Analyses of association between PPAR gamma and EPHX1 polymorphisms and susceptibility to COPD in a Hungarian cohort, a case-control study

RESEARCH ARTICLE Open Access Analyses o associa ion be ween PPAR gamma and EPHX1 polymo phisms and suscep ibili y o COPD in a Hunga ian coho , a case-con ol s udy And as Penyige 1*† , Szila d Poliska 2,3† , Esz e Csanky 5,9 , Bea a Schol z 3 , Balazs Dezso 6 , I an Schmelcze 1 , Iain Kil y 7 , Laszlo Takacs 8 , Laszlo Nagy 2,3,4* Abs ac Backg ound: In addi ion o smoking, gene ic p edisposi ion is belie ed o play a majo ole in he pa hogenesis o ch onic obs uc i e pulmona y disease (COPD). Gene ic associa ion s udies o new candida e genes in COPD may lead o imp o ed unde s anding o he pa hogenesis o he disease. Me hods: Two p oposed casual single nucleo ide polymo phisms (SNP) ( s1051740, s2234922) in mic osomal epoxide hyd olase (EPHX1) and h ee SNPs ( s1801282, s1800571, s3856806) in pe oxisome p oli e a o -ac i a ed ecep o gamma (PPARG), a new candida e gene, we e geno yped in a case-con ol s udy (272 COPD pa ien s and 301 con ols subjec s) in Hunga y. Allele equencies and geno ype dis ibu ions we e compa ed be ween he wo coho s and end es was also used o e alua e associa ion be ween SNPs and COPD. To es ima e he s eng h o associa ion, odds a ios (OR) (wi h 95% CI) we e calcula ed and po en ial con ounding a iables we e es ed in logis ic eg ession analysis. Associa ion be ween haplo ypes and COPD ou come was also assessed. Resul s: The dis ibu ion o impu ed EPHX1 pheno ypes was signi ican ly di e en be ween he COPD and he con ol g oup (P = 0.041), OR o he slow ac i i y pheno ype was 1.639 (95% CI = 1.08- 2.49; P = 0.021) in ou s udy. In logis ic eg ession analysis adjus ed o bo h a ian s, also age and pack-yea , he a e allele o His447His o PPARG showed signi ican associa ion wi h COPD ou come (OR = 1.853, 95% CI = 1.09-3.14, P = 0.0218). In haplo ype analysis he GC haplo ype o PPARG (OR = 0.512, 95% CI = 0.27-0.96, P = 0.035) con e ed educed isk o COPD. Conclusions: The “slow”ac i i y-associa ed geno ypes o EPHX1 we e associa ed wi h inc eased isk o COPD. The mino His447His allele o PPARG signi ican ly inc eased; and he haplo ype con aining he mino P o12Ala and he majo His447His polymo phisms o PPARG dec eased he isk o COPD. Backg ound Ch onic obs uc i e pulmona y disease (COPD) is an inc easing and se ious public heal h p oblem ep esen - ing he ou h leading cause o dea h globally. COPD is a complex human disease, associa ed wi h pe sis en ai - way in lamma ion, p o ease-an i-p o ease imbalance, oxida i e s ess, ch onic obs uc i e b onchi is and emphysema, esul ing in p og essi e ai low limi a ion ha is no subs an ially e e sed by b onchodila o s. Impo an ly, i is a smoking- ela ed diso de and ciga - e e smoking is he majo en i onmen al isk ac o o de elopmen o COPD.Howe e , he ac ha onlya subse o smoke s (15-20%) de elops clinically signi i- can symp oms sugges s ha gene ic p edisposi ion also plays ole in he de elopmen o COPD [1,2]. P e ious gene ic associa ion and genome-wide linkage s udies ha e iden i ied se e al candida e genes ha migh be in ol ed in he pa hogenesis o COPD [3-7]. In ou case-con ol s udy, i e pu a i e causal single nucleo ide polymo phisms (SNPs) in wo genes - * Co espondence: [email p o ec ed]b.hu; [email p o ec ed] †Con ibu ed equally 1 Depa men o Human Gene ics, Uni e si y o Deb ecen, Deb ecen, Hunga y 2 Depa men o Biochemis y and Molecula Biology, Resea ch Cen e o Molecula Medicine, Uni e si y o Deb ecen, Deb ecen, Hunga y Full lis o au ho in o ma ion is a ailable a he end o he a icle Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 © 2010 Penyige e al; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. mic osomal epoxide hyd olase (EPHX1) and pe oxisome p oli e a o -ac i a ed ecep o gamma (PPARG)-we e chosen o analyze hei associa ion wi h COPD. PPARG is a membe o nuclea ho mone ecep o s, implica ed in adipocy e di e en ia ion and in ol ed in mac ophage ac i a ion and dend i ic cell biology [8]. An an i-in lamma o y ole o PPARG was also epo ed based on i s inhibi o y e ec on p o-in lamma o y an- sc ip ion ac o s such as NF-BandAP-1[9,10].The NCBI SNP da abase ea u es mo e han 700 polymo ph- isms o PPARG, many o hem in onic o synonymous a ian and gene ally lack o in o ma ion ega ding popula ion di e si y. Se e al s udies on he polymo phisms o PPARG such as P o12Ala and His447His we e p e iously pe o med in ela ion o in lamma o y diseases such as IBD, ype 2 diabe es and ecen ly i has been sugges ed ha PPARG polymo phisms we e associa ed wi h he isk o as hma [11]. Since PPARG migh bein ol edin he egula ion o p o-in lamma o y signaling pa hways and i is gene - ally accep ed ha COPD is associa ed wi h an abno mal in lamma o y esponse, gene ic polymo phisms in PPARG could be implica ed in he suscep ibili y o COPD isk [12]. Tha may p o ide a heo e ical basis o he s udy o his new candida e gene in COPD de elopmen . Consequen ly, we examined he associa- ion be ween h ee SNP polymo phisms o PPARG gene and COPD. EPHX1 is an enzyme associa ed wi h he me abolism and de oxi ica ion o xenobio ic chemicals; i plays an impo an ole in he gene al oxida i e de ense o lung. Se e al polymo phisms a e known in EPHX1 including wo ela i ely common SNPs, he exon 3 Ty 113His ( s1051740) and exon 4 His139A g ( s2234922) a ian s. These wo a ian alleles ha e been sugges ed o be associa ed wi h al e ed EPHX1 enzyme ac i i y [13]. Subs i u ion o Ty 113 o His dec eases EPHX1 ac i i y (slow allele), whe eas subs i u ion o His139 o A g inc eases EPHX1 ac i i y ( as allele). I was ound ha he slow me abolizing o m o EPHX1 was associa ed wi h an inc eased isk o COPD and e idence suppo - ing his associa ion has been eplica ed in se e al case- con ol gene ic associa ion s udies [7,14,15]. Howe e , he e idence suppo ing his associa ion has no been consis en , se e al gene ic associa ion s udies ailed o show associa ion be ween hese polymo phisms and COPD [16,17]. In his s udy we ha e chosen he Ty 113His and His139A g polymo phisms o alida e ou Hunga ian popula ion o COPD associa ion s udies. All oge he a o al o i e SNPs in wo candida e genes we e geno yped in o de o examine hei asso- cia ion wi h COPD suscep ibili y by using Hunga ian coho s, a p e iously unin es iga ed popula ion wi h he highes COPD mo ali y a e among men in Eu - ope [18]. Me hods S udy popula ions The s udy is a case-con ol gene ic associa ion s udy in a Cen al-Eu opean Caucasian popula ion. In ou analy- sis he cases included 272 and he con ol subjec s included 301 age-ma ched Hunga ian indi iduals. The ec ui men and he clinical analyses o pa ien s we e conduc ed a he Depa men o Pulmonology, Medical and Heal h Science Cen e , Uni e si y o Deb ecen, acco ding o he Global Ini ia i e o Ch onic Obs uc- i e Lung Disease (GOLD) c i e ia. The Resea ch E hics Commi ee o Uni e si y o Deb ecen Medical and Heal h Science Cen e app o ed he clinical p o ocol and he s udy. W i en in o med consen was ob ained be o e he subjec s en e ed he s udy. The in es iga o explained he na u e, pu pose and isk o he s udy and p o ided he subjec wi h a copy o he in o ma ion shee . The subjec s we e hen gi en ime o conside he s udy’s implica ion be o e deciding o pa icipa e. Be o e s a ing sample collec ion, we de ined inclusion and exclusion c i e ia o diseased and heal hy pa ien s. Inclusion c i e ia o COPD subjec s we e age 40 o 65 yea s old. Pa ien s mus ha e p edic ed alue o o ced expi a o y olume a 1 second (FEV1) 50%-80% and FEV1/FVC% <70% (s age 2 acco ding o GOLD c i e ia). Inclusion c i e ia o con ol pa ien s we e age be ween 40 and 65 yea s; no mal spi ome y, FEV1 ≥90% (p e- dic ed alue) and FEV1/FVC% ≥80%. All pa ien s mus be cu en o ex-smoke (minimum 15 pack yea s). Exclusion c i e ia o all pa ien s: plasma IgE le el >70 U/ml, alpha1-an ipsin de iciency, e idence o as hma, a opic disease, his o y o lung diso de s, espi a o y in ec ion in he pas 3 mon hs, o he in lamma o y dis- eases (e.g. in lamma o y bowel disease, heuma oid a h i is, pso iasis e c.), au oimmune diseases (lupus, scle osis), cance , posi i e plasma es o HIV, Hepa i is B o Hepa i is C. Geno yping Genomic DNA was ex ac ed om pe iphe al blood using E.Z.N.A. Blood DNA Midi Ki (Peqlab Bio echno- logie) acco ding o he manu ac u e ’s p o ocol. Quali y o he DNA samples was checked by aga ose gel- elec opho esis and quan i a ed by NanoD op1000. All SNPs we e geno yped using TaqMan geno yping assays (Addi ional File 1, Table S1). Samples we e measu ed in duplica es and nuclease- ee wa e was used as no- empla e con ol. Following PCR ampli ica ion he end- poin luo escence was ead wi h he ABI 7900 HT ins umen and geno ypes we e assigned using SDS Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 2 o 9 so wa e (Applied Biosys ems). The a e age geno yping success a e o a leas 95% was a ained o each SNP. Al eola mac ophage (AM) and pe iphe al blood monocy e (MO) collec ion B onchoal eola la age luid (BALF) samples we e col- lec ed by ibe -op ic b onchoscopy om heal hy and COPD pa ien s. AMs we e sepa a ed by Pe coll (Ame - sham Biosciences) g adien cen i uga ion. To al cell numbe was de e mined by coun ing in hemocy ome e . Di e en ial cell coun was assessed on hema oxylin- eosin s ained cy ospin slides be o e and a e he g adi- en sepa a ion. O e 95% AM pu i y was eached a e sepa a ion. 50 ml hepa in ea ed enous blood was collec ed om heal hy and diseased pa ien s. MOs we e sepa a ed by Ficoll g adien cen i uga ion using an i-CD14 conju- ga ed mic obeads (>98% MO) (Va ioMACS, Mil enyi Bio ec.). 5-5 con ol and COPD pa ien s we e ec ui ed in his expe imen . TaqMan RT-QPCR To al RNA was isola ed om pe iphe al blood mono- cy es, al eola mac ophages, lung issue and adipose is- sue using T izol eagen (In i ogen). Fi s s and cDNA was gene a ed om 5 ug o al RNA using cDNA A chi e Ki (Applied Biosys ems). Fo RT-QPCR eac- ion 200 ng cDNA/sample and 2X TaqMan PCR mix (Applied Biosys ems) was used. Reac ions we e un in ABI P ism HT 7900 ins umen . We ha e designed p i- me pai and p obe o measu ing PPARG mRNA le el. Fo wa d p ime : 5’GATGACAGCGACTTGGCAA, e e se p ime : 5’CTTCAATGGGCTTCACATTCA, p obe: 5’FAM-CAAACCTGGGCGGTCTCCACTGAG- 3’TAMRA. The housekeeping gene cyclophilin A was used as a no malize gene: o wa d p ime : 5’ACGGC- GAGCCCTTGG, e e se p ime : 5’TTTCTGCTG TCTTTGGGACCT, p obe: 5’FAM-CGCGTCTCC TTTGAGCTGTTTGCA-3’TAMRA. Rela i e gene exp ession le els we e calcula ed by compa a i e C me hod. S a is ical analysis was pe o med in G aph- Pad P ism using non-pa ame ic es (Mann-Whi ney U- es ). Immunohis ochemis y Tissues o mo phology and immunos ainings we e ob ained om he iles o he Pa hology Depa men o Uni e si y o Deb ecen and we e eshly ixed in 10% neu al o malinandembeddedinpa a in ollowedby hema oxyline and eosine (HE) s aining using s anda d me hods. Immunohis ochemis y (IHC) o al eola mac ophages (AM) was ca ied ou using PPARG, CD68 and DCSign (San a C uz) monoclonal an ibodies by means o immunope oxidase s aining as desc ibed ea lie [19,20]. Double immuno luo escence s aining was pe o med as desc ibed ea lie [21] using CSAII de ec- ion ki wi h FITC-labeled y amine ollowed by an immuno luo escen s aining o DCSign wi h s ep a i- din- exas ed luo och ome. Nuclea coun e s aining was made wi h DAPI (blue luo escence). S a is ical analysis Di e ences be ween cases and he con ol g oup con- ce ning demog aphic and main clinical da a we e ana- lyzed by Mann-Whi ney U- es and Pea son c 2 es . Geno ype da a o each SNP we e es ed o depa u es om Ha dy-Weinbe g equilib ium (HWE) sepa a ely in case and con ol popula ions using a goodness-o - i c 2 - es o he exac es o es ima e P alues. HWE cal- cula ionswe edonebyusing heHWE oolh p://ihg. gs .de/cgi-bin/hw/hwa1.pl. The signi icance o di e - ences in geno ype and allele equencies be ween pa ien s and con ols we e es ed by using ei he c 2 ana- lyses o Fishe ’s exac es whe e app op ia e [22]. To assess he deg ee o associa ion be ween each o he SNPs and COPD odds a ios wi h 95% con idence in e - als (OR, 95% CI) we e calcula ed using logis ic eg es- sion analysis; he model was adjus ed o SNPs, pack- yea , age and gende . All single locus associa ion es s we e pe o med using he STATA 9.0 s a is ical package (excep whe e o he wise s a ed). Haplo ype equencies we e es ima ed o con ol and pa ien g oups sepa a ely wi h he Full-P ecise-I e a ion algo i hm implemen ed in he SHEsis so wa e h p:// analysis.bio-x.cn/myAnalysis.php. The ex en o linkage disequilib ium be ween pai s o biallelic ma ke s was de e mined using bo h he s anda dized disequilib ium and co ela ion coe icien s (gi en as Lewon in’sD’and 2, espec i ely) and associa ion be ween haplo ypes and COPD was assessed by he c 2 - es o he exac es as implemen ed in he p og am SHEesis [23,24]. Co ec- ion o mul iple es ing was no used in he analysis o he associa ion geno ype and allele equencies because: (i) he EPHX1 polymo phisms we e known o be unc- ional and (ii) he gene is conside ed a suscep ibili y gene o COPD; and (iii) in case o he PPARG poly- mo phisms he s udied indi idual alleles we e no inde- penden . A pos e io i es ima es o s udy powe we e assessed by means o Quan o so wa e h p://hyd a.usc. edu. We ha e es ima ed he powe o ou s udy wi h he ollowing pa ame e s: sample size o 272 cases; con ol/ case a io o 1.1; mino allele equencies (MAF) a e in he ange om 0.12 o 0.3; log-addi i e model; disease p e alence o COPD 5%. Assuming hese pa ame e s ou s udy had ~50% powe o de ec a geno ype ela i e isk (GRR) o 1.4 o MAF = 0.12, o ~79% powe o de ec aGRRo 1.6 o hesameMAF;whilei had ~76% powe o de ec a GRR o 1.4, and ~96% powe o Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 3 o 9 de ec a GRR o 1.6 o MAF = 0.3 a an a= 0.05 signi - icance le el. Resul s O he 573 subjec s geno yped, 61.8% o he subjec s we e men. The p opo ion o males was highe among cases (69.85% o 54.48%) bu he e is no signi ican di - e ence in he mean age o con ols and cases. Cases had been exposed o mo e obacco smoke as e idenced by he di e ence in pack-yea s bu he di e ence is no signi ican , COPD pa ien s had a much la ge educ ion in lung unc ion, ypical o a clinical COPD popula ion (Table 1). All geno ype equencies we e consis en wi h Ha dy- Weinbe g equilib ium o bo h SNPs o he EPHX1 gene, and he geno ype and allele equencies did no di e signi ican ly be ween cases and he con ol g oup (Table 2). The assessmen o he associa ion o indi i- dual SNPs wi h COPD showed ha homozygosi y o he mino allele inc eased he isk o disease in case o Ty 113His polymo phism ("slow”allele) (OR = 1.345; 95% CI = 0.96 - 1.91; P = 0.095) educed i in case o he His139A g SNP (" as ”allele) (OR = 0.675; 95% CI = 0.27 - 1.69; P = 0.399). Howe e , none o hese SNPs we e signi ican ly associa ed wi h COPD e en a e adjus ing he model o gende , age and pack-yea s in logis ic eg ession. F equencies o he ou SNP based haplo ypes we e es ima ed o cases and con ols. Al hough he “slow ac i i y”CA (His 113 -His 139 )haplo ypewasmo e e- quen among cases (24.8% e sus he 20.8% in con ols), he o e all dis ibu ion did no di e signi ican ly be ween he wo g oups (P = 0.736). Fu he mo e, alleles o he wo loci in EPHX1 a e in comple e linkage equilib ium as shown by he pai -wise s anda dized dise- quilib ium coe icien (D’= 0.036). Due o he p esence o hese coding a ian s, ma ked a ia ions in EPHX1 ac i i y ha e been epo ed p e- iously. The e o e we ha e assessed he associa ion o he p edic ed ( apid”,“no mal”,“slow”and “ e y slow”) EPHX1 pheno ypes wi h he de elopmen o COPD [25,26]. The dis ibu ion o p edic ed EPHX1 ac i i y was signi ican ly di e en be ween con ol subjec s and COPD pa ien s (P = 0.041). In he analysis o p edic ed pheno ypes he COPD g oup had highe p opo ion o he p edic ed “slow”pheno ype. Consequen ly he slow pheno ype signi ican ly aises he isk o de eloping COPD [OR = 1.639; 95% CI = 1.06-2.49; P = 0.021)] in ou case con ol s udy (Table 3). High exp ession le el o PPARG mRNA in he lung has been epo ed p e iously [27,28]. We examined PPARG exp ession a p o ein le el in su gical lung issue samples. PPARG p o ein is exp essed in he lung and mos o he PPARG p o ein de i ed signals co-localized wi h he exp ession o a ypical mac ophage ma ke CD68 and he dend i ic cell ma ke DCSign (Figu e 1). PPARG exp ession was also measu ed a mRNA le el. mRNA exp ession is en iched in AM ela i e o o al lung issue, howe e we did no ind di e ences in exp ession le el be ween COPD and heal hy indi iduals (Figu e 2). Thus, gene ic a ian s, a he han he exp ession o he PPARG gene could be associa ed wi h he de elopmen o COPD. We ha e geno yped h ee exonic SNPs ( s10801282 (P o12Ala), s3856806 (His447His) and s1800571 (P o113Gln))inPPARG gene, bu he s1800571 locus was le ou om he analysis because i was homozy- gous o he majo C allele in all indi iduals. The o he wo SNPs we e in HWE and showed linkage disequili- b ium, he ex en o LD be ween s10801282 and s3856806 was ound o be D’= 0.673, al hough he pai showed lowe LD wi h espec o hei co ela ion coe - icien ( 2 = 0.42). The single loci allelic and geno ypic analysis ound no signi ican associa ion be ween he wo coding a - ian s o PPARG and COPD. In logis ic eg ession applying a model adjus ed o bo h SNPs, age and pack-yea s, he a e a ian o His447His polymo ph- ism was signi ican ly associa ed wi h inc eased odds o COPD (OR = 1.853, 95% CI = 1.09 - 3.14, P = 0.021). The mino Ala allele o he P o12Ala a ian had an OR o 0.679, sugges ing a p o ec i e e ec , howe e i did no each signi icance (95% CI = 0.40 - 1.14, P = 0.145) (Table 4). We ha e es ima ed he equency o he possible wo-SNP haplo ypes o PPARG and assessed he asso- cia ion be ween haplo ypes and COPD de elopmen . The e was a signi ican di e ence in he equency o he GC haplo ype in ol ing he a e G a ian o he P o12Ala locus be ween he wo g oups (P = 0.035). The s eng h o associa ion was also assessed, o his haplo ype (OR = 0.512; 95% CI = 0.27-0.96). This ind- ing sugges s a p o ec i e e ec o he GC (12Ala/ 447His) haplo ype o he PPARG gene o COPD ou - come (Table 5). Table 1 Clinical ea u es o he s udy popula ion Pa ame e Cases (N = 272) Con ols (N = 301) P alue Male (%) 190 (69.85) 164 (54.48) 0.781 Age (±SD)* 63.87 (±8.96) 64.29 (±9.07) 0.577 Pack-Yea s (±SD)* 38.75 (±19.91) 34.76 (±14.71) 0.120 FEV1% p edic ed (±SD)* 47.17 (±13.69) 99.28 (±9.76) <0.0001 FEV1/FVC% p edic ed (±SD) * 57.57 (±10.12) 87.10 (±35.24) <0.001 Da a p esen ed as mean ± SD. FEV1: o ced expi a o y olume in one second, FVC: o ced i al capaci y, *Mann-Whi ney U- es was used. Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 4 o 9 Discussion The aims o his s udy we e o in es iga e associa ion o EPHX1 polymo phisms o COPD in a Hunga ian popu- la ion and o assess possible associa ion be ween SNPs o PPARG, a new candida e gene, and COPD ou come. In ou s udy o indi idual SNPs in EPHX1 ( he exon-3 Ty 113His and exon-4 His139A g a ian s), he homozy- gous mino Th 113His a ian showed only a bo de line associa ion wi h he COPD pheno ype (P = 0.095), how- e e he le el o associa ion was u he educed when he model was adjus ed o age, sex and pack-yea s. The e is comple e linkage equilib ium be ween Ty 113His and His139A g SNPs and no s a is ical sig- ni icance was ound in haplo ype equency dis ibu ion and hei associa ion wi h COPD pheno ype. Since EPHX1 is in ol ed in he de oxi ica ion o epox- ide in e media es in obacco smoke, he a e o con e - sion o hese highly eac i e compounds could a ec an indi idual’s abili y o cope wi h he oxic e ec o ciga - e e smoke [29]. We ha e econs uc ed he equency o p edic ed EPHX1 ac i i y in pa ien and con ol g oups [25,26]. The di e ence in dis ibu ion o p edic ed phe- no ypes was signi ican be ween he wo g oups. The COPD g oup had a highe p opo ion o p edic ed slow and lowe p opo ion o p edic ed no mal and apid EPHX1 ac i i y. In e es ingly he con ol g oup showed an excess o e y slow pheno ypes, bu i s equency was low in bo h g oups. In ou analysis he slow ac i i y a - ian o EPHX1 enzyme was associa ed wi h a signi ican ly inc eased he isk o COPD. This esul p o ides addi- ional suppo o he no ion ha EPHX1 is likely o be in ol ed in COPD pa hogenesis. P e ious s udies abou he po en an i-in lamma o y p ope ies o he PPARG agonis s sugges he use o PPARG ligands in COPD he apy [12,30,31] and associa- ion o PPARG gene polymo phisms wi h he de elopmen o as hma has been epo ed [11]. These lines o e idence p omp ed us o in es iga e PPARG gene associa ion exis ed be ween he p esence o ce ain PPARG gene poly- mo phisms and COPD ou come. Using IHC s aining and RT-QPCR measu emen s, we ha e con i med PPARG mRNA and p o ein exp ession in lung issues and pa icu- la in AM [32]. Al hough gene exp ession le el is compa - able in pa ien s and heal hy indi iduals, polymo phisms o he gene s ill could be po en ial candida e ma ke s o he disease i a unc ionally al e ed PPARG has a ole in he de elopmen in COPD. Among he h ee SNPs we ha e geno yped in PPARG - s1801282, s3856806 and s1800571 - he s1800571 was excluded since all indi iduals ca ied an iden ical homozygous geno ype [33]. Ou single-ma ke es s o he o he wo coding a ian s yielded a signi ican asso- cia ion o he mino allele o His447His polymo phism in logis ic eg ession adjus ed o bo h SNPs, age, sex and pack-yea s (OR = 1.853; 95% CI = 1.09-3.14; P = 0.02). The His447His a ian did no cause amino acid change and i has no known unc ion. Howe e , se e al pape s poin ed ou ha exonic synonymous SNPs can a ec mRNA splicing o s abili y [34,35]. Any change in hese p ocesses could exe a signi ican e ec on p o- ein unc ion, he e o e he e was a eason o in es iga e he associa ion o his SNP wi h COPD. O cou se he possibili y, ha he His447His SNP is igh ly linked wi h an unknown unc ional a ian ha de e mine COPD suscep ibili y can no be excluded. A modes pai -wise LD was ound be ween s1801282 and s3856806. Since he use o SNP-based haplo ypes Table 3 The dis ibu ion o he p edic ed EPHX1 pheno ypes P edic ed EPHX1 ac i i y Con ols n (%) Cases n (%) § P alue Con ingency ables OR (95% CI) P alue No mal 144 (53.1) 123 (48.4) 1 ( e e ence) Slow 55 (20.2) 77 (30.3) 1.64 (1.08-2.49) 0.021 Ve y slow 17 (6.2) 9 (3.5) 0.041 0.62 (0.27-1.44) 0.306 Rapid 55 (20.2) 45 (17.7) 0.96 (0.59-3.19) 0.855 No mal: exon3 Ty /Ty and exon 4 His/His o exon 3 Ty /His and exon 4 His/ A g; Slow: exon 3 Ty /Ty and exon 4 His/His; Ve y slow: exon 3 His/His and exon 4 His/His; Rapid: exon 3 y /Ty and exon 4 A g/A g o His/A g § c 2 - es was used o he dis ibu ion o p edic ed pheno ypes, OR: odds a io, CI: con idence in e al Table 2 Allele and geno ype equencies o examined EPHX1 gene polymo phisms Gene Symbol SNP ID Allele equency Geno ype equency § Ha dy-Weinbe g Equilib ium OR (95% CI) s1051740 (Ty 113His) T C TT (%) TC (%) CC (%) P alue Con ols 0.723 0.277 154 (53.3) 110 (38.1 25 (8.7) 0.401 1.11 (0.86-1.44) Cases 0.701 0.299 127 (47.4) 122 (45.5) 19 (7.1) 0.154 s2234922 (His139A g) A G AA (%) AG (%) GG (%) Con ols 0.779 0.221 171 (60.0) 102 (35.8) 12 (4.2) 0.507 0.88 (0.66-1.18) Cases 0.799 0.201 169 (62.8) 92 (64.2) 8 (2.9) 0.280 § c 2 - es was used, OR: odds a io, CI: con idence in e al Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 5 o 9 in gene ic associa ion s udies may o e a mo e powe ul app oach han he use o indi idual SNPs, a haplo ype analysis was also pe o med. A signi ican di e ence was ound in he equency o GC haplo ype (con aining he mino G allele o P o12Ala and majo C o His447His a ian ) be ween he con ol and COPD g oups and he associa ion o his haplo ype o COPD ou come was also de e mined. The GC haplo ype con e s a signi ican lowe isk o COPD, poin ing o a po en ial unc ional p o ec i e e ec o his haplo ype. In e es ingly he mino allele o P o12Ala polymo ph- ism ha lowe s he binding a ini y o PPARG p o ein o Figu e 1 Mo phology and immunohis ochemis y o COPD-associa ed lung lesions.A: Ch onic b onchi is. B: Cen iacina emphysema. C: Ad anced ac i e b onchi is wi h ib osis and emphysema. A-C, hema oxylin-eosin s aining; D: CD68-PPARG coexp ession wi h double IHC s aining using alkaline phospha ase [ ed cy oplasm-CD68] and diamino-benzidine [b own nuclei-PPARG]. E: Cells wi h ed luo escence and g een nuclei DCSign-PPARG double luo escence s aining. Nuclea coun e -s aining is DAPI. Indica ions:b, b onchus; a, al eola spaces; a ows, al eola mac ophages. O iginal magni ica ions: A-D 20×; E40×. Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 6 o 9 Figu e 2 mRNA exp ession o PPARG. PPARG showed as high as mRNA exp ession le el in al eola mac ophages as in subcu an a and also showed exp ession in he whole lung issue wi h signi ican ly lowe le el (Mann-Whi ney U es ). Howe e i was no exp essed in pe iphe al blood monocy es. Da a p esen ed no malized alues o RT-QPCR measu emen s; Cyclophilin A was used as housekeeping gene. G ey ba s ep esen mean alues o 5 con ol pa ien s; whi e ba s ep esen mean alues o 5 COPD pa ien s. ** p < 0.01. Table 4 Allele and geno ype equencies o examined PPARG gene polymo phisms Gene Symbol SNP ID Allele equency Geno ype equency § Ha dy-Weinbe g Equilib ium Logis ic Analysis s1801282 (P o12Ala) C G CC (%) CG (%) GG (%) P alue OR (95% CI) # P alue Con ols 0.864 0.136 217 (75.2) 64 (22.4) 7 (2.4) 0.398 0.68 (0.40-1.14) 0.15 Cases 0.874 0.126 199 (76.2) 67 (22.3) 4 (1.5) 0.869 s3856806 (His447His) C T CC (%) CT (%) TT (%) P alue OR (95% CI) # P alue Con ols 0.882 0.118 224 (78.8) 53 (18.7) 8 (2.5) 0.09 1.85 (1.09-3.14) 0.02 Cases 0.862 0.138 199 (74.0) 65 (24.5) 5 (1.5) 0.57 § c 2 - es was used, OR: odds a io, CI: con idence in e al # Adjus ed o age and pack-yea Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 7 o 9 pe oxisome p oli e a o esponse elemen was ound o be associa ed wi h lowe body mass index, imp o ed insulin sensi i i y, dec eased isk o ype 2 diabe es [36,37] and educed isk o de elop colo ec al cance [38]. Highe equency o T allele o he His447His C/T polymo phism was obse ed in colon cance [39,40]. Wea eawa eo he ac ha signi ican esul s could p o e o be alse posi i es, and a clea limi a ion o ou s udy is he ela i ely low sample size. The s udy has o he limi a ions such as ha popula ion s a i ica ion should be in es iga ed in hese kinds o s udies and we did no analyze a second coho o eplica e ou esul s. Possible gene-gene and gene- en i onmen in e ac ions pose a di icul y o gene ic analysis o COPD associa ion s udies, oo. Fu he s u- dies using la ge popula ions a e needed and o he a - ian s in he PPARG gene should be in es iga ed in o de o cla i y he associa ion o PPARG and indi i- dual suscep ibili y o he de elopmen o COPD. Conclusions In summa y, ou s udy p o ided suppo o he sug- ges ed causa i e ole o EPHX1 polymo phisms and phe- no ypes impu ed om exon 3 and exon 4 geno ype da a in COPD ou come in a Hunga ian popula ion. We ha e ca ied ou he i s in es iga ion o PPARG gene polymo phisms in a case-con ol COPD s udy and cha ac e ized he associa ion be ween indi idual SNPs and haplo ypes in PPARG and suscep ibili y o COPD. Al hough he GC haplo ype has a modes p o ec i e e ec , i migh poin owa d he po en ial impo ance o common alleles wi h weak e ec in he e ogeneous dis- eases, like COPD. The documen a ion o PPARG haplo- ype associa ion wi h COPD iden i ies his impo an gene as a a ge o u he in es iga ion o he pa ho- genesis o COPD and as a po en ial a ge o he apy. Addi ional ma e ial Addi ional File 1: Table S1. Cha ac e is ics o es ed SNP. Cha ac e is ics, NCBI e e ence numbe s and ABI assays code o examined single nucleo ide polymo phisms. Abb e ia ions COPD: ch onic obs uc i e pulmona y disease; EPHX1: mic osomal epoxide hyd olase; FEV1: o ced expi a o y olume in one second; FVC: o ced i al capaci y; PPARG: pe oxisome p oli e a o -ac i a ed ecep o gamma. Acknowledgemen s The au ho s a e indeb ed o Zsuzsa Bodná o clinical coo dina ion, D A ila Vaskó, D Pé e Szabó, D Sándo Sz. Kiss, Ti anilla Tölgyesi, Má ia Ráduly, o he help wi h clinical sample collec ion. The au ho s would like o hank he expe echnical assis ance o Júlia Buslig, Ibolya Fü ös and Ma a Béládi. This wo k was suppo ed by g an s om he Hunga ian Na ional O ice o Resea ch and Technology (NKFP 1/007/01, NKFP 1A/008/04), Hunga ian g an o he Na ional Resea ch Fund (NI 67877 and TAMOP-4.2.2/08/1). Au ho de ails 1 Depa men o Human Gene ics, Uni e si y o Deb ecen, Deb ecen, Hunga y. 2 Depa men o Biochemis y and Molecula Biology, Resea ch Cen e o Molecula Medicine, Uni e si y o Deb ecen, Deb ecen, Hunga y. 3 Clinical Genomics Cen e , Medical and Heal h Science Cen e , Resea ch Cen e o Molecula Medicine, Uni e si y o Deb ecen, Deb ecen, Hunga y. 4 Apop osis and Genomics Resea ch G oup o he Hunga ian Academy o Sciences, Resea ch Cen e o Molecula Medicine, Uni e si y o Deb ecen, Deb ecen, Hunga y. 5 Depa men o Pulmonology, Medical and Heal h Science Cen e , Uni e si y o Deb ecen, Deb ecen, Hunga y. 6 Depa men o Pa hology, Uni e si y o Deb ecen, Medical and Heal h Science Cen e , Deb ecen, Hunga y. 7 P ize Global Resea ch and De elopmen , Sandwich, UK. 8 Biosys ems In e na ional SAS, E y, F ance. 9 Depa men o Pulmonology, Semmelweis Heal h Ca e Cen e o Miskolc, Miskolc, Hunga y. Au ho s’con ibu ions LN is an In e na ional Schola o HHMI and holds a Wellcome T us Senio Resea ch Fellowship in Biomedical Sciences, he planned and di ec ed he s udy. AP pe o med da a analyses and p epa ed he manusc ip . SP pe o med he geno yping measu emen s and RT-QPCR analyses and p epa ed igu es and ables. ECs o ganized pa ien ec ui men and sample collec ion. BS o ganized sample p epa a ion. BD pe o med IHC s aining. IS pe o med da a analyses. IK and LT pa icipa ed in he planning and he design o he s udy and helped p epa e he manusc ip . All au ho s ha e ead and app o ed he inal e sion o he manusc ip . Compe ing in e es s LN has no con lic s o in e es s (COIs) o disclose. AP has no COIs o disclose. SP has no COIs o disclose. ECs has no COIs o disclose. BS has no COIs o disclose. BD has no COIs o disclose. IS has no COIs o disclose. IK is di ec o o P ize and holde o P ize s ock. LT has no COIs o disclose. Recei ed: 10 May 2010 Accep ed: 2 No embe 2010 Published: 2 No embe 2010 Re e ences 1. Sand o d AJ, Sil e man EK: Ch onic obs uc i e pulmona y disease. 1: Suscep ibili y ac o s o COPD he geno ype-en i onmen in e ac ion. Tho ax 2002, 57(8):736-741. 2. Sei a C, Plagens A: Gene ics o ch onic obs uc i e pulmona y disease. In J Ch on Obs uc Pulmon Dis 2007, 2(4):541-550. 3. Joos L, Pa e PD, Sand o d AJ: Gene ic isk ac o s o ch onic obs uc i e pulmona y disease. Swiss Med Wkly 2002, 132(3-4):27-37. 4. Sand o d AJ, Joos L, Pa e PD: Gene ic isk ac o s o ch onic obs uc i e pulmona y disease. Cu Opin Pulm Med 2002, 8(2):87-94. 5. Mol ino NA: Gene ics o COPD. Ches 2004, 125(5):1929-1940. 6. He sh CP, DeMeo DL, Sil e man EK: Na ional Emphysema T ea men T ial s a e o he a : gene ics o emphysema. P oc Am Tho ac Soc 2008, 5(4):486-493. 7. He sh CP, Demeo DL, Lange C, Li onjua AA, Reilly JJ, Kwia kowski D, Lai d N, Syl ia JS, Spa ow D, Speize FE: A emp ed eplica ion o epo ed ch onic obs uc i e pulmona y disease candida e gene associa ions. Am J Respi Cell Mol Biol 2005, 33(1):71-78. 8. Sza ma i I, Nagy L: Nuclea ecep o signalling in dend i ic cells connec s lipids, he genome and immune unc ion. EMBO J 2008, 27(18):2353-2362. Table 5 The dis ibu ion and s eng h o associa ion o PPARG haplo ypes Haplo ypes COPD ( eq.) Con ol ( eq.) § P alue OR (95% CI) CC 0.834 0.832 0.914 1.02 (0.74-1.40) CT 0.039 0.030 0.424 1.31 (0.68-2.50) GC 0.028 0.053 0.035 0.51 (0.27-0.96) GT 0.098 0.085 0.429 1.18 (0.78-1.77) § c 2 - es was used, OR: odds a io, CI: con idence in e al Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 8 o 9 9. Asada K, Sasaki S, Suda T, Chida K, Nakamu a H: An iin lamma o y oles o pe oxisome p oli e a o -ac i a ed ecep o gamma in human al eola mac ophages. Am J Respi C i Ca e Med 2004, 169(2):195-200. 10. Spea s M, McSha y C, Thomson NC: Pe oxisome p oli e a o -ac i a ed ecep o -gamma agonis s as po en ial an i-in lamma o y agen s in as hma and ch onic obs uc i e pulmona y disease. Clin Exp Alle gy 2006, 36(12):1494-1504. 11. Oh SH, Pa k SM, Lee YH, Cha JY, Lee JY, Shin EK, Pa k JS, Pa k BL, Shin HD, Pa k CS: Associa ion o pe oxisome p oli e a o -ac i a ed ecep o - gamma gene polymo phisms wi h he de elopmen o as hma. Respi Med 2009, 103(7):1020-1024. 12. S andi o d TJ, Keshamouni VG, Reddy RC: Pe oxisome p oli e a o - ac i a ed ecep o -{gamma} as a egula o o lung in lamma ion and epai . P oc Am Tho ac Soc 2005, 2(3):226-231. 13. Kiyoha a C, Yoshimasu K, Takayama K, Nakanishi Y: EPHX1 polymo phisms and he isk o lung cance : a HuGE e iew. Epidemiology 2006, 17(1):89-99. 14. He sh CP, Demeo DL, Laza us R, Celedon JC, Raby BA, Bendi JO, C ine G, Make B, Ma inez FJ, Scanlon PD: Gene ic associa ion analysis o unc ional impai men in ch onic obs uc i e pulmona y disease. Am J Respi C i Ca e Med 2006, 173(9):977-984. 15. He sh CP, DeMeo DL, Reilly JJ, Sil e man EK: Xenobio ic me abolizing enzyme gene polymo phisms p edic esponse o lung olume educ ion su ge y. Respi Res 2007, 8:59. 16. B ogge J, S een VM, Eiken HG, Guls ik A, Bakke P: Gene ic associa ion be ween COPD and polymo phisms in TNF, ADRB2 and EPHX1. Eu Respi J2006, 27(4):682-688. 17. Chappell S, Daly L, Mo gan K, Gue a-Ba anes T, Roca J, Rabino ich R, Lo ya J, Milla AB, Donnelly SC, Kea ings V: Gene ic a ian s o mic osomal epoxide hyd olase and glu ama e-cys eine ligase in COPD. Eu Respi J 2008, 32(4):931-937. 18. Hu d S: The impac o COPD on lung heal h wo ldwide: epidemiology and incidence. Ches 2000, 117(2 Suppl):1S-4S. 19. Sza ma i I, Gogolak P, Im JS, Dezso B, Rajna olgyi E, Nagy L: Ac i a ion o PPARgamma speci ies a dend i ic cell sub ype capable o enhanced induc ion o iNKT cell expansion. Immuni y 2004, 21(1):95-106. 20. Tsaki is I, Soos G, Nemes Z, Kiss SS, And as C, Szan o J, Dezso B: The p esence o ca boxypep idase-M in umou cells signi ies epide mal g ow h ac o ecep o exp ession in lung adenoca cinomas: he coexis ence p edic s a poo p ognosis ega dless o EGFR le els. J Cance Res Clin Oncol 2008, 134(4):439-451. 21. Gogolak P, Re hi B, Sza ma i I, Lanyi A, Dezso B, Nagy L, Rajna olgyi E: Di e en ia ion o CD1a- and CD1a+ monocy e-de i ed dend i ic cells is biased by lipid en i onmen and PPARgamma. Blood 2007, 109(2):643-652. 22. Chiano MN, Clay on DG: Geno ypic ela i e isks unde o de ed es ic ion. Gene Epidemiol 1998, 15(2):135-146. 23. Shi YY, He L: SHEsis, a powe ul so wa e pla o m o analyses o linkage disequilib ium, haplo ype cons uc ion, and gene ic associa ion a polymo phism loci. Cell Res 2005, 15(2):97-98. 24. Li Z, Zhang Z, He Z, Tang W, Li T, Zeng Z, He L, Shi Y: A pa i ion-liga ion- combina ion-subdi ision EM algo i hm o haplo ype in e ence wi h mul iallelic ma ke s: upda e o he SHEsis [h p://analysis.bio-x.cn]. Cell Res 2009, 19(4):519-523. 25. Smi h CA, Ha ison DJ: Associa ion be ween polymo phism in gene o mic osomal epoxide hyd olase and suscep ibili y o emphysema. Lance 1997, 350(9078):630-633. 26. Benhamou S, Reinikainen M, Boucha dy C, Daye P, Hi onen A: Associa ion be ween lung cance and mic osomal epoxide hyd olase geno ypes. Cance Res 1998, 58(23):5291-5293. 27. Huang TH, Razmo ski-Naumo ski V, Ko a BP, Lin DS, Rou ogalis BD: The pa hophysiological unc ion o pe oxisome p oli e a o -ac i a ed ecep o -gamma in lung- ela ed diseases. Respi Res 2005, 6:102. 28. Bel isi MG, Hele DJ, Bi ell MA: Pe oxisome p oli e a o -ac i a ed ecep o gamma agonis s as he apy o ch onic ai way in lamma ion. Eu J Pha macol 2006, 533(1-3):101-109. 29. Omiecinski CJ, Hasse C, Hosag aha a V: Epoxide hyd olase– polymo phism and ole in oxicology. Toxicol Le 2000, 112-113:365-370. 30. Cai o S, Yang SR, Kode A, Edi isinghe I, Rajend asozhan S, Phipps RP, Rahman I: Rosigli azone and 15-deoxy-Del a12,14-p os aglandin J2, PPARgamma agonis s, di e en ially egula e ciga e e smoke-media ed p o-in lamma o y cy okine elease in monocy es/mac ophages. An ioxid Redox Signal 2008, 10(2):253-260. 31. Pa el HJ, Bel isi MG, Bishop-Bailey D, Yacoub MH, Mi chell JA: Ac i a ion o pe oxisome p oli e a o -ac i a ed ecep o s in human ai way smoo h muscle cells has a supe io an i-in lamma o y p o ile o co icos e oids: ele ance o ch onic obs uc i e pulmona y disease he apy. J Immunol 2003, 170(5):2663-2669. 32. Reddy RC: Immunomodula o y ole o PPAR-gamma in al eola mac ophages. J In es ig Med 2008, 56(2):522-527. 33. Ris ow M, Mulle -Wieland D, P ei e A, K one W, Kahn CR: Obesi y associa ed wi h a mu a ion in a gene ic egula o o adipocy e di e en ia ion. N Engl J Med 1998, 339(14):953-959. 34. Capon F, Allen MH, Ameen M, Bu den AD, Tillman D, Ba ke JN, T emba h RC: A synonymous SNP o he co neodesmosin gene leads o inc eased mRNA s abili y and demons a es associa ion wi h pso iasis ac oss di e se e hnic g oups. Hum Mol Gene 2004, 13(20):2361-2368. 35. Sauna ZE, Kimchi-Sa a y C, Ambudka SV, Go esman MM: Silen polymo phisms speak: how hey a ec pha macogenomics and he ea men o cance . Cance Res 2007, 67(20):9609-9612. 36. Deeb SS, Fajas L, Nemo o M, Pihlajamaki J, Mykkanen L, Kuusis o J, Laakso M, Fujimo o W, Auwe x J: A P o12Ala subs i u ion in PPARgamma2 associa ed wi h dec eased ecep o ac i i y, lowe body mass index and imp o ed insulin sensi i i y. Na Gene 1998, 20(3):284-287. 37. Al shule D, Hi schho n JN, Klannema k M, Lindg en CM, Vohl MC, Nemesh J, Lane CR, Scha ne SF, Bolk S, B ewe C: The common PPARgamma P o12Ala polymo phism is associa ed wi h dec eased isk o ype 2 diabe es. Na Gene 2000, 26(1):76-80. 38. Landi S, Mo eno V, Gioia-Pa icola L, Guino E, Na a o M, de Oca J, Capella G, Canzian F: Associa ion o common polymo phisms in in lamma o y genes in e leukin (IL)6, IL8, umo nec osis ac o alpha, NFKB1, and pe oxisome p oli e a o -ac i a ed ecep o gamma wi h colo ec al cance . Cance Res 2003, 63(13):3560-3566. 39. Jiang J, Gajalakshmi V, Wang J, Ku iki K, Suzuki S, Nakamu a S, Akasaka S, Ishikawa H, Tokudome S: In luence o he C161T bu no P o12Ala polymo phism in he pe oxisome p oli e a o -ac i a ed ecep o -gamma on colo ec al cance in an Indian popula ion. Cance Sci 2005, 96(8):507-512. 40. Siezen CL, an Leeuwen AI, K am NR, Luken ME, an K anen HJ, Kampman E: Colo ec al adenoma isk is modi ied by he in e play be ween polymo phisms in a achidonic acid pa hway genes and ish consump ion. Ca cinogenesis 2005, 26(2):449-457. P e-publica ion his o y The p e-publica ion his o y o his pape can be accessed he e: h p://www.biomedcen al.com/1471-2350/11/152/p epub doi:10.1186/1471-2350-11-152 Ci e his a icle as: Penyige e al.: Analyses o associa ion be ween PPAR gamma and EPHX1 polymo phisms and suscep ibili y o COPD in a Hunga ian coho , a case-con ol s udy. BMC Medical Gene ics 2010 11:152. Submi you nex manusc ip o BioMed Cen al and ake ull ad an age o : • Con enien online submission • Tho ough pee e iew • No space cons ain s o colo igu e cha ges • Immedia e publica ion on accep ance • Inclusion in PubMed, CAS, Scopus and Google Schola • Resea ch which is eely a ailable o edis ibu ion Submi you manusc ip a www.biomedcen al.com/submi Penyige e al.BMC Medical Gene ics 2010, 11:152 h p://www.biomedcen al.com/1471-2350/11/152 Page 9 o 9