comparing affected and nonaffected children in the subcohort.
Ethical approval and informed consent were obtained in each participating study center.
Definition of Outcome Phenotypes
Allergen-specific serum IgE was measured against the Phadiatop mix of eight common inhalant allergens and the fx5 mix of six common food allergens (Pharmacia CAP System; Pharmacia Diagnostics AB, Uppsala, Sweden). Asthma was defined as a physician's diagnosis of asthma according to the questionnaire response (see online supplement for questionnaire details). For the 4-year-old BAMSE children, the phenotype "current asthma," defined as physician-diagnosed asthma and one or more episodes of wheezing in the last 12 months at 4 years, was also used. Allergic sensitization was defined as an IgE level greater than or equal to 0.35 kU/L by either Phadiatop or fx5. Allergic asthma (or atopic asthma) was derived from the combination of physician's diagnosis of asthma and sensitization. Allergic rhinoconjunctivitis was defined as a questionnaire response of having had rhinitis and conjunctivitis symptoms without concurrent cold within the last 12 months, in combination with sensitization to inhalant allergens (Phadiatop) in PARSIFAL only. The young age of the participants and the questionnaire data available in BAMSE did not allow for a comparable diagnosis.
SNP Genotyping and Data Checking
DNA was extracted from peripheral blood leukocytes using standard techniques. Seven of the SNPs investigated in the Laitinen and coworkers study were selected to tag the previously observed haplotypes (Figure 1A) (4). Primers were designed by the SpectroDesigner software (Sequenom Inc., San Diego, CA) and genotyping was performed using MALDI-TOF technology (see online supplement for details). All samples were analyzed in a sex-specific assay (AMELX/Y; OMIM 410000 / 300391) to evaluate the accuracy in sample and database management.
Data Analysis
Population Hardy-Weinberg equilibrium was evaluated for each SNP in both the PARSIFAL (combined reference groups) and BAMSE subcohort samples, with a chi square test for the distribution of genotypes. Allele frequencies were analyzed for cases versus control subjects in Haploview 2.05 software (http://www.broad.mit.edu/personal/jcbarret/haploview) using a chi square test. Mantel-Haenszel combined odds ratios (ORs) and 95% CIs for comparison of allele frequency in cases versus control subjects were calculated using standard procedures. Linkage disequilibria between the genotyped SNPs were obtained in Haploview, which estimates the recombination history between SNPs by evaluating the normalized measure of allelic association, D'. Haplotype block generation was performed using the default algorithm defined by Gabriel and coworkers (14). Two separate approaches were used for estimation of haplotype association with the defined outcomes in the PARSIFAL and BAMSE study groups and the combined material. First, haplotype association was tested in the statistical software R (15) using the haplo.score algorithm, which estimates probabilities for each possible haplotype for each individual based on the full study populations (cases and control subjects combined) and then generates both global and haplotype-specific score statistics by comparing estimated haplotype frequencies in cases and control subjects (16). Adjustment was made for a variable with one category for each combination of country-of-origin and sampling group, and 10,000 simulations were performed to obtain empirical p values for the observed data. All p values are nominal two-sided values. Second, for estimation of ORs and CIs attributed to each haplotype, the frequency of haplotypes was again estimated in the full study populations (cases and control subjects combined) using the expectation maximization algorithm (17), implemented in the Haploblocks program (M. Zucchelli and J. Kere, unpublished data). Each individual contributed two haplotypes (one from each chromosome) to the analyses and the frequency and estimated counts of each haplotype were then assessed against all others using standard procedures for OR calculations.
Phylogenetic Analyses
The genomic DNA of seven Finnish individuals who were each homozygous for one of the haplotypes H1eCH7 was sequenced across the 133 kb region (4). Repeat regions such as SINE, LINE, LTR, MER1, and MER2 elements covered 60% of the sequence. Polymerase chain reaction assays were designed to overlap. Repeat regions were resequenced when it was possible to design primers in unique sequence. We identified 152 polymorphic sites (SNPs and deletion/insertion polymorphisms (DIPs) and created the phylogenetic analysis of the seven haplotypes spanning the locus using PHYLIP (the PHYLogeny Inference Package, http://www.hgmp.mrc.ac.uk/). All 152 recognized polymorphisms were included in this analysis, which thus substantially refines the analysis of only 40 SNPs in the region from the seven individuals presented earlier (4).
RESULTS
Baseline characteristics showed that the two study populations were similar in terms of sex distribution and prevalence of asthma and allergic sensitization, whereas mean age at study differed by design (Table 1). The overall success rates for each studied SNP ranged from 93.2 (rs324377) to 98.1% (rs324396). Results from the AMELXY assay indicated 3.1% sex discrepancies. None of the assayed SNPs deviated significantly from population Hardy-Weinberg equilibrium in the PARSIFAL reference groups (p > 0.01), but Hopo546333 in BAMSE was slightly off Hardy-Weinberg equilibrium (p = 0.001 for controls with no sensitization). Further investigation revealed that 9 homozygous AA were observed instead of the expected 3. Because this difference between observed and expected genotypes was minor and only affected the smaller BAMSE study group, the SNP was also included in the analysis.
Pairwise comparisons between the genotyped SNPs indicated no historical recombination and the SNPs were thus defined to be contained within one preserved 70-kb haplotype block for all groups (Figures 1A and 1B and Table 2). Haplotype frequencies in the different countries are presented in Table E4 (see the online supplement). Extended phylogenetic analyses in Finnish subjects, using 152 instead of 40 sequence changes, revealed a closer relationship between haplotypes H6 and H7 than that previously presented (Figure 1C) (4).
We initially tested the hypothesis that GPR154 haplotypes are associated with childhood sensitization and asthma, based on previously reported findings in adults (4). The analysis for the combined studies revealed a global association of the haplotype block to sensitization (p = 0.022) as a result of haplotype-specific associations for H1 (p = 0.038), H5 (p = 0.019), and H6 (p = 0.015), whereas H1 and H5 were similarly, but not significantly, associated with asthma in the whole sample (Table 3). In the PARSIFAL population the H5 haplotype was significantly associated with asthma (p = 0.033) (Table 3). Similarly, with "current asthma" as the outcome for the BAMSE children, in an attempt to reduce misclassification by excluding early childhood transient wheezers from the asthma group, H5 was significantly associated with asthma in the combined sample (p = 0.026, data not shown).
To further characterize risk of disease attributed to each haplotype in the GPR154 gene, we performed additional analyses for estimation of haplotype specific ORs (Table 3). Separate analyses of the PARSIFAL and BAMSE samples showed reasonably concordant results, notably the ORs for H1 and asthma or H5 and sensitization, and the overall pattern of haplotype effects.
