UBC Faculty Research and Publications

Contribution of alpha- and beta-defensins to lung function decline and infection in smokers: an association… Wallace, Alison M; He, Jian-Qing; Burkett, Kelly M; Ruan, Jian; Connett, John E; Anthonisen, Nicholas R; Paré, Peter D; Sandford, Andrew J May 15, 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12931_2005_Article_467.pdf [ 507.43kB ]
Metadata
JSON: 52383-1.0074624.json
JSON-LD: 52383-1.0074624-ld.json
RDF/XML (Pretty): 52383-1.0074624-rdf.xml
RDF/JSON: 52383-1.0074624-rdf.json
Turtle: 52383-1.0074624-turtle.txt
N-Triples: 52383-1.0074624-rdf-ntriples.txt
Original Record: 52383-1.0074624-source.json
Full Text
52383-1.0074624-fulltext.txt
Citation
52383-1.0074624.ris

Full Text

ralssBioMed CentRespiratory ResearchOpen AcceResearchContribution of alpha- and beta-defensins to lung function decline and infection in smokers: an association studyAlison M Wallace1, Jian-Qing He1, Kelly M Burkett1, Jian Ruan1, John E Connett2, Nicholas R Anthonisen3, Peter D Paré1 and Andrew J Sandford*1Address: 1James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research, University of British Columbia, Vancouver, Canada, 2Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, USA and 3Faculty of Medicine, University of Manitoba, Winnipeg, CanadaEmail: Alison M Wallace - aw2264@columbia.edu; Jian-Qing He - jhe@mrl.ubc.ca; Kelly M Burkett - kburkett@sfu.ca; Jian Ruan - jruan@mrl.ubc.ca; John E Connett - john-c@ccbr.umn.edu; Nicholas R Anthonisen - nanthonisen@exchange.hsc.mb.ca; Peter D Paré - ppare@mrl.ubc.ca; Andrew J Sandford* - asandford@mrl.ubc.ca* Corresponding author    AbstractBackground: Alpha-defensins, which are major constituents of neutrophil azurophilic granules,and beta-defensins, which are expressed in airway epithelial cells, could contribute to thepathogenesis of chronic obstructive pulmonary disease by amplifying cigarette smoke-induced andinfection-induced inflammatory reactions leading to lung injury. In Japanese and Chinesepopulations, two different beta-defensin-1 polymorphisms have been associated with chronicobstructive pulmonary disease phenotypes. We conducted population-based association studies totest whether alpha-defensin and beta-defensin polymorphisms influenced smokers' susceptibility tolung function decline and susceptibility to lower respiratory infection in two groups of whiteparticipants in the Lung Health Study (275 = fast decline in lung function and 304 = no decline inlung function).Methods: Subjects were genotyped for the alpha-defensin-1/alpha-defensin-3 copy numberpolymorphism and four beta-defensin-1 polymorphisms (G-20A, C-44G, G-52A and Val38Ile).Results: There were no associations between individual polymorphisms or imputed haplotypesand rate of decline in lung function or susceptibility to infection.Conclusion: These findings suggest that, in a white population, the defensin polymorphisms testedmay not be of importance in determining who develops abnormally rapid lung function decline oris susceptible to developing lower respiratory infections.BackgroundChronic obstructive pulmonary disease (COPD) is charac-ing, which is the main environmental risk factor. Longitu-dinal studies show that only a minority of cigarettePublished: 15 May 2006Respiratory Research 2006, 7:76 doi:10.1186/1465-9921-7-76Received: 24 October 2005Accepted: 15 May 2006This article is available from: http://respiratory-research.com/content/7/1/76© 2006 Wallace et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 8(page number not for citation purposes)terized by irreversible airflow obstruction due to chronicinflammation. COPD is closely related to cigarette smok-smokers develop airflow limitation,[1] suggesting thatother environmental or genetic factors are important.Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76Family and twin studies provide further evidence thatgenetic factors play a key role in the etiology of this dis-ease.[2,3] Inherited deficiency of alpha-1 antitrypsin isassociated with COPD; other genetic risk factors remain tobe identified.Defensins are small (29–47 amino acids) cationic micro-bicidal peptides that form part of both the innate andadaptive immune responses. Defensins show activitiesagainst Gram-positive and Gram-negative bacteria, fungi,yeast, and enveloped viruses.[4] Defensins are known pri-marily for their antimicrobial activities; however, thescope of their activities extends beyond immuneresponses and some of these functions could contribute tolung injury. [5-9]Based on a characteristic three-dimensional fold and a 6-cysteine/3-disulfide pattern, human defensins are dividedinto two groups, alpha-defensins and beta-defensins.[10]Three closely related alpha-defensins, alpha-defensin-1(DEFA1), DEFA2, and DEFA3, are major constituents ofneutrophil azurophilic granules (30–50% of the total pro-tein content)[11] and play a role in antimicrobial activityand inflammation in the lung. Beta-defensin-1 (DEFB1) isconstituatively expressed in airway epithelia and plays animportant role in mucosal immunity in the lung.[12,13]In humans, five alpha-defensins and at least four beta-defensins are clustered in a 450 kb region on chromo-some 8p23.[14,15] DEFA1, DEFA2, and DEFA3 differ byonly one amino acid. DEFA2 is the proteolytic product ofDEFA1 and/or DEFA3; DEFA1 and DEFA3 are encoded byseparate genes.[16] Due to a copy number polymor-phism, individuals can inherit variable copy numbers ofthe DEFA1 and DEFA3 genes, and the DEFA3 gene is oftenabsent.[16] Genetic variation in the DEFB1 gene is associ-ated with COPD phenotypes in Japanese [17] and Chi-nese [18] populations.We determined whether inheritance of both DEFA1 andDEFA3 genes, rather than the DEFA1 gene only, was asso-ciated with fast decline in lung function. In addition, wedetermined the frequency of four DEFB1 polymorphismsin the Lung Health Study participants. We also investi-gated the relationship between the defensin polymor-phisms and smokers' susceptibility to lower respiratoryinfection.MethodsStudy subjectsSubjects were selected from participants in Phase I of theNational Heart, Lung, and Blood Institute (NHLBI) LungHealth Study. Details of the study have been previouslyairflow obstruction (FEV1 55–90% predicted and FEV1/FVC ≤ 0.70). The primary outcome variable was rate ofdecline in post-bronchodilator FEV1 over a follow-upperiod of five years. Lower respiratory infection rate wasquantified using the number of self-reported visits to aphysician and days in bed for lower respiratory infectionat each follow-up. Of the 3216 continuing smokers in thiscohort, 275 were chosen with a fast decline in lung func-tion (decline in percent predicted FEV1 > 3.0% per year),and 304 were selected with no decline in lung functionover the same period (increase in percent predicted FEV1> 0.4% per year). All 579 selected participants were whiteand non-Hispanic. In addition, 27 African American LungHealth Study participants were genotyped; the data wereused to determine the prevalence of genotypes acrossracial groups.GenotypingRestriction fragment length polymorphism analysis wasperformed in order to distinguish the DEFA1 and DEFA3genes.[14] An amplicon of 950 bp was generated by 35cycles of PCR using the sense primer 5'-CAGCGGACATC-CCAGAAGTGG and the antisense primer 5'-GCGTTTT-GGTACGTGTATCC. PCRs were performed in a totalreaction volume of 20 μL with 100 ng of genomic DNA,0.5 U Taq polymerase (Invitrogen), 10X PCR buffer (Inv-itrogen), 3 mM Mg2+, 0.4 μM forward and reverse primers,and 200 μM dNTPs. After the initial denaturation at 95°Cfor 15 min, the reaction mixture was subjected to 35 cyclesof 94°C for 30 s, annealing for 30 s at 57°C, and 72°C for30 s followed by the final extension at 72°C for 5 min.After PCR, 20 μL of the reaction mixture was digested with1.25 U Hae III restriction endonuclease (New EnglandBioLabs Inc., Beverly, MA). The digest mixture wasresolved on a 2% agarose gel stained with ethidium bro-mide. DNA from individuals with DEFA1 only producedtwo bands, one at 300 bp and one at 650 bp, and individ-uals with both DEFA1 and DEFA3 produced all threebands. Genotyping of the DEFB1 polymorphisms at posi-tions -20, -44, and -52 in the 5' untranslated region wereperformed by restriction fragment length polymorphismanalysis.[14] An amplicon of 260 bp was generated by 35cycles of PCR using the sense primer 5'-GTGGCACCTC-CCTTCAGTTCCG and the antisense primer 5'-CAGCCCT-GGGGATGGGAAACTC. PCRs were performed in a totalreaction volume of 60 μL with 100 ng of genomic DNA,0.5 U Taq polymerase (Invitrogen), 10X PCR buffer (Inv-itrogen), 1.5 mM Mg2+, 0.75 μM forward and reverseprimers, and 200 μM dNTPs. After the initial denaturationat 95°C for 15 min, the reaction mixture was subjected to35 cycles of 94°C for 30 s, annealing for 30 s at 67°C, and72°C for 30 s followed by the final extension at 72°C for5 min. After PCR, 20 μL of the reaction mixture wasPage 2 of 8(page number not for citation purposes)published.[19] Briefly, study participants were currentsmokers, 35–60 years of age, who had mild to moderatedigested with 2 U Scr FI restriction endonuclease (NewEngland BioLabs Inc.) to detect variation at position -20.Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76The digest mixture was resolved on a 3% agarose gelstained with ethidium bromide. DNA from individualswith the homozygous G genotype (GG) produced threebands at 136 bp, 118 bp, and 6 bp; the homozygous Agenotype (AA) produced two bands, one at 254 bp andone at 6 bp; and the heterozygous genotype (GA) pro-duced all four bands. 20 μL of the reaction mixture wasdigested with 4 U Hga I restriction endonuclease (NewEngland BioLabs Inc.) to detect variation at position -44.The digest mixture was resolved on a 3% agarose gelstained with ethidium bromide. DNA from individualswith the homozygous C genotype (CC) produced twobands, one at 71 bp and one at 189 bp; the homozygousG genotype (GG) produced three bands at 71 bp, 30 bp,and 159 bp; and the heterozygous genotype (CG) pro-duced all three bands. 20 μL of the reaction mixture wasdigested with 2 U Nla IV restriction endonuclease (NewEngland BioLabs Inc.) to detect variation at position -52.The digest mixture was resolved on a 3% agarose gelstained with ethidium bromide. DNA from individualswith the homozygous G genotype (GG) produced threebands at 108 bp, 122 bp, and 30 bp; the homozygous Agenotype (AA) produced two bands, one at 230 bp andone at 30 bp; and the heterozygous genotype (GA) pro-duced all four bands. The DEFB1 Val38Ile polymorphismwas analyzed as previously described.[17] Template-freecontrols and known genotype controls were included ineach experiment. Twenty samples were selected at randomand sequenced to confirm the genotyping protocols. Gen-otypes were assigned by two independent investigatorswho were unaware of the patients' identities and pheno-types. Inconsistencies were resolved by two additionalgenotyping reactions.Statistical analysisHardy-Weinberg equilibrium tests and linkage disequilib-rium estimation were performed using Arlequin version2.0.[20] Haplotype frequencies were estimated usingPHASE version 2.0.[21,22] The chi-square test was used toadjust for the potential confounding factors of age, sex,smoking history (pack-years), methacholine responsive-ness, and initial level of lung function (pre-bronchodila-tor FEV1 percent predicted). Both analyses wereperformed using JMP Version 5.1. The infection outcomesof physician visits/year and days in bed/year are highlyright-skewed, with a high proportion of subjects havingno infection outcomes. To handle this type of data, a two-part model was used.[23,24] The first part deals with thespike of observations at zero and models which factorscontribute to a person developing infection-related out-comes. The second part models the severity, duration(days in bed/year), or frequency (physician visits/year) ofthe infection related outcomes in those who had an out-come. For the first part of the model, logistic regressionwas used to test for association with the presence of atleast one infectious event in a year. The data for the secondpart consists only of those with an infection related out-come and is highly skewed. Generalized Linear Modelregression, assuming gamma distributed observations anda log link,[25] was then used to test for association of gen-otype with the average number of infectious events peryear in those having any such events. For both regressionanalyses, rate of decline was included as a factor and bothwere performed using R http://www.r-project.org. All hap-lotype analyses were performed using a contributed R pro-gram called Hapassoc, which tests for haplotype-phenotype association when haplotype phase isunknown.[26,27] The power calculator used is availablethrough the UCLA Department of Statistics http://calculators.stat.ucla.edu/powercalc/. Power calculations for asample size of n = 275 cases and n = 304 controls wereperformed using a two sided test with the observed allelefrequencies, alpha = 0.05, and beta = 0.80. Statistical sig-nificance was defined at the 5% level.ResultsGeneral characteristicsThe study population consisted of 579 white and non-Table 1: Description of the study population.Variable Fast decline (n = 275) No decline (n = 304) p Value*Age, yr† 49.5 (6.4) 47.6 (6.9) 0.0007Sex, n (%) 163 men (59) 203 men (67) 0.06Smoking history, pk yrs†‡ 43.3 (19.1) 38.3 (18.1) 0.0005Baseline FEV1, % predicted†§ 72.7 (8.9) 75.7 (8.1) <0.0001Methacholine response†ll -23.4 (32.7) -7.5 (14.0) <0.0001* p Values derived from Wilcoxon test or chi-square analysis.† Mean (SD).‡ Number of packs of cigarettes smoked per day × number of years of smoking.§ Lung function at the start of the study as measured percent predicted FEV1 (postbronchodilator).ll Measurement of the responsiveness of the airways to methacholine expressed as percent decline in FEV1 per final cumulative dose of methacholine administered[36] Methacholine response is strongly associated with rate of decline in lung function[37,38]Page 3 of 8(page number not for citation purposes)test for association of genotype with decline in lung func-tion status. Logistic regression was then performed toHispanic smokers; 275 smokers with a fast decline in lungfunction and 304 smokers with no decline in lung func-Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76tion over the five year study period. The characteristics ofthe study group are described in Table 1. Age, sex, smok-ing history (pack-years), baseline FEV1, and metha-choline response differed significantly between the twogroups.DEFA1 and DEFA3 genes and rate of decline in lung functionThe number of individuals who had a copy of the DEFA1gene only (n = 64; 31 fast decline and 33 no decline),rather than copies of both DEFA1 and DEFA3 genes (n =515; 244 fast decline and 271 no decline), was not signif-icantly different in the fast decline in lung function group(11.8%) compared with the no decline in lung functiongroup (10.9%), p = 0.87. No individuals in the studygroup inherited a copy of the DEFA3 gene alone. Logisticregression modeling to adjust for confounding variablesconfirmed the lack of association (p = 0.99).DEFB1 gene polymorphisms and rate of decline in lung functionGenotypic frequencies of the three DEFB1 SNPs (G-20A,C-44G, and G-52A) analyzed are given in Table 2. Theobserved distribution of all three genotypes was consist-ent with Hardy-Weinberg equilibrium. Univariate analy-sis did not suggest any significant associations with thegenotypes and rate of decline in lung function. Logisticregression modeling to adjust for confounding variablesconfirmed the lack of association (G-20A, p = 0.82; C-44G, p = 0.66; and G-52A, p = 0.76). Subjects were alsogenotyped for the DEFB1 Val38Ile SNP. Only one hetero-zygous subject was detected, suggesting that this polymor-phism is very rare (<1%), at least in whites.DEFB1 linkage disequilibrium and haplotype association analysisThere was strong linkage disequilibrium between thethree DEFB1 SNPs in the 5' untranslated region (Table 3).Seven haplotypes were revealed. The estimated haplotypefrequencies in the two Lung Health Study groups are givenin Table 4. Univariate analysis did not suggest any signifi-cant association between haplotypes and rate of decline inlung function. The three most common haplotypes (GCA,ACG, and GGG) were detected in two other studies andwere the only haplotypes reported by those stud-ies.[17,28]Secondary outcomes analysisStatistical analysis did not suggest any significant associa-tions between genotypes (Table 5) or haplotypes (Table6) and infection outcomes.Alpha-defensin and beta-defensin polymorphisms in different racial groupsWe determined the prevalence of genotypes across whiteand African American participants of the Lung HealthStudy and found statistically significant differencesbetween racial groups (Table 7).DiscussionWe investigated the role of alpha-defensin and beta-Table 2: Beta-defensin-1 genotypic frequencies by fast decline or no decline in lung function status. Frequencies shown, n (%).Polymorphism Phenotype GenotypeG-20A GG GA AA(rs11362) Fast decline 83 (30.2) 131 (47.6) 61 (22.2)No decline 91 (29.9) 150 (49.3) 63 (20.7)*p = 0.89C-44G CC CG GG(rs1800972) Fast decline 177 (64.4) 85 (30.9) 13 (4.7)No decline 198 (65.1) 91 (29.9) 15 (4.9)*p = 0.97G-52A GG GA AA(rs1799946) Fast decline 107 (38.9) 128 (46.6) 40 (14.6)No decline 130 (42.8) 128 (42.1) 46 (15.1)*p = 0.55* p Values derived from chi-square test.Table 3: Beta-defensin-1 linkage disequilibrium.D' (D/Dmax) r2 p Value*G-20A and G-52A 0.75 0.29 <0.00001G-20A and C-44G 0.88 0.17 <0.00001C-44G and G-52A 0.93 0.13 <0.00001Page 4 of 8(page number not for citation purposes)defensin polymorphisms in promoting FEV1 decline insmokers with COPD and the influence on smokers' sus-* p Values derived from chi-square test.Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76ceptibility to infection. Mars and coworkers[16] haveshown that individuals can inherit variable copy numbersof the DEFA1 and DEFA3 genes, and that the DEFA3 geneis often absent. Several polymorphisms of beta-defensin-1 have been reported. [28-31] The frequency of polymor-phisms in the coding region is low compared with poly-morphisms in the promoter and untranslated regions. Ofthe reported polymorphisms, three polymorphisms in the5' untranslated region at positions -20, -44, and -52 of theDEFB1 gene have minor allele frequencies greater thantwenty percent in a white population.[28] The SNP atposition -20 (G-20A) results in the formation of annuclear factor-kappaB transcription factor-bindingsequence; however, as DEFB1 is constituatively expressed,the functional impact is unclear.[30] It has been demon-strated that DEFB1 function is compromised in cysticfibrosis.[32] The SNP at position -44 (C-44G) is associ-ated with Candida carriage in type I diabetics.[33] Varia-tion in the DEFB1 gene is associated with asthma,[31] acondition that shares some underlying characteristics withCOPD. The frequency of a SNP in exon 2 (Val38Ile) of theDEFB1 gene is significantly greater in male JapaneseTable 4: Beta-defensin-1 haplotypes and association with rate of decline in lung function.Imputed haplotypes (-20/-44/-52)Fast decline, n (%) No decline, n (%) p Value*ACA 24 (4.4) 18 (3.0) 0.26GCA 181 (32.9) 201 (33.1)GGA 3 (1.0) 1 (<1.0)ACG 223 (40.6) 257 (42.3)GCG 11 (2.0) 11 (1.8)AGG 6 (1.1) 1 (<1.0)GGG 102 (18.6) 119 (19.6)* p Value for global haplotype association calculated with a Wald statistic.Table 5: Defensin genotypes and infection outcomes.PolymorphismGenotype Physician visits‡§Part 1 p Value*Part 2 p Value†Days in bed‡ll Part 1 p Value*Part 2 p Value†DEFA1/DEFA3DEFA1 only 0.22 (0.07) 0.23 0.05 0.38 (0.16) 0.60 0.36DEFA1/DEFA3 0.28 (0.02) 0.51 (0.06)DEFB1 G-20AGG 0.27 (0.04) 0.85 0.88 0.49 (0.10) 0.86 0.98GA 0.26 (0.03) 0.48 (0.08)AA 0.30 (0.05) 0.54 (0.11)DEFB1 C-44GCC 0.28 (0.03) 0.81 0.42 0.53 (0.07) 0.93 0.14CG 0.27 (0.04) 0.44 (0.10)GG 0.18 (0.10) 0.28 (0.24)DEFB1 G-52AGG 0.29 (0.04) 0.67 0.62 0.47 (0.08) 0.81 0.29GA 0.25 (0.03) 0.47 (0.08)AA 0.30 (0.06) 0.62 (0.14)Definition of abbreviations: DEFA = alpha-defensin; DEFB = beta-defensin.*Part 1 of the two-part model involves modeling whether an individual has any physician visits or any days in bed due to lower respiratory infection. P values for global haplotype association derived from logistic regression analysis. Decline in lung function status was included as a factor in the analysis.†Part 2 of the two-part model involves modeling the average number of physician visits or days in bed due to lower respiratory infection given that it is greater than zero. P values for global haplotype association derived from gamma regression analysis. Decline in lung function status was included as a factor in the analysis.‡Mean (SE).Page 5 of 8(page number not for citation purposes)§Mean number of visits to a physician for lower respiratory infections per year.llMean number of days kept in bed for lower respiratory infections per year.Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76patients with COPD compared with healthy controls.[17]Our data indicate that this polymorphism is very rare inwhites. More recently, a C-44G polymorphism, but notthe Val38Ile polymorphism, was associated with COPD ina Chinese population.[18]Defensins could play a role in the pathogenesis of COPDand are excellent candidate genes for COPD associationstudies due to their expression patterns, location, andfunction. First, DEFA1–3 are expressed in neutrophils andDEFB1 is expressed in airway epithelial cells. Second,members of the alpha-defensin and beta-defensin fami-lies are located on chromosome 8p23, a genomic regionlinked to the development of airflow obstruction in smok-ers.[34] Finally, defensins may contribute to the patho-genesis of COPD by amplifying the cigarette smoke-induced and the infection-induced inflammatory reac-tions, which may lead to lung tissue injury.Table 6: Beta-defensin-1 haplotypes and infection outcomes.Haplotype Physician visits‡§Part 1 p Value* Part 2 p Value† Days in bed‡ll Part 1 p Value* Part 2 p Value†ACA 0.26 (0.08) 0.84 0.88 0.60 (0.19) 0.40 0.10GCA 0.27 (0.03) 0.50 (0.06)GGA 0.50 (0.27) 2.50 (0.63)ACG 0.28 (0.02) 0.50 (0.06)GCG 0.45 (0.11) 1.00 (0.27)AGG 0.23 (0.20) 0.60 (0.48)GGG 0.24 (0.04) 0.36 (0.08)*Part 1 of the two-part model involves modeling whether an individual has any physician visits or any days in bed due to lower respiratory infection. P values derived from the Wald statistic based on logistic regression results from Hapassoc. Decline in lung function status was included as a factor in the analysis.†Part 2 of the two-part model involves modeling the average number of physician visits or days in bed due to lower respiratory infection given that it is greater than zero. P values derived from the Wald statistic based on gamma regression results from Hapassoc. Decline in lung function status was included as a factor in the analysis.‡Mean (SE).§Mean number of visits to a physician for lower respiratory infections per year.llMean number of days kept in bed for lower respiratory infections per year.Table 7: Frequency of defensin polymorphisms by racial group.Polymorphism Genotype White (n = 579)† African American (n = 27)†p Value*DEFA1/DEFA3DEFA1 only 11.1 22.2 0.08DEFA1/DEFA3 88.9 77.8DEFB1 G-20AGG 30.1 44.4 0.21GA 48.5 44.5AA 21.4 11.1DEFB1 C-44GCC 64.8 100.0 0.0008CG 30.4 0.0GG 4.8 0.0DEFB1 G-52AGG 40.9 22.2 0.001GA 44.2 37.1AA 14.9 40.7Definition of abbreviations: DEFA = alpha-defensin; DEFB = beta-defensin.Page 6 of 8(page number not for citation purposes)* p Value derived from chi-square test.†Overall observed SNP frequency (%).Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76We conducted a population-based association study todetermine whether alpha-defensin and beta-defensin pol-ymorphisms influence smokers' susceptibility to lungfunction decline in a large population of white individu-als. The present genetic association study fulfills the majorcriteria for robust association studies[35] and has the fol-lowing strengths: 1) defensins are excellent COPD candi-date genes because defensins mediate inflammation andresponse to infection in the lung; 2) we included onlywhite subjects in the association study in order to mini-mize the potential effects of population stratification; 3)our study is a large association study involving 579 indi-viduals; 4) linkage data from a previous study suggeststhat 8p23 contains COPD susceptibility genes;[34] and 5)we have data on the frequency of lower respiratory infec-tions in the study subjects.The major concern with unrecognized population stratifi-cation is that of inflation of the type 1 error rate. However,there is also a possibility of type 2 error and this may haveinfluenced the lack of association observed in this study.The use of unlinked 'genomic control' markers may haveallowed this issue to be examined further. Unfortunately,genotyping of the requisite number of neutral, genomiccontrol loci to measure stratification was beyond thescope of this study.Analysis of individual polymorphisms and haplotypesdid not suggest any associations between the polymor-phisms/haplotypes and an increased rate of decline inlung function in smokers. In addition, the polymor-phisms were not important predictors of susceptibility toinfection, suggesting that they do not influence DEFA1/DEFA3-mediated or DEFB1-mediated responses to infec-tion.Power analysis indicates that our study was not under-powered. We estimated that for the DEFA1/DEFA3 copynumber polymorphism, the study design has eighty per-cent power to detect an association with a relative risk of1.98. For the DEFB1 -20, -44, and -52 SNPs, the studydesign has eighty percent power to detect associationswith relative risks of 1.60, 1.72, and 1.61, respectively.We found seven DEFB1 haplotypes with estimated fre-quencies ranging from <1.0% to 42.3% in the subjectgroup with no decline in lung function. These results aresimilar to previously reported data http://innateimmunity.net/IIPGA2/index_html. Some of the haplotypesreported in our study are very rare. In the studies by Dorket al.[28] and Matsushita et al.[17] where there was a totalof 103 German blood donors and 60 Japanese patientswith COPD, respectively, they may not have had largetion, racial/ethnic differences may account for the differ-ences observed.It is known that the frequencies of DEFB1 SNPs differbetween racial/ethnic groups.[30] We determined theprevalence of the genotypes across white and AfricanAmerican subjects and found that the DEFB1 G-52A SNPdiffered significantly between white and African Americanpopulations. In addition, we did not detect the G allele ofthe DEFB1 C-44G SNP in twenty-seven African Americansubjects suggesting that this allele is very rare or non-exist-ent in the African American population. It has beenshown that the C-44G SNP is monomorphic in the Afri-can American population http://innateimmunity.net/IIPGA2/index_html.In interpreting our data, a few limitations should be con-sidered. To our knowledge, this is the first time theDEFA1/DEFA3 copy number polymorphism and thethree DEFB1 SNPs have been tested for association withrate of decline in lung function in any population.Although we did not find association between the DEFA1/DEFA3 copy number polymorphism or DEFB1 polymor-phisms and rate of decline in lung function, these poly-morphisms may influence other COPD-relatedphenotypes. Furthermore, as we have only examined afew polymorphisms, we have not ruled out the possibilitythat other genetic variation in the alpha- and beta-defensins genes may be associated with the disease. Thegenetic basis of COPD may differ between racial/ethnicgroups. We limited the association study to white subjectsand some of the genotype frequencies differ significantlybetween white and African American racial groups. There-fore, it may be of interest to investigate the role of thesepolymorphisms in other racial/ethnic populations.ConclusionVariation in the DEFB1 gene is associated with COPD inboth Japanese and Chinese populations. Our results indi-cate that in a white population, the DEFA1/DEFA3 copynumber polymorphism and the three common DEFB1SNPs at positions -20, -44, and -52 do not influencesmokers' susceptibility to lung function decline or suscep-tibility to lower respiratory infection.Competing interestsThe author(s) declare that they have no competing inter-ests.Authors' contributionsAW carried out the molecular genetic studies and draftedthe manuscript. JH and JR carried out some of the geno-typing assays. KB performed the statistical analysis andPage 7 of 8(page number not for citation purposes)enough sample sizes to detect rare haplotypes. In addi- helped revise the manuscript. JC, NA, PP, and AS contrib-uted to the study conception and design, sample acquisi-Respiratory Research 2006, 7:76 http://respiratory-research.com/content/7/1/76tion, analysis and interpretation of the data, and helpedrevise the manuscript. All authors read and approved thefinal manuscript.AcknowledgementsThe authors thank Dr. David Erle for his insightful comments and sugges-tions. The authors gratefully acknowledge the National Heart, Lung, and Blood Institute for the recruitment and characterization of the study pop-ulation. This study was supported by a grant from the Canadian Institutes of Health Research and National Institutes of Heath grant 5R01HL064068-04. The Lung Health Study was supported by contract N01-HR-46002 from the Division of Lung Diseases of the National Heart, Lung, and Blood Insti-tute. Alison Wallace is supported by a Doctoral Research Award from the Michael Smith Foundation for Health Research. Andrew Sandford is the recipient of a Canada Research Chair.References1. Fletcher C, Peto R: The natural history of chronic airflowobstruction.  Br Med J 1977, 1:1645-1648.2. McCloskey SC, Patel BD, Hinchliffe SJ, Reid ED, Wareham NJ, LomasDA: Siblings of patients with severe chronic obstructive pul-monary disease have a significant risk of airflow obstruction.Am J Respir Crit Care Med 2001, 164:1419-1424.3. Sandford AJ, Pare PD: Genetic risk factors for chronic obstruc-tive pulmonary disease.  Clin Chest Med 2000, 21:633-643.4. Ganz T, Lehrer RI: Defensins.  Pharmacol Ther 1995, 66:191-205.5. Panyutich AV, Hiemstra PS, van Wetering S, Ganz T: Human neu-trophil defensin and serpins form complexes and inactivateeach other.  Am J Respir Cell Mol Biol 1995, 12:351-357.6. Van Wetering S, Mannesse-Lazeroms SP, Van Sterkenburg MA, DahaMR, Dijkman JH, Hiemstra PS: Effect of defensins on interleukin-8 synthesis in airway epithelial cells.  Am J Physiol 1997,272:L888-96.7. Paone K: Human neutrophil peptides stimulate am to produc-tion of LTB4 and IL-8 [abstract].  Am J Respir Crit Care Med 1999,159:A511.8. van Wetering S, Rahman I, Hiemstra PS, Macnee W: Role of intrac-ellular glutathione in neutrophil defensin-induced IL-8 syn-thesis and cytotoxicity in airway epithelial cells [abstract].Eur Respir J 1998, 12:420s.9. Lillard JWJ, Boyaka PN, Chertov O, Oppenheim JJ, McGhee JR:Mechanisms for induction of acquired host immunity by neu-trophil peptide defensins.  Proc Natl Acad Sci U S A 1999,96:651-656.10. Diamond G, Bevins CL: beta-Defensins: endogenous antibioticsof the innate host defense response.  Clin Immunol Immunopathol1998, 88:221-225.11. Ganz T, Selsted ME, Szklarek D, Harwig SS, Daher K, Bainton DF,Lehrer RI: Defensins. Natural peptide antibiotics of humanneutrophils.  J Clin Invest 1985, 76:1427-1435.12. McCray PBJ, Bentley L: Human airway epithelia express a beta-defensin.  Am J Respir Cell Mol Biol 1997, 16:343-349.13. Singh PK, Jia HP, Wiles K, Hesselberth J, Liu L, Conway BA, Green-berg EP, Valore EV, Welsh MJ, Ganz T, Tack BF, McCray PBJ: Pro-duction of beta-defensins by human airway epithelia.  ProcNatl Acad Sci U S A 1998, 95:14961-14966.14. Linzmeier R, Ho CH, Hoang BV, Ganz T: A 450-kb contig ofdefensin genes on human chromosome 8p23.  Gene 1999,233:205-211.15. Taudien S, Galgoczy P, Huse K, Reichwald K, Schilhabel M, SzafranskiK, Shimizu A, Asakawa S, Frankish A, Loncarevic IF, Shimizu N, Sid-diqui R, Platzer M: Polymorphic segmental duplications at8p23.1 challenge the determination of individual defensingene repertoires and the assembly of a contiguous humanreference sequence.  BMC Genomics 2004, 5:92.16. Mars WM, Patmasiriwat P, Maity T, Huff V, Weil MM, Saunders GF:Inheritance of unequal numbers of the genes encoding thehuman neutrophil defensins HP-1 and HP-3.  J Biol Chem 1995,270:30371-30376.obstructive pulmonary disease.  Biochem Biophys Res Commun2002, 291:17-22.18. Hu RC, Xu YJ, Zhang ZX, Ni W, Chen SX: Correlation ofHDEFB1 polymorphism and susceptibility to chronicobstructive pulmonary disease in Chinese Han population.Chin Med J (Engl) 2004, 117:1637-1641.19. Anthonisen NR, Connett JE, Kiley JP, Altose MD, Bailey WC, BuistAS, Conway WAJ, Enright PL, Kanner RE, O'Hara P: Effects ofsmoking intervention and the use of an inhaled anticholiner-gic bronchodilator on the rate of decline of FEV1. The LungHealth Study.  Jama 1994, 272:1497-1505.20. Schneider S, Roessli D, Excoffier L: Arlequin ver. 2.000: A soft-ware for population genetics data analysis.  Switzerland, Genet-ics and Biometry Laboratory, University of Geneva; 2000. 21. Stephens M, Smith NJ, Donnelly P: A new statistical method forhaplotype reconstruction from population data.  Am J HumGenet 2001, 68:978-989.22. Stephens M, Donnelly P: A comparison of bayesian methods forhaplotype reconstruction from population genotype data.Am J Hum Genet 2003, 73:1162-1169.23. Blough DK, Madden CW, Hornbrook MC: Modeling risk usinggeneralized linear models.  J Health Econ 1999, 18:153-171.24. Manning WG, Mullahy J: Estimating log models: to transform ornot to transform?  J Health Econ 2001, 20:461-494.25. McCullough P, Nelder JA: Generalized Linear Models.  BocaRaton, Chapman & Hall/CRC Press; 1989. 26. Burkett K, McNeney B, Graham J: A note on inference of traitassociations with SNP haplotypes and other attributes ingeneralized linear models.  Hum Hered 2004, 57:200-206.27. Lake SL, Lyon H, Tantisira K, Silverman EK, Weiss ST, Laird NM,Schaid DJ: Estimation and tests of haplotype-environmentinteraction when linkage phase is ambiguous.  Hum Hered2003, 55:56-65.28. Dork T, Stuhrmann M: Polymorphisms of the human beta-defensin-1 gene.  Mol Cell Probes 1998, 12:171-173.29. Vatta S, Boniotto M, Bevilacqua E, Belgrano A, Pirulli D, Crovella S,Amoroso A: Human beta defensin 1 gene: six new variants.Hum Mutat 2000, 15:582-583.30. Jurevic RJ, Chrisman P, Mancl L, Livingston R, Dale BA: Single-nucle-otide polymorphisms and haplotype analysis in beta-defensingenes in different ethnic populations.  Genet Test 2002,6:261-269.31. Levy H, Raby BA, Lake S, Tantisira KG, Kwiatkowski D, Lazarus R, Sil-verman EK, Richter B, Klimecki WT, Vercelli D, Martinez FD, WeissST: Association of defensin beta-1 gene polymorphisms withasthma.  J Allergy Clin Immunol 2005, 115:252-258.32. Goldman MJ, Anderson GM, Stolzenberg ED, Kari UP, Zasloff M, Wil-son JM: Human beta-defensin-1 is a salt-sensitive antibiotic inlung that is inactivated in cystic fibrosis.  Cell 1997, 88:553-560.33. Jurevic RJ, Bai M, Chadwick RB, White TC, Dale BA: Single-nucle-otide polymorphisms (SNPs) in human beta-defensin 1: high-throughput SNP assays and association with Candida car-riage in type I diabetics and nondiabetic controls.  J Clin Micro-biol 2003, 41:90-96.34. Palmer LJ, Celedon JC, Chapman HA, Speizer FE, Weiss ST, SilvermanEK: Genome-wide linkage analysis of bronchodilator respon-siveness and post-bronchodilator spirometric phenotypes inchronic obstructive pulmonary disease.  Hum Mol Genet 2003,12:1199-1210.35. Weiss ST: Association studies in asthma genetics.  Am J RespirCrit Care Med 2001, 164:2014-2015.36. O'Connor GT, Sparrow D, Weiss ST: A prospective longitudinalstudy of methacholine airway responsiveness as a predictorof pulmonary-function decline: the Normative Aging Study.Am J Respir Crit Care Med 1995, 152:87-92.37. Rijcken B, Schouten JP, Xu X, Rosner B, Weiss ST: Airway hyper-responsiveness to histamine associated with accelerateddecline in FEV1.  Am J Respir Crit Care Med 1995, 151:1377-1382.38. Tracey M, Villar A, Dow L, Coggon D, Lampe FC, Holgate ST: Theinfluence of increased bronchial responsiveness, atopy, andserum IgE on decline in FEV1. A longitudinal study in the eld-erly.  Am J Respir Crit Care Med 1995, 151:656-662.Page 8 of 8(page number not for citation purposes)17. Matsushita I, Hasegawa K, Nakata K, Yasuda K, Tokunaga K, KeichoN: Genetic variants of human beta-defensin-1 and chronic

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.52383.1-0074624/manifest

Comment

Related Items