and Immunity (2001) 2, 455–460  2001 Nature Publishing Group All rights reserved 1466-4879/01 $15.00 www.nature.com/gene Identification of novel polymorphisms in the ␤7 : family-based association studies in inflammatory bowel disease

DA van Heel1,2, AH Carey3 and DP Jewell2 1Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN, UK; 2Gastroenterology Unit, Gibson Laboratories, Radcliffe Infirmary, Oxford, OX2 6HE, UK; 3Oxagen Ltd, 91 Milton Park, Abingdon, OX14 4RY, UK

Linkage studies from five groups worldwide have confirmed the presence of an inflammatory bowel disease susceptibility locus on 12q. Beta 7 integrin is a strong candidate gene within this region, and is involved in lymphocyte homing to the gut and retention of intra-epithelial lymphocytes. Monoclonal antibodies to ␤7 integrin ameliorate colitis in animal models. We obtained genomic sequence for ␤7 integrin, and screened all 16 exons and 1.7 kb of 5Ј promoter region for polymorphisms in 24 individuals. Fourteen single nucleotide polymorphisms were identified in total and, of these, two common (frequency у10%) intronic and two amino acid changing polymorphisms were assessed for potential disease associations. Data were available from 102 multiply affected inflammatory bowel disease families (affected sibling pairs) and 362 simplex (one affected proband) families containing 254 ulcerative colitis, 13 indeterminate colitis and 300 Crohn’s disease trios (parents + affected child). No significant associations with any disease phenotype were found with the transmission disequilibrium test. Beta 7 integrin is unlikely to be involved in the genetic susceptibility to inflammatory bowel disease, and therefore future studies on should focus on other positional candidate genes. Genes and Immunity (2001) 2, 455–460.

Keywords: inflammatory bowel disease; Crohn’s disease; ulcerative colitis; ITGB7; integrin; genetic susceptibility

Introduction all lymphocytes to the gut and associated lymphoid tissues. Gut-homing effector cells (expressing high levels Inflammatory bowel disease (IBD), comprising Crohn’s of ␣4␤7 integrin) attach directly to mucosal cell disease (CD) and ulcerative colitis (UC), is likely to result adhesion molecule 1 on , whilst from inherited mutations in disease susceptibility genes naı¨ve cells attach initially via L-selectin.11 ␣E␤7 integrin is interacting with environmental factors. Epidemiological expressed on gut intraepithelial lymphocytes, which are evidence from twin studies, familial clustering and intra- retained in the gut by binding to the E-cadherin familial disease concordance suggests that IBD has a present on epithelial cells.12,13 1 strong genetic component. Estimates of the sibling rela- Other evidence that ITGB7 plays a key role in the ␭ tive risk s vary from 10 to 35, depending on IBD pheno- development of bowel inflammation comes from animal 2–5 type used. In the Oxford IBD genome-wide scan the and pharmacological studies. Monoclonal antibodies to greatest evidence for linkage was found on chromosome ␣4␤7 ameliorate colitis in both the cotton top tamarin and 6 12q. Subsequently four other groups worldwide have CD45RBhigh CD4+ T cell reconstituted severe combined 7–10 confirmed this IBD2 linkage region (OMIM: 601458). immunodeficiency syndrome (SCID) mouse models of The comprise a large family of transmem- IBD.14,15 Knockout mice deficient for the ␤7 integrin sub- ␣ brane adhesion glycoproteins, each composed of an unit exhibit severely impaired development of the gut ␤ ␤ and subunit. The 7 integrin gene (ITGB7) maps to the associated lymphoid tissue, and attenuated immune IBD2 linkage region on chromosome 12. It is an excellent responses against helminth infection.16,17 candidate gene for IBD susceptibility, both because of its Mutations in the ␤2, ␤3 and ␤4 integrins cause leuko- position and the fact that it is involved in leucocyte traf- cyte adhesion deficiency, thromboasthenia and epiderm- ␣ ␤ ficking and recruitment to the gut. The 4 7 integrin het- olysis bullosa respectively.18–20 We therefore hypoth- erodimer is an adhesion molecule involved in homing of esised that similar functional mutations in the ␤7 integrin gene may play a role in susceptibility to IBD. Potential mechanisms for pathological mutations include increased Correspondence: Dr DA van Heel, Wellcome Trust Centre for Human leukocyte ␤7 integrin expression, changes in ligand affin- Genetics, University of Oxford, Roosevelt Drive, Oxford, OX3 7BN, UK. ity and alterations in the complex integrin signalling and Ȱ E-mail: david.vanheel well.ox.ac.uk lymphocyte activation pathway.21 No polymorphisms in Part of this work was funded by Oxagen Ltd, and a Medical ITGB7 have been reported, and therefore we screened all Research Council LINK grant between the University of Oxford and Ј Oxagen Ltd. exons and the 5 promoter region in order to identify Received 20 August 2001; revised 12 September 2001; accepted 13 markers to test for association. The transmission disequi- September 2001 librium test (TDT) was chosen to overcome the potential ␤7 integrin polymorphisms in IBD DA van Heel et al 456 Table 1 Primer pairs for ITGB7 mutation detection

Region Primers (5Ј to 3Ј)

promoter (−1688 to −790) CCCCTCCTAGCAGCTACACC GGGATTACAGGCACACTCG promoter (−1011 to −282) CCAGAGCCTATGTAGTCAAGCAC CCAAGGTCCCACAGCTAGTAAG promoter (−434 to −59) CTGGGAAACAGAGCAAGACC CTAGAAGGTGGTGCAGATAGGG promoter (−169) exon 1 ATGTTACGTTCAACACATGACAGG CAGATGACAGCACTCATATACATCC exon 2 AAAGCCGCCAAGTATGTCAG TTCGAAACCAGCCTAGTCAAC exon 3 CACATGTCCCCCTTTATATCC TGTGCACAAACTCAGTCACAC exon 4 ATCAGGGACTCAAGGAGTGGAC TCAGGTCCATAAGGTAGTACAGGTC exon 5 GCGGCCTGGTGAGTTAGG ATGTTGGCAGAGGCTAGGG exon 6 CCACCCCACTTCCCATAGAG GTTCCCAAACAGACCTCCAG exon 7 CCCAGGTGCATAGCCTACC AAGAGGTGTGGCTGAAATGG exon 8 ATGTTGTGGGGGAAACCTG ATATGGGGGACGGAGAATG exon 9 GGAGCCCATTTTCCTAGAGC CAGTTAATGACAGCCACATGC exon 10 GGACATAAGGTGGGGTTGG TTCCTGCCTGCTTAATTTCC exon 11 TAACATGACCCCATCCCTTC GTTGTTGGGAGCCAGGTG exon 12 CCTGGGAGAGACAAGTCAGG AGCCTAAGTGCCTTGGGAAG exon 13 GGGTGTGGTTGGCATACTTC GACGGAGAGTAGGCAGATGG exon 14 CTGCTTGTAGTTGGGCAAGG CAAACTTCCAGGGTTTGTGG exon 15 GGTTCCCTTTGCTTTCAGTG GAATCAGGGCTGGTCTTGTG exon 16 AGGAATCCTGGGATTTTTGC GCAAGAGGAGGATGGACAAG

problems of false positive results that can arise in case- sequence (SNP13 and SNP18) had allele frequencies control based association studies. у10%.

Association analysis Results The two amino acid changing SNPs were genotyped in the IBD families, as we hypothesised that these may be Polymorphisms directly disease causative. Two intronic SNPs, common Genomic sequence was obtained, enabling all ITGB7 Ј in the healthy population, were also genotyped as these exons and 1.7 kb of 5 promoter region to be screened may represent markers in linkage disequilibrium with a for polymorphisms (primers listed in Table 1). Revised true disease allele. Genotypes were available on 567 IBD Ј 22,23 sequence was obtained for the 5 promoter region. trios (254 ulcerative colitis, 13 indeterminate colitis and Fourteen single nucleotide polymorphisms (SNPs) were 300 Crohn’s disease) from 464 families. No significant identified by mutation detection and sequencing (Table results were obtained with SNP3 (H672Y), SNP13 or 2). We developed genotyping assays and confirmed two SNP18 for the IBD, UC or CD phenotype (Table 4). Four potentially functional SNPs and eight intronic SNPs haplotypes (Ͼ1% frequency) of the three SNPs were (Table 3). SNP3, a histidine to tyrosine change at amino observed, no significant haplotype TDT results were F acid 672 (GenBank accession number NP 000880), results obtained for any phenotype. No significant results were in a charge change in the region between the extracellular found when the multiplex and simplex families were cysteine repeats and the transmembrane domain. SNP4, analysed separately for the three phenotypes, and allele a threonine to methionine change at amino acid 200, frequencies did not differ between affecteds and controls results in a size and hydrophobicity change in the (data not shown). mucosal addressin molecule-1 binding 24 The T200M mutation was found in only two families region. Two out of eight SNPs confirmed in intronic (one of which was used for mutation detection), and equal patterns of transmission and non-transmission were observed in these families. Table 2 SNPs discovered by mutation detection and sequencing

SNP Sequence Position Discussion Linkage analysis has localised an IBD susceptibility gene 18 TTCCC(T/A)CCTCC intronic 171 bp 3Ј of exon 1 Ј to chromosome 12q. The ITGB7 gene, within this IBD2 11 GAGGC(T/C)CAGTC intronic 39 bp 3 of exon 2 locus, plays a key role in the induction of intestine spe- 12 TCCCC(C/A)AATCC intronic 48 bp 3Ј of exon 2 13 GTGCA(C/T)GCCAC intronic 274 bp 3Ј of exon 2 cificinflammation (through leukocyte recruitment and 14 TCTTG(C/T)TTTGT intronic 158 bp 3Ј of exon 2 retention). We therefore sought potential disease causing 4 CAAAA(C/T)GGTGC T200M mutations in this positional candidate gene, encouraged 6 GGGGC(G/A)GGGAT intronic 16 bp 3Ј of exon 6 by the discovery of ␤ integrin abnormalities in other con- Ј 5 AGGTC(T/A)GTTTG intronic 27 bp 3 of exon 6 ditions. 7 CCCTG(T/C)CCCAG intronic 6 bp 5Ј of exon 8 Ј We established genomic sequence for ITGB7, and 8 GTGTG(T/C)GCATG intronic 41 bp 5 of exon 11 Ј 9 CTTCA(A/C)CCACC intronic 24 bp 3Ј of exon 11 screened the entire coding region and 5 region for 3 GTGCC(C/T)ATACC H672Y sequence variants. We had hypothesised that altered leu- 2 CAAGT(G/A)TGTCT intronic 21 bp 5Ј of exon 15 kocyte ITGB7 expression levels may play a role in IBD 1 CAACT(G/A)ATGAT intronic 58 bp 5Ј of exon 15 susceptibility, thus it was important to screen the actual ITGB7 promoter region for polymorphisms. As our

Genes and Immunity ␤7 integrin polymorphisms in IBD DA van Heel et al 457 h controls size (bp)255 size (bp) 223, 32 ) Major allele Minor allele Ј to 3 Ј ) Reverse primer (5 Ј to 3 Ј ITGB7 single nucleotide polymorphism assays, fragment sizes after restriction enzyme digestion, and minor allele frequencies observed in 90 healt Table 3 SNP Freq.SNP18SNP11 Enzyme 0.10SNP12 0.03SNP13 Forward BseRI primerSNP4 (5 0.03 DdeISNP6 0.14SNP8 0.00 BstXI AGGGCTCTGACCTGGATACCSNP3 0.04 ApaLISNP2 AAAGCCGCCAAGTATGTCAG 0.01SNP1 NlaIII 0.02 AAAGCCGCCAAGTATGTCAG AciI AAAGCCGCCAAGTATGTCAG 0.05 HhaI 0.04 Bsp1286I TCCCAGGTTTTGGTTCCTTTGTGGACAACA BsmAI CACCATGTGCTGTCCCTGACG CTGCTTGTAGTTGGGCAAGG PvuII TAACATGACCCCATCCCTTC AGAACTGTGAATCCAGGACAAAG ATGGTGAAAGCTGAATGGTGACT CACTCATGGGAAACTGATGC TTGGACAGGCTGGACACACTGTGAGCAGCT TTCGAAACCAGCCTAGTCAAC TGGCCAAGGTGGGCAAAATGGCCTACCAAAGG TTCGAAACCAGCCTAGTCAAC CTCCTTCTCAAAGCGACTGTATTC 441, 79 GAACTGTGCACTACACAACC CTGCTCCCTCTGTGAACAAGAAACCAGAGA CAAACTTCCAGGGTTTGTGG GTTGTTGGGAGCCAGGTG 93, 191, 74 520 25 210, 122, 99, 80, 9 195, 28 290, 125 122, 99, 9 216 168 101, 67, 43 223 140, 122, 57 353 95, 30 168, 43 262, 57 280, 73

Genes and Immunity ␤7 integrin polymorphisms in IBD DA van Heel et al 458 Table 4 ITGB7 SNP minor allele and SNP haplotype transmission present at low allele frequencies (р10%) in the Caucasian data in combined multiplex and simplex families healthy control population. We did not identify any com- mon functional polymorphisms in ITGB7, despite using Single SNP CD UC IBD an efficient and powerful mutation detection method with sensitivity and specificity reported to range from 3 (H672Y) 11 TR:13 NT 11 TR: 10 NT 22 TR: 24 NT 96–100%.30–32 We were, however, unable to assess poly- 13 42 TR: 39 NT 34 TR: 26 NT 77 TR: 67 NT morphisms in as yet unknown genomic regions which 18 44 TR: 48 NT 38 TR: 30 NT 83 TR: 78 NT may regulate ITGB7 expression. SNP Haplotype Beta 7 integrin is likely to be central to leukocyte traf- 13.1/3.1/18.1 64 TR: 62 NT 51 TR: 68 NT 116 TR: 130 NT ficking in other inflammatory diseases of the intestine. 13.2/3.1/18.1 34 TR: 31 NT 32 TR: 24 NT 66 TR: 56 NT 13.1/3.1/18.2 30 TR: 32 NT 33 TR: 25 NT 64 TR: 57 NT We have facilitated the assessment of ITGB7 as a candi- 13.1/3.2/18.1 9 TR: 12 NT 11 TR: 9 NT 20 TR: 22 NT date gene for gastrointestinal disease through discovery of novel polymorphisms and design of primers to screen Count of TR, transmissions and NT, non-transmissions of minor for promoter and exonic variation. Although our study SNP alleles or haplotypes from heterozygous parent to affected cannot completely exclude linkage disequilibrium child shown. All transmissions vs non-transmissions non-significant between IBD and ITGB7 polymorphisms, ITGB7 is (P Ͼ 0.05). Significant partial linkage disequilibrium was observed Ј = ෂ unlikely to be the IBD2 gene and future IBD genetic stud- between SNP13 and SNP18 (D 0.77, P 0.003, 2 kb apart), but ies should concentrate on other transcripts in the chromo- not between other polymorphisms. some 12 linkage region.

revised 5Ј region human sequence showed homology with the mouse promoter (shown to be functional in Materials and methods luciferase reporter gene assays), we screened 1.7 kb for promoter polymorphisms.25 We deliberately chose to Genomic sequencing screen parents of IBD families, who are more likely to be In order to optimise primer design and allow for uneven heterozygous for a disease allele in the case of recessive melting at the ends of DNA fragments during mutation or additive inheritance and carry with detection, at least 50 bp of intron sequence 5Ј and 3Ј of healthy as well as disease variants. Although there is each exon was needed. This was not available from the some evidence that UC may play a greater role in the published ITGB7 gene structure or sequence database IBD2 locus, linkage to this region has also been observed searches.33 A bacterial artificial chromosome clone in pure CD families.8,26 The existence of genetic hetero- (482a8) was screened, using primers for the 5Ј and 3Ј geneity may complicate gene identification in IBD, but as untranslated region of ITGB7, from the human CITB B & evidence exists for linkage of IBD2 to both UC and CD C libraries (Research Genetics, Huntsville, TX, USA). Pri- we screened families of both disease types for polymor- mers were designed for inverse PCR or PCR between phisms. exons.23 PCR products were sequenced using dye termin- Although the primary aim of this study was to identify ator chemistry and standard protocols on ABI 377 disease causing mutations (likely to be in exons or sequencers. To avoid confusion we have numbered the promoter/regulatory regions) SNPs within introns were ITGB7 exons from 1 to 16, with 1 corresponding to the 5Ј also discovered. These polymorphisms, whilst unlikely to exon reported after the initial paper describing the ITGB7 be functional, can be used as markers for potential dis- gene structure.22 Working draft sequence (GenBank ease alleles when linkage disequilibrium exists. Thus accession number AC073573) containing part of ITGB7 common intronic SNPs may be used to confirm or rule has been recently published by the Pro- out a gene containing disease causing alleles, as observed ject, and agrees with the sequence obtained in this study. in a study in late-onset Alzheimer disease which found evidence of association for six out of 12 SNPs spanning Mutation detection 40 kb around the APOE-4 susceptibility allele.27 How- Unrelated parents of multiply affected IBD families were ever, no evidence for association was observed for either screened for polymorphisms in ITGB7. We screened 24 coding or intronic SNPs genotyped in our study. Whilst individuals from seven CD, four UC and one mixed dis- it is possible that we may have missed a rare causative ease family. This design has 90% power to detect alleles mutation in the screening of ITGB7, the data from of Ͼ5% frequency. Nineteen primer pairs were designed intronic SNPs reduce this possibility and provide further to amplify all 16 exons and the 5Ј region (Table 1). PCR evidence against ITGB7 as the IBD2 gene. fragments (size 238–898 bp) were analysed for purity by The use of a large cohort of family-based controls in agarose gel electrophoresis, heated and slowly cooled to association studies provides both the power to detect a allow hybridisation, and injected into a denaturing high true association and a test robust to population hetero- pressure liquid chromatography column (Transgenomic geneity. Simulation studies in IBD have suggested WAVE).34 Optimal denaturing temperature(s) for between 500–800 triplets are necessary to identify modest mutation detection were calculated by experimentally risk loci.28 The 567 IBD trios we analysed had 80% power determined melting curves and use of WAVEMAKER to detect a common disease allele with a genotype rela- software (Transgenomic, San Jose, USA). Samples were tive risk of 1.5 or above. Our study, therefore, was pow- run at several temperatures if different melting domains ered to exclude a substantial role for ITGB7 polymor- existed. Fragments were eluted in a triethyl ammonium phisms in IBD susceptibility, although to exclude acetate buffer, with a linear acetonitrile gradient (as mutations of weak effect would require thousands of manufacturer’s protocol). PCR products showing hetero- families.29 duplex traces were direct sequenced in both forward and The majority of ITGB7 SNPs we discovered were reverse strands.

Genes and Immunity ␤7 integrin polymorphisms in IBD DA van Heel et al 459 Study subjects families using Simwalk2 and the GOLD software pack- Northern European Caucasian patients and families were age.41 recruited from the IBD clinics in Oxford, High Wycombe, Plymouth (UK) and from UK consultant gastroenterolog- ists. Case notes of patients were reviewed and a diagnosis Acknowledgements of ulcerative colitis, indeterminate colitis or Crohn’s dis- David van Heel is a Medical Research Council Clinical ease confirmed by standard endoscopic, histopatho- Training Fellow. We thank all participating families, our logical and radiological criteria. Questionnaires were sent research nurses Heather Holt, Daphne Lever and Sue to all participating family members and informed consent Goldthorpe, the National Association for Crohn’s and was obtained. Ethical approval for this study was Colitis, and consultants and general practitioners who obtained from the Central Oxford Research Ethics Com- helped ascertain families. mittee. Genotypes were obtained for 567 trios (both par- ents and affected child) from 102 multiply affected famil- Electronic Database Information ies (ascertained as affected sibling pairs) and 362 simplex ASPEX program package, ftp://lahmed.stanford.edu/ families (one affected child only). Ninety Caucasian heal- pub/aspex thy controls were recruited from staff of the Wellcome Trust Centre for Human Genetics and Oxford hospitals. Sequencing protocols, http://www.well.ox.ac.uk/ genomics/sequencing.html Genotyping Online Mendelian Inheritance in Man, http://www.ncbi. Genomic DNA was extracted from peripheral blood leu- nlm.nih.gov:80/entrez/query.fcgi?db=OMIM kocytes and quantified by fluorometric assay. Restriction enzyme digestion assays were designed using the MIS- MATCH program.35 PCR products were digested and References restriction fragment length polymorphisms analysed by 1 van Heel DA, Satsangi J, Carey AH, Jewell DP. Inflammatory agarose gel electrophoresis (assay conditions available on bowel disease: progress toward a gene. Can J Gastroenterol 2000; request). Allele frequencies were established in a panel 14: 207–218. of 90 healthy controls. Potentially functional polymor- 2 Probert CSJ, Jayanthi V, Hughes AO et al. Prevalence and family phisms, and any other polymorphisms of frequency risk of ulcerative colitis and Crohn’s disease: an epidemiological у10% were genotyped in the simplex and multiplex fam- study among Europeans and South Asians in Leicestershire. Gut ilies. Common intronic polymorphisms, although 1993; 34: 1547–1551. unlikely to be directly disease causing, may act as mark- 3 Peeters M, Nevens H, Baert F et al. Familial aggregation in Crohn’s disease: increased age adjusted risk and concordance ers when in linkage disequilibrium with a true disease in clinical characteristics. Gastroenterology 1996; 111: 597–603. causing allele. Mendelian inheritance within each family 4 Orholm M, Munkolm P, Langholz E et al. Familial occurrence was checked. of inflammatory bowel disease. N Engl J Med 1991; 324:84–88. Mutation detection and genotyping reactions were car- 5 Satsangi J, Rosenberg WMC, Jewell DP. The prevalence of ried out using 20 ng genomic DNA, 10 pM each forward inflammatory bowel disease in relatives of patients with Crohn’s and reverse primer, Amplitaq Gold, PCR buffer II disease. Euro J Gastro Hepatol 1994; 6: 413–416. (Applied Biosystems, Warrington, UK) and 2.5 mM Mg++ 6 Satsangi J, Parkes M, Louis E et al. Two-stage genome-wide per reaction. Annealing temperatures and number of search in inflammatory bowel disease: evidence for suscepti- cycles were optimised for each primer pair. bility loci on chromosomes 3, 7 and 12. Nat Genet 1996; 14: 199–202. 7 Duerr RH, Barmada MM, Zhang L et al. Linkage and association Analysis between inflammatory bowel disease and a locus on chromo- The TDT measures transmission of a marker allele to some 12. Am J Hum Genet 1998; 63:95–100. affected offspring from parents who are heterozygous for 8 Yang H, Ohmen J, Ma Y et al. Additional evidence of linkage the allele.36 The transmitted and non-transmitted alleles between Crohn’s disease and a putative locus on chromosome are compared by a standard chi-squared test. In contrast 12. Genet Med 1999; 1: 194–199. to the case-control association study (where cases and 9 Hampe J, Schreiber S, Shaw SH et al. A genomewide analysis controls are unrelated) the TDT uses internal controls provides evidence for novel linkages in inflammatory bowel dis- from each family tested. This provides a more robust ease in a large European cohort. Am J Hum Genet 1999; 64: association test, unaffected by population stratification. 808–816. 10 Akolkar P, Gulwani B, Lin X et al. Fine mapping of regions We estimated that 244 TDT trios were needed to have linked to Crohn’s disease on chromosomes 12 and 16. Am J Hum 80% power to detect association (assuming significance P = Genet 1998; 63: A279 (abst. A279). 0.05, a multiplicative model, minor allele frequency 11 Salmi M, Jalkanen S. Molecules controlling lymphocyte 37,38 у10%, genotype relative risk 2.0). ASPEX was used migration to the gut. Gut 1999; 45: 148–153. to calculate a classical TDT using both parents and one 12 Cepek KL, Shaw SK, Parker CM et al. Adhesion between epi- or more affected probands.36 This test has properties of thelial cells and T lymphocytes mediated by E-cadherin and the both linkage and association if multiple affected siblings alpha E beta 7 integrin. Nature 1994; 372: 190–193. are used from the same family. To make the tests of 13 Higgins JM, Mandlebrot DA, Shaw SK et al. Direct and regulated association independent of any linkage present, the level interaction of integrin alphaEbeta7 with E-cadherin. J Cell Biol of significance was calculated empirically using simul- 1998; 140: 197–210. 39 14 Hesterberg PE, Winsor-Hines D, Briskin MJ et al. Rapid resol- ation. Multiple markers were analysed for association ution of chronic colitis in the cotton-top tamarin with an anti- with an unbiased haplotype TDT, corrected for multiple body to a gut-homing beta 7. Gastroenterology 40 affected siblings. Analyses were performed for CD, UC 1996; 111: 1373–1380. and combined IBD phenotypes. Standardized disequilib- 15 Picarella D, Hurlbut P, Rottman J et al. Monoclonal antibodies rium coefficents (DЈ) were calculated from the combined specific for beta 7 integrin and mucosal addressin cell adhesion

Genes and Immunity ␤7 integrin polymorphisms in IBD DA van Heel et al 460 molecule-1 (MAdCAM-1) reduce inflammation in the colon of for polygenic disorders: considerations for study design in the scid mice reconstituted with CD45RBhighCD4+ T cells. J Immu- complex trait of inflammatory bowel disease. Hum Hered 2000; nol 1997; 158: 2099–2106. 50:91–101. 16 Artis D, Humphreys NE, Potten CS et al. Beta7 integrin-deficient 29 Risch N, Merikangas K. The future of genetic studies of human mice: delayed leukocyte recruitment and attenuated protective complex diseases. Science 1996; 273: 1516–1517. immunity in the small intestine during enteric helminth infec- 30 O’Donovan MC, Oefner PJ, Roberts SC et al. Blind analysis of tion. Eur J Immunol 2000; 30: 1656–1664. denaturing high-performance liquid chromatography as a tool 17 Wagner N, Lohler J, Kunkel EJ et al. Critical role for beta7 inte- for mutation detection. Genomics 1998; 52:44–49. grins in formation of the gut-associated lymphoid tissue. Nature 31 Jones AC, Austin J, Hansen N et al. Optimal temperature selec- 1996; 382: 366–370. tion for mutation detection by denaturing HPLC and compari- 18 Etzioni A, Harlan J. Adhesion molecules and leukocyte adhesion son to single-stranded conformation polymorphism and hetero- defects. In: Smith E, Puck J, Ochs H (eds). Primary Immunodefi- duplex anlaysis. Clin Chem 1999; 45: 1133–1140. ciency Diseases, a Molecular and Genetic Approach. Oxford Univer- 32 Buyse IM, Fang P, Hoon KT et al. Diagnostic testing for Rett sity Press: Oxford, 1998, pp 435–452. syndrome by DHPLC and direct sequencing analysis of the 19 Georg JN, Caen JP, Nurden AT. Glanzmann’s thrombasthenia: MECP2 gene: identification of several novel mutations and poly- the spectrum of clinical disease. Blood 1990; 75: 1383–1395. morphisms. Am J Hum Genet 2000; 67: 1428–1436. 20 Takizawa Y, Shimizu H, Nishikawa T et al. Novel ITGB4 33 Jiang WM, Jenkins D, Yuan Q et al. The gene organization of mutations in a patient with junctional epidermolysis bullosa- the human beta 7 subunit, the common beta subunit of the leu- pyloric atresia syndrome and altered basement membrane zone kocyte integrins HML-1 and LPAM-1. Int Immunol 1992; 4: immunofluorescence for the alpha6beta4 integrin. J Invest 1031–1040. 34 Kuklin A, Munson K, Gjerde D, Haefele R, Taylor P. Detection Dermatol 1997; 108: 943–946. of single-nucleotide polymorphisms with the WAVE DNA frag- 21 Giancotti FG. Complexity and specificity of integrin signalling. ment analysis system. Genet Test 1997; 1: 201–206. Nat Cell Biol 2000; 2: E13–E14. 35 Davidow LS. Selecting PCR designed mismatch primers to cre- 22 Mead PE, Leung E, Yang Y et al. Isolation of the 5Ј region of the ated diagnostic restriction sites. Comput Appl Biosci 1992; 8: human ITGB7 integrin gene. Immunogenetics 1994; 39: 375–376. 193–194. 23 van Heel DA, Allen M, Jewell DP, Carey AH. A revised 36 Spielman RS, McGinnis RE, Ewens WJ. Transmission test for sequence of the human beta7 integrin gene (ITGB7) promoter linkage disequilibrium: the insulin gene region and insulin- region obtained by inverse PCR. Immunogenetics 2000; 51: 863– dependent diabetes mellitus (IDDM). Am J Hum Genet 1993; 52: 865. 506–516. 24 Tidswell M, Pachynski R, Wu SW et al. Structure-function analy- 37 Altshuler D, Hirschhorn JN, Klannemark M et al. The common sis of the subunit: identification of domains PPARgamma Pro12Ala polymorphism is associated with involved in adhesion to MAdCAM-1. J Immunol 1997; 159: decreased risk of type 2 diabetes. Nat Genet 2000; 26:76–80. 1497–1505. 38 Mitchell LE. Relationship between case-control studies and the 25 Lim SP, Leung E, Krissansen GW. The beta7 integrin gene (Itgb- transmission/disequilibrium test. Genet Epidemiol 2000; 19: 7) promoter is responsive to TGF-beta1: defining control 193–201. regions. Immunogenetics 1998; 48: 184–195. 39 Risch N, Spiker D, Lotspeich L et al. A genomic screen of autism: 26 Parkes M, Barmada MM, Satsangi J et al. The IBD2 locus shows evidence for a multilocus etiology. Am J Hum Genet 1999; 65: linkage heterogeneity between ulcerative colitis and Crohn dis- 493–507. ease. Am J Hum Genet 2000; 57: 1605–1610. 40 Dudbridge F, Koeleman BP, Todd JA, Clayton DG. Unbiased 27 Martin ER, Lai EH, Gilbert JR et al. SNPing away at complex application of the transmission/disequilibrium test to multilo- diseases: analysis of single-nucleotide polymorphisms around cus haplotypes. Am J Hum Genet 2000; 66: 2009–2012. APOE in Alzheimer disease. Am J Hum Genet 2000; 67: 383–394. 41 Abecasis GR, Cookson WO. GOLD—graphical overview of link- 28 Hampe J, Wienker T, Nurnberg P, Schreiber S. Mapping genes age disequilibrium. Bioinformatics 2000; 16: 182–183.

Genes and Immunity