This page has been archived and is no longer updated
Genetic variants regulating ORMDL3 expression contribute to the risk of childhood asthma.
Author: M. F. Moffatt
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"LETTERS Genetic variants regulating ORMDL3 expression contribute to the risk of childhood asthma Miriam F. Moffatt 1 *, Michael Kabesch 2 *, Liming Liang 3 *, Anna L. Dixon 4 , David Strachan 5 , Simon Heath 6 , Martin Depner 2 , Andrea von Berg 7 , Albrecht Bufe 8 , Ernst Rietschel 9 , Andrea Heinzmann 10 , Burkard Simma 11 , Thomas Frischer 12 , Saffron A. G. Willis-Owen 1 , Kenny C. C. Wong 1 , Thomas Illig 13 , Christian Vogelberg 14 , Stephan K. Weiland 15 , Erika von Mutius 2 , Gonc�alo R. Abecasis 3 , Martin Farrall 4 , Ivo G. Gut 6 , G. Mark Lathrop 6 & William O. C. Cookson 1 Asthma is caused by a combination of poorly understood genetic and environmental factors 1,2 . We have systematically mapped the effectsofsinglenucleotidepolymorphisms(SNPs)onthepresence of childhood onset asthma by genome-wide association. We char- acterized more than 317,000 SNPs in DNA from 994 patients with childhood onset asthma and 1,243 non-asthmatics, using family andcase-referentpanels.Hereweshowmultiplemarkersonchro- mosome 17q21 to be strongly and reproducibly associated with childhood onset asthma in family and case-referent panels with a combinedPvalueofP,10 212 .Inindependentreplicationstudies the17q21locusshowedstrongassociationwithdiagnosisofchild- hood asthma in 2,320 subjects from a cohort of German children (P50.0003) and in 3,301 subjects from the British 1958 Birth Cohort (P50.0005). We systematically evaluated the relation- ships between markers of the 17q21 locus and transcript levels ofgenesinEpstein?Barrvirus(EBV)-transformedlymphoblastoid cell lines from children in the asthma family panel used in our association study. The SNPs associated with childhood asthma were consistently and strongly associated (P,10 222 )incis with transcript levels of ORMDL3, a member of a gene family that encodes transmembrane proteins anchored in the endoplasmic reticulum 3 . The results indicate that genetic variants regulating ORMDL3 expression are determinants of susceptibility to child- hood asthma. Our study design aimed to provide internal replication of poten- tially positive results with family and case-referent panels of subjects with childhood asthma (Fig. 1). This structure also allowed compar- ison for consistency between case-control and family-based tests of association. To systematically investigate the effects of polymorph- isms on the transcription of positional candidate genes, we further measured global gene expression in B-cell-derived EBV-transformed lymphoblastoid cell lines (EBVL) in probands and siblings of the family panel. The study subjects included a panel of 207 predominantly (99%) nuclear families (MRC-A). These were recruited through a proband with severe (Step 3) childhood onset asthma and contained 295 sib pairs, 11 half-sib pairs and 3 singletons (counting all possible sibs). Four-hundred and thirty-seven non-asthmatic Caucasian UK controls (UK-C) were studied by the same protocols. We also geno- typed 728 children of German origin recruited in the Multicentre Asthma Genetics in Childhood Study (MAGICS) study with physi- cian-diagnosed asthma for comparison with 694 reference children recruited in the cross sectional International Study of Asthma and Allergies in Childhood (ISAAC) study 4 . Wegenotypedallchildrenintheprimaryassociationstudywiththe IlluminaSentrixHumanHap300BeadChip.Afterqualitycontroland elimination of markers with low minor-allele frequencies (,0.5%), extreme Hardy?Weinberg equilibrium statistics (x 2 .25) or low genotyping call rates (,95%), we retained 307,328 SNPs and 684 million genotypes (99.4% call rate) for analysis. Amongst the 2,236,212 common (minimum allele frequency.0.05) SNPs in the HapMappanel,79%weretaggedwithcorrelationR 2 .0.8and90.9% with R 2 .0.5 with our successfully typed markers. We also typed the parents and children in the MRC-A panel with the Illumina Sentrix Human-1 Genotyping BeadChip (concentrated on genes and surrounding sequences) (Fig. 1), producing an *These authors contributed equally to this work. 1 National Heart and Lung Institute, Imperial College, London SW3 6LY, UK. 2 University Children?s Hospital, Ludwig Maximilians University, D80337 Munich, Germany. 3 Center for StatisticalGenetics,DepartmentofBiostatistics,SPHII,AnnArbor,Michigan48109-2029,USA. 4 WellcomeTrustCentreforHumanGenetics,UniversityofOxford,OxfordOX37BN, UK. 5 DivisionofCommunityHealthScience,StGeorge?s,UniversityofLondon,LondonSW170RE,UK. 6 CentreNationaldeGe�notypage,InstitutGe�nomique,Commissariata`l?E � nergie Atomique, 91057 Evry, France. 7 Research Institute for the Prevention of Allergic Diseases, Children?s Department, Marien-Hospital, D46483 Wesel, Germany. 8 Department of Experimental Pneumology, Ruhr-University, D44789 Bochum, Germany. 9 University Children?s Hospital, University of Cologne, D50924 Cologne, Germany. 10 University Children?s Hospital, Albert Ludwigs University, D79106 Freiburg, Germany. 11 Children?s Department, Feldkirch Hospital, A6800 Feldkirch, Austria. 12 University Children?s Hospital Vienna, A1090 Vienna, Austria. 13 Institute of Epidemiology, GSF-Research Centre for Environment and Health, D85764 Neuherberg, Germany. 14 University Children?s Hospital, Technical University Dresden, D01307 Dresden, Germany. 15 Institute of Epidemiology, Ulm University, D89081 Germany. UK-C MAGICS Total Subjects Parents Children Non-asthmatic 112 437 694 Asthmatic 266 0 728 994 734873504latoT Measurements 300K Illumina 100K Illumina Affy Hu133A 2.0 EBVL Analyses Transmit Regression model Expression QTL MRC-A 100K 300K Family Case-referent Children Children 1,422 2,642 1,243 Figure 1 | Study design. The subjects were recruited from family (MRC-A) and case-control panels (MAGICS and UK-C). All children were genotyped with the Illumina Sentrix HumanHap300 BeadChip. The children and parents in the MRC-A panel were in addition genotyped with the Illumina Sentrix Human-1 Genotyping BeadChip. Gene expression in lymphoblastoid cell lines (EBVL) was measured in the affected and unaffected children of the MRC-A panel. Replication of positive results was sought in two independent panels of subjects from the ISAAC Phase II and 1958 British Birth Cohort studies. Vol 448|26 July 2007|doi:10.1038/nature06014 470 Nature �2007 Publishing Group additional 91,293 SNPs with 36.0% average heterozygosity and 89,815,992 genotypes (99.0% call rate). We found only 0.412 men- delianerrorsperSNP:thesewereexcludedfromsubsequentanalyses. We tested for association of the 300K panel to childhood onset asthmainthecombinedprimarydatasetof994asthmaticsand1,243 non-asthmatics (Fig. 2). We calculated the 1% false-discovery rate (FDR) threshold 5 to be P#6.8310 27 and the 5% FDR to be P#5.0310 26 . We tested for population stratification in the com- bined data set, finding a genomic control parameter of 1.07. This minor degree of background stratification reflected small differences in allele frequencies within and between the two nationally defined European populations that contributed to the study. We further tested the effects of stratification on our top results by analysingthe34SNPsthatsurpassedthe5%FDRthresholdwith102 randomly selected SNPs as covariates in a backward stepwise logistic regression procedure 6 . The results of the stepwise logistic regression analysisshowedonlymodestreductionofthesupportforassociation for 31 of the SNPs. Of 34 SNPs, 16 continued to surpass the 5% FDR and 12 of 20 SNPs remained above the 1% FDR threshold (Sup- plementary Table 1). Strikingly, 7 of the 12 markers still below the 1% FDR threshold mapped to a 112 kb interval on chromosome 17q21. Several other markers in this interval also showed strong evidence of association (Supplementary Table 2, Fig. 2 and Fig. 3a). The SNP with the stron- gestevidenceofassociationwithintheintervalwasrs7216389(uncor- rected P59310 211 ). With the exception of the 17q21 locus, none of the markers below the 5%FDR, after controlling for stratification, were within 1Mb of each other (Supplementary Table 1). The patterns of association for the chromosome 17q21 markers were similar and significant in both the UK family panel and the German case-referent panel (Table 1 and Supplementary Table 2). There was no evidence of heterogeneity of the association, or of significant allele frequency differences in cases or controls from the UK and Germany. We selected 27 markers from the NCBI database of genetic vari- ation(dbSNP)thatwerewithinoradjacenttothestronglyassociated interval for genotyping. These exhibited similar patterns and strengthsofassociationasthegenome-wideassociation(GWA)mar- kers (Supplementary Table 2). SNPs from the region that we had typed in whole families in the MRC-Apanelshowedsignificantassociationsinafamily-basedasso- ciation test (Supplementary Table 3). The most strongly associated markerwasrs8067378(248informativetransmissions,P53310 26 ; odds ratio,1.84; 95% confidence interval, 1.43?2.42). The pattern of alleles transmitted in excess to affected offspring was consistent with the case-referent association results, further indicating that the asso- ciation was robust to population stratification. The trait-associated markers (with P,10 26 in the combined data set) from chromosome 17q21 fall within a 206.5 kb core containing three consecutive haplotype blocks (linkage disequilibrium co- efficient (D?).0.94 for adjacent markers, D?,0.82 between blocks), as well as one ?isolated? marker in moderate (D?50.7) link- age disequilibrium with an adjacent marker (Supplementary Fig. 1). Low linkage disequilibrium between markers that showed strong association with the disease trait indicated that multiple variants may independently confer disease susceptibility. We evaluated this hypothesis statistically in a forward stepwise regression potentially incorporating all the genotyped markers between 34.5 and 36.0Mb on chromosome 17. This identified three SNPs (rs7216389, rs11650680 and rs3859192), which jointly showed strong association to childhood asthma (P,10 212 ) and contributed statistically inde- pendent significant effects (Supplementary Table 4). This result is consistent with the possibility that more than one functional SNP underlies the locus or (less likely given the SNP density across this region)thepresenceofasinglefunctionalSNPinincompletelinkage disequilibriumwiththetypedmarkers.Furtherfine-mappingstudies will be required to resolve these alternatives. We further examined the impact of the chromosome 17q21 locus on childhood asthma with the aid of 200 asthmatic cases and 2,120 non-asthmatic controls from within the cross-sectional study popu- lation of ISAAC Phase II, recruited in Dresden and Munich. The corresponding DNAs were genotyped for a series of nine markers from across the locus that had shown evidence of association in the GWA (Table 1). These showed the same trends as observed in the GWAsamples(oddsratiosrangingfrom1.52to1.11),andmostwere significantly associated with disease with P,0.001 (4 out of 9 mar- kers) or P,0.01 (5 out of 9 markers). One chromosome 17q21 SNP associated with childhood asthma in the GWA analysis (rs3894194, Table 1) had been previously examined in 3,301 subjects from the UK 1958 birth cohort. The genotype data for this marker were deposited for public use in the database maintained by the investigators responsible for the cohort study (http://www.b58cgene.sgul.ac.uk/). Although data were not available for the other disease-associated chromosome 17q21 mar- kers, we were able to confirm association to childhood asthma with rs3894194, which is in linkage disequilibrium with the other markers at the locus (linkage disequilibrium block 3 in Supple- mentary Fig. 1). Restricting analyses to cohort members of Caucasian ethnicity, we found that the 398 cases recalling ?asthma ever?atage42,showedasignificantassociation(oddsratio,1.21,95% confidence interval, 1.04?1.40, P50.012). Ninety-three individuals were reported to have ?asthma attacks? in the first seven years of life (that is during 1958 to 1965), and these were strongly associated to rs3894194 (odds ratio51.68, 95% confidence interval, 1.25?2.26, P50.0005). The disease-associated chromosome 17q21 markers had consist- ent odds ratios and directions of association effects in the GWA panels and the two replication cohorts (P50.19 for heterogeneity) 1514131211109 Chromosome 87654321 9 11 10 8 7 6 5 4 3 2 1 0 1% FDR 5% FDR ?log 10 ( P -value) 17 X2119 Figure 2 | Genome-wide association of 317,447 SNPs and asthma in 994 asthmatic children and 1,243 non-asthmatic children. Position in the genome, divided by chromosome, is depicted along the xaxis. Strength of association is shown on the yaxis. The result for each individual marker is depicted as a black circle. The genome-wide thresholds for 1% and 5% false discoveryrates(FDR)areshownashorizontalredlines.Numerousmarkers on chromosome 17q21 show association to asthma above the 1% FDR threshold in the region of maximum association. NATURE|Vol 448|26 July 2007 LETTERS 471 Nature �2007 Publishing Group (Table 1), suggesting that we have identified a robust risk factor for childhood asthma. Variation in gene transcription is an important mechanism in mediating susceptibility to asthma and other diseases, and the transcript abundances of genes may be directly modified by poly- morphisms in regulatory elements 7,8 . We therefore measured global gene expression in EBVL from children in the MRC-A panel (Fig. 1). These same subjects had been typed with the Illumina Sentrix HumanHap300 and Sentrix Human-1 Genotyping BeadChips (Fig. 1). EBVLs represent the B-celllineage, andare consequently ofdirect relevance to asthma. Cells were harvested at log-phase in the first growth after EBV transformation. Global transcript abundance was measuredwithAffymetrixHG-U133Plus2.0chips.Weusedquantile normalization after Robust Multi-Array Average (RMA) to enforce normality and reduce outlier leverage. A complete description of the global results will be presented in another paper. Expression data were available for 14 of 19 annotated genes in the region from 35.0 to 35.5Mb on chromosome 17 (that is, within or near the 206 kb region of association on 17q21.1) (Fig. 3d). We found that transcripts in one gene, ORMDL3, were strongly (P,10 222 for rs7216389) and consistently positively associated to exactly the same SNPs from the Illumina Sentrix HumanHap300 BeadChipaschildhoodasthma(Fig.3c).Thedisease-associatedmar- kers accounted for 29.5% of the variance of expression. No other markers were significantly correlated to ORMDL3 expression after adjustment for genome-wide multiple testing. After accounting for the effects of the disease-associated markers, the residual heritability of the ORMDL3 expression was not significant (P50.29 compared to P50.0009 before adjustment). Thesefindingssuggested thatthe 17q21disease-associated locusis the principal genetic determinant of ORMDL3 expression. None of the other transcripts from the region or elsewhere in the genome showed a significant relationship to the disease-associated markers in our data set. Despite the strength of these effects, the presence of several SNPs independently associated to asthma nevertheless makes itpossiblethatORMDL3maynotbetheexclusivedeterminantofthe disease susceptibility at this locus. Inthesubsetofindividualsforwhomexpressiondataareavailable, the T nucleotide allele at rs7216389 (the marker most strongly assoc- iated with disease in the combined GWA analysis) has a frequency of 62% amongst asthmatics compared to 52% in non-asthmatics (P50.005 in this sample). The additive effect of this allele corre- sponds to a change of 0.78 standard deviation units in ORMDL3 expression (P,10 222 ). We saw the expected increase of ,0.064 standard deviation units in ORMDL3 expression among asthmatics, but this was not significant given the sample size. Patterns of ORMDL3 expression by genotype in asthmatics and non-asthmatics are shown in Supplementary Fig. 2. ORMDL3isthethirdmemberofanovelclassofgenesofunknown function that encode transmembrane proteins anchored in the endoplasmic reticulum (ER) 3 . We examined multiple tissue com- plementary (MTC) DNA panels by PCR with reverse transcription (RT?PCR), and found ORMDL3 to be expressed in many tissues, particularly liver and peripheral blood lymphocytes (Fig. 3g). The SNPs showing the strongest association to asthma and ORMDL3 transcript abundances are contained within an island of linkage dis- equilibrium between 35.2 and 35.4Mb on chromosome 17q21 (Fig. 3b and 3c). The one-lod support unit for SNPs showing maxi- mum association to ORMDL3 levels lies within the first intron of the neighbouring GSDML gene. This non-coding sequence shows sig- nificant homology between species (Fig. 3e and 3f), and contains an element with high homology to the pro-inflammatory transcription factor C/EBPb (transcription factor score, 86.8; http://www.cbrc.jp/ research/db/TFSEARCH.html). Genomic regions other than the 17q21 locus did not contain multiple markers with significant evidence of association at the 1% Location (Mb) 0 4020060 2 4 6 8 10 ORMDL3 and asthma 0 8 6 4 2 0 10 5 10 15 20 25 b a c 35.835.635.435.235.034.834.6 36.0 35.835.635.435.235.034.834.6 36.0 ?log 10 ( P ) [ ORMDL 3] ?log 10 ( P ) asthma ?log 10 ( P ) asthma d g 500 bp Lymph node Tonsil Thymus Activated CD4 + cells Activated CD19 + cells Activated CD8 + cells Activated mononuclear cells Resting CD19 + cells Resting CD8 + cells Resting CD14 + cells Mononuclear cells Colon Lung Kidney Placenta Liver Skeletal muscle BrainHeart Peripheral blood leukocytes 1,000 1,517 f e GSDML Conservation Rhesus Chimp Mouse Rat Dog Rabbit Cow Armadillo Elephant Tenrec Lod=19: Chr17: No template control Figure 3 | Association to asthma and transcript abundances of ORMDL3 on chromosome 17q21. a,Mappingofassociationtoasthma on chromosome 17. b, Detail of association to SNPs on chromosome 17q21. c, Association to ORMDL3 transcript abundance with the same markers. A GOLD plot 22 of linkage disequilibrium betweenmarkersisalsoshown,with red indicating high linkage disequilibrium and blue denoting low. The central island of linkage disequilibrium, which contains maximum association to ORMDL3 and asthma, is contained within the grey rectangle. d, Genes contained within the associated interval. e,Homologyplotfromtheregionof maximum association. f, Sequence homologyfromintronIofGSDML. g, RT?PCR (34 cycles) of ORMDL3 cDNA from representative tissues (Clontech). LETTERS NATURE|Vol 448|26 July 2007 472 Nature �2007 Publishing Group FDRthresholdineitheroftheindividualcollectionsorthecombined samples. Several other loci (Supplementary Table 1) passed a less stringent 5% FDR threshold after adjustment for population strati- fication, indicating that other susceptibility loci for childhood asthma could be present, perhaps having smaller effects than those found on chromosome 17q21. These results will be explored in large scale replication studies in multiple centres and population samples (the GABRIEL project: http://www.gabriel-fp6.org/project). METHODS SUMMARY Children and their parents from the UK panels were administered a standard questionnaire 9 byanurseordoctor.Asthmawasdefinedasapositiveresponseto the question ??Has your doctor ever told you that you have asthma??? Probands had Step 3 asthma or worse according to British Thoracic Society guidelines 10 . Siblings were included regardless of asthma status 11 . Asthma cases from the MAGICS were diagnosed by a paediatric pulmonologist or allergologist on the basis of clinical examination, history and objective tests of lung function. The transformation of peripheral blood lymphocytes in all children in the MRC-A panel was carried out by the ECACC (http://www.ecacc.org.uk). Microarray hybridization of EBVL RNA to the U133 Plus 2.0 GeneChips (Affymetrix) was under standard conditions. Whole-genome genotyping was carried out using Illumina Sentrix Human-1 Genotyping BeadChip 12 and Sentrix HumanHap300 Genotyping BeadChips 13 (Illumina, San Diego). Genotyping of additional mar- kers was performed on an ABI7900HT Sequence Detection System using TaqMan probes (Applied Biosystems, Foster City, California). German replica- tion samples were genotyped using matrix-assisted laser desorption/ionization? time of flight (http://www.sequenom.com) 14 . Logistic regression models with robust sandwich estimation of the variance 15 implemented in the Stata logit function were used to detect association to asthma in the combined panels. The TRANSMIT program 16 was used to analyse nuclear family data (including parental genotypes). The false-discovery rate (FDR) method 5 was used to assess the overall statistical significance of the genome-wide association results, and FDR thresholds were calculated by applying the QVALUE (http://faculty. washington.edu/,jstorey/qvalue/) package 17 . Gene expression data were nor- malized with the RMA package 18,19 to remove any technical or spurious back- groundvariation.Aninversenormalizationtransformationstepwasalsoapplied to each trait to avoid any outliers. Association analysis was applied with Merlin (FASTASSOC option) 20 , after probabilistically inferring missing genotypes 21 . Full Methods and any associated references are available in the online version of the paper at www.nature.com/nature. Received 24 April; accepted 14 June 2007. Published online 4 July 2007. 1. Cookson, W. The immunogenetics of asthma and eczema: a new focus on the epithelium. Nature Rev. Immunol. 4, 978?988 (2004). 2. Ober, C. & Hoffjan, S. Asthma genetics 2006: the long and winding road to gene discovery. Genes Immun. 7, 95?100 (2006). 3. Hjelmqvist, L. et al. ORMDL proteins are a conserved new family of endoplasmic reticulum membrane proteins. Genome Biol. 3, RESEARCH0027 (2002). 4. Weiland, S. K. et al. Phase II of the International Study of Asthma and Allergies in Childhood (ISAAC II):rationale and methods. Eur. Respir. J. 24, 406?412 (2004). 5. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Statist. Soc. B. 57, 289?300 (1995). 6. Setakis, E., Stirnadel, H. & Balding, D. J. Logistic regression protects against population structure in genetic association studies. Genome Res. 16, 290?296 (2006). 7. Schadt,E.E.etal.Geneticsofgeneexpressionsurveyedinmaize,mouseandman. Nature 422, 297?302 (2003). 8. Morley, M. et al. Genetic analysis of genome-wide variation in human gene expression. Nature 430, 743?747 (2004). 9. Standards. for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, November 1986. Am. Rev. Respir. Dis. 136, 225?244 (1987). 10. British. guideline on the management of asthma. Thorax 58 (Suppl 1), i1?i94 (2003). 11. Abecasis, G., Cardon., L. & Cookson, W. Selection strategies for disequilibrium mapping of quantitative traits in nuclear families. Am. J. Hum. Genet. 65, A245 (1999). 12. Gunderson, K.L.,Steemers, F.J.,Lee,G.,Mendoza,L. G.&Chee,M.S.A genome- wide scalable SNP genotyping assay using microarray technology. Nature Genet. 37, 549?554 (2005). 13. Steemers, F. J. et al. Whole-genome genotyping with the single-base extension assay. Nature Methods 3, 31?33 (2006). 14. Buetow, K. H. et al. High-throughput development and characterization of a genomewide collection of gene-based single nucleotide polymorphism markers by chip-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Proc. Natl Acad. Sci. USA 98, 581?584 (2001). 15. Williams, R. L. A note on robust variance estimation for cluster-correlated data. Biometrics 56, 645?646 (2000). 16. Clayton, D. A generalization of the transmission/disequilibrium test for uncertain-haplotype transmission. Am. J. Hum. Genet. 65, 1170?1177 (1999). 17. Storey, J. D. & Tibshirani, R. Statistical significance for genomewide studies. Proc. Natl Acad. Sci. USA 100, 9440?9445 (2003). 18. Irizarry, R. A. et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4, 249?264 (2003). 19. Bolstad, B. M., Irizarry, R. A., Astrand, M. & Speed, T. P. A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics 19, 185?193 (2003). 20. Abecasis, G. R., Cherny, S. S., Cookson, W. O. & Cardon, L. R. Merlin?rapid analysis of dense genetic maps using sparse gene flow trees. Nature Genet. 30, 97?101 (2002). 21. Burdick, J. T., Chen, W. M., Abecasis, G. R. & Cheung, V. G. In silico method for inferring genotypes in pedigrees. Nature Genet. 38, 1002?1004 (2006). 22. Abecasis, G. R. & Cookson, W. O. GOLD?graphical overview of linkage disequilibrium. Bioinformatics 16, 182?183 (2000). Supplementary Information is linked to the online version of the paper at www.nature.com/nature. Acknowledgements The study was funded by the Wellcome Trust, the Medical Research Council, the French Ministry of Higher Education and Research, the GermanMinistryofeducationandresearch(BMBF),thenationalgenomeresearch network (NGFN), the National Institutes of Health (NHGRI and NHLBI; G.R.A.), and the European Commission as part of GABRIEL (a multidisciplinary study to identify the genetic and environmental causes of asthma in the European Community). We acknowledge use of genotype data from the British 1958 Birth CohortDNAcollection,fundedbytheMedicalResearchCouncilandtheWellcome Trust. We thank J. Todd for genotyping rs3894194 in the 1958 British Birth cohort. Author Information Microarray and chromosome 17 genotyping data have been deposited in the GEO database, with accession number GSE8052. Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Correspondence and requests for materials should be addressed to W.O.C.C. (w.cookson@imperial.ac.uk). Table 1 | Association of asthma and ORMDL3 to chromosome 17q21 in combined and replication panels Primary GWA study eQTL ISAAC II replication cohort Location (Mb) 2log 10 (P) asthma 2log 10 (P)[ORMDL3] Ref.allele Alt. allele Odds ratio Lower 95% CI Upper 95%CI 2log 10 (P) asthma Marker MRC-A MAGICS Combined MRC-A rs9303277 35.230 4.45.08.821.9 CT1.41 1.14 1.76 2.8 rs11557467 35.282 4.65.29.122.5 GT1.45 1.16 1.82 3.1 rs8067378 35.305 4.35.39.02.7 rs2290400 35.320 4.06.49.822.4 AG1.47 1.18 1.82 3.2 rs7216389 35.323 4.36.410.02.4 TC1.45 1.17 1.81 3.1 rs4795405 35.342 4.15.38.71.6 CT1.52 1.20 1.89 3.6 rs8079416 35.346 2.46.58.210.9 1.30 1.05 1.61 1.9 rs4795408 35.361 2.16.17.51.3 AG1.25 1.01 1.54 1.4 rs3894194 35.376 1.96.67.711.0 TC1.22 0.98 1.51 1.2 rs3859192 35.382 2.06.07.33.5 1.11 0.89 1.37 0.4 SNPs are from the Illumina 300K panel. Results were calculated with logistic regression models. Association to ORMDL3 transcript abundance is shown for comparison and is based on the family panel only. Replication results are shown for the ISAAC II population. CI, confidence interval; Ref., reference; Alt., alternative; eQTL, expression quantitative trait locus. NATURE|Vol 448|26 July 2007 LETTERS 473 Nature �2007 Publishing Group METHODS Subjects.ChildrenandtheirparentsfromtheUKpanelswererecruitedaspartof the MRC UK National family collection and were administered a standard questionnaire (based on the American Thoracic Society and International Study of Asthma and Allergies in Childhood (ISAAC) questionnaires 9 )bya nurse practitioner or a doctor. Asthma was defined as a positive response to the question ??Has your doctor ever told you that you have asthma??? Probands had Step 3 asthma or worse according to the British Thoracic Society guidelines (high-dose inhaled steroids, or low-dose inhaled steroids and a long-acting b-agonist) 10 . Siblings were included regardless of asthma status 11 . Asthma cases from the Multicentre Asthma Genetics in Childhood Study (MAGICS) were diagnosed by a paediatric pulmonologist or allergologist on the basis of clinical examination,casehistoryandobjectivetestsoflungfunction.Asthmatics(mean age 10.95yr) were recruited from 7 centres located in Germany and Austria (Wesel, Bochum Cologne, Freiberg, Munich, Feldkirch and Vienna), and as a reference, 800 German children (mean age 9.62yr) from Dresden(n5400)and Munich (n5400) were randomly drawn from all German children with DNA availableinthecrosssectionalISAACPhaseIIstudy 4 . Furthercasesandcontrols forreplicationweredrawnfromtheGermanISAACpopulationinwhichasthma wasdiagnosedusingstandardizedquestionnairesandvalidatedbylungfunction and bronchial hyper-responsiveness testing 4 . All study methods were approved by the appropriate ethics committees. EBV. The transformation of the Peripheral Blood Lymphocytes (PBL) in all children in the MRC-A panel was carried out by the ECACC (http://www.ecacc. org.uk). Previously transformed cryo-preserved EBV cell lines were grown as 500ml roller cultures. Once log phase had been obtained, cells were pelleted, media discarded and a mixture of RLT buffer and b-mercaptoethanol added. Pellets were vortexed to ensure thorough re-suspension, after which they were frozen at270uC and stored at280uC. RNA was extracted in batches after cell homogenization using RNeasy Maxi Kits (Quiagen), and quality and quantity assessed. Microarrayhybridization.RNA(10mg)wasusedtosynthesizedouble-stranded cDNA using the One-cycle cDNA synthesiskit (Affymetrix). Usingthe cDNA as a template, in vitro transcription of cRNA was carried out using the IVT kit (Affymetrix), following the manufacturer?s protocol. A hybridization cocktail was made according to protocol, using 15mg of labelled, fragmented cRNA, and hybridized to U133 Plus 2.0 GeneChips (Affymetrix) for 16h at 45uCina