HLA-DRB1-DQB1 Haplotypes Confer Susceptibility and Resistance to Multiple Sclerosis in Sardinia

Abstract

IntroductionGenetic predisposition to multiple sclerosis (MS) in Sardinia (Italy) has been associated with five DRB1*-DQB1* haplotypes of the human leukocyte antigen (HLA). Given the complexity of these associations, an in-depth re-analysis was performed with the specific aims of confirming the haplotype associations; establishing the independence of the associated haplotypes; and assessing patients' genotypic risk of developing MS.Methods and ResultsA transmission disequilibrium test (TDT) of the DRB1*-DQB1* haplotypes in 943 trio families, confirmed a higher than expected transmission rate (over-transmission) of the *13:03-*03:01 (OR = 2.9, P = 7.6×10−3), *04:05-*03:01 (OR = 2.4, P = 4.4×10−6) and *03:01-*02:01 (OR = 2.1, P = 1.0×10−15) haplotype. In contrast, the *16:01-*05:02 (OR = 0.5, P = 5.4×10−11) and the *15:02-*06:01 (OR = 0.3, P = 1.5×10−3) haplotypes exhibited a lower than expected transmission rate (under-transmission). The independence of the transmission of each positively and negatively associated haplotype was confirmed relative to all positively associated haplotypes, and to the negatively associated *16:01-*05:02 haplotype. In patients, carriage of two predisposing haplotypes, or of protective haplotypes, respectively increased or decreased the patient's risk of developing MS. The risk of MS followed a multiplicative model of genotypes, which was, in order of decreasing ORs: *04:05-*0301/*03:01-*02:01 (OR = 4.5); *03:01-*02:01/*03:01-*02:01 (OR = 4.1); and the *16:01-*05:02/*16:01-*0502 (OR = 0.2) genotypes. Analysis of DRB1 and DQB1 protein chain residues showed that the Val/Gly residue at position 86 of the DRB1 chain was the only difference between the protective *16:01- *15:02 alleles and the predisposing *15:01 one. Similarly, the Ala/Val residue at position 38 of the DQB1 chain differentiated the positively associated *06:02 allele and the negatively associated *05:02, *06:01 alleles.ConclusionsThese findings show that the association of specific, independent DRB1*-DQB1* haplotypes confers susceptibility or resistance to MS in the MS-prone Sardinian population. The data also supports a functional role for specific residues of the DRB1 and DQB1 proteins in predisposing patients to MS.

Genetic risk variants in African Americans with multiple sclerosis

“…8,20,31 The role of HLA-DRB1*04:05 observed in this study was also reported in other populations, including Japanese and Sardinian. 33,34 We also confirmed the protective association of HLA-A*02:01 in African Americans, independent of the HLA-DRB1 risk alleles. Also consistent with previous reports, 35 a lower frequency for HLA-DRB1*11 was observed.…”

Enrichment for Northern European-derived multiple sclerosis risk alleles in Sardinia

“…26 HLA serotypes were generated on 129 Sardinian samples; imputed HLA genotypes were used for samples without serotype data (total 247 controls and 73 cases). The concordance rate between serotype and imputed HLA risk haplotypes for samples that had both (N = 174) was 0.82.…”

The Role of HLA in MS Susceptibility and Phenotype

“…(Table 3). Overall, in addition to DRB1*15:01, other DRB1 alleles that appear to be linked to development of MS across a number of different ethnic groups include DRB1*03:01 (Barcellos et al 2006;Cocco et al 2012Cocco et al , 2013Dyment et al 2005;Field et al 2010;Isobe et al 2013;Marrosu et al 1998;Okensberg et al 2004;Patsopoulos et al 2013;Sawcer et al 2011;Zhang et al 2011), DRB1*08:01 (particularly when carried together with DRB1*15:01) (Barcellos et al 2006;Chao et al 2010;Dyment et al 2005), some alleles of DRB1*04 (Brassat et al 2005;Cocco et al 2012Cocco et al , 2013Isobe et al 2013;Patsopoulos et al 2013), and DRB1*13:03 (Cocco et al 2012(Cocco et al , 2013Patsopoulos et al 2013;Sawcer et al 2011). Although carriage of DRB1*15:01 is often significantly higher in patients with MS compared to healthy controls matched for ethnicity, in many ethnic groups, this still only equates to a relatively small percentage of MS patients who carry this allele.…”

“…In individuals of African descent in South (Balnyte et al 2012(Balnyte et al , 2013 DRB1*15; DRB1*08 (in RR-MS) Italian (Brassat et al 2005;Cocco et al 2012Cocco et al , 2013Laroni et al 2006 (Benedek et al 2010) DRB1*0301-DQB1*0201 Maltese (Dean et al 2008) DRB1*15; DRB1*11 Pakistani (Wasay et al 2013) No statistically significant linkages African-American (Cree et al 2009;McElroy et al 2010 America, it has been suggested that DQB1*06:02 is of greater relevance to disease susceptibility (Caballero et al 1999), although studies on African-American patients with MS have not found any effects of DQB1*06:02 that are independent of DRB1*15 (Okensberg et al 2004). It has also been suggested that DRB1, DQA1 and DQB1 alleles contribute to MS susceptibility via epistatic interactions through haplotypic, rather than alleleic, associations (Lincoln et al 2009), and even that they may act in trans with other HLA molecules carried by an individual, so that alleles which by themselves are not risk alleles can nevertheless increase disease risk when they combine with DRB1*15:01 (Lincoln et al 2009).…”

“…There is also a robust literature showing that certain class II HLA types occur less frequently in MS patients than that in healthy individuals: these include HLA-DR1, HLA-DR4, HLA-DR6, HLA-DR7, HLA-DR9, HLA-DR11 and HLA-DR14 types (Balnyte et al 2012;Barcellos et al 2006;Cocco et al 2012Cocco et al , 2013DeLuca et al 2007;Dyment et al 2005;Fernandez et al 2009;Greer and Pender 2005;Kaimen-Maciel et al 2009;Link et al 2012;Masterman et al 2000;McDonnell et al 1999;Qiu et al 2011a;Stankovich et al 2009;Wu et al 2010a;Zhang et al 2011). It should be noted, however, that many of the studies from which this information is derived only reported HLA typing at the serotyping or two digit level and that some of these "protective" HLA types consist of many different alleles with markedly different amino acid sequences in regions forming the antigen-binding grooves of the HLA molecules, as discussed further below.…”