INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

A genome-wide screen for asthma susceptibility alleles in the Hutterites, a founder population of European origins, was recently reported (1). A marker in the 5q23-31 cytokine gene cluster, D5S1480, showed evidence of linkage to asthma by both the likelihood ratio (LR) ² test (2) and the transmission disequilibrium test (TDT) (3, 4) (p = 0.0079 and 0.000091, respectively) in a large, multigenerational Hutterite kindred (1). Markers in this region have been linked to or associated with asthma and asthma-related phenotypes in other populations (5). D5S1480 is located less than 3 centimorgans (cM) from the gene encoding the ₂-adrenergic receptor (₂AR), and polymorphisms in the ₂AR gene have been shown to influence receptor function in cellular assays (13, 14) and to be associated with asthma phenotypes (15). Thus, we considered this gene to be an excellent positional and functional candidate for the 5q-linked asthma-susceptibility gene in the Hutterites.

In particular, two amino acid substitutions in the ₂AR gene (Arg16Gly and Gln27Glu) are common variants in both normal and asthmatic subjects (15). In vitro studies using human ₂AR transfected into Chinese hamster fibroblasts and cultured human airway smooth muscle cells demonstrated that the Gly16 allele confers enhanced agonist-induced downregulation whereas the Glu27 allele imparts resistance to downregulation (13, 14). Neither polymorphic site has been associated with asthma per se (15, 18, 19), but among asthmatic subjects, the Gly16 allele has been associated with oral steroid use (15), nocturnal asthma (16), and decreased responsiveness and desensitization to ₂-agonists (19, 21). In addition, the Gly16-Gln27 haplotype has been associated with moderate as compared with mild asthma (20) and with bronchial hyperresponsiveness (BHR) in a cohort unselected for asthma or atopy (24), whereas the Glu27 allele has been associated with both reduced airway reactivity (17, 25) and elevated IgE levels (18).

We report the results of a population-based study of the ₂AR polymorphisms in Hutterite individuals who participated in our genetic studies of asthma (1). In addition to investigating associations between ₂AR genotypes and asthma- associated phenotypes, the availability of a large number of nonasthmatic individuals derived from the same population provided a unique opportunity to explore the role of ₂AR genotypes in normal lung function.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

The Hutterites are an Anabaptist religious sect that originated in the Tyrolean Alps in the 1500s. After migrations across eastern Europe into Russia to escape religious persecution, a group of approximately 900 Hutterites settled in the United States in what is now South Dakota in the 1870s (26). The Hutterites practice a communal, agrarian lifestyle. Smoking is prohibited (and rare) in the population. As a result, environmental exposures within this population are similar, and exposure to first- or second-hand tobacco smoke is unusual. We studied 361 individuals living on communal farms (colonies) in South Dakota. Four colonies were selected because previous studies documented an asthma prevalence of 10.9% (27). The present study includes all colony members age 6 yr and older who were present at the time of data collection. In addition, we evaluated 55 Hutterites from neighboring colonies who participated in our study owing to a previous diagnosis of asthma in their family. The study participants are descended from an estimated 64 Hutterite progenitors (28) and represent one large, multigenerational pedigree (referred to as the primary sample described in Reference 1).

Clinical Assessment

The Collaborative Study on the Genetics of Asthma (CSGA) respiratory health and medication history questionnaire (8) was modified to take into account cultural differences among the Hutterites and was administered to all individuals older than 15 yr of age and to the mothers of children younger than age 15. Baseline spirometry was performed (M1-01001-001 Spirometer; Multispiro, Tempe, AZ) following guidelines of the American Thoracic Society (29). Airway reactivity was assessed by methacholine (MCh) bronchial provocation test in those subjects with a baseline FEV₁  70% predicted n = 341), using a modified, abbreviated screening protocol (30). After inhalation of diluent, MCh at a concentration of 25 mg/dl was delivered by the tidal breathing method via a nebulizer (model 646; Devilbiss, Somerset, PA) with a breath-activated dosimeter. Subjects with a predicted FEV₁  70% (n = 5) underwent reversibility studies with 200 µg albuterol via metered-dose inhaler. A  20% decrease from baseline FEV₁ after 25 mg/dl MCh inhalation or a  15% increase in FEV₁ after administration of bronchodilator was considered to be diagnostic of BHR. Provocative concentration of MCh causing a 20% reduction in FEV₁ (PC₂₀) could not be extrapolated in this sample because the asthmatic subjects had a  20% decrease from baseline FEV₁ values after the first inhaled dose of MCh. Therefore, the percent drop in FEV₁ at the maximal inhaled dose of MCh was used as a measure of airway reactivity. This measure was available on all but the five individuals who had reversibility studies. Furthermore, on subsequent visits 2 to 3 yr later, 53 of the 101 subjects with a  12% decrease in baseline FEV₁ at a dose of 25 mg/dl MCh were available for retesting with serial administration of 1 mg/dl, 5 mg/dl, and 25 mg/dl MCh, from which the PC₂₀ was extrapolated. We chose 12% as a cutoff for these follow-up studies to allow for disease progression in previously undiagnosed individuals. Among the subjects who were tested, 34 had a PC₂₀ (median = 5.79 mg/dl; range 1.5 to 19.8 mg/dl). Participants were asked to withhold using any asthma medications for 12 h before testing; none were using long-acting -agonists or were taking corticosteroids. Total serum concentrations of IgE were determined at Sanofi Diagnostics (Minneapolis, MN).

Diagnostic criteria for asthma included self-reported symptoms of asthma and results of MCh provocation testing and reversibility studies (1). Our study sample included 50 subjects with unambiguous asthma (one with reversibility studies only), 30 subjects with asymptomatic BHR (two with reversibility studies only), 35 subjects with self-reported asthma symptoms but who did not meet the criteria for BHR, and 246 subjects with neither a current or past history of asthma symptoms and normal MCh challenge studies. The latter group was considered to be unaffected.

₂AR Genotyping

The ₂AR genotypes at codons 16 (Arg Gly) and 27 (Gln Glu) were determined either by dideoxynucleotide sequencing or by restriction fragment length polymorphism (RFLP) analysis. The 50 asthmatic individuals, along with their parents (and siblings, if DNA from one or both parents was not available), were genotyped by the sequencing of polymerase chain reaction (PCR) products. These were generated using primers designed from the published sequence (31): 5'-ATG-GGGCAACCCGGGAAC-3' (sense) and 5'-CAGGCCAGTGAAGTGATGAA-3' (antisense). Each PCR reaction contained 200 ng DNA, 10 pmol each of sense and antisense primer, 1.25 units Taq polymerase (Promega, Madison, WI), and 200 µM deoxyribonucleoside triphosphate (dNTP) in a buffer of 100 mM Tris (pH 8.3), 500 mM KCl, and 15 mM MgCl₂. PCR conditions included denaturation at 95° C for 5 min, then 30 cycles of 95° C for 1 min, 55° C for 1 min, and 72° C for 2 min, followed by a final 10-min extension step at 72° C. PCR products were sequenced using fluorescent-labeled primers, solid-phase sequencing kits (Pharmacia Biotech, Uppsala, Sweden), and an automated sequencer (ALF Sequencer; Pharmacia Biotech, Uppsala, Sweden), according to the manufacturer's instructions.

All other individuals in the pedigree were genotyped by RFLP analysis using the primers and protocol detailed by Martinez and colleagues (19), with some modifications. PCR reactions contained 200 ng of DNA, 1.5 mM MgCl₂, 200 µM dinucleotide triphosphate mix, 1 µmol of each primer, and 0.8 U Amplitaq Gold polymerase in buffer supplied by the manufacturer (Perkin-Elmer, Norwalk, CT). After denaturation at 94° C for 10 min, PCR conditions were 94° C for 40 s, 64° C for 40 s, and 72° C for 50 s times 40 cycles followed by extension at 72° C for 5 min. A volume of 8 µl of the 168-bp PCR product was digested with 2 U NcoI or 0.5 U BbvI enzyme and 5 µl of 10 × buffer provided by the manufacturer (New England BioLab, Boston, MA) in a total volume of 50 µl at 37° C for 2 h. The digest was then electrophoresed on a 12% polyacrylamide gel in 1× TBE buffer at 30 milliamps for 2 h. Polymorphisms were identified by visualization of bands under ultraviolet illumination after ethidium bromide staining. A total of 340 individuals were genotyped at codons 16 and 27 of the ₂AR gene. Haplotypes were assigned by observation of the pattern of allele segregation in the pedigree. Forty-one individuals could not be typed because DNA was no longer available, two or more PCR attempts failed, or incompatibilities in the genotypes between family members could not be resolved.

Statistical Analysis

The TDT (3, 4) was used as a measure of linkage, considering the Arg16Gly and Gln27Glu polymorphisms separately and as a 2-locus haplotype. The transmission pattern of each allele or haplotype from heterozygous parents to affected children was scored using the different diagnostic categories of asthma, as previously described (1). The TDT is a robust test that assesses whether one allele is more likely to be transmitted from heterozygous parent to affected child, testing simultaneously for linkage and linkage disequilibrium. However, because these studies were conducted in a single large pedigree, a significant TDT result may merely reflect linkage and not necessarily association. In addition, the semiparametric LR ² test was used as a second measure of linkage (2). Finally, analysis of variance (ANOVA) and t tests were used to assess differences among mean values of the quantitative traits of log total serum IgE, FEV₁ percentage of predicted, and FVC percentage of predicted in groups stratified according to genotype or haplotype. Because our quantitative measure of airway reactivity (percent decrease of FEV₁ at  25 mg/dl MCh administration) and PC₂₀ were not normally distributed, these measures were compared by a nonparametric test, the Kruskal-Wallis one-way ANOVA on ranks.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

₂AR allele, genotype, and haplotype frequencies for the Arg16Gly and Gln27Glu polymorphisms are shown in Table 1. Genotypes at both sites were in Hardy-Weinberg equilibrium (goodness-of-fit ² = 0.096 at amino acid 16 and 0.104 at amino acid 27; p = 0.757 and 0.747, respectively). Four haplotypes were present in the population. However, the Arg16/ Glu27 haplotype was rare (frequency 0.7; Table 1), similar to the findings in an outbred population (19, 24).

There was no evidence for linkage to asthma at either of the two polymorphic sites, whether considered individually or on a single haplotype. TDT ² (1 degree of freedom [df]) ranged from 0 to 3.00, p > 0.10 and LR ²_{(1 df)} ranged from 0.07 to 1.86, p > 0.10 for the following qualitative traits: asthma, asthma symptoms, and BHR. When quantitative variables were analyzed, neither log total serum IgE nor percent decrease drop in FEV₁ varied with ₂AR genotype in the full sample (p > 0.10). Similarly, PC₂₀ did not vary with ₂AR genotype among 34 individuals with BHR (p > 0.10) (Table 2).

In contrast to the analyses of asthma and asthma associated phenotypes, significant differences in mean FEV₁ percentage of predicted and FVC percentage of predicted were detected among the three genotypic groups at the Arg16Gly locus (p = 0.008 and 0.021, respectively) (Table 3). Overall, subjects with one or two Gly16 alleles had a greater mean FEV₁ percentage of predicted (103.8 ± 14.9%) compared with Gly16 negative individuals (98.3 ± 13.2%; p = 0.003). FVC percentage of predicted was also higher among subjects with one or two Gly16 alleles (108.3 ± 13.2%) compared with Gly16 negative individuals (104.2 ± 12.3%; p = 0.02). When the 50 asthmatic subjects were removed from the analysis, the difference in mean FEV₁ percentage of predicted remains significant (105.2 ± 14.2% versus 99.7 ± 13.5%; p = 0.015), and FVC percentage nearly so (110.0 ± 11.0% versus 106.1 ± 12.6; p = 0.054). No associations were detected between alleles at amino acid 27 and lung volumes (Table 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This is the first population-based study of the ₂AR polymorphisms in a large, multigenerational pedigree that shows evidence for linkage of asthma to marker loci in the 5q23-31 cytokine gene cluster (1). This sample was evaluated for asthma, asthma-associated phenotypes, and lung function. We did not find evidence that the ₂AR polymorphisms at amino acids 16 or 27 were linked to or associated with asthma, BHR, total IgE, or airway reactivity. These data indicate that variation in the ₂AR gene does not account for the observed linkages between asthma phenotypes and markers in the 5q23-31 region in the Hutterites.

The lack of association between the ₂AR polymorphisms and asthma in the Hutterites is consistent with results of case-control and cohort studies in outbred populations (15, 19). However, our study did not find an association between the Gly16-Gln27 haplotype and BHR, which was reported in an outbred, Italian cohort that was also unselected for asthma or atopy (24). Furthermore, although Dewar and associates reported a significant association between the Gln27 allele and log total serum IgE in outbred English asthma families (18), this association was not present in the Hutterites or in outbred Germans (32) or Italians (24). The reasons for the differences are not clear but may be due to heterogeneity between the samples studied or differences in ascertainment strategies. For example, the Dewar study ascertained subjects through asthmatic individuals, the German study ascertained subjects through atopic individuals, and the Hutterite and Italian subjects were unselected for disease phenotypes.

Associations between the Gly16 allele and asthma severity (15, 16, 20) and the Glu27 allele and reduced airway reactivity in asthmatics with moderate disease (17) could not be adequately assessed in this study because we had few objective measures of severe asthma and, in general, this population is characterized by mild asthma. For example, none of the asthmatics in this study population are dependent on oral steroids and medical attention is rarely sought for acute exacerbations.

Surprisingly, a significant association between the Gly16 allele and higher FEV₁ percentage of predicted and FVC percentage of predicted was present in this sample, suggesting that ₂AR genotypes may influence adult lung volumes. Martinez and colleagues (19) examined these parameters and did not find an association between ₂AR genotypes and FEV₁ percentage of predicted or FVC percentage of predicted. However, the study of Martinez and colleagues included only children (mean age 10.8 yr ± 0.6), whereas our study also included adults (83% of sample was over the age of 12). Indeed, when only children  age 12 were considered in our analysis, no association between the Gly16 allele and FEV₁ percentage of predicted and FVC percentage of predicted was detected. Rather, the associations were present only in the older (> 12 yr) sample (Table 4). Thus, the role of the ₂AR gene in lung growth and development may become manifest during maturation and ultimately influence the size of the adult lung. Alternatively, the ₂AR polymorphisms may influence the rate of decline of lung function which normally occurs with age.

In conclusion, this study demonstrated that polymorphisms in the ₂AR gene are not linked to or associated with asthma or asthma-associated phenotypes in the Hutterites, despite evidence for linkage of asthma to genes in the 5q23-31 region. On the other hand, our data suggest that this gene may play a role in determining lung volumes at maturity, or in the rate of decline of lung function with aging.