3.2.1. Prph2<sup>307</sup>/+ and Prph2<sup>307</sup>/<sup>307</sup>

A deletion of a single base pair at codon 307 in human *PRPH2* results in a slow progressing form of ADRP [106]. A mouse model in which a targeted single base deletion at codon 307 in *Prph2* was introduced and showed a severe decline in photoreceptor survival and function [107]. ERG revealed a decrease in photoreceptor function in heterozygous animals starting at two months of age, while no ERG responses were detected in the homozygous animals even at one month of age. Retinal phenotypes in the heterozygous or homozygous mice were more severe than in age-matched *Prph2*+/− and *Prph2*−/− mice, respectively. Thus, the deletion mutation resulted in a strong dominant gain-of-function e ffect. This knockin model represents a drastic case of a retinal phenotype that di ffers greatly from that observed in the patients whereby the structural and functional decline is slow, starting in the fifth or sixth decades of life. The mouse model, on the other hand, displays a rapid degeneration with an early onset. The deletion at codon 307 in the human *PRPH2* gene causes a frameshift and creates a stop codon after the addition of 16 amino acids. If the resulting mutant protein is stable, it is expected to be 26 amino acid shorter than the wildtype. It is likely the case since the phenotype seen in patients is mild due to gain-of-function defect, but if the mutant protein is unstable, then the phenotype would be loss-of-function defect similar to that of haploinsu fficiency. However, in the mouse genome, such one base deletion at codon 307 is predicted to result in the alteration of the last 40 amino acids of the C terminus of the protein and the addition of 11 extra amino acids. This results in the translation of 51 amino acids after codon 307 in the mouse [107]. The di fferent e ffects on the translated protein caused by the deletion in codon 307 observed between human and mouse explains

the di fferences in the severity of the retinal phenotypes. It is important to note that retinal phenotype in the homozygous mouse is reported to be worse than the complete-loss-of-function phenotype seen in the Prph2 null mice. This observation suggests that the mouse phenotype is likely a combination of lossand gain-of-function defects and that the toxicity beyond that seen in the Prph2 knockout is probably the outcome of the latter. Since the mouse model was not assessed for the presence of the mutant protein, at this point, it is unclear whether the toxicity arises from the absence of the protein, the 51 altered amino acids at the C-terminus or the lack of endogenous 40 amino acids at the C-terminus. Obviously, the addition of 16 altered amino acids in the human PRPH2 and shortening the protein by 26 amino acids at the C-terminus have less drastic e ffect on the retina than those modifications occurred in the mouse protein.
