Intraperitoneal chromophore injections delay early-onset and rapid retinal cone degeneration in a mouse model of Leber congenital amaurosis

Highly expressed in the retinal pigment epithelium (RPE), the RPE-specific 65-kDa (RPE65) enzyme is indispensable to generate 11-cis-retinal (11cRAL), a chromophore for rhodopsin and cone photopigments. RPE65 deficiency can lead to Leber congenital amaurosis type 2 (LCA2), in which the isomerization of photobleached all-trans-retinal into photosensitive 11cRAL is blocked, ultimately causing severe retinal dysfunction and degeneration. The related mouse models, which are constructed through gene knockout or caused by spontaneous mutations, morphologically present with early-onset and rapid retinal cone cells degeneration, including loss of short-wavelength-sensitive cone opsins (S-opsins) and mislocalization of medium-wavelength-sensitive cone opsins (M-opsins). Studies have shown that routine Rpe65 gene replacement therapy, mediated by an adeno-associated virus (AAV) vector, can restore RPE65 protein. However, AAV transfection and Rpe65 transgene expression require at least one to two weeks, and the treatment cannot fully block the early-onset cone degeneration. To determine the feasibility of delaying cone degeneration before gene therapy, we investigated the impact of 11cRAL treatment in an early-age LCA2 retinal degeneration 12 (rd12) mouse model. Similar to human patients, the mouse model carries a spontaneous mutation in the Rpe65 gene, which results in disrupted endogenous 11cRAL regeneration. We found that RPE65 deficiency did not notably affect rodent retinal vessels. Under red light illumination, the rd12 mice were intraperitoneally injected with exogenous 11cRAL from postnatal day (P) 14 to P21. Three days after the last injection, a notable recovery of retinal function was observed using scotopic and photopic electroretinograms. Using optical coherence tomography and histological analyses of the deficient retinas, we found changes in the thickness of the photoreceptor outer segment (OS); this change could be rescued by early 11cRAL treatment. In addition, the treatment notably preserved M- and S-opsins, both of which maintained appropriate localization inside cone cells, as shown by the wild-type mice. In contrast, the age-matched untreated rd12 mice were characterized by retinal S-opsin loss and M-opsin mislocalization from the photoreceptor OS to the inner segment, outer nuclear layer, or outer plexiform layer. Notably, 11cRAL treatment could not maintain retinal function for a long time. Ten days after the last injection, the rod and M-cone electroretinograms significantly decreased, and S-cone responses almost extinguished. Our findings suggest that early 11cRAL treatment is useful for restoring retinal function and rescuing morphology in the rd12 mouse model, and the early-onset and rapid cone degeneration can be delayed before gene therapy.