Interaction between HLA-DRB1-DQB1 haplotypes in Sardinian multiple sclerosis population

Abstract

We performed a case-control study in 2,555 multiple sclerosis (MS) Sardinian patients and 1,365 healthy ethnically matched controls, analyzing the interactions between HLA-DRB1-DQB1 haplotypes and defining a rank of genotypes conferring a variable degree of risk to the disease. Four haplotypes were found to confer susceptibility (*13:03-*03:01 OR = 3.3, Pc 5.1 × 10(-5), *04:05-*03:01 OR = 2.1, Pc 9.7 × 10(-8), *15:01-*06:02 OR = 2.0, Pc = 9.1 × 10(-3), *03:01-*02:01 OR = 1.7 Pc = 7.9 × 10(-22)) and protection (*11, OR = 0.8, Pc = 2.7 × 10(-2), *16:01-*05:02 OR = 0.6, Pc = 4.8 × 10(-16), *14:01-4-*05:031 = OR = 0.5, Pc = 9.8 × 10(-4) and *15:02-*06:01 OR = 0.4, Pc = 5.1 × 10(-4)). The relative predispositional effect method confirms all the positively associated haplotypes and showed that also *08 and *04 haplotypes confers susceptibility, while the *11 was excluded as protective haplotype. Genotypic ORs highlighted two typologies of interaction between haplotypes: i) a neutral interaction, in which the global risk is coherent with the sum of the single haplotype risks; ii) a negative interaction, in which the genotypic OR observed is lower than the sum of the OR of the two haplotypes. The phylogenic tree of the MS-associated DRB1 alleles found in Sardinian patients revealed a cluster represented by *14:01, *04:05, *13∶03, *08:01 and *03:01 alleles. Sequence alignment analysis showed that amino acids near pocket P4 and pocket P9 differentiated protective from predisposing alleles under investigation. Furthermore, molecular dynamics simulation performed on alleles revealed that position 70 is crucial in binding of MBP 85-99 peptide. All together, these data suggest that propensity to MS observed in Sardinian population carried by the various HLA-DRB1-DQB1 molecules can be due to functional peculiarity in the antigen presentation mechanisms.

Figure 1. Phylogenetic tree of the MS associate DRB1 alleles.

Figure 2. Binding region for the MHC-peptide complex for DRB1*03∶01 allele.

MHC binding region is shown in cartoon representation (black), MBP peptide backbone is shown in ball-stick representation. The residues in pocket P4, and P9 are shown in surface representation and are colored based on residue type (blue: basic, red: acidic, green: polar).

Figure 3. Binding region for the MHC-peptide complex for DRB1*14∶01 allele.

MHC is shown in cartoon representation (black), MBP peptide backbone is shown in ball-stick representation. The residues in pocket P4, and P9 are shown in surface representation and are colored based on residue type (blue: basic, red: acidic, green: polar).

Figure 4. MBP-MHC H-bonds.

Percentual duration time of MBP-established H-bond, during 3 ns MD simulation, for the residues of DRB1*03∶01 (in blue) and DRB1*14∶01 (green) binding site.

Figure 5. Area calculation near Pockets.

Total available area (in unit Ų) near (left histogram) the P4 region (right histogram) P9 region for the alleles DRB1*03∶01 (in blue) and DRB1*14∶01 (green), in the absence of MBP peptide.

Figure 6. Polar and apolar area calculation.

In percentage, polar and apolar area available near (A) P4 region and (B) residue 60 (close to pocket P9 region), for the alleles DRB1*03∶01 and DRB1*14∶01 respectively.