|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Channing Laboratory (Drs. Barr, Speizer, and Camargo) and Division of Pulmonary and Critical Care Medicine (Dr. Drazen), Department of Medicine, Brigham and Womens Hospital, Harvard Medical School, and General Medicine Division (Dr. Barr) and Department of Emergency Medicine (Dr. Camargo), Massachusetts General Hospital, Boston, MA; and Respiratory Center for Children (Dr. Cooper), Childrens Medical Center, Atlantic Health Systems, Morristown, NJ.
Correspondence to: Carlos A. Camargo; Jr, MD, DrPH, FCCP, Channing Laboratory, 181 Longwood Ave, Boston, MA 02115; e-mail: carlos.camargo{at}channing.harvard.edu
| Abstract |
|---|
|
|
|---|
Design: Nested case-control study.
Setting: Nurses Health Study, a large, prospective cohort study with participants throughout the United States.
Participants: Among lifelong nonsmokers, 171 women with adult-onset, medication-requiring asthma and 137 age-matched control subjects.
Measurements: Physical activity and BMI were self-reported by previously validated questionnaire items. Genomic DNA was obtained from buccal brushings collected via first-class mail.
Results: Of 76 sedentary women, the adjusted odds ratios of Gly16 allele were 7.4 (p = 0.047) for asthma and 13.8 (p = 0.02) for steroid-requiring asthma. No similar associations were observed among 232 active women (p = 0.91). Sedentary individuals with both Gly16 and Glu27 alleles had a less elevated risk for asthma. BMI was associated with asthma and Glu27 allele among sedentary women.
Conclusion: This exploratory analysis suggests an important gene/environment interaction for asthma involving physical activity level. Further study in larger populations is warranted to confirm if sedentary lifestyle unmasks a genetic risk for asthma.
Key Words: asthma ß2-adrenoceptor polymorphism gene/environment interaction obesity physical activity
| Introduction |
|---|
|
|
|---|
In addition to effects on the airways, the ß2-adrenoceptor influences fat metabolism by promoting lipolysis in human adipose tissue.7 The Glu27 allele, which appears to reduce risk of asthma in children, is associated with obesity in healthy women.8 Expression of this possible genetic predisposition for obesity may be modified by environmental factors. A highly significant interaction has been shown between physical activity and polymorphism at codon 27; in a preliminary report, the Gln27 allele was associated with obesity in sedentary men but not in active men.9
A genetic polymorphism that may predispose to obesity and modify asthma severity is particularly interesting in light of reports that obesity and weight gain prospectively predict a diagnosis of asthma in the Nurses Health Study10 and that a sedentary lifestyle is associated with adult-onset asthma.11 12 We therefore hypothesized that activity status might modify the influence of genetic polymorphisms and obesity on adult-onset asthma. We performed a nested case-control study within the Nurses Health Study in which we stratified women into sedentary and active groups based on self-reported physical activity, anticipating that ß2-adrenoceptor polymorphisms and body mass index (BMI) would predispose to asthma in sedentary individuals only.
| Materials and Methods |
|---|
|
|
|---|
From 1988 onwards, participants were asked about a physician diagnosis of asthma. Participants who reported a physician diagnosis of asthma in 1988 and 1990 were sent a supplemental questionnaire in 1994 requesting information on date of asthma diagnosis, symptom severity, and medication use. Ninety-seven percent responded to this supplemental questionnaire, and 86% confirmed a physician diagnosis of asthma.
Cases for the present study were randomly selected from these confirmed cases, after restricting the group to lifelong nonsmokers who reported asthma medication use in the preceding year. Participants who reported a concurrent diagnosis of emphysema or chronic bronchitis were excluded. Age-matched control subjects were selected from lifelong nonsmokers in the overall cohort who did not report asthma or asthma medication use in the preceding year. The Human Research Committee of Brigham and Womens Hospital approved the study design and accrual methods.
Genotyping
Buccal brushing kits were sent via first-class mail to a total
of 469 patients and control subjects in 19951996. In all, 392 samples
(84%) were returned for analysis, and gene amplification of both sites
of interest was successfully performed for 330 samples (84%). Genomic
DNA was prepared for genotypic analysis using standard
techniques.14
Genotypes at B16 and B27 were assessed by
the amplification refractory mutation system,15
as
previously described.1
Genotype was verified in 10% of
individuals by oligonucleotide-specific hybridization as a quality
control measure throughout the study. Individuals who performed
genotyping were blinded to participants assignment as a patient or
control subject and to activity status.
Activity Status and Covariates
Weight and activity status in 1994 and adult height were
self-reported. Both of these questionnaire-based measures have been
previously validated in this cohort.16
17
BMI was
calculated as weight divided by height squared. Activity level was
measured as the number of metabolic equivalent (MET)-hours per week of
recreational physical activity. Sedentary status was defined a
priori as < 4.5 MET-hours per week of leisure-time physical
activity (equivalent to 1 h of medium-strain exercise per
week).18
Total caloric intake in 1994 was calculated from
a food frequency questionnaire, which has also been validated
extensively.19
Statistical Analysis
Since sequence variants were highly interrelated, analyses were
performed across three groups: Arg16 reference (B16 Arg/Arg), Gly16
allele (B16 Arg/Gly or Gly/Gly, and B27 Gln/Gln), and Gly16-Glu27
alleles (B16 Arg/Gly or Gly/Gly, and B27 Gln/Glu or Glu/Glu).
Associations were tested with
2, t
test, and Wilcoxon rank sum, as appropriate. Statistical significance
was defined at p < 0.05 (two tailed). Analyses were stratified
a priori by activity status, and multiplicative interaction
terms were tested in logistic regression models. Odds ratios (ORs) were
calculated with 95% confidence intervals (CIs). Within strata of
activity status, unconditional logistic regression models were
specified a priori to include the following variables:
dummy-coded allele groups, BMI, caloric intake, and age. Statistical
analysis was performed using SAS software (version 7.0; SAS Institute;
Cary, NC).
| Results |
|---|
|
|
|---|
|
|
|
|
In multivariate analyses (Fig 1) , cases of asthma were associated with the presence of any Gly16 allele among sedentary women (OR, 7.4 ;p = 0.047) after controlling for BMI, caloric intake, and age. Sedentary women with both Gly16 and Glu27 alleles had a risk similar to bivariate analyses. By contrast, physically active women harboring any Gly16 allele were not at increased risk of asthma (OR, 1.1; 95% CI, 0.4 to 2.7; p = 0.91), nor were active women with both Gly16 and Glu27 alleles (OR, 0.9; 95% CI, 0.4 to 1.9; p = 0.70).
In sedentary women, the association for steroid-dependent asthma and presence of any Gly16 allele (OR, 13.8; p = 0.02) was stronger than the association for other medication-requiring asthma. Among active women, results for steroid-dependent asthma were similar to those for medication-requiring asthma, except that the negative association for BMI became nonsignificant (p = 0.07). Too few observations were available to make reliable inferences about Gly16 homozygotes; however, their risk of asthma appeared to be higher than Gly16 heterozygotes in sedentary and active strata.
| Discussion |
|---|
|
|
|---|
Previous analyses of genetic risk of asthma have not, to our knowledge, been stratified by activity status; nevertheless, our observations in sedentary women are consistent with the available prior literature. Most studies to date suggest that Gly16 is associated with more severe symptoms of asthma, evidenced by greater nocturnal symptoms,1 greater-than-expected frequency of steroid-dependent asthma,2 poorer response to therapy,20 21 22 and increased atopy.3 Conversely, a relative protective effect of Glu27 has been shown in children5 but not in other populations.4 23
Prior studies3 4 8 24 of Gly16 and Glu27 polymorphisms have noted linkage disequilibrium between the two loci. Such disequilibrium, or nonindependence, is suggestive of multilocus effects (gene/gene interactions); therefore, analyses should be stratified by the loci. This approach was used in our analysis and also in a population-based study of persistent bronchial hyperresponsiveness. In that study, the Gly16-Gln27 combination was strongly associated with persistent bronchial hyperresponsiveness but the Gly16-Glu27 combination was not.24 In addition, the Gln27 allele occurred only in the presence of a Gly16 allele. The mechanism whereby these DNA sequence variants influence these observations is not known, but a number of studies25 26 27 have demonstrated varied phenotypic effects of these variants in vitro and in vivo.
