Tai Wai Yeo

A high-density screen for linkage in multiple sclerosis.

To provide a definitive linkage map for multiple sclerosis, we have genotyped the Illumina BeadArray linkage mapping panel (version 4) in a data set of 730 multiplex families of Northern European descent. After the application of stringent quality thresholds, data from 4,506 markers in 2,692 individuals were included in the analysis. Multipoint nonparametric linkage analysis revealed highly significant linkage in the major histocompatibility complex (MHC) on chromosome 6p21 (maximum LOD score [MLS] 11.66) and suggestive linkage on chromosomes 17q23 (MLS 2.45) and 5q33 (MLS 2.18). This set of markers achieved a mean information extraction of 79.3% across the genome, with a Mendelian inconsistency rate of only 0.002%. Stratification based on carriage of the multiple sclerosis-associated DRB1*1501 allele failed to identify any other region of linkage with genomewide significance. However, ordered-subset analysis suggested that there may be an additional locus on chromosome 19p13 that acts independent of the main MHC locus. These data illustrate the substantial increase in power that can be achieved with use of the latest tools emerging from the Human Genome Project and indicate that future attempts to systematically identify susceptibility genes for multiple sclerosis will have to involve large sample sizes and an association-based methodology.

A second major histocompatibility complex susceptibility locus for multiple sclerosis

Variation in the major histocompatibility complex (MHC) on chromosome 6p21 is known to influence susceptibility to multiple sclerosis with the strongest effect originating from the HLA‐DRB1 gene in the class II region. The possibility that other genes in the MHC independently influence susceptibility to multiple sclerosis has been suggested but remains unconfirmed.

HLA-DRB1 and multiple sclerosis in Malta

Background: By comparison with the neighboring island of Sicily, the frequency of multiple sclerosis (MS) in Malta is remarkably low. Methods: To explore whether the relative rarity of MS in Malta might be the result of lower population frequencies of major histocompatibility complex susceptibility alleles, we genotyped the HLA-DRB1 locus in 77 Maltese-born patients (97% of the prevalent unrelated native cases) and 206 Maltese controls. We made comparisons with previously published data for Sicily and other European countries. Results: The anticipated association with HLA-DRB1*15, the main susceptibility allele in most other populations, was confirmed (pc = 0.009) but, in addition, we also observed an equally strong, and apparently protective, effect of the HLA-DRB1*11 allele (pc = 0.016). In comparison with previously published data from Sicily, we found that all HLA-DRB1 risk alleles were more common in Malta, whereas HLA-DRB1*11 was slightly less common. Conclusions: The difference in prevalence seen between the neighboring islands of Malta and Sicily cannot be explained by differences in background HLA-DRB1 population allele frequencies, which if anything would predict a higher rate of disease in Malta than in Sicily.

Four single nucleotide polymorphisms from the Vitamin D Receptor Gene in UK Multiple Sclerosis

Sirs: Multiple sclerosis is a disabling disease of unknown aetiology affecting the central nervous system. Epidemiological studies indicate that pathogenesis involves both genetic and environmental factors. The close correlation between latitude and disease frequency in many parts of the world suggests a strategy for identifying environmental conditions that correlate with the frequency of multiple sclerosis. In 1974, Goldberg [4] proposed that the reduced levels of vitamin D found in higher latitudes might contribute to an increased risk of developing multiple sclerosis. This hypothesis was supported by Freedman et al. [2] who found that mortality from multiple sclerosis was negatively correlated to sunlight exposure. More recently, van der Mei et al. [11] also concluded that higher exposure to sunlight during childhood correlated inversely with the risk of multiple sclerosis. Interest in the role of vitamin D in multiple sclerosis pathogenesis was further increased by the observation that peripheral mononuclear blood cells express vitamin D receptors (VDRs) [9] and therefore that vitamin D might influence the development of autoimmunity [6]. Furthermore, vitamin D treatment has been shown to prevent the development of murine experimental autoimmune encephalomyelitis, an animal model of multiple sclerosis [5]. More recently, Meehan and DeLuca [7] demonstrated that VDR is the receptor by which vitamin D acts to suppress autoimmunity. Genetic analysis of markers in and around the VDR gene in multiple sclerosis patients has provided contradictory results. Fukazawa et al. [3] and Niino et al. [8] reported association in their study of 77 patients and 95 controls. However, this result was not reproduced by Steckley et al. [10] in a linkage study involving 236 sibling pairs from 187 families. In order to explore this candidate further, we tested the VDR gene for association with susceptibility to multiple sclerosis by genotyping four single nucleotide polymorphisms (SNPs) in 380 trio families (an affected index patient and both parents). Informed consent for genetic analysis was obtained from each family member. Each individual provided a venous blood sample from which DNA was extracted using standard methods. Genotyping was performed using the Taqman method (Applied Biosystems). Statistical analysis of each individual SNP and of the four marker haplotype was performed using the TRANSMIT program [1]. None of the individual SNP’s assays showed evidence for association with susceptibility to multiple sclerosis. Analysis of the four marker haplotypes was equally disappointing (Table 1). We have found no evidence for association of multiple sclerosis with these variants in the VDR gene. However, the variants considered, although informative, do not represent an exhaustive list and it remains possible that other untested variants are important. Similarly the sample size employed can not exclude a minor effect from the variants considered although any major effect has safely been excluded.

SELPLG and SELP single-nucleotide polymorphisms in multiple sclerosis

P-Selectin (SELP) and P-selectin glycoprotein ligand-1 (SELPLG) constitute a receptor/ligand complex involved in the recruitment of activated lymphocytes, a critical event in the pathogenesis of multiple sclerosis (MS). In order to determine whether genetic variation in these pivotal molecules influences susceptibility to MS, we genotyped 214 Italian patients compared with 220 Italian controls for three single-nucleotide polymorphisms (SNPs): SELPLG Met62Ile, SELP C-2123G and SELP Thr715Pro. No significant differences in both SELP SNPs were found between patients and controls, whereas a decreased frequency of the Met62Ile SNP was found in patients versus controls in the Italian population (P = 0.025). To confirm these preliminary findings, the Met62Ile SNP was analysed in 938 UK trio families. This SNP did not show evidence for association with susceptibility to MS in the larger UK cohort. Therefore, none of the SNPs investigated is associated with MS, although this analysis does not conclusively exclude SELPLG and SELP as genetic risk factors for MS as much variation remains untested.

Ultraconserved regions in multiple sclerosis.

In their recent publication, Bejerano et al identified 481 ‘ultraconserved elements’ in the human genome. They defined these elements as genomic segments longer than 200 bp, showing 100% sequence homology with rat and mouse, but excluding ribosomal RNA regions. Many of these elements also show extremely high levels of homology with the chicken and dog genomes. Overall, these elements are more highly conserved than protein coding regions. The mechanisms responsible for maintaining these sequences through evolution are unclear but seem likely to include profound negative selection, suggesting that these segments have important, if not vital, functions. Whether these functions will necessarily be relevant to the pathogenesis of multiple sclerosis (MS) is speculative. The high levels of homology are reflected in correspondingly low levels of variation within the general human population. Bejerano et al identified just six validated single-nucleotide polymorphisms (SNPs) lying within these ultraconserved elements (rs1538101, rs1861100, rs2056116, rs7092999, rs7143938 and rs9572903), a total of 106 767 bp of sequence. While there remains limited data on the functional effects of SNPs within ultraconserved regions, several pathogenic mutations in the region that regulates sonic hedgehog (SHH) expression are associated with preaxial polydactyly, a congential limb malformation. Given the likely biological importance of these ultraconserved regions, it is reasonable to infer that these six variants have a high prior probability of conferring functional significance; however, the nature of this altered function is currently unpredictable. We tested these six SNPs from ultraconserved elements for evidence of association with MS, in which genetic factors are implicated, but their nature and identity are largely unknown. Association was sought by typing each variant in 938 trio families (an affected individual and both parents). All patients satisfied Poser diagnostic criteria with the majority of patients having relapsing-remitting disease (64.8%), 28.7% were secondary progressive and 6.5% were primary progressive. These assays were performed using TaqMan methodology (the primers used are listed in Supplementary Table S1) on a 7900HT Sequence Detection System, according to the manufacturer’s standard conditions. Marker rs9572903 was found to be monomorphic. Results for the other five SNPs, analysed using the TRANSMIT program to search for evidence of transmission distortion, are shown in Table 1. In each case, 163 samples were typed in duplicate to ensure genotyping accuracy; no inconsistency was observed for any marker. Only one Mendelian inconsistency was observed across all five markers, indicating an approximate genotyping error rate of 0.02%. None of the markers showed deviation from Hardy–Weinberg equilibrium. After appropriate Bonferroni correction, none of the SNPs show any statistically significant evidence for association. SNP rs7143938 showed reduced transmission of the minor allele, which just reached nominal significance and therefore could reflect a modest effect. This SNP lies in the second intron of the Mirror Image Polydactyly Gene 1 (MIPOL1) on chromosome 14q13. This is a developmental gene, and it is not inconceivable that the gene product influences the immune or nervous systems. The observed under-transmission could indicate negative selection of the minor allele but, if so, it is unlikely that this allele would have reached a population frequency of

Novel Olig 1-coding variants and susceptibility to multiple sclerosis

Olig1 is a basic helix–loop–helix (bHLH) transcription factor expressed in cells of the oligodendrocyte lineage in the nervous system. Its role during normal development has not yet been fully resolved, but it is known that in adult life the protein is crucial in the process of remyelination after injury.1–3 Olig1 translocates from the cytoplasm to the nucleus in early remyelinating lesions in rodent models of demyelinating disease as well as in oligodendrocyte precursor cells at the edge of multiple sclerosis lesions.1 Olig1 specifically regulates the expression of the major myelin-specific genes during oligodendrocyte maturation in the brain,2 and remyelination after injury is impaired in Olig1-/- mice.1 In patients with multiple sclerosis, remyelinating capacity is limited even though oligodendrocyte precursor cells are often efficiently recruited.4 These findings raise the question whether genetic variants in the Olig1 gene (Olig1) influence remyelinating capacity and vulnerability to …

No evidence for association of the protein kinase C alpha gene with multiple sclerosis.

Sirs: In their recent article, Barton et al. (2004) [1] tested genetic variation in the protein kinase C alpha (PRKCA) gene for evidence of association with multiple sclerosis. The authors selected this gene for study because it lies in a region (17q22–24) which has been suggested by a number of previous studies [3–6] and because of the immunomodulatory function of the gene [7]. In a comprehensive analysis they typed 35 single nucleotide polymorphisms (SNPs) from across the gene in 184 patients and 340 healthy controls. Association with a strikingly small Bonferroni corrected p-value (p = 1.1 x 10–5) was reported with the haplotype formed between adjacent SNPs, C_2866405_10 and C_2866396_10. Given the apparent strength of this effect it was rather surprising that neither of these SNPs showed any trend towards association when analysed individually. However in order to follow up this interesting observation we typed the two identified SNPs using the same methodology (Applied Biosystems Assay on Demand TaqMan system) in 947 trio families from the UK (each trio family consisting of an affected individual and both their parents). All individuals taking part in the study gave informed consent. Results from the analysis of the individual markers are summarised in the table. As in the analysis performed by Barton et al. (2004) neither SNP showed any evidence for association with MS. In order to test for evidence of haplotypic association data from both SNPs were analysed together using the TRANSMIT program [2]. No evidence for association was seen with any haplotype (the best p-value for any haplotype being 0.42) or for the marker overall (p = 0.39, tested on 2 degrees of freedom, see below). Significance was determined using TRANSMIT’s bootstrap function with 10,000 samples. Substantial, and highly statistically significant, linkage disequilibrium (LD) was found between the two studied SNPs (D’ = 1.0 and r2 = 0.91). In fact the LD was so nearly complete that we did not observe the C_2866405_10 allele1 – C_2866396_10 allele2 (abbreviated to 1–2) haplotype in any of our families (alleles labeled according to the convention followed by Barton et al. 2004). The frequency of this haplotype is thus strikingly different from that reported by Barton et al. (2004), where a frequency of 6.3 % was seen in the controls and 0.2 % in the cases. As phase cannot be determined directly in unrelated cases and controls, we wondered to what extent the frequency of the 1–2 haplotype could have been over inferred from individuals that were doubly heterozygous? As many groups investigating complex diseases are now embarking upon large case-control studies with the intent of employing haplotype analysis to increase power, the potential to observe false positive association as a result of the difficulty in reliably estimating the frequency of rare alleles is of concern and needs to be remembered.

Cis acting expression loci in multiple sclerosis.

In their recent publication, Morley et al. (Morley et al., 2004) combined microarray expression analysis with whole genome linkage screening in order to localise genetic determinants influencing gene expression (bregulatorsQ). Using a stringent linkage significance threshold, they identified a total of 142 such regulator regions, 27 of which are located within 5Mb of the regulated gene (referred to by the authors as cis acting regulators). Extended analysis of these cis acting loci revealed 14 where the tested marker also showed evidence of association with expression, indicating that the typed variants either directly influence expression or are in linkage disequilibrium with nearby variants primarily responsible for this functional effect. Seven of these loci (rs1061810, rs1506440, rs2030364, rs2280714, rs4128767, rs4869311 and rs6547625) regulate genes of immunological importance and could therefore be expected to influence functions of the immune system. Given the higher prior probability that functionally important variants influence susceptibility to complex disease such as multiple sclerosis (Tabor et al., 2002), we decided to study the relevance of these seven SNPs in the context of multiple sclerosis. Association was sought by typing each variant in 938 trio families (an affected