Multiple Sclerosis, Susceptibility To
A number sign (#) is used with this entry because of evidence that susceptibility to multiple sclerosis-1 (MS1) is associated with variation in certain HLA genes on chromosome 6p21, including HLA-A (142800), HLA-DRB1 (142857), HLA-DQB1 (604305), HLA-DRA (142860), on chromosome 6p21.3.
An HLA-DRB1*1501-DQB1*0602 haplotype (HLA-DR15) has been repeatedly demonstrated in high-risk populations of northern European descent.
Additional MS susceptibility loci include MS2 (612594) on chromosome 10p15, MS3 (612595) on chromosome 5p13, MS4 (612596) on chromosome 1p36, and MS5 (614810), influenced by variation in the TNFRSF1A gene (191190) on chromosome 12p13.
Svejgaard (2008) provided a detailed review of the immunogenetics of multiple sclerosis, with special emphasis on the association with HLA molecules.
InheritanceFamilial aggregation in this disease is not strong; however, in a series of 91 cases, Bas (1964) found 3 instances of affected mother and daughter. From an extensive review, McAlpine (1965) concluded that the risk to a first-degree relative of a patient with multiple sclerosis is at least 15 times that for a member of the general population but that no definite genetic pattern is discernible. MacKay and Myrianthopoulos (1966) found that concordance is slightly higher in monozygotic than in dizygotic twins and that multiple sclerosis is about 20 times more frequent among relatives of probands than in the general population. The frequency declined as the relationship to the proband became more remote. They concluded that the family data were consistent with autosomal recessive inheritance with reduced penetrance but that exogenous factors must be very strong.
Ebers et al. (1986) surveyed 10 multiple sclerosis clinics across Canada and found 27 monozygotic and 43 dizygotic twin pairs with multiple sclerosis in at least 1 of each pair. Seven (25.9%) of the monozygotic pairs and 1 (2.3%) of the dizygotic pairs were concordant for multiple sclerosis. The concordance rate for nontwin sibs was 1.9%. Kinnunen et al. (1987) also reported a nationwide series of twins. The higher concordance rate in monozygotic twins despite the low recurrence risk in families is consistent with a polygenic model (Ebers, 1988). The situation may be the same as that for Hodgkin disease; see 236000.
Ebers et al. (1995) concluded that familial aggregation in MS is genetically determined. They could detect no effect of shared environment in a study of adopted index cases and MS cases with adopted relatives. Waksman (1995), in a commentary, reviewed evidence suggesting that environmental factors are not completely excluded.
Sadovnick et al. (1996) studied familial aggregation of multiple sclerosis in a sample of 16,000 multiple sclerosis cases in Canada. The age-adjusted multiple sclerosis rate in half sibs of index cases was 1.32%, compared with 3.46% in full sibs. There were similar risks in half sibs raised together and those raised apart. The risk for maternal and paternal half sibs was similar. They quoted previous studies which indicated a 300-fold increase of risk for monozygotic twins of index cases (Ebers et al., 1986) and 20- to 40-fold increase for biologic first-degree relatives (Mumford et al., 1994). Together, these studies suggested that familial aggregation in multiple sclerosis is genetic. However, since most monozygotic twins remain discordant, nongenetic risk factors are clearly important.
Steinman (1996) stated that the concordance rate among monozygotic twins is 30%, a 10-fold increase over that in dizygotic twins or first-degree relatives. The higher incidence rate among monozygotic twins empasizes the importance of genetic factors, but the discordance rate of 70% among identical twins illuminates the role of nongenetic factors on disease penetrance.
From a review of genomic screens, Dyment et al. (1997) concluded that a number of genes with interacting effects are likely and that no single region has a major influence on familial risk. An HLA haplotype associated with the disease has been identified, but HLA contributes only modestly to overall susceptibility.
The Multiple Sclerosis Genetics Group (1998) reported demographic and clinical characteristics of 89 multiplex families. The mean difference in age of onset between probands and affected sibs was 8.87 years. There was a higher concordance rate among sister pairs than among brother pairs, but there was no difference in affection rate between sons and daughters of either affected mothers or affected fathers.
Chataway et al. (1998) reported a follow-up on the studies in progress in the United Kingdom for a systematic genome screen to determine the genetic basis of MS. They stated that a gene of major effect had been excluded from 95% of the genome and one with a moderate role from 65%. The results to date suggested that multiple sclerosis depends on independent or epistatic effects of several genes, each with small individual effects, rather than a very few genes of major biologic importance.
Sadovnick et al. (1999) provided familial risk data in a practical format for use during genetic counseling for MS.
Noseworthy et al. (2000) included genetic factors in an extensive review of multiple sclerosis.
Marrosu et al. (2002) examined the recurrence risk in sibs of 901 Sardinian MS patients and factors influencing risk, such as patient and sib sex, patient age at onset, sib birth cohort, and presence of affected relatives other than sibs. To evaluate the presence of distant familial relationships among patients, extended pedigrees were traced for all patients who were born in 1 Sardinian village. The authors found that 23 brothers and 36 sisters of the 2,971 sibs were affected with MS. Recurrence risk was greater in sibs of index patients with onset age less than 30 years (increased risk 2.33 times) and with a relative with MS other than a sib or parent (increased risk 2.90 times). Pedigree analysis of patients from the 1 village showed that all 11 patients descended from 3 pairs of ancestors, whereas no cases occurred in the remaining 2,346 inhabitants. In descendants from the 3 couples, MS prevalence was dramatically greater than the regional average and 1.5 times greater than that observed in sibs of affected cases.
