Neuropeptide S Receptor 1 Gene Polymorphism Is Associated With Susceptibility to Inflammatory Bowel Disease

Neuropeptide S Receptor 1 Gene Polymorphism Is Associated With Susceptibility to Inflammatory Bowel Disease
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Neuropeptide S Receptor 1 Gene Polymorphism Is Associated WithSusceptibility to Inflammatory Bowel Disease MAURO D’AMATO,* ,‡ SARA BRUCE, § FRANCESCA BRESSO,* ,‡ MARCO ZUCCHELLI, § SINI EZER,  ,¶  VILLE PULKKINEN,  CECILIA LINDGREN, § MARCO ASTEGIANO, # MARIO RIZZETTO, # PAOLO GIONCHETTI,** GABRIELE RIEGLER, ‡‡ RAFFAELLO SOSTEGNI,  MARCO DAPERNO,  SANDRA D’ALFONSO, ¶¶ PATRICIA MOMIGLIANO–RICHIARDI, ¶¶ LEIF TORKVIST, ## PAULI PUOLAKKAINEN,*** MAARIT LAPPALAINEN, ‡‡‡,§§§ PAULINA PAAVOLA–SAKKI, ‡‡‡,  LEENA HALME, ¶¶¶ MARTTI FARKKILA,  ULLA TURUNEN,  KIMMO KONTULA, ‡‡‡ ROBERT LOFBERG, ###, ****SVEN PETTERSSON,* ,‡,‡‡‡‡ and JUHA KERE §,  ,¶ *Strategic Research Center IRIS,  ‡ Department of Microbiology, Tumor and Cell Biology,  § Department of Biosciences and Nutrition,  ## Division of Colorectal Surgery,IBD Unit, and ****Department of Medicine, Karolinska Institutet, Stockholm, Sweden;   Department of Medical Genetics, ***Department of Surgery,  ‡‡‡ Department of Medicine,   Department of Gastroenterology,  ¶¶¶ Department of Transplantation and Liver Surgery, University of Helsinki, Helsinki, Finland;  §§§ Research program for Molecular Medicine, Biomedicum Helsinki, Helsinki, Finland;  ¶ Folkhalsan Institute of Genetics, Helsinki, Finland;  # Clinic of Gastrohepatology DU, Molinette Hospital,Torino, Italy;   Gastroenterology Division, ASO Ordine Mauriziano, Torino, Italy; **Division of Internal Medicine, S. Orsola-Malpighi Hospital, Bologna, Italy;  ‡‡ Gastroenterology Unit, 2nd University of Naples, Naples, Italy;  ¶¶ Department of Medical Sciences, University of Eastern Piedmont and IRCAD, Novara, Italy;  ### IBDUnit, HMQ Sophia Hospital, Stockholm, Sweden; and the  ‡‡‡‡ Genome Institute of Singapore, Singapore Background & Aims:   The neuropeptide S receptor(NPSR1) gene has been associated recently withasthma and maps in a region of chromosome 7 pre- viously linked also to inflammatory bowel disease(IBD). NPSR1 is expressed on the epithelia of several organs including the intestine, and appears to beup-regulated in inflammation. We tested NPSR1 genepolymorphism for association with IBD and veri-fied whether the expression of its 2 major isoforms(NPSR1-A and NPSR1-B) is altered in the intestineof IBD patients.  Methods:   Eight NPSR1 polymor-phisms were genotyped in 2490 subjects from 3 co-horts of IBD patients and controls from Italy,Sweden, and Finland. Real-time polymerase chain re-action and immunohistochemistry were used toquantify NPSR1 messenger RNA (mRNA) and pro-tein expression in intestinal biopsy specimens fromIBD patients and controls.  Results:   Global analysisof the whole dataset identified strong association of aNPSR1 haplotype block with IBD (  P   .0018) and its2 major forms: Crohn’s disease (CD) (  P     .026) and ulcerative colitis (UC) (  P     .003). Genetic effectscaused by individual haplotypes were identified mainly for the predisposing haplotype H2 in CD (  P   .0005) and the protective haplotype H8 in UC (  P    .003). NPSR1 mRNA and protein levels were in-creased in IBD patients compared with controls, and the risk haplotype H2 correlated with higher expres-sion of both NPSR1-A (  P   .024) and NPSR1-B (  P   .047) mRNAs.  Conclusions:   NPSR1 polymorphismis associated with IBD susceptibility. Specific NPSR1alleles might act as genetic risk factors for chronicinflammatory diseases of the epithelial barrier or-gans. C rohn’s disease (CD) and ulcerative colitis (UC) areinflammatory bowel diseases (IBDs) characterizedby a chronic relapsing inflammation of the gastrointes-tinal tract. 