Fine mapping of quantitative trait loci (QTL) for osteochondrosis in Hanoverian warmblood horses

Previous genome and chromosome scans revealed quantitative trait loci (QTL) for osteochondrosis (OC) in Hanoverian warmblood horses on equine chromosomes (ECA) 2, 4, 5, 16, 18 and 21. The subject of this study was to refine detected QTL on ECA2 and 4 in order to identify potential candidate genes involved in the etiopathogenesis of OC. The release of the horse genome assembly EquCab2 enabled the development of new microsatellites which decreased the marker distance of less than 1 Mb in significant regions. 104 Hanoverian warmblood horses of 14 paternal half-sib families, their dams and 8 sires were genotyped. Linkage and association analyses were performed for six phenotypic traits: (1) OC (fetlock and/or hock joints), (2) OCD (fetlock and/or hock joints), (3) fetlock OC, (4) fetlock OCD, (5) hock OC, and (6) hock OCD. On ECA2, altogether 37 newly developed microsatellites and 24 Single Nucleotide Polymorphisms (SNPs) were genotyped. The QTL for the different traits of OC could be confirmed and narrowed down: for OC to an interval at 20.08 to 30.94 Mb, for OCD at 26.89 to 29.47 Mb, for fetlock OC at 15.65 to 30.94 Mb, for fetlock OCD at 21.15 to 31.91 Mb, and for hock OC between 26.89 to 33.05 Mb. On ECA4, a total of 41 microsatellites and 11 SNPs were newly developed and used for linkage and association analyses. The QTL for OC could be delimited between 4.92 and 39.76 Mb, the fetlock OC QTL could be narrowed down to an interval at 7.42 to 13.10 Mb, at 27.15 and 28.79 Mb and at 56.15 to 59.84 Mb. Furthermore, a QTL for hock OC could be detected between 3.62 and 6.34 Mb. The further aim of this study was to approve identified QTL of previous studies and to detect new potential QTL by performing a whole genome scan with SNPs using the equine SNP50 BeadChip. This comprehensive and powerful microarray allows genotyping of more than 50,000 SNPs simoultaneously. The whole genome association analysis was performed with 154 unrelated Hanoverian warmblood horses, 39 of them already genotyped in the genome scan. Regarding detected QTL on chromosomes 2, 4, 5, 16, 18 and 21, a total of 313 significant associated SNPs for the six different phenotypic traits of OC were observed. Multiple variance analysis was performed in order to focus on meaningful associated SNPs. The QTL on ECA2 could be confirmed at 17.55 Mb and was therefore consistent with results of genome scan and fine mapping studies (15.65 - 33.05 Mb all phenotypic traits). On ECA4, variance analysis revealed two QTL at 7.61 Mb and additionally at 39.26 Mb which were located within the QTL of previous linkage studies as well. Furthermore, a total of ten QTL on ECA3, 5, 7, 16, 19, 20, 22, 26, and 29 were detected. The whole genome scan and adjacent fine mapping studies were the first steps towards the identification of significant regions harbouring potential candidate genes which might be involved in the development of OC. The equine BeadChip implied a great progress in molecular genetic research and permitted the verification of identified QTL on ECA2 and 4. Furthermore, this microarray enabled the detection of new potential QTL previously unknown. Following studies have to approve current results of the BeadChip in a large sample of horses. The use of the next generation of equine microarray technologies with a much denser marker set could be helpful to clarify the relevance of numerous associated SNPs. Furthermore, mutation and expression analyses of detected potential candidate genes nearby identified QTL should be performed.