MHC Ancestral Haplotypes Research Articles

G.P.61 Influence of HLA-DRB3 on susceptibility to sporadic inclusion body myositis

In Caucasians, sIBM susceptibility is associated with the HLA-DRB1*0301 allele and the extended 8.1 MHC ancestral haplotype (HLA-A1, B8, DRB3*0101, DRB1*0301, DQB1*0201). Previous studies suggested that HLA-DRB1*0301 itself may not be the risk-conferring allele, but is in linkage disequilibrium with the susceptibility allele. In a recent recombinant mapping study Scott et al. (Neuromuscular Disorders 2011;235:77–83) refined the susceptibility region to a 172kb segment including HLA-DRB3, HLA-DRA and BTNL2. In this study we analysed the carrier frequency of HLA-DRB3 and its alleles, DRB3*0101, DRB3*0202 and DRB3*0301, in 75 biopsy-confirmed sIBM patients of largely Anglo-Celtic origin who had previously undergone high-resolution HLA-DRB1 genotyping. HLA-DRB3 genotyping was performed using PCR–RFLP analysis. As the second expressed HLA-DRB locus, when present, can be either HLA-DRB3 or HLA-DRB4 or HLA-DRB5, carriage of HLA-DRB4 and HLA-DRB5 was also determined, by the presence or absence of amplified DRB4 or DRB5 specific amplicons. The percentage of sIBM individuals who carried a second DRB locus was higher than published figures from a normal Welsh population study (95.2% vs 74.2%, p<0.001; Dorak et al., Genes & Immunity 2002; 3:263–9), as was the frequency of HLA-DRB3 homozygosity (65% vs 14%; p<0.001). Fifty-one of the 75 sIBM patients carried the HLA-DRB3*0101 allele, compared with 54/75 who were HLA-DRB1*0301 carriers; 50/75 carried both DRB1*0301 and DRB3*0101. Thirty of the 51 DRB3*0101 carriers were homozygous, and the allelic frequency of DRB3*0101 was significantly higher than that of DRB1*0301 (52.7% vs 39%, p=0.02). Our findings suggest that preferential carriage of HLA-DRB3 as the second expressed HLA-DRB locus, and homozygosity for HLA-DRB3*0101, which is in linkage dysequilibrium with HLA-DRB1*0301, are associated with increased susceptibility to sIBM.

G.P.63 High-resolution analysis of HLA-DRB1 alleles and diplotypes in an Australian inclusion body myositis cohort

In Caucasians, susceptibility to sporadic inclusion body myositis (s-IBM) is associated with the HLA-DR3 serotype and the HLA-DRB1∗0301 allele which encodes it, and with the extended 8.1 MHC ancestral haplotype. Most previous HLA studies in s-IBM have used serological typing or low-resolution (2-digit) genotyping to identify disease-associated alleles, and the contribution of alleles other than HLA-DRB1∗0301 and of the complementary in trans allele at the HLA-DRB1 locus has received little attention. In the present study we performed high-resolution (4-digit) HLA-DRB1 genotyping in an Australian cohort of 105 s-IBM patients and 189 controls from the Busselton Population Study and analysed risk-associated alleles and effects on age-at-onset (AAO) of the disease. Case–control logistic regression analysis showed that, apart from the strongly associated HLA-DRB1∗0301 allele (OR 13.6), HLA-DRB1∗0101 (OR 3.65) and HLA-DRB1∗1301 (OR 3.31) are independent risk alleles. A number of other alleles, HLA-DRB1∗0401, ∗0404, ∗0701, ∗0901, ∗1101 and ∗1501 were reduced in frequency in s-IBM cases and may be protective. We evaluated the influence of allele combinations on disease risk and found that the strongest estimated risk is with the HLA-DRB1∗0301/∗0101 diplotype, while other allele combinations have diminishing risk effects. HLA-DRB1∗0301 was associated with an AAO 4.3 years earlier than other alleles and DRB1∗1301 with an AAO 4.5 years later, despite it being a risk factor. The findings indicate that the influence of HLA-DRB1 in s-IBM is complex, influencing both disease risk and AAO, and that allelic interactions at the HLA-DRB1 locus contribute to disease susceptibility.

The association of sporadic inclusion body myositis and Sjögren's syndrome in carriers of HLA-DR3 and the 8.1 MHC ancestral haplotype

Sporadic inclusion body myositis (sIBM) usually occurs as an isolated condition, but it may occur in association with another autoimmune disorder such as Sjögren's syndrome. We reviewed sIBM cases with Sjögren's syndrome (sIBM/SS) from the Perth Inflammatory Myopathies Database to determine whether they are distinguishable from other sIBM cases. Six such cases were identified, representing 12% of all sIBM cases. Muscle biopsies confirmed the presence of an inflammatory myopathy with rimmed vacuoles and the characteristic muscle fibre inclusions of sIBM. Five of the six were females, contrasting with a 2:1 male preponderance in the rest of the sIBM cohort. The mean age-at-onset and the pattern of muscle weakness were similar in the two groups. Four out of five sIBM/SS patients treated with immune therapies had improvement in muscle strength lasting for 6–24 months, whereas only 27% of other sIBM patients improved. All 6 patients with sIBM/SS carried the HLA-DRB1*0301 allele, or its equivalent HLA-DR3 serological specificity, compared with 83% of other sIBM cases and all carried some or all of the major markers of the 8.1 MHC ancestral haplotype which is also associated with Sjögren's syndrome. Patients with sIBM/SS represent a subgroup of sIBM cases who are more likely to be female and carriers of HLA-DR3 and the 8.1 MHC ancestral haplotype, and are more likely to respond to treatment. The association of sIBM and Sjögren's syndrome is likely to be due to a common genetic predisposition linked to the MHC and supports the notion that sIBM has an autoimmune basis.

Sporadic inclusion body myositis: HLA-DRB1 allele interactions influence disease risk and clinical phenotype

Susceptibility to sIBM is strongly associated with the HLA-DRB1*03 allele and the 8.1 MHC ancestral haplotype (HLA-A1, B8, DRB1*03) but little is known about the effects of allelic interactions at the DRB1 locus or disease-modifying effects of HLA alleles. HLA-A, B and DRB1 genotyping was performed in 80 Australian sIBM cases and the frequencies of different alleles and allele combinations were compared with those in a group of 190 healthy controls. Genotype–phenotype correlations were also investigated. Amongst carriers of the HLA-DRB1*03 allele, DRB1*03/*01 heterozygotes were over-represented in the sIBM group ( p < 0.003) while. DRB1*03/*04 heterozygotes were under-represented ( p < 0.008). The mean age-at-onset (AAO) was 6.5 years earlier in DRB1*03/*01 heterozygotes who also had more severe quadriceps muscle weakness than the rest of the cohort. The findings indicate that interactions between the HLA-DRB1*03 allele and other alleles at the DRB1 locus can influence disease susceptibility and the clinical phenotype in sIBM.

G.P.16.09 Epistatic interactions between DRB1 alleles influence susceptibility and clinical phenotype in sporadic inclusion body myositis (sIBM)

Background: Susceptibility to sIBM is strongly associated with HLA-DRB1*03 and the 8.1 MHC ancestral haplotype but little is known about the effect of allelic interactions at the DRB1 locus or disease-modifying effects of HLA alleles. Objectives: To determine whether DRB1 allele interactions influence susceptibility to sIBM and disease phenotype. Patients & Methods: HLA-DRB1 genotyping was carried out in 80 biopsy-proven cases of sIBM using high-resolution DNA sequencing in 55 cases and serological typing in 25 cases.

Polymorphic MHC ancestral haplotypes affect the activity of tumour necrosis factor-alpha.

It remains unclear which MHC loci are involved in susceptibility to autoimmune diseases and immune deficiencies. We have chosen to evaluate whether different alleles of tumour necrosis factor-alpha (TNF-alpha) are important, as TNF has been implicated in the etiology of many immunological disorders. We have shown previously that a restriction fragment length polymorphism in the TNF region correlates with MHC ancestral haplotypes. We therefore examined the effect of ancestral haplotype on the activity of TNF-alpha in culture supernatants of lymphoblastoid cell lines. The results demonstrate that TNF-alpha activity in supernatants of 8.1 (A1, B8, DR3) cell lines was higher than that present in the supernatants from cells homozygous for eight different MHC ancestral haplotypes, and indicate that polymorphisms in TNF-alpha, or in other MHC genes that regulate TNF, may be responsible for the 8.1 phenotype.

Frequency of Carriers of 8.1 Ancestral Haplotype and its Fragments in Two Caucasian Populations

Within the human MHC region larger stretches of conserved DNA, called conserved ancestral haplotypes exist. However, many MHC haplotypes contain only fragments of an ancestral haplotype. Little is known, however, on relative distribution of the ancestral haplotypes to their fragments. Therefore we determined the frequency of carriers of the whole ancestral haplotype 8.1 (AH8.1) and its fragments in 127 healthy Hungarian people, 101 healthy Ohioian females, and in nine Hungarian families. The HLA-DQ2, HLA-DR3(17), RAGE -429C allele, the mono-S-C4B genotype, the HSP70-2 1267G allele and the TNF -308A (TNF2) allele were used as markers of the AH8.1. Frequency of carriers of the whole AH8.1 and its fragments was similar in the both populations. 18% of the subjects carried the whole AH8.1 in at least one chromosome, while 17–20%, 36–39%, and 24–29%, respectively carried two or three constituents of the haplotype, only one constituent or none of them. Similar results were obtained in the family study. In addition, marked differences were found in the relationship of the constituents' alleles to the whole AH8.1. In both populations, 29%, 50–59%, 52–56% and 76–96%, respectively of the carriers of HSP70-2 1267G, RAGE-429C, TNF2, and mono-S carriers carried the whole 8.1 haplotype. These findings may have important implications for studies of the disease associations with different MHC ancestral haplotypes.

Familial inclusion body myositis in a mother and son with different ancestral MHC haplotypes

An Ashkenazi Jewish family in which the mother and a son both have inclusion body myositis (IBM) is reported. The condition developed at an earlier age and was more rapidly progressive and less responsive to treatment in the son than in the mother or other IBM patients in our clinic. Genetic analysis showed that the mother carried alleles of the 8.1 MHC ancestral haplotype (AH; HLA-B8, DRB1 *0301), which is found in 85% of IBM patients in Western Australia. The son did not inherit this haplotype, but carried alleles characteristic of the 52.1AH (HLA-B5, DRB1 *1502) of paternal origin. The findings indicate that in this family either the 8.1AH or 52.1AH may carry susceptibility for IBM and that the 52.1AH is associated with a more severe and treatment-resistant form of the disease.

Sporadic inclusion body myositis in Japanese is associated with the MHC ancestral haplotype 52.1

In Caucasians, sporadic inclusion body myositis has been associated with the MHC ancestral haplotypes; HLA-A1, B8, DR3 (8.1AH) and HLA-B35, DR1 (35.2AH). It is not known whether these haplotypes carry susceptibility for the disease in other ethnic groups. We report here the results of HLA-B and -DRB1 typing using a high-resolution sequence-based technique in a cohort of 31 Japanese patients with definite sIBM. Patient allele frequencies were 40.3% for HLA-B*5201 (10.7% in controls: p<0.001) and 37.1% for HLA-DRB1*1502 (10% in controls: p<0.001). Both alleles were found together as part of a conserved haplotype (52.1AH) at a frequency of 37.1% in patients (8.4% in controls: p<0.001). This is the first description of a haplotypic MHC association with sporadic inclusion body myositis in Japanese patients. These findings indicate that different MHC ancestral haplotypes are associated with sIBM in different ethnic groups and further emphasize the importance of genetic factors in this condition.

Genetics of human complement component C4 and evolution the central MHC

The two classes of human complement component C4 proteins C4A and C4B manifest differential chemical reactivities and binding affinities towards target surfaces and complement receptor CR1. There are multiple, polymorphic allotypes of C4A and C4B proteins. A complex multiplication pattern of C4A and C4B genes with variations in gene size, gene dosage and flanking genes exists in the population. This is probably driven by the selection pressure to respond to a great variety of parasites efficiently and effectively, which the bony fish achieved through the multiplication and diversification of the related complement C3 proteins. Complement C4, C3 and C5 belong to the alpha2 macroglobulin protein family but acquired specific features that include an anaphylatoxin domain, a netrin (NTR) domain, and stretches of basic residues for proteolytic processings to form multiple chain structures. Complement C3 and C4 are important in the innate immune response as they opsonize parasites for phagocytosis. The emergence of complement C3 predates proteins involved in the adaptive immune response as C3 is present in deuterostome invertebrates such as echinoderms. The human C4 genes are located in the central MHC at chromosome 6p21.3. C3 and C5 are located at chromosome 19 and 9, respectively, with representatives of the other groups of genes paralogous to the MHC at 19p13.1-p13.3, 1q21-25, and 9q33-34. The central MHC also contains genes for complement components C2 and Bf. These genes appear to have similar evolutionary histories to C3/C4/C5 and are used here to illustrate stepwise processes resulting in co-location of diverse domains, chromosomal duplication, local segmental duplication and divergence of sequence and function. This model of evolution is useful in the investigation of innate and acquired immunity and in seeking explanations for diseases associated with MHC ancestral haplotypes.

The primate MHC contains sequences related to the fibroblast growth factor receptor gene family.

We have cloned and sequenced a genomic region centromeric of the HLA-B locus from different MHC ancestral haplotypes. These haplotypes are associated with several diseases. The sequences were analyzed for coding potential and their relevance to disease associations were assessed with respect to the level of polymorphism. Analysis of sequences located approximately 25kb centromeric of HLA-B reveals the existence of fibroblast growth factor receptor related sequences. These sequences designated PERB1 (FGFR6) reveal 80% homology, at both nucleic acid and amino acid level, to the immunoglobulin domain 1 (Ig-1) of the human fibroblast growth factor receptor 3 (FGFR3) gene. Amino acid comparison of the Ig-1 domain of PERB1 to those of other FGFR molecules indicates that PERB1 is more closely related to FGFR3 and FGFR5 than to FGFR1, FGFR2 or FGFR4. Genomic sequence analysis, however, reveals no consensus splice sites and indicates the existence of inframe premature stop codons in the putative coding sequences. The results suggest that these sequences may represent FGFR gene fragments existing within the central MHC. Sequence analysis of the Mhc in 6 chimpanzee and one orangutan indicates that the existence of PERB1 predates the speciation of the three species. The fact that the MHC contains a mixture of functional and nonfunctional (pseudo) genes suggests that a functional copy of PERB1 (FGFR6) may exist within or in close proximity to the MHC.

Analysis of MHC Genomic Structure and Gene Content between HLA-B and TNF Using Yeast Artificial Chromosomes

To characterize the genomic structure and gene content between HLA-B and TNF we studied three overlapping yeast artificial chromosome (YAC) clones derived from the Boleth cell line carrying the 62.1 MHC ancestral haplotype (AH). In particular, we have evaluated the stability of haplospecific sequence motifs (haplospecific geometric elements) in YACs. By PCR and gene scanning we show that the two larger YAC clones (318G11 and 8F2) exhibit features characteristic of the 62.1 AH carried by the parental cell line as well as those of a separate cell line (BSM) which also carries 62.1. It is concluded that a highly polymorphic duplicated region is represented faithfully in spite of the fact that it contains sequences that could be interpreted as simple repeats liable to mutations. In contrast, microsatellites AFM031 and AFM158, which are also located in the MHC, showed greater instability and the structure in the YACs was not consistent with genomic structure. Likewise, different examples of the same ancestral haplotype show differences at these microsatellite loci. We conclude that polymorphic repeat sequences may exhibit a range of stabilities in vivo and that this may be reflected in their stability in YACs. Northern blotting using the largest YAC clone revealed several novel transcripts which await localization.

Ancestral haplotypes carry haplotypic and haplospecific polymorphisms of BAT1: possible relevance to autoimmune disease.

The human BAT1 gene, located in the central MHC region (approximately 170 kb centrometric of HLA-B), is polymorphic and the polymorphism correlates with MHC ancestral haplotypes. Allelic RFLP patterns have been assigned to several ancestral haplotypes and have been shown to be 'haplotypic' (i.e. found on all examples of the same ancestral haplotype) and in some cases 'haplospecific' (i.e. unique to one ancestral haplotype). The relevance of the BAT1 polymorphism to susceptibility to Myasthenia Gravis (MG) has been investigated. The frequency of the BAT1 B allelic pattern is increased in patients with MG (n = 16) compared to an equal number of control subjects. The increase is due to the association between MG and the 8.1 ancestral haplotype (HLA A1, Cw7, B8, BfS, C4AQ0, C4B1, DR3, DQw2).

Differences in gene copy number carried by different MHC ancestral haplotypes. Quantitation after physical separation of haplotypes by pulsed field gel electrophoresis.

We have examined the hypothesis that MHC ancestral haplotypes have a specific content of genes regulating the extent of autoimmune reactions. Gene copy number was quantitated by objective densitometry after PFGE was used to separate heterozygous AHs of different lengths. Initially we analyzed examples of known gene copy number at the C4 and 21 hydroxylase loci and showed that the approach provides predictable results. We then studied heterozygotes containing one characterized and one uncharacterized AH with particular attention to the gene copy number at the C4, Cyp21, and DRB loci. Each AH studied has a characteristic gene copy number at each locus studied. The same may be true of TNF, but other possibilities must be considered. AHs are markers for extensive chromosomal segments including particular numbers of several functional genes. Since AHs mark susceptibility to autoimmune disease, differences in gene copy number may be implicated.

Pulsed field electrophoresis identifies the possible locations of diabetogenic genes contained within MHC ancestral haplotypes

Pulsed field electrophoresis identifies the possible locations of diabetogenic genes contained within MHC ancestral haplotypes