In a longitudinal population-based study of twins with MS in Canada, Willer et al. (2003) analyzed 370 index cases from 354 pairs and obtained a probandwise concordance rate of 25.3% in monozygotic twin pairs, 5.4% in dizygotic pairs, and 2.9% for their nontwin sibs. The excess concordance in monozygotes was derived primarily from female pairs with a probandwise concordance rate of 34% for female monozygotic pairs compared to 3.8% for female dizygotic pairs. Willer et al. (2003) did not demonstrate a monozygotic/dizygotic difference in males, but they noted that the sample size was small.
Ristori et al. (2006) analyzed data from 216 Italian twin pairs in which at least 1 twin had MS, including 198 pairs from continental Italy and 18 pairs from Sardinia. These regions have estimated disease prevalences of 61.1 and 147.1 per 100,000 individuals, respectively. They found a twinning rate of 0.62% among MS patients, which was significantly less than the twinning rate of the general population. In continental Italy, concordance for MS was 14.5% and 4.0% for mono- and dizygotic twins, respectively. In Sardinia, concordance for MS was 22.2% for monozygotic twins and zero for dizygotic twins. Results from a questionnaire on nonheritable risk factors given to a subset of patients suggested a link to infection. Ristori et al. (2006) concluded that nonheritable variables play a role in the development of MS in Mediterranean regions, and they suggested a role for protective factors in particular.
In a study of 79 MS-discordant monozygotic twin pairs, Islam et al. (2007) found that childhood sun exposure offered protection against disease development. Depending on sun exposure, the odds ratio ranged from 0.25 to 0.57. The authors concluded that early sun exposure is protective against MS, independent of genetic susceptibility. The effect was significant only for female twins; however, there were only 13 male twin pairs. Islam et al. (2007) hypothesized that exposure to ultraviolet radiation may induce immunosuppression via several mechanisms.
In a cohort of 807 avuncular MS families with 938 affected aunt/uncle-niece/nephew pairs ascertained from a longitudinal, population-based Canadian database, Herrera et al. (2008) observed an increased number of avuncular pairs connected through unaffected mothers compared to unaffected fathers (p = 0.008). To restrict confounders introduced by families with multiple pairs, the overall number of maternal and paternal families were compared, and the comparison revealed a significantly higher number of maternal families (p = 0.038). The findings indicated a maternal parent-of-origin effect in susceptibility to MS.
Baranzini et al. (2010) reported the genome sequences of one MS-discordant monozygotic twin pair, and mRNA transcriptome and epigenome sequences of CD4+ lymphocytes from 3 MS-discordant, monozygotic twin pairs. No reproducible differences were detected between cotwins among approximately 3.6 million SNPs or approximately 0.2 million insertion-deletion polymorphisms. Nor were any reproducible differences observed between sibs of the 3 twin pairs in HLA haplotypes, confirmed MS susceptibility SNPs, copy number variations, mRNA and genomic SNP and insertion-deletion genotypes, or the expression of approximately 19,000 genes in CD4+ T cells. Only 2 to 176 differences in the methylation of approximately 2 million CpG dinucleotides were detected between sibs of the 3 twin pairs, in contrast to approximately 800 methylation differences between T cells of unrelated individuals and several thousand differences between tissues or between normal and cancerous tissues. In the first systematic effort to estimate sequence variation among monozygotic cotwins, Baranzini et al. (2010) did not find evidence for genetic, epigenetic, or transcriptome differences that explained disease discordance. Baranzini et al. (2010) noted that these were the first female, twin, and autoimmune disease individual genome sequences reported.
Clinical ManagementIn patients with multiple sclerosis, treatment with interferon-beta reduces clinical exacerbations and disease burden via multiple immunomodulatory actions, including augmentation of apoptosis. In 10 of 18 patients with MS who responded to interferon-beta therapy, Sharief and Semra (2002) found a significant decline in cellular survivin expression after 6 and 12 months. Specifically, T-cell susceptibility to etoposide-induced apoptosis was increased in these patients, findings that were confirmed by in vitro experiments. These results suggested at least 1 mechanism by which interferon-beta treatment is effective in some patients with MS.
Miller et al. (2003) and Ghosh et al. (2003) reported clinical trials of natalizumab, a recombinant anticlonal antibody against alpha-4-integrins (192975), for the treatment of multiple sclerosis and Crohn disease (see 266600), respectively. Miller et al. (2003) reported that a group of patients with multiple sclerosis who received monthly injections of natalizumab had significantly fewer new inflammatory central nervous system lesions than the placebo group (a reduction of approximately 90%) and had approximately half as many clinical relapses. Ghosh et al. (2003) reported that patients with Crohn disease also had a favorable response to natalizumab, with remission rates that were approximately twice as high in patients who received 2 injections of the antibody as in patients from the placebo group. The rate of adverse events did not differ significantly between the natalizumab and placebo groups in either trial. Von Andrian and Engelhardt (2003) stated that natalizumab probably has therapeutic effects because it blocks the ability of alpha-4/beta-1 and alpha-4/beta-7 to bind to their respective endothelial counter-receptors, VCAM1 (192225) and MADCAM1 (102670). In both disorders, lesions result from autoimmune responses involving activated lymphocytes and monocytes. Alpha-4-integrin is expressed on the surface of these cells and plays an integral part in their adhesion to the vascular endothelium and migration into the parenchyma.
Williams and Johnson (2004) reported that 3 (8.6%) of 35 consecutive patients with neuroretinitis had previously been diagnosed with MS, suggesting that neuroretinitis is a late finding in MS rather than an initial presenting event. All 3 patients had been treated with interferon-beta before or concurrently with the development of neuroretinitis, which raised the question of whether interferon-beta might have been a causative agent of neuroretinitis in the patients.
Hoffmann et al. (2008) used high-resolution HLA class I and II typing to identify 2 HLA class II alleles associated with the development of antibodies to interferon-B in the treatment of multiple sclerosis. In 2 independent continuous and binary-trait association studies, HLA-DRB1*0401 and HLA-DRB1*0408 (odds ratio: 5.15), but not other HLA alleles, were strongly associated with the development of binding and neutralizing antibodies to interferon-B. The associated HLA-DRB1*04 alleles differ from nonassociated HLA-DRB1*04 alleles by a glycine-to-valine substitution in position 86 of the epitope-binding alpha-helix of the HLA class II molecule. The peptide-binding motif of HLA-DRB1*0401 and *0408 might promote binding and presentation of an immunogenic peptide, which may eventually break T cell tolerance and facilitate antibody development to interferon-beta. In summary, Hoffmann et al. (2008) identified genetic factors determining the immunogenicity of interferon-beta, a protein-based disease-modifying agent for the treatment of MS.
Kumpfel et al. (2008) identified 20 patients with MS who carried a heterozygous variant (R92Q) in the TNFRSF1A gene (191190) and had clinical features consistent with late-onset of the tumor necrosis factor receptor 1-associated periodic syndrome (TRAPS; 142680), including myalgias, arthralgias, headache, fatigue, and skin rashes. Most of these patients experienced severe side effects during immunomodulatory therapy for MS. Kumpfel et al. (2008) concluded that patients with coexistence of MS and features of TRAPS should be carefully observed during treatment.
Comabella et al. (2009) performed a genomewide association study in 53 MS patients who responded to beta-interferon treatment and 53 nonresponders in an attempt to identify a genetic basis influencing the variable response observed in patients. The discovery study and a replication study in 49 additional responders and 45 additional nonresponders pointed to 18 SNPs in various genes that showed a possible association (uncorrected p values of less than 0.05). The findings indicated that response to beta-interferon is a complex and polygenic trait.
Hla and Brinkmann (2011) and Soliven et al. (2011) provided reviews of the neurobiology of sphingosine 1-phosphate (S1P) signaling in the CNS via the S1P receptors (S1PRs), of which there are 5 subtypes (see, e.g., S1PR1; 601974), and discussed the benefit of the S1PR modulator, fingolimod (FTY720), in the treatment of MS. FTY720 was approved in 2010 as the first oral treatment for relapsing MS in the U.S. One effect of FTY720 is to downmodulate S1PR1 to retain circulating naive and central memory T and B lymphocytes in lymph nodes, while sparing effector memory T cells. The result is to reduce the infiltration of autoreactive lymphocytes into the CNS, causing a slowing of the disease process (Hla and Brinkmann, 2011). In addition, S1PR1 is expressed in oligodendrocytes, astrocytes, neurons, and microglia, where it may modulate cell survival, process dynamics, migration, differentiation, activation, and crosstalk. The presence of S1PRs on multiple cell lines in the CNS may represent a mechanism by which FTY720 may contribute to observed neurologic benefit in patients with MS via neuroprotective and regenerative effects (Soliven et al., 2011).
Population GeneticsSteinman (1996) stated that multiple sclerosis is the most common autoimmune disease involving the nervous system and that approximately 250,000 individuals in the United States have MS.
Pugliatti et al. (2002) demonstrated a hotspot of MS in the southwestern part of Sassari province in Sardinia, bordering with the commune of Macomer, where MS was once hypothesized as having occurred as an epidemic. These areas of MS clustering comprised the Common Logudorese linguistic domain. The Catalan area, which is linguistically and genetically distant from the remaining Sardinian domains, did not show such high estimates.
MappingBell and Lathrop (1996) reviewed the work on linkage analysis in multiple sclerosis.
MS1 Locus Associated with HLA on Chromosome 6p21.3
Terasaki et al. (1976) described a high frequency of a B-lymphocyte antigen (group 4) in multiple sclerosis. Associations with HLA-A3, HLA-B7, and HLA-Dw2 have been demonstrated also. The association with Dw2 seems to be especially strong and probably indicates an immune-response mechanism.
Zipp et al. (1995) compared the production of lymphotoxin (tumor necrosis factor-beta (TNFB; 153440) and tumor necrosis factor-alpha (TNFA; 191160)) by T-cell lines isolated from multiple sclerosis patients in normal controls. There was greater production in those lines derived from HLA-DR2-positive donors than from those that were HLA-DR2-negative. Although both lymphotoxin and tumor necrosis factor-alpha are encoded within the HLA region, there was no significant association of cytokine production with individual lymphotoxin or TNF alleles. The authors suggested that the association of multiple sclerosis with HLA-DR2 results from a propensity of T cells to produce increased amounts of lymphotoxic TNF, controlled by a polymorphic gene in this region.
In a linkage analysis of 72 pedigrees, Tiwari et al. (1980) found evidence of linkage between HLA and a hypothesized multiple sclerosis susceptibility gene (MSSG) for both dominant and recessive models of inheritance and for a wide range of penetrance values. They suggested that the MSSG is located 15-20 recombination units from HLA, probably on the B-D side. The analysis showed no evidence of linkage heterogeneity, and the lod scores appeared not to be inflated artificially by the association of multiple sclerosis with HLA-B7. In linkage studies with HLA, Haile et al. (1980) assumed a dominant model of inheritance. With a penetrance value of 0.05, a maximal lod score of 2.411 was obtained for recombination fraction of 0.10. With high penetrance values, lod scores did not support linkage. Francis et al. (1987) did a study of familial MS: 10 affected sib pairs and 4 instances of affected parent and offspring, together with 1 family with 3 affected sibs and another with 2 affected sibs and an affected parent. They concluded that an MS susceptibility gene exists in the HLA complex in linkage disequilibrium with HLA-D.
In a 2-stage genome screen, Sawcer et al. (1996) found 2 principal regions of linkage with multiple sclerosis: 17q22 and the HLA region on 6p21. The results were considered compatible with genetic models involving epistatic interaction between these and several additional genes. A similar complete genomic screen by the Multiple Sclerosis Genetics Group (1996) yielded results suggesting a multifactorial etiology, including both environmental and multiple genetic factors of moderate effect. The results supported a role for the MHC region on 6p.
Ebers et al. (1996) found maximum lod scores (MLS) greater than 1 for MS at 5 loci on chromosomes 2, 3, 5, 11, and X. Two additional datasets containing 44 and 78 sib pairs respectively, were used to further evaluate the HLA region on 6p21 and a locus on chromosome 5 with an MLS of 4.24. Markers within 6p21 gave an MLS of 0.65. However, D6S461, just outside the HLA region, showed significant evidence for linkage disequilibrium by the transmission disequilibrium test (TDT), in all 3 datasets, suggesting to the investigators a modest susceptibility locus in this region. The chromosome 5p results from 3 datasets (222 sib pairs) yielded a multipoint MLS of 1.6. Ebers et al. (1996) concluded that the results support the genetic epidemiologic evidence that several genes interact epistatically to determine heritable susceptibility.
In a collaborative study, Haines et al. (1998) studied a data set of 98 multiplex MS families to test for an association to the HLA-DR2 allele in familial MS and to determine if genetic linkage to the major histocompatibility complex (MHC) was due solely to such an association. Three highly polymorphic markers (HLA-DR, D6S273, and TNF-beta) in the MHC demonstrated strong genetic linkage (parametric lod scores of 4.60, 2.20, and 1.24, respectively) and a specific association with the HLA-DR2 allele was confirmed; the transmission/disequilibrium test (TDT) yielded a P value of less than 0.001. Stratifying the results by HLA-DR2 status showed that the linkage results were limited to families segregating HLA-DR2 alleles. These results demonstrated that genetic linkage to the MHC can be explained by the HLA-DR2 allelic association. They also indicated that sporadic and familial MS share a common genetic susceptibility. In addition, preliminary calculations suggested that the MHC explains between 17% and 62% of the genetic etiology of MS. This heterogeneity is also supported by the minority of families showing no linkage or association with loci within the MHC. In a study of the Sardinian population, Marrosu et al. (1998) tested the role of other class II HLA loci in MS predisposition.
Fernandez-Arquero et al. (1999) found a significant correlation between a TNFA-376 promoter polymorphism with susceptibility to multiple sclerosis in a study of 238 patients and 324 controls. This association was independent of HLA class II association and synergistically increased risk in the presence of HLA-DRB1*1501. In a follow-up case-control study of 241 Spanish patients with MS, Martinez et al. (2004) confirmed an association between MS and the TNFA-376 polymorphism. Noting that another study (Weinshenker et al., 2001) had failed to replicate the findings in a mostly northern European population, Martinez et al. (2004) concluded that the positive association is specific to the Spanish white population or that only studies in this population have sufficient power because of the higher frequency of the TNFA-376 allele.
Ligers et al. (2001) assessed the importance of the HLA-DR locus to multiple sclerosis susceptibility in 542 sib pairs with MS and in their families. By genotyping 1,978 individuals for HLA-DRB1 (142857) alleles, they confirmed the well-established association of MS with HLA-DRB1*15 (HLA-DRB1*1501 and HLA-DRB5*0101, 604776), by the transmission/disequilibrium test. They obtained significant evidence of linkage throughout the whole dataset (mlod = 4.09; 59.9% sharing). Surprisingly, similar sharing was also observed in 58 families in which both parents lacked the DRB1*15 allele (mlod = 1.56; 62.7% sharing; p = 0.0081). The findings suggested that the notion that HLA-DRB1*15 is the sole MHC determinant of susceptibility in northern European populations with MS may be incorrect. The possibility remained that the association of MS with HLA-DRB1*15 is due to linkage disequilibrium with a nearby locus and/or to the presence of disease-influencing allele(s) in DRB1*15-negative haplotypes.
Lang et al. (2002) examined the association of MS with HLA-DRB1*1501 and -DRB5*0101 polymorphisms by determining the antigen-recognition profile of an MS patient with a relapsing-remitting disease course. A T-cell receptor (TCR) from the patient recognized both DRB1*1501-restricted myelin basic protein (MBP; 159430) (residues 85 to 99) and DRB5*0101-restricted Epstein-Barr virus DNA polymerase peptide. The crystal structure of both DRB-antigen complexes revealed a marked degree of structural equivalence at the surface presented for TCR recognition, with 4 identical TCR-peptide contacts. Lang et al. (2002) concluded that these similarities support the concept of molecular mimicry (in structural terms, a similarity of charge distribution) involving HLA molecules and suggested that these structural details may explain the preponderance of MHC class II associations in HLA-associated diseases. They noted the findings of Madsen et al. (1999) with transgenic mice, which also showed that MBP(85 to 99) associated with HLA-DRB1*1501 was involved in the development of an MS-like disease.
Models of disease susceptibility in MS often assume a dominant action for the HLA-DRB1*1501 (see 142857) allele and its associated haplotype, DRB1*1501-DQB1*0602, also known as DR2. Barcellos et al. (2003) found a dosage effect of HLA-DR2 haplotypes on MS susceptibility. Two copies of a susceptibility haplotype further increased disease risk. They also reported that DR2 haplotypes modify disease expression. There was a paucity of benign MS and an increase of severe MS in individuals homozygous for DR2.
Mattila et al. (2001) genotyped 97 patients with MS and 100 healthy controls and found an association between the pp polymorphism in the ESR1 (133430) gene on chromosome 6q25 in combination with the previously described association of HLA-DR2 in women with MS (odds ratio for MS in women with both ESR1pp and HLA-DR2 was 19.4 vs 5.1 with DR2 alone).
Marrosu et al. (2001) scanned an 11.4-Mb region encompassing the whole HLA complex on chromosome 6p21.3 for MS association in the founder population of Sardinia. Using 19 microsatellite markers, single-nucleotide polymorphisms (SNPs) within 12 candidate genes, and the extended transmission disequilibrium test (ETDT), a peak of association represented by the 3 adjacent DRB1, -DQA1, and -DQB1 loci was detected in the class II region. Two additional less significant areas of association were detected, respectively, in the centromeric side of the class II region at the DPB1 locus and, telomeric of the classically defined class I loci, at the D6S1683 microsatellite. Conditional ETDT analysis indicated that these regions of association could be independent of each other. Within the main peak of association, DRB1 and DQB1 contributed to the disease association independently of each other, whereas DQA1 had no detectable primary genetic effects. Five DQB1-DRB1 haplotypes positively associated with MS in Sardinia, which consistently included all the haplotypes previously found associated with MS in the various human populations. The authors concluded that their results are consistent with a multilocus model of the MHC-encoded susceptibility to MS.
In 30 patients with relapsing-remitting MS, which the authors termed 'benign,' and 25 patients with secondary-progressive MS, which the authors termed 'malignant,' from a region in northeast Italy, Perini et al. (2001) found a positive association between the HLA-DR13 haplotype (particularly the DRB1*1302 allele) and 'benign' MS. The DR13 haplotype was detected in 40% of patients with 'benign' MS, in 4% with 'malignant' MS, and in 16% of normal controls.
Association of MS with the HLA-DRB1*1501-DQB1*0602 haplotype has been repeatedly demonstrated in high-risk (northern European) populations. African populations are characterized by greater haplotypic diversity and distinct patterns of linkage disequilibrium compared with northern Europeans. To better localize the HLA gene responsible for MS susceptibility, Oksenberg et al. (2004) performed case-control and family-based association studies for the DRB1 and DQB1 loci in a large and well-characterized African American dataset. 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 was 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.
Genetic susceptibility to multiple sclerosis is associated with genes of the major histocompatibility complex (MHC), particularly HLA-DRB1 and HLA-DQB1. To clarify whether HLA-DRB1 itself, nearby genes in the region encoding the MHC, or combinations of these loci underlie susceptibility to multiple sclerosis, Lincoln et al. (2005) genotyped 1,185 Canadian and Finnish families with multiple sclerosis with a high-density SNP panel spanning the genes encoding the MHC and flanking genomic regions. Strong associations in Canadian and Finnish samples were observed with blocks in the HLA-II genomic region, but the strongest association was with HLA-DRB1. Conditioning on either HLA-DRB1 or the most significant HLA class II haplotype block found no additional block or SNP association independent of the HLA class II genomic region. This study therefore indicated that MHC-associated susceptibility to multiple sclerosis is determined by HLA class II alleles, their interactions, and closely neighboring variants.
Dyment et al. (2004) reported a multistage genome scan of 552 sib pairs from 442 MS families. Only markers at chromosome 6p showed significant evidence for linkage (MLOD = 4.40), while other regions were only suggestive. The replication analysis involving all 552 affected sib pairs confirmed suggestive evidence for 5 locations, namely, 2q27, 5p15, 18p11, 9q21, and 1p31. The overall excess allele sharing observed for the entire sample was due to increased allele sharing within the DRB1*15 negative subgroup alone. The authors concluded that their observations supported a model of genetic heterogeneity between HLA and other genetic loci.
Gregersen et al. (2006) reported that the MHC HLA-DR2 haplotype comprised of DRB1*1501 (DR2b) and DRB5*0101 (DR2a), which predisposes to multiple sclerosis, shows more extensive linkage disequilibrium than other common Caucasian HLA haplotypes in the DR region and thus seems likely to have been maintained through positive selection. Characterization of 2 multiple sclerosis-associated HLA-DR alleles at separate loci by a functional assay in humanized mice indicates that the linkage disequilibrium between the 2 alleles may be due to a functional epistatic interaction, whereby 1 allele modifies the T-cell response activated by the second allele through activation-induced cell death. This functional epistasis is associated with a milder form of multiple sclerosis-like disease. Gregersen et al. (2006) suggested that such epistatic interaction might prove to be an important general mechanism for modifying exuberant immune responses that are deleterious to the host and could also help to explain the strong linkage disequilibrium in this and perhaps other HLA haplotypes.
The International Multiple Sclerosis Genetics Consortium (2007) found evidence that variation in the HLA-C gene (142840) influences susceptibility to MS independent of the HLA-DRB1 gene. Using a combination of microsatellite, SNP, and HLA typing in a family-based and case-control cohort beginning with a sample of 1,201 MS patients, the authors analyzed 264 patients without the common DRB1*1501, DRB1*03, and DRB1*0103 alleles. Significant association was found with the HLA-C locus (p = 5.9 x 10(-5)). Specifically, the HLA-C*05 allele was underrepresented in patients compared to controls (p = 3.3 x 10(-5)), suggesting a protective effect.
In a multistage genomewide association study involving a total of 1,540 multiple sclerosis family trios, 2,322 case subjects, and 5,418 control subjects, the International Multiple Sclerosis Genetics Consortium (2007) used the HLA-DRA (142860) A/G SNP rs3135388 as a proxy for the DRB1*1501 allele (complete concordance between the rs3135388 A allele and DRB1*1501 was found in 2,730 of 2,757 subjects for whom data were available) and confirmed unequivocally that the HLA-DRA locus was associated with MS (p = 8.94 X 10(-81); OR, 1.99).
Baranzini et al. (2009) conducted a genomewide association study in 978 well-characterized individuals with MS and 883 group-matched controls. The authors compared allele frequencies and assessed genotypic influences on susceptibility, age of onset, disease severity, as well as brain lesion load and normalized brain volume from MRI exams. Top SNPs were located in the MHC class-II subregion likely reflecting linkage disequilibrium with the HLA-DRB1*1501 allele. Logistic regression analysis adjusting for gender, study site, and DRB1*1501 suggested an independent association in the HLA-class I region localized around TRIM26 (600830), TRIM15, and TRIM10 (605701).
In a collaborative GWAS involving 9,772 cases of European descent collected by 23 research groups working in 15 different countries, the International Multiple Sclerosis Genetics Consortium and Wellcome Trust Case Control Consortium 2 (2011) replicated almost all of the previously suggested associations and identified at least a further 29 novel susceptibility loci for multiple sclerosis. Within the MHC the International Multiple Sclerosis Genetics Consortium and Wellcome Trust Case Control Consortium 2 (2011) refined the identity of the HLA-DRB1 risk alleles as DRB1*1501 (142857.0002) and DRB1*1303, and confirmed that variation in the HLA-A gene (142800) underlies the independent protective effect attributable to the class I region. Immunologically relevant genes were significantly overrepresented among those mapping close to the identified loci and particularly implicated T helper cell differentiation in the pathogenesis of multiple sclerosis.
Disanto et al. (2011) found that 64 (24%) of 266 children with an initial attack of demyelination (acquired demyelinating syndrome, ADS) met criteria for a diagnosis of MS during a mean follow-up of 3.2 years. ADS children with 1 or more DRB1*15 alleles were more likely to be diagnosed with MS (OR of 2.7) compared to children without this allele. The association was most apparent in those children of European descent (OR of 3.3). Presence of DRB1*15 did not convey an increased risk for MS in ADS children of non-European descent. The findings indicated that DRB1*15 alleles confer increased susceptibility to pediatric-onset MS, supporting a fundamental similarity in genetic contribution to risk of chronic MS in both pediatric- and adult-onset disease.
Associations Pending Confirmation
Mycko et al. (1998) found an increased frequency of the K469 allele of intercellular adhesion molecule-1 (ICAM1; 147840) in 79 Polish multiple sclerosis patients compared with 68 ethnically matched controls (68% vs 49%). Homozygosity for this variant was also increased (53% vs 34%).
Vandenbroeck et al. (1998) found evidence that the interferon-gamma gene (IFNG; 147570) on chromosome 12q14 is a susceptibility factor for multiple sclerosis in those Sardinians who are at low risk by virtue of their HLA status.
In a genomewide association study (GWAS) involving 1,618 MS patients and 3,413 controls, with replication in an independent set of 2,256 cases and 2,310 controls, the Australia and New Zealand Multiple Sclerosis Genetics Consortium ANZgene (2009) identified several risk-associated SNPs on chromosome 12q13-14, including rs703842 in the METTL1 gene (604466) (p = 5.4 x 10(-11)); rs10876994, p = 2.7 x 10(-10); and rs12368653, p = 1.0 x 10(-7). The region encompassed 17 putative genes. Gandhi et al. (2010) determined that the MS-associated SNP rs703842 identified by the Australia and New Zealand Multiple Sclerosis Genetics Consortium ANZgene (2009) was also associated with expression of the FAM119B gene (615258), the MS susceptibility allele being the low-expressor of FAM119B.
Schrijver et al. (1999) found that patients with multiple sclerosis who were carriers of the IL1RN*2 allele (see 147679) and noncarriers of the IL1B*2 allele (see 147720) had a higher rate of progression than those with other allele combinations.
In 3 of 4 independent case-control studies, Jacobsen et al. (2000) demonstrated an association of a SNP in the PTPRC gene (151460) with MS. Furthermore, they found that the PTPRC mutation was linked to and associated with the disease in 3 MS nuclear families. However, studies by Vorechovsky et al. (2001) Barcellos et al. (2001), Cocco et al. (2004), and Szvetko et al. (2009) found no association between the PTPRC SNP and multiple sclerosis.
Dyment et al. (2001) analyzed and performed genotyping in 219 sib pairs assembled in connection with 4 published genome screens that had identified a number of markers with increased sharing in MS families but which did not reach statistical significance.
Dyment et al. (2001) used 105 markers previously identified as showing increased sharing in genome screens of Canadian, British, Finnish, and American MS families, but which did not reach statistical significance for linkage, in a genotype analysis of a Canadian sample of 219 sibs pairs. None of the markers met the criteria for significant linkage. Markers located at 5p14 and 17q22 were analyzed in a total of 333 sib pairs and attained maximum lod scores of 2.27 and 1.14, respectively. The known HLA-DRB1 association with MS was confirmed (p less than 0.0001). A significant transmission disequilibrium was also observed for D17S789 at 17q22 (p = 0.0015). The authors noted that the study highlighted the difficulty of searching for genes with only mild to moderate effects on susceptibility, although large effects of specific loci may still be present in individual families. They suggested that progress in the genetics of this complex trait may be helped by (1) focusing on more ethnically homogeneous samples, (2) using an increased number of MS families, and (3) using transmission disequilibrium analysis in candidate regions rather than the affected relative pair linkage analysis.
Xu et al. (2001) investigated 27 microsatellite markers from 8 chromosomal regions syntenic to loci of importance for experimental autoimmune diseases in the rat in 74 Swedish MS families. Possible linkage was observed with markers in the 7q35 (highest NPL score of 1.16) and 12p13.3 (highest NPL score of 1.16) regions, which are syntenic to the rat Cia3 (collagen-induced arthritis) and Oia2 (oil-induced arthritis) loci, respectively. Both regions overlapped with areas showing evidence for linkage in previous MS genomic screens.
The prevalence of MS in Sardinia (approximately 140 per 100,000) is significantly higher than in surrounding Mediterranean countries, suggesting that the isolated growth of this population has concentrated genetic susceptibility factors for the disease. Coraddu et al. (2001) performed a genomewide screen for linkage in 49 Sardinian multiplex families (46 sib pairs and 3 sib trios) using 327 markers. Nonparametric multipoint linkage analysis revealed suggestive linkage (MLS greater than 1.8) to chromosome regions 1q31, 10q23, and 11p15. Coraddu et al. (2001) concluded that the individual effects of genes determining susceptibility to MS are modest.
Pericak-Vance et al. (2001) reviewed linkage studies in multiple sclerosis. Genomic screens had suggested over 50 regions that might harbor MS susceptibility genes, but there had been little agreement between studies. The one region suggested by all 4 screens resided within chromosome 19q13. They examined this region in detail in an expanded dataset of MS families from the United States. Genetic linkage and association were tested with multiple markers in this region using both parametric and nonparametric analyses. Additional support for an MS susceptibility locus was observed, primarily in families with the MS-associated HLA-DR2 allele. While consistent, this effect appeared to be modest, probably representing no more than 10% of the overall genetic effect in MS.
Haines et al. (2002) studied a population of 266 individuals with MS belonging to 98 multiplex families. Their analysis continued to support linkage to chromosomes 6p21, 6q27, and 19q13 with lod scores higher than 3.0, and suggested that regions on chromosomes 12q23-q24 and 16p13 may also harbor susceptibility loci for MS. Analysis taking into account the known HLA-DR2 association identified additional potential linkage regions on chromosomes 7q21-22 and 13q33-34.
Vitale et al. (2002) identified a pedigree of Pennsylvania Dutch extraction in which MS segregated with an autosomal dominant inheritance pattern. Eighteen individuals, of whom 7 were affected, were serotyped for HLA class I and II and also analyzed by a genomewide screen for linkage analysis. There was suggestive linkage to markers on 12p12 with a maximum multipoint lod score of 2.71, conditional on the presence of HLA-DR15*DQ6. Contingency table analysis showed that all MS affected individuals had both the DR15*DQ6 allele and the 12p12 haplotype, whereas the unaffected individuals had either 1 or neither of these markers (P = 0.00011). The authors concluded that both HLA-DR15*DQ6 and a novel locus on chromosome 12p12 may be necessary for development of MS in this family.
He et al. (2002) studied a genetically isolated population in the Overkalix community of northern Sweden, which demonstrates a high incidence of MS. This ethnically homogeneous population was probably founded in the 17th century by a few couples. A genealogic analysis established that 19 of the MS patients originated from a single common ancestral couple. Five affected individuals from 4 nuclear families were selected for genomewide genotyping with 390 microsatellite markers. Seven shared haplotypes in 6 different chromosomal regions were identified. Only 1 of the suggested haplotypes was confirmed to be identical-by-descent after analysis of additional markers in 15 MS patients, and the identified region at 17p11 consisted of 4 markers spanning 7 cM. A significant excess of transmission of alleles to affected individuals (p less than 0.05) was observed for 3 of the markers by TDT. No increased sharing of haplotypes was observed for the HLA-DR and -DQ loci. The results suggested the presence of a susceptibility gene for MS in chromosome 17p11.
Saarela et al. (2002) carried out linkage analyses in 22 Finnish multiplex MS families originating from a regional subisolate that showed an exceptionally high prevalence of MS. The authors identified a 4-cM region flanked by the markers D17S1792 and ATA43A10 in 17 of 22 families. Using the combined power of linkage, association, and shared haplotype analyses, the authors restricted the MS locus on chromosome 17q to a region corresponding to a physical interval of 2.5 Mb.
By genomewide analysis of 779 Finnish MS patients and 1,165 controls, including those from an isolate in Southern Ostrobothnia, Jakkula et al. (2010) found an association between multiple sclerosis and the A allele of rs744166 in the STAT3 gene (102582) on 17q21; the A allele was protective. The findings were replicated in a total of 3,859 cases and 9,110 controls from various populations, including Norway, Denmark, the Netherlands, Switzerland, and the United States, yielding an overall p value of 2.75 x 10(-10) and an odds ratio of 0.87 (CI, 0.83-0.91). To validate the findings of Jakkula et al. (2010), Lill et al. (2012) performed a genetic association study of 2 SNPs in the STAT3 gene in a German case-control sample of 2,932 MS patients and 2,972 controls. There was a nominally significant association between the G allele of rs744166 and MS (OR of 1.09, p = 0.012), and no association with rs2293152. Lill et al. (2012) noted that rs744166 occurs in an intron and is not likely to have functional significance.
Kenealy et al. (2004) used a panel of 390 microsatellite markers for a genome screen in 245 U.S. and French multiplex families (the largest genomic screen for MS to that time). Four regions were thought to warrant further study.
Admixture mapping is a method for scanning the genome for gene variants that affect the risk for common, complex disease. The method has high statistical power to detect factors that differ markedly in frequency across human populations. Multiple sclerosis was an excellent candidate for admixture mapping because it is more prevalent in European Americans than in African Americans (Kurtzke et al., 1979, Wallin et al. (2004)). Reich et al. (2005) performed a high-powered admixture scan, focusing on 605 African American cases of multiple sclerosis and 1,043 African American controls. The individuals in their study had, on average, 21% European and 79% African ancestry. The goal was to identify genetic regions where individuals with multiple sclerosis tended to have an unusually high proportion of ancestry from either Europeans or Africans, indicative of the presence of a multiple sclerosis risk variant that differs in frequency between the ancestral populations. Reich et al. (2005) hypothesized that if there are genetic risk factors for multiple sclerosis that explain the epidemiology, they should be identifiable as regions with a high proportion of European ancestry in African Americans with multiple sclerosis compared with the average. They reported a locus on chromosome 1 that is significantly associated with multiple sclerosis. The 95% credible interval on chromosome 1 was estimated to be between 114.9 Mb and 144.7 Mb from 1pter, a region containing 68 known genes.
Among 242 patients with multiple sclerosis and 207 controls from a central Ohio population, Zhou et al. (2003) found that homozygosity for an ala57-to-val (A57V) SNP in the CD24 gene (600074) on chromosome 6q21 was associated with a 2-fold increased risk of MS in the general population, and the V57 allele was preferentially transmitted to affected individuals among familial MS cases. Most V57 homozygotes reached an expanded disability status within 5 years, whereas heterozygotes and A57 homozygotes reached this milestone in 16 and 13 years, respectively. Flow cytometric analysis demonstrated that CD24 was more highly expressed on T cells of V57 homozygous patients than A57 homozygous patients. Zhou et al. (2003) concluded that the A57V CD24 polymorphism genetically modifies susceptibility and progression of MS, perhaps by affecting the efficiency of CD24 expression. However, Goris et al. (2006) were unable to confirm the association between the A57V SNP and multiple sclerosis in a combined cohort of 1,180 cases and 1,168 unrelated and family-based controls from Belgium and the United Kingdom. Among 135 Spanish Basque patients with MS and 285 controls, Otaegui et al. (2006) found evidence for trend of association between the V56 allele and MS, but the results did not reach significance for an association study.
By fine mapping of a candidate locus at chromosome 1p13 in 1,278 trio families with MS and replication in an additional 3,341 MS patients, De Jager et al. (2009) observed a significant association between protection against MS and the G allele of rs2300747 in the CD58 gene (153420) (combined p of 1.1 x 10(-6); OR of 0.82). The protective G allele was associated with a dose-dependent increase in CD58 mRNA expression in lymphoblastic cells lines from MS patients (p = 1.1 x 10(-10)), suggesting a functional effect. De Jager et al. (2009) found that CD58 mRNA expression was higher in MS patients during clinical remission.
In a metaanalysis of genomewide association studies including 2,624 patients with MS and 7,220 controls, followed by replication in an independent set of 2,215 patients with MS and 2,116 controls, De Jager et al. (2009) identified loci for MS susceptibility on chromosome 12p13 in the TNFRSF1A gene (191190) (rs1800693; see MS5, 614810), on chromosome 16 (rs17445836) near