1  Among others, symptoms include chronicdiarrhea, rectal bleeding, abdominal pain, and extrain-testinal manifestations such as arthritis, uveitis, and skinmanifestations. Although showing a largely overlappingphenotype, which hampers a differential diagnosis in upto 20% of IBD cases, CD and UC have some distinctiveclinical features. CD is characterized by a more penetrat-ing, transmural inflammation that can affect any tract of the mucosa in the gut, whereas UC shows progressiveinflammation that does not extend beyond the largeintestine. These diseases appear to result from an impair-ment of gut epithelial barrier and a dysregulated immuneresponse toward intestinal bacteria, with the latter lead-ing to the establishment of chronic inflammation. 2 Thefundamental role of the intestinal flora in IBD also ishighlighted by the study of animal models of disease, inwhich the phenotype is strictly dependent on the pres-ence of bacteria in the gut, and does not develop inanimals raised in germ-free conditions. 3 Similar to other inflammatory disorders of the barrierorgans such as asthma, eczema (or atopic dermatitis),and psoriasis, there are environmental, immunologic,and genetic factors that play a major role in the predis-position to IBD. Remarkably, several loci of genetic link-age have been identified that appear to be relevant to  Abbreviations used in this paper:  CI, confidence interval; FDR, falsediscovery rate; LD, linkage disequilibrium; NPSR1, neuropeptide Sreceptor; OR, odds ratio; PCR, polymerase chain reaction; SNP, singlenucleotide polymorphism. © 2007 by the AGA Institute0016-5085/07/$32.00doi:10.1053/j.gastro.2007.06.012  C L  I   NI    C AL  –AL  I   ME  NT  AR Y T  R A C T   GASTROENTEROLOGY 2007;133:808–817  more than one of these diseases, 4 and the existence of common predisposing genes has therefore been postu-lated. In particular, genetic predisposition to IBD hasbeen recognized since the late 1980s, when familiar clus-tering was shown 5–7 and a higher level of disease concor-dance among monozygotic twins was first reported. 8 More recently, genome-wide linkage scans in IBD fami-lies with different ethnic backgrounds have providedevidence of the contribution of several chromosomalregions to IBD predisposition, 9–13 and a number of sus-ceptibility genes have been identified by positional clon-ing such as NOD2/CARD15, DLG5, and SLC22A4/SLC22A5. 14 The neuropeptide S receptor (NPSR1) gene (also called GPR154  or  G-protein–coupled receptor for asthma susceptibil-ity ), which codes for a 7-transmembrane receptor of stillpoorly characterized function, recently has received muchattention because of its identification as a susceptibility locus for asthma and related traits. 15–19 The NPSR1 genemaps to chromosome 7p14, within a large region of chromosome 7 previously linked to IBD in a genome-widelinkageanalysisperformedonUKfamilies. 13 NPSR1,aswellas its ligand neuropeptide S, is found on the epithelia of several organs including the small and large intestine, andits expression seems to increase with inflammation, as pre- viously shown in asthmatic patients. 15 Based on the data, we considered NPSR1 as a candi-date gene for IBD and sought to determine whethergenetic variation at its locus can influence susceptibility to this disease. NPSR1 polymorphisms have been impli-cated previously in the pathogenesis of asthma and re-lated traits in several populations. 15–18  Although no in-dividual single nucleotide polymorphism (SNP) wasfound to be responsible for the observed associations, inall these studies the genetic information was containedwithin a haplotype block formed by 7 SNPs (rs323917,rs323922, rs324377, hopo546333, rs324384, rs324396,and rs740347), mapping within a 70-kb region betweenexons 2 and 3 of the NPSR1 gene (see the Appendix forelectronic database information). In addition, a recentanalysis of NPSR1 polymorphism in Chinese asthmaticpatients provided some evidence of association of a func-tional SNP (Asn107Ile, dbSNP rs324981) with airway hyperresponsiveness (a hallmark of asthma). 19 We thusselected these 8 SNPs to test our working hypothesis, andgenotyped them in 3 independent cohorts of IBD pa-tients and controls from Italy, Sweden, and Finland.Moreover, we studied NPSR1 expression in intestinaltissue from IBD patients and controls, and analyzedthese results in relation to genotype.Our results provide evidence of an involvement of NPSR1 in the genetic susceptibility to both CD and UC,and therefore extend previous findings on this gene inasthma by implicating NPSR1 polymorphism in the pre-disposition to another chronic inflammatory disorder of the barrier organs. Materials and Methods Study Subjects  Three independent cohorts of IBD patients andethnically matched controls from Italy, Sweden, and Fin-land were included in this study. The Italian cohortconsisted of a consecutive series of 474 IBD patients (277CD and 197 UC) recruited in 2004–2005 at IBD clinics of the Molinette Hospital in Torino, the S. Orsola-MalpighiHospital in Bologna, and the Department of ClinicalExperimental Medicine of the 2nd University of Naples,and 259 unrelated blood donors as controls. The Swedishstudy population was composed of 500 IBD patients (221CD and 279 UC) from the IBD units of the Sophia Hospital and the Department of Medicine at Karolinska Institute in Stockholm, and a random sample of 278ethnically matched but otherwise unselected controls.Consecutive Finnish patients with sporadic and familialIBD (235 CD and 455 UC) were recruited from thecapital area of Helsinki. Part of the familial cases srci-nated from IBD families that were recruited by collabo-rators at the University Hospitals in Finland. A total of 291 unrelated controls were healthy voluntary blood do-nors from the Finnish Red Cross Blood Service in Hel-sinki (kindly provided by Dr T. Krusius).Diagnosis of IBD (CD or UC) was assessed in all casesaccording to established clinical criteria including endo-scopic, radiologic, and histopathologic examination. Pa-tients with indeterminate colitis were excluded. Data on familial history of disease were available for the Italianand Finnish cohorts (14% and 23% of IBD patients,respectively).For NPSR1 expression analysis, intestinal biopsy spec-imens were collected from macroscopically affected sitesin 22 IBD patients (10 CD and 12 UC) and unaffectedareas in 12 asymptomatic controls (undergoing screeningfor colorectal cancer), during planned routine colonos-copy. Most of the biopsy specimens were from descend-ing colon in both IBD patients and controls.Informed consent was obtained from all participantsincluded in this study, which was approved by the re-sponsible local Italian, Swedish, and Finnish ethics com-mittees. Genotyping  Genomic DNA was isolated from peripheral bloodleukocytes and biopsy specimens using standard proce-dures (DNA extraction kits from Qiagen, Hilden, Ger-many). Eight NPSR1 SNPs (dbSNPs rs323917, rs323922,rs324377, rs324384, rs324396, rs740347, rs324981, andSNP hopo546333) were genotyped using matrix-assistedlaser desorption/ionization time-offlight mass spectrom-etry (SEQUENOM Inc., San Diego, CA) as previously described. 17  All samples were analyzed in a sex-specificassay to evaluate the accuracy in sample and databasemanagement and, for each SNP, 2 independent scorers     C    L    I    N    I    C    A    L   –    A    L    I    M    E    N    T    A    R    Y    T    R    A    C    T September 2007 NPSR1 AND IBD 809  confirmed all genotypes, which were produced with anaverage success rate of greater than 92%. Hardy–Weinbergcalculations were performed to ensure that each markerwas within allelic population equilibrium, and no signif-icant deviation was observed for any of the SNPs tested inthe Italian, Swedish, and Finnish controls.  NPSR1 Messenger RNA Expression Analysis  NPSR1-A and NPSR1-B messenger RNA (mRNA)expression levels were measured in biopsy specimens andthe Human Blood Fractions multiple tissue cDNA(MTC) panel (Clontech, Mountain View, CA) by perform-ing real-time polymerase chain reaction (PCR) in an ABIPrism 7500 Sequence Detection System (Applied Biosys-tems, Foster City, CA), according to the manufacturer’sinstructions. PCR primers were designed with PrimerExpress 2.0 software (Applied Biosystems), and real-timePCR reactions were performed in triplicate for each gene.HEK293 cells stably transfected with either NPSR1-A orNPSR1-B complementary DNAs (cDNAs) were used ascontrols to confirm the specificity of the PCR. Afternormalization to the endogenous housekeeping gene, thelevel of expression of NPSR1-A and NPSR1-B isoforms ineach sample was determined by the comparative C T method of relative quantification, 20 and expressed in ar-bitrary units relative to a randomly chosen referencesample.For the analysis of biopsy specimens, total RNA wasisolated from specimens kept in RNAlater (Ambion, Aus-tin, TX) by a 2-step purification method using Trizolreagent (Invitrogen, Carlsbad, CA) and subsequentclean-up with RNeasy kits (Qiagen), according to manu-facturers’ instructions. One microgram of total RNA wasused for cDNA synthesis, which was performed withthe SuperScript First-Strand Synthesis System (In- vitrogen) and an oligo(dT) 12-18  in the reaction mixture.Real-time PCR primers were CCCCCTCATCTACTGT-GTCTTCA (forward) and TCTCTCCCGGAACGTCAT-TCT (reverse) for NPSR1-A, CCCCCTCATCTACTGT-GTCTTCA (forward) and TCGTTGAGGGCAGAGCATTA(reverse) for NPSR1-B, and CCACATCGCTCAGACACCAT(forward) and GCGCCCAATACGACCAAAT (reverse) forthe endogenous control GAPDH.The Human Blood Fractions MTC Panel, containingnormalized cDNA from resting and activated leukocytes,was used to measure NPSR1 mRNA expression in differ-ent cell types. Real-time PCR reactions were performed aspreviously described, 21 and relative gene expression dif-ferences are presented as fold-changes in resting vs acti- vated cells.  Immunohistochemistry NPSR1 protein expression was analyzed by stain-ing formalin-fixed, paraffin-embedded biopsy specimensof intestinal tissue from IBD patients and controls with2 antibodies specific for the alternative carboxy termini of NPSR1-A and NPSR1-B. The characterization of theseantibodies and their specificities have been reported ear-lier. 21,22 The slides were heated in a microwave oven in 10mmol/L citrate buffer, pH 6.0 for 5 minutes, and immu-nohistochemistry was performed using the ABC methodwith Vectastain Elite ABC kit (Vector Laboratories, Bur-lingame, CA). Omission of primary antibodies and stain-ing with pre-immune sera were used as negative controls. Statistical Analysis   A total of 2490 individuals from 3 populations of Italian, Swedish, and Finnish IBD patients and controlswas analyzed in this study. With an association  P   value of .005 (to take into account multiple testing), we estimatedthat this sample size was sufficient to explore relativerisks as low as 1.3 at a power of about 90%. Differences ingender proportions between cases and controls weretested with    2 statistics. Linkage disequilibrium (LD),haplotype block structure (solid spine of LD), and allelicassociations were investigated in Haploview 3.2. 23 Globaland haplotype-specific association tests were performedusing the software haplo.stats package (v1.2) imple-mented in R (see the Appendix for software informa-tion). 24,25 This program first infers haplotype pair prob-abilities for each individual, using a standard EMalgorithm, and then calculates both global and haplo-type-specific score statistics for the distribution of prob-ability-weighted haplotype frequencies between cases andcontrols. Only haplotypes with frequencies greater than2% were considered, to avoid spurious results, and bothglobal and haplotype-specific  P   values were estimatedempirically with 10,000 permutations of the disease sta-tus. The effects of individual haplotypes and alleles in all3 populations combined were investigated with a Man-tel–Haenszel analysis in STATA 9.0, with haplotypecounts derived from the frequencies estimated in hap-lo.stats. Multiple testing was taken into account for thecombined analysis of the whole dataset by correctingempiric  P   values with the false discovery rate (FDR)method, 26 as implemented in the package multtest 1.8.2in R. Significant associations were investigated furtherfor recessive effects, using a logistic regression for theSNP rs324981, and    2 analysis on the number of ho-mozygous carriers in patients and controls (estimatedfrom posterior haplotype probabilities) for the haplotypeH2.Statistical analysis of NPSR1 expression was per-formed by comparing its mRNA levels in intestinal bi-opsy specimens from IBD patients and controls with a Mann–Whitney test as implemented in R. For the corre-lation with genotype, NPSR1 haplotypes were first in-ferred from individual SNP genotypes using the EM-based algorithm SNPHAP (see Appendix), and only haplotypes identified with a probability of greater than90% were considered for the analysis. After stratificationaccording to haplotype carrier status, differences in  C L  I   NI    C AL  –AL  I   ME  NT  AR Y T  R A C T   810 D’AMATO ET AL GASTROENTEROLOGY Vol. 133, No. 3  NPSR1 expression were analyzed with a Mann–Whitney test as described previously.Reported  P   values are based on 2-sided tests, and wereconsidered significant when less than .05 in all analyses. Results  Association of NPSR1 Polymorphism With IBD Susceptibility The demographic characteristics of the Italian,Swedish, and Finnish study populations are reported inTable 1. After genotype data were collected, a Haploview anal-ysis of LD in the entire sample showed a single haplotypeblock (block 1) to be formed by the 7 noncoding SNPs(rs323917, rs323922, rs324377, hopo546333, rs324384,rs324396, and rs740347), corresponding to 8 differenthaplotypes (haplotypes H1–H8), whereas the Asn107IleSNP rs324981 lay outside this block (Figure 1).  A global test of association between the NPSR1 locusand IBD on the entire dataset of 1664 IBD patients and826 controls from Italy, Sweden, and Finland providedstrong evidence of association with IBD (global  P    .0018), CD (global  P   .026), and UC (global  P   .003). As reported in Table 2, significant associations also were found when individual NPSR1 variants were tested fortheir specific contribution to IBD susceptibility. In par-ticular, the haplotypes H2 and H8 were associated, re-spectively, with predisposing and protective effects inIBD (  P   .003 and  P   .010). A separate analysis of UCand CD patients showed that these effects mainly wereaccounted for by the strong associations of the predis-posing haplotype H2 with CD (  P     .0005), and theprotective haplotype H8 with UC (  P     .003). These as-sociations remained significant after FDR correction formultiple comparisons (Table 2) and, as previously re- ported also for NPSR1 haplotypes associated withasthma, 15–18 they were not driven by any particular SNPin the haplotype block 1 (not shown). In an attempt toinvestigate whether recessive effects also could be de-tected, we compared the number of H2 homozygotes inpatients and controls (no H8 homozygotes were identi-fied). Although some trend was observed, H2 homozy-gosity was not associated significantly with an increasedrisk of disease (  P     .231; odds ratio [OR], 1.29; 95%confidence interval [CI], 0.84–2.02).The Asn107Ile polymorphism (SNP rs324981) showedevidence of a marginal effect on CD (  P   .040) and IBD(  P     .026), although these did not retain statistical sig-nificance after correction for multiple comparisons (Ta-ble 2). Of note, however, further analysis of this SNPusing a recessive model reinforced its association with Table 1.  Demographics of IBD Patients and Controls NMean age  SD, y Sex a  M (%) F (%)Italy CD 277 35.3  12.9 155 (55.9) 125 (45.1)UC 197 32.9  13.1 115 (58.4) 82 (41.6)Controls 257 30.8  10.2 108 (42.0) 149 (58.0)SwedenCD 221 34.4  12.1 117 (52.5) 105 (47.5)UC 279 33.7  15.6 160 (57.4) 119 (42.6)Controls 278 41.3  11.1 129 (46.4) 149 (53.6)FinlandCD 235 40.5  13.1 100 (42.6) 135 (57.4)UC 455 44.7  13.9 242 (53.2) 213 (46.8)Controls 291 45.5  3.1 156 (53.6) 135 (46.4) a  Male and female counts were significantly different in Italian CD ( P   .003) and UC patients ( P   .0008), Swedish UC patients ( P   .012),and Finnish CD patients ( P   .015). Figure 1.  LD of the studied NPSR1SNPs. (  A  ) LD and haplotype block structure obtained by Haploview 3.2analysis of the whole dataset (Italian,Swedish, and Finnish samples com-bined). The numbers in each box cor-respond to LD coefficient D’ betweenrespectiveSNPs(onlyvaluesforLD  100%arereported).TheSNPsareor-deredaccordingtotheirpositioninthegene, and NPSR1 direction of trans-cription is shown on top. ( B  ) Controlfrequencies for the NPSR1 variants(haplotypes from block 1 and SNPrs324981 alleles) in the 3 individualpopulations, and in the entire dataset.     C    L    I    N    I    C    A    L   –    A    L    I    M    E    N    T    A    R    Y    T    R    A    C    T September 2007 NPSR1 AND IBD 811  both CD (  P   .007; OR, 1.42; 95% CI, 1.10–1.83) and IBD(  P   .009; OR, 1.34; 95% CI, 1.07–1.67).To estimate the relative contributions of associated variants in individual populations, we also performedan association analysis on the 3 separate samples of IBD patients and controls from Italy, Sweden, andFinland, which confirmed and extended the findingsobtained on the entire dataset. The CD-predisposingeffect of the haplotype H2 was detected in Italy (  P    .016; OR, 1.33; 95% CI, 0.98–1.81) and Sweden (  P    .018; OR, 1.44; 95% CI, 1.06–1.97), and a similar trend,although not significant, was observed in Finland (  P   .342; OR, 1.16; 95% CI, 0.85–1.58). Although in theFinnish samples the frequency of the haplotype H8was too low to draw statistical conclusions, this hap-lotype was associated significantly with a reduced riskof UC both in Italy (  P     .016; OR, 0.13; 95% CI,0.02–0.56) and Sweden (  P     .015; OR, 0.40; 95% CI,0.17–0.89). Interestingly, some population-specific as-sociations also were found for the haplotype H1 andCD in Italy (  P     .005; OR, 0.75; 95% CI, 0.58–0.98)and the haplotype H5 and UC in Finland (  P     .022;OR, 0.67; 95% CI, 0.47–0.95). No significant associa-tions were found for the Asn107Ile polymorphism(SNP rs324981) in individual populations, althoughsome trends were observed in a recessive model of association (OR in IBD vs controls was 1.29 [95% CI,0.87–1.92] for Italy; 1.47 [95% CI, 0.99–2.19) for Swe-den; and 1.26 [95% CI, 0.89–1.8] for Finland).Finally, although patient and control groups hadsignificantly different proportions of men and women(Table 1), no changes in the genetic effect estimates were found when sex was included as a covariate in ourstatistical analyses (not shown).Taken together, and similar to previous studies inasthma and asthma-related traits, 15–19 these resultstherefore strongly implicate NPSR1 polymorphismalso in the genetic susceptibility to IBD.  Increased NPSR1 Expressionin Inflammation The expression of NPSR1 has been shown previ-ously to increase with inflammation in the bronchi of asthmatic patients. 15 To evaluate if a similar phenome-non takes place also in IBD, we analyzed the expressionof NPSR1 in intestinal biopsy specimens from 22 IBDpatients (10 CD and 12 UC) and 12 controls. Alternativesplicing occurs at the NPSR1 locus, giving rise to a number of different isoforms previously described. 22  Among these, NPSR1-A and NPSR1-B are the most abun-dant and best characterized, and encode the only NPSR1full-length proteins that reach the cell membrane intransfection experiments. 22 We therefore measured by quantitative reverse-transcriptase PCR the level of expres-sion of these 2 NPSR1 mRNAs. As shown in Figure 2,although a large variability was found among individualsthe expression of both NPSR1-A and NPSR1-B trans-cripts generally was higher in IBD patients than in con-trols. A Mann–Whitney analysis of the results confirmedmost of these differences to be statistically significant,with  P   values of .002 and .006 for the expressionof NPSR1-A, and .004 and .06 for the expression of NPSR1-B in CD and UC patients vs controls, respectively.To further characterize the expression of NPSR1 inIBD, we also performed an immunohistochemicalanalysis of intestinal specimens from 5 IBD patientsand 4 controls with 2 antibodies specific for the alter-native carboxy termini of NPSR1-A and NPSR1-B pro-teins. 15,21,22 This confirmed and extended the results ob-tained at the mRNA level. As previously shown in otherstudies, 15,22 both isoform-specific antibodies gave a clearpositive staining at the level of the epithelial cell layer of the intestinal mucosa, with NPSR1-A generally moreabundant than NPSR1-B (Figure 3). Remarkably, how- ever, NPSR1 epithelial staining was stronger in the tissuefrom IBD patients, where its expression also was detected Table 2.  Associations Between NPSR1 Haplotypes and IBD in a Combined Analysis of 3 Cohorts of IBD Patients andControls Controls(n  826)CD(n  733)UC(n  931)IBD(n  1664) a  Freq Freq OR (95% CI)  P   P FDR  Freq OR (95% CI)  P   P FDR  Freq OR (95% CI)  P   P FDR Block 1H1 0.347 0.316 0.88 (0.75–1.02) .080 .177 0.328 0.90 (0.78–1.04) .144 .348 0.323 0.88 (0.78–0.99) .050 .133H2 0.196 0.248 1.36 (1.14–1.61)  .0005 .003  0.223 1.16 (0.98–1.37) .076 .305 0.233 1.25 (1.08–1.45)  .003 .024 H3 0.124 0.114 0.90 (0.73–1.12) .359 .387 0.127 1.04 (0.85–1.28) .693 .886 0.121 0.99 (0.83–1.19) .924 .960H4 0.069 0.067 1.02 (0.77–1.35) .884 .596 0.075 1.03 (0.79–1.33) .839 .886 0.071 1.03 (0.82–1.30) .798 .960H5 0.088 0.071 0.80 (0.62–1.04) .098 .177 0.078 0.85 (0.66–1.08) .174 .348 0.075 0.83 (0.67–1.03) .089 .178H6 0.085 0.090 1.06 (0.83–1.36) .647 .582 0.078 1.02 (0.80–1.31) .855 .886 0.083 1.03 (0.83–1.27) .792 .960H7 0.053 0.053 0.97 (0.71–1.33) .845 .596 0.056 1.02 (0.76–1.38) .886 .886 0.054 1.01 (0.77–1.31) .960 .960H8 0.031 0.023 0.73 (0.47–1.13) .154 .208 0.015 0.49 (0.30–0.79)  .003 .024  0.019 0.61 (0.42–0.89)  .010 .039 rs324981T 0.431 0.468 1.16 (1.01–1.34)  .040  .325 0.462 1.13 (0.99–1.30) .079 .325 0.465 1.15 (1.02–1.29)  .027  .325P FDR , False Discovery Rate method-corrected  P   value.Significant  P   values ( P   .05) are reported in bold. a  CD patients  UC patients.  C L  I   NI    C AL  –AL  I   ME  NT  AR Y T  R A C T   812 D’AMATO ET AL GASTROENTEROLOGY Vol. 133, No. 3
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks