IMPROVED THERAPEUTIC METHOD FOR RARE OCULAR DISEASES BY GENE REPLACEMENT

Abstract

A recombinant adeno-associated virus (AAV) vector that carries a nucleic acid sequence encoding the retinal transcription factor cone-rod homeobox (CRX) for its use in treating CRX-associated IRDs in a subject in need thereof or for use in treating inherited retinal dystrophies caused by hypomorphic mutations in the CRX target genes.

Claims

1-13. (canceled)

14. A method for treating a CRX-associated inherited retinal disease (IRD) in a subject in need thereof, comprising administering to the subject a recombinant adeno-associated virus (AAV) vector comprising a polynucleotide encoding the retinal transcription factor cone-rod homeobox (CRX).

15. The method according to claim 14, wherein the AAV vector is an AAV2 serotype.

16. The method according to claim 15, wherein the AAV vector is an AAV2/5 or AAV2/8 serotype.

17. The method according to claim 14, wherein the polynucleotide is under the control of a promoter that drives expression in rod and cone photoreceptors chosen among Rhodopsin (Rho), beta-phosphodiesterase (PDE), retinitis pigmentosa 1 (RP1) or the human Rhodopsin kinase 1 (GRK1) promoters.

18. The method according to claim 17, wherein the polynucleotide is under the control of the GRK1 promoter.

19. The method according to claim 14, wherein the CRX-associated IRD is chosen among a retinopathy resulting from a dominant mutation in the CRX gene or results in symptoms due to a hypomorphic mutation cured by the expression of CRX.

20. The method according to claim 19, wherein the CRX-associated IRD is chosen among retinitis pigmentosa, Leber's congenital amaurosis or cone-rod dystrophies or to hypomorphic mutations in CRX target genes such as PDE6B, NMNAT1, ARL13b, AIPL1 or ABCA4 genes.

21. The method according to claim 14, wherein the AAV vector is administered to the subject in a therapeutically effective amount.

22. The method according to claim 21, wherein the quantity of the vector is between 10.sup.8 and 10.sup.12 vg/eye.

23. The method according to claim 22, wherein the quantity of the vector is between 1×10.sup.9 and 1×10.sup.12 vg/eye.

24. The method according to claim 14, wherein the vector is administered before disease onset.

25. The method according to claim 14, wherein the vector is administered after initiation of photoreceptor degeneration.

26. The method according to claim 14, wherein the vector is administered as long as there are functional cone and/or rod photoreceptors.

27. The method according to claim 14, wherein the vector is included in a composition.

28. The method according to claim 27, wherein the vector is included in a pharmaceutical composition.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0086] FIG. 1: 822-pKL.AAV.GRK1.hCRX final vector plasmid map comprising the human Rhodopsin kinase 1 promoter driving the expression of human CRX. The promoter was subcloned using the restriction sites SpeI and EcoRI. The cDNA coding for human CRX was subcloned using KpnI digestion generating a cohesive site, which was blunted, and follow-up to XbaI digestion.

[0087] FIG. 2: The selected AAV vector combined with the selected promoter allows efficient transduction of WT photoreceptors and Crx.sup.Rip/+ immature cone-like photoreceptors.

[0088] WT or Crx.sup.Rip/+ mice received a single subretinal injection of 2.5×10.sup.10 AV-GFP at P30. Eyes were harvested 14 days after for sectioning and immunolabelled with an anti-GFP antibody. Nuclei were stained with DAPI.

[0089] ONL: Outer Nuclear Layer; INL: Inner Nuclear Layer; GCL: Ganglion Cell Layer. Scale bar, 20 μm.

[0090] FIG. 3: Overexpression of CRX following subretinal injection of AAV-CRX in Crx.sup.Rip/+.

[0091] Crx.sup.Rip/+ mice received or not a single subretinal injection of three different doses of AAV-CRX (0.5×10.sup.10, 1×10.sup.10 and 2.5×10.sup.10 vg per eye) at P30. CRX expression was assessed 14 days after by immunoblotting.

[0092] FIG. 4: Quantification of CRX expression following subretinal injection of AAV-GFP or AAV-CRX.

[0093] Crx.sup.Rip/+ mice received a single subretinal injection of three different doses of AAV-GFP or AAV-CRX at 0.5×10.sup.10, 1×10.sup.10 or 2.5×10.sup.10 vg per eye between P30 and P40. CRX expression was quantified 14 days after by immunoblotting. Expression was normalized using Tubulin. All quantifications were reported to endogenous CRX expression in animals injected with 0.5.Math.10.sup.10 vg of AAV-GFP.

[0094] Black bars: endogenous CRX expression in Crx.sup.Rip/+ injected AAV-GFP; white bars: CRX expression in Crx.sup.Rip/+ mice injected AAV-CRX.

[0095] FIG. 5: Differentiated PR can be observed in Crx.sup.Rip/+ retina using AAV-CRX vector.

[0096] Crx.sup.Rip/+ mice received a single subretinal injection of 2.5×10.sup.10 AAV-CRX at P30. The other eye was not injected and used as control. Eyes were harvested 1 months after for sectioning and labelled with anti-Rhodopsin and anti-cone arrestin antibodies. Nuclei were stained with DAPI.

[0097] ONL, outer nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer Scale bar, 20 μm.

[0098] FIG. 6: Differentiated PR with outer segment can be observed in Crx.sup.Rip/+ retina using a lower dose of AAV-CRX vector (0.5.Math.10.sup.19 vg).

[0099] (A) Both eyes of Crx.sup.Rip/+ mice received a subretinal injection of 0.5×10.sup.10 vg of either AAV-CRX or AAV-GFP at P30. Eyes were harvested 2 months after and flat mounted after labelling with anti-Rhodopsin and anti-cone arrestin antibodies. GFP expression following AAV-GFP subretinal injection was directly observed. Scale bar, 20 μm. (B) Crx.sup.Rip/+ mice received a single subretinal injection of 0.5×10.sup.10 AAV-CRX at P30, as well as a small dose (<1.Math.10.sup.9 vg) of AAV-GFP to sparsely labelled transduced photoreceptors with GFP and visualize their morphology. Confocal acquisition of multiple z sections allowed 3D reconstruction.

[0100] 1, outer segment; 2, nucleus; 3, synaptic terminal Scale bar, 5 μm.

[0101] FIG. 7: AAV-CRX subretinal injection allowed restoration of photoreceptor light sensitivity.

[0102] Crx.sup.Rip/+ mice received or a single subretinal injection of 0.5×10.sup.10 AAV-CRX or AAV-GFP at P30 in both eyes. Light response was assessed 4 months after injection. (A) The increase of scotopic b-wave response was statistically significant in AAV-CRX treated animals compared to AAV-GFP controls at the stimulus intensity of 2.2 log cd.Math.sec/m.sup.2. (B) The increase of photopic b-wave response was statistically significant in AAV-CRX treated animals compared to AAV-GFP controls at the stimulus intensity of 1.6 log cd.Math.sec/m.sup.2. (C) The light/dark choice test showed that the number of time than Crx.sup.Rip/+ mice injected with AAV-CRX crossed the line toward the light compartment is comparable to non-injected WT mice. In contrast, Crx.sup.Rip/+ mice injected with AAV-GFP crossed the line more often indicating a default in their ability to detect the light.

[0103] Grey bar, non-injected WT; black bars, Crx.sup.Rip/+ mice injected AAV-GFP; white bars, Crx.sup.Rip/+ mice injected AAV-CRX *p<0.05, **p<0.01.

[0104] FIG. 8: Photoreceptor degeneration was reduced in rd10 mice following AAV-CRX vector injection at the dose of 2.5.Math.10.sup.10 vg.

[0105] (A) AAV-CRX vector was injected or not in P14 Rd10 mice. Retinas were collected two months after injection. Retinal sections were immunolabelled with anti-Rhodopsin (Rho) and anti-cone arrestin antibodies (CA). Nuclei were stained with DAPI.

[0106] (B) Histograms represent the measurement of rd10 ONL thickness injected with AAV-CRX (white bar) or not (NI, black bar).

[0107] ONL, outer nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer. Scale bar, 20 μm.

[0108] FIG. 9: Photoreceptor degeneration was reduced in rd10 mice following AAV-CRX vector injection at the dose of 0.5.Math.10.sup.10 and 1.Math.10.sup.10 vg.

[0109] Histograms represent the measurement of rd10 ONL thickness at 2 months after injection at P14 with AAV-CRX. The specific detection of human CRX on paraffin sections allowed the discrimination of a transduced or non-transduced area.

[0110] Black bars, AAV-CRX non-transduced area; white bars, AAV-CRX transduced area.

[0111] FIG. 10: Expression of CRX.sup.R41W in a cone-only retina led to cone degeneration which can be prevented using AAV-CRX.