These findings encouraged us to test whether GPR154 showed effects for other allergic disease phenotypes by investigating association with allergic asthma and allergic rhinoconjunctivitis. Haplotypes H1 and H5 showed stronger associations with allergic asthma for the whole group (p = 0.045, OR = 0.78, 95% CI, 0.63eC0.96 and p = 0.023, OR = 1.47, 95% CI, 1.01eC2.14, respectively) and in the PARSIFAL population H3 was found to be associated with allergic rhinoconjunctivitis (p = 0.046, OR = 0.68, 95% CI, 0.48eC0.96, data not shown).
Although the seven SNPs included in this study were specifically selected for tagging the previously identified haplotypes (4), we also examined possible allelic associations for each SNP and the outcomes mentioned above. The strongest associations were observed for rs324384 (minor allele C) with sensitization (OR = 0.88, 95% CI, 0.80eC0.98) and for rs324396 (minor allele T) with sensitization (OR = 0.88, 95% CI, 0.79eC0.99) and asthma (OR = 0.83, 95% CI, 0.70eC0.97) (see Table E2 in online supplement). Both SNPs were also significantly associated with allergic asthma (OR = 0.81, 95% CI, 0.66eC1.00 and OR = 0.78, 95% CI, 0.62eC0.98, respectively). The strongest allelic association with allergic rhinoconjunctivitis was OR = 0.83, 95% CI, 0.66eC1.04 for rs324384 (C).
DISCUSSION
Large well characterized epidemiologic studies are key to verifying results from complex disease genetic mapping and to assessing the relevance of findings on the population level. In this study we investigated a haplotype region in the novel G proteineCcoupled receptor, GPR154, in two separate datasets of European children and demonstrated an association between its genetic variants and allergic diseases and asthma in childhood. Detailed analyses enabled the effective dissection of haplotype effects of GPR154; both revealing new and confirming previous relationships to asthma and IgE-mediated allergic disorders.
Recently, performing association studies on haplotypes rather than on individual SNPs has gained popularity with the appreciation that a limited number of haplotypes in regions of high linkage disequilibrium (haplotype blocks) capture most of human genetic heterogeneity in that region. Such haplotypes can be tagged by a small number of SNPs and may increase the power of genetic association studies (14, 18, 19). Evaluating haplotype association was central in the present study, because this was the core of the findings in the Laitinen and coworkers study (4) from which we selected SNPs tagging the observed risk and nonrisk haplotypes.
In all, the expectation maximization algorithm used predicted only seven haplotypes with frequencies over 3% from the 7,826 chromosomes investigated. These haplotypes were identical to those previously reported (4). Our observations emphasize the high linkage disequilibrium existing between the investigated SNPs and the remarkable preservation of the GPR154 haplotype block in the outbred populations of the study. The similar haplotype distribution in each country also supports the finding of a well preserved genetic region across several populations.
We observed association between allergic disease and a distinct set of risk (H5/H6) and nonrisk (H1/H3) haplotypes, which was supported by allelic association of the SNPs rs324384 and rs324396, differentiating the nonrisk haplotypes. Whereas haplotype H6 conferred risk of sensitization in this study, no association to high serum IgE was reported in the original study (4). Of note, the H6 haplotype occurred more frequently in the PARSIFAL population (10.8%, Table 2) than in the Finnish and Canadian populations (5.1 and 8.7%, respectively). Thus, our observation could result from a higher power to detect effects of H6. The notion of H6 as a risk haplotype is supported by the refined phylogenetic analyses. The SNP 522363[G/C] (rs 323922) was initially proposed as a tagging SNP for risk (C) and nonrisk haplotypes (G) (4), but these new data show a somewhat different pattern. Further, we did not find any associations between the SNP in question and the studied outcomes, and a recently published Korean study also showed lack of association for this SNP to asthma and high serum IgE (20).
In our data, we were not able to detect any associations between the outcomes of interest and the H2 and H4 haplotypes, which were suggested as risk haplotypes for asthma and elevated total IgE, respectively (4). This could clearly reflect true differences between the genetic effects of GPR154 variants or other genetic variants in linkage disequilibrium with the different haplotypes that may vary in the different populations of the studies, but some aspects concerning study design also deserve attention. Laitinen and coworkers reported haplotype associations with asthma and high serum IgE-levels in mostly adults, whereas we attempted to replicate this finding in children 4 to 13 years old. Thus, some phenotypic heterogeneity between the studies is likely. Longitudinal studies have shown that approximately 50% of children with asthma will outgrow their symptoms after a number of years (21), whereas some individuals develop asthma only in adulthood. This natural development makes the allocation of children into asthma or nonasthma groups both age dependent and complicated in terms of previous asthma history. Evidence also suggests that childhood and adult asthma differ in a number of parameters, including sex preponderance, airflow obstruction, sensitization and structural changes of the airways, although remodelling may occur already in children with severe asthma (22eC24). In our study we defined sensitization with IgE antibody levels against food allergens or aeroallergens, which are correlated with total IgE to some extent, but total IgE has also been associated with asthma independently of specific IgE levels (25, 26). Given the possible phenotypic heterogeneity between the studies, the replication of associations between asthma/sensitization and GPR154 variants in children suggests a general role for this gene in respiratory allergic disease, and the present results do not exclude that H2 and H4 may be associated with increased risk in well established, adult asthma and elevated serum IgE phenotypes.
The PARSIFAL and BAMSE samples each have unique study design features. The PARSIFAL samples represent a varied set of Western European children selected on the basis of farming and anthroposophic lifestyles, and their respective reference groups from rural and sub-urban/urban communities. The four groups were exposed to differential environments known to influence the development of IgE-mediated allergic disorders (11, 27), and these aspects are being investigated in separate analyses. We intentionally focused on the effect across the total sample, as the purpose was to replicate and explore the original GPR154 associations in terms of a significant overall genetic association with asthma and allergic phenotypes rather than addressing specific geneeCenvironment interaction hypotheses. Such analyses are planned when the environmental aspects have been explored. Considering that the PARSIFAL children represent a heterogeneous population from five European countries, the significant association estimated for the four groups reinforce the notion of an association between GPR154 variants and allergic disease in general. With the younger BAMSE cohort, which differs from the PARSIFAL study in terms of population and study design, we aimed to extend the analyses to increase representativity and power, and to detect general associations across different populations and age groups. The results from the BAMSE samples separately did not reach statistical significance for several potential reasons. First, the children in BAMSE were around 4 years of age, and although we used questionnaire reported doctor's diagnosis of asthma as common outcome, this may to a greater extent represent a phenotype of transient wheezing, often triggered by infections, that does not share the characteristics of asthma later in childhood or adult asthma. Restriction of the analyses to children with current asthma at the age of 4 years, which is equivalent to a definition used in, for example, the Home Allergens and Asthma Study (28), also showed that the associations became stronger. Second, the BAMSE study is smaller, resulting in more limited statistical power. Nevertheless, the point estimates for association largely agree with those from the PARSIFAL study, and the combined analyses show that the associations are valid with this dataset included.
The quality assessments performed and the high success rate imply that the genotyping data we obtained are robust and reliable. For the SNP that deviated from Hardy-Weinberg equilibrium in the BAMSE samples, homozygous individuals were only minimally overrepresented compared with the expected number, and this difference should have only marginal, if any, influence on the results. Nevertheless, apart from the possibility of true deviation in the population, possible reasons for such Hardy-Weinberg equilibrium deviation include nonspecific assays (although primers were controlled), genotyping errors, DNA quality, unknown issues, or simply deviation from equilibrium by chance (29).
When we compared those selected for genotyping with the remaining subjects for BAMSE (13) and PARSIFAL, the genotyped subjects appeared to be essentially representative, with no important differences noted. Furthermore, selection bias will only occur if selection factors differ by genotype, which is mostly unlikely.
Although population stratification may often be relatively unlikely to cause bias in real-life settings (30), it is difficult to completely rule out in creating false positive results. In this study, we note that haplotype frequencies were quite similar in the five countries. In addition, the haplotype association p values estimated in haplo.score were adjusted for country of origin and study group, which should rule out any major residual effect of population stratification.
In summary, we report that GPR154 variants are significantly associated with allergic sensitization, asthma, and allergic rhinoconjunctivitis in children from 4 to 13 years of age from two Western European datasets, with the H5 and to some extent H6 haplotypes conferring risk and the H1 and H3 haplotypes exerting protective effects. We observed the previously reported haplotype block and were able to identify the described haplotypes, supporting the original finding (4). We also report refined phylogenetic analyses that support subdivision of the observed haplotypes into risk and nonrisk variants. Further research is needed to clarify the effect of the observed haplotypes on, for example, GPR154 protein levels and the immune response. Some functional data on GPR154 supporting the reported associations were given in the original report and are also available elsewhere (4, 31).
Our study and a parallel, independent study by Kormann and coworkers (32) together provide robust replication for GPR154 haplotypes as risk determinants for asthma and related disorders, although the details differ. In addition, our results and those of Kormann and coworkers (32) also extend the original findings of Laitinen and coworkers identified by positional cloning in an inbred population (4) to children with asthma in a wide spectrum of Western European outbred populations, and define not only risk and neutral haplotypes, but also protective haplotype variants.
Acknowledgments
The authors thank all children and parents in the PARSIFAL and BAMSE studies, as well as fieldworkers in all centers. They also thank Dr. Marco Zucchelli of the Department of Biosciences at Karolinska Institutet for statistical assistance.
E.M. and S.B. contributed equally to the study.
The PARSIFAL Genetics Study Group: Gran Pershagen, Fredrik Nyberg, Annika Scheynius, Marianne van Hage-Hamsten, Magnus Wickman (Sweden); Charlotte Braun-Fahrlnder, Marco Waser, Roger Lauener (Switzerland); Bert Brunekreef, Gert Doekes, Dieneke Schram (the Netherlands); Erika von Mutius, Marcus Ege (Germany); Josef Riedler, Waltraud Eder, Ellen elagger (Austria), Michael Kabesch (Munich lab), Juha Kere, Cecilia Lindgren, Erik Meleen (Stockholm lab), Fernando Martinez (Tucson lab).
This article has an online supplement, which is accessible from this issue's table of content online at www.atsjournals.org
REFERENCES
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, et al. Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;418:426eC430.
Allen M, Heinzmann A, Noguchi E, Abecasis G, Broxholme J, Ponting CP, Bhattacharyya S, Tinsley J, Zhang Y, Holt R, et al. Positional cloning of a novel gene influencing asthma from chromosome 2q14. Nat Genet 2003;35:258eC263.
Zhang Y, Leaves NI, Anderson GG, Ponting CP, Broxholme J, Holt R, Edser P, Bhattacharyya S, Dunham A, Adcock IM, et al. Positional cloning of a quantitative trait locus on chromosome 13q14 that influences immunoglobulin E levels and asthma. Nat Genet 2003;34:181eC186.
Laitinen T, Polvi A, Rydman P, Vendelin J, Pulkkinen V, Salmikangas P, Makela S, Rehn M, Pirskanen A, Rautanen A, et al. Characterization of a common susceptibility locus for asthma-related traits. Science 2004;304:300eC304.
Weiss ST, Raby BA. Asthma genetics 2003. Hum Mol Genet 2004;13 Spec No 1:R83eC89. Epub 2004 Feb 05. Review.
Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG. Replication validity of genetic association studies. Nat Genet 2001;29:306eC309.
Altshuler D, Hirschhorn JN, Klannemark M, Lindgren CM, Vohl MC, Nemesh J, Lane CR, Schaffner SF, Bolk S, Brewer C, et al. The common PPARgamma Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes. Nat Genet 2000;26:76eC80.
Asher MI, Keil U, Anderson HR, Beasley R, Crane J, Martinez F, Mitchell EA, Pearce N, Sibbald B, Stewart AW, et al. International study of asthma and allergies in childhood (ISAAC): rationale and methods. Eur Respir J 1995;8:483eC491.
Melen E, Kere J, Pershagen G, Svartengren M, Wickman M. Influence of male sex and parental allergic disease on childhood wheezing: role of interactions. Clin Exp Allergy 2004;34:839eC844.
Riedler J, Braun-Fahrlander C, Eder W, Schreuer M, Waser M, Maisch S, Carr D, Schierl R, Nowak D, von Mutius E. Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey. Lancet 2001;358:1129eC1133.
Alm JS, Swartz J, Lilja G, Scheynius A, Pershagen G. Atopy in children of families with an anthroposophic lifestyle. Lancet 1999;353:1485eC1488.
Lannero E, Kull I, Wickman M, Pershagen G, Nordvall SL. Environmental risk factors for allergy and socioeconomic status in a birth cohort (BAMSE). Pediatr Allergy Immunol 2002;13:182eC187.
Melen E, Gullsteen H, Zucchelli M, Lindstedt A, Nyberg F, Wickman M, Pershagen G, Kere J. Sex-specific protective effects of interleukin-9 receptor haplotypes in childhood wheezing and sensitisation. J Med Genet 2004;41:e123.
Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Higgins J, DeFelice M, Lochner A, Faggart M, et al. The structure of haplotype blocks in the human genome. Science 2002;296:2225eC2229.
R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2004.
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;70:425eC434.
Excoffier L, Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol 1995;12:921eC927.
Daly MJ, Rioux JD, Schaffner SF, Hudson TJ, Lander ES. High-resolution haplotype structure in the human genome. Nat Genet 2001;29:229eC232.
Reich DE, Cargill M, Bolk S, Ireland J, Sabeti PC, Richter DJ, Lavery T, Kouyoumjian R, Farhadian SF, Ward R, et al. Linkage disequilibrium in the human genome. Nature 2001;411:199eC204.
Shin HD, Park KS, Park CS. Lack of association of GPRA (G protein-coupled receptor for asthma susceptibility) haplotypes with high serum IgE or asthma in a Korean population. J Allergy Clin Immunol 2004;114:1226eC1227.
Barbee RA, Murphy S. The natural history of asthma. J Allergy Clin Immunol 1998;102:S65eCS72.
Jenkins HA, Cherniack R, Szefler SJ, Covar R, Gelfand EW, Spahn JD. A comparison of the clinical characteristics of children and adults with severe asthma. Chest 2003;124:1318eC1324.
Kjellman B, Gustafsson PM. Asthma from childhood to adulthood: asthma severity, allergies, sensitization, living conditions, gender influence and social consequences. Respir Med 2000;94:454eC465.
Payne DN, Rogers AV, Adelroth E, Bandi V, Guntupalli KK, Bush A, Jeffery PK. Early thickening of the reticular basement membrane in children with difficult asthma. Am J Respir Crit Care Med 2003;167:78eC82.
Johansson SG, Hourihane JO, Bousquet J, Bruijnzeel-Koomen C, Dreborg S, Haahtela T, Kowalski ML, Mygind N, Ring J, van Cauwenberge P, et al. A revised nomenclature for allergy: an EAACI position statement from the EAACI nomenclature task force. Allergy 2001;56:813eC824.
Sunyer J, Anto JM, Castellsague J, Soriano JB, Roca J. Total serum IgE is associated with asthma independently of specific IgE levels: the Spanish group of the European study of asthma. Eur Respir J 1996;9:1880eC1884.
Von Ehrenstein OS, Von Mutius E, Illi S, Baumann L, Bohm O, von Kries R. Reduced risk of hay fever and asthma among children of farmers. Clin Exp Allergy 2000;30:187eC193.
Celedon JC, Litonjua AA, Ryan L, Weiss ST, Gold DR. Lack of association between antibiotic use in the first year of life and asthma, allergic rhinitis, or eczema at age 5 years. Am J Respir Crit Care Med 2002;166:72eC75.
Hosking L, Lumsden S, Lewis K, Yeo A, McCarthy L, Bansal A, Riley J, Purvis I, Xu CF. Detection of genotyping errors by Hardy-Weinberg equilibrium testing. Eur J Hum Genet 2004;12:395eC399.
Wacholder S, Rothman N, Caporaso N. Population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias. J Natl Cancer Inst 2000;92:1151eC1158.
Kere J, Laitinen T. Positionally cloned susceptibility genes in allergy and asthma. Curr Opin Immunol 2004;16:689eC694.
Kormann M, Carr D, Klopp N, Illig T, Leupold W, Fritzsch C, Weiland S, Von Mutius E, Kabesch M. G-Protein coupled receptor polymorphisms are associated with asthma in a large German population. Am J Respir Crit Care Med (In press)(Erik Meleen, Sara Bruce, )
Department of Biosciences, Karolinska Institutet
Clinical Allergy Research Unit and Clinical Immunology and Allergy Unit, Department of Medicine, Karolinska Institutet and University Hospital
Department of Occupational and Environmental Health, Stockholm County Council, Stockholm
Clinical Research Centre, Karolinska University Hospital, Huddinge
AstraZeneca R&D Mlndal, Mlndal, Sweden
Institute for Risk Assessment Sciences, University of Utrecht, The Netherlands
University Children's Hospital, Ludwig Maximilian's University Munich, Munich, Germany
GeneOS Limited, Helsinki
Department of Medical Genetics, University of Helsinki, Helsinki, Finland
Division for Immunology, Zurich University Children's Hospital, Zurich, Switzerland
Children's Hospital, Schwarzach, Austria
ABSTRACT
Rationale: Allergic diseases are influenced by both genes and environment. A 70-kb haplotype block in the G proteineCcoupled receptor for asthma susceptibility gene (GPR154; alias GPRA) on chromosome 7p was recently identified to influence susceptibility to asthma and elevated total serum IgE levels in adults. Objectives: To assess the impact of GPR154 on childhood allergic disease, including allergic sensitization, asthma, and rhinoconjunctivitis, in study populations with diverse environmental backgrounds. Methods: We studied farm children, Steiner school children, and two reference groups from five Western European countries in the cross-sectional PARSIFAL (Prevention of Allergy Risk factors for Sensitization In children related to Farming and Anthroposophic Lifestyle) study and a sample of children from the Swedish birth cohort study BAMSE. DNA samples from 3,113 PARSIFAL and 800 BAMSE children were genotyped for 7 GPR154 polymorphisms and haplotypes were inferred. The proportions of alleles and haplotypes (H1eCH7) were compared in affected children with their healthy counterparts. Results: Data indicate a global association of the haplotype block to sensitization (allergen-specific serum IgE 0.35 kU/L, p = 0.022), with significant haplotype-specific associations for H1, H5, and H6. Haplotypes H1 and H5 were also significantly associated with childhood allergic asthma (p = 0.045 and p = 0.023, respectively), and H5 to asthma regardless of sensitization. A broader involvement of GPR154 in allergic diseases was further supported in allergic rhinoconjunctivitis (H3: p = 0.046). The associated haplotypes could be allocated into risk (H5/H6) and nonrisk (H1/H3) groups, a pattern supported by allelic association of single nucleotide polymorphisms (SNPs) rs324384 and rs324396. Conclusions: Our results indicate that polymorphisms and haplotypes in the haplotype block of GPR154 are associated with asthma, rhinoconjunctivitis, and sensitization in European children.
Key Words: asthma children genetic association GPRA gene IgE
Linkage analyses for asthma susceptibility and IgE regulation loci performed some years ago initially held promise in the search for a major gene, but strong candidate genes have been identified only recently (1eC4). The G proteineCcoupled receptor for asthma susceptibility gene (GPR154; alias GPRA), an orphan G proteineCcoupled receptor gene on chromosome 7p, which was identified through positional cloning, is the latest candidate gene suggested to influence the susceptibility to asthma and elevated total IgE levels (4). The genetic evidence was supported by single nucleotide polymorphism (SNP) and haplotype associations, a distinct distribution of protein isoforms between bronchial biopsies from healthy adults and adults with asthma, and increased expression of the GPR154 gene in experimentally induced lung inflammation in mice. The observed 77-kb haplotype block showed a similar risk and nonrisk pattern in all three populations studied. Significant relative risks were reported for elevated total IgE among carriers of particular haplotypes (H4 or H5) and for asthma among homozygous H2 carriers (1.4, 95% confidence interval [CI] 1.1eC1.9 and 2.5, 95% CI 2.0eC3.1, respectively) (4). Given the complex pathophysiology and hereditary patterns of asthma and allergy, these results suggest that the GPR154 gene or other genetic variants in linkage disequilibrium with the identified haplotype block may be associated more broadly with allergic disease, including asthma, allergic sensitization, and rhinoconjunctivitis.
Replication is important as original genetic findings often correlate only moderately well with subsequent research on the same association, with the strongest effect estimate typically occurring in the first study (5, 6). Large, well powered original studies are naturally desirable to avoid spurious associations. Systematic metaanalysis and pooling of data are also powerful tools for assessing effects across several studies and different datasets (7).
Here we present results from GPR154 analyses exploring association in the same haplotype region to childhood asthma, allergic sensitization, and allergic rhinoconjunctivitis using two large and well characterized datasets, the Western European cross-sectional PARSIFAL (Prevention of Allergy Risk factors for Sensitization In children related to Farming and Anthroposophic Lifestyle) study and the Swedish birth cohort study BAMSE. The study population represents a broad spectrum of European children of different ages and lifestyles: farmers, children with an anthroposophic lifestyle, and subsamples of the general population from small communities and large cities.
METHODS
Participants and Questionnaire
The study included 3,113 school children 5 to 13 years old from the cross-sectional PARSIFAL study and 800 children at about 4 years of age from the birth cohort study BAMSE. PARSIFAL children originally included a total of 14,893 from 5 Western European countries. It was designed to investigate the role of different lifestyles and environmental exposures in farm children, Steiner school children (mainly from anthroposophic families), and two corresponding reference groups to identify protective factors for development of asthma and allergic disorders. In Austria, Germany, the Netherlands, and Switzerland farm children were recruited from schools in rural areas known to have a high percentage of farmers; in Sweden through the Farming Registry at the National Bureau of Statistics. Children with an anthroposophic lifestyle were recruited from classes in Steiner schools. The respective reference groups were recruited with similar methods from the same geographical areas, to match for environmental exposures not related to the lifestyles of interest. Parents completed a detailed questionnaire on allergic diseases, infectious history, and environmental exposures, largely based on questions from the International Study of Asthma and Allergies in Childhood (ISAAC) (8), BAMSE (9), the ALEX study (10), and a Swedish study focused on the anthroposophic lifestyle (11). A subsample of children (n = 4,854) and their parents were asked for consent to collect venous blood for allergen specific IgE analyses and DNA extraction. The present study includes 3,113 children (1,579 boys, 1,534 girls) with complete questionnaire data, adequate DNA material, and consent to genetic analyses.
The BAMSE study is based on a well characterized, representative cohort of children followed prospectively from birth, as previously described (9, 12). In brief, 4,089 newborn infants were recruited between 1994 and 1996 and data on parental allergic diseases and residential characteristics were obtained by questionnaire. When the children were 1, 2, and 4 years old, additional mailed questionnaires included questions on symptoms related to wheezing and allergic disease (largely following ISAAC [8]). Blood samples drawn at 4 years of age (mean 4.3) were available from 2,298 children with questionnaire data and informed consent for genetics. In a caseeCcohort sampling design, a sequential random sample of 709 children was selected as a primary subcohort for genetic analyses (13), from whom 686 blood samples provided adequate DNA. Children with reported doctor's diagnosis of asthma up to age four were identified as asthma cases (n = 62), and the remaining children with no asthma reported constituted random controls. From among the 2,298 children not included in the subcohort, a further 114 children with reported asthma were also selected, resulting in a total of 176 asthma cases and 800 samples altogether. The randomly sampled subcohort also allows for unbiased analysis of other outcomes, e.g., sensitization, by comparing affected and nonaffected children in the subcohort.
Ethical approval and informed consent were obtained in each participating study center.
Definition of Outcome Phenotypes
Allergen-specific serum IgE was measured against the Phadiatop mix of eight common inhalant allergens and the fx5 mix of six common food allergens (Pharmacia CAP System; Pharmacia Diagnostics AB, Uppsala, Sweden). Asthma was defined as a physician's diagnosis of asthma according to the questionnaire response (see online supplement for questionnaire details). For the 4-year-old BAMSE children, the phenotype "current asthma," defined as physician-diagnosed asthma and one or more episodes of wheezing in the last 12 months at 4 years, was also used. Allergic sensitization was defined as an IgE level greater than or equal to 0.35 kU/L by either Phadiatop or fx5. Allergic asthma (or atopic asthma) was derived from the combination of physician's diagnosis of asthma and sensitization. Allergic rhinoconjunctivitis was defined as a questionnaire response of having had rhinitis and conjunctivitis symptoms without concurrent cold within the last 12 months, in combination with sensitization to inhalant allergens (Phadiatop) in PARSIFAL only. The young age of the participants and the questionnaire data available in BAMSE did not allow for a comparable diagnosis.
SNP Genotyping and Data Checking
DNA was extracted from peripheral blood leukocytes using standard techniques. Seven of the SNPs investigated in the Laitinen and coworkers study were selected to tag the previously observed haplotypes (Figure 1A) (4). Primers were designed by the SpectroDesigner software (Sequenom Inc., San Diego, CA) and genotyping was performed using MALDI-TOF technology (see online supplement for details). All samples were analyzed in a sex-specific assay (AMELX/Y; OMIM 410000 / 300391) to evaluate the accuracy in sample and database management.
Data Analysis
Population Hardy-Weinberg equilibrium was evaluated for each SNP in both the PARSIFAL (combined reference groups) and BAMSE subcohort samples, with a chi square test for the distribution of genotypes. Allele frequencies were analyzed for cases versus control subjects in Haploview 2.05 software (http://www.broad.mit.edu/personal/jcbarret/haploview) using a chi square test. Mantel-Haenszel combined odds ratios (ORs) and 95% CIs for comparison of allele frequency in cases versus control subjects were calculated using standard procedures. Linkage disequilibria between the genotyped SNPs were obtained in Haploview, which estimates the recombination history between SNPs by evaluating the normalized measure of allelic association, D'. Haplotype block generation was performed using the default algorithm defined by Gabriel and coworkers (14). Two separate approaches were used for estimation of haplotype association with the defined outcomes in the PARSIFAL and BAMSE study groups and the combined material. First, haplotype association was tested in the statistical software R (15) using the haplo.score algorithm, which estimates probabilities for each possible haplotype for each individual based on the full study populations (cases and control subjects combined) and then generates both global and haplotype-specific score statistics by comparing estimated haplotype frequencies in cases and control subjects (16). Adjustment was made for a variable with one category for each combination of country-of-origin and sampling group, and 10,000 simulations were performed to obtain empirical p values for the observed data. All p values are nominal two-sided values. Second, for estimation of ORs and CIs attributed to each haplotype, the frequency of haplotypes was again estimated in the full study populations (cases and control subjects combined) using the expectation maximization algorithm (17), implemented in the Haploblocks program (M. Zucchelli and J. Kere, unpublished data). Each individual contributed two haplotypes (one from each chromosome) to the analyses and the frequency and estimated counts of each haplotype were then assessed against all others using standard procedures for OR calculations.
Phylogenetic Analyses
The genomic DNA of seven Finnish individuals who were each homozygous for one of the haplotypes H1eCH7 was sequenced across the 133 kb region (4). Repeat regions such as SINE, LINE, LTR, MER1, and MER2 elements covered 60% of the sequence. Polymerase chain reaction assays were designed to overlap. Repeat regions were resequenced when it was possible to design primers in unique sequence. We identified 152 polymorphic sites (SNPs and deletion/insertion polymorphisms (DIPs) and created the phylogenetic analysis of the seven haplotypes spanning the locus using PHYLIP (the PHYLogeny Inference Package, http://www.hgmp.mrc.ac.uk/). All 152 recognized polymorphisms were included in this analysis, which thus substantially refines the analysis of only 40 SNPs in the region from the seven individuals presented earlier (4).
RESULTS
Baseline characteristics showed that the two study populations were similar in terms of sex distribution and prevalence of asthma and allergic sensitization, whereas mean age at study differed by design (Table 1). The overall success rates for each studied SNP ranged from 93.2 (rs324377) to 98.1% (rs324396). Results from the AMELXY assay indicated 3.1% sex discrepancies. None of the assayed SNPs deviated significantly from population Hardy-Weinberg equilibrium in the PARSIFAL reference groups (p > 0.01), but Hopo546333 in BAMSE was slightly off Hardy-Weinberg equilibrium (p = 0.001 for controls with no sensitization). Further investigation revealed that 9 homozygous AA were observed instead of the expected 3. Because this difference between observed and expected genotypes was minor and only affected the smaller BAMSE study group, the SNP was also included in the analysis.
Pairwise comparisons between the genotyped SNPs indicated no historical recombination and the SNPs were thus defined to be contained within one preserved 70-kb haplotype block for all groups (Figures 1A and 1B and Table 2). Haplotype frequencies in the different countries are presented in Table E4 (see the online supplement). Extended phylogenetic analyses in Finnish subjects, using 152 instead of 40 sequence changes, revealed a closer relationship between haplotypes H6 and H7 than that previously presented (Figure 1C) (4).
We initially tested the hypothesis that GPR154 haplotypes are associated with childhood sensitization and asthma, based on previously reported findings in adults (4). The analysis for the combined studies revealed a global association of the haplotype block to sensitization (p = 0.022) as a result of haplotype-specific associations for H1 (p = 0.038), H5 (p = 0.019), and H6 (p = 0.015), whereas H1 and H5 were similarly, but not significantly, associated with asthma in the whole sample (Table 3). In the PARSIFAL population the H5 haplotype was significantly associated with asthma (p = 0.033) (Table 3). Similarly, with "current asthma" as the outcome for the BAMSE children, in an attempt to reduce misclassification by excluding early childhood transient wheezers from the asthma group, H5 was significantly associated with asthma in the combined sample (p = 0.026, data not shown).
To further characterize risk of disease attributed to each haplotype in the GPR154 gene, we performed additional analyses for estimation of haplotype specific ORs (Table 3). Separate analyses of the PARSIFAL and BAMSE samples showed reasonably concordant results, notably the ORs for H1 and asthma or H5 and sensitization, and the overall pattern of haplotype effects.
These findings encouraged us to test whether GPR154 showed effects for other allergic disease phenotypes by investigating association with allergic asthma and allergic rhinoconjunctivitis. Haplotypes H1 and H5 showed stronger associations with allergic asthma for the whole group (p = 0.045, OR = 0.78, 95% CI, 0.63eC0.96 and p = 0.023, OR = 1.47, 95% CI, 1.01eC2.14, respectively) and in the PARSIFAL population H3 was found to be associated with allergic rhinoconjunctivitis (p = 0.046, OR = 0.68, 95% CI, 0.48eC0.96, data not shown).
Although the seven SNPs included in this study were specifically selected for tagging the previously identified haplotypes (4), we also examined possible allelic associations for each SNP and the outcomes mentioned above. The strongest associations were observed for rs324384 (minor allele C) with sensitization (OR = 0.88, 95% CI, 0.80eC0.98) and for rs324396 (minor allele T) with sensitization (OR = 0.88, 95% CI, 0.79eC0.99) and asthma (OR = 0.83, 95% CI, 0.70eC0.97) (see Table E2 in online supplement). Both SNPs were also significantly associated with allergic asthma (OR = 0.81, 95% CI, 0.66eC1.00 and OR = 0.78, 95% CI, 0.62eC0.98, respectively). The strongest allelic association with allergic rhinoconjunctivitis was OR = 0.83, 95% CI, 0.66eC1.04 for rs324384 (C).
DISCUSSION
Large well characterized epidemiologic studies are key to verifying results from complex disease genetic mapping and to assessing the relevance of findings on the population level. In this study we investigated a haplotype region in the novel G proteineCcoupled receptor, GPR154, in two separate datasets of European children and demonstrated an association between its genetic variants and allergic diseases and asthma in childhood. Detailed analyses enabled the effective dissection of haplotype effects of GPR154; both revealing new and confirming previous relationships to asthma and IgE-mediated allergic disorders.
Recently, performing association studies on haplotypes rather than on individual SNPs has gained popularity with the appreciation that a limited number of haplotypes in regions of high linkage disequilibrium (haplotype blocks) capture most of human genetic heterogeneity in that region. Such haplotypes can be tagged by a small number of SNPs and may increase the power of genetic association studies (14, 18, 19). Evaluating haplotype association was central in the present study, because this was the core of the findings in the Laitinen and coworkers study (4) from which we selected SNPs tagging the observed risk and nonrisk haplotypes.
In all, the expectation maximization algorithm used predicted only seven haplotypes with frequencies over 3% from the 7,826 chromosomes investigated. These haplotypes were identical to those previously reported (4). Our observations emphasize the high linkage disequilibrium existing between the investigated SNPs and the remarkable preservation of the GPR154 haplotype block in the outbred populations of the study. The similar haplotype distribution in each country also supports the finding of a well preserved genetic region across several populations.
We observed association between allergic disease and a distinct set of risk (H5/H6) and nonrisk (H1/H3) haplotypes, which was supported by allelic association of the SNPs rs324384 and rs324396, differentiating the nonrisk haplotypes. Whereas haplotype H6 conferred risk of sensitization in this study, no association to high serum IgE was reported in the original study (4). Of note, the H6 haplotype occurred more frequently in the PARSIFAL population (10.8%, Table 2) than in the Finnish and Canadian populations (5.1 and 8.7%, respectively). Thus, our observation could result from a higher power to detect effects of H6. The notion of H6 as a risk haplotype is supported by the refined phylogenetic analyses. The SNP 522363[G/C] (rs 323922) was initially proposed as a tagging SNP for risk (C) and nonrisk haplotypes (G) (4), but these new data show a somewhat different pattern. Further, we did not find any associations between the SNP in question and the studied outcomes, and a recently published Korean study also showed lack of association for this SNP to asthma and high serum IgE (20).
In our data, we were not able to detect any associations between the outcomes of interest and the H2 and H4 haplotypes, which were suggested as risk haplotypes for asthma and elevated total IgE, respectively (4). This could clearly reflect true differences between the genetic effects of GPR154 variants or other genetic variants in linkage disequilibrium with the different haplotypes that may vary in the different populations of the studies, but some aspects concerning study design also deserve attention. Laitinen and coworkers reported haplotype associations with asthma and high serum IgE-levels in mostly adults, whereas we attempted to replicate this finding in children 4 to 13 years old. Thus, some phenotypic heterogeneity between the studies is likely. Longitudinal studies have shown that approximately 50% of children with asthma will outgrow their symptoms after a number of years (21), whereas some individuals develop asthma only in adulthood. This natural development makes the allocation of children into asthma or nonasthma groups both age dependent and complicated in terms of previous asthma history. Evidence also suggests that childhood and adult asthma differ in a number of parameters, including sex preponderance, airflow obstruction, sensitization and structural changes of the airways, although remodelling may occur already in children with severe asthma (22eC24). In our study we defined sensitization with IgE antibody levels against food allergens or aeroallergens, which are correlated with total IgE to some extent, but total IgE has also been associated with asthma independently of specific IgE levels (25, 26). Given the possible phenotypic heterogeneity between the studies, the replication of associations between asthma/sensitization and GPR154 variants in children suggests a general role for this gene in respiratory allergic disease, and the present results do not exclude that H2 and H4 may be associated with increased risk in well established, adult asthma and elevated serum IgE phenotypes.
The PARSIFAL and BAMSE samples each have unique study design features. The PARSIFAL samples represent a varied set of Western European children selected on the basis of farming and anthroposophic lifestyles, and their respective reference groups from rural and sub-urban/urban communities. The four groups were exposed to differential environments known to influence the development of IgE-mediated allergic disorders (11, 27), and these aspects are being investigated in separate analyses. We intentionally focused on the effect across the total sample, as the purpose was to replicate and explore the original GPR154 associations in terms of a significant overall genetic association with asthma and allergic phenotypes rather than addressing specific geneeCenvironment interaction hypotheses. Such analyses are planned when the environmental aspects have been explored. Considering that the PARSIFAL children represent a heterogeneous population from five European countries, the significant association estimated for the four groups reinforce the notion of an association between GPR154 variants and allergic disease in general. With the younger BAMSE cohort, which differs from the PARSIFAL study in terms of population and study design, we aimed to extend the analyses to increase representativity and power, and to detect general associations across different populations and age groups. The results from the BAMSE samples separately did not reach statistical significance for several potential reasons. First, the children in BAMSE were around 4 years of age, and although we used questionnaire reported doctor's diagnosis of asthma as common outcome, this may to a greater extent represent a phenotype of transient wheezing, often triggered by infections, that does not share the characteristics of asthma later in childhood or adult asthma. Restriction of the analyses to children with current asthma at the age of 4 years, which is equivalent to a definition used in, for example, the Home Allergens and Asthma Study (28), also showed that the associations became stronger. Second, the BAMSE study is smaller, resulting in more limited statistical power. Nevertheless, the point estimates for association largely agree with those from the PARSIFAL study, and the combined analyses show that the associations are valid with this dataset included.
The quality assessments performed and the high success rate imply that the genotyping data we obtained are robust and reliable. For the SNP that deviated from Hardy-Weinberg equilibrium in the BAMSE samples, homozygous individuals were only minimally overrepresented compared with the expected number, and this difference should have only marginal, if any, influence on the results. Nevertheless, apart from the possibility of true deviation in the population, possible reasons for such Hardy-Weinberg equilibrium deviation include nonspecific assays (although primers were controlled), genotyping errors, DNA quality, unknown issues, or simply deviation from equilibrium by chance (29).
When we compared those selected for genotyping with the remaining subjects for BAMSE (13) and PARSIFAL, the genotyped subjects appeared to be essentially representative, with no important differences noted. Furthermore, selection bias will only occur if selection factors differ by genotype, which is mostly unlikely.
Although population stratification may often be relatively unlikely to cause bias in real-life settings (30), it is difficult to completely rule out in creating false positive results. In this study, we note that haplotype frequencies were quite similar in the five countries. In addition, the haplotype association p values estimated in haplo.score were adjusted for country of origin and study group, which should rule out any major residual effect of population stratification.
In summary, we report that GPR154 variants are significantly associated with allergic sensitization, asthma, and allergic rhinoconjunctivitis in children from 4 to 13 years of age from two Western European datasets, with the H5 and to some extent H6 haplotypes conferring risk and the H1 and H3 haplotypes exerting protective effects. We observed the previously reported haplotype block and were able to identify the described haplotypes, supporting the original finding (4). We also report refined phylogenetic analyses that support subdivision of the observed haplotypes into risk and nonrisk variants. Further research is needed to clarify the effect of the observed haplotypes on, for example, GPR154 protein levels and the immune response. Some functional data on GPR154 supporting the reported associations were given in the original report and are also available elsewhere (4, 31).
Our study and a parallel, independent study by Kormann and coworkers (32) together provide robust replication for GPR154 haplotypes as risk determinants for asthma and related disorders, although the details differ. In addition, our results and those of Kormann and coworkers (32) also extend the original findings of Laitinen and coworkers identified by positional cloning in an inbred population (4) to children with asthma in a wide spectrum of Western European outbred populations, and define not only risk and neutral haplotypes, but also protective haplotype variants.
Acknowledgments
The authors thank all children and parents in the PARSIFAL and BAMSE studies, as well as fieldworkers in all centers. They also thank Dr. Marco Zucchelli of the Department of Biosciences at Karolinska Institutet for statistical assistance.
E.M. and S.B. contributed equally to the study.
The PARSIFAL Genetics Study Group: Gran Pershagen, Fredrik Nyberg, Annika Scheynius, Marianne van Hage-Hamsten, Magnus Wickman (Sweden); Charlotte Braun-Fahrlnder, Marco Waser, Roger Lauener (Switzerland); Bert Brunekreef, Gert Doekes, Dieneke Schram (the Netherlands); Erika von Mutius, Marcus Ege (Germany); Josef Riedler, Waltraud Eder, Ellen elagger (Austria), Michael Kabesch (Munich lab), Juha Kere, Cecilia Lindgren, Erik Meleen (Stockholm lab), Fernando Martinez (Tucson lab).
This article has an online supplement, which is accessible from this issue's table of content online at www.atsjournals.org
REFERENCES
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, et al. Association of the ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;418:426eC430.
Allen M, Heinzmann A, Noguchi E, Abecasis G, Broxholme J, Ponting CP, Bhattacharyya S, Tinsley J, Zhang Y, Holt R, et al. Positional cloning of a novel gene influencing asthma from chromosome 2q14. Nat Genet 2003;35:258eC263.
Zhang Y, Leaves NI, Anderson GG, Ponting CP, Broxholme J, Holt R, Edser P, Bhattacharyya S, Dunham A, Adcock IM, et al. Positional cloning of a quantitative trait locus on chromosome 13q14 that influences immunoglobulin E levels and asthma. Nat Genet 2003;34:181eC186.
Laitinen T, Polvi A, Rydman P, Vendelin J, Pulkkinen V, Salmikangas P, Makela S, Rehn M, Pirskanen A, Rautanen A, et al. Characterization of a common susceptibility locus for asthma-related traits. Science 2004;304:300eC304.
Weiss ST, Raby BA. Asthma genetics 2003. Hum Mol Genet 2004;13 Spec No 1:R83eC89. Epub 2004 Feb 05. Review.
Ioannidis JP, Ntzani EE, Trikalinos TA, Contopoulos-Ioannidis DG. Replication validity of genetic association studies. Nat Genet 2001;29:306eC309.
Altshuler D, Hirschhorn JN, Klannemark M, Lindgren CM, Vohl MC, Nemesh J, Lane CR, Schaffner SF, Bolk S, Brewer C, et al. The common PPARgamma Pro12Ala polymorphism is associated with decreased risk of type 2 diabetes. Nat Genet 2000;26:76eC80.
Asher MI, Keil U, Anderson HR, Beasley R, Crane J, Martinez F, Mitchell EA, Pearce N, Sibbald B, Stewart AW, et al. International study of asthma and allergies in childhood (ISAAC): rationale and methods. Eur Respir J 1995;8:483eC491.
Melen E, Kere J, Pershagen G, Svartengren M, Wickman M. Influence of male sex and parental allergic disease on childhood wheezing: role of interactions. Clin Exp Allergy 2004;34:839eC844.
Riedler J, Braun-Fahrlander C, Eder W, Schreuer M, Waser M, Maisch S, Carr D, Schierl R, Nowak D, von Mutius E. Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey. Lancet 2001;358:1129eC1133.
Alm JS, Swartz J, Lilja G, Scheynius A, Pershagen G. Atopy in children of families with an anthroposophic lifestyle. Lancet 1999;353:1485eC1488.
Lannero E, Kull I, Wickman M, Pershagen G, Nordvall SL. Environmental risk factors for allergy and socioeconomic status in a birth cohort (BAMSE). Pediatr Allergy Immunol 2002;13:182eC187.
Melen E, Gullsteen H, Zucchelli M, Lindstedt A, Nyberg F, Wickman M, Pershagen G, Kere J. Sex-specific protective effects of interleukin-9 receptor haplotypes in childhood wheezing and sensitisation. J Med Genet 2004;41:e123.
Gabriel SB, Schaffner SF, Nguyen H, Moore JM, Roy J, Blumenstiel B, Higgins J, DeFelice M, Lochner A, Faggart M, et al. The structure of haplotype blocks in the human genome. Science 2002;296:2225eC2229.
R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2004.
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;70:425eC434.
Excoffier L, Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol 1995;12:921eC927.
Daly MJ, Rioux JD, Schaffner SF, Hudson TJ, Lander ES. High-resolution haplotype structure in the human genome. Nat Genet 2001;29:229eC232.
Reich DE, Cargill M, Bolk S, Ireland J, Sabeti PC, Richter DJ, Lavery T, Kouyoumjian R, Farhadian SF, Ward R, et al. Linkage disequilibrium in the human genome. Nature 2001;411:199eC204.
Shin HD, Park KS, Park CS. Lack of association of GPRA (G protein-coupled receptor for asthma susceptibility) haplotypes with high serum IgE or asthma in a Korean population. J Allergy Clin Immunol 2004;114:1226eC1227.
Barbee RA, Murphy S. The natural history of asthma. J Allergy Clin Immunol 1998;102:S65eCS72.
Jenkins HA, Cherniack R, Szefler SJ, Covar R, Gelfand EW, Spahn JD. A comparison of the clinical characteristics of children and adults with severe asthma. Chest 2003;124:1318eC1324.
Kjellman B, Gustafsson PM. Asthma from childhood to adulthood: asthma severity, allergies, sensitization, living conditions, gender influence and social consequences. Respir Med 2000;94:454eC465.
Payne DN, Rogers AV, Adelroth E, Bandi V, Guntupalli KK, Bush A, Jeffery PK. Early thickening of the reticular basement membrane in children with difficult asthma. Am J Respir Crit Care Med 2003;167:78eC82.
Johansson SG, Hourihane JO, Bousquet J, Bruijnzeel-Koomen C, Dreborg S, Haahtela T, Kowalski ML, Mygind N, Ring J, van Cauwenberge P, et al. A revised nomenclature for allergy: an EAACI position statement from the EAACI nomenclature task force. Allergy 2001;56:813eC824.
Sunyer J, Anto JM, Castellsague J, Soriano JB, Roca J. Total serum IgE is associated with asthma independently of specific IgE levels: the Spanish group of the European study of asthma. Eur Respir J 1996;9:1880eC1884.
Von Ehrenstein OS, Von Mutius E, Illi S, Baumann L, Bohm O, von Kries R. Reduced risk of hay fever and asthma among children of farmers. Clin Exp Allergy 2000;30:187eC193.
Celedon JC, Litonjua AA, Ryan L, Weiss ST, Gold DR. Lack of association between antibiotic use in the first year of life and asthma, allergic rhinitis, or eczema at age 5 years. Am J Respir Crit Care Med 2002;166:72eC75.
Hosking L, Lumsden S, Lewis K, Yeo A, McCarthy L, Bansal A, Riley J, Purvis I, Xu CF. Detection of genotyping errors by Hardy-Weinberg equilibrium testing. Eur J Hum Genet 2004;12:395eC399.
Wacholder S, Rothman N, Caporaso N. Population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias. J Natl Cancer Inst 2000;92:1151eC1158.
Kere J, Laitinen T. Positionally cloned susceptibility genes in allergy and asthma. Curr Opin Immunol 2004;16:689eC694.
Kormann M, Carr D, Klopp N, Illig T, Leupold W, Fritzsch C, Weiland S, Von Mutius E, Kabesch M. G-Protein coupled receptor polymorphisms are associated with asthma in a large German population. Am J Respir Crit Care Med (In press)(Erik Meleen, Sara Bruce, )