The ovine MHC class II region : polymorphism, organisation, and association with footrot

The major histocompatibility complex (MHC) consists of a long array of genetic loci which are central to the immune response. Numerous studies have demonstrated associations between the genes in this complex and economically important traits, such as disease resistance in farm animals. However, in several species including the sheep, little is known of the organisation and structure of this genetic region. The main aim of this study was to investigate polymorphism within the ovine MHC class II region by Southern blot hybridisation using sheep-specific probes. The DQA, DQB, DRA and DRB loci were investigated, resulting in identification of thirty-two alleles (8 DQA1, 16 DQA2, 5 DQB, and 3 DRA). Little information was derived from the DRB subregion due to the poor quality of the autoradiographs generated with the DRB probe and extensive cross-hybridisation with the DQB locus. Allele frequencies were found to vary both between breeds and within different flocks of the same breed. Recombination within the ovine DQ-DR subregion was investigated in twelve families (9 paternal half-sib and 3 full-sib) by linkage analysis. Lod scores were derived for a range of recombination fractions (θ) between 0.5 and 0.000001 for linkage between pairs of markers within the DQ-DR subregion. For each comparison, maximum lod scores were identified at the recombination fraction closest to zero (θ= 0.000001) of those evaluated, indicating the absence of recombination and the close linkage of genes in the DQ-DR subregion. Individuals were identified which did not generate hybridising bands with the DQA1 exon 2 probe, suggesting a deletion in these sheep. This postulate was supported by a similar finding when hybridisation was performed with a DQA1 probe which covered the third and forth exons. The apparent deletion of the DQA1 gene was termed the DQA1 null allele (DQA1-N). The frequency of DQA1-N homozygotes was comparable between distantly related breeds, suggesting that the deletion occurred before the establishment of present breeds of sheep. The usefulness of PCR-RFLP analysis for the identification of polymorphisms within the DQA1 and DQA2 genes was demonstrated. With the DQA1-specific primer pair DQA1(top)/DQA1(bot) a conserved exon 2 BglII restriction site and a polymorphic BglII site in the 5' non-coding region of the DQA1 gene were identified. However, this primer pair selectively amplified certain DQAl alleles, indicating that additional primers would be required to enable PCR-RFLP analysis of all DQA1 types. PCR-RFLP analysis of animals of known DQA2 genotype with the primer pair DQA1(oz)/DQA2(oz) revealed association of polymorphic and conserved restriction sites with different DQA2 alleles. One allele (DQA2-E) appeared to be associated with extensive polymorphism compared to the other DQA2 alleles. Two of the four restriction endonucleases used for RFLP analysis of the PCR product from DQA2-E homozygotes identified polymorphic restriction sites within the second exon. Southern blot hybridisation and PCR-RFLP analysis were utilised to provide evidence for the detection of extra DQA2-like genetic material in individuals which possessed the DQA1 null allele. Banding patterns were consistent with the detection of a second DQA2 gene, suggesting that this gene may have been duplicated in individuals which lost the DQA1 gene. The composition of the ovine DQA region therefore appears to vary between haplotypes, some possessing DQA1 and DQA2 while others possess DQA2 and a second DQA2-like gene. Evidence was also obtained for the presence of extra DQB-like genetic material in individuals which lack the DQA1 gene. Whether the extra DQA2 and DQB genetic material constitute a third DQA/DQB locus remains to be determined. Association between genetic variation within the ovine MHC class II region and resistance to footrot was investigated. Half-sib families of the Broomfield Corriedale Breeding Group were subjected to footrot challenge by the introduction of footrot infected animals into the flock. Based on their response to challenge, the progeny of the different families were classified as resistant, susceptible or self-curing. Each sire's two MHC haplotypes, comprising alleles at the DQA1, DQA2, DQB and DRA loci, were inferred from the alleles present in his offspring. The progeny in each family were divided two groups based on the haplotype inherited from the sire and associations between paternal haplotype and disease status investigated by chi-square analysis. A highly significant association (P= 0.005) was observed within one family sired by ram 20/88 following challenge conducted in 1992. The significance of this association increased (P= 0.002) when the self-curing animals were combined with the resistant animals and became even more significant (P= 0.001) when the self-curing animals were excluded from analysis. No associations were observed in the other three half-sib families subjected to challenge in 1992. In a second challenge conducted in 1993 the association of MHC haplotype with resistance to footrot in the half-sib family sired by ram 20/88 was not reproduced. However, this challenge was adversely affected by weather conditions which did not favour the transmission of footrot resulting in unreliable assignments of disease status. This study, while not conclusive, augurs well for future investigations of association between the ovine MHC and footrot resistance or other bacterial infections