Polymorphisms in A Disintegrin and Metalloprotease 33 (ADAM33) Predict Impaired Early-Life Lung Function
http://www.100md.com
美国呼吸和危急护理医学 2005年第7期
Academic Division of Medicine and Surgery South
Centre for Integrated Genomic Medical Research, University of Manchester, Manchester
Infection, Inflammation
Repair
Human Genetics Divisions, School of Medicine, University of Southampton, Southampton General Hospital, Southampton, United Kingdom
ABSTRACT
Rationale: Asthma commonly originates in early life in association with impaired lung function, which tracks to adulthood. Objectives: Within the context of a prospective birth cohort study, we investigated the association between single nucleotide polymorphisms (SNPs) in a disintegrin and metalloprotease 33 (ADAM33) gene and early-life lung function. Methods: Children were genotyped for 17 SNPs in ADAM33. Lung function at age 3 (n = 285) and 5 years (n = 470) was assessed using plethysmographic measurement of specific airway resistance (sRaw). At age 5, we also measured FEV1. SNPs were analyzed individually using logistic regression, followed by linkage disequilibrium mapping to identify the causal locus. Main Results: Carriers of the rare allele of F+1 SNP had reduced lung function at age 3 years (p = 0.003). When the recessive model was considered, four SNPs (F+1, S1, ST+5, V4) showed association with sRaw at age 5 years (p < 0.04). Using linkage disequilibrium mapping, we found evidence of a significant causal location between BC+1 and F1 SNPs, at the 5' end of the gene. Four SNPs were associated with lower FEV1 (F+1, M+1, T1, and T2; p 0.04). The risk of transient early wheezing more than doubled among children homozygous for the A allele of F+1 (odds ratio, 2.39; 95% confidence intervals, 1.18eC4.86; p = 0.02), but there was no association between any SNP and allergic sensitization or physician-diagnosed asthma. Conclusions: Polymorphisms in ADAM33 predict impaired early-life lung function. The functionally relevant polymorphism is likely to be at the 5' end of the gene.
Key Words: ADAM33; asthma; genetics; lung function
Asthma is the most common chronic disease of childhood (1), and on a worldwide scale has dramatically increased in prevalence over the last 30 years (2). Twin and other family-based studies confirm a strong genetic component of asthma (3). The recent identification of a putative asthma susceptibility gene underlying bronchial hyperresponsiveness, a disintegrin and metalloprotease 33 (ADAM33) (4), emphasizes the possibility that a primary end-organ determinant together with the systemic immune response underlie the manifestation of asthma.
Associations between polymorphisms in ADAM33 and asthma were reported in African Americans, Hispanics, and whites in the United States, and Dutch populations, but none of the associated single nucleotide polymorphisms (SNPs) were common to all populations studied (5). One SNP was subsequently associated with excess decline in lung function in Dutch adults with asthma (6), and associations with asthma were reported in German (7) and Korean (8) populations. A recent study of children with asthma identified only modest associations with ADAM33 SNPs (9), whereas no association was seen in Puerto Rican or Mexican populations (10). Inconsistencies in genetic association studies are common, particularly for genes in which the functionally relevant polymorphism has not been established. Such inconsistencies could be attributed to genetic heterogeneity between populations, differences in phenotype definition, and insufficient power. However, it is worth emphasizing that the support for a role of ADAM33 in asthma also comes from mRNA expression analysis (11, 12).
The majority of asthma originates in early life in association with impaired lung function, which tracks into later childhood (13) and adulthood (14eC16). Therefore, it is important to elucidate factors influencing lung function in early life. Although it is often difficult to assess standard lung function in preschool children, measurement of specific airway resistance (sRaw) is feasible (17, 18). sRaw is a measure of airway caliber independent of lung size, and narrowing of the airways results in elevated values. Bronchodilator response measured by sRaw has been shown to be useful in identifying preschool children with asthma, with sensitivity at least as good as spirometry in older children (19).
We have analyzed 17 SNPs in the ADAM33 gene in a prospective population-based birth cohort (the Manchester Asthma and Allergy Study) (18, 20, 21) in which assessment of lung function was completed at age 3 and 5 years. Given that the functionally relevant polymorphisms have not been clearly identified, we used the recently described method of association mapping by linkage disequilibrium (LD) (22, 23). This approach uses single-marker tests within a composite likelihood framework to identify the most likely location of the causal SNP on the LD map, which assigns a location in LD units (LDU) for each marker (24). Our hypothesis was that variation in the ADAM33 gene contributes to impaired lung function during the first few years of life. Some of the results of this study were reported in the form of an abstract (25).
METHODS
A detailed description of methods is presented in the online supplement. The Manchester Asthma and Allergy Study is a population-based birth cohort study (18, 20, 21). Participants were recruited prenatally (20), and attended review at age 3 and 5 years (± 4 weeks). Study was approved by the Local Research Ethics Committee. Informed consent was obtained from all parents.
Outcomes
sRraw was measured using whole-body plethysmography as previously described (18) by a single-step procedure from the simultaneously measured changes of respiratory flow and plethysmographic pressure, omitting the measurement of thoracic gas volume. Measurements were performed during normal tidal breathing, and we used the mean of three measurements of effective sRaw.
Dynamic lung volumes and expiratory flow were measured using incentive animation software. The test was repeated at intervals of 30 seconds until three technically acceptable traces were obtained and the highest FEV1 was recorded.
Symptoms.
A validated questionnaire (26) was interviewer-administered to collect information on parentally reported symptoms and physician-diagnosed illnesses. According to parentally reported wheeze at two follow-ups, children were assigned as follows (27): no wheezing (never reported wheeze; n = 241), transient early wheeze (wheezing during the first 3 years, no wheezing in the previous 12 months at age 5; n = 104), late-onset wheezing (no wheeze during the first 3 years, wheezing in the previous 12 months at age 5 years; n = 25), and persistent wheezing (wheezing during the first 3 years, wheezing in the previous 12 months at age 5; n = 63).
Allergic sensitization was ascertained at age 5 years by measurement of specific serum IgE to common inhalant and food allergens (ImmunoCAP; Pharmacia, Uppsala, Sweden). Sensitization was defined as allergen-specific IgE greater than 0.2 kU/L.
Genotyping
Seventeen SNPs previously associated with an asthma phenotype, spanning 10 kb of ADAM33 were selected for genotyping (Table 1). We sought to provide good coverage of the gene and to examine SNPs with a range of allele frequencies. SNPs were genotyped using TaqMan allelic discrimination assay (28) (Applied Biosystems, Foster City, CA). Polymerase chain reaction was performed according to the manufacturer's instructions in 5-e reaction volumes in 96-well plates and contained 10 ng genomic DNA per reaction. Four control samples were included on each polymerase chain reaction plate (one negative and three of known genotype). One SNP (S2) was genotyped using Sequenom (see online supplement for method).
Statistical Analysis
Each SNP was tested for deviation from Hardy-Weinberg equilibrium using a 2 test. Genotype comparisons of symptoms were performed using the 2 test. sRaw followed a log-normal distribution, hence was subject to a loge transformation. Univariate linear regression analysis was used to compare sRaw between genotype groups for each SNP. For SNPs showing a trend for association with lung function by genotype (p < 0.1), dominant and recessive genetic models were tested. Analysis was performed in Stata 8 (Stata Corporation, College Station, TX).
LDU Map Construction and Association Mapping
The LDU map was created using the LDMAP program (22eC24). These metric maps have additive distances and are analogous to linkage maps, but superior in representing the pattern of LD rather than just recombination (29, 30). Composite likelihood analysis of single-marker tests was used to identify the most likely location of the causal SNP on the LDU map, which avoids the heavy Bonferroni correction through multiple testing.
RESULTS
Participant Flow
The study profile is shown in Figure 1; children who were prenatally randomized to the environmental control study and 26 children who were not of mixed European descent were excluded from this analysis. Acceptable sRaw results were obtained in 285 children at age 3 and 470 at age 5 years (both 3- and 5-year lung function data were available in 280 children; Figure 1). By age 5 years, 34% of children were sensitized, 38.6% had at least one episode of wheeze, 21.7% had wheezed in the last 12 months, and 20% had received a physician diagnosis of asthma at some time during the first 5 years of life. In keeping with our previous report (18), symptomatic children had significantly higher sRaw values compared with asymptomatic children (e.g., wheeze ever, p < 0.0001).
Genotyping success rates were high (97.5eC99.5%), and allele frequencies (Table 1) were consistent with those seen in other mixed European cohorts (4, 5). Sixteen of the 17 SNPs were in Hardy-Weinberg equilibrium at the p > 0.05 significance level. SNP V-3 marginally deviated from Hardy-Weinberg equilibrium (p = 0.03) because of an observed increase in heterozygotes (252 vs. 229 expected), but was included in the analysis because other quality-control checks (including genotyping of samples with known genotypes) suggested this deviation was not caused by genotyping error.
sRaw at Age 3 and 5 Years
F+1 was the only SNP associated with sRaw at age 3 years by genotype (p = 0.01; Table 1, Figure 2a). Carriers of the A allele had significantly higher sRaw values (i.e., reduced lung function) than homozygotes for the G allele (geometric mean [GM], 95% confidence intervals [CI]; kPa · 1.12, 1.09eC1.16 vs. 1.04, 1.01eC1.08; p = 0.003). The F+1 polymorphism explained 3% of the variance in the lung function at age 3.
At age 5 years, four SNPs (F+1, S1, ST+5, V4) showed association with sRaw by genotype (Table 1, Figure 2b). The strength of association increased substantially when a recessive model was considered for all four SNPs (Table 2).
LD Mapping across the ADAM33 Gene
The LD map describes the underlying structure of LD in the form of a metric map with additive distances in LDU. Therefore, every SNP was assigned two locations: one in kilobases and the other in LDU (Table 3). Figure 3 presents graphically the block-step structure of the ADAM33 region by plotting the LDU locations against the kilobase map. As shown in Figure 3, there is one main block of high LD with invariant LDU spanning 3.7 kb (SNPs M+1 to V2), which is flanked by two short SNP blocks.
For association mapping, we used the method developed by Maniatis and colleagues (22, 23) (see online supplement for the detailed results). Using different subhypotheses of the model, the analysis first determines whether there is adequate power to test for association in the genomic region, and then provides the estimated causal location. Our calculations suggested sufficient power to localize a causal polymorphism within the ADAM33 gene that is associated with lung function at age 5 years using the kilobase map under an additive model (p < 0.001). Tests for a disease determinant provided a significant causal location at 1.34 kb (p < 0.05), which is positioned in a step between BC+1 and F1 SNPs (Table E2, online supplement).
Analysis of sRaw at age 5 years under the recessive model increased the error variance and decreased the power by approximately 50%. Nevertheless, the analysis suggested sufficient power to localize a causal polymorphism and showed a significant association within the region (p < 0.005), but no significant association at any location (p > 0.05).
We also examined lung function at age 3 years, but did not have sufficient power to localize a causal polymorphism, probably because of a smaller sample size (285, compared with 470 at age 5).
Other Lung Function Measures at Age 5 Years
FEV1.
Four SNPs were associated with FEV1 by genotype (Table 1). The rare alleles of four SNPs were associated with lower FEV1 percent-predicted values: F+1—GM, 106; 95% CI, 101eC110 versus GM, 111; 95% CI, 109eC113; p = 0.04; M+1—GM, 106; 95% CI, 102eC110 versus GM, 112; 95% CI, 110eC114; p = 0.005; T1—GM, 107; 95% CI, 103eC111 versus GM, 112; 95% CI, 110eC114; p = 0.01; and T2—GM, 107, 95% CI, 103eC111 versus GM, 112; 95% CI, 110eC114; p = 0.02 (Table E3).
Symptoms
Five SNPs associated with impaired lung function were also associated with early-life wheezing. Children who were homozygous for the A allele of F+1 were more than twice as likely to have wheezed in the first year of life (odds ratio [OR], 2.56; 95% CI, 1.36eC4.82; p = 0.004). The presence of the A allele of M+1 (OR, 2.02; 95% CI, 1.25eC3.26; p = 0.004), the C allele of T1 (OR, 2.05; 95% CI, 1.28eC3.29; p = 0.003) and the T allele of T2 (OR, 2.01; 95% CI, 1.26eC3.21; p = 0.004), and homozygosity for the G allele of ST+5 (OR, 1.93; 95% CI, 1.24eC2.97; p = 0.003) were all associated with an increased risk of infant wheeze.
We assigned 433 children into one of four wheeze phenotypes (37 children could not be classified, e.g., no history of wheeze, but receiving asthma medication). Among children who had never wheezed, there was a smaller proportion homozygous for the A allele of F+1 (7.5%) compared with transient early wheezers (16.2%, p = 0.01), and persistent wheezers (14.3%, p = 0.09).
Using never wheezers as a reference, the risk of transient early wheezing more than doubled among children homozygous for the A allele of F+1 (OR, 2.39; 95% CI, 1.18eC4.86; p = 0.02). There was no association between physician-diagnosed asthma by age 5 years and any of the SNPs.
Allergic Sensitization
No association was observed between ADAM33 SNPs and allergic sensitization at age 5 years whether using specific IgE or skin-prick test results.
DISCUSSION
The current study provides the first evidence that the association of ADAM33 with asthma may be mediated via its effects on lung function in childhood. Our data support the hypothesis that reduced early-life lung function is in part a genetically determined trait. Polymorphisms were associated with reduced lung function at both 3 and 5 years of age and the more significant associations at age 5 years may reflect the increased sample size. All of the SNPs associated with altered lung function in the current study have previously been associated with asthma and bronchial hyperresponsiveness in adults in U.K., U.S., Dutch, and German populations (4, 5, 7). This study was unable to show any association between ADAM33 SNPs and allergic sensitization. This suggests that separate genetic factors contribute to disordered airway function and the immunologic component of asthma (31, 32), and further consolidates the role of ADAM33 in primary end-organ pathology rather than the systemic immune response.
Birth cohort studies provide an opportunity to observe the development of wheezing symptoms throughout childhood. We have shown an association between polymorphisms in ADAM33 and transient early wheeze, with a trend for persistent wheeze but not late-onset wheeze. Our findings are consistent with the notion that ADAM33 polymorphisms may be associated with smaller airways. The contribution of airway size to the expression of symptoms may be greater earlier in childhood. This supports the hypothesis that childhood asthma is a heterogeneous disease, and provides additional evidence that different genetic and environmental factors may be important in the different wheezing phenotypes. Thus, ADAM33 may be a "lung function gene" rather than an "asthma gene," variants of which increase the risk of wheeze in early life.
Although the selection of the 17 SNPs in the current study was based on the results of previous studies, they represent only a proportion of the total variation of the gene. The regression analysis showed that F+1 is the most strongly associated SNP, and even when we applied an overconservative Bonferroni correction, F+1 remained nominally significant (see online supplement). However, single regression analysis tests each SNP assuming it is the functional polymorphism. The LD method we present in this study is based on composite likelihood where all the information from all single-marker tests is considered simultaneously, removing the need for any correction for multiple tests. The maximum likelihood estimation is an estimate of the causal location under the assumption that the functional polymorphism may not be one of those tested. Using simulated (22) and real (23) data, we have previously shown that greater power is achieved when mapping within an LDU map compared with a map in kilobases, especially in a densely typed region (22). This analysis identified that the ADAM33 locus is associated with lung function and that the functional polymorphism most likely lies at the 5' end of the gene (between BC+1 and F1). The original positional cloning study identified a higher density of SNPs at the 3' end of the gene, and so replication studies have focused on this region, with variable results (6eC10). Although it is plausible that more than one functional polymorphism exists and that these may have different effects on lung function and asthma, there is also biological evidence that the majority of splice variation occurs at the 5' end of the gene (33).
ADAM proteins are transmembrane multidomain proteins with potential for cell adhesion, fusion, communication, signaling, and protease activities (34). The exact function of ADAM33 remains unknown and the functional nature of the associated polymorphisms is yet to be established. Expression of ADAM33 mRNA has been detected in all tissues except liver, with high levels in lung and placenta (35). Gene expression studies indicate that ADAM33 RNA is expressed in lung fibroblasts and bronchial smooth muscle, but not in bronchial epithelial cells or lymphocytes (4, 12, 33). Furthermore, studies of alternative splicing of ADAM33 transcripts suggest that the 3' domains (the epidermal growth factoreClike, transmembrane, and cytoplasmic domains corresponding to SNPs Q+1 through V4) are present in all transcripts, but that the variability in exon usage occurs at the 5' end of the gene (the metalloprotease and prodomains, corresponding to SNPs BC+1 through M+1) (33). It is tempting to speculate that the functional SNP located between F1 and BC+1 may alter which isoform is expressed. Recent work from our laboratory failed to find differences in total levels of ADAM33 transcripts, or relative expression of different splice forms, in bronchial biopsies from normal individuals and individuals with asthma (36). However, these studies were performed on a relatively small number of subjects without genotype information for the relevant polymorphisms and will need to be followed up in larger cohorts where genotypic information is available. Clearly, additional research is required to establish the exact function of ADAM33 and how it is altered by polymorphisms within the gene.
This demonstration that ADAM33 polymorphisms influence lung function in early life may support the hypothesis that epithelial-mesenchymal dysfunction in the airways may predispose individuals toward asthma (37), being present in early childhood before asthma becomes clinically expressed (38, 39). In this context, it is interesting to note that ADAM33 expression has been shown to occur during murine embryonic development, suggesting that it may regulate tissue morphogenesis (40). Preliminary studies have found that murine ADAM33 expression is induced in embryonic lungs at the start of branching morphogenesis, increases with gestation and remains present into adulthood (41). This raises the possibility that polymorphisms which modulate ADAM33 expression and/or function may modulate lung growth and development. We speculate that, in addition to being "remodeled" as a consequence of asthma, the airways could be "premodeled" as one of the prerequisites for subsequent development of wheeze as a genetically determined trait (18).
Sex, maternal smoking, and high allergen exposure in sensitized individuals are also associated with early-life lung function (42). We have undertaken a multivariate analysis that includes ADAM33 polymorphisms and these three factors and have shown that the association of ADAM33 polymorphisms and preschool lung function is independent of sex, smoking, or allergen exposure in sensitized individuals (see online supplement). Consistent with this gene influencing lung function, a recent study has shown that SNPs in ADAM33 predict a more rapid decline in lung function in individuals with chronic asthma followed over 20 years (6).
Thorough knowledge of lung physiology in early life and its genetic component could refine our understanding of the causes of asthma. If the pattern of lung function in early childhood is predictive of that in adult life and can help identify individuals at risk, this helps our understanding of the causes of wheezing illness and targeting at-risk individuals for primary prevention strategies.
Acknowledgments
The authors thank the children and parents of the Manchester Asthma and Allergy Study for their continued dedication to the study.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Kaur B, Anderson HR, Austin J, Burr M, Harkins LS, Starchan DP, Warner JO. Prevalence of asthma symptoms, diagnosis and treatment in 12eC14 year old school children across Great Britain (International Study of Asthma and Allergies in Childhood, ISAAC, UK). BMJ 1998;316:118eC124.
Omran M, Russell G. Continuing increase in respiratory symptoms and atopy in Aberdeen schoolchildren. BMJ 1996;312:34.
Duffy DL, Martin NG, Battistutta D, Hopper JL, Mathews JD. Genetics of asthma and hay fever in Australian twins. Am Rev Respir Dis 1990;42:1351eC1358.
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, et al. Association of ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;418:426eC430.
Howard TD, Postma DS, Jongepier H, Moore WC, Koppelman GH, Zheng SL, Xu J, Bleecker ER, Meyers DA. Association of a disintegrin and metalloprotease 33 (ADAM33) gene with asthma in ethnically diverse populations. J Allergy Clin Immunol 2003;112:717eC722.
Jongepier H, Boezen HM, Dijkstra A, Howard TD, Vonk JM, Koppleman GH, Zheng SL, Meyers DA, Bleecker ER, Postma DS. Polymorphisms of the ADAM33 gene are associated with accelerated lung function decline in asthma. Clin Exp Allergy 2004;34:757eC760.
Werner M, Herbon N, Gohlke H, Altmuller J, Knapp M, Heinrich J, Wjst M. Asthma is associated with single-nucleotide polymorphisms in ADAM33. Clin Exp Allergy 2004;34:26eC31.
Lee JH, Park HS, Park SW, Jang AS, Uh ST, Rhim T, Park CS, Hong SJ, Holgate ST, Holloway JW, et al. ADAM33 polymorphism: association with bronchial hyper-responsiveness in Korean asthmatics. Clin Exp Allergy 2004;34:860eC865.
Raby BA, Silverman EK, Kwiatkowski DJ, Lange C, Lazarus R, Weiss ST. ADAM33 polymorphisms and phenotype associations in childhood asthma. J Allergy Clin Immunol 2004;113:1071eC1078.
Lind DL, Choudry S, Ung N, Ziv E, Avila PC, Salari K, Ha C, Lovins EG, Coyle NE, Nazario S, et al. ADAM33 is not associated with asthma in Puerto Rican or Mexican populations. Am J Respir Crit Care Med 2003;168:1312eC1316.
Garlisi CG, Zou J, Devito KE, Tian F, Zhu FX, Liu J, Shah H, Wan Y, Motasim Billah M, Egan RW, et al. Human ADAM33: protein maturation and localization. Biochem Biophys Res Commun 2003;301:35eC43.
Umland SP, Garlisi CG, Shah H, Wan Y, Zou J, Devito KE, Huang WM, Gustafson EL, Ralson R. Human ADAM33 mRNA expression profile and post- transcriptional regulation. Am J Respir Cell Mol Biol 2003;29:571eC582.
Turner SW, Palmer LJ, Rye PJ, Gibson NA, Judge PK, Cox M, Young S, Goldblatt J, Landau LI, LeSouef PN. The relationship between infant airway function, childhood airway responsiveness and asthma. Am J Respir Crit Care Med 2004;169:921eC927.
Phelan PD, Robertson CF, Olinski A. The Melbourne Asthma Study: 1964eC1999. J Allergy Clin Immunol 2002;109:189eC194.
Rasmussen F, Taylor DR, Flannery EM, Cowan JO, Greene JM, Herbison GP, Sears MR. Risk factors for airway remodelling in asthma manifested by a low postbronchodilator FEV1/vital capacity ratio: a longitudinal population study from childhood to adulthood. Am J Respir Crit Care Med 2002;165:1480eC1488.
Sears MR, Greene JM, Willan AR, Wiecek AM, Taylor DR, Flannery EM, Cowan JO, Herbison GP, Silva PA, Poulton R. A longitudinal, population-based, cohort study of childhood asthma followed to adulthood. N Engl J Med 2003;349:1414eC1422.
Klug B, Bisgaard H. Measurement of specific airway resistance by plethysmography in young children accompanied by an adult. Eur Respir J 1997;10:1599eC1605.
Lowe L, Murray CS, Custovic A, Simpson BM, Kissen PM, Woodcock A. Specific airway resistance in three year-old children: a prospective cohort study. Lancet 2002;359:1904eC1908.
Nielsen KG, Bisgaard H. Discriminative capacity of bronchodilator response measured with three different lung function techniques in asthmatic and healthy children aged 2 to 5 years. Am J Respir Crit Care Med 2001;164:554eC559.
Simpson BM, Custovic A, Simpson A, Hallam CL, Walsh D, Marolia H, Campbell J, Woodcock A. NAC Manchester Asthma and Allergy Study (NACMAAS): risk factors for asthma and allergic disorders in adults. Clin Exp Allergy 2001;31:391eC399.
Custovic A, Simpson BM, Simpson A, Kissen P, Woodcock A. Effect of environmental manipulation in pregnancy and early life on respiratory symptoms and atopy during first year of life: a randomized trial. Lancet 2001;358:188eC193.
Maniatis N, Collins A, Gibson J, Zhang W, Tapper W, Morton NE. Positional cloning by linkage disequilibrium. Am J Hum Genet 2004;74:846eC855.
Maniatis N, Morton NE, Gibson J, Xu C-F, Hosking LK, Collins A. The optimal measure of linkage disequilibrium minimizes error in association mapping of affection status. Hum Mol Genet 2005;14:145eC153.
Maniatis N, Collins A, Xu CF, McCarthy LC, Hewett DR, Tapper W, Ennis S, Ke X, Morton NE. The first linkage disequilibrium (LD) maps: delineation of hot and cold blocks by diplotype analysis. Proc Natl Acad Sci USA 2002;99:2228eC2233.
Simpson A, Jury F, Custovic A, Holloway JW, Woodcock A, Cakebread JA, Lowe LA, Ollier WER, Holgate ST, John S. Manchester Asthma and Allergy Study: polymorphisms in ADAM33 predict lung function at age 5 years. J Allergy Clin Immunol 2004;113:S340.
Pearce N, Weiland S, Keil U, Langridge P, Anderson HR, Strachan D, Bauman A, Young L, Gluyas P, Ruffin D, et al. Self-reported prevalence of asthma symptoms in children in Australia, England, Germany and New Zealand: an international comparison using the ISAAC protocol. Eur Respir J 1993;6:1455eC1461.
Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan WJ. Asthma and wheezing in the first six years of life. N Engl J Med 1995;332:133eC138.
Livak KG. Alleic discrimination using fluoregenic probes and the 5' nuclease assay. Genet Anal 1999;14:143eC149.
Tapper WJ, Maniatis N, Morton NE, Collins A. A metric linkage disequilibrium map of a human chromosome. Ann Hum Genet 2003;67:487eC494.
Ke X, Hunt S, Tapper W, Lawrence R, Stavrides G, Ghori J, Whittaker P, Collins A, Morris AP, Bentley D, et al. The impact of SNP density on fine-scale patterns of linkage disequilibrium. Hum Mol Genet 2004;13:577eC588.
Gerrard JW, Vickers P, Gerrard CD. The familial incidence of allergic disease. Ann Allergy 1976;36:10eC15.
von Mutius E, Nicolai T. Familial aggregation of asthma in a South Bavarian population. Am J Respir Crit Care Med 1996;153:1266eC1272.
Powell RM, Wicks J, Holloway JW, Holgate ST, Davies DE. The splicing and fate of ADAM33 transcripts in primary human airways fibroblasts. Am J Respir Cell Mol Biol 2004;31:13eC21.
Shapiro SD, Owen CA. ADAM33 surfaces as an asthma gene. N Engl J Med 2002;347:936eC938.
Yoshinaka T, Nishii K, Yamada K, Sawada H, Nishiwaki E, Smith K, Yoshino K, Ishiguro H, Higashiyama S. Identification and characterization of novel mouse and human ADAM33s with potential metalloprotease activity. Gene 2002;282:227eC236.
Haitchi HM, Powell RM, Shaw TJ, Howarth PH, Wilson S, Wilson DI, Holgate ST, Davies DE. A disintegrin and metalloprotease 33 expression in asthmatic airways and human embryonic lung. Am J Respir Crit Care Med 2005;171:958eC965.
Davies DE, Wicks J, Powell RM, Puddicombe SM, Holgate ST. Airway remodeling in asthma: new insights. J Allergy Clin Immunol 2003;111:215eC225.
Payne DN, Rogers AV, Adelroth E, Bandi V, Guntapalli KK, Bush A, Jeffery PK. Early thickening of the reticular basement membrane in children with difficult asthma. Am J Respir Crit Care Med 2003;167:78eC82.
Warner JO, Marquet C, Rao R, Roche WR, Pohunek P. Inflammatory mechanisms in childhood asthma. Clin Exp Allergy 1998;28:71eC75.
Gunn TM, Azarani A, Kim PH, Hyman RW, Davis RW, Barsh GS. Identification and preliminary characterization of mouse ADAM33. BMC Genet 2002;3:2.
Haitchi HM, Powell RM, Wilson DI, Holgate ST, Davies DE. ADAM 33 expression in embryonic mouse lung . Am J Respir Crit Care Med 2003;167:A377.
Lowe LA, Woodcock A, Murray CS, Morris J, Simpson A, Custovic A. Lung function at age 3 years: effect of pet ownership and exposure to indoor allergens. Arch Pediatr Adolesc Med 2004;158:996eC1001.(Angela Simpson, Nikolas M)
Centre for Integrated Genomic Medical Research, University of Manchester, Manchester
Infection, Inflammation
Repair
Human Genetics Divisions, School of Medicine, University of Southampton, Southampton General Hospital, Southampton, United Kingdom
ABSTRACT
Rationale: Asthma commonly originates in early life in association with impaired lung function, which tracks to adulthood. Objectives: Within the context of a prospective birth cohort study, we investigated the association between single nucleotide polymorphisms (SNPs) in a disintegrin and metalloprotease 33 (ADAM33) gene and early-life lung function. Methods: Children were genotyped for 17 SNPs in ADAM33. Lung function at age 3 (n = 285) and 5 years (n = 470) was assessed using plethysmographic measurement of specific airway resistance (sRaw). At age 5, we also measured FEV1. SNPs were analyzed individually using logistic regression, followed by linkage disequilibrium mapping to identify the causal locus. Main Results: Carriers of the rare allele of F+1 SNP had reduced lung function at age 3 years (p = 0.003). When the recessive model was considered, four SNPs (F+1, S1, ST+5, V4) showed association with sRaw at age 5 years (p < 0.04). Using linkage disequilibrium mapping, we found evidence of a significant causal location between BC+1 and F1 SNPs, at the 5' end of the gene. Four SNPs were associated with lower FEV1 (F+1, M+1, T1, and T2; p 0.04). The risk of transient early wheezing more than doubled among children homozygous for the A allele of F+1 (odds ratio, 2.39; 95% confidence intervals, 1.18eC4.86; p = 0.02), but there was no association between any SNP and allergic sensitization or physician-diagnosed asthma. Conclusions: Polymorphisms in ADAM33 predict impaired early-life lung function. The functionally relevant polymorphism is likely to be at the 5' end of the gene.
Key Words: ADAM33; asthma; genetics; lung function
Asthma is the most common chronic disease of childhood (1), and on a worldwide scale has dramatically increased in prevalence over the last 30 years (2). Twin and other family-based studies confirm a strong genetic component of asthma (3). The recent identification of a putative asthma susceptibility gene underlying bronchial hyperresponsiveness, a disintegrin and metalloprotease 33 (ADAM33) (4), emphasizes the possibility that a primary end-organ determinant together with the systemic immune response underlie the manifestation of asthma.
Associations between polymorphisms in ADAM33 and asthma were reported in African Americans, Hispanics, and whites in the United States, and Dutch populations, but none of the associated single nucleotide polymorphisms (SNPs) were common to all populations studied (5). One SNP was subsequently associated with excess decline in lung function in Dutch adults with asthma (6), and associations with asthma were reported in German (7) and Korean (8) populations. A recent study of children with asthma identified only modest associations with ADAM33 SNPs (9), whereas no association was seen in Puerto Rican or Mexican populations (10). Inconsistencies in genetic association studies are common, particularly for genes in which the functionally relevant polymorphism has not been established. Such inconsistencies could be attributed to genetic heterogeneity between populations, differences in phenotype definition, and insufficient power. However, it is worth emphasizing that the support for a role of ADAM33 in asthma also comes from mRNA expression analysis (11, 12).
The majority of asthma originates in early life in association with impaired lung function, which tracks into later childhood (13) and adulthood (14eC16). Therefore, it is important to elucidate factors influencing lung function in early life. Although it is often difficult to assess standard lung function in preschool children, measurement of specific airway resistance (sRaw) is feasible (17, 18). sRaw is a measure of airway caliber independent of lung size, and narrowing of the airways results in elevated values. Bronchodilator response measured by sRaw has been shown to be useful in identifying preschool children with asthma, with sensitivity at least as good as spirometry in older children (19).
We have analyzed 17 SNPs in the ADAM33 gene in a prospective population-based birth cohort (the Manchester Asthma and Allergy Study) (18, 20, 21) in which assessment of lung function was completed at age 3 and 5 years. Given that the functionally relevant polymorphisms have not been clearly identified, we used the recently described method of association mapping by linkage disequilibrium (LD) (22, 23). This approach uses single-marker tests within a composite likelihood framework to identify the most likely location of the causal SNP on the LD map, which assigns a location in LD units (LDU) for each marker (24). Our hypothesis was that variation in the ADAM33 gene contributes to impaired lung function during the first few years of life. Some of the results of this study were reported in the form of an abstract (25).
METHODS
A detailed description of methods is presented in the online supplement. The Manchester Asthma and Allergy Study is a population-based birth cohort study (18, 20, 21). Participants were recruited prenatally (20), and attended review at age 3 and 5 years (± 4 weeks). Study was approved by the Local Research Ethics Committee. Informed consent was obtained from all parents.
Outcomes
sRraw was measured using whole-body plethysmography as previously described (18) by a single-step procedure from the simultaneously measured changes of respiratory flow and plethysmographic pressure, omitting the measurement of thoracic gas volume. Measurements were performed during normal tidal breathing, and we used the mean of three measurements of effective sRaw.
Dynamic lung volumes and expiratory flow were measured using incentive animation software. The test was repeated at intervals of 30 seconds until three technically acceptable traces were obtained and the highest FEV1 was recorded.
Symptoms.
A validated questionnaire (26) was interviewer-administered to collect information on parentally reported symptoms and physician-diagnosed illnesses. According to parentally reported wheeze at two follow-ups, children were assigned as follows (27): no wheezing (never reported wheeze; n = 241), transient early wheeze (wheezing during the first 3 years, no wheezing in the previous 12 months at age 5; n = 104), late-onset wheezing (no wheeze during the first 3 years, wheezing in the previous 12 months at age 5 years; n = 25), and persistent wheezing (wheezing during the first 3 years, wheezing in the previous 12 months at age 5; n = 63).
Allergic sensitization was ascertained at age 5 years by measurement of specific serum IgE to common inhalant and food allergens (ImmunoCAP; Pharmacia, Uppsala, Sweden). Sensitization was defined as allergen-specific IgE greater than 0.2 kU/L.
Genotyping
Seventeen SNPs previously associated with an asthma phenotype, spanning 10 kb of ADAM33 were selected for genotyping (Table 1). We sought to provide good coverage of the gene and to examine SNPs with a range of allele frequencies. SNPs were genotyped using TaqMan allelic discrimination assay (28) (Applied Biosystems, Foster City, CA). Polymerase chain reaction was performed according to the manufacturer's instructions in 5-e reaction volumes in 96-well plates and contained 10 ng genomic DNA per reaction. Four control samples were included on each polymerase chain reaction plate (one negative and three of known genotype). One SNP (S2) was genotyped using Sequenom (see online supplement for method).
Statistical Analysis
Each SNP was tested for deviation from Hardy-Weinberg equilibrium using a 2 test. Genotype comparisons of symptoms were performed using the 2 test. sRaw followed a log-normal distribution, hence was subject to a loge transformation. Univariate linear regression analysis was used to compare sRaw between genotype groups for each SNP. For SNPs showing a trend for association with lung function by genotype (p < 0.1), dominant and recessive genetic models were tested. Analysis was performed in Stata 8 (Stata Corporation, College Station, TX).
LDU Map Construction and Association Mapping
The LDU map was created using the LDMAP program (22eC24). These metric maps have additive distances and are analogous to linkage maps, but superior in representing the pattern of LD rather than just recombination (29, 30). Composite likelihood analysis of single-marker tests was used to identify the most likely location of the causal SNP on the LDU map, which avoids the heavy Bonferroni correction through multiple testing.
RESULTS
Participant Flow
The study profile is shown in Figure 1; children who were prenatally randomized to the environmental control study and 26 children who were not of mixed European descent were excluded from this analysis. Acceptable sRaw results were obtained in 285 children at age 3 and 470 at age 5 years (both 3- and 5-year lung function data were available in 280 children; Figure 1). By age 5 years, 34% of children were sensitized, 38.6% had at least one episode of wheeze, 21.7% had wheezed in the last 12 months, and 20% had received a physician diagnosis of asthma at some time during the first 5 years of life. In keeping with our previous report (18), symptomatic children had significantly higher sRaw values compared with asymptomatic children (e.g., wheeze ever, p < 0.0001).
Genotyping success rates were high (97.5eC99.5%), and allele frequencies (Table 1) were consistent with those seen in other mixed European cohorts (4, 5). Sixteen of the 17 SNPs were in Hardy-Weinberg equilibrium at the p > 0.05 significance level. SNP V-3 marginally deviated from Hardy-Weinberg equilibrium (p = 0.03) because of an observed increase in heterozygotes (252 vs. 229 expected), but was included in the analysis because other quality-control checks (including genotyping of samples with known genotypes) suggested this deviation was not caused by genotyping error.
sRaw at Age 3 and 5 Years
F+1 was the only SNP associated with sRaw at age 3 years by genotype (p = 0.01; Table 1, Figure 2a). Carriers of the A allele had significantly higher sRaw values (i.e., reduced lung function) than homozygotes for the G allele (geometric mean [GM], 95% confidence intervals [CI]; kPa · 1.12, 1.09eC1.16 vs. 1.04, 1.01eC1.08; p = 0.003). The F+1 polymorphism explained 3% of the variance in the lung function at age 3.
At age 5 years, four SNPs (F+1, S1, ST+5, V4) showed association with sRaw by genotype (Table 1, Figure 2b). The strength of association increased substantially when a recessive model was considered for all four SNPs (Table 2).
LD Mapping across the ADAM33 Gene
The LD map describes the underlying structure of LD in the form of a metric map with additive distances in LDU. Therefore, every SNP was assigned two locations: one in kilobases and the other in LDU (Table 3). Figure 3 presents graphically the block-step structure of the ADAM33 region by plotting the LDU locations against the kilobase map. As shown in Figure 3, there is one main block of high LD with invariant LDU spanning 3.7 kb (SNPs M+1 to V2), which is flanked by two short SNP blocks.
For association mapping, we used the method developed by Maniatis and colleagues (22, 23) (see online supplement for the detailed results). Using different subhypotheses of the model, the analysis first determines whether there is adequate power to test for association in the genomic region, and then provides the estimated causal location. Our calculations suggested sufficient power to localize a causal polymorphism within the ADAM33 gene that is associated with lung function at age 5 years using the kilobase map under an additive model (p < 0.001). Tests for a disease determinant provided a significant causal location at 1.34 kb (p < 0.05), which is positioned in a step between BC+1 and F1 SNPs (Table E2, online supplement).
Analysis of sRaw at age 5 years under the recessive model increased the error variance and decreased the power by approximately 50%. Nevertheless, the analysis suggested sufficient power to localize a causal polymorphism and showed a significant association within the region (p < 0.005), but no significant association at any location (p > 0.05).
We also examined lung function at age 3 years, but did not have sufficient power to localize a causal polymorphism, probably because of a smaller sample size (285, compared with 470 at age 5).
Other Lung Function Measures at Age 5 Years
FEV1.
Four SNPs were associated with FEV1 by genotype (Table 1). The rare alleles of four SNPs were associated with lower FEV1 percent-predicted values: F+1—GM, 106; 95% CI, 101eC110 versus GM, 111; 95% CI, 109eC113; p = 0.04; M+1—GM, 106; 95% CI, 102eC110 versus GM, 112; 95% CI, 110eC114; p = 0.005; T1—GM, 107; 95% CI, 103eC111 versus GM, 112; 95% CI, 110eC114; p = 0.01; and T2—GM, 107, 95% CI, 103eC111 versus GM, 112; 95% CI, 110eC114; p = 0.02 (Table E3).
Symptoms
Five SNPs associated with impaired lung function were also associated with early-life wheezing. Children who were homozygous for the A allele of F+1 were more than twice as likely to have wheezed in the first year of life (odds ratio [OR], 2.56; 95% CI, 1.36eC4.82; p = 0.004). The presence of the A allele of M+1 (OR, 2.02; 95% CI, 1.25eC3.26; p = 0.004), the C allele of T1 (OR, 2.05; 95% CI, 1.28eC3.29; p = 0.003) and the T allele of T2 (OR, 2.01; 95% CI, 1.26eC3.21; p = 0.004), and homozygosity for the G allele of ST+5 (OR, 1.93; 95% CI, 1.24eC2.97; p = 0.003) were all associated with an increased risk of infant wheeze.
We assigned 433 children into one of four wheeze phenotypes (37 children could not be classified, e.g., no history of wheeze, but receiving asthma medication). Among children who had never wheezed, there was a smaller proportion homozygous for the A allele of F+1 (7.5%) compared with transient early wheezers (16.2%, p = 0.01), and persistent wheezers (14.3%, p = 0.09).
Using never wheezers as a reference, the risk of transient early wheezing more than doubled among children homozygous for the A allele of F+1 (OR, 2.39; 95% CI, 1.18eC4.86; p = 0.02). There was no association between physician-diagnosed asthma by age 5 years and any of the SNPs.
Allergic Sensitization
No association was observed between ADAM33 SNPs and allergic sensitization at age 5 years whether using specific IgE or skin-prick test results.
DISCUSSION
The current study provides the first evidence that the association of ADAM33 with asthma may be mediated via its effects on lung function in childhood. Our data support the hypothesis that reduced early-life lung function is in part a genetically determined trait. Polymorphisms were associated with reduced lung function at both 3 and 5 years of age and the more significant associations at age 5 years may reflect the increased sample size. All of the SNPs associated with altered lung function in the current study have previously been associated with asthma and bronchial hyperresponsiveness in adults in U.K., U.S., Dutch, and German populations (4, 5, 7). This study was unable to show any association between ADAM33 SNPs and allergic sensitization. This suggests that separate genetic factors contribute to disordered airway function and the immunologic component of asthma (31, 32), and further consolidates the role of ADAM33 in primary end-organ pathology rather than the systemic immune response.
Birth cohort studies provide an opportunity to observe the development of wheezing symptoms throughout childhood. We have shown an association between polymorphisms in ADAM33 and transient early wheeze, with a trend for persistent wheeze but not late-onset wheeze. Our findings are consistent with the notion that ADAM33 polymorphisms may be associated with smaller airways. The contribution of airway size to the expression of symptoms may be greater earlier in childhood. This supports the hypothesis that childhood asthma is a heterogeneous disease, and provides additional evidence that different genetic and environmental factors may be important in the different wheezing phenotypes. Thus, ADAM33 may be a "lung function gene" rather than an "asthma gene," variants of which increase the risk of wheeze in early life.
Although the selection of the 17 SNPs in the current study was based on the results of previous studies, they represent only a proportion of the total variation of the gene. The regression analysis showed that F+1 is the most strongly associated SNP, and even when we applied an overconservative Bonferroni correction, F+1 remained nominally significant (see online supplement). However, single regression analysis tests each SNP assuming it is the functional polymorphism. The LD method we present in this study is based on composite likelihood where all the information from all single-marker tests is considered simultaneously, removing the need for any correction for multiple tests. The maximum likelihood estimation is an estimate of the causal location under the assumption that the functional polymorphism may not be one of those tested. Using simulated (22) and real (23) data, we have previously shown that greater power is achieved when mapping within an LDU map compared with a map in kilobases, especially in a densely typed region (22). This analysis identified that the ADAM33 locus is associated with lung function and that the functional polymorphism most likely lies at the 5' end of the gene (between BC+1 and F1). The original positional cloning study identified a higher density of SNPs at the 3' end of the gene, and so replication studies have focused on this region, with variable results (6eC10). Although it is plausible that more than one functional polymorphism exists and that these may have different effects on lung function and asthma, there is also biological evidence that the majority of splice variation occurs at the 5' end of the gene (33).
ADAM proteins are transmembrane multidomain proteins with potential for cell adhesion, fusion, communication, signaling, and protease activities (34). The exact function of ADAM33 remains unknown and the functional nature of the associated polymorphisms is yet to be established. Expression of ADAM33 mRNA has been detected in all tissues except liver, with high levels in lung and placenta (35). Gene expression studies indicate that ADAM33 RNA is expressed in lung fibroblasts and bronchial smooth muscle, but not in bronchial epithelial cells or lymphocytes (4, 12, 33). Furthermore, studies of alternative splicing of ADAM33 transcripts suggest that the 3' domains (the epidermal growth factoreClike, transmembrane, and cytoplasmic domains corresponding to SNPs Q+1 through V4) are present in all transcripts, but that the variability in exon usage occurs at the 5' end of the gene (the metalloprotease and prodomains, corresponding to SNPs BC+1 through M+1) (33). It is tempting to speculate that the functional SNP located between F1 and BC+1 may alter which isoform is expressed. Recent work from our laboratory failed to find differences in total levels of ADAM33 transcripts, or relative expression of different splice forms, in bronchial biopsies from normal individuals and individuals with asthma (36). However, these studies were performed on a relatively small number of subjects without genotype information for the relevant polymorphisms and will need to be followed up in larger cohorts where genotypic information is available. Clearly, additional research is required to establish the exact function of ADAM33 and how it is altered by polymorphisms within the gene.
This demonstration that ADAM33 polymorphisms influence lung function in early life may support the hypothesis that epithelial-mesenchymal dysfunction in the airways may predispose individuals toward asthma (37), being present in early childhood before asthma becomes clinically expressed (38, 39). In this context, it is interesting to note that ADAM33 expression has been shown to occur during murine embryonic development, suggesting that it may regulate tissue morphogenesis (40). Preliminary studies have found that murine ADAM33 expression is induced in embryonic lungs at the start of branching morphogenesis, increases with gestation and remains present into adulthood (41). This raises the possibility that polymorphisms which modulate ADAM33 expression and/or function may modulate lung growth and development. We speculate that, in addition to being "remodeled" as a consequence of asthma, the airways could be "premodeled" as one of the prerequisites for subsequent development of wheeze as a genetically determined trait (18).
Sex, maternal smoking, and high allergen exposure in sensitized individuals are also associated with early-life lung function (42). We have undertaken a multivariate analysis that includes ADAM33 polymorphisms and these three factors and have shown that the association of ADAM33 polymorphisms and preschool lung function is independent of sex, smoking, or allergen exposure in sensitized individuals (see online supplement). Consistent with this gene influencing lung function, a recent study has shown that SNPs in ADAM33 predict a more rapid decline in lung function in individuals with chronic asthma followed over 20 years (6).
Thorough knowledge of lung physiology in early life and its genetic component could refine our understanding of the causes of asthma. If the pattern of lung function in early childhood is predictive of that in adult life and can help identify individuals at risk, this helps our understanding of the causes of wheezing illness and targeting at-risk individuals for primary prevention strategies.
Acknowledgments
The authors thank the children and parents of the Manchester Asthma and Allergy Study for their continued dedication to the study.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Kaur B, Anderson HR, Austin J, Burr M, Harkins LS, Starchan DP, Warner JO. Prevalence of asthma symptoms, diagnosis and treatment in 12eC14 year old school children across Great Britain (International Study of Asthma and Allergies in Childhood, ISAAC, UK). BMJ 1998;316:118eC124.
Omran M, Russell G. Continuing increase in respiratory symptoms and atopy in Aberdeen schoolchildren. BMJ 1996;312:34.
Duffy DL, Martin NG, Battistutta D, Hopper JL, Mathews JD. Genetics of asthma and hay fever in Australian twins. Am Rev Respir Dis 1990;42:1351eC1358.
Van Eerdewegh P, Little RD, Dupuis J, Del Mastro RG, Falls K, Simon J, Torrey D, Pandit S, McKenny J, Braunschweiger K, et al. Association of ADAM33 gene with asthma and bronchial hyperresponsiveness. Nature 2002;418:426eC430.
Howard TD, Postma DS, Jongepier H, Moore WC, Koppelman GH, Zheng SL, Xu J, Bleecker ER, Meyers DA. Association of a disintegrin and metalloprotease 33 (ADAM33) gene with asthma in ethnically diverse populations. J Allergy Clin Immunol 2003;112:717eC722.
Jongepier H, Boezen HM, Dijkstra A, Howard TD, Vonk JM, Koppleman GH, Zheng SL, Meyers DA, Bleecker ER, Postma DS. Polymorphisms of the ADAM33 gene are associated with accelerated lung function decline in asthma. Clin Exp Allergy 2004;34:757eC760.
Werner M, Herbon N, Gohlke H, Altmuller J, Knapp M, Heinrich J, Wjst M. Asthma is associated with single-nucleotide polymorphisms in ADAM33. Clin Exp Allergy 2004;34:26eC31.
Lee JH, Park HS, Park SW, Jang AS, Uh ST, Rhim T, Park CS, Hong SJ, Holgate ST, Holloway JW, et al. ADAM33 polymorphism: association with bronchial hyper-responsiveness in Korean asthmatics. Clin Exp Allergy 2004;34:860eC865.
Raby BA, Silverman EK, Kwiatkowski DJ, Lange C, Lazarus R, Weiss ST. ADAM33 polymorphisms and phenotype associations in childhood asthma. J Allergy Clin Immunol 2004;113:1071eC1078.
Lind DL, Choudry S, Ung N, Ziv E, Avila PC, Salari K, Ha C, Lovins EG, Coyle NE, Nazario S, et al. ADAM33 is not associated with asthma in Puerto Rican or Mexican populations. Am J Respir Crit Care Med 2003;168:1312eC1316.
Garlisi CG, Zou J, Devito KE, Tian F, Zhu FX, Liu J, Shah H, Wan Y, Motasim Billah M, Egan RW, et al. Human ADAM33: protein maturation and localization. Biochem Biophys Res Commun 2003;301:35eC43.
Umland SP, Garlisi CG, Shah H, Wan Y, Zou J, Devito KE, Huang WM, Gustafson EL, Ralson R. Human ADAM33 mRNA expression profile and post- transcriptional regulation. Am J Respir Cell Mol Biol 2003;29:571eC582.
Turner SW, Palmer LJ, Rye PJ, Gibson NA, Judge PK, Cox M, Young S, Goldblatt J, Landau LI, LeSouef PN. The relationship between infant airway function, childhood airway responsiveness and asthma. Am J Respir Crit Care Med 2004;169:921eC927.
Phelan PD, Robertson CF, Olinski A. The Melbourne Asthma Study: 1964eC1999. J Allergy Clin Immunol 2002;109:189eC194.
Rasmussen F, Taylor DR, Flannery EM, Cowan JO, Greene JM, Herbison GP, Sears MR. Risk factors for airway remodelling in asthma manifested by a low postbronchodilator FEV1/vital capacity ratio: a longitudinal population study from childhood to adulthood. Am J Respir Crit Care Med 2002;165:1480eC1488.
Sears MR, Greene JM, Willan AR, Wiecek AM, Taylor DR, Flannery EM, Cowan JO, Herbison GP, Silva PA, Poulton R. A longitudinal, population-based, cohort study of childhood asthma followed to adulthood. N Engl J Med 2003;349:1414eC1422.
Klug B, Bisgaard H. Measurement of specific airway resistance by plethysmography in young children accompanied by an adult. Eur Respir J 1997;10:1599eC1605.
Lowe L, Murray CS, Custovic A, Simpson BM, Kissen PM, Woodcock A. Specific airway resistance in three year-old children: a prospective cohort study. Lancet 2002;359:1904eC1908.
Nielsen KG, Bisgaard H. Discriminative capacity of bronchodilator response measured with three different lung function techniques in asthmatic and healthy children aged 2 to 5 years. Am J Respir Crit Care Med 2001;164:554eC559.
Simpson BM, Custovic A, Simpson A, Hallam CL, Walsh D, Marolia H, Campbell J, Woodcock A. NAC Manchester Asthma and Allergy Study (NACMAAS): risk factors for asthma and allergic disorders in adults. Clin Exp Allergy 2001;31:391eC399.
Custovic A, Simpson BM, Simpson A, Kissen P, Woodcock A. Effect of environmental manipulation in pregnancy and early life on respiratory symptoms and atopy during first year of life: a randomized trial. Lancet 2001;358:188eC193.
Maniatis N, Collins A, Gibson J, Zhang W, Tapper W, Morton NE. Positional cloning by linkage disequilibrium. Am J Hum Genet 2004;74:846eC855.
Maniatis N, Morton NE, Gibson J, Xu C-F, Hosking LK, Collins A. The optimal measure of linkage disequilibrium minimizes error in association mapping of affection status. Hum Mol Genet 2005;14:145eC153.
Maniatis N, Collins A, Xu CF, McCarthy LC, Hewett DR, Tapper W, Ennis S, Ke X, Morton NE. The first linkage disequilibrium (LD) maps: delineation of hot and cold blocks by diplotype analysis. Proc Natl Acad Sci USA 2002;99:2228eC2233.
Simpson A, Jury F, Custovic A, Holloway JW, Woodcock A, Cakebread JA, Lowe LA, Ollier WER, Holgate ST, John S. Manchester Asthma and Allergy Study: polymorphisms in ADAM33 predict lung function at age 5 years. J Allergy Clin Immunol 2004;113:S340.
Pearce N, Weiland S, Keil U, Langridge P, Anderson HR, Strachan D, Bauman A, Young L, Gluyas P, Ruffin D, et al. Self-reported prevalence of asthma symptoms in children in Australia, England, Germany and New Zealand: an international comparison using the ISAAC protocol. Eur Respir J 1993;6:1455eC1461.
Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan WJ. Asthma and wheezing in the first six years of life. N Engl J Med 1995;332:133eC138.
Livak KG. Alleic discrimination using fluoregenic probes and the 5' nuclease assay. Genet Anal 1999;14:143eC149.
Tapper WJ, Maniatis N, Morton NE, Collins A. A metric linkage disequilibrium map of a human chromosome. Ann Hum Genet 2003;67:487eC494.
Ke X, Hunt S, Tapper W, Lawrence R, Stavrides G, Ghori J, Whittaker P, Collins A, Morris AP, Bentley D, et al. The impact of SNP density on fine-scale patterns of linkage disequilibrium. Hum Mol Genet 2004;13:577eC588.
Gerrard JW, Vickers P, Gerrard CD. The familial incidence of allergic disease. Ann Allergy 1976;36:10eC15.
von Mutius E, Nicolai T. Familial aggregation of asthma in a South Bavarian population. Am J Respir Crit Care Med 1996;153:1266eC1272.
Powell RM, Wicks J, Holloway JW, Holgate ST, Davies DE. The splicing and fate of ADAM33 transcripts in primary human airways fibroblasts. Am J Respir Cell Mol Biol 2004;31:13eC21.
Shapiro SD, Owen CA. ADAM33 surfaces as an asthma gene. N Engl J Med 2002;347:936eC938.
Yoshinaka T, Nishii K, Yamada K, Sawada H, Nishiwaki E, Smith K, Yoshino K, Ishiguro H, Higashiyama S. Identification and characterization of novel mouse and human ADAM33s with potential metalloprotease activity. Gene 2002;282:227eC236.
Haitchi HM, Powell RM, Shaw TJ, Howarth PH, Wilson S, Wilson DI, Holgate ST, Davies DE. A disintegrin and metalloprotease 33 expression in asthmatic airways and human embryonic lung. Am J Respir Crit Care Med 2005;171:958eC965.
Davies DE, Wicks J, Powell RM, Puddicombe SM, Holgate ST. Airway remodeling in asthma: new insights. J Allergy Clin Immunol 2003;111:215eC225.
Payne DN, Rogers AV, Adelroth E, Bandi V, Guntapalli KK, Bush A, Jeffery PK. Early thickening of the reticular basement membrane in children with difficult asthma. Am J Respir Crit Care Med 2003;167:78eC82.
Warner JO, Marquet C, Rao R, Roche WR, Pohunek P. Inflammatory mechanisms in childhood asthma. Clin Exp Allergy 1998;28:71eC75.
Gunn TM, Azarani A, Kim PH, Hyman RW, Davis RW, Barsh GS. Identification and preliminary characterization of mouse ADAM33. BMC Genet 2002;3:2.
Haitchi HM, Powell RM, Wilson DI, Holgate ST, Davies DE. ADAM 33 expression in embryonic mouse lung . Am J Respir Crit Care Med 2003;167:A377.
Lowe LA, Woodcock A, Murray CS, Morris J, Simpson A, Custovic A. Lung function at age 3 years: effect of pet ownership and exposure to indoor allergens. Arch Pediatr Adolesc Med 2004;158:996eC1001.(Angela Simpson, Nikolas M)