[0112] (A) Histogram representing the decreased photopic b-wave amplitude in 2-months old Nrl.sup.−/−; Tg(cRX.sup.R41W) compared to Nrl.sup.−/− at the light stimulus of 2.8 log cd.Math.sec/m.sup.2. (B) Histogram representing the preservation two months post-injection of the photopic b-wave amplitude in Nrl.sup.−/−; Tg(CRX.sup.R41W) treated at 2-months with AAV-CRX compared to AAV-GFP. *p<0.05.

EXAMPLES

[0113] The present invention is further illustrated by the following examples.

Example 1

[0114] Material and Methods Animals

[0115] Male and female adult (3-8 weeks old) C57BL/6J, Crx.sup.Rip/+ and Tg(CRX.sup.R41W) mices, and male and female young (2 weeks old) rd10 mices were produced in the laboratory animal facilities (Animalerie central-campus CNRS Gif sur Yvette).

[0116] All animal experiments were carried out according to European guidelines for the care and use of experimental animals, and were approved by the regional ethics committee (CEEA59).

Vectors

[0117] rAAV2/5 vectors expressing green fluorescent protein (GFP) or human cone-rod homeobox protein (CRX) under the control of the human rhodopsin kinase (GRK1) promotor were produced.

1. Vector Plasmid Cloning (822-pKL.AAV.GRK1.hCRX Vector #6556 Batch)

[0118] The vector plasmid was constructed in 2 phases as described below: [0119] First, the promoter GRK1 was extracted from the 754-pAAV.GRK1.eGFP.wpre.bGH by double digestion SpeI (position 185) and EcoR1 (position 427) and inserted in the 778-pKL.AAV.CAG.bGH vector plasmid in advance double digested by SpeI (position 188)-EcoR1 (position 1947) to extract the CAG promoter and allow to the intermediate plasmid pKL.AAV.bGH. The GRK1 promoter was inserted in the pKL.AAV.bGH generating the 821-pKLAAV.GRK1.bGH backbone plasmid. [0120] Second step: the 821-pKLAAV.GRK1.bGH backbone plasmid was linearized by double digestion EcoR1 (position 531)/XbaI (position 569). The hCRX gene was extracted from the pcDNA4c.hCRX plasmid by KpnI digestion (position 1171), generating a cohesive site, which was blunted, and follow-up to XbaI digestion (position 2120). The extracted hCRX gene was inserted in the linearized pKL.AAV.hCRX.bGH at position 531 to generate the 822-pKL.AAV.GRK1.hCRX final vector plasmid (FIG. 1). [0121] The 822-pKL.AAV.GRK1.hCRX final vector plasmid has been sequencing from position 106 to 2745 bp corresponding to the expression cassette.

2. AAV2/5 Manufacturing and Purification Operating Procedure

Cell Amplification, Transfection, Harvest and Supernatant PEG-Precipitation

[0122] The HEK293 cells seeding in CellStack Cells 5 chambers with vent caps (CS5) are cultured with DMEM supplemented with 10% FBS and 1% Pen/Strep. At approx. 80% of confluency, cells are co-transfected with the vector plasmid and helper plasmid (containing helper genes from adenovirus (E2A, E4 and VA RNA) and the rep cap genes corresponding to the serotype-2 for rep and capsid serotype-5) using the CaPO.sub.4 precipitate technique. The culture medium is removed from the CS5 and exchanged with the transfection medium; the cells are then incubated 6 to 15 hours at 37+/−1° C. and 5+/−1% CO.sub.2. The transfection medium is then removed from the CS5 and replaced by fresh exchange medium (DMEM, 1% Pen/Strep) prior to a 3 days incubation at 37+/−1° C. and 5+/−1% CO.sub.2. The cells of the CS5 transfected are then harvested. The supernatant is only precipitated at 5+/−3° C. overnight with PEG. The precipitated supernatant is then centrifuged. The supernatant is discarded and the PEG-pellet is resuspended in TBS before benzonase digestion.

Vector Purification by Double CsCl-Gradient Ultracentrifugation

[0123] The viral suspension is centrifuged and the vector-containing supernatant is loaded on a step density CsCl gradient in Ultra-Clear tube for SW28 rotor. The gradient is centrifuged at 28 000 rpm for 24 hours at 15° C. The full particles band is collected and transferred to a new Ultra-Clear tube for SW41 rotor. The 2.sup.nd gradient is centrifuged at 38 000 rpm for 48 hours. The enriched-full particles band is then collected. The viral suspension is then subjected to 4 successive rounds of dialysis in a Slide-a Lyzer cassette against buffer solution for ophthalmic preparation. The purified vector is finally collected, sampled for vg titer and purity assay, and stored at <−70° C. in polypropylene low-binding cryovials.

