Portal de Periódicos da CAPES

Mapping Multiple Sclerosis Susceptibility to the HLA-DR Locus in African Americans

2004; Elsevier BV; Volume: 74; Issue: 1 Linguagem: Inglês

10.1086/380997

1537-6605

Jorge R. Oksenberg, Lisa F. Barcellos, Bruce Cree, Sergio E. Baranzini, Teodorica L. Bugawan, Omar Khan, Robin Lincoln, A. Swerdlin, Emmanuel Mignot, Ling Lin, Douglas S. Goodin, M Kellis, Silke Schmidt, Glenys Thomson, David Reich, Margaret A. Pericak‐Vance, Jonathan L. Haines, Stephen L. Hauser,

Immunotherapy and Immune Responses

An underlying complex genetic susceptibility exists in multiple sclerosis (MS), and an association with the HLA-DRB1*1501-DQB1*0602 haplotype has been repeatedly demonstrated in high-risk (northern European) populations. It is unknown whether the effect is explained by the HLA-DRB1 or the HLA-DQB1 gene within the susceptibility haplotype, which are in strong linkage disequilibrium (LD). African populations are characterized by greater haplotypic diversity and distinct patterns of LD compared with northern Europeans. To better localize the HLA gene responsible for MS susceptibility, case-control and family-based association studies were performed for DRB1 and DQB1 loci in a large and well-characterized African American data set. A selective association with HLA-DRB1*15 was revealed, indicating a primary role for the DRB1 locus in MS independent of DQB1*0602. This finding is unlikely to be solely explained by admixture, since a substantial proportion of the susceptibility chromosomes from African American patients with MS displayed haplotypes consistent with an African origin. An underlying complex genetic susceptibility exists in multiple sclerosis (MS), and an association with the HLA-DRB1*1501-DQB1*0602 haplotype has been repeatedly demonstrated in high-risk (northern European) populations. It is unknown whether the effect is explained by the HLA-DRB1 or the HLA-DQB1 gene within the susceptibility haplotype, which are in strong linkage disequilibrium (LD). African populations are characterized by greater haplotypic diversity and distinct patterns of LD compared with northern Europeans. To better localize the HLA gene responsible for MS susceptibility, case-control and family-based association studies were performed for DRB1 and DQB1 loci in a large and well-characterized African American data set. A selective association with HLA-DRB1*15 was revealed, indicating a primary role for the DRB1 locus in MS independent of DQB1*0602. This finding is unlikely to be solely explained by admixture, since a substantial proportion of the susceptibility chromosomes from African American patients with MS displayed haplotypes consistent with an African origin. Evidence of disease risk heritability in multiple sclerosis (MS [MIM 126200]) is supported by familial aggregation of cases (Ebers et al. Ebers et al., 1995Ebers GC Sadovnick AD Risch NJ A genetic basis for familial aggregation in multiple sclerosis.Nature. 1995; 377: 150-151Crossref PubMed Scopus (527) Google Scholar; Robertson et al. Robertson et al., 1996Robertson NP Fraser M Deans J Clayton D Walker N Compston DAS Age-adjusted recurrence risks for relatives of patients with multiple sclerosis.Brain. 1996; 119: 449-455Crossref PubMed Scopus (180) Google Scholar; Sadovnick et al. Sadovnick et al., 1996Sadovnick AD Ebers GC Dyment DA Risch NJ Evidence for genetic basis of multiple sclerosis.Lancet. 1996; 347: 1728-1730Abstract PubMed Scopus (234) Google Scholar, Sadovnick et al., 2000Sadovnick AD Yee IM Ebers GC Factors influencing sib risks for multiple sclerosis.Clin Genet. 2000; 58: 431-435Crossref PubMed Scopus (28) Google Scholar) and by the high prevalence in some ethnic populations (particularly those of northern European origin) compared with others (African and some Asian groups) irrespective of geographic location (Compston Compston, 1998Compston A Distribution of multiple sclerosis.in: Compston A Ebers H Lassman H McDonald I Matthews B Wekerle H McAlpine’s multiple sclerosis. Churchill Livingstone, London1998: 63-100Google Scholar). Modeling of the available data predicts that the MS-prone genotype results from multiple independent or interacting polymorphic genes, each exerting a small—or, at most, a moderate—effect on the overall risk. The HLA-DR2 haplotype (DRB1*1501-DQB1*0602) within the major histocompatibility complex on chromosome 6p21 has consistently demonstrated both linkage and association in family and case-control studies (Multiple Sclerosis Genetics Group Multiple Sclerosis Genetics Group, 1998Multiple Sclerosis Genetics Group Linkage of the MHC to familial multiple sclerosis suggests genetic heterogeneity.Hum Molec Genet. 1998; 7: 1229-1234Crossref PubMed Google Scholar). However, fine-mapping studies have not settled whether the effect is explained by the DRB1 gene itself (MIM 142857); by another closely spaced gene within the class II HLA region, such as DQB1 (MIM 604305); or by some other nearby gene in strong disequilibrium with the HLA-DR locus. Analyses have been made more complex by extensive linkage disequilibrium (LD) occurring across the region and by the presence of >240 genes within this superlocus, many of which have roles in immune function and are thus plausible disease candidates. Results suggesting that genes of interest for MS exist within the class III (de Jong et al. de Jong et al., 2002de Jong BA Huizinga TW Zanelli E Giphart MJ Bollen EL Uitdehaag BM Polman CH Westendorp RG Evidence for additional genetic risk indicators of relapse-onset MS within the HLA region.Neurology. 2002; 59: 549-555Crossref PubMed Scopus (52) Google Scholar; Palacio et al. Palacio et al., 2002Palacio LG Rivera D Builes JJ Jimenez ME Salgar M Anaya JM Jimenez I Camargo M Arcos-Burgos M Sanchez JL Multiple sclerosis in the tropics: genetic association to STR’s loci spanning the HLA and TNF.Mult Scler. 2002; 8: 249-255Crossref PubMed Scopus (20) Google Scholar) and/or telomeric to the class I regions (Shinar et al. Shinar et al., 1998Shinar Y Pras E Siev-Ner I Gamus D Brautbar C Israel S Achiron A Analysis of allelic association between D6S461 marker and multiple sclerosis in Ashkenazi and Iraqi Jewish patients.J Mol Neurosci. 1998; 11: 265-269Crossref PubMed Scopus (7) Google Scholar; Fogdell-Hahn et al. Fogdell-Hahn et al., 2000Fogdell-Hahn A Ligers A Gronning M Hillert J Olerup O Multiple sclerosis: a modifying influence of HLA class I genes in an HLA class II associated autoimmune disease.Tissue Antigens. 2000; 55: 140-148Crossref PubMed Scopus (211) Google Scholar; Marrosu et al. Marrosu et al., 2001Marrosu MG Murru R Murru MR Costa G Zavattari P Whalen M Cocco E Mancosu C Schirru L Solla E Fadda E Melis C Porru I Rolesu M Cucca F Dissection of the HLA association with multiple sclerosis in the founder isolated population of Sardinia.Hum Mol Genet. 2001; 10: 2907-2916Crossref PubMed Scopus (127) Google Scholar; Rubio et al. Rubio et al., 2002Rubio JP Bahlo M Butzkueven H van Der Mei IA Sale MM Dickinson JL Groom P Johnson LJ Simmons RD Tait B Varney M Taylor B Dwyer T Williamson R Gough NM Kilpatrick TJ Speed TP Foote SJ Genetic dissection of the human leukocyte antigen region by use of haplotypes of Tasmanians with multiple sclerosis.Am J Hum Genet. 