Although a well-described subgroup of asthmatics has exercised-induced asthma,28 recent studies11 12 suggest that adult-onset asthma may also be associated with a lack of physical activity, particularly among older subjects. Our findings suggest an additional subgroup of women who might be at risk of asthma related to inactivity. Inactivity, or sedentary lifestyle, was defined quite strictly in this study. For example, a woman who walked at a brisk pace for 1 h/wk was defined as active. It is possible that at the low levels of physical activity, mechanisms related to lack of deep inspirations could promote bronchospasm.29
Given previously published relationships of asthma and obesity,10 sequence variants at B27 with asthma5 and obesity among sedentary men,9 we had anticipated that the association of obesity and asthma would be mediated by the Glu27 allele. We actually found the opposite: sedentary women were more likely to have asthma if they harbored Gln27 or if they had a higher BMI. That is, genetic associations with asthma were negatively confounded by BMI due to the negative association of Gln27 and BMI in sedentary women. This latter finding is consistent with the strong association between Glu27 and obesity shown in women8 but is the opposite of findings in sedentary men.9
Our study has several potential limitations. Most importantly, the data and genetic analysis were not collected to investigate asthma risk stratified by activity status. Our results should therefore be considered exploratory. Nevertheless, we established thresholds and analyses a priori based on the prior literature and found generally consistent results. Second, although the overall data set was large compared with many prior studies,1 2 21 the number of individuals in the sedentary group was relatively small. In addition, phenotype was defined by clinical not spirometric parameters. We recognize that small changes in the measurement of phenotype could alter the statistical significance of the results. Third, our data were cross-sectional, therefore the presence of asthma may have influenced activity status. It is hard to imagine, however, how asthma would influence genotype.
This exploratory study suggests that the presence of the Gly16 allele is associated with adult-onset asthma among sedentary women, implying that the relationship between ß2-adrenoceptor polymorphisms and adult-onset asthma is modifiable by environmental factors. A similar gene-environment interaction may exist between ß2-adrenoceptor polymorphisms and smoking.30 These interactions may explain some of the contradictions in the prior literature on ß2-adrenoceptor polymorphisms. Physical activity levels have plummeted in the general population in the United States over the last 2 decades,18 concurrent with a large increase in asthma incidence. Further study in larger populations is warranted to confirm if a sedentary lifestyle unmasks genetic risk for asthma in addition to contributing to obesity-related asthma.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by National Institutes of Health grants PE-11001, HL-07427, HL-03533 and CA-87969, and Allen and Hanburys Respiratory Institute Fellowship (Dr. Cooper).
In addition to National Institutes of Health funding, some of the authors receive pharmaceutical support for other projects.
Received for publication February 27, 2001. Accepted for publication May 17, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Thakkinstian, M. McEvoy, C. Minelli, P. Gibson, B. Hancox, D. Duffy, J. Thompson, I. Hall, J. Kaufman, T.-f. Leung, et al. Systematic Review and Meta-Analysis of the Association between {beta}2-Adrenoceptor Polymorphisms and Asthma: A HuGE Review Am. J. Epidemiol., August 1, 2005; 162(3): 201 - 211. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. F. Thomsen, C. S. Ulrik, K. O. Kyvik, K. Larsen, L. R. Skadhauge, I. Steffensen, and V. Backer The Incidence of Asthma in Young Adults Chest, June 1, 2005; 127(6): 1928 - 1934. [Abstract] [Full Text] [PDF] |
||||
![]() |
K Wickens, D Barry, A Friezema, R Rhodius, N Bone, G Purdie, and J Crane Obesity and asthma in 11-12 year old New Zealand children in 1989 and 2000 Thorax, January 1, 2005; 60(1): 7 - 12. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Iribarren, I. V Tolstykh, and M. D Eisner Are patients with asthma at increased risk of coronary heart disease? Int. J. Epidemiol., August 1, 2004; 33(4): 743 - 748. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |