|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Anesthesiology (Drs. McQuitty, Emala, and Eleff), Pediatrics (Dr. Levine), and Environmental Health Sciences (Dr. Hirshman), The Johns Hopkins Medical Institutions, Baltimore, MD; the Department of Internal Medicine (Drs. Hopkins-Price and Lawyer), Division of Pulmonary Medicine, The Southern Illinois University School of Medicine, Springfield, IL; the Division of Biostatistics (Dr. Hoh), The Joseph L. Mailman School of Public Health, Columbia University, New York, NY; and the Laboratory of Statistical Genetics (Dr. Ott), Rockefeller University, New York, NY.
Correspondence to: Charles Emala, MD, Department of Anesthesiology, Columbia University College of Physicians and Surgeons, 630 W. 168th St, PH 525, New York, NY 10032; e-mail: cwe5{at}columbia.edu
| Abstract |
|---|
|
|
|---|
Design: The current study used six distinct intragene markers within the ß2-adrenoceptor gene, and evaluated genetic linkage between the ß2-adrenoceptor and asthma, allergy, or methacholine airway hyperresponsiveness in eight multiplex families.
Patients: Forty-nine members of eight multiplex families with a high incidence of asthma.
Interventions: Phenotypes were characterized by history, physical examination, skin testing, pulmonary function tests, and methacholine inhalational challenge. Genetic loci were identified using restriction fragment length polymorphisms, denaturing gradient gel electrophoresis, and restriction enzyme digest of polymerase chain reaction-amplified fragments of the ß2-adrenoceptor gene.
Measurements and results: Nonparametric analysis using computer analysis software found no evidence for linkage between these markers within the ß2-adrenoceptor gene and asthma. Parametric exclusion analysis using a dominant inheritance model resulted in large negative lod scores (- 6.74, - 19.44, and - 49.9, respectively) for tight linkage between asthma, allergy, or methacholine airway hyperresponsiveness and these polymorphic markers.
Conclusions: These results indicate that asthma, allergy, and methacholine airway hyperresponsiveness are not linked to a dominant ß2-adrenoceptor gene with strong effect in these eight families with an inherited pattern of asthma.
Key Words: airway hyperresponsiveness allergy asthma ß2adrenoceptor gene methacholine polymorphism
| Introduction |
|---|
|
|
|---|
Since Szentivanyi10 proposed in 1968 that asthma may be due to an inherited or acquired deficit in ß-adrenoceptor function, much research has focused on the ß2-adrenoceptor and its signaling pathway. Several lines of evidence continue to suggest that the ß2-adrenoceptor may be abnormal in asthma, making the ß2-adrenoceptor gene an attractive candidate gene in this disease. Administration of ß-adrenoceptor antagonists increases airway tone and responsiveness in patients with asthma,11 but is without effect in normal individuals. Asthmatic patients require higher doses of ß-adrenoceptor agonists to achieve bronchodilation than do nonasthmatics.12 Bronchial or tracheal smooth muscle obtained at either autopsy or surgery from asthmatic patients show a deficit in ß-adrenoceptor function.13 14 15 16
A large number of polymorphisms or point mutations have been described in the human ß2-adrenoceptor gene. Lentes et al17 first reported a restriction fragment length polymorphism (RFLP) of this gene using the restriction enzyme Ban I. Another biallelic polymorphism was found by McQuitty et al18 in our laboratory using the restriction enzyme Fnu4HI, while Reihsaus et al19 found nine different point mutations within the coding region, four of which result in changes in amino-acid residues 16, 27, 34, and 164. Moreover, cells transfected with ß2-adrenoceptor complementary DNA containing the mutations at amino-acid positions 27 or 164 showed altered ß-adrenoceptor function.20
Potter et al21 studied the distribution of the Ban I polymorphisms in a population of 72 South Africans comparing asthmatic patients, allergic patients, allergic rhinitic patients, and nonallergic control subjects. Both Ban I alleles were present in the population, but the genotypes were found with similar frequencies in allergic and nonallergic subjects. Reihsaus et al19 directly sequenced the ß2-adrenoceptor gene in 51 unrelated asthmatic patients and 56 nonasthmatic patients and identified nine separate point mutations or polymorphisms, but found no significant difference in the frequency of alleles between the asthmatic and nonasthmatic patients. Familial asthma, however, may have a different mode of inheritance than nonfamilial asthma.
In a study of 56 family members of four Japanese asthmatic families, Ohe et al22 found a higher prevalence of asthma in family members who lacked the 3.1 kb Ban I RFLP, but this study lacked the power to exclude genotypic linkage to either methacholine responsiveness or allergy. A subsequent study23 by these investigators of 77 Japanese subjects also showed that the distribution of these alleles was not different between asthmatic and nonasthmatic subjects, but again was unable to exclude an association between the ß2-adrenoceptor gene and asthma, airway hyperresponsiveness, or allergy.
Because previous studies have been unable to either conclusively prove or disprove a causal relationship between the ß2-adrenoceptor gene and asthma, methacholine hyperresponsiveness, or allergy, we selected for study eight families with a high prevalence of asthma. Each of the families was specifically chosen for this study because most generations contained affected and unaffected individuals. We used six polymorphic markers found within the coding region of the ß2-adrenoceptor gene to analyze linkage of ß2-adrenoceptor alleles to asthma, allergy, and methacholine responsiveness in these highly selected multiplex families with asthma. With this experimental design, we were finally able to conclusively exclude linkage between the ß2-adrenoceptor gene and asthma, allergy, and airway hyperresponsiveness.
| Materials and Methods |
|---|
|
|
|---|
Clinical Evaluation
All subjects recruited for this study underwent a screening evaluation with a respiratory questionnaire, physical examination, assessment of allergy by skin testing, assessment of airway hyperresponsiveness by pulmonary function testing, and methacholine inhalational challenge.
A history of asthma was obtained using a questionnaire, which included questions about shortness of breath, cough, wheeze, chest tightness, as well as a history of asthma diagnosed by a physician and the use of current and past medications. A clinical diagnosis of asthma was confirmed after an interview and physical examination by a pulmonary physician. Pulmonary function tests were not used to classify patients as asthmatic or nonasthmatic. One patient with a questionable history of asthma was listed as unknown for the purpose of analysis.
Pulmonary Function Testing and Methacholine Challenge
Spirograms were performed in triplicate using a spirometer (Stead-Wells; Warren E. Collins; Braintree, MA). FEV1 and FVC were determined. The procedure was performed according to the recommendations of the American Thoracic Society.24
Current medications were not discontinued.
Methacholine challenge was performed using standard techniques.25
In brief, the subjects were instructed to take five maximal inspirations from functional residual capacity. Methacholine solutions (J.T. Baker; Phillipsburg, NJ) were nebulized for 0.6 s by triggering a breath-activated solenoid valve timing circuit (dosimeter). This apparatus delivered an average of 1.028 ± 0.222 mL (mean ± SEM) with every five breaths. After first inhaling a control saline solution, all subjects received increasing concentrations of methacholine (0.625, 1.25, 2.5, 5.0, 10.0, and 25.0 mg/mL). During the 5-min intervals between inhalations, spirometry was performed. When a
20% fall in FEV1 occurred, the challenge was terminated. Dose-response curves for methacholine were then constructed whereby the cumulative dose of methacholine required to produce a 20% fall in FEV1 (PC20). This value was used for analysis of methacholine reactivity as a quantitative trait in the determination of lod scores. Patients were considered to be methacholine positive if the PC20 was
8 mg/mL.
Allergy and Skin Testing
Skin tests were performed with the prick method using the following commercial allergens: timothy grass, mixed trees, mixed grasses, ragweed, alternaria, cladosporium, hormodendrum, Dermatophagoides farinae, Dermatophagoides pteronyssinus, cat, dog, and cockroach. Histamine hydrochloride (1 mg/mL) and allergen diluent served as controls. Papules > 3 mm were considered positive. Patients were considered to be allergic if they had a history of allergic rhinitis or allergic drug reactions or two or more positive skin test results. Five patients who had questionable histories of allergies were listed as unknown for the purpose of analysis.
DNA Isolation From Peripheral Blood
High-molecular-weight nuclear DNA was isolated from the leukocytes contained in 4 mL of peripheral blood of all subjects as previously described,26
and purified by centrifugation to equilibrium in a cesium chloride density gradient.27
Identification of Polymorphisms
Six polymorphisms within the human ß2-adrenoceptor gene were evaluated in genomic DNA. Two previously described RFLPs that occur in noncoding regions of the gene (BanI and BsoFI [or Fnu4HI])17
18
were identified by Southern blot analysis using a radiolabeled 2.0-kb ß2-adrenoceptor complementary DNA probe released by EcoRI digestion of pTF3 (American Type Culture collection).9
Denaturing gradient gel electrophoresis was used to identify a single-base polymorphism (A
C at base 1786)28
in the ß2-adrenoceptor gene. DNA was amplified by polymerase chain reaction (PCR) using a primer pair in which the sense primer included a 45 base GC-rich extension at the 5' end.29
PCR reactions (25 µL) were assembled using 250 ng genomic DNA, 20 mM Tris, pH 8.4, 50 mM KCl, 1.5 mM MgCl2, 0.1 mM each deoxynucleoside triphosphate, 1 µM each primer, and 1.25 U Taq polymerase. Forty cycles of PCR were performed with a denaturing temperature of 94°C, an annealing temperature of 60°C, and an extension temperature of 72°C. The primer pair (sense primer: 5'-cgcccgccgcgccccgcgcccgccccgccgcccccgcccgcagagcctgctgaccaagaataag-3' [italics indicates GC clamp]; antisense primer: 5'-ctgaaagaccctggagtagacgaa-3') corresponding to amino-acids 142224 of the human ß2-adrenoceptor sequence generated a 249-base pair (bp) fragment. Denaturing gradient gel electrophoresis was performed as previously described30
31
using a 50 to 80% gradient of denaturants (100% denaturant = 7 mol/L urea and 40% formamide).
Three additional polymorphisms in the 5' coding region of the human ß2-adrenoceptor gene were identified by restriction enzyme digestion of PCR-amplified DNA. The sense primer 5'-cccgggaacggcagcgccttcttgctggcaccccat-3' and the antisense primer 5'-cataagaatatgggcggccccaaagggcaccactg-3' amplified a 279-bp fragment corresponding to bases 1273
1551 of the human ß2-adrenoceptor gene sequence.28
The sense primer was designed with an intentional mismatch (in italics) of a single base, three bases from the 3' end of the primer. This allowed for the creation of an NcoI restriction enzyme recognition site if base 1309 were a G.28
Restriction enzyme digests of the PCR product with Fnu4HI or MaeI were used separately to detect the presence or absence of a C
G or G
A polymorphism at bases 1342 or 1515, respectively.28
These three polymorphisms corresponded to single-base polymorphisms that encode amino-acid changes of Arg
Gly at position 16, Gln
Glu at position 27, and a neutral polymorphism at Leu 84.19
Linkage Analysis: Nonparametric Analysis
The Genehunter program (Laboratory of Statistical Genetics, Rockefeller University, New York, NY)32
was used to evaluate genetic linkage between asthma and six tightly linked markers inside the ß2-adrenoceptor gene. Since this initial analysis yielded no significant evidence for linkage (empirical significance level, p = 0.29) we wanted to evaluate the evidence against linkage in a parametric analysis.
Linkage Analysis: Parametric Exclusion Analysis
Asthma and Allergy Phenotype:
To compute lod scores, we employed the dominant inheritance model originally proposed by Cookson et al.33
Although these phenotypes may be complex traits, it is appropriate statistical practice to employ a suitable monogenic analysis model even though a trait may be multifactorial.34
35
The model postulates penetrances of 0.01 and 0.99 for nongenetic and genetic cases, respectively, with a disease allele frequency of 0.20. Within this model, a 7-point linkage analysis was carried out (Mlink program; Laboratory of Statistical Genetics, Rockefeller University36
) between the hypothesized disease locus and the six intragenetic marker loci, which were assumed to be tightly linked with each other. Lod scores were computed for recombination fraction values,
, between the trait and the cluster of marker loci. Additional preliminary analyses were performed using two other models, the one originally proposed by Cookson et al33
but applied to individuals expressing the phenotypes only, and the marginal recessive and dominant models derived from the two-locus model proposed by Folster-Holst et al.37
Methacholine Reactivity Scores:
Using the methacholine reactivity values (PC20) as quantitative trait loci (QTL), we first estimated means and variances from the data. Initial analyses were done assuming independence of all observations (family members) and assuming a mixture of two distributions. The resulting estimates of means and variances were then used as starting values in the iterative estimation for these parameter values. This latter estimation was carried out with the Ilink program (Laboratory of Statistical Genetics, Rockefeller University)36
and allowed for the relationships among family members. The estimated mean for heterozygotes was essentially equal to that for homozygous normal individuals, that is, the data suggested a recessive mode of inheritance for the methacholine QTL. With these estimates, lod scores between the QTL and the set of six tightly linked markers were again calculated in a 7-point analysis with the recombination fraction,
= 0.
| Results |
|---|
|
|
|---|
8 mg/mL). Twenty-eight patients (57%) exhibited allergy.
|
|
|
1551 (279 bp) of the human ß2-adrenoceptor gene. An allele lacking a polymorphism at base 1309 (ie, guanine) did not generate an NcoI site and was arbitrarily referred to as R. A polymorphism at base 1309 (ie, adenosine) created a restriction site for NcoI and was arbitrarily referred to as J. The fifth polymorphism was identified by the presence or absence of a Fnu4HI restriction site in the 279-bp PCR product due to a cytosine vs guanosine residue, respectively, at base 1342, and were arbitrarily referred to as T and U, respectively. The sixth polymorphism was identified by the presence or absence of a MaeI restriction site in the 279-bp PCR product due to an adenosine vs guanosine residue, respectively, at base 1515, and were arbitrarily referred to as K and E, respectively.
Linkage Analyses
An initial nonparametric analysis was performed using the Genehunter program32
to evaluate linkage between asthma and markers within the ß2-adrenoceptor gene. No evidence of linkage was found (empirical significance level, p = 0.29); therefore, parametric analyses were performed to exclude linkage between the ß2-adrenoceptor gene and asthma. Assuming a dominant model of inheritance as originally proposed by Cookson et al33
and
= 0, a lod score of - 6.74 was obtained that excludes the candidate gene as the trait locus with odds of 1:5.5 x 106 (Table 2
). Lod scores remained < - 2 for recombination fractions up to
= 0.12.
|
= 0, which excludes the trait locus with odds of 1:2.8 x 1019 (Table 2)
. Lod scores remained < - 2 for recombination fractions up to
= 0.25.
Analysis of PC20 scores as QTL was performed. Lod scores between the QTL and the markers within the ß2-adrenoceptor gene with
= 0 was equal to - 49.9, which excludes the ß2-adrenoceptor gene as a trait locus with odds of 1:7.9 x 1049 (Table 2)
. Lod scores remained < - 2 for recombination fractions up to
= 0.09.
| Discussion |
|---|
|
|
|---|
Rationale for Evaluating the ß2-Adrenoceptor Gene as a Candidate Gene in Asthma, Allergy, and Methacholine Airway Hyperresponsiveness
There are many reasons to evaluate the ß2-adrenoceptor gene as a candidate gene in asthma. Clinical and laboratory functional data show that ß2-adrenoceptor dysfunction is a hallmark of asthma. ß2-Adrenergic agonists are a primary treatment of asthma, and numerous in vitro studies in man13
14
15
and animal models38
of asthma show ß2-adrenoceptor dysfunction in airway smooth muscle. Additionally, several studies5
7
39
40
41
have suggested the existence of genetic linkage between asthma, allergy, and airway hyperresponsiveness and a region of human chromosome 5q that includes the ß2-adrenoceptor gene. These combinations of functional and genetic data make the ß2-adrenoceptor gene an attractive candidate gene for asthma, particularly within families in which a genetic cause of asthma is suspected.
Polymorphisms of the ß2-Adrenoceptor Gene and Asthma
Previous genetic evaluations of the ß2-adrenoceptor gene in patients with asthma identified several sequences that alter receptor activity. Mutation of Thr164 to Ile164 leads to altered coupling to adenylyl cyclase,20
mutation of Arg16 to Gly16 leads to enhanced receptor downregulation,42
while mutation of Gln27 to Glu27 leads to a resistance to receptor downregulation.42
Previous attempts to correlate ß2-adrenoceptor polymorphisms with the severity of asthma or responses to ß-agonist therapy have been inconclusive. Turki et al43
correlated the Gly16 allele, which is associated with enhanced receptor downregulation with nocturnal asthma, and Martinez et al44
found that children who were homozygous for Arg16 were more responsive to inhaled albuterol. In contrast, Lipworth et al45
demonstrated that the protection afforded by inhaled formoterol to a subsequent methacholine challenge was not influenced by the ß2-adrenoceptor genotype. These previous studies evaluated an association between ß2-adrenoceptor genotypes and either the severity of asthma or the response to ß-agonist therapy. In contrast, the present study used ß2-adrenoceptor polymorphisms as markers to compare the inheritance of ß2-adrenoceptor alleles from parent to child with the inheritance of a disease phenotype. The polymorphisms were not used to access severity of disease, but instead were used in an attempt to establish linkage between the inheritance of these polymorphisms and the inheritance of asthma, allergy, or methacholine hyperresponsiveness established by clinical criteria.
Approaches to Identify Genetic Loci Associated With Asthma
Three types of genetic studies have been performed in an attempt to determine the genetic basis of asthma and allergy. These approaches include genome-wide linkage analyses using widely spaced genetic markers, more focused analyses using markers within defined chromosomal regions, or narrowly focused searches using markers within candidate genes. Genome-wide linkage studies have identified a number of genetic loci with potential linkage to asthma, bronchial hyperresponsiveness, allergy, or elevated IgE levels.5
7
39
40
41
Genetic analyses focusing down from a genome-wide search to specific chromosomal regions have found genetic linkages between areas of chromosome 5q and bronchial hyperresponsiveness. Chromosome 5q has been of particular interest because it contains genes encoding numerous cytokines and the ß2-adrenoceptor. In studies of Dutch7
46
and Japanese5
populations, linkage of asthma to 5q was found, whereas linkage of asthma to 5q was not confirmed in two Australian studies.41
47
Conflicting results may in part be due to difficulties in defining the asthmatic phenotype. Interpretations of phenotypes between studies likely differ for many reasons: (1) a limited understanding of the interplay between allergy, asthma, and airway hyperresponsiveness; (2) asthma may be a collection of diseases with a common clinical manifestation of wheezing; (3) although allergy commonly accompanies asthma, not all asthmatics are allergic; (4) although airway hyperresponsiveness is considered a hallmark of asthma, this may be altered by therapy; (5) not all patients with airway hyperresponsiveness exhibit clinical asthma; and (6) studies of different ethnic groups may yield different genetic linkage results.
Although several previous studies5 21 22 have used a candidate gene approach using intragenic polymorphic markers within the ß2-adrenoceptor gene to search for linkage between this gene and asthma, bronchial hyperresponsiveness, or allergy, none of the studies had the statistical power to make any definitive statements about the absence of linkage. In contrast, our study, using a relatively small sample size, strongly demonstrates that no linkage exists between a dominant ß2-adrenoceptor gene with strong effect and asthma, allergy, or methacholine airway hyperresponsiveness based on our very negative lod scores. Although large samples are required to demonstrate the presence of linkage, small sample sizes can be used to demonstrate the absence of linkage.48
This study does not agree with the study of Ohe et al,22 who found linkage between the biallelic Ban I RFLP and asthma in four Japanese families,22 but does agree with a subsequent study from the same group23 and with a study of Potter et al,21 which failed to find linkage between the ß2-adrenoceptor gene and asthma or allergy. The current study also agrees with a study by Dewar et al,49 in which no association was found between the arginine-glycine 16 or glutamine-glutamate 27 ß2-adrenoceptor polymorphism and asthma or allergy in families with asthma. Furthermore, our study agrees with a recent study performed in a large inbred kindred of Hutterites in which no linkage was found between the Arg16Gly or Gln27Glu polymorphisms and asthma.50
Genetic linkage studies of asthma have certain limitations related to defining the phenotype.51 More specifically, it has not been possible to develop a quantitative model for a phenotype that would improve statistical linkage analysis. The difficulty in defining the asthmatic phenotype may in part explain the inconsistent results of other studies that sought to link asthma, airway hyperreactivity, or allergy to gene loci.
Despite these limitations, remarkable progress has been made in genome-wide searches for genetic loci linked to the phenotypes of asthma, allergy, and bronchial hyperresponsiveness.40 41 Consistent with the current study, these genome-wide searches have suggested that asthma is likely a polygenic disorder modified by environmental influences. Although mutations in the ß2-adrenoceptor gene alone cannot account for asthma, airway hyperresponsiveness, or allergy in the vast majority of affected subjects, mutations in the ß2-adrenoceptor gene may influence the severity of the disease in those individuals who have the necessary complement of genetic and/or environmental characteristics to manifest the disease.42 52
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by National Institutes of Health grants HL-45974, HL-02693, and HG-00008.
Received for publication November 22, 1999. Accepted for publication November 9, 2000.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |