Restoration of vision in the pde6β-deficient dog, a large animal model of rod-cone dystrophy

Abstract

Defects in the β subunit of rod cGMP phosphodiesterase 6 (PDE6β) are associated with autosomal recessive retinitis pigmentosa (RP), a childhood blinding disease with early retinal degeneration and vision loss. To date, there is no treatment for this pathology. The aim of this preclinical study was to test recombinant adeno-associated virus (AAV)-mediated gene addition therapy in the rod-cone dysplasia type 1 (rcd1) dog, a large animal model of naturally occurring PDE6β deficiency that strongly resembles the human pathology. A total of eight rcd1 dogs were injected subretinally with AAV2/5RK.cpde6β (n = 4) or AAV2/8RK.cpde6β (n = 4). In vivo and post-mortem morphological analysis showed a significant preservation of the retinal structure in transduced areas of both AAV2/5RK.cpde6β- and AAV2/8RK.cpde6β-treated retinas. Moreover, substantial rod-derived electroretinography (ERG) signals were recorded as soon as 1 month postinjection (35% of normal eyes) and remained stable for at least 18 months (the duration of the study) in treated eyes. Rod-responses were undetectable in untreated contralateral eyes. Most importantly, dim-light vision was restored in all treated rcd1 dogs. These results demonstrate for the first time that gene therapy effectively restores long-term retinal function and vision in a large animal model of autosomal recessive rod-cone dystrophy, and provide great promise for human treatment.

Figures

Figure 1

Schematic structure of recombinant adeno-associated virus (AAV) vectors. AAV2/5RK.cpde6β and AAV2/8RK.cpde6β vectors encode the canine pde6β cDNA under the control of a human RK promoter (–112 bp to +87 bp region of the proximal promoter). BGHpA, bovine growth hormone polyadenylation signal; cpde6β, canine pde6β cDNA; ITR, inverted terminal repeats from AAV2; RK, human rhodopsin kinase promoter.

Figure 2

Detection of transgene (wild-type) and endogenous (mutant) pde6β transcripts in A5- and A9-treated retinas at 4 mpi. (a) Reverse transcription-PCR (RT-PCR) amplification of retinal cDNA using 2056F- and 2561R-specific primers generates a 405 bp product encompassing a portion of exons 21 and 22 of transgene and native pde6β transcripts. Mismatched 2561R specifically creates a unique BtsCI restriction site in PCR products arising from pde6β transgene transcripts, dividing the 405 bp product into 385 and 20 bp fragments. This BtsCI restriction site is not present in PCR products derived from mutant RNA templates. Arrows indicate the positions of the 2056F and 2561R primers used in RT-PCR. (b) Agarose gel electrophoresis of RT-PCR products after complete BtsCI digestion. Lines 1 and 2 show products from Pde6β+/+ and Pde6β+/− (NA2) untreated control retinas, respectively. Lines 3 and 4 show products from untreated and AAV2/5RK.cpde6β-treated retinas of dog A5 at 4 mpi, respectively. Line 5 shows the AAV2/8RK.cpde6β-treated right retina of dog A9. +/+, Pde6β+/+ retina; AAV, adeno-associated virus; bp, base pairs; ladder, 1 kb molecular size ladder; T, treated retina; U, untreated retina.

Figure 3

In vivo assessment of retinal morphology in dogs A3 and A7 at 4 and 18 mpi. (a) Fundus photographs and retinal cross-sectional images obtained from control nonaffected, untreated dog NA1 at 18 months of age. (b) Fundus photographs and retinal cross-sectional images obtained from dog A3 treated with AAV2/5RK.cpde6β at 4 and 18 mpi. (c) Fundus photographs and retinal cross-sectional images obtained from A7 treated with AAV2/8RK.cpde6β at 4 and 18 mpi. Dark circles on fundus photographs schematically represent areas of the treated retinas exposed to recombinant adeno-associated virus (rAAV) vectors. Optical coherence tomography (OCT) scans were acquired on a horizontal line shown on the fundus images (dark-line). The localization and the size of the dark-lines represent the localization and the size of the OCT scans. Retinal thicknesses at the same location were measured using calibrated calipers and indicated on the OCT scan. mpi, months postinjection; µm, micrometers.

Figure 4

Post-mortem assessment of retinal morphology in dog A5 at 4 months postinjection. (a–c) Nasal retinal cryosections from the treated and untreated eyes of dog A5 at 4 months post subretinal delivery of rAAV2/5RK.cpde6β in the nasal superior retina. (a) Wide A5 retinal section displaying vector exposed and unexposed-areas. Retinal layers remained intact while the choroid has been partially detached from the retina during the embedding process. (b) Fundus photograph representing the A5 retina exposed to the rAAV2/5 vector (dark circle) and the localization of the wide retinal section (red line). (c) Nasal retinal cryosections from unaffected, untreated dog NA2 at 5 months of age and from the untreated and treated eyes of dog A5. Serial retinal cryosections were processed for hematoxylin and eosin coloration (top) and for immunohistochemistry using antibodies against rod PDE6 (middle) or GNAT1 (bottom). Primary antibodies were detected with Alexa 488-conjugated goat anti-rabbit IgG (green). Cell nuclei were counterstained with DAPI (blue). Bar = 10 µm. Vertical dark lines indicate ONL thickness (mean ± SEM, n = 10). GC, ganglion cells; INL, inner nuclear layer; IS, inner segments; ONL, outer nuclear layer; OS, outer segments; rAAV, recombinant adeno-associated virus; RPE, outer retinal pigment.

Figure 5

Bilateral full-field electroretinographic traces from dogs NA1, A1, A2, and A6 at 1, 9, and 18 months following subretinal injection. (a) Electroretinographic trace from control nonaffected, untreated dog NA1 at 18 months of age. (b) Electroretinographic traces from control affected, untreated dog A1 at 1 and 18 months of age. (c) Electroretinographic traces from dog A2 treated with AAV2/5RK.cpde6β. (d) Electroretinographic traces from dog A6 treated with AAV2/8RK.cpde6β. The top two recordings are low- and high-intensity dark-adapted responses, whereas the bottom two recordings show light-adapted responses (responses to single flash and 30 Hz flicker stimuli, respectively). AAV, adeno-associated virus; mpi, month(s) postinjection; T, treated eye; U, untreated eye.

Figure 6

Kinetics of retinal function recovery in treated dogs A2 and A6. (a) Amplitudes of electroretinography (ERG) responses for control affected, untreated dog A1 from 1 to 18 months of age. (b) Amplitudes of ERG responses for dog A2 treated with AAV2/5RK.cpde6β from 1 to 18 mpi. (c) Amplitudes of ERG responses for dog A6 treated with AAV2/8RK.cpde6β from 1 to 18 mpi. The left and middle panels show scotopic rod and mixed cone-rod-mediated b-wave amplitudes, respectively. The right panel shows photopic 30 Hz flicker amplitude. Right eyes are shown in dark, left eyes in white. AAV, adeno-associated virus; mpi, month(s) postinjection.