Molecular genetics of autosomal recessive retinitis pigmentosa

Autosomal recessive retinitis pigmentosa (arRP) is one of the commonest forms of monogenic retinal degeneration (RD). To date, 24 loci have been implicated in the pathogenesis of arRP. The genes for five of these loci (RP22, RP25, RP28, RP29 and RP32), still remain to be identified. This thesis mainly focused on the cloning of a major gene (RP25) however identifying novel loci for recessive RP constituted a significant objective. Originally the RP25 locus was mapped to chromosome 6pl2.1-ql5, a region that spans 34 Mb, by our collaborators in Seville in seven Spanish families. Initially, a whole-genome scan in these families was undertaken using GeneChip 10K array. The data obtained confirmed the initial findings of linkage to the RP25 region. To date, 61 out of 111 genes within the interval (-55%) have been excluded as disease causing by direct sequence analysis. A large number of single nucleotide polymorphisms (SNPs), of which a significant percentage was novel were identified. We have also postulated that both RP25 and Leber congenital amaurosis 5 (LCA5), a severe form of RD, could be due to the same genetic defect since they genetically overlap. Therefore, seventeen LCA families were genotyped to identify new LCA5 families that may further refine the RP25 interval by identifying novel crossovers. However, the gene for LCA5 has been recently cloned and sequence analysis of the RP25 families rules out this gene as causative of RP25. To investigate if copy number variations (CNVs) exist within the RP25 interval, a comparative genome hybridisation (CGH) was performed on one of the RP25 families (RP5). A clone from the tiling path, chr6tp-19C7, within 6ql2 was observed to be deleted in all affected members of this family indicating that one of the genes within this interval could be responsible for the RP25 phenotype. A novel approach utilising the 10K GeneChip for identifying the disease locus in three non-consanguineous Chinese families with arRP was implemented. The studied families were probably linked to the RP25 locus proposing that this approach could be a useful tool for genetic mapping in cases of rare and genetically heterogeneous recessive traits. Finally, in parallel, a genomewide linkage search in a consanguineous family with arRP was undertaken. Linkage to a 10-cM interval on chromosome 10q23.1-23.3 was observed where a good candidate gene, protocadherin-21 (PCDH21), is located. A homozygous 1-bp deletion was identified in this family in addition to two other novel mutations in two different patients raising the possibility that PCDH21 is likely to be a novel gene for RP.