rotatingoven.GeneChipswerewashedand stainedaccordingtomanufacturer?s protocols and scanned on a high-resolution scanner (Affymetrix). Genotyping. Whole-genome genotyping (WGGT) was carried out using Illumina Sentrix Human-1 Genotyping BeadChip 12 and Sentrix HumanHap300 Genotyping BeadChips 13 (Illumina,SanDiego), according tothemanufacturer?s instructions in a BeadLab with full automation at the Centre National de Genotypage.AllDNAsamplesweresubjectedtorigorousqualitycontroltocheck for fragmentation andamplification. Twenty microlitres of DNA at a concentra- tion of 50ngml 21 was used for each array. DNA samples were tracked using a Laboratory Information Management System. The HumanHap300 Genotyping BeadChip was used with an Illumina LIMS, whereas the Sentrix Human-1 Genotyping BeadChip was tracked through the Illumina process by hand. Groups of 24 samples were batched. Five percent of the samples were selected from different batches, re-genotyped and the results compared to the original data. No sample discrepancies were detected. Raw data were analysed using GTS Image and extracted for statistical analysis. Genotyping of additional markers on chromosome 17q21 was performed on an ABI7900HT Sequence Detection System using TaqMan probes (Applied Biosystems, Foster City, California). German replication samples were genotyped using matrix-assisted laser desorp- tion/ionization time-of flight (MALDI-TOF) mass spectrometry (http://www. sequenom.com) 14 . Primer extension products were analysed by a MassARRAY massspectrometer(http://www.bdal.de)andresultingmassspectrawereanalysed using the SpectroTYPER RT 2.0 software. Association testing. Tests of Hardy?Weinberg equilibrium were performed in cases and controls using the genhw procedure (http://www.biostat-resources. com/stata/) and Stata version 9.2, and SNPs showing Hardy?Weinberg disequi- librium incontrols(x 2 .25) wereexcluded.As the datacompriseda mixtureof unrelatedandrelatedcasesandcontrols,weusedlogisticregressionmodelswith robust sandwich estimation of the variance 15 as implemented in the Stata logit functiontomodelclusteringofsiblings?genotypes.SimulationsusingtheMRC- A family structures (data available on request) confirmed that this method appropriately controls the Type I error. Heterogeneity of association between the two main strata (UK and Germany) was assessed by a weighted linear com- bination test using the results of an additive-effects-only regression analysis within each stratum. X-linked markers were analysed by fitting an additive- effects-only logit model that equates the risks of male hemizygotes with female homozygotes. The TRANSMIT program 16 was used to analyse nuclear family data (including parental genotypes), using the sandwich variance estimation option to robustly incorporate information from multiple affected siblings; confidence intervals for odds ratio estimates were computed as described. The false-discovery rate (FDR) method 5 was used to assess the overall statistical significance of the genome-wide association results, taking into account the multiple hypothesis testing implications inherent in the analysis of more than 300K SNPs. The FDR thresholds were calculated by applying the QVALUE (http://faculty.washington.edu/,jstorey/qvalue/) software package 17 . Association to transcript abundances. Data from the gene expression experi- ment were normalized together using the RMA package 18,19 to remove any technical or spurious background variation. An inverse normalization trans- formation step was also applied to each trait to avoid any outliers. Association analysis was applied with Merlin (FASTASSOC option) 20 . We estimated an additive effect for each SNP and tested its significance using a score test that adjusts for familiality and takes into account uncertainty in the inference of missing genotypes. In the absence of a positive genomic control test, we did not adjust for stratification. We probabilistically inferred missing genotypes 21 and adjusted for familiality, but not for linkage signal. doi:10.1038/nature06014 Nature �2007 Publishing Group "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
No comments yet.
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment