Regulation of retinal function but nonrescue of vision in RPE65-deficient dogs treated with doxycycline-regulatable AAV vectors

Abstract

In previous studies, we demonstrated that recombinant adeno-associated virus (rAAV)-mediated gene transfer of the doxycycline (Dox)-regulatable system allows for the regulation of erythropoietin (EPO) expression in the retina of nonhuman primates after intravenous or oral administration of Dox. In addition, it was shown that administrating different amounts of Dox resulted in a dose-response dynamic of transgene expression. Adeno-associated viral gene therapy has raised hope for the treatment of patients with Leber congenital amaurosis, caused by mutations in the retinal pigment epithelium (RPE)-specific gene RPE65. The preliminary results of three clinical trials suggest some improvement in visual function. However, further improvements might be necessary to optimize vision recovery and this means developing vectors able to generate transgene expression at physiological levels. The purpose of this study was to investigate the ability of the Dox-regulatable system to regulate retinal function in RPE65(-/-) Briard dogs. rAAV vectors expressing RPE65 under the control of either the TetOff and TetOn Dox-regulated promoters or the cytomegalovirus (CMV) constitutive promoter were generated and administered subretinally to seven RPE65-deficient dogs. We demonstrate that the induction and deinduction of retinal function, as assessed by electroretinography (ERG), can be achieved using a Dox-regulatable system, but do not lead to any recovery of vision.

Figure 1

Structure of rAAV vectors. (a) The AAV2/4TetOff.rpe65 vector encodes the human RPE65 cDNA under the control of the Dox-inducible TetO-CMV promoter, and the tTA chimeric transactivator (tTA2) under the control of the CAG promoter (CMV enhancer/chicken β-actin promoter). (b) The AAV2/4TetOn.rpe65 vector encodes the human RPE65 cDNA under the control of the Dox-inducible TetO-CMV promoter, and the rtTA chimeric transactivator (rtTA-M2) under the control of the CAG promoter. (c) The AAV2/4CMV.rpe65 vector encodes the human RPE65 cDNA under the control of the CMV promoter.

Figure 2

Assessment of retinal morphology and evaluation of humoral immune response against the tTA2 and rtTA.M2 proteins. (a,b) Fundus photography and OCT image of the AAV2/4TetOff.rpe65 treated eye of A2 before (a) and 6 months postinjection (b). (c,d) Fundus photography and OCT image of the AAV2/4TetOn.rpe65 treated eye of A4 before (c) and 6 months postinjection (d). (e,f) Fundus photography and OCT image of the AAV2/4CMV.rpe65 treated eye of A6 before (e) and 6 months postinjection (f). The white circle on the fundus photography indicates the zone of the bleb. The white line on the fundus photography indicates the OCT scanning path. The white arrow on the OCT image indicates the retinal vessel. (g–i) Evaluation of antibody generation against tTA2 and rtTA.M2 before and at 2, 4, and 6 months postinjection in A2 (g), A4 (h), and A6 (i). bi, before injection; mpi, months postinjection; OCT, optical coherence tomography.

Figure 3

Bilateral full-field electroretinographic (ERG) recordings of AAV2/4TetOff.rpe65 and AAV2/4TetOn.rpe65 treated dogs in the ON and OFF states. (a) Affected dog A2 treated with the AAV2/4TetOff.rpe65 vector before and at different time points postinjection, in the ON₁, OFF₁, and ON₂ states. (b) Affected dog A5 treated with the AAV2/4TetOn.rpe65 vector before and at different time points postinjection, in the ON₁, OFF₁, and ON₂ states. The top two recordings are low- and high-intensity scotopic responses, whereas the bottom two recordings show photopic responses (responses to light-adapted single flash and 30 Hz flicker stimuli, respectively). bi, before injection; mpi, months postinjection; R, right eye; L, left eye.

Figure 4

Long-term regulation of retinal function in RPE65-deficient dogs following subretinal injection of the AAV2/4TetOff.rpe65 vector. Affected dogs (a) A1, (b) A2, and (c) A3 before injection and at different time points postinjection, in the ON and OFF states. The b-wave amplitudes of the ERG max. (scotopic) of the treated and untreated eyes were measured and reported in the graphs. Gray, treated eye; black, untreated contralateral eye. The ON state was obtained in the absence of Dox, whereas the OFF state was obtained upon oral administration of Dox. bi, before injection; ERG, electroretinography; mpi, months postinjection.

Figure 5

Long-term regulation of retinal function in RPE65-deficient dogs following subretinal injection of the AAV2/4TetOn.rpe65 vector. Affected dogs (a) A4 and (b) A5 before injection and at different time points postinjection, in the ON and OFF states. The b-wave amplitudes of the ERG max. (scotopic) of the treated and untreated eyes were measured and reported in the graphs. Gray, treated eye; black, untreated contralateral eye. The ON state was obtained upon oral administration of Dox, whereas the OFF state was obtained in the absence of Dox. bi, before injection; ERG, electroretinography; mpi, months postinjection.

Figure 6

Retinal function in control dogs. (a) Bilateral full-field electroretinographic recordings of control dogs. Nonaffected dog NA at 12 months of age, affected dog A7 treated with the AAV2/4RPE65.rpe65 vector at 3 months postinjection, and affected dog A6 treated with the AAV2/4CMV.rpe65 vector at 3 months postinjection. (b) Kinetics of retinal function in control dogs. Nonaffected dog NA at different time points after birth, affected dog A7 treated in the right eye with the AAV2/4RPE65.rpe65 vector, and affected dog A6 treated in the right eye with the AAV2/4CMV.rpe65 vector at different time points postinjection. The b-wave amplitudes of the ERG max. (scotopic) of both eyes were measured at different times and reported in the graphs. Gray, right eye; black, left eye. bi, before injection; ERG, electroretinography; mpi, months postinjection.