AAV2/4-RS1 gene therapy in the retinoschisin knockout mouse model of X-linked retinoschisis

Abstract

Purpose : Human gene therapy for X-linked retinoschisis (XLRS) has shown promise using intravitreal injection of adeno-associated virus AAV2/8-RS1 vector, which was originally tested in the retinoschisin knockout mouse (1). We tested the hypotheses that subretinal and intravitreal injections of a novel vector, AAV2/4-RS1, could correct photoreceptor cells in the Rs1-KO mouse detectable by electroretinography (ERG) and swim assay. AAV2/4 transduces most layers of the human retina in explants (2) and, thus, is a logical choice for Rs1 gene therapy. We also studied whether brinzolamide, which reduces cysts in humans, potentiates the effects of AAV2/4-RS1.

Methods : AAV2/4-RS1 with a CMV promoter (2x10¹² viral genomes/mL) was delivered to the right eyes of Rs1-KO mice (National Eye Institute) either via intravitreal (N= 5; 1µL) or subretinal (N=5; 2µL) injections in vivo. Six additional mice received subretinal gene therapy plus topical brinzolamide two times a day for the first two weeks after AAV2/4-RS1 injection. Injections were performed at P60-90. Full field ERG dark adapted latency/intensity protocol was performed at day 50-60 post-injection for all Rs1-KO eyes. The treated groups were compared to untreated Rs1-KO mice (N=3) and wild-type (WT) mice (N=2) using a visually guided swim assay. Synapse labeling was performed on Rs1-KO and WT mice using anti-PSD95.

Results : Compared to untreated eyes, treated eyes tended to have higher ERG amplitudes in subretinal gene therapy, subretinal gene therapy plus brinzolamide, or intravitreal gene therapy. Treated eyes had a less electronegative b-wave on dark adapted 10 cd.sec/m² flash than untreated eyes. The b/a ratio was significantly higher in the subretinal plus brinzolamide (p=0.016) and intravitreal (p=0.049) eyes compared to untreated eyes, but not the subretinal group without brinzolamide (p=0.58). There was a significantly higher b/a ratio in the subretinal plus brinzolamide group compared to those treated with subretinal gene therapy alone (p=0.025). Rs1-KO eyes had retraction of synaptic terminals into the ONL compared to WT. Swim assay showed no difference between treated and untreated mice (p=0.14) but a significant difference between WT and Rs1-KO (p=0.0003 to p=0.05).

Conclusions : In the Rs1-KO mouse, subretinal and intravitreal injections of AAV2/4-RS1 improved ERG b/a-wave ratio, and brinzolamide potentiated the effect of subretinal therapy.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.