3. Quality Control: Vector Genome Quantification

[0124] The basic unit to determine the vector dose is vector genome (vg). The vector genome corresponds to the concentration of the particles containing the gene of interest and is quantified by quantitative PCR using ITR-2 specific primers, described in D'Costa et al., 2016 paper.

In Vivo Sub-Retinal AAV Injections

[0125] Adult/young mice were anesthetized with intraperitoneal injection, using 50 μL/10 μg per mice of a solution composed of xylazine (1 mg/mL) and ketamine (10 mg/mL) in phosphate-buffered saline (PBS). Iris were dilated using specific eye drops (Mydriaticum (Tropicamide 5 mg 0.5%) and Neosynephrine (Phenylephrine hydrochloride 2.5%)). Eyes were anesthetizing using Cebesine 0.4%.

[0126] Once anesthetized, C57BL/6J, rd10 and Crx.sup.Rip/+ mice were injected into the subretinal space with: [0127] AAV2/5.hGRK1.GFP (AAV-GFP) at: [0128] 5×10.sup.9 vg per eye (n=20 Crx.sup.Rip/+ mice; n=3 rd10 mice; n=2 Nrl.sup.−/− Tg(CRX.sup.R41W) [0129] 1×10.sup.10 vg per eye (n=8 Crx.sup.Rip/+ mice, n=2 rd10 mice) [0130] 2.5×10.sup.10 vg per eye (n=3 C57BL/6J mice; n=10 Crx.sup.Rip/+ mice; n=2 rd10 mice) or [0131] AAV2/5.hGRK1.hCRX (AAV-CRX) at: [0132] b 5×10.sup.9 vg per eye (n=3 C57BL/6J mice; n=21 Crx.sup.Rip/+ mice; n=3 rd10 mice; n=2 Nrl.sup.−/−) [0133] 1×10.sup.10 vg per eye (n=9 Crx.sup.Rip/+ mice; n=4 rd10 mice) [0134] 2.5×10.sup.10 vg per eye (n=6 C57BL/6J mice; n=10 Crx.sup.Rip/+ mice; n=14 rd10 mice)

[0135] A 33 G beveled needle attached to a 5 μL Hamilton syringe was inserted into the ventral subretinal space passing through the RPE/choroid membrane, near to the optic nerve. 2 μL were slowly injected to detach the retina from RPE cells (during approximately 30 seconds).

Protein Extraction and Immunoblot Analysis

[0136] Mice were euthanized using cervical dislocation. Retinas were dissected and frozen in dry ice, and then stored at −80° C. Proteins were extracted using a lysis buffer (20 mM Na.sub.2HPO.sub.4, 250 mM NaCl, 5% DTT, 30 mM NaPPi, 0.1% NP-40, 5 mM EDTA) with protease inhibitors (cOmplete™). Mechanical disruption was achieved by sonication and proteins were isolated after centrifugation (10 minutes, 13200 rpm, 4° C.). 20 μg of protein from each eye studied were loaded on a 7.5%-12% gradient acrylamide gel (Biorad). The run was carried out at constant voltage (100 V). Upon completion, western blot transfer was carried out using an iBlot 2 Dry Blotting System on nitrocellulose membrane (ThermoFisher Scientific).

[0137] After transfer, the nitrocellulose membrane was incubated with blocking buffer (PB ST [PBS, 0.05% Triton X-100], 5% milk) for 1 hour at room temperature. Primary antibody was then added in fresh blocking buffer, either overnight at 4° C. (mouse anti-Crx (A-9), ref.: sc-377138 [Santa Cruz Biotechnology, Inc.], batches J1116 and H2918, dilution 1/5000) or for 1 hour at room temperature (mouse anti-tubulin clone DM1A, ref.: T9026 [Sigma-Aldrich], batch 052M4837, dilution 1/10000). Excess of primary antibody was rinsed 3 times with PBST for 10 minutes.

[0138] Secondary antibody was added in fresh blocking buffer for 2 hours at room temperature (goat anti-mouse IgG-peroxidase, ref.: A4416 [Sigma-Aldrich], batch SLBH3692, dilution 1/5000), and then rinsed 3 times for 10 minutes with PBST.

[0139] ECL revelation kit (Supersignal™ West Dura Extended Duration Substrate: 34076 [ThermoFisher Scientific]; batch SK256986) was added on the nitrocellulose membrane 5 minutes before revelation in dark room (Carestream® Kodak® BioMax® light film, ref.: Z373508 [Sigma-Aldrich]).

[0140] Membrane can be stripped with stripping buffer (1.5 g glycine, 1 mL SDS 10%, 1 mL Tween-20, milliQ water qs. 100 mL, pH 2.2). As such, the membrane was incubated twice for 10 minutes at room temperature, followed by two PBS washes for 10 minutes, and finally washed twice with PBST for 5 minutes. Membranes were incubated with primary and secondary antibodies as previously described. After revelation, films were scanned and analyzed with Image J.

Histological Analysis

[0141] Mice were euthanized using cervical dislocation. Eyes were collected and fixed in 4% paraformaldehyde (PFA) for 1 hour, rinsed in PBS before dehydratation and paraffin inclusion. 7 μm-thick sections were cut with a microtome (Rotatory microtom, ref.: HM 340E [ThermoFisher Scientific]), placed overnight at 37° C., and then stored at room temperature. Prior to immunohistochemistry (IHC), slides were deparaffinized and then incubated in hot citrate buffer (pH 6, 0.1 M) for 20 minutes at 500 W. Once cooled down, slides were placed in PBS. For immunostaining (IHC) on flat mount retina, the retina was dissected and fixed in 4% PFA for 1 hour, rinsed in PBS and incubated directly with the primary antibody.

[0142] Primary antibodies in Dako REAL™ Antibody diluent (ref 52022 lot 20060091) supplemented with 0.3% Triton X-100 were added to the slides overnight at 4° C. in a dry chamber (mouse anti-rhodopsin clone 4D2, ref.: MABN15 [Merck], batch 2935495, dilution 1/1000; rabbit anti-cone arrestin, ref.: AB15282 [Merck], batch 2802590, dilution 1/1000; goat anti-GFP, ref.: ab6673 [Abcam], dilution 1/1000). Flat mount retinas, were incubated with the primary antibodies for 2 days.

[0143] They were rinsed 3 times with PBST, 5 minutes before adding secondary antibodies (donkey anti-mouse IgG Alexa Fluor 555, ref.: A31570 [ThermoFisher Scientific], dilution 1/1000; donkey anti-rabbit IgG Alexa Fluor 488, ref.: A21206 [ThermoFisher Scientific], dilution 1/1000; donkey anti-goat Alexa Fluor 488, ref.: A11055 [ThermoFisher Scientific], dilution 1/1000) in Dako REAL™ Antibody diluent (ref S2022 lot 20060091) supplemented with 0.3% Triton X-100. After 2 hours of incubation at room temperature in the dark, slides were rinsed 3 times with PBST for 5 minutes, and DAPI (ref.: 62248 [ThermoFisher Scientific], batch RJ2279362, dilution 1/1000) was added for 20 minutes in PBST. Finally, slides were rinsed 3 times with PBST. Flat mount retinas were incubated for 1 night with the secondary antibody.

[0144] For flat mount retina, the retina was positioned on the slide with photoreceptor side up. FluorSave™ Reagent (ref.: 345789, Millipore, batch 3034632) was used to add a coverslip. Slides were stored at 4° C. Pictures were acquired using an Imager M2 microscope (Zeiss) or a LSM710 confocal microscope (Zeiss) and analyzed using Zen software and Image J.

Electroretinogram

[0145] Electroretinogram (ERG) recordings were performed using a focal ERG module attached to Micron IV (Phoenix Research Laboratory). Briefly, mice were dark-adapted overnight and prepared for the experiment under dim-red light. The mice were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg) and received topical proparacaine hydrochloride (0.5%, Alcon) via eye drops. Pupils were dilated with tropicamide (1%, Alcon) and phenylephrine (2.5%, Alcon) and lightly coated with GONAK hypromellose ophthalmic demulcent solution (2.5%, Akorn). Lens of the Micron IV was placed directly on the cornea, and a reference electrode was placed on the mouse head. Scotopic responses were elicited with a series of flashes of increasing light intensities from −1.7 to 2.2 cd.Math.s/m2. Photopic responses were elicited under rod-desensitizing background light with a series of flashes of increasing light intensities from −0.5 to 2.8 cd.Math.s/m2. Values of a- and b-wave were extracted and plotted for comparisons between groups of interest.

Light-Dark Choice.

[0146] The apparatus had 20 cm-high Plexiglas walls and consisted of a brightly lit white box (40×15 cm) connected by a trap door (6×6 cm) to a dark box (15×15 cm). Illumination in the dark box was <10 Lux. In the light box, the light was placed at the end of the 40 cm compartment, thus providing a gradient of illumination increasing from 600 Lux at the entrance to 1500 Lux at end of compartment.

[0147] Testing was performed between 9 am and 12:00. Each mouse was placed in the dark compartment for 10s, the trap door was then opened and mice allowed to freely explore the whole apparatus for 5 min. Step through latency, number of entries and total time spent in the light compartment were scored.

Example 2

[0148] Selection of the promoter and the serotypes of the AAV vector to be used to efficiently transduced photoreceptors in CRX-associated retinopathy mouse models.

[0149] Several AAV serotypes and promoters have been tested and showed transduction of mature photoreceptors in adults. In the case of Crx.sup.Rip/+ retina, the outer nuclear layer contains only immature cone-like photoreceptors and therefore AAV vectors commonly used to treat differentiated photoreceptors may not be as efficient. Therefore, we tested several serotypes expressing GFP driven by a photoreceptor-specific promoter, which directs gene expression in both rod and cone photoreceptors. The human Rhodopsin kinase 1 (GRK1) promoter was selected because i) it is already well characterized to express genes in both rod and cone photoreceptors ii) it is already used in clinical trial (PDE6B) iii) It is active in our mouse models based on our published transcriptomic data. For the serotype, we selected the AAV2/5 vector described as transducing specifically mouse photoreceptors.

[0150] Different doses of GFP-expressing AAV vectors, ranging from 1×10.sup.9 to 5×10.sup.10 vg per Crx.sup.Rip/+ eye, were injected into the subretinal space at post-natal day 30 (P30). The quality of the subretinal injection was confirmed immediately after by the presence of a retinal detachment observed by Optical Coherence Tomography (OCT). Two weeks later, retinal detachment disappeared and the efficacy of photoreceptor transduction was assessed using fluorescent fundus imaging followed by histological analysis. A GFP immunofluorescence analysis performed on sections from injected retina revealed a strong GFP expression in the injected area restricted to the outer nuclear layer were photoreceptors are located (FIG. 2).

[0151] In conclusion, the AAV vector serotype selected for future therapy is an AAV2/5 vector including a GRK1 promoter.

Example 3

[0152] Quantification of CRX expression driven by GRK1 promoter using an AAV2/5 vector.

[0153] In order to verify that the product (CRX-expressing AAV vector or AAV-CRX) is able to produce CRX protein, a series of subretinal injections was performed in Crx.sup.Rip/+ retina with three different doses (0.5.Math.10.sup.10, 1.Math.10.sup.10 and 2.5.Math.10.sup.10 vg) and CRX expression was assessed 14 days later. At all three tested doses, increased expression of CRX was observed. The amount produced was more dependent of the injection than the dose used (FIG. 3). This experiment was repeated when the subretinal injection technic was fully mastered but this time GFP-expressing vector (AAV-GFP) was injected as control (FIG. 4) at three different doses (0.5.Math.10.sup.10, 1.Math.10.sup.10 and 2.5.Math.10.sup.10 vg). CRX relative expression was quantified after tubulin normalization and reported to endogenous CRX expression following AAV-GFP injection at 0.5.Math.10.sup.10 vg. CRX expression was clearly significantly increased after AAV-CRX injection proportionally to the dose used.

Example 4

[0154] Restoration of photoreceptor differentiation in Crx.sup.Rip/+ retina following AAV-CRX injection.

[0155] To assess the ability of the product (CRX-expressing AAV vector or AAV-CRX) to restore photoreceptor differentiation, AAV-CRX was injected at 2.5×10.sup.10 vg per eye at P30. Retinas were collected two months after for histological analysis Immunohistochemistry analysis was performed using anti-rhodopsin and anti-cone arrestin to label rod and cone photoreceptors, respectively. The result clearly revealed a large area of Rhodopsin- and Cone Arrestin-positive cells around the injected area (FIG. 5). Imaging of Rhodopsin- and Cone-Arrestin-positive cells on flat mount retina four months after AAV-CRX subretinal injection in Crx.sup.Rip/+ demonstrated fully differentiated rod and cone photoreceptors at the dose of 0.5.Math.10..sup.10 vg (FIG. 6A).

[0156] In contrast, AAV-GFP injected eyes did not show positive cells for these markers. To better visualize the morphology of the differentiated photoreceptors four months after AAV-CRX treatment, a low amount of AAV-GFP (<1.Math.10.sup.9) was co-injected to obtain a presence of sparse GFP-positive cells easier to image. Higher magnification of GFP-positive photoreceptors on flat-mount retina transduced with AAV-CRX demonstrated the presence of well-formed outer segments, as well as synaptic terminus (FIG. 6B).

Example 5

[0157] Partial functional recovery in Crx.sup.Rip/+ retina following AAV-CRX injection.

[0158] In order to assess the functionality of the newly differentiated rod and cone photoreceptors following AAV-CRX treatment in Crx.sup.Rip/+, scotopic and photopic ERG recording were performed. a statistically significant increase of the scotopic B-wave amplitude at the maximum stimulus of 2.2 Log cd sec/m.sup.2 (FIG. 7A) was recorded. For the photopic response, it was overall slightly improved but statistically significant only at the stimulus of 1.6 Log cd sec/m.sup.2 compared to AAV-GFP treated animals (FIG. 7B). To estimate the ability of the Crx.sup.Rip/+ newly differentiated photoreceptors to transduce a visual signal to the brain area dedicated to vision, a behavior named “Light/Dark choice” was used. WT mice spent a vast majority of their time in the dark side of the box whereas untreated Crx.sup.Rip/+ mice will go in both side indifferently due to their complete blindness. In contrast, AAV-CRX treated Crx.sup.Rip/+ mice spent more time in the dark side by crossing significantly less the line leading to the bright-side of the box (FIG. 7C). Altogether, these results demonstrated the ability of the AAV-CRX treatment to trigger Crx.sup.Rip/+ photoreceptor differentiation as well as a partial functional recovery, especially by the ability of the AAV-CRX treated mice to detect light.

Example 6

[0159] AAV-CRX protects rd10 photoreceptors from cell death.

[0160] A potential interest of overexpressing CRX.sup.WT is to stimulate the expression of its target genes. Among them, many encodes enzymes associated with IRD due to hypomorphic mutations leading to lower expression or to a low activity such as certain subtype of PDE6B mutations. Enhancing CRX target gene expression by overexpressing CRX.sup.WT will allow to reach a certain threshold of expression and/or activity allowing proper phototransduction and long-term PR maintenance. This was done using the rd10 mice carrying a hypomorphic recessive mutation in Pde6b. Of note, retinal degeneration starts around P15 in this mouse model and by P30, most of the rods are lost. A sufficient increased expression level allows reaching a certain threshold of enzymatic activity sufficient to reduce cGMP levels and therefore restore the signal transduction cascade. As such, the therapy product of the invention will have a broader spectrum of application with more patients potentially treatable.

[0161] To assess the potential therapeutic effect of AAV-CRX vector for hypomorphic mutations, subretinal injection in rd10 mice at P14 (2.5×10.sup.10 vg) was performed and the retina were harvested 1 month after. IHC analysis revealed good preservation of the ONL in the treated eyes compared to controls. In contrast to non-injected eyes where outer segments are dramatically reduced, rod and cone PRs displayed well preserved outer segments where Rhodopsin and Cone Arrestin are localized such as in healthy retina (FIG. 8 (A)). Measurement of the ONL thickness confirmed the preservation (FIG. 8 (B)). In addition, the efficacy 1 months after injection at P14 with two additional doses of AAV-CRX: 0.5.Math.10.sup.10 vg and 1.Math.10.sup.10 vg was tested. Using paraffin section, the transduced area was determined using an anti-CRX recognizing specifically the human CRX produced by the AAV and not the endogenous mouse Crx. Measurement of the ONL thickness demonstrated that the dose of 1.Math.10.sup.10 vg led to a better preservation of the ONL compared to the non-transduced area (FIG. 9). These results clearly demonstrated the therapeutic value of the use of the product of the invention for certain subtypes of IRD.

[0162] However due to the kinetic of degeneration starting at P14 and the time of injection, AAV-CRX injection has to be done earlier in rd10 life to improve the efficacy assessment of the therapy by giving more time for CRX production by the AAV before the onset of the degeneration.

[0163] The correct expression of the human wild-type CRX cDNA specifically in photoreceptors was demonstrated. Preliminary subretinal injections of AAV-CRX in one-month-old Crx.sup.Rip/+ mice demonstrated i) the absence of toxicity at the dose of 0.5.Math.10.sup.10 vg, ii) increased CRX expression levels, and iii) reinitiation of PR differentiation four months post-injection with a large area of Rhodopsin- and Cone Arrestin-positive cells compared to non-treated mice. Notably, the gene therapy product of the invention leads to a much higher differentiation of cones compared to rod photoreceptors. Finally, a slight recovery of the scotopic and photopic electroretinogram (ERG) response 4-months after subretinal injection was observed. Together, these very promising results clearly demonstrate the validity of the product of the invention to restore PR differentiation in adult Crx.sup.Rip/+ retina.

Example 6

[0164] A second mouse model was established based on the identification of two families in France each with a CORD proband carrying the CRX.sup.R41W mutation. This dominant mutation does not alter CRX nuclear localization but reduces its DNA-binding properties leading to a decreased transcriptional activity of its target genes. This mutation may lead to a dominant-negative effect or to a loss-of-function preventing CRX.sup.R41W from binding to target promoters. Therefore, a transgenic line, Tg(CRX.sup.R41W) has been generated using a lentiviral approach (collaboration P. Charneau and L. Vives, Institut Pasteur, Paris). This line carries the recurrent human CRX.sup.R41W mutation fused with a myc tag under the control of the Crx promoter. A first group of Tg(CRX.sup.R41W) founders with multiple copies of the transgene showed loss of scotopic and photopic ERG responses due to photoreceptor degeneration. Successive crossing with wild-type (WT) mice to decrease the number of copies led to a partial recovery of the ERG response demonstrating the importance of the ratio between the amount of CRX.sup.WT and CRX.sup.R41W in the development of the disease. These observations clearly demonstrate the therapeutic interest of increasing the amount of CRX.sup.WT to compensate the negative effect of the mutated CRX. Three lines with one copy of the transgene CRX.sup.R41W were established wherein a normal differentiation of the photoreceptors with a slight decrease in the ERG response was found. The expression of the transgene in photoreceptors was verified by IHC.

[0165] Mice with two copies of Tg(CRX.sup.R41W) have been obtained (from the line with one copy). If the line with one copy of Tg(CRX.sup.R41W) had not defects in the ERG response, mice with two copies showed reduced photopic response at 6 months of age. Therefore, this line a good model of cone-rod dystrophy to test the present therapy efficacy although the degeneration is not massive and slow. To circumvent this issue, a line expressing this mutation on the Nrl.sup.−/− background in which all photoreceptors are cones was established. In this line, decreased photopic response at the higher light stimulus could be observed at 2 months (FIG. 10 (A)). Preliminary injection in 2 months old animals of AAV-CRX or AAV-GFP showed preservation of the photopic response in Nrl.sup.−/−;Tg(CRX.sup.R41W) treated with AAV-CRX compared to AAV-GFP treated animals 2 months after injection (FIG. 10 (B)).

Phenotypic Characterization of Photoreceptor Cells Differentiated from CRX-Mutated iPSC Lines of Patients and Assessment of the Efficacy of Gene Replacement.

[0166] As iPSC can be reprogrammed from fibroblasts of patients with IRDs, it is possible to generate disease-specific retinal cell models that represent excellent models for proof-of-concept gene therapy studies. iPSC-derived photoreceptor models were generated from three patients carrying CRX mutations causing CORD (p.Arg41Trp); LCA (p.Pro232Argfs*139) and RP (p.Asp65His). iPSC for the CORD and LCA patients were generated and confirmed their pluripotency and genetic stability. Furthermore, iPSC from a control and the CORD individual were differentiated into retinal organoids containing photoreceptors. Using qPCR analysis at different times of differentiation, similar expression pattern of early photoreceptor markers (SIX3, VXVX2, RAX, OTX2, NEUROD1, NRL) up to day 160 between the two lines was observed. CRX expression appeared similar until day 160, when a decrease in the CORD line compared to the control was detected. By contrast, a decreased NR2E3 expression was detected at all time points tested in the CORD line. This is particularly interesting as NR2E3 interacts with CRX for rod development. Furthermore, a decreased expression in the more mature cone (OPN1SW, CAR, OPSN1MW) and rod (RK, RCVRN) markers in the CORD line compared to control was detected Moreover, no rhodopsin expression could be detected at either the mRNA or protein level in the CORD line. These results show a delayed or altered photoreceptor differentiation in the case of the p.Arg41Trp/R41W mutation.

[0167] Patient-specific iPSC-derived photoreceptors in retinal organoids were transduced with AAV-CRX to increase wild type CRX expression, and the differentiation profile was monitored over time and compared to non-treated and control cells.

[0168] As such, i) a genotype-phenotype correlation was obtained and hence elucidated the basis for the differential clinical profiles associated with each mutation, and ii) it was determined whether all forms can benefit from a gene replacement strategy and whether there are toxic effects linked with CRX overexpression. These studies allow to assess, in the absence of in vivo administration challenges, the true effect of exogenous CRX expression.