2002; 70: 1125-1137Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar) have been reported as well. Because patterns of LD differ between populations, the most direct and practical approach to resolving this complex genetic obstacle will be to scrutinize and compare a large number of MS haplotypes in well-characterized data sets from distinct populations. However, studies of HLA and MS in low-risk ethnic groups have generally utilized small cohorts, and the results have been inconclusive or conflicting (Serjeantson et al. Serjeantson et al., 1992Serjeantson SW Gao X Hawkins BR Higgins DA Yu YL Novel HLA-DR2-related haplotypes in Hong Kong Chinese implicate the DQB1*0602 allele in susceptibility to multiple sclerosis.Eur J Immunogenet. 1992; 19: 11-19Crossref PubMed Scopus (51) Google Scholar; Kelly et al. Kelly et al., 1995Kelly MA Jacobs KH Penny MA Mijovic CH Nightingale S Barnett AH Francis DA An investigation of HLA-encoded genetic susceptibility to multiple sclerosis in subjects of Asian Indian and Afro-Caribbean ethnic origin.Tissue Antigens. 1995; 45: 197-202Crossref PubMed Scopus (30) Google Scholar; Caballero et al. Caballero et al., 1999Caballero A Alves-Leon S Papais-Alvarenga R Fernandez O Navarro G Alonso A DQB1*0602 confers genetic susceptibility to multiple sclerosis in Afro-Brazilians.Tissue Antigens. 1999; 54: 524-526Crossref PubMed Scopus (71) Google Scholar; Kwon et al. Kwon et al., 1999Kwon OJ Karni A Israel S Brautbar C Amar A Meiner Z Abramsky O Karussis D HLA class II susceptibility to multiple sclerosis among Ashkenazi and non-Ashkenazi Jews.Arch Neurol. 1999; 56: 555-560Crossref PubMed Scopus (63) Google Scholar; Quelvennec et al. Quelvennec et al., 2003Quelvennec E Bera O Cabre P Alizadeh M Smadja D Jugde F Edan G Semana G Genetic and functional studies in multiple sclerosis patients from Martinique attest for a specific and direct role of the HLA-DR locus in the syndrome.Tissue Antigens. 2003; 61: 166-171Crossref PubMed Scopus (26) Google Scholar). Compared with northern Europeans and white Americans, African Americans are at a lower risk for MS (Kurtzke et al. Kurtzke et al., 1979Kurtzke JF Beebe GW Norman Jr, JE Epidemiology of multiple sclerosis in U.S. veterans: 1. Race, sex, and geographic distribution.Neurology. 1979; 29: 1228-1235Crossref PubMed Google Scholar), but genetic studies of HLA and other loci in this population can be extremely valuable for disease gene identification, because of the presence of distinctive patterns of disequilibrium shaped by the population history of the group (Just et al. Just et al., 1997Just JJ King MC Thomson G Klitz W African-American HLA class II allele and haplotype diversity.Tissue Antigens. 1997; 49: 547-555Crossref PubMed Scopus (40) Google Scholar; Zachary et al. Zachary et al., 1997Zachary AA Leffell MS Johnson A Rose SM Bias WB HLA antigens, alleles, and haplotypes among African-Americans.Transplant Proc. 1997; 29: 3706Abstract Full Text PDF PubMed Scopus (3) Google Scholar). Some combinations in cis of DRB1 and DQB1 alleles are unique to African Americans, and, more specifically, the DRB1*1501 and DQB1*0602 alleles do not display the high degree of LD and haplotype fixity in African Americans that is characteristic of Europeans. To localize the HLA gene responsible for MS susceptibility, we performed case-control and family-based association studies for DRB1 and DQB1 loci in a large African American cohort consisting of 1,003 genotyped individuals. The data set comprised 336 unrelated patients with MS (female:male ratio 4.6:1; mean age at onset 32.7±9.4 years; mean disease duration 9.1±7.1 years), available nuclear family members (n=357), and unrelated control individuals (n=310, primarily patient spouses). Families included 33 complete trios (both parents and the affected individual), 162 families with one parent, and 100 discordant sib pairs (one affected and one unaffected individual). All study participants were self-reported African Americans, but European ancestry in patients and controls was documented on the basis of genotyping results of 186 SNPs. Based on the use of two parent populations, West Africans and Europeans, these SNPs have a mean 54% allele frequency differences between the parental populations and are spaced at least 10 cM from each other across the genome (Parra et al. Parra et al., 1998Parra EJ Marcini A Akey J Martinson J Batzer MA Cooper R Forrester T Allison DB Deka R Ferrell RE Shriver MD Estimating African American admixture proportions by use of population-specific alleles.Am J Hum Genet. 1998; 63: 1839-1851Abstract Full Text Full Text PDF PubMed Scopus (632) Google Scholar). Global estimation of European ancestry in 50 patients and 50 controls indicated 19.1% admixture in both groups (data not shown). Institutional review board approval and informed consent were obtained from all study participants. Medical records of all patients were reviewed by one of the authors (S.L.H. or B.A.C.C.) and, in all cases, diagnosis was confirmed using standard criteria (Poser et al. Poser et al., 1983Poser CM Paty DW Scheinberg L McDonald WI Davis FA Ebers GC Johnson KP Sibley WA Silberberg DH Tourtellotte WW New diagnostic criteria for multiple sclerosis: guidelines for research protocols.Ann Neurol. 1983; 13: 227-231Crossref PubMed Scopus (6891) Google Scholar). A total of 92% of patients had relapsing-remitting MS at onset. Global testing for class II HLA-DRB1 and -DQB1 in the African American patient-control data set revealed differences in allele distributions (P=4.0×10−4 and P=.02, respectively). Whereas the DQB1 effect was restricted to the DQB1*0602 allele only (odds ratio [OR] 1.4; P=.01), disease association at the DRB1 locus was found for both DRB1*15 and DRB1*0301 alleles (OR 1.7, P=6.0×10−4 for DRB1*15; and OR 1.7, P=.01 for DRB1*0301; see table 1). As observed in white MS data sets, there was an increased risk, albeit not as pronounced, for the DRB1*1501 allele (OR 1.7; P=.03) and the DRB1*1501-DQB1*0602 haplotype (OR 1.6; P=.049) in African Americans (table 1). In addition, associations were observed for both the DRB1*1503 allele (OR 1.6; P=.015) and the common African haplotype DRB1*1503-DQB1*0602 (OR 1.5; P=.02). Two-locus DRB1-DQB1 haplotypes were deduced on the basis of known associations (Fernandez-Vina et al. Fernandez-Vina et al., 1991Fernandez-Vina M Moraes JR Moraes ME Miller S Stastny P HLA class II haplotypes in Amerindians and in black North and South Americans.Tissue Antigens. 1991; 38: 235-237Crossref PubMed Scopus (57) Google Scholar; Lin et al. Lin et al., 1997Lin L Jin L Kimura A Carrington M Mignot E DQ microsatellite association studies in three ethnic groups.Tissue Antigens. 1997; 50: 507-520Crossref PubMed Scopus (43) Google Scholar). As an additional measure of validation, DR-DQ haplotype frequencies in patients and controls were also estimated from genotype data via the method of maximum likelihood, through the use of the expectation-maximization algorithm (Arlequin version 2.0 [Arlequin's Home on the Web]) under the assumption of Hardy-Weinberg equilibrium (Excoffier and Slatkin Excoffier and Slatkin, 1995Excoffier L Slatkin M Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population.Mol Biol Evol. 1995; 12: 921-927PubMed Google Scholar; Long et al. Long et al., 1995Long JC Williams RC Urbanek M An E-M algorithm and testing strategy for multiple locus haplotypes.Am J Hum Genet. 1995; 56: 799-810PubMed Google Scholar), and were confirmed, whenever possible, by available family data.Table 1HLA-DR2– and HLA-DR3–Related Alleles and Haplotypes in African American Patients with MS and ControlsNo. (%) ofAllele or HaplotypePatients (2N=672)Controls (2N=620)OR95% CIPValueAlleleaReference group: all other DRB1 or DQB1 alleles. P values (χ2 or Fisher’s exact test, when appropriate), ORs, and CIs were derived using PROC FREQ (SAS version 8.2).: DRB1*030172 (10.7)42 (6.8)1.71.1–2.5.01 DRB1*15bIncludes DRB1*1501 and *1503 alleles only. DRB1*1502 was also present in patients and controls but was very rare (.1% and .8%, respectively).149 (22.2)91 (14.7)1.71.2–2.2.0005 DRB1*150149 (7.3)27 (4.4)1.71.1–2.8.03 DRB1*1503100 (14.9)64 (10.3)1.51.1–2.1.014 DQB1*0602158 (23.5)111 (17.9)1.41.1–1.9.01DRB1-DQB1 haplotypecThe reference group for DRB1*0301-DQB1*0200 haplotype comparison was all other haplotypes. To distinguish the HLA-DR2 related haplotype effect, all non-DRB1*15 and non-DQB1*0602 haplotypes were used as the reference group for DRB1*15- and DQB1*0602-related haplotypes.: 0301-0200dPrimarily DRB1*0301-DQB1*0201 haplotypes. Does not include DRB1*03021-DQB1*0402 haplotypes.72 (10.7)41 (6.6)1.71.1–2.5.009 1501-060241 (6.1)25 (4.0)1.71.0–2.8.049 1503-060290 (13.4)61 (9.8)1.51.1–2.1.02 15-0602131 (19.5)86 (13.9)1.51.1–2.1.004 X-060227 (4.0)25 (4.0)1.1.6–1.9.74 15-X18 (2.7)5 (.8)3.71.3–9.9.007 1501-X8 (1.2)2 (.3)4.1.9–19.2.06 1503-X10 (1.5)3 (.5)3.4.9–12.4.05Note.—Molecular typing of HLA-DRB1 and -DQB1 loci was performed using a nonradioactive PCR-based sequence-specific oligonucleotide probe reverse line-blot assay (Dynal). DRB1*15/DRB1*03/DQB1*06 subtyping was performed for all individuals, through use of a similar approach, after allele-specific amplification of the DR2 allele (Bugawan et al. Bugawan et al., 2002Bugawan TL Klitz W Alejandrino M Ching J Panelo A Solfelix CM Petrone A Buzzetti R Pozzilli P Erlich HA The association of specific HLA class I and II alleles with type I diabetes among Filipinos.Tissue Antigens. 2002; 59: 452-469Crossref PubMed Scopus (64) Google Scholar). For other DRB1 and DQB1 alleles not distinguished by this method, the resolution is reported at the lowest level common to any particular allele.a Reference group: all other DRB1 or DQB1 alleles. P values (χ2 or Fisher’s exact test, when appropriate), ORs, and CIs were derived using PROC FREQ (SAS version 8.2).b Includes DRB1*1501 and *1503 alleles only. DRB1*1502 was also present in patients and controls but was very rare (.1% and .8%, respectively).c The reference group for DRB1*0301-DQB1*0200 haplotype comparison was all other haplotypes. To distinguish the HLA-DR2 related haplotype effect, all non-DRB1*15 and non-DQB1*0602 haplotypes were used as the reference group for DRB1*15- and DQB1*0602-related haplotypes.d Primarily DRB1*0301-DQB1*0201 haplotypes. Does not include DRB1*03021-DQB1*0402 haplotypes. Open table in a new tab Note.— Molecular typing of HLA-DRB1 and -DQB1 loci was performed using a nonradioactive PCR-based sequence-specific oligonucleotide probe reverse line-blot assay (Dynal). DRB1*15/DRB1*03/DQB1*06 subtyping was performed for all individuals, through use of a similar approach, after allele-specific amplification of the DR2 allele (Bugawan et al. Bugawan et al., 2002Bugawan TL Klitz W Alejandrino M Ching J Panelo A Solfelix CM Petrone A Buzzetti R Pozzilli P Erlich HA The association of specific HLA class I and II alleles with type I diabetes among Filipinos.Tissue Antigens. 2002; 59: 452-469Crossref PubMed Scopus (64) Google Scholar). For other DRB1 and DQB1 alleles not distinguished by this method, the resolution is reported at the lowest level common to any particular allele. It is important that the DRB1*1501 and *1503 disease associations were independent of DQB1*0602, as evidenced by the fact that chromosomes carrying the *0602 allele with other DRB1 alleles were present at identical frequencies in patients and controls (4.0%; P=.77). Conversely, DRB1*15 (including *1501 and *1503) chromosomes carrying other DQB1 alleles (non-*0602 and denoted by “X”) were significantly increased in patients with MS (OR 3.7; P=.01; see table 1). Patients and controls were also grouped, by genotype, according to whether they carried at least one DRB1*15 and/or one DQB1*0602 allele, to rule out the possibility of trans allelic effects (table 2). In these analyses, for example, the effect in an individual of DRB1*15 on one chromosome in trans with DQB1*0602 on the other chromosome is eliminated (see the DRB1*15−/DQB1*0602+ genotypic category), and an independent role for DQB1*0602 unmasked by DRB1*15 can be assessed. As shown in table 2, our results do not indicate the involvement of DQB1*0602 in MS susceptibility. An increased risk in individuals carrying the DRB1*15−/DQB1*0602+ genotype was not observed (OR 1.0; P=.99). Furthermore, the analyses of DRB1*15+/DQB1*0602− and DRB1*15−/DQB1*0602+ haplotypes did not reveal additional susceptibility alleles acting in an asymmetric fashion. A diverse array of DRB1 alleles, including DRB1*03, *08, *11, *13, *14 and *16, were present in similar proportions on DRB1*15−/DQB1*0602+ haplotypes in patients and controls (data not shown). DRB1*15+/DQB1*0602− haplotypes also displayed marked heterogeneity and included DQB1*02, *03, *0501, *502, *601 *0603, *0609, *0615, and *0617 alleles distributed equally in both groups. Nonetheless, much larger data sets will be required to exclude the possibility of minor heterogeneous multiallelic DRB1 and DQB1 interactions influencing disease susceptibility in MS.Table 2DRB1*15 and DQB1*0602 Genotypes in African American Patients with MS and ControlsNo. (%) ofGenotypePatientsControlsHWbNumber and percent expected under Hardy-Weinberg conditions.ORaThe reference group for all OR estimates comprised X/X (DRB1*15−/DQB1*0602−) individuals.95% CIP ValueDRB1*0602*15+/DQB1*0602+123 (36.6)76 (24.5)86 (27.7)1.61.1–2.2.01DRB1*15+/DQB1*0602−11 (3.3)4 (1.3)4 (1.3)3.0.9–9.6.05DRB1*15−/DQB1*0602+23 (6.8)24 (7.7)25 (8.1)1.0.5–1.9.99Note.—Patients and controls were grouped according to whether they carried at least one DRB1*15 and/or one DQB1*0602 allele. For example, the genotypic category DRB1*15+/DQB1*0602+ contains individuals carrying at least one DRB1*15 and one DQB1*0602 allele, regardless of chromosomal phase, and would include the following combination of genotypes: DRB1*15-DQB1*0602/DRB1*15-DQB1*0602, DRB1*15-DQB1*0602/DRB1*15-X (where “X” = non-DQB1*0602 alleles), DRB1*15-DQB1*0602/ X-DQB1*0602 (where “X” = non-DRB1*15 alleles), DRB1*15-DQB1*0602/X (where “X” = all other non-DRB1*15–non-DQB1*0602 haplotypes), and DRB1*15/X-DQB1*0602 (where “X” = non-DQB1*0602 alleles). Similarly, the DRB1*15+/DQB1*0602− category comprised individuals with DRB1*15-X/X genotypes (where “X” = non-DQB1*0602 alleles and “X” = all other non-DRB1*15–non-DQB1*0602 haplotypes, respectively) and DRB1*15-X/DRB1*15-X genotypes (where “X” = non-DQB1*0602 alleles). For these analyses, the Hardy-Weinberg expected genotype frequencies for the control subjects were used, rather than the actual observed genotype counts, to obtain greater precision in the OR estimates.a The reference group for all OR estimates comprised X/X (DRB1*15−/DQB1*0602−) individuals.b Number and percent expected under Hardy-Weinberg conditions. Open table in a new tab Note.— Patients and controls were grouped according to whether they carried at least one DRB1*15 and/or one DQB1*0602 allele. For example, the genotypic category DRB1*15+/DQB1*0602+ contains individuals carrying at least one DRB1*15 and one DQB1*0602 allele, regardless of chromosomal phase, and would include the following combination of genotypes: DRB1*15-DQB1*0602/DRB1*15-DQB1*0602, DRB1*15-DQB1*0602/DRB1*15-X (where “X” = non-DQB1*0602 alleles), DRB1*15-DQB1*0602/ X-DQB1*0602 (where “X” = non-DRB1*15 alleles), DRB1*15-DQB1*0602/X (where “X” = all other non-DRB1*15–non-DQB1*0602 haplotypes), and DRB1*15/X-DQB1*0602 (where “X” = non-DQB1*0602 alleles). Similarly, the DRB1*15+/DQB1*0602− category comprised individuals with DRB1*15-X/X genotypes (where “X” = non-DQB1*0602 alleles and “X” = all other non-DRB1*15–non-DQB1*0602 haplotypes, respectively) and DRB1*15-X/DRB1*15-X genotypes (where “X” = non-DQB1*0602 alleles). For these analyses, the Hardy-Weinberg expected genotype frequencies for the control subjects were used, rather than the actual observed genotype counts, to obtain greater precision in the OR estimates. All DRB1*0301-associated chromosomes in African American patients and controls carried DQB1*02 alleles; the vast majority were the common autoimmune disease-associated haplotype DRB1*0301-DQB1*0201. The effect for the DRB1*0301 allele considered alone or in combination with DQB1*02 was identical (OR 1.7; P=.01), and no evidence for association in patients with non-DRB1*0301 haplotypes bearing DQB1*02 alleles (or X-DQB1*02) was observed (OR 0.8; P=.14; data not shown). These results further support an independent role for DRB1*0301 in MS, rather than variation at the DQB1 locus. The patient data set was also compared with another large published African American control group (N=242 individuals) (Just et al. Just et al., 1997Just JJ King MC Thomson G Klitz W African-American HLA class II allele and haplotype diversity.Tissue Antigens. 1997; 49: 547-555Crossref PubMed Scopus (40) Google Scholar), with similar results (data not shown). Significant associations for DRB1*15 (OR 1.4; 95% CI 1.1–1.9; P=.02) and DRB1*0301 (OR 1.6; 95% CI 1.1–2.5; P=.03) and no independent association with DQB1*0602 (OR 1.0; 95% CI 0.8–1.3; P=.94) or X-*0602 haplotypes (OR 0.6; 95% CI 0.3–1.0; P=.04) were observed. Further, the frequencies of predominantly African DRB1*03021-DQB1*0402 and DRB1*1304-DQB1*0301 and of northern European DRB1*0800-DQB1*0402 haplotypes in patient and control groups were statistically indistinguishable (6.6%–5.2%, P=.57 for DRB1*03021- DQB1*0402; 0.5%–1.2%, P=.28 for DRB1*1304- DQB1*0301; and 0.5%–0.9%, P=.64 for DRB1 *0800-DQB1*0402). These observations, together with the SNP genotyping results mentioned above, indicate that the patients with MS and the controls used in this study were adequately matched. To further address the possibility that the case-control results might be influenced by population stratification, family-based analyses of the HLA-DRB1 and DQB1 loci were performed in a subgroup of patients for whom informative family controls were available (table 3). Evidence for excess transmission of the DRB1*1501 allele to affected individuals was observed when the pedigree disequilibrium test (PDT) was used, for all families (P=.04) and for trios only (P=.025). TRANSMIT (available from David Clayton's Web site) also yielded significant results for DRB1*1501 (P=.0006). Associations with other DRB1 or DQB1 alleles were not observed. Although only a few DRB1*1501, DQB1*X haplotypes were identified in the family data set, TRANSMIT results for this haplotypic classification were also supportive of an independent DRB1*1501 association, but not for DQB1*0602 (data not shown). The failure to detect evidence for statistically significant excess transmission of DRB1*1503 and DRB1*0301 alleles was likely due to the small number of informative trios available for analysis.Table 3African American MS Family-Based Association Test Results for HLA-DRB1 and HLA-DQB1TransmissionAllelePDT (All)aPDT analysis utilized 33 trios and 100 discordant sib pairs.PDT (Trios Only)aPDT analysis utilized 33 trios and 100 discordant sib pairs.TRANSMITbTRANSMIT analysis utilized 162 families.DRB1 global.33.73.057DRB1*1501.04.025.0006DRB1*1503.70.67.56DRB1*0301.30.56.39DQB1 global.47.84.34DQB1*0602.61.53.40Note.—Two complementary approaches were used to look for evidence of excess transmission of DRB1 and DQB1 alleles in African American families. The PDT (Martin et al. Martin et al., 2000Martin ER Monks SA Warren LL Kaplan NL A test for linkage and association in general pedigrees: the pedigree disequilibrium test.Am J Hum Genet. 2000; 67: 146-154Abstract Full Text Full Text PDF PubMed Scopus (518) Google Scholar, Martin et al., 2001Martin ER Bass MP Kaplan NL Correcting for a potential bias in the pedigree disequilibrium test.Am J Hum Genet. 2001; 68: 1065-1067Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar; Pedigree Disequilibrium Test Analysis Program Web site) can utilize both discordant sib pairs and nuclear families from extended pedigrees. TRANSMIT (Clayton and Jones Clayton and Jones, 1999Clayton D Jones H Transmission/disequilibrium tests for extended marker haplotypes.Am J Hum Genet. 1999; 65: 1161-1169Abstract Full Text Full Text PDF PubMed Scopus (235) Google Scholar) is restricted to parent-offspring triads but can consider transmission of alleles or haplotypes even in the presence of phase uncertainty and missing parental genotypes.a PDT analysis utilized 33 trios and 100 discordant sib pairs.b TRANSMIT analysis utilized 162 families. Open table in a new tab Note.— Two complementary approaches were used to look for evidence of excess transmission of DRB1 and DQB1 alleles in African American families. The PDT (Martin et al. Martin et al., 2000Martin ER Monks SA Warren LL Kaplan NL A test for linkage and association in general pedigrees: the pedigree disequilibrium test.Am J Hum Genet. 2000; 67: 146-154Abstract Full Text Full Text PDF PubMed Scopus (518) Google Scholar, Martin et al., 2001Martin ER Bass MP Kaplan NL Correcting for a potential bias in the pedigree disequilibrium test.Am J Hum Genet. 2001; 68: 1065-1067Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar; Pedigree Disequilibrium Test Analysis Program Web site) can utilize both discordant sib pairs and nuclear families from extended pedigrees. TRANSMIT (Clayton and Jones Clayton and Jones, 1999Clayton D Jones H Transmission/disequilibrium tests for extended marker haplotypes.Am J Hum Genet. 1999; 65: 1161-1169Abstract Full Text Full Text PDF PubMed Scopus (235) Google Scholar) is restricted to parent-offspring triads but can consider transmission of alleles or haplotypes even in the presence of phase uncertainty and missing parental genotypes. In summary, we report a selective association with DRB1*15, indicating a primary role for the DRB1 locus in MS independent of DQB1. An approach similar to that reported here was previously used to identify the primary class II region gene associated with the sleep disorder narcolepsy, another HLA-DR2-related disorder (Mignot et al. Mignot et al., 2001Mignot E Lin L Rogers W Honda Y Qiu X Lin X Okun M Hohjoh H Miki T Hsu S Leffell M Grumet F Fernandez-Vina M Honda M Risch N Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups.Am J Hum Genet. 2001; 68: 686-699Abstract Full Text Full Text PDF PubMed Scopus (443) Google Scholar). In striking contrast to MS, the specific HLA association in African American narcoleptic patients was with DQB1*0602 rather than with DRB1*1501. A primary role for DRB1*15 in susceptibility to MS is consistent with a pathogenesis model that involves a T cell–mediated autoimmune response against the 85-99 peptide of myelin basic protein (MBP) (Allegretta et al. Allegretta et al., 1990Allegretta M Nicklas JA Sriram S Albertini RJ T cells responsive to myelin basic protein in patients with multiple sclerosis.Science. 1990; 247: 718-721Crossref PubMed Scopus (406) Google Scholar; Pette et al. Pette et al., 1990Pette M Fujita K K