Genetic variation in immune signaling genes differentially expressed in asthmatic lung tissues

Article Outline

• Abstract
• Methods
  • Association study samples
  • Phenotype definitions
  • Genotyping
  • Statistical analyses
• Results
• Discussion
• Acknowledgment
• Methods
  • The SLSJ familial sample
  • The CAPPS
  • The SAGE
  • The Busselton Health Study population (Busselton cohort)
  • Pairwise LD pattern
• References
• Fig E1. 
• Fig E2. 
• Fig E3. 
• Fig E4. 
• Fig E5. 
• Fig E6. 
• Fig E7. 
• Fig E8. 
• Table E1. 
• Table E2. 
• Table E3. 
• Table E4. 
• References
• Copyright

Background

Eight genes in the immune signaling pathway shown to be differentially expressed in asthmatic lung biopsy specimens in a previous microarray experiment were selected as candidate genes for asthma susceptibility.

Objective

We sought to perform an association study with these genes and asthma-related phenotypes in 3 independent Canadian familial asthma collections and 1 Australian asthma case-control study.

Methods

Tagging single nucleotide polymorphisms were selected by using the HapMap public database (r² > 0.8; minor allele frequency >0.10) and genotyped with the Illumina platform. Family-based association and trend tests for asthma, atopy, airway hyperresponsiveness, and allergic asthma phenotypes were done in each sample, correcting for multiple testing.

Results

Uncorrected associations with polymorphisms within 7 genes were detected with 1 or more of the phenotypes in 1 or more of the 4 populations (.001 < P < .05). After correction, the 15-lipoxygenase (15-LO) associations with airway hyperresponsiveness and allergic asthma remained significant in 2 Canadian samples (corrected P = .022 and .049, respectively), and the association of the CD14 antigen with asthma remained significant in 1 Canadian sample (corrected P = .042). In both cases a protective effect of the minor alleles was observed.

Conclusion

Expression profiling studies are a useful way to identify candidate genes for asthma because this approach has led to the first report of an association with 15-LO in 2 independent populations. Because 15-LO is involved in anti-inflammatory processes, further functional and clinical investigation of the role of this biologic pathway in asthma is warranted.

Key words: Asthma, microarrays expression profiling, candidate genes, genetic association study, tagging single nucleotide polymorphism, 15-lipoxygenase

Abbreviations used: AHR, Airway hyperresponsiveness, ALOX15, Arachidonate 15-lipoxygenase gene, CAPPS, Canadian Asthma Primary Prevention Study, CXCL12, Chemokine (C-X-C motif) ligand 12 (stromal cell–derived factor 1) gene, FBAT, Family-based Association Test, IL2RB, IL-2 receptor β gene, IL7R, IL-7 receptor gene, LD, Linkage disequilibrium, 15-LO, 15-Lipoxygenase, NOS2A, Nitric oxide synthase 2A gene, SAGE, Study of Asthma Genetics and Environment Cohort, SFRP1, Secreted frizzled-related protein 1 gene, 15(S)-HETE, 15(S)-hydroxyeicosatetraenoic acid, SLSJ, Saguenay–Lac-Saint-Jean, SNP, Single nucleotide polymorphism

Asthma is a chronic inflammatory disease of the airways associated with variable airflow obstruction and airway hyperresponsiveness (AHR).¹ Asthma is also recognized as a complex heritable disease, implying the interaction of numerous genetic and environmental factors.² Identification of the genes that influence asthma susceptibility and severity has been the subject of intense research over the last decade.³, ⁴ This knowledge has significantly contributed to our understanding of the pathophysiology of asthma, but much work is needed to replicate reported associations, to characterize the functional effects of the genes already identified,⁵ and to identify new candidate genes for asthma.

Expression profiling studies in which transcript levels in tissues of affected and control subjects are compared allow the identification of genes related to a disease process.⁶ These genes, which might or might not have known function, can be selected as candidate genes in subsequent genetic association studies on the basis of their differential expression in affected tissues.⁵ We used this approach in the selection of the CX3CR1 gene, the fractakline receptor, for a previous asthma association study⁷ because CX3CR1 was underexpressed in bronchial biopsy specimens of asthmatic subjects compared with control subjects.⁸

CX3CR1 is related to immune signaling, and 8 other genes that we reported as differentially expressed in asthmatic bronchial biopsy specimens are also part of this biologic cluster.⁸ Three of these have been previously suggested as asthma genes (CD14, chemokine [C-X-C motif] ligand 12 [CXCL12], and nitric oxide synthase 2A [NOS2A]),⁴ but 5 have not been recognized as asthma genes or clearly implicated in asthma pathophysiology (arachidonate 15-lipoxygenase [ALOX15], CD27, IL-2 receptor β [IL2RB], IL-7 receptor [IL7R], and secreted frizzled-related protein 1 [SFRP1]; see Table I for a description of the genes). Because the approach used for CX3CR1 allowed us to find a positive association that has been replicated in an independent cohort,⁷ we have applied a similar strategy to these 8 additional differentially expressed genes. We postulated that these genes represent asthma biomarkers, and if genetic variation explains the difference in gene expression, then we should observe an association between variants of these candidate genes and asthma and related phenotypes in a collection of asthmatic subjects. Specifically, we hypothesized that, based on their differential expression in asthmatic bronchial biopsy specimens⁸ and their biologic role in immunity, ALOX15, CD14, CD27, CXCL12, IL2RB, IL7R, NOS2A, and SFRP1 represent candidate genes for asthma susceptibility.

Table I. Gene and TagSNP characteristics

UTR, Untranslated region; NF, nuclear factor; MAPK, mitogen-activated protein kinase; JNK, c-jun N-terminal kinase; NO, nitric oxide.

The objectives of the present study are to report tests for association between asthma-related phenotypes and the 8 candidate genes in 4 independent asthmatic collections (differentiated by their recruitment, ascertainment of the subjects, and their genetic background) included in the Canadian AllerGen network (AllerGen NCE, Inc). Three of these are Canadian family studies, and the fourth is an Australian asthma case-control study. Our results demonstrated the usefulness of the expression profiling studies in the identification of asthma candidate genes because they allow us to report the first asthma genetic association for ALOX15 in 2 independent Canadian population samples.

Back to Article Outline

Methods 

Association study samples 

In the context of the AllerGen (Allergy, Genes, and Environment) NCE, Inc, Canadian network, 4 well-characterized samples of individuals affected with asthma and their family members, as well as case-control data, were combined as a resource to investigate the genetics of asthma and related phenotypes (see the complete description in this article's Methods section in the Online Repository at www.jacionline.org). Briefly, these include 3 Canadian independent studies: (1) the Saguenay–Lac-Saint-Jean (SLSJ) founder population family collection⁹, ¹⁰, ¹¹, ¹², ¹³, ¹⁴; (2) the Canadian Asthma Primary Prevention Study (CAPPS) high-risk birth cohort¹⁵, ¹⁶, ¹⁷; and (3) the Study of Asthma Genes and the Environment Cohort (SAGE), a nested case-control sample from a population-based birth cohort. These 3 samples have been fully described in a recent article by Bégin et al.¹⁸ The fourth asthma sample is the Australian Busselton Health Study, which is a nested case-control sample selected from a population-based cohort that has a longitudinal design and a standardized phenotype classification.¹⁹, ²⁰, ²¹, ²², ²³ Informed consent was approved by each center's research ethics board and collected for each study participant.

Phenotype definitions 

Four dichotomous asthma related-phenotypes were analyzed. First was asthma, which was defined as doctor-diagnosed asthma in SLSJ, CAPPS, and SAGE and as a positive response to the question “Has a doctor every told you that you had asthma, or bronchial asthma?” at either survey (1981 and 1994) in the Busselton cohort. Second was atopy, which was defined as at least 1 positive response (wheal diameter ≥3 mm at 10 minutes) on skin prick testing for any allergen tested. Third was AHR, which was assessed by means of methacholine challenge testing. A positive response was defined as a 20% decrease in FEV₁ (PC₂₀ or PD₂₀). For the SLSJ, CAPPs, and SAGE samples, AHR was defined as a PC₂₀ value of less than 8 mg/mL and for the Busselton cohort, it was defined as a PD₂₀ value of 3.9 μmol or less. The fourth phenotype was allergic asthma, which was defined as being present in subjects given diagnoses of both asthma and atopy.

Genotyping 

DNA samples were available for each studied sample. AllerGen TagSNP genotyping was performed with the Illumina BeadLab (http://www.illumina.com; Illumina, Inc, San Diego, Calif) platform, as described previously.²⁴ TagSNPs were selected from the International HapMap Project (http://www.hapmap.org) release 16c from among single nucleotide polymorphisms (SNPs) located in the vicinity of the candidate genes (if they were not located in neighboring genes) and included the flanking regions 10 kb upstream and 10 kb downstream of the first and the last exon, respectively. TagSNPs were selected in a manner similar to LD Select²⁵ by using an r² cutoff of 0.8. If equivalent TagSNPs were identified, those with the highest Illumina design score were kept. Only SNPs with a design score of greater than 0.4 were retained for the panel. Samples with call rates of less than 95% were excluded from the final dataset. Assays with poor cluster separation and with call rates of less than 90% were removed from the final dataset. All primers and conditions used are available on request to the corresponding author. For the 8 studied candidate genes, a panel of 77 TagSNPs was tested for association with asthma, atopy, AHR, and allergic asthma phenotypes in each sample. SNPs described in this report are cited by using their National Center for Biotechnology Information reference sequence numbers.

Statistical analyses 

Allele distribution patterns and Mendelian errors for the 3 familial samples (SLSJ, CAPPS, and SAGE) were assessed by using Family-based Association Test (FBAT) software (version 1.7.2; Harvard University, Boston, Mass)²⁶, ²⁷, ²⁸ under an additive genetic model. An empiric estimate of the variance has been used for the SLSJ sample.²⁶ FBAT results are reported for the minor alleles. For the Busselton cohort, association testing was performed with a general allelic likelihood ratio test, χ² test (1 df), as implemented in the UNPHASED program (Cambridge, United Kingdom [UK]).²⁹ This approach uses a standard retrospective case-control likelihood for case-control designs. Odds ratios were estimated by using the most common allele as the referent and are reported for each minor allele. For each gene and for each SNP within that gene, we applied a correction to the significance level that accounts for both the effective number of independent SNPs tested in that gene and the effective number of phenotypes tested. A Sidak correction was applied with respect to that number (independent SNPs × independent phenotypes). Because the TagSNPs are correlated, the effective number of independent SNPs was calculated by using the method proposed by Li and Ji³⁰ implemented in the Single Nucleotide Polymorphism Spectral Decomposition Web resource (http://gump.qimr.edu.au/general/daleN/SNPSpD/).³¹ We used a similar procedure, the Matrix Spectral Decomposition approach, to estimate the number of independent phenotypes (n = 3) in each sample. Hardy-Weinberg equilibrium P values were calculated using Haploview software (version 3.32; Broad Institute of MIT and Harvard University, Boston, Mass)³² for the SLSJ, CAPPS, and SAGE samples and by using a χ² test for the Busselton cohort, considering 1 df and a significance threshold of .001. The strength of linkage disequilibrium (LD) between pairs of SNPs was evaluated with Haploview software (version 3.32).³² LD plots for each gene in each population are presented in Fig E1, Fig E2, Fig E3, Fig E4, Fig E5, Fig E6, Fig E7, Fig E8 (available in this article's Online Repository at www.jacionline.org).

Back to Article Outline

Results 

For the SLSJ sample, the probands had a male/female ratio of 1:1.2, a mean age of 18 years, and a mean PC₂₀ of 2.66 mg/mL. For the CAPPs sample, the children were 7 years old, the sex ratio was 1.05:1, and they had a mean PC₂₀ of 2.99 mg/mL. For the SAGE sample, the children were aged between 8 and 10 years, had a sex ratio of 1.26:1, and had a mean PC₂₀ of 2.72 mg/mL. Finally, for the Busselton cohort, the age range was from 17 to 91 years, the sex ratio was 1:1.35, and the mean PD₂₀ was 1.14 μmol. The number of individuals genotyped and the phenotypic distribution in each studied sample are presented in Table II. Genotyping was achieved with a mean success rate of 99.9% for the SLSJ, 99.6% for the CAPPS, 99.7% for the SAGE, and 99.7% for the Busselton samples. A total of 17 Mendelian errors were observed. Families with Mendelian errors were reset to zero for FBAT analyses. All SNPs were in Hardy-Weinberg equilibrium, and none of them showed more than 2 Mendelian errors.

Table II. Distribution of asthma, atopy, AHR, and allergic asthma phenotypes among the 4 studied samples

The family-based association analyses for the SLSJ, CAPPS, and SAGE samples are presented in Table III, Table IV, Table V, respectively. Odds ratios for the Busselton cohort are presented in Table VI. For the sake of brevity, only results showing a P value less than the .05 threshold for 1 or more phenotypes are presented, with their respective corrected P values indicated in parentheses. The significance values for each population for all TagSNPs that generated significant associations are available in Table E1, Table E2, Table E3, Table E4 (available in this article's Online Repository at www.jacionline.org). After correction for multiple testing, only 4 associations remained significant. In the CAPPS sample, ALOX15 rs7217186 showed a positive association for the AHR phenotype, suggesting a protective effect of the minor allele because it is undertransmitted to the hyperresponsive probands (corrected P = .022). In the SAGE sample, ALOX15 showed a positive association for rs2664593 and allergic asthma. The undertransmission of the minor allele to the allergic asthmatic probands suggests a protective effect (corrected P = .049). In this sample, a protective association was also found for the CD14 gene (corrected P = .042). For all other TagSNPs tested, no association remained significant after correcting for multiple testing, regardless of the studied phenotype in a given sample.

Table III. Significant FBAT results for TagSNPs of the studied candidate genes and asthma, atopy, AHR, and allergic asthma phenotypes under an additive genetic model in the SLSJ familial sample

MAF, Minor allele frequency; N, number of families contributing the statistic; Z, Z score.

Table IV. Significant FBAT results for TagSNPs of the studied candidate genes and asthma, atopy, AHR, and allergic asthma phenotypes under an additive genetic model in the CAPPS familial sample

P values highlighted in bold and underlined indicate that they remain significant after correction for multiple testing.

MAF, Minor allele frequency; N, number of families contributing the statistic; Z, Z score.

Table V. Significant FBAT results for TagSNPs of the studied candidate genes and asthma, atopy, AHR, and allergic asthma phenotypes under an additive genetic model in the SAGE familial sample

P values highlighted in bold and underlined indicate that they remain significant after correction for multiple testing.

MAF, Minor allele frequency; N, number of families contributing the statistic; Z, Z score.

Table VI. Significant odds ratios for TagSNPs of the studied candidate genes and asthma, atopy, AHR, and allergic asthma phenotypes in the Busselton case-control cohort

MAF, Minor allele frequency; OR, 95% CI odds ratio.

Back to Article Outline

Discussion 

This study reports association analyses of 8 candidate genes selected based on our previous microarray results⁷ in 4 independent asthma samples. Before correction for multiple testing, this strategy identified 7 associations with asthma or related phenotypes in 1 or more of the samples studied. The genes that showed positive associations included CD14, CXCL12, and NOS2A, which are well known in asthma genetics,⁴ as well as novel genes, such as ALOX15, CD27, IL2RB, and SFRP1. However, we applied a correction of our P values to reduce the possibility of false-positive associations. Thus, without minimizing the potential effect of the 5 genes that did not survive correction, only the ALOX15 and CD14 associations remained significant. Taken together, these results demonstrate the usefulness, as well as the efficiency, of the microarray approach to identify genes relevant to asthma.

The ALOX15 association remained significant in 2 of the 4 studied samples and is the first report of its association with asthma phenotypes, thus accentuating our interest in this gene and its biologic pathway. In the CAPPS sample the association found was for rs7217186 and AHR, and in the SAGE sample it was for the rs2664593 and allergic asthma. In both cases a protective effect of the minor allele was suggested. The ALOX15 gene, located on chromosome 17p13.3, encodes arachidonate 15-lipoxygenase (15-LO). This enzyme is expressed in several lung cell types, including human airway epithelial cells,³³ alveolar macrophages,³⁴, ³⁵ and neutrophils,³⁶ and converts arachidonic acid into 15(S)-hydroxyeicosatetraenoic acid (15[S]-HETE), a major metabolite of the 15-LO pathway. Alterations in the 15-LO pathway have been suggested as potential contributors to the pathogenesis of asthma,³⁴, ³⁷ but a protective anti-inflammatory role has also been proposed.

First, 15-LO could act through the uptake of soluble 15(S)-HETE into the cellular membrane phospholipids, contributing to a decrease in the pericellular concentration of this proinflammatory mediator.³³, ³⁴ Second, 15-LO could have a protective effect through lipoxin A₄ biosynthesis (from 15[S]-HETE), which exerts diverse inhibitory mechanisms to protect against AHR and asthmatic inflammation.³⁵, ³⁸ Finally, 15-LO could act through the inhibition of the chemotactic mediator leukotriene B₄, which modulates neutrophilic trafficking and infiltration.³⁶, ³⁹ Thus the protective association we report for ALOX15 rs2664593 and rs7217186 with respect to allergic asthma and AHR could be explained by these anti-inflammatory roles of the 15-LO pathway. Obviously, this hypothesis remains speculative because further functional investigations of the ALOX15 gene in inflammation are needed to delineate its role in the pathophysiology of asthma.

The variability in association patterns among the 4 studies might reflect the differences in their respective designs,¹⁸ as well as genetic and environmental heterogeneity. Indeed, the recruitment of the subjects differed between the studies: the SLSJ probands were recruited at a mean age of 18 years; the CAPPS subjects were recruited prenatally based on a family history of asthma, allergy, or both; the SAGE subjects were ascertained as part of a population-based sample; and the Busselton cohort is a study of adult asthmatic subjects. Because the natural history of asthma shows considerable variability implying complex pathologic mechanisms,⁴⁰, ⁴¹ the variable recruitment strategies could have resulted in subjects who are at different disease stages or actual phenotypic subsets (eg, childhood transient vs persistent asthma). Because different genes might predispose to these different phenotypes, this could be a cause of heterogeneous results. Another potential source of different association patterns is variable LD patterns for the studied SNPs across the populations.⁴² The SLSJ sample is recognized as a founder population,¹¹, ¹², ⁴³, ⁴⁴, ⁴⁵, ⁴⁶ whereas for the CAPPS, SAGE, and Busselton studies, individuals come from admixed Canadian or Australian populations. These populations present variable ALOX15 LD patterns, such that a polymorphism might be in LD with a nearby disease allele in the CAPPS or SAGE samples but not in the SLSJ or Busselton studies. A last possible factor that could explain, in part, the different associations found for ALOX15 is population-specific gene-gene and gene-environment interactions.⁴² This possibility is also likely to explain the unique CD14 association found in the SAGE sample. In fact, a literature review reports CD14 as a classical example of gene-environment interactions for asthma and atopy phenotypes, and to date, numerous positive and negative associations have been reported across diverse populations.⁴⁷

In conclusion, we believe that hypothesis-driven candidate gene selection from microarray experiments is a methodological approach that improves the likelihood of finding positive and novel asthma associations within a limited number of tests performed. We also think that studying numerous independent asthma samples has the advantage of being able to document the biologic relevance of candidate genes across diverse populations living in different environments. Based on the results presented here, we believe that ALOX15 represents an interesting potential candidate for further study.

Back to Article Outline

We thank the participants in each asthma study presented in this report.

Back to Article Outline

Methods 

The SLSJ familial sample 

The SLSJ region of northeastern Quebec, Canada, is inhabited by 287,000 individuals who descend predominantly from approximately 2500 founders originating from France who settled in Quebec in the 17th century.^E1 The population grew at a high rate with little admixture over 12 to 14 generations and is an example of a young founder population.^E2 Recruitment occurred through media advertisement (newspaper and radio) and from the Chicoutimi Hospital's specialized clinics (pediatric and respiratory). To be included, probands were required to fulfill at least 2 of the following criteria: (1) a minimum of 3 clinic visits for acute asthma within 1 year; (2) 2 or more asthma-related hospital admissions within 1 year; or (3) steroid dependency, as defined by either 6 months of oral or 1 year of inhaled corticosteroid use. Families were included in the study if parents were available for phenotypic assessment, if at least 1 parent was unaffected, and if all 4 grandparents were of French-Canadian origin. The family participation rate was about 60%. A total of 223 independent families with family size ranging from 3 to 17 and the number of affected family members ranging from 1 to 10 were analyzed.

The affection status of the SLSJ subjects was determined by means of clinical evaluation and the completion of a standard respiratory questionnaire that was modified to include questions about asthma and atopy severity; family history of asthma, atopy, or both; age of onset; and presence of other respiratory diseases.^E3 In 41 cases the age of onset described by parents was less than 2 years; because of the uncertainty of this information, we used a default class of 2 years. Measurements of expiratory flow were done with a Morgan spirometer (Morgan Spiro 232; PK Morgan Ltd), according to the American Thoracic Society's recommendations.^E4 Bronchodilator response was measured as the increase in FEV₁ at 15 minutes after a 200-mg dose of salbutamol was inhaled. Peak expiratory flow rates were measured in the morning and evening over a period of 2 weeks by using a mini-Wright peak flowmeter (Armstrong Medical, Lincolnshire, Ill). The best of 3 repeated measurements was recorded on a diary card. Skin prick tests were performed for 24 common airborne allergens, which were divided into 6 main categories: (1) cat, dog, horse, cow, and feathers; (2) dust; (3) Dermatophagoides farinae and Dermatophagoides pteronyssinus; (4) grass, weed, ambrosia, timothy, and ryegrass; (5) tree mix, birch, maple, oak, and elm; and (6) Cladosporium species, Hormodendrum species, Alternaria alternata, Alternaria tenuis, Aspergillus species, and Penicillium species. Serum IgE levels were measured with enzyme immunofluorometry (QuantiCLONE total IgE kit; Kallestad Diagnostics, Chaska, Minn). Methacholine-based bronchoprovocation tests were performed for participants 12 years and older, according to the method described by Juniper et al.^E5 The dose of methacholine that resulted in a 20% decrease in FEV₁ was recorded as the PC₂₀. All the tests were performed by the investigators for every participant (except for bronchoprovocation) at the University of Montreal Community Genomic Medicine Centre in Chicoutimi, Canada. Medical charts were also reviewed for any previous function tests and medication. AHR was defined as a PC₂₀ of less than 8 mg/mL. If PC₂₀ results were unavailable, AHR was defined as either an increase in the postbronchodilator FEV₁ of greater than 15% from baseline or a variation of the morning-evening peak expiratory flow rate of greater than 12% over a 2-week period. Participants were defined as asthmatic if (1) they had a reported history of asthma (validated by a physician) or (2) they had asthma-related symptoms and a positive PC₂₀ result. Subjects were deemed atopic if they had at least 1 positive response (wheal diameter ≥3 mm at 10 minutes) on skin prick testing. The Chicoutimi University Hospital Ethics Committee approved the study, and subjects provided informed consent.

The CAPPS 

The CAPPS was initiated in 1995 to assess the effectiveness of a multifaceted intervention program applied in the first 12 months of life on the primary prevention of asthma and other atopic disorders in high-risk infants.^E6 Infants at high risk for asthma and other allergic disorders were identified, and their mothers were recruited during their third trimester of pregnancy. Infants at high risk were defined as those who had a parent with asthma or 2 first-degree relatives with allergies or atopic disorders. The study had 2 recruitment centers in Canada (Vancouver and Winnipeg). The diagnosis of asthma at the 7-year time point^E7 was made by a pediatric allergist in each center without knowledge of the group allocation status of the children and who did not provide health care services to the families. The physician conducted a structured interview with parents by using a standardized form to record symptoms and physical findings. Spirometry and methacholine challenge testing were performed after obtaining parental consent. The diagnoses of asthma and other atopic disorders were clinical decisions made by the pediatric allergists without knowledge of the results of the questionnaire, allergy skin tests, spirometry, or methacholine challenge tests. Atopy was defined as at least 1 positive response (wheal diameter ≥3 mm at 10 minutes) on skin prick tests. Allergy skin tests were performed with the epicutaneous method by using a prick lanceter (Hollister-Stier, Omega Laboratories LTD, Montreal, Quebec, Canada) with the following allergens (Hollister-Stier, Omega Laboratories LTD): house dust mite (D pteronyssinus and D farinae), cat, dog, cockroach, Alternaria species, Cladosporium species, tree, grass and weed (ragweed) pollens, cow's milk, egg white, wheat, soy, and peanut. Histamine (1 mg/mL) was used as the positive control, and saline was used as the negative control. The largest wheal diameter and its perpendicular were measured at 15 minutes after testing. A mean wheal diameter of 3 mm or greater than that elicited by the negative control was considered a positive reaction. Atopy was defined as a positive skin test reaction to 1 or more of these common allergens.

Methacholine (Methapharm, Inc, Brantford, Ontario, Canada) challenge testing was carried out according to the protocol of Cockcroft et al.^E8 Two children who had FEV₁ values of less than 70% of predicted value were excluded from methacholine testing, but both of them had a postbronchodilator change in FEV₁ of 12% or greater, and they were considered to have AHR. The methacholine PC₂₀ value was determined. The Ethics Committees of the University of British Columbia and the University of Manitoba approved the study, and parents provided written informed consent for participation.

The SAGE 

The SAGE is derived from a population-based cohort of 16,320 children born in the province of Manitoba, Canada, between January 1, 1995, and December 31, 1995. A survey on child health and home environment exposures was mailed to the parents of this birth cohort in 2002. After parent response to the mailed survey and approval to be contacted, a subset of children was invited to join the study at age 8 to 10 years. This included all children with parent-declared asthma and a comparable number of children without asthma who were randomly selected after stratification of the received surveys by urban/rural location and family income. All recruited children underwent clinical assessment for asthma by a pediatric allergist (AB) on the basis of the Canadian Asthma Consensus Guidelines. To aid in the diagnosis, a standardized history was used, including questions on cough, wheeze, shortness of breath, response to current medications (ie, bronchodilators and corticosteroids), and the presence of other allergic conditions (eg, allergic rhinitis, atopic dermatitis, and food allergies). Physical examination included examination for chest symptoms (hyperinflation, wheeze, prolonged expiration, and decreased breath sounds). Hospitalization and medical visits for breathing difficulty in the past year were also noted. All findings were recorded and a diagnosis of asthma in the last 12 months was made. In total, 723 study trios (child and both parents) were recruited, 247 of which have a child affected with asthma. A child affected with asthma was defined as follows: pediatric allergist diagnosis of asthma based on history and physical examination. AHR was assessed in all children after a methacholine challenge test (Methapharm, Inc, Brantford, Ontario Canada) carried out according to the protocol of Cockcroft et al.^E8 The children with FEV₁ values of less than 70% of predicted value had reversibility testing with an aerosol bronchodilator. Reversibility was defined as an improvement in FEV₁ by greater than 12% after inhalation of a β₂-adrenergic agonist. Bronchoprovocation was not attempted for these children. Atopic status was determined for all children; skin prick tests to common relevant allergens were performed (tree pollen mix, weed pollen mix, ragweed [separate], grass pollen mix, Alternaria species, Cladosporium species, Penicillium species, house dust mites [D pteronyssinus and D farinae], cockroach, cat, dog, feathers, and peanut), and atopy was defined as at least 1 positive response (wheal diameter ≥3 mm at 10 minutes). For children in Winnipeg and the immediate surroundings, the assessment was carried out in the investigator's laboratories in the Manitoba Institute of Child Health. For children in remote and northern communities, a team travelled to the communities to assess children in and surrounding those communities. The study was approved by the Ethics Committee of the University of Manitoba, and parents provided written informed consent for participation.

The Busselton Health Study population (Busselton cohort) 

Residents of the town of Busselton in the southwest of Western Australia have been involved in a series of health surveys since 1966. The population is predominantly of European origin. In the present nested case-control study, all subjects (n = 1599, of which 1549 were sent for genotyping [679 cases and 870 control subjects]) who attended any of 6 cross-sectional surveys from 1996 to 1981 and the follow-up survey in 1994 and who had available DNA were included. Subjects were considered to have asthma if they answered yes to the question “Has your doctor ever told you that you had asthma/bronchial asthma?” at any survey. All other subjects were considered control subjects. The panel includes a sample of allergic and nonallergic individuals, as determined based on skin prick tests to common allergens. Methacholine challenge tests using a modification of the Yan method^E9 were performed on all participants of the 1994 follow-up survey. The early dose-response slope with PD₂₀ was determined.

Pairwise LD pattern 

The strength of LD between pairs of SNPs in each gene has been measured as D′^E10 by using Haploview software (3.32 version).^E11 Fig E1, Fig E2, Fig E3, Fig E4, Fig E5, Fig E6, Fig E7, Fig E8 present these LD patterns for each studied gene in each studied sample.

Back to Article Outline

References 

Back to Article Outline

Fig E1. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for ALOX15. A, ALOX15 LD pattern in the SLSJ familial sample. B, ALOX15 LD pattern in the CAPPS cohort. C, ALOX15 LD pattern in the SAGE cohort. D, ALOX15 LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E2. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for CD14. A, CD14 LD pattern in the SLSJ familial sample. B, CD14 LD pattern in the CAPPS cohort. C, CD14 LD pattern in the SAGE cohort. D, CD14 LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E3. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for CD27. A, CD27 LD pattern in the SLSJ familial sample. B, CD27 LD pattern in the CAPPS cohort. C, CD27 LD pattern in the SAGE cohort. D, CD27 LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E4. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for CXCL12. A, CXCL12 LD pattern in the SLSJ familial sample. B, CXCL12 LD pattern in the CAPPS cohort. C, CXCL12 LD pattern in the SAGE cohort. D, CXCL12 LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E5. 

• 
  • View Large Image
  • Download to PowerPoint
• 
  • View Large Image
  • Download to PowerPoint
• 
  • View Large Image
  • Download to PowerPoint
• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for IL2RB. A, IL2RB LD pattern in the SLSJ familial sample. B, IL2RB LD pattern in the CAPPS cohort. C, IL2RB LD pattern in the SAGE cohort. D, IL2RB LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E6. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for IL7R. A, IL7R LD pattern in the SLSJ familial sample. B, IL7R LD pattern in the CAPPS cohort. C, IL7R LD pattern in the SAGE cohort. D, IL7R LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E7. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for NOS2A. A, NOS2A LD pattern in the SLSJ familial sample. B, NOS2A LD pattern in the CAPPS cohort. C, NOS2A LD pattern in the SAGE cohort. D, NOS2A LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Fig E8. 

• 
  • View Large Image
  • Download to PowerPoint

• LD pattern for SFRP1. A, SFRP1 LD pattern in the SLSJ familial sample. B, SFRP1 LD pattern in the CAPPS cohort. C, SFRP1 LD pattern in the SAGE cohort. D, SFRP1 LD pattern in the Busselton cohort. The location of each tested TagSNP along the chromosome is indicated in the upper part of the figure. The number in each diamond indicates the magnitude of LD (D′) as a percentage between respective pairs of SNPs. The strength of LD is depicted by progression of color: the color moves from red to white as D′ runs from 1 to 0.

Back to Article Outline

Table E1. 

FBAT values for all TagSNPs of the SLSJ sample that generated significant associations in any 1 population for 1 or more of the studied phenotypes

P values of .05 or less are marked in bold. Gray shading indicates that the TagSNP has at least 1 P value of less than .05. Corrected P values are indicated in parentheses.

MAF, Minor allele frequency; N, number of families contributing the statistic; Z, Z score.

Back to Article Outline

Table E2. 

FBAT values for all TagSNPs of the CAPPS sample that generated significant associations in any 1 population for 1 or more of the studied phenotypes

P values of .05 or less are marked in bold and underlined if they remain significant after correction for multiple testing. Gray shading indicates that the TagSNP has at least 1 P value of less than .05. Corrected P values are indicated in parentheses.

MAF, Minor allele frequency; N, Number of families contributing the statistic; Z, Z score; NA, not available because the number of families was too low (<10) to contribute to the test.

Back to Article Outline

Table E3. 

FBAT values for all TagSNPs of the SAGE sample that generated significant associations in any 1 population for 1 or more of the studied phenotypes

P values of .05 or less are marked in bold and underlined if they remain significant after correction for multiple testing. Gray shading indicates that the TagSNP has at least 1 P value of less than .05. Corrected P values are indicated in parentheses.

MAF, Minor allele frequency; N, number of families contributing the statistic; Z, Z score; NA, not available because the number of families is too low (<10) to contribute to the test.

Back to Article Outline

Table E4. 

Odds ratios for all TagSNPs of the Busselton cohort that generated significant associations in any 1 population for 1 or more of the studied phenotypes

P values of .05 or less are marked in bold. Gray shading indicates that the TagSNP has at least 1 P value of less than .05. Corrected P values are indicated in parentheses.

MAF, Minor allele frequency; OR, 95% CI odd ratio.

Back to Article Outline

References 

1. Bousquet J, Jeffery PK, Busse WW, Johnson M, Vignola AM. Asthma. From bronchoconstriction to airways inflammation and remodeling. Am J Respir Crit Care Med. 2000;161:1720–1745
   • View In Article
   • MEDLINE
2. Sandford A, Weir T, Pare P. The genetics of asthma. Am J Respir Crit Care Med. 1996;153:1749–1765
   • View In Article
   • MEDLINE
3. Hoffjan S, Nicolae D, Ober C. Association studies for asthma and atopic diseases: a comprehensive review of the literature. Respir Res. 2003;4:14
   • View In Article
   • CrossRef
4. Ober C, Hoffjan S. Asthma genetics 2006: the long and winding road to gene discovery. Genes Immun. 2006;7:95–100
   • View In Article
   • MEDLINE
   • CrossRef
5. Holloway JW, Koppelman GH. Identifying novel genes contributing to asthma pathogenesis. Curr Opin Allergy Clin Immunol. 2007;7:69–74
   • View In Article
   • MEDLINE
6. Schmidt-Weber CB. Gene expression profiling in allergy and asthma. Chem Immunol Allergy. 2006;91:188–194
   • View In Article
   • MEDLINE
7. Tremblay K, Lemire M, Provost V, Pastinen P, Renaud Y, Sandford AJ, et al. Association study between the CX3CR1 gene and asthma. Genes Immun. 2006;7:632–639
   • View In Article
   • MEDLINE
   • CrossRef
8. Laprise C, Sladek R, Ponton A, Bernier MC, Hudson TJ, Laviolette M. Functional classes of bronchial mucosa genes that are differentially expressed in asthma. BMC Genomics. 2004;5:21–31
   • View In Article
   • MEDLINE
   • CrossRef
9. Laitinen T, Daly MJ, Rioux JD, Kauppi P, Laprise C, Petays T, et al. A susceptibility locus for asthma-related traits on chromosome 7 revealed by genome-wide scan in a founder population. Nat Genet. 2001;28:87–91
   • View In Article
   • MEDLINE
   • CrossRef
10. Heyer E, Tremblay M, Desjardins B. Seventeenth-century European origins of hereditary diseases in the Saguenay population (Quebec, Canada). Hum Biol. 1997;69:209–225
    • View In Article
    • MEDLINE
11. Labuda M, Labuda D, Korab-Laskowska M, Cole DE, Zietkiewicz E, Weissenbach J, et al. Linkage disequilibrium analysis in young populations: pseudo-vitamin D-deficiency rickets and the founder effect in French Canadians. Am J Hum Genet. 1996;59:633–643
    • View In Article
    • MEDLINE
12. Scriver CR. Human genetics: lessons from Quebec populations. Annu Rev Genomics Hum Genet. 2001;2:69–101
    • View In Article
    • MEDLINE
    • CrossRef
13. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, November 1986. Am Rev Respir Dis. 1987;136:225–244
    • View In Article
    • MEDLINE
14. Lemire M, Roslin NM, Laprise C, Hudson TJ, Morgan K. Transmission-ratio distortion and allele sharing in affected sib pairs: a new linkage statistic with reduced bias, with application to chromosome 6q25.3. Am J Hum Genet. 2004;75:571–586
    • View In Article
    • MEDLINE
    • CrossRef
15. Chan-Yeung M, Manfreda J, Dimich-Ward H, Ferguson A, Watson W, Becker A. A randomized controlled study on the effectiveness of a multifaceted intervention program in the primary prevention of asthma in high-risk infants. Arch Pediatr Adolesc Med. 2000;154:657–663
    • View In Article
    • MEDLINE
16. Chan-Yeung M, Ferguson A, Watson W, Dimich-Ward H, Rousseau R, Lilley M, et al. The Canadian Childhood Asthma Primary Prevention Study: outcomes at 7 years of age. J Allergy Clin Immunol. 2005;116:49–55
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (154 KB)
    • CrossRef
17. Becker A, Watson W, Ferguson A, Dimich-Ward H, Chan-Yeung M. The Canadian asthma primary prevention study: outcomes at 2 years of age. J Allergy Clin Immunol. 2004;113:650–656
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (174 KB)
    • CrossRef
18. Bégin P, Tremblay K, Daley D, Lemire M, Claveau S, Salesse C, et al. Association of urokinase-type plasminogen activator with asthma and atopy. Am J Respir Crit Care Med. 2007;175:1109–1116
    • View In Article
    • MEDLINE
    • CrossRef
19. James AL, Palmer LJ, Kicic E, Maxwell PS, Lagan SE, Ryan GF, et al. Decline in lung function in the Busselton Health Study: the effects of asthma and cigarette smoking. Am J Respir Crit Care Med. 2005;171:109–114
    • View In Article
    • MEDLINE
    • CrossRef
20. Palmer LJ, Knuiman MW, Divitini ML, Burton PR, James AL, Bartholomew HC, et al. Familial aggregation and heritability of adult lung function: results from the Busselton Health Study. Eur Respir J. 2001;17:696–702
    • View In Article
    • MEDLINE
    • CrossRef
21. Moffatt MF, Schou C, Faux JA, Abecasis GR, James A, Musk AW, et al. Association between quantitative traits underlying asthma and the HLA-DRB1 locus in a family-based population sample. Eur J Hum Genet. 2001;9:341–346
    • View In Article
    • MEDLINE
    • CrossRef
22. Hill MR, James AL, Faux JA, Ryan G, Hopkin JM, le Souef P, et al. Fc epsilon RI-beta polymorphism and risk of atopy in a general population sample. BMJ. 1995;311:776–779
    • View In Article
    • MEDLINE
    • CrossRef
23. Daniels SE, Bhattacharrya S, James A, Leaves NI, Young A, Hill MR, et al. A genome-wide search for quantitative trait loci underlying asthma. Nature. 1996;383:247–250
    • View In Article
    • MEDLINE
    • CrossRef
24. Lincoln MR, Montpetit A, Cader MZ, Saarela J, Dyment DA, Tiislar M, et al. A predominant role for the HLA class II region in the association of the MHC region with multiple sclerosis. Nat Genet. 2005;37:1108–1112
    • View In Article
    • MEDLINE
    • CrossRef
25. Carlson CS, Eberle MA, Rieder MJ, Yi Q, Kruglyak L, Nickerson DA. Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet. 2004;74:106–120
    • View In Article
    • MEDLINE
    • CrossRef
26. Lake SL, Blacker D, Laird NM. Family-based tests of association in the presence of linkage. Am J Hum Genet. 2000;67:1515–1525
    • View In Article
    • MEDLINE
    • CrossRef
27. Horvath S, Xu X, Lake SL, Silverman EK, Weiss ST, Laird NM. Family-based tests for associating haplotypes with general phenotype data: application to asthma genetics. Genet Epidemiol. 2004;26:61–69
    • View In Article
    • MEDLINE
    • CrossRef
28. Laird NM, Horvath S, Xu X. Implementing a unified approach to family-based tests of association. Genet Epidemiol. 2000;19(suppl 1):S36–S42
    • View In Article
    • CrossRef
29. Dudbridge F. Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol. 2003;25:115–121
    • View In Article
    • MEDLINE
    • CrossRef
30. Li J, Ji L. Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix. Heredity. 2005;95:221–227
    • View In Article
    • MEDLINE
    • CrossRef
31. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet. 2004;74:765–769
    • View In Article
    • MEDLINE
    • CrossRef
32. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–265
    • View In Article
    • MEDLINE
    • CrossRef
33. Profita M, Vignola AM, Sala A, Mirabella A, Siena L, Pace E, et al. Interleukin-4 enhances 15-lipoxygenase activity and incorporation of 15(S)-HETE into cellular phospholipids in cultured pulmonary epithelial cells. Am J Respir Cell Mol Biol. 1999;20:61–68
    • View In Article
    • MEDLINE
34. Profita M, Sala A, Riccobono L, Paterno A, Mirabella A, Bonanno A, et al. 15-Lipoxygenase expression and 15(S)-hydroxyeicoisatetraenoic acid release and reincorporation in induced sputum of asthmatic subjects. J Allergy Clin Immunol. 2000;105:711–716
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (74 KB)
    • CrossRef
35. Levy BD, Romano M, Chapman HA, Reilly JJ, Drazen J, Serhan CN. Human alveolar macrophages have 15-lipoxygenase and generate 15(S)-hydroxy-5,8,11-cis-13-trans-eicosatetraenoic acid and lipoxins. J Clin Invest. 1993;92:1572–1579
    • View In Article
    • MEDLINE
    • CrossRef
36. Profita M, Sala A, Riccobono L, Pace E, Paterno A, Zarini S, et al. 15(S)-HETE modulates LTB(4) production and neutrophil chemotaxis in chronic bronchitis. Am J Physiol Cell Physiol. 2000;279:C1249–C1258
    • View In Article
    • MEDLINE
37. Chu HW, Balzar S, Westcott JY, Trudeau JB, Sun Y, Conrad DJ, et al. Expression and activation of 15-lipoxygenase pathway in severe asthma: relationship to eosinophilic phenotype and collagen deposition. Clin Exp Allergy. 2002;32:1558–1565
    • View In Article
    • MEDLINE
    • CrossRef
38. Levy BD, De Sanctis GT, Devchand PR, Kim E, Ackerman K, Schmidt BA, et al. Multi-pronged inhibition of airway hyper-responsiveness and inflammation by lipoxin A(4). Nat Med. 2002;8:1018–1023
    • View In Article
    • MEDLINE
    • CrossRef
39. Takata S, Matsubara M, Allen PG, Janmey PA, Serhan CN, Brady HR. Remodeling of neutrophil phospholipids with 15(S)-hydroxyeicosatetraenoic acid inhibits leukotriene B4-induced neutrophil migration across endothelium. J Clin Invest. 1994;93:499–508
    • View In Article
    • MEDLINE
    • CrossRef
40. Reed CE. The natural history of asthma. J Allergy Clin Immunol. 2006;118:543–550
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (130 KB)
    • CrossRef
41. Koh MS, Irving LB. The natural history of asthma from childhood to adulthood. Int J Clin Pract. 2007;61:1371–1374
    • View In Article
    • CrossRef
42. Hirschhorn JN, Lohmueller K, Byrne E, Hirschhorn K. A comprehensive review of genetic association studies. Genet Med. 2002;4:45–61
    • View In Article
    • MEDLINE
    • CrossRef
43. Heyer E, Tremblay M. Variability of the genetic contribution of Quebec population founders associated to some deleterious genes. Am J Hum Genet. 1995;56:970–978
    • View In Article
    • MEDLINE
44. Engert JC, Berube P, Mercier J, Dore C, Lepage P, Ge B, et al. ARSACS, a spastic ataxia common in northeastern Quebec, is caused by mutations in a new gene encoding an 11.5-kb ORF. Nat Genet. 2000;24:120–125
    • View In Article
    • MEDLINE
    • CrossRef
45. Richter A, Rioux JD, Bouchard JP, Mercier J, Mathieu J, Ge B, et al. Location score and haplotype analyses of the locus for autosomal recessive spastic ataxia of Charlevoix-Saguenay, in chromosome region 13q11. Am J Hum Genet. 1999;64:768–775
    • View In Article
    • MEDLINE
    • CrossRef
46. Lander ES, Schork NJ. Genetic dissection of complex traits. Science. 1994;265:2037–2048
    • View In Article
    • MEDLINE
47. Simpson A, John SL, Jury F, Niven R, Woodcock A, Ollier WE, et al. Endotoxin exposure, CD14, and allergic disease: an interaction between genes and the environment. Am J Respir Crit Care Med. 2006;174:386–392
    • View In Article
    • MEDLINE
    • CrossRef