Functional studies of mutations found in PRPF3 pre-MRNA splicing factor causing RP18

Purpose: Genes encoding pre–mRNA splicing factors (PRPF31, PRPF8, PRPF3) were found mutated in three forms of autosomal dominant RP (adRP, RP11, RP13 and RP18). The link between mutant splicing factors and retinal degeneration is completely obscure. So far, only 2 amino acids (P493S and T494M) were found mutated in PRPF3. Analysis of the subcellular localization and interaction with other splicing factors will be fundamental in defining why mutation of PRPF3 causes RP and whether RP18 is due to a dominant effect of the mutant protein or haploinsufficiency. Methods: Analyses of the subcellular localization of the endogenous protein by immunofluorescence using specific polyclonal and monoclonal antibodies in cell lines and in the murine retina were undertaken. Transfection experiments using cell lines to define subcellular localization of the mutant forms of PRPF3. Interaction of wild type and mutant PRPF3 with known partner proteins and snRNA was performed by co–immunoprecipitation experiments. Results: We analyzed expression of the Prpf3 gene in the developing and adult murine eye by RNA in situ hybridization. We found that Prpf3 transcripts are ubiquitously distributed but higher amounts could be detected in the central nervous system and neural retina. Using in vitro transfection in HeLa, NIH3T3, Cos7 cells and primary retinal cultures, PRPF3 localizes in the nuclei. Using similar approaches we found the two different mutations (P493S and T494M) have similar intracellular distribution as found for the wild–type protein. Mutant PRPF3 also maintains its co–localization with spliceosome specific proteins such as SC35. Conclusions: We analyzed Prpf3 transcript distribution in vivo and found no direct correlation with photoreceptors, the cells that undergo degeneration in patients with mutations in this gene. Co–localization with splicing factors confirms that the protein is part of the spliceosome. Finally, mutant PRPF3 has a subcellular distribution similar to the wild type, suggesting that RP18 is caused either by a dominant effect of the mutation or by haploinsufficiency due to lack of function.