Products and methods for use in treating NDP-related disorders

Abstract

The present invention relates to constructs and AAV particles comprising wild-type or codon-optimized NDP nucleic acid sequences. The present invention further relates to products and methods for treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases.

Description

[Technical Field]

[0001] FIELD OF THE INVENTION The present invention relates to gene therapy constructs expressing the NDP gene and AAV particles for use in gene therapy. More specifically, the present invention relates to constructs expressing a codon-optimized NDP open reading frame for the treatment of NDP-associated diseases. [Background technology]

[0002] Introduction Norrie disease (ND) is a rare, recessive, X-linked, dual sensory disorder caused by mutations in the NDP gene, resulting in congenital blindness and progressive hearing loss (Fradkin 1971; Holmes 1971; Berger et al. 1992a; Berger et al. 1992b; Chen et al. 1992). Visual impairment is caused by underdevelopment of the deep retinal vasculature, leading to retinal detachment (Apple et al. 1974; Drenser et al. 2007). Hearing is typically normal in infancy but begins to decline gradually, on average, from age 12, often beginning at specific frequency ranges (Smith et al. 2012). Some patients with Norrie disease have cognitive impairment and other neurological symptoms, or peripheral vascular disease and erectile dysfunction (Smith et al. 2012; Michaelides et al. 2004; Rehm et al. 1997; Cacao et al. 2018). There is no cure for Norrie disease. However, late onset of hearing loss provides an opportunity for early intervention to preserve hearing.

[0003] NDP encodes the secreted, soluble WNT analogue, norrin (norrin cystine knot growth factor; NDP), which binds to a receptor complex consisting of FZ-4, LRP-5 / 6, and TSPAN-12, inducing intracellular β-catenin signaling (Xu et al. 2004; Junge et al. 2009; Chang et al. 2015). This pathway is essential for deep retinal angiogenesis and blood-vessel barrier maintenance (Xu et al. 2004; Apple et al. 1974). Hearing loss in Norrin patients is traced back to the cochlea, with both the vasculature and cochlear hair cells affected (Nadol et al. 1990; Parving et al. 1978), consistent with Norrin signaling being important for the development or maintenance of these structures (Bryant et al. 2022; Hayashi et al. 2021; Rehm et al. 2002; Ye et al. 2011).

[0004] The Ndp-KO mouse model recapitulates human Norrie disease (Nadol et al. 1990; Rehm et al. 2002; Berger et al. 1996). These mice have been demonstrated to exhibit early cochlear vascular morphology and barrier malformations, as well as reduced endocochlear potentials. Subsequently, between 1 and 2 months of age, outer hair cell (OHC) degeneration in "sensitive" frequency regions and corresponding mid-frequency hearing loss occurred (Bryant et al. 2022). This sequence of events suggests that outer hair cell degeneration may be the tissue correlate of auditory dysfunction and that vascular pathology may be the primary cause of hearing loss in Norrie disease (Bryant et al. 2022).

[0005] Norrie disease is a good candidate for gene replacement therapy due to its small size (402 bp coding sequence) and simple structure (Ohlmann and Tamm 2012). Importantly, there is evidence that Norrin is secreted, does not exhibit concentration gradient effects, and does not provide directional cues, making targeting NDP expression to specific cell types unnecessary (Wang et al. 2012; Ohlmann et al. 2005). Adeno-associated virus (AAV) vectors are advantageous for clinical application due to their low immunogenicity and genotoxicity (Verdoodt et al. 2021). AAV9 has been shown to cross the blood-brain barrier (Merkel et al. 2017) and transduce a wide range of cells, including the retina and cochlea (Massaro et al. 2020; Shibata et al. 2017), and has already been approved for clinical use (Aslesh and Yokota 2022). Summary of the Invention [Problem to be solved by the invention]

[0006] The objective of the present invention is to provide an early treatment that preserves both the structure and function of the cochlea and retina. Treatment using the present invention achieves complete or partial rescue of the structure and function of the cochlea and eye. This is the first AAV-mediated gene therapy treatment for NDP-related diseases, particularly Norrie's disease, at a clinically relevant stage of disease progression. [Means for solving the problem]

[0007] Summary of the Invention The present invention relates to products and methods for the treatment of one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-related diseases, or other β-catenin signaling-related retinal diseases.

[0008] According to a first aspect of the present invention, there is provided a construct comprising a wild-type or codon-optimized NDP nucleic acid sequence.

[0009] Codon optimization is important because it allows for a reduction in viral dose, which is advantageous because it reduces the risk of uveitis when injected into the vitreous, and possibly also the risk of an immune response when injected into the cochlea.

[0010] In one embodiment, the wild-type or codon-optimized NDP nucleic acid sequence is a human NDP open reading frame.

[0011] In one embodiment, the wild-type NDP nucleic acid sequence comprises the nucleotide sequence of SEQ ID NO:1, or the codon-optimized NDP nucleic acid sequence comprises the nucleotide sequence of SEQ ID NO:3.

[0012] In one embodiment, the NDP nucleic acid sequence comprises a nucleotide sequence having at least about 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:1, or the NDP nucleic acid sequence comprises a nucleotide sequence having at least about 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:3.

[0013] In one embodiment, the wild-type NDP nucleic acid sequence or the codon-optimized NDP nucleic acid sequence is selected from the group consisting of a CAG promoter, a CBA promoter, a CMV promoter, an EF1a promoter, a PGK promoter, a TRE promoter, a U6 promoter, a UAS promoter, an EFS promoter, a SFFV promoter, an MSCV promoter, an SV40 promoter, a UBC promoter, a Pro1A promoter, a hRHO promoter, a hBEST1 promoter, a Grm6 promoter, a GJB2 promoter, a GJB6 promoter, a SLC26A4 promoter, a TECTA promoter, a DFNA5 promoter, a COCH promoter, an NDP promoter, a SYN1 promoter, a The gene is under the control of a GFP gene, a GFAP promoter, a PLP promoter, a TAK1 promoter, a SOX21 promoter, a SOX2 promoter, a FGFR3 promoter, a PROX1 promoter, a GLAST1 promoter, a LGR5 promoter, a HES1 promoter, a HES5 promoter, a NOTCH1 promoter, a JAG1 promoter, a CDKN1A promoter, a CDKN1B promoter, a SOX10 promoter, a P75 promoter, a CD44 promoter, a HEY2 promoter, a LFNG promoter, a S100b promoter, a CLDN11 promoter, a NDP promoter, or a synthetic modification or combination of these promoters.

[0014] In one embodiment, downstream of the wild-type or codon-optimized NDP nucleic acid sequence is a WPRE or oPRE element, which is advantageous for enhancing expression in the vitreous humor and reducing immunogenic effects.

[0015] In one embodiment, the construct further comprises one or more of the following elements: 5' and 3' inverted terminal repeats, a self-cleaving P2A linker, a FLAG epitope sequence for C-terminal tagging, a simian virus 40 polyA (SV40 late polyA) sequence, a bovine growth hormone polyadenylation signal (BGHpA), an EGFP open reading frame sequence, and a pUC ori.

[0016] In one embodiment, the wild-type NDP nucleic acid sequence or the codon-optimized NDP nucleic acid sequence is incorporated into one or more of the following vectors: AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, AAV-rh8, AAV-rh10, AAV-rh39, AAV-rh43, AAVAnc80, AAV 2 / ShH10, AAV2 / 9, AAV-S vector, AAV-ie, or AAV-PHP.eB.

[0017] In one embodiment, the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV 2 / ShH10 vector, an AAV 2 / 9 vector, or an AAV-S vector.

[0018] In one embodiment, the construct is formulated for at least one of intravenous, intraocular, and / or intracochlear administration.

[0019] According to a second aspect of the present invention, there is provided a construct comprising the wild-type or codon-optimized NDP nucleic acid sequence of claims 1 to 10 for use in the treatment of one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases.

[0020] According to a third aspect of the present invention there is provided an AAV particle (NDP.AAV) comprising a construct of the first aspect of the invention.

[0021] According to a fourth aspect of the invention, there is provided an AAV particle according to the third aspect of the invention for use in the treatment of one or more of Norrie's disease, age-related or diabetic macular degeneration and retinopathy, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), and other NDP-associated diseases.

[0022] In one embodiment, the AAV particles are adapted to target cochlear fibrocytes, basal and marginal cells of the lateral spiral ligament (SL) and stria vascularis, and glial cells of the modiolus, retinal Müller cells, retinal ganglion cells, retinal pigment epithelial cells and photoreceptors, retinal ganglion cells alone and / or Müller cells alone.

[0023] In one embodiment, the AAV particles according to the third aspect of the invention are administered intravenously.

[0024] In one embodiment, the AAV particles are AAV 2 / 9 particles.

[0025] In one embodiment, the AAV particles according to the third aspect of the invention are administered intraocularly.

[0026] In one embodiment, the AAV particles are AAV 2 / ShH10 particles.

[0027] In one embodiment, the AAV particles according to the third aspect of the invention are administered intracochlearly.

[0028] In one embodiment, the AAV particles according to the third aspect of the invention are administered by at least one of intravenous, intraocular, and / or intracochlear administration.

[0029] In further embodiments, NDP.AAV administration can be by any combination of intravenous, intraocular, and intracochlear administration, for example, NDP.AAV administration is by intravenous and intraocular administration, or intravenous and intracochlear administration, or intraocular and intracochlear administration, or intravenous, intraocular, and intracochlear administration.

[0030] According to a fifth aspect of the present invention, there is provided ASV particles according to the third aspect of the present invention for use in the treatment of one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-related diseases, or other beta-catenin signaling-related retinal diseases, wherein a single dose is administered intraocularly neonatally or prenatally, preferably at about 14-17 post-conception weeks (pcws), more preferably at about 15-16 pcws, or a single dose is administered intraocularly, preferably at about 22-26 post-conception weeks (pcws), more preferably at about 24 pcws, or a single dose is administered intraocularly at any time after birth.

[0031] According to a sixth aspect of the present invention, there is provided an AAV particle according to the third aspect of the present invention for use in treating one or more of Norrie's disease, age-related hearing loss, and other NDP-related diseases, wherein a single dose is administered intracochlearly neonatally or prenatally, preferably a single dose administered at about 13-20 post-conception weeks (pcws), more preferably at about 15-18 pcws, optionally a single dose administered intracochlearly during adolescence, preferably at about 12 years of age or younger, and optionally a single dose administered intracochlearly during adulthood, preferably at 12 years of age or older.

[0032] According to a seventh aspect of the present invention, there is provided an AAV particle according to the third aspect of the invention for use in the treatment of one or more of Norrie's disease, age-related or diabetic macular degeneration and retinopathy, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), and other NDP-associated diseases, wherein a single dose is administered intravenously, the dose being administered prenatally, or postnatally, in the neonatal period, in childhood, or in adulthood.

[0033] According to an eighth aspect of the present invention, there is provided a method for rescuing the vasculature of the ear and eye and / or the vascular barrier function of the ear and eye, and / or for preventing or treating ocular neovascularization, comprising administering NDP.AAV particles by at least one of intravenous, intraocular, and intracochlear administration during early development, preferably prenatal development, or during the first year of life, preferably neonatally. Treatment at an early developmental stage rescues the vasculature of the ear and eye and the blood-retinal barrier. In the ear, this prevents sensory hair cell death and progressive hearing loss. Administration of a prenatal human-equivalent treatment to neonatal mice corrects vasculature, prevents neovascularization, or restores normal vasculature.

[0034] According to a ninth aspect of the present invention, there is provided a method for rescuing the blood-retinal barrier and / or the cochlea and / or cochlear potential barrier, comprising administering NDP.AAV particles at a later stage of development, preferably prenatal, by at least one of intravenous, intraocular, and intracochlear administration. The present invention has been shown to prevent sensory hair cell death and progressive hearing loss even when administered after development is complete, suggesting the possibility of postnatal treatment. Treatment of juvenile / young adult mice improves progressive hearing loss and retinal vascular leakage. Barrier rescue includes restoration of tight junctions and transcellular permeability and normalization of pericyte arrangement on blood vessels. Rescuing the blood-receptor barrier in rescuing the cochlea and endocochlear potential protects sensory hair cells and hearing without rescuing vascular morphology. Thus, the present invention achieves long-term improvement in hearing loss (15-month efficacy, i.e., data from aged mice). Late-stage application of this invention reduces retinal vascular leakage and prevents exudates, vision loss, secondary glaucoma, and blindness as a result of tuberculosis. Thus, administering NDP.AAV to postnatal individuals offers significant advantages, and the ability to rescue retinal vascular integrity after transduction of immature tissue and after retinal vascular development is complete is a truly surprising finding, as it demonstrates that hair cell rescue can be achieved independently of vasculature pathology.

[0035] In all aspects, sequence identity is determined by comparing two aligned, substantially complementary sequences over their length, and the overall identity is expressed as a percentage. Measurement of nucleotide sequence identity is well known in the art and uses specialized computer programs such as "BLAST."

[0036] The nucleic acid sequence may be DNA, RNA, cDNA, or PNA, and may be recombinant or synthetic. The nucleic acid sequence may be single-stranded or double-stranded.

[0037] Nucleic acid sequences can be obtained by cloning, for example, using standard molecular cloning techniques including restriction digestion, ligation, and gel electrophoresis (e.g., as described in Sambrook et al.; Molecular Cloning: A laboratory manual, Cold Spring Harbour Laboratory Press). Nucleic acid sequences can be isolated or amplified by PCR techniques. Such techniques can use primers based on the sequence of the nucleic acid to be amplified. With the sequence information provided, one skilled in the art can use available cloning techniques to generate nucleic acid sequences or vectors suitable for introduction into cells.

[0038] The invention is further illustrated in the following non-limiting figures. [Brief explanation of the drawings]

[0039] [Figure 1-1]Gene Therapy Construct Evaluation and Study Design. A. Schematic of the AAV expression construct. Expression is driven by the ubiquitous CAG promoter, and all transduced cells are EGFP-tagged. The human NDP cDNA, containing its native secretion signal and a C-terminal Flag tag, is linked via a P2A linker, allowing self-cleavage from the same transcript during translation, resulting in secretion of the tagged Norrin from the cell. (B-B'') Expression of the construct in HEK293 cells. Endogenous EGFP largely colocalizes with anti-Flag immunostaining (yellow). Scale bar: 50 μm. C. Detection of GFP and NDP monomer bands in transfected HEK293 cell lysates. A larger version of this data is available. D. Schematic of β-catenin signaling and the TopFlash assay. NDP dimers bind to the FZ4 complex, which contains the essential coreceptors LRP5 or LRP6 and the signal-amplifying coreceptor TSPAN12, inducing β-catenin binding to TCF / LEF sites and downstream gene transcription. The TopFlash plasmid used in the β-catenin activation assay encodes firefly luciferase under a promoter containing TCF / LEF binding sites. The TopFlash and mCherry plasmids served as transfection controls. LiCl mimics destruction complex inhibition and was used as a positive control. E. Activity of transgenic NDP in the TopFlash assay against NDP and HEK293 cells cotransfected with endogenous members of its receptor complex and the TopFlash plasmid. The TopFlash plasmid used in the β-catenin activation assay encodes firefly luciferase under a promoter containing TCF / LEF binding sites. The TopFlash and mCherry plasmids served as transfection controls. LiCl mimics destruction complex inhibition and was used as a positive control. [Figure 1-2] Evaluation of Gene Therapy Constructs and Study Design F. Study Design, Readouts, and Approximate Corresponding Stages in Human Development for Therapeutic Administration. Histological analysis of eyes and ears in all groups was performed at 2 months of age, with separate groups of mice analyzed for visual function at 1.5 months of age and hearing at 3 months of age. [Figure 2-1]Expression of constructs in the eye and ear at 2 months. A. Body weights of male mice and age-matched controls before and after AAV administration. Data are shown as mean ± SD. Number of animals: P2-L group and littermate controls, n(WT) = 5, n(Ndp-KO) = 2, n(P2-L) = 16. P30-H group and littermate controls, n(WT) = 15, n(Ndp-KO) = 5, n(P30-H) = 14. B-C. Transduction levels in P2- and P21-treated retinas: (B) P2-L group, n=4; (C) P30-H group, n=4. Staining: anti-GFP antibody and DAPI. GCL - ganglion cell layer, ONL - outer nuclear layer, INL - inner nuclear layer. D. Schematic diagram of an axial section of one rotation of the cochlea. SL - spiral ligament, SV - stria vascularis, OoC - organ of Corti. Orange rectangle outlines the area marked F. E. Transduction in cochlear cross section from P2-treated mouse. SGN - spiral ganglion neuron. [Figure 2-2] Expression of constructs in the eye and ear at 2 months. F-G. Transduction of the lateral cochlear wall after treatment at P2 (G) and P30 (H). Scale bar: 500 μm, n=4. H-I. Transduction of the organ of Corti after treatment at P2 (H) and P30 (I). Scale bar: 100 μm. [Figure 3-1]Effects of early and late treatment on retinal vascular morphology and barrier function, as well as visual function. While rescue of retinal angiogenesis and visual function is limited to early treatment, expression of barrier factors shows a long-term response. A, B. (A) Cross-section and (B) Schematic of retinal vasculature in whole mount. A' shows the depth projection color scheme of C-F. C-F. Whole mount of three vascular plexuses in the central retina, pseudocolor depth projection of the central retina. Vasculature stained with isolectin B4 (IB4). Depth color scheme is shown in A'. Treatment groups: C. WT, D. Ndp-KO, E. P2-L, F. P30-H. Arrows indicate deep plexus vessels. Numbers 1, 2, and 3 indicate vascular plexuses. GCL - ganglion cell layer, INL - inner nuclear layer, ONL - outer nuclear layer. G-J. Vasculature: IB4, nuclei: DAPI. K-N. Colocalization of the tight junction marker claudin-5 (CLDN5) and PLVAP. O-S. Assessment of visual function recovery by scotopic electroretinogram (ERG) at 1.5 months of age. O. Example ERG waveforms during a 10,000 mcd / sm^2 flash stimulation in WT, Ndp-KO, P2-L, and P21-H groups, n=1. Data are shown as individual traces in O and as mean ± SD in P-R. Sample size for C-N: n=4 per group. Post-hoc test values: *P≤0.05, **P≤0.01, ***P≤0.001, ****P≤0.0001; ns=not significant. Sample size for ERG analysis: n(WT)=10, n(Ndp-KO)=10, n(P2-L)=7, n(P21-H)=10. [Figure 3-2]Effects of early and late treatment on retinal vascular morphology and barrier function, and visual function. While rescue of retinal angiogenesis and visual function is limited to early treatment, expression of barrier factors shows a long-term response. O-S. Assessment of visual function recovery by dark-adapted electroretinogram (ERG) at 1.5 months of age. P. Oscillatory potentials. Q-R. ERG B-wave during increasing stimulus flash intensity in the P2-L group (Q) and the P21-H group (R). Data information: Data are shown as individual traces in O and as mean ± SD in P-R. P was analyzed by one-way ANOVA with Sidak's post-hoc test, and Q-R by two-way repeated measures ANOVA with Tukey's test; all values ​​are compared with WT (blue asterisk) and Ndp-KO (red asterisk). Post-hoc test values: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns - not significant. Number of samples for ERG analysis: n(WT) = 10, n(Ndp-KO) = 10, n(P2-L) = 7, n(P21-H) = 10. [Figure 4-1]Recovery of pathology-related gene expression in the cochlea by 2 months of age as determined by RNA-seq and RT-PCR. A. Heatmap of the expression of 45 pathology-related DEGs in the cochleae of WT, Ndp-KO, and PL groups. This gene set showed no significant differences between WT and P2-L, indicating 1) no over-induction of gene expression due to the treatment and 2) the treatment restored expression of all genes to levels similar to WT (sidebar, dark blue). Of the 45 DEGs, 16 genes were also significantly different from Ndp-KO, indicating that expression was restored to normal levels (purple), and 29 genes were significantly different only from WT (light blue). B–E. qRT-PCR analysis of pathology-related gene expression in all treatment groups. (B) Quality control of transgene (EGFP primer) expression. (C) Quality control of mouse genotype by mouse Ndp expression. Expression of pathology-related genes (D) Abcb1a and (E) Cldn5. {Note that despite transgene levels being at least 10-fold higher in the P2-L group, pathology-related genes were not upregulated to levels significantly higher than in the WT group.} Data information: B–E data are shown as mean ± SD. Sample numbers: (A): n(WT) = 4, n(Ndp-KO) = 3, n(P2-L) = 4; (B–E): n(WT) = 9, n(Ndp-KO) = 12, n(P2-L) = 10, n(P21-L) = 8, n(P21-H) = 6, n(P30-H) = 8. Statistical analysis: D–E were analyzed by one-way ANOVA with Sidak's post-hoc test, and all values ​​were compared with WT (blue asterisk) and Ndp-KO (red asterisk): *P≦0.05, **P≦0.01, ***P≦0.001, ****P≦0.0001, ns—not significant. [Figure 4-2]Restoration of pathology-associated gene expression in the cochlea by 2 months of age as determined by RNA-seq and RT-PCR. B-E qRT-PCR analysis of pathology-associated gene expression in all treatment groups. (B) Quality control of transgene (EGFP primers) expression. (C) Quality control of mouse genotype by mouse Ndp expression. Expression of pathology-associated genes (D) Abcb1a, (E) Cldn5. {Note that despite transgene levels being at least 10-fold higher in the P2-L group, pathology-associated genes were not upregulated to significantly higher levels than in the WT group.} Data information: B-E data are shown as mean ± SD. Number of samples: (A): n(WT) = 4, n(Ndp-KO) = 3, n(P2-L) = 4. (B-E): n(WT) = 9, n(Ndp-KO) = 12, n(P2-L) = 10, n(P21-L) = 8, n(P21-H) = 6, n(P30-H) = 8. Statistical analysis: D-E were analyzed by one-way ANOVA with Sidak's post-hoc test. All values ​​were compared with WT (blue asterisk) and Ndp-KO (red asterisk): *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns - not significant. [Figure 5]Differences in the efficacy of early (P2) and late (P30) treatments for salvaging the cochlear vasculature by 2 months of age. A. Schematic of the lateral wall vasculature. SL - spiral ligament, SV - stria vascularis capillaries. B. Quantification of the number of capillary branch points per region in eight consecutive regions along the stria vascularis. Note that branching improved in the P2-L group but not in the P30-H group. No significant morphological differences were detected in other parts of the cochlea between the WT and Ndp-KO groups or between the other groups. C-F. Capillary network density and morphology of the apex of the stria vascularis, stained with anti-endomucin. Scale bar: 100 μm, n = 4 per group. (C) WT, n = 4; (D) Ndp-KO, n = 4; (E) P2-L, n = 5; (F) P30-H, n = 4. Arrows indicate decreased capillary network density and increased vessel diameter. G-J: (G-J) Anti-endomucin immunostaining and (G'-J) anti-claudin-5 immunostaining demonstrate capillary network density and expression of claudin-5 as a tight junction marker in the spiral ligament. Scale bar: 100 μm, n = 4 per group. (G, G') WT, (H, H') Ndp-KO, (I, I') P2-L, (J, J') P30-H. Arrows: In WT and P2-L samples, the vasculature is evenly branched and claudin-5 and endomucin are evenly expressed. Arrows: Diversified blood vessel types. Red arrows: "reticular" vessels. Orange arrows: "barrier" vessels. Data source: Data shown as mean ± SD. Statistical analysis: Two-way repeated measures ANOVA with Tukey's post-hoc test. All values ​​are compared to WT (blue) and Ndp-KO (red). Post-hoc test values: *P ≤ 0.05, **P ≤ 0.01, ns - not significant. [Figure 6-1]Rescue or improvement of outer hair cell survival in all treatment groups by 2 months. A–E: Examples of hair cell survival in the corresponding "sensitive" regions along the frequency axis from different treatment groups. (A) WT, n=7; (B) Ndp-KO, n=6; (C) P2-L, n=5; (D) P21-H, n=6; (E) P30-H, n=8. F–K: Quantification of surviving hair cells from samples similar to A–E and P21-L groups. Samples were compared to WT by repeated measures two-way ANOVA with Tukey's post-hoc test. (F) WT, n=3; (G) Ndp-KO, n=6; (H) P2-L, n=5; (I) P21-L, n=3; (J) P21-H, n=6; (K) P30-H, n=8. Data information: Quantitative data are shown as mean ± SD. Analyzed by two-way repeated measures ANOVA with Tukey's post-hoc test. Significant effects of region (P<0.0001), treatment group (P<0.0001) and their interaction (P<0.0001). Post-hoc test values: *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns-not significant. [Figure 6-2] Rescue or improvement of outer hair cell survival in all treatment groups by 2 months. F-K: Quantification of surviving hair cells from samples similar to those in A-E and P21-L groups. Analyzed by repeated measures two-way ANOVA with Tukey's post-hoc test, samples were compared to WT. (F) WT, n=3; (G) Ndp-KO, n=6; (H) P2-L, n=5; (I) P21-L, n=3; (J) P21-H, n=6; (K) P30-H, n=8. Data information: Quantitative data are presented as mean ± SD. Analyzed by two-way repeated measures ANOVA with Tukey's post-hoc test. Significant effects of region (P<0.0001), treatment group (P<0.0001), and their interaction (P<0.0001). Post-hoc test values: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns - not significant. [Figure 7]Hearing measurements and hearing rescue or improvement at 3 months. (A) Endocochlear potentials (mean ± SD) analyzed by one-way ANOVA with Sidak's post-hoc test, comparing each group to WT. P (Ndp-KO vs. WT) = 0.0107. (B) Overlay of DPOAE threshold means. The gray area indicates the area of ​​WT affected by C57BL / 6-related degeneration. (C) Click ABR thresholds (mean ± SD) analyzed by one-way ANOVA with Sidak's post-hoc test, comparing each group to WT. (D) Overlay of pure-tone ABR threshold means across all groups. The gray area indicates the area of ​​WT affected by C57BL / 6-related degeneration. (E) Schematic diagram of frequency domains and corresponding hearing measurements. The length of the cochlea is divided into eight equal regions (1 / 8–8 / 8). Black arrows indicate the frequency (kHz) to which the specific points correspond. The red line indicates the frequency region sensitive to degeneration in Ndp-KO in C57BL / 6. The gray line indicates the frequency region where ABR and DPOAEs were recorded. The gray region indicates the region associated with degeneration in WT C57BL / 6. (F) Schematic of the putative mechanism of Norrie phenotype rescue by gene therapy. Green indicates typical transduced areas across all treatment groups, and red arrows indicate putative targets of NDP produced in and secreted from the transduced areas. Data information: Data are shown as mean ± SD in A and C, and mean in B and D. Numbers are as indicated in the respective legends. Number of animals for auditory analysis: Endocochlear potential: n(WT) = 8, n(Ndp-KO) = 6, n(P2-L) = 6, n(P21-H) = 6, n(P30-H) = 7. DPOAE: n(WT) = 11, n(Ndp-KO) = 6, n(P2-L) = 9, n(P30-H) = 8. Click and pure tone ABR: n(WT) = 12, n(Ndp-KO) = 7, n(P2-L) = 10, n(P21-H) = 8, n(P30-H) = 8. Statistical analyses in B and D are shown on the pavement images. Post-hoc test values: *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, ns - not significant. [Figure 8-1]Detection of transgenic proteins by Western blot in lysates of HEK293 cells transfected with the constructs. Samples were treated under different conditions: HI: heat inactivation, R: reduction, Tr: transfection. (A) EGFP is detected as a band of approximately 29 kDa under non-reducing conditions and 25 kDa under reducing conditions (arrows). (B) Loading control (GAPDH). (C) Anti-FLAG antibody. Two NDP bands (arrows) on the same blot: 15 kDa represents the monomer, and >250 kDa represents oligomers or aggregates. [Figure 8-2] Detection of transgenic proteins by Western blot in lysates of HEK293 cells transfected with the constructs. Samples were treated under different conditions: HI: heat inactivation, R: reduction, Tr: transfection. (D) NDP monomers or oligomers are also detected with anti-NDP antibodies. Note that anti-flag or anti-NDP staining did not colocalize with GFP, and the molecular weight of the monomer was of the expected size, indicating that no fusion protein was produced. (E) Loading control (GAPDH). [Figure 9] A-B. Body weights of mice before and after AAV administration on P21 and age-matched controls. Data are shown as mean ± SD. (A) Ctrl N=7, P21-L administration N=14. (B) Ctrl N=14, P21-H administration N=9. [Figure 10] The degree of vascular coverage at the time of treatment (P2) and the resulting transduction of AAV9 vectors in the retina. (A, B) Flat mounts of WT and Ndp-KO retinas at P2, with vasculature immunostained with anti-endomucin. (C, D) Frozen sections of WT and Ndp-KO retinas at P21, with vasculature immunostained with isolectin-B4. Scale bar = 50 μm. (E, F) Rat mounts of WT and Ndp-KO retinas at P30, with vasculature immunostained with anti-endomucin. Scale bar = 500 μm. (G, H) GFP staining in retinas 1 month after treatment at P2 (G) and P21 (H). [Figure 11]Transduction of AAV9 vectors in the cochlea. Whole mounts of the organ of Corti and lateral wall were stained with anti-GFP antibody. (A-A'') Untreated cochlea. (B-B'') P2-L treatment. (C-C'') P21-L treatment. (D-D'') P21-H treatment. (E-E'') P30-H treatment. Scale bar = 500 μm (A, B, C, D, E). 500 μm (A', B', C', D', E'). [Figure 12] Electroretinograms showing the effects of early and late treatment. (A) Representative ERG traces in response to single light flashes of increasing intensity. (B, C) Flash ERG, b-wave amplitude / a-wave amplitude ratio. (D, E) Flash ERG, a-wave amplitude. WT N=10, Ndp-KO N=10, P2-L N=7, P21-H N=10. [Figure 13-1] Differential gene expression in the cochlea by RNA sequencing and qPCR. (A) Venn diagram showing genes differentially expressed between Ndp-KO and WT cochleae, and between Ndp-KO and P2-L cochleae. (B) Genes found to be significantly differentially expressed between Ndp-KO and P2-L cochleae. Note that these genes showed similar expression patterns between Ndp-KO and WT cochleae, but the fold changes did not reach significance. No genes were found to be significantly differentially expressed between WT and P2-L cochleae. [Figure 13-2] Differential gene expression in the cochlea by RNA sequencing and qPCR. (C, D) Expression patterns of Plvap and Sox17 in all groups by qPCR. WT N=9, Ndp-KO N=12, P2-L N=10, P21-L N=8, P21-H N=6, P30-H N=8. [Figure 14] Hair cell survival along the apical-basal axis in whole mounts of the organ of Corti. Cochlear regions and corresponding frequencies are indicated above the figure. (A) WT N=3 (B) Ndp-KO N=6 (C) P2-L N=5 (D) P21-L N=3 (E) P21-H N=6 (F) P30-H N=8 Scale bar = 500 [Figure 15-1]Statistical analysis of DPOAE and ABR thresholds for all groups. (A, B) DPOAE thresholds for P2-L, P-30H, and control groups. P values ​​indicate statistically significant differences from WT. (C-E) ABR thresholds for P2-L, P-30H, and control groups. P values ​​indicate statistically significant differences from WT. Data information: Data are presented as mean ± SD. N numbers are indicated in each legend. Statistical analysis was performed separately for each group, comparing treated samples with the same WT and Ndp-KO control groups. Post-hoc test values: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. [Figure 15-2] Statistical analysis of DPOAE and ABR thresholds for all groups. (C-E) ABR thresholds for P2-L, P-30H, and control groups. P values ​​indicate statistically significant differences from WT. Data information: Data are presented as mean ± SD. N numbers are indicated in each legend. Statistical analysis was performed separately for each group, comparing treated samples with the same WT and Ndp-KO control groups. Post-hoc test values: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. [Figure 16] Study design for systemic expression testing of codon-optimized NDP after intravenous injection. [Figure 17] Study design for intraocular expression study of codon-optimized NDP after intraocular injection. [Figure 18] Intravitreal injection of AAV2 / ShH10 / NDP in p22 rescues the blood-retinal barrier (test (c0) and clinical (c6) (codon-optimized)) constructs). A: BSA-AF 549 concentration is linearly related to fluorescence in the dynamic range relevant to this study. B: Boxplots show the degree of transcellular permeability of the blood-retinal barrier (BRB) for WT, Ndp, KO, and AAV2 / ShH10.NDP KO. Each circle represents a single data point. [Figure 19] Cross section of retina from p30 Ndp KO mouse treated with p22 intravitreal construct 6 (SEQ ID NO: 21, SEQ ID NO: 22) (3.47 x 109 vector genomes per eye). [Figure 20] Cross sections of retinas from WT and Ndp KO mice at p30. [Figure 21]Section showing retinol vasculature - WT had three layers, whereas Ndp-KO had only one layer. [Figure 22] Intravitreal injection of AAV2 / ShH10.NDP at p8 rescues the intermediate and deep retinal capillary layers. [Figure 23] Methods of vascular analysis - retinal flat mounts. [Figure 24] Peripheral blood vessel density assessed by p30 in p8 or p21 BSS-treated WT mice, BSS-treated Ndp KO mice, and AAV2 / ShH10-treated Ndp KO mice. [Figure 25] Quantification of peripheral angiogenesis radius. [Figure 26] Peripheral retinal vessel radii were increased by intravitreal treatment with construct 0 (SEQ ID NO: 18, SEQ ID NO: 19) at P8, but not at P21. [Figure 27] The b-wave height can be rescued to some extent by intravenous delivery of AAV2 / 9.NDP at p2, but not at p21. [Figure 28] B-wave height can be rescued by intravenous delivery of AAV2 / 9.NDP (construct 0 (SEQ ID NO: 18, SEQ ID NO: 19)) at p2, but not at p21. [Figure 29] Schematic diagram of the progression of Norrie disease. [Figure 30] Schematic diagram showing the cochlear (A) and retinal (B, C) vasculature. [Figure 31] Retinal images of WT and Ndp KO mice at P2 (A, B), as well as retinal images (C, D) and retinal cross sections (E, F) at 1 month. [Figure 32-1] Development of vascular pathology in WT and Ndp KO mice. [Figure 32-2] Development of vascular pathology in WT and Ndp KO mice. [Figure 33] Schematic diagram of the therapeutic window. [Figure 34] Schematic diagram of NDP-transduced and recipient cells. [Figure 35]Schematic of construct packaging into AAV vectors using the triple transfection protocol. [Figure 36] The construct labeled transduced HEK293 cells, and anti-flag tag immunostaining colocalized with endogenous GFP. [Figure 37] Western blot analysis detected the expression of NDP protein monomer and GFP as distinct proteins. [Figure 38] A) Schematic showing that Norrin is a secreted signaling protein that induces canonical WNT / β-catenin signaling through the FZ4 complex with LRP5 or LRP6 and TSPAN12. B) Graph showing that the ability of the construct to induce β-catenin signaling through these receptors was demonstrated in vitro using the Top flash assay. [Figure 39] The graph shows that Ndp-KO mice treated at neonatal age grew and developed normally. [Figure 40] Retinal vascular morphology of treated Ndp-KO mice compared to WT. [Figure 41] Cochlear vascular morphology of treated Ndp-KO (C) mice compared to WT. [Figure 42] HEK293 cells transduced with the NDP construct (CE10). DETAILED DESCRIPTION OF THE INVENTION

[0040] Detailed Description The following describes further embodiments of the present invention. Various embodiments are described below. Each aspect defined in this manner can be combined with one or more other aspects, unless expressly indicated otherwise.

[0041] Generally, the nomenclature used in connection with, and techniques of, cell and tissue culture, pathology, oncology, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry and hybridization described herein are well known and commonly used in the art. The methods and techniques of this disclosure are performed according to conventional methods well known in the art and as described in the various general and more specific references cited and discussed throughout this specification, unless otherwise indicated.

[0042] Enzymatic reactions and purification techniques are performed according to manufacturer's specifications, or as commonly accomplished in the art, or as described herein. The nomenclature used in connection with analytical chemistry, synthetic organic chemistry, and medicinal and pharmaceutical chemistry, as well as the laboratory procedures and techniques described herein, are well known and commonly used in the art. Standard techniques are used for chemical synthesis, chemical analysis, pharmaceutical preparation, formulation, and delivery, and for the treatment of patients. Suitable assays for measuring the properties of the molecules disclosed herein are also described in the Examples.

[0043] The present invention is a construct comprising a wild-type NDP nucleic acid sequence or a codon-optimized NDP nucleic acid sequence.

[0044] Human patients with mutations in the gene NDP can develop retinal and cochlear disease, leading to vision and hearing loss.

[0045] Norrie disease, caused by mutations in the NDP gene, is a severe X-linked disorder that causes blindness and progressive hearing loss; approximately 30% of patients also develop cognitive impairment and peripheral vascular disease. Blindness results from hypoxia, ischemia, persistence of the fetal vasculature, and impaired retinal vascular development, resulting in compensatory neovascularization. Hearing is normal at birth, and boys with Norrie disease pass newborn hearing screening. However, nearly all boys develop progressive hearing loss, usually during puberty.

[0046] The present invention is an NDP gene therapy construct that incorporates a functional copy of the NDP gene within an adeno-associated viral vector (NDP.AAV), which may prevent vision and / or hearing loss.

[0047] The use of AAV as a vector for gene therapy is commonly used experimentally and is also the basis for several commercially available gene therapies (e.g., for RPE65-associated reverse congenital amaurosis).

[0048] Vision and hearing loss result from (i) impaired vascularization of the retina and cochlea during fetal development and (ii) impaired function of the blood-brain barrier. As demonstrated herein, administration of the NDP.AAV of the present invention during vascular development in NDP knockout mice can partially rescue peripheral vascularization in the retina and reduce the loss of outer hair cells in the cochlea (which causes hearing loss). Furthermore, administration after vascular development is complete can partially rescue the blood-retinal barrier in NDP knockout mice.

[0049] Delivery of NDP.AAV was effective via both intravenous and intravitreal (intraocular) routes. The adenovirus used for intravenous injection was AAV 2 / 9, while AAV 2 / ShH10 was used for intraocular injection. AAV 2 / ShH10 was chosen for intraocular delivery because it previously demonstrated high transfection efficiency into Müller cells (the site of native NDP expression in the retina).

[0050] Expression of NDP leads to the secretion of functional Norrin protein, a signaling protein within the extracellular matrix of the retina and cochlea. NDP is required for normal retinal and cochlear vascularization, proper function of the blood-retinal barrier in the eye, and function of the WNT signaling pathway, which maintains the endocochlear potential in the cochlea.

[0051] In one embodiment, the construct can express a wild-type NDP nucleic acid sequence. In another embodiment, the construct can express a codon-optimized NDP nucleic acid sequence. The wild-type and codon-optimized NDP nucleic acid sequences can be human NDP nucleic acid sequences. For example, in one embodiment, the wild-type NDP nucleic acid sequence can be the nucleotide sequence of SEQ ID NO: 1. The polypeptide sequence expressed from SEQ ID NO: 1 is set forth in SEQ ID NO: 2. In another embodiment, the codon-optimized NDP nucleic acid sequence can be the nucleotide sequence of SEQ ID NO: 3. The polypeptide sequence expressed from SEQ ID NO: 3 is set forth in SEQ ID NO: 4.

[0052] In the present invention, the codon-optimized NDP sequence of SEQ ID NO: 3 has been shown to be particularly effective in rescuing vision and hearing loss as part of an AAV gene therapy construct.

[0053] In yet another embodiment, the construct may comprise a nucleic acid sequence having at least about 90%, or 91%, or 92%, or 93%, or 94%, or 95%, or 96%, or 97%, or 98%, or 99% sequence identity to SEQ ID NO:1.

[0054] In yet another embodiment, the construct may comprise a nucleic acid sequence having at least about 90%, or 91%, or 92%, or 93%, or 94%, or 95%, or 96%, or 97%, or 98%, or 99% sequence identity to SEQ ID NO:3.

[0055] In some embodiments, the NDP nucleic acid sequence can be DNA, RNA, cDNA, or PNA, can be recombinant or synthetic, and can be single-stranded or double-stranded.

[0056] In some embodiments, the NDP polypeptide expressed from the NDP nucleic acid sequence is a secreted form of the NDP polypeptide.

[0057] In one embodiment, the construct may comprise at least a promoter sequence, for example, a CAG promoter sequence or a CBA promoter sequence, a full-length human NDP coding sequence (402 bp including the native secretion signal), and a WPRE sequence.

[0058] In one embodiment, the construct may comprise at least a promoter sequence, such as a CAG promoter sequence or a CBA promoter sequence, a full-length human NDP coding sequence (402 bp including the native secretion signal), a WPRE sequence, and an SV40 late polyA sequence at the 3' end.

[0059] In other embodiments, the promoter is a CAG promoter, a CBA promoter, a CMV promoter, an EF1a promoter, a PGK promoter, a TRE promoter, a U6 promoter, a UAS promoter, an EFS promoter, a SFFV promoter, a MSCV promoter, a SV40 promoter, a UBC promoter, a Pro1A promoter, a hRHO promoter, a hBEST1 promoter, a Grm6 promoter, a GJB2 promoter, a GJB6 promoter, a SLC26A4 promoter, a TECTA promoter, a DFNA5 promoter, a COCH promoter, a NDP promoter, a SYN1 promoter, a GF AP promoter, PLP promoter, TAK1 promoter, SOX21 promoter, SOX2 promoter, FGFR3 promoter, PROX1 promoter, GLAST1 promoter, LGR5 promoter, HES1 promoter, HES5 promoter, NOTCH1 promoter, JAG1 promoter, CDKN1A promoter, CDKN1B promoter, SOX10 promoter, P75 promoter, CD44 promoter, HEY2 promoter, LFNG promoter, SlOOb promoter, CLDN11 promoter, NDP promoter, or synthetic modifications or combinations of these promoters.

[0060] In the embodiment shown in Figure 1A, the construct may comprise the following elements: a strong ubiquitous CAG promoter upstream of EGFP (to label transduced cells), a self-cleaving P2A linker, a full-length human NDP coding sequence (402 bp including the native secretion signal) with a FLAG epitope sequence tagging the C-terminus to aid in detection of transgenic Norrin, followed by a WPRE sequence and an SV40 late polyA sequence at the 3' end.

[0061] In some embodiments, the construct may comprise an AAV vector. For example, the construct may comprise one or more of the following: AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, AAV-rh8, AAV-rh10, AAV-rh39, AAV-rh43, AAVAnc80, AAV 2 / ShH10, AAV-S vector.

[0062] In a preferred embodiment, the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV 2 / ShH10 vector. NDP expressed from AAV 2 / ShH10 and AAV-S vectors is advantageous for intraocular administration.

[0063] In another preferred embodiment, the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV 2 / 9 vector. NDP expressed from AAV 2 / 9 and AAV-S vectors is advantageous for intracochlear administration.

[0064] AAV particles (NDP.AAV) comprising this construct can be used in the treatment of one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases.

[0065] AAV virions carrying the NDP gene can be injected into the ear or eye of a patient to induce production of the NDP gene product.

[0066] In one embodiment, the NDP.AAV gene therapy construct can be administered intravenously. Intravenous administration is advantageous because it is the least invasive means of delivering the NDP.AAV construct. Intravenous administration can be performed in utero, such that the construct is administered to the fetus at a time when the eye and ear structures are beginning to form. Thus, intravenous delivery allows for a simple means of achieving NDP gene therapy without the need for more complex procedures on the fetus. Intravenous administration can also be performed at any time after birth, with minimal intervention by medical professionals. Intravenous administration can be performed in combination with other forms of administration of the NDP.AAV construct.

[0067] In one embodiment, the NDP.AAV gene therapy construct can be administered intraocularly. To administer NDP.AAV intraocularly, intraocular injection can be used using known techniques. Intraocular administration is advantageous because it delivers NDP.AAV directly to the site of retinal degeneration. Intraocular administration can include any one of intravitreal, subretinal, or suprachoroidal injection. Intraocular administration can be performed in utero, and the use of intrauterine AAV injection for early gene expression has previously been shown to be feasible in mice (Yasuda et al. 2021). Intraocular administration can also be performed at any time after birth. Intraocular administration can be performed in combination with other forms of administration of the NDP.AAV construct.

[0068] In one embodiment, the NDP.AAV gene therapy construct can be administered intracochlearly. To administer NDP.AAV intracochlearly, intracochlear injection can be used using known techniques. Intraocular administration is advantageous because it delivers NDP.AAV directly to the site of cochlear degeneration, preventing sensory hair cell death and progressive hearing loss. Intracochlear administration can be performed in utero, and the use of intrauterine AAV injection for early gene expression has been shown to be feasible in mice (Chin-Ju et al. 2020) (Yasuda et al. 2021). Intracochlear administration can also be performed at any time after birth. Intracochlear administration can be performed in combination with other forms of NDP.AAV construct administration.

[0069] In further embodiments, NDP.AAV administration can be any combination of intravenous, intraocular, and intracochlear administration. For example, NDP.AAV can be administered intravenously and intraocularly, or intravenously and intracochlearly, or intraocularly and intracochlearly, or intravenously, intraocularly, and intracochlearly.

[0070] The use of NDP.AAV gene therapy provides both an early and late treatment to rescue the vasculature of the ear and eye.

[0071] As described in the Examples, early treatment is important for complete resolution of ocular vascular pathology.

[0072] Also contemplated are embodiments in which NDP.AAV gene therapy is used to increase the activity of β-catenin signaling in situations where other related pathways (e.g., NDP, LRP5, TSPAN5, and FRZ6) are affected or other components (e.g., TSPAN5) are mutated, or in ocular pathologies. Furthermore, as previously described, NDP binds to a receptor complex consisting of FZD4, LRP-5 / 6, and TSPAN-12 to induce β-catenin signaling in cells. In some embodiments, NDP.AAV gene therapy can be used to correct pathologies caused by mutations in other WNT signaling genes (e.g., FZD4, TSPAN12, LRP5 / 6).

[0073] In further embodiments, NDP.AAV gene therapy can be used to enhance the activity of β-catenin signaling in situations where genes in other related pathways are affected or in ocular pathologies. As demonstrated in the Examples, NDP.AAV gene therapy has shown efficacy in preventing vascular leakage and can therefore be provided to Coats patients in place of steroids, even in the absence of mutations.

[0074] NDP.AAV gene therapy treatment may also be beneficial for individuals suffering from progressive hearing loss with a vascular component, such as age-related and / or diabetic maculopathy and retinopathy, as well as retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), and other NDP-related diseases. RNAseq data can also be used to assess treatment outcomes. These include novel druggable targets, such as Clu (or Fabp3)—clusterin confers resistance to age-related hearing loss—where NDP.AAV regulates Clu or other markers to protect against progressive or age-related hearing loss. Additionally, embodiments include the use of NDP.AAV gene therapy to prevent exudates, retinal detachment, secondary glaucoma (resulting in pain), or bulbar palsy (resulting in eye loss).

[0075] Certain embodiments relate to delivery sites required for efficacy, such as fibrocytes of the spiral ligament (SL), stria vascularis and lateral wall basal and marginal cells, glial cells of the modiolus; retinal Müller cells, RGCs, RPE cells, and evidence of rescue after transduction of PRS-RGCs alone and Müller cells alone. This is evident from human RNAseq analysis of human fetal cochleae, which found human NDP expression sites in POM, fibrocytes, and glial cells.

[0076] It should be noted that, as described in the Examples, signaling to hair cells is not required. Thus, the target in the ear is not the hair cell. This discovery provides the ability to appropriately target the cause of hearing loss at various points in development.

[0077] In one embodiment, NDP.AAV is used to treat one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-related diseases, or other β-catenin signaling-related retinal diseases, by administering an initial dose intraocularly during the neonatal period. Delivery of the initial dose to the neonate can rescue the ocular vasculature and vision. The rescue of vasculature includes the deep vasculature. In one embodiment of the present invention, the initial dose is administered at about 14-17 postconceptional weeks (pcws), more preferably at about 15-16 pcws. In another embodiment, a second dose is administered intraocularly during the neonatal period. The second dose can be administered to the neonate in utero and may be beneficial for generating sufficient angiogenesis. In one embodiment, the second dose is administered at about 22-26 postconceptional weeks (pcws), more preferably at about 24 pcws. In another embodiment, the second dose is administered intraocularly at any time after birth. In yet another embodiment, the third dose can be administered intraocularly at any time after the neonatal administration of the first two doses. Intraocular administration of DP.AAV at later stages of development prevents retinal vascular leakage. Late application prevents retinal vascular leakage, resulting in exudation and loss of the growth zone of the eye. Cell rescue is possible even after birth with intraocular administration, indicating that the effect is independent of vascular pathology, which is completed during the neonatal period. Intraocular administration can be performed using known techniques and may be intravitreal injection.

[0078] In one embodiment, NDP.AAV is used to treat one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-related diseases, or other β-catenin signaling-related retinal diseases, with the first dose administered intracochlearly during the neonatal period. Delivery of the first dose to a newborn can rescue the vasculature of the ear and prevent hearing loss. Thus, sensory hair cell death and progressive hearing loss are prevented using intracochlear administration. In one embodiment of the present invention, the first dose is administered at about 13-20 postconception weeks (pcws), more preferably about 15-18 pcws. In another embodiment, a second dose is administered intracochlearly during the neonatal period. In one embodiment, the second dose is administered intracochlearly during adolescence, more preferably at about 12 years of age or younger. In another embodiment, the second dose is administered intracochlearly in adulthood, more preferably at age 12 or older. In yet another embodiment, the third dose can be administered intracochlearly at any time after birth following receipt of the first two doses as a newborn. The third dose can be administered intracochlearly in adulthood, more preferably at any time after age 12 or older. Using the NDP.AAV of the present invention, sensory hair cell death and progressive hearing loss can be prevented even when administered after development is complete, making lifelong treatment feasible. Rescue of the cochlear blood-vessel barrier and endocochlear potential preserves sensory hair cells and hearing without rescuing vascular morphology. Thus, the present invention achieves long-term improvement in hearing loss.

[0079] In one embodiment, NDP.AAV can be used to treat one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-related diseases, or other β-catenin signaling-related retinal diseases, and the first dose of the drug is administered intravenously. In one embodiment, the first dose is administered neonatally or postnatally. In one embodiment, the second dose is administered neonatally or postnatally. In one embodiment, the third dose is administered neonatally or postnatally.

[0080] Further embodiments involve intravenous and / or intraocular and / or intracochlear administration of more than three doses. For example, 4, 5, 6, 7, 8 or more doses may be administered to an individual. This administration may be administered prenatally or postnatally. Embodiments also include periodic administration via intravenous and / or intraocular and / or intracochlear administration throughout the individual's life.

[0081] In some embodiments, the constructs can be specifically formulated for delivery of DNA or RNA molecules using exosomes, nanoparticles, or liposomes (e.g., lipofectamine) or non-viral vectors such as viral vectors, e.g., retroviral vectors, vaccinia virus vectors, adenoviral vectors, or herpes simplex virus vectors, or conjugated lipids.

[0082] In the present invention, each construct can be packaged in any of the AAV serotypes. Construct 1 is a codon-optimized NDP open reading frame with a tag (SEQ ID NO: 19 (ITR-to-ITR sequence) and SEQ ID NO: 20 (full-length plasmid sequence). Construct 6 is a codon-optimized open reading frame without a tag (SEQ ID NO: 21 (ITR-to-ITR sequence) and SEQ ID NO: 22 (full-length plasmid sequence). For reference, construct 0 (SEQ ID NO: 17 (ITR-to-ITR sequence) and SEQ ID NO: 18 (full-length plasmid sequence)) is used in Example 1 below and is located within AAV2 / 9. SEQ ID NO: 23 is construct 2, which comprises the NDP promoter. SEQ ID NO: 24 is construct 3. [Example]

[0083] The present invention is further illustrated by the following non-limiting examples. Example 1 - Study demonstrating that systemic AAV0.NDP rescues retinal pathology and hearing loss in a Norrie disease model Norrie disease (Ndp) is a rare, recessive, X-linked, dual sensory disorder characterized by congenital blindness and progressive hearing loss. It is caused by mutations in the NDP gene, which encodes Norrin, a secreted Wnt analog protein that induces canonical Wnt / β-catenin signaling via the FZ4 / LRP5 / 6 / TSPAN12 complex. Currently, no treatment is available for Norrie disease. The visual loss caused by Norrie disease results from early developmental defects in the deep retinal vascular plexus during late gestation, making therapeutic intervention difficult. We recently demonstrated that structural abnormalities and barrier defects in the cochlear vasculature in Ndp-KO mice begin early in development and precede sensory outer hair cell degeneration and hearing loss, potentially contributing to these defects through reduced endocochlear potential or metabolic stress. Importantly, patients with Norrie disease have normal hearing in early childhood, with the onset of hearing loss often beginning in the teens to mid-twenties, suggesting a potential therapeutic window for intervention to prevent sensory cell degeneration and hearing loss. To explore the potential application of gene therapy to the treatment of Norrie disease, we generated a proof-of-concept tagged Norrin expression construct and studied the distribution and detection of transgenic Norrin. Despite the tag, this construct produced the expected protein conformation and was shown to interact with the primary receptor complex in functional assays. Finally, in vivo functionality of the construct in the eye and ear was demonstrated in an Ndp-KO mouse model.

[0084] This example of the present invention demonstrates the efficacy of an AAV9 vector carrying a human NDP gene therapy construct delivered intravenously to an Ndp-KO mouse model at three clinically relevant stages of disease progression. This example of the present invention demonstrates that early postnatal treatment preserved both cochlear and retinal structure and function. Treatment of young mice achieved complete or partial rescue of cochlear structure and hearing function, but not retinal rescue.

[0085] Testing design and function of NDP gene therapy constructs in vitro To evaluate gene therapy in Ndp-KO mice, we designed an experimental construct expressing the human NDP gene. This construct consisted of a strong, ubiquitous CAG promoter upstream of EGFP (to label transduced cells), a self-cleaving P2A linker, and the full-length human NDP coding sequence (402 bp, including the native secretion signal) with a FLAG epitope sequence at the C-terminus to aid in the detection of transgenic Norrin, followed by a WPRE sequence and an SV40 late poly(A) sequence at the 3' end (Figure 1A). The expression and function of the construct were characterized in vitro in HEK293 cells. Figure 1B'-B'' shows cytoplasmic EGFP in transfected HEK293 cells, which coincided with anti-FLAG immunostaining and labeled the cell surface (Figure 1B'', yellow). EGFP and NDP / Norrin proteins were detected by Western blot of transfected HEK293 cell lysates (Figure 1C, Figure 8). Recombinant NDP formed high molecular weight oligomers or aggregates (>250 kDa) that were reducible to a band the size of the NDP monomer (16 kDa) (Fig. 1C, Fig. 8).

[0086] Using a TopFlash luciferase reporter assay in HEK293 cells, we confirmed the ability of recombinant NDP to interact with its cognate receptor complex and activate β-catenin signaling (Figure 1D) (Chang et al., 2015). The NDP expression construct induced luciferase activity when cotransfected with human FZ4, LRP6, and TSPAN12 expression plasmids (Chang et al., 2015), but not alone (Figure 1E), consistent with previous evidence demonstrating the interaction of NDP with its receptor complex (Lai et al., 2017). Addition of lithium chloride (lithium choride), which is known to stabilize β-catenin by inhibiting GSK3 (Zeilbeck, 2014), induced luciferase activity, as expected, even in the absence of NDP or receptor. Together, these data indicate that this construct expresses biologically active recombinant NDP.

[0087] Safety and transduction of the eye and ear following systemic delivery of AAV9.NDP To examine the expression of NDP gene therapy constructs in mouse models, the AAV9 serotype was chosen for packaging because it can cross the blood-brain barrier but does not transduce vascular endothelial cells (Merkel et al. 2017). Other studies have achieved widespread transduction by administering intravascular AAV9 (Shibata et al. 2017; Massaro et al. 2020; Merkel et al. 2017). We anticipated that the use of the ubiquitous CAG promoter and AAV9 would provide delivery of the construct to the retina and cochlea after intravenous injection while avoiding direct eye and ear damage caused by local administration. Based on our previous findings, the goal was to transduce cells near blood vessels at the site of angiogenesis in the cochlea—the modiolus and lateral wall—so that the secreted NDPs could target endothelial cells of adjacent blood vessels and maintain a microenvironment conducive to sensory hair cell survival (Figure 1F).

[0088] To test the efficacy of treatment at time points relevant to patients with progressive hearing loss in Norrie disease and informed by our previous characterization of the Ndp-KO mouse (on C57BL / 6J) background (Bryant et al. 2022), we selected three time points for vector administration that represent different stages of maturation and pathology in the eye and cochlea (Figure 1F): 1) neonatal period (postnatal day P2), prior to the development of vision and hearing, when retinal vasculature begins to form and the endocochlear potential is established; 2) juvenile-pre-regression period (P21), when the ocular vasculature and cochlea have just matured and there is no hair cell death; and 3) juvenile-regression period (P30), when progressive hair cell death begins in the cochlea and neovascularization occurs in the eye.

[0089] These correspond to the developmental timeframe for prenatal therapeutic delivery in children and young adults. Figure 1F summarizes the experimental design. AAV.GFP was delivered intravenously to groups of newborn mice (2.73E+13vg / kg; dose P2-L), two groups of P21 juvenile mice (5.45E+12vg / kg; dose P21-L and 2.74E+13vg / kg; dose P2-H), and two groups of P30 juvenile mice (1.37E+13vg / kg; dose P30-H). Treated mice were monitored regularly and demonstrated normal weight and health compared to controls. Control samples were pooled between the WT and Ndp-KO groups because no differences were observed in measurements from PBS-injected control mice at any time point (Figure 2A, Figure 9).

[0090] At 2 months of age, transduction of the retina and cochlea was confirmed by GFP immunostaining (Figure 2). P2-injected retinas were most efficiently transduced in the central vascularized area at P2 (Supplementary Figure Y) (Figure 2B, Figure 10). P21 administration resulted in widespread transduction of the retina (Figure 2C), consistent with the complete coverage of the inner retinal vasculature in Ndp-KO mice (Figure 10). Retinal ganglion cells were efficiently transduced in both early- and late-treated mice, with low expression in Müller glia, the known site of Ndp expression (Ye et al. 2009) (Figure 2B, C).

[0091] In the cochlea, transduction was achieved in the modiolus and lateral wall (Figure 2F), which are in close proximity to putative targets of NDP signaling, i.e., the lateral wall vasculature and hair cells (Hayashi et al. 2021; Rehm et al. 2002). Spiral ganglion neurons were transduced in all treatment groups (Figure S4), as were fibrocytic cells in the lateral wall (presumably type II and type IV fibrocytes) and modiolus (Figure 2F-I). Inner hair cells were sparsely transduced only in the P2-L group; no transduction was observed in outer hair cells, vascular endothelial cells, pericytes, or astrocytes.

[0092] In summary, the AAV.NDP vector achieved transduction of both neonatal and juvenile cochleae and retina, although at lower levels in juvenile compared to neonatal administration. Vascular cells were not transduced.

[0093] Neonatal treatment with AAV9.NDP rescues retinal vasculature and vision The effect of AAV NDP gene delivery on retinal pathology was assessed by analyzing the retinal vasculature (Figure 3A, B) of mouse eyes 2 months after treatment on P2-L and P21-H using retinal cryosections and whole mounts. (Previous studies in Ndp-KO mice have shown that the development of the superficial retinal vasculature is slower than normal but complete by P20, whereas the deep retinal vasculature fails to form (Richter et al. 1998; Luhmann et al. 2005a). Figure 10 shows the appearance of Norrie diseased retinas at the time of treatment.)

[0094] In whole-mount retinal sections from WT mice, vasculature was observed in three vascular plexuses (Fig. 3C), whereas in Ndp-KO mice, only the superficial vascular plexus was observed (Fig. 3D). Treatment at P2 rescued all three vascular plexuses, but not at P21 (Fig. 3E, F). Formation of vascular networks in the vascular plexuses was confirmed by color-coded Z-stack depth projections of retinal whole-mounts (Fig. 3C-F). Three vascular networks at different depths were also detected in cryosections from WT and P2-treated mice (Fig. 3G, I). However, in Ndp-KO and P21-H mice, abnormal neovascular plexuses were present only in the superficial retina, with individual non-branching neovascular tufts (arrows in Fig. 3H, J). The blood-retinal barrier is normally established by P17–P20 (Fruttiger 2002). Immunostaining of Ndp-KO retinas by 2 months revealed decreased expression of claudin-5, a structural component of endothelial tight junctions, and increased expression of PLVAP, a structural component of endothelial openings, compared to WT retinas. These have previously been reported as early markers of abnormal Norrie disease retinal vasculature (Wang et al. 2012). In both the P2-L and P21-H groups, claudin-5 expression was restored, and PLVAP staining typical of Ndp-KO disappeared (Figure 3K-N), consistent with rescue of the vascular phenotype.

[0095] To evaluate the effects of AAV NDP delivery on visual function, dark-adapted electroretinograms (ERGs) were recorded at 1.5 months of age from the P2 and P21 treatment groups and the control group. Figure 3O shows typical dark-adapted ERG traces in response to a bright flash of 10,000 mcd / s-2 from WT, Ndp-KO, P2-L, and P21-H groups. The prominent b-wave in WT, indicating signaling from photoreceptors to bipolar cells, was nearly flattened in Ndp-KO (Figure 3O). P2-L-treated animals were similar to WT and showed some recovery of oscillatory potentials, although the amplitude did not reach its full height, while ERG traces from P21-H were similar to Ndp-KO (Figure 3O). Supplementary Figure 12 shows the full set of average traces for each group (Figure 12A) and the a-wave to b-wave amplitude ratios (Figure 12B, C). Although the difference did not reach significance in either group, an improvement in oscillatory potential amplitude was observed (Fig. 3P).

[0096] There were no differences in a-wave parameters between WT and Ndp-KO or between treatment groups (Fig. 12D-E). There was a significant difference in b-wave amplitude between WT and Ndp-KO with a large effect size (Fig. 3Q,R). The b-wave at P2-L showed significant improvement, consistent with deep retinal revascularization (Fig. 3Q). At P21-H, b-wave amplitude was partially restored in response to the highest intensity flash (Fig. 3R).

[0097] Collectively, these data demonstrated the efficacy of intravenous delivery of AAV.NDP vectors in ameliorating retinal pathology. Treatment before retinal vascular maturation rescued deep retinal vascular pathology, but not at later time points. Previous studies using genetically engineered mice showed that restoration of the deep vascular plexus in Norrie disease was not possible after maturation (P17–P20) (Wang et al. 2012).

[0098] Biomarkers of Norrie disease in the cochlear response to AAV.NDP treatment Because our understanding of the downstream molecular mechanisms leading to cochlear damage in Norrie disease is limited, and to provide biomarkers for evaluating therapeutic efficacy in the cochlea, we compared gene expression patterns in WT and Ndp-KO cochleae and in P2-treated mice. NDP downstream targets and transcriptional activity associated with Norrie disease in adult mouse cochleae have not previously been analyzed. RNA-seq analysis of whole cochleae from 2-month-old male WT mice (n = 4), Ndp-KO mice (n = 3), and P2-L mice (n = 4) identified dysregulated gene expression profiles.

[0099] Differential gene expression analysis revealed 45 genes significantly differentially expressed between WT and Ndp-KO cochleae (adjusted p ≤ 0.05) and 35 genes significantly differentially expressed between Ndp-KO and treated Ndp-KO P2-L samples (Figure 13). No genes were significantly differentially expressed between the WT and Ndp-KO P2-L groups, suggesting rescue by treatment. Of these 45 and 35 significantly differentially expressed genes (DEGs), 16 genes overlapped, indicating rescue to normal expression levels. Furthermore, unsupervised clustering of all samples based on the expression of the 45 differentially expressed genes between WT and Ndp-KO revealed that treated Ndp-KO P2-L samples clustered with WT samples, but not with untreated Ndp-KO samples (Figure 4). Slc7a1, Flt1, Abcb1a, and Cldn5 were downregulated in Ndp-KO mice and returned to normal levels upon treatment (Figure 4A). These are likely downstream targets of Ndp signaling and are associated with the normal function of the cochlear microvasculature: the barrier (Cldn5), pericyte-driven vascular branching and barrier (Flt1) (Eilken et al. 2017; Wang et al. 2019; Zhang et al. 2021), and molecular transport (Abcb1a, Slc7a1). Abcb1a has been associated with hearing loss and increased susceptibility to ototoxicity in mice (Li et al. 2019; Zhang et al. 2000). Slc7a1 is an amino acid transporter typical of a normal BBB (Yahyaoui and Perez-Frias 2019). The set of genes upregulated in Ndp-KO due to their functions may be related to stress response in cochlear pathology ( Figure 4A ).

[0100] These findings were consistent with microvasculature being a primary site of pathology. Of note, genes expressed in hair cells were not differentially expressed, despite a significant loss of hair cells becoming evident by 2 months (Bryant et al., 2022). The dysregulation of barrier markers and transporters supports the hypothesis that microvascular disruption creates an inappropriate microenvironment for hair cell survival in the cochlea in Norrie disease.

[0101] The expression of the newly identified set of pathology biomarkers was analyzed by qRT PCR at later time points after treatment, along with analysis of expression levels of transgenic NDP GFP and the pericellular permeability gene Plvap, which we previously showed to be dysregulated in 2-month-old Ndp-KO cochleae ( Bryant et al. 2022 ).

[0102] Expression levels of the transgene NDP-GFP in the cochlea were highest after neonatal treatment. In juvenile pups, expression corresponded to the dose, with higher expression at P21 and P30 compared with the low-dose treatment at P21 (Figure 4B). These patterns are consistent with the patterns of GFP transduction observed in cochlear whole mounts (Figure S11), demonstrating age-dependence of transduction levels (see groups P2-L and P21-H, injected with the same amount of vg / kg).

[0103] After 2 months, gene expression returned to WT expression levels in the P2-L and P21-H groups, and at least partial / significant improvement was observed in the P30-H sample, but not in the P21-L group (Figure 4B–E). Two-way ANOVA and Tukey's post-hoc test demonstrated that Cldn5 was downregulated in the Ndp-KO group but fully restored in the P2-L group. Treatment of young mice revealed Cldn5 levels comparable to those of WT only at the high dose of P21 (P21-H) (Figure 4). Plvap was highly sensitive to AAV NDP treatment, even at the lowest dose, and increased Plvap expression was successfully downregulated to WT levels in all treatment groups (Figure 13). The rescue of both markers at P21 suggests that even mature cochlear vasculature in Norrie disease is still responsive to restoration of the pericellular tight junction barrier.

[0104] Together, these data demonstrate that dysregulated gene expression levels are restored to wild-type after treatment of not only newborn mice but also subsequent juvenile mice, and the expression of selected hallmarks of molecular microvascular pathology is ameliorated, suggesting that gene therapy delivery of NDP maintains barrier and transport function.

[0105] Effects of AAV.NDP on lateral wall vasculature and organ of Corti sensory hair cell survival after treatment in newborn and young mice To assess whether biomarker gene expression patterns were consistent with rescue of tissue pathology, we analyzed whole mount cochleae from AAV.NDP-treated mice by immunohistochemistry to assess the effects on the lateral wall microvasculature and hair cell survival in the organ of Corti. Having previously confirmed that malformations of the cochlear microvasculature are an early site of pathology (Bryant et al. 2022), we compared the effects of AAV.NDP gene therapy on the morphology of the lateral wall vasculature by endomucin immunostaining after treatment in newborn and young mice (Figure 5A). Quantification of branching points in the stria vascularis (and spiral ligament) along the lateral wall revealed reduced branching in Ndp-KO mice, which was most pronounced in the apical region (1 / 8 and 2 / 8 regions of the apical-basal axis) compared to WT mice (analyzed by repeated measures two-way ANOVA with Dunnett's post-hoc test, P < 0.05) (Figure 5B, C–F).

[0106] We also found that treatment improved the malformation in the capillary network of the spiral ligament in neonatal pups, but not in later time points. While endomucin and claudin-5 staining were uniformly distributed in WT vessels (Fig. 5G-G'), claudin-5 was low or absent in most vessels in Ndp-KO (Fig. 5H-H'). As previously mentioned, some atypical vessels had high levels of claudin-5 and abnormally low levels of endomucin.

[0107] In the P2-L treatment group, the appearance of the vascular network was comparable to that of WT, with endomucin and claudin-5 staining uniformly distributed (Figure 5I-I'), indicating good rescue. In the P30-H treatment group, vascular diversification was prominent, and some of the vasculature formed atypical network structures similar to those in Ndp-KO (Figure 5J-J'), suggesting that rescue was not achieved. These data suggest that NDP is required for the early development of the apical stria vascularis, which has not yet formed at P2, and that this pathology is irreversible by subsequent treatment after the stria vascularis has fully matured (P20) (Ando and Takeuchi 1998).

[0108] AAV.NDP prevents OHC loss even after the onset of degeneration We analyzed hair cell survival levels at 2 months of age in WT mice compared with untreated Ndp-KO mice. At 2 months, hair cells in the WT cochlea were largely intact (Figure 6A-A''). However, Ndp-KO mice showed severe degeneration of OHCs in two to four of the eight mid-frequency regions corresponding to 6 to 18 kHz (Figure 6B-B''). This was consistent with our previous study. In the treated groups, OHCs were either completely preserved, as in the P2-L and P21-H groups (Figure 6C-C'' and D-D''), or partially preserved in the P30-H group (Figure 6E-E''). Surviving OHCs were quantified in whole mounts of the organ of Corti (Figure 6F-K), and mapped to equidistant regions along the apical-basal axis. For each treatment group, data were analyzed using a two-way repeated measures ANOVA with Dunnett's post-hoc test, comparing the corresponding region with WT (Figure 6F-K). The analysis confirmed complete OHC rescue in the P2-L and P21-H (Figure 5I, K) groups and a significant improvement in the "sensitive" region (apical 2 / 8 to 4 / 8) in the P21-L and P30-H samples (Figure 5J, L). Figure 14 shows hair cell survival along the apical-basal axis of the organ of Corti in all groups.

[0109] Whole-mount analysis of the cochlea demonstrated that neonatal treatment improved the microvasculature and prevented the onset of hair cell degeneration, and importantly, subsequent treatment in young pups also inhibited progressive hair cell degeneration.

[0110] AAV9.NDP treatment of Ndp-KO young mice rescues hearing function {referring to advanced stages}. Finally, to evaluate the therapeutic effects on hearing, we performed audiodynamic and electrophysiological assessments of cochlear function at 3 months of age (Figure 7). Endocochlear potentials were significantly reduced in Ndp-KO mice compared to WT mice, but recovered after treatment in both newborn and juvenile mice (Figure 7A).

[0111] To estimate hair cell function, we analyzed DPOAE thresholds between 6 and 30 kHz. The mean thresholds are overlaid in Figure 7B, and the statistical analysis is shown in Figure 15. Thresholds in the mid-frequency region between 6 and 18 kHz were significantly elevated in Ndp-KO compared to WT, consistent with the loss of hair cell integrity in these regions of the organ of Corti after 2 months (Figure 6) and our previous analysis (Bryant et al., 2022). Unexpectedly, in the high-frequency region where little hair cell degeneration was observed in Ndp-KO, thresholds in WT mice were elevated, demonstrating significantly poorer function at 30 kHz than Ndp-KO (Figure 7B, arrowheads). This likely reflects the rapid onset of age-related hearing loss in these WT control mice, a known feature of later stages of aging in C57BL / 6 mice (Johnson et al., 1997). For completeness, audiological and electrophysiological analyses were performed at all frequencies, marking the "crossing point" of the frequency range where WT controls exhibit paradoxical auditory dysfunction.

[0112] The thresholds in the mid-frequency range of 6–18 kHz in Ndp-KO compared with WT were significantly elevated in Ndp-KO but were completely rescued after neonatal treatment (P2-L). Treated juveniles also showed indistinguishable function from WT at 12–18 kHz (Figure 7B–B′′′), consistent with hair cell survival in the organ of Corti after treatment.

[0113] At the highest frequency (30 kHz), both treated Ndp-KO and untreated mice exhibited significantly lower DPOAE thresholds than WT (Fig. 7B-B''').

[0114] To assess hearing recovery, auditory brainstem responses (ABRs) to broadband click and pure tone stimuli were recorded. Click thresholds were significantly elevated in Ndp-KO compared with WT, but returned to WT values ​​in all treatment groups (Fig. 7C), indicating that overall good hearing rescue was achieved.

[0115] Next, we analyzed frequency-specific ABR responses to pure tone stimuli ranging from 3 to 42 kHz. Ndp-KO mice exhibited elevated thresholds between 3 and 18 kHz compared to WT mice. Similar to DPOAE analysis above 24 kHz, Ndp-KO mice maintained normal hearing (?REF) despite elevated thresholds in WT mice (Figures 7D-D'', 15). After AAV.NDP treatment of both newborn and juvenile mice, thresholds between 30 and 18 Hz decreased, with complete recovery to WT thresholds achieved only after neonatal treatment.

[0116] In the high-frequency range (30–42 kHz), Ndp-KO mice had better hearing than WT mice, and both P21-H and P30-H mice showed even better hearing than Ndp-KO mice (Figure 7D′′–D′′′). However, the mean thresholds of the P2-L group showed slightly worse hearing than Ndp-KO mice (Figure 7D′). This difference was significant.

[0117] Collectively, these EP, DPOAE, and ABR results demonstrate that AAV9.NDP gene therapy can preserve hearing function after treatment at various stages of disease. Taken together, the concordance of affected and rescued frequency regions in DPOAE and tone-ABR measurements, as well as in the degenerated and rescued regions of OHCs, suggests that rescue likely occurs via prevention of progressive hair cell loss and preservation of OHC function (Figure 7E).

[0118] Consideration AAV.NDP treatment prevented disease progression and improved response and duration of treatment In this proof-of-concept study, we explored the feasibility of applying gene replacement therapy to treat Norrie disease. For the first time, we demonstrated that early treatment provided complete rescue of cochlear pathology and retinal vasculature, proving that our vector delivered functional human Norrin. The time point of early treatment corresponds to the fetal stage of human development. Fetal gene therapy is still in its infancy, and in the case of NDP, there may be a risk of adverse effects on the placental vasculature or the developing fetus (Ye et al. 2009). However, we also demonstrate that Norrie disease cochleae respond to treatment across various stages of the disease, including after the onset of degenerative changes. These changes occur after sensory system development is complete, corresponding to childhood or young adulthood in Norrie disease patients. This suggests that treatment of these patients is feasible and achievable. The low cell turnover in the cochlea suggests that NDP.AAV gene replacement therapy may provide a long-term supply of Norrin to maintain cochlear blood-vessel barrier function throughout life.

[0119] In the eye, Norrie deficiency results in underdevelopment of two deep layers of the retinal vasculature. Visual impairment in Norrie disease is usually present from birth (Redmond et al. 1993). Our study demonstrated that defects in the deep retinal vasculature cannot be regenerated by NDP after maturation of the vascular network. However, consistent with other studies (Wang et al. 2012), the retinal vascular barrier responds to AAV NDP after birth and remains responsive to treatment for the long term. This provides a potential treatment route for NDP-associated ocular pathologies in patients.

[0120] The role of the vasculature in the cochlear phenotype in Norrie disease The cochlear responsiveness to treatment at later time points is consistent with a pathology mediated by vascular barrier dysfunction. Deficiency of NDP does not result in gross morphological abnormalities or defects in the cochlear vasculature, but only disrupts the blood-labyrinth barrier (Bryant et al., 2022). Maintenance of the blood-labyrinth barrier is essential for maintaining the endocochlear potential and, through it, hair cell survival and function (Liu et al., 2016). RNA-seq and qPCR analysis of whole cochlear lysates demonstrated significant dysregulation of the vascular barrier and transport factors at early and late time points, followed by recovery after AAV-NDP treatment. Furthermore, downregulation of the expression of the transporters Abcb1a and Slc7a1 may have caused metabolic stress. While recovery of vascular gene expression coincided with recovery of the endocochlear potential, OHCs, and hearing, morphological defects in the cochlear vessels were not. Even as the disease progresses, the cochlear vascular barrier appears to be reversible, allowing the restoration of a normal hair cell environment and its normal function.

[0121] In the cochlea, Norrin may directly affect HCs in regulating maturation and gene expression through the transcription factor Pou4f3 (Hayashi et al.). Constitutive overexpression of Ndp in supporting cells or stabilization of neonatal b-catenin in hair cells using Atoh1-Cre expression have both been reported to maintain OHC survival (Hayashi et al.). Despite the observed loss of OHCs in Ndp-KO cochleae, RNA-seq analysis at 2 months of age did not detect a decrease in Pou4f3 or hair cell gene expression (e.g., Myo7a), demonstrating the resolution of bulk RNA-seq analysis. Our study specifically demonstrates that OHCs differentiated and matured in the absence of Ndp can survive (Figure 6) and function (Figure 7, DPOAE) for a long period of time if Ndp is restored at P21 or 1 month of age.

[0122] Clinical AAV.NDP Gene Therapy Development and Safety Gene therapy is a rapidly evolving field. While the majority of treatments are in preclinical trials, three AAV-mediated therapies—Luxturna®, Zolgensma®, and Glybera®—have been approved for the treatment of severe genetic disorders. Secreted signaling protein gene therapy has not yet been extensively studied, and there is potential for side effects due to unregulated, prolonged expression. The viral dose used to achieve cochlear rescue in this study was low, approximately 5–25 times lower than that used in Zolgensma®, which is clinically approved for the treatment of spinal muscular atrophy, a life-threatening condition, using a similar AAV9 vector (1.1 × 10 14 vg / kg). Rescue with our ubiquitously expressed NDP construct means that specific cell targeting is not required, as the NDP is secreted and reaches the required target cells. This is consistent with previous reports in which rescue was achieved by ectopic overexpression of NDP in transgenic mice (Ohlmann et al. 2005; Bassett et al. 2016). The AAV.NDP gene therapy construct transduced cells near the cochlear lesion site, but hardly transduced affected vasculature endothelial cells or sensory hair cells.

[0123] Because systemic delivery of AAV can cause side effects, direct administration to the eye and ear may be more suitable for clinical application. Full longitudinal toxicological studies are needed to establish the safety of AAV.NDP gene therapy. This proof-of-concept study demonstrates for the first time that Norrie disease pathology responds to gene replacement therapy, paving the way for targeted gene delivery to treat progressive hearing loss and retinal vascular leakage. Such NDP gene delivery may be useful for palliating ocular disease in mild FEVR-like Norrie disease cases (Wawrzynski et al., 2022) or in conjunction with planned preterm birth in cases of prenatal diagnosis of Norrie disease (Sisk et al., 2014). AAV.NDP gene replacement may also have potential for treating peripheral vascular disease symptoms in patients with Norrie disease. To our knowledge, this is the first application of systemic AAV9 delivery to treat progressive cochlear damage (Shibata et al., 2017).

[0124] method The gene expression plasmid: CAG>EGFP-P2A-NDP-FLAG pAAV gene therapy construct was designed using vectorbuilder.com and purchased as the E. coli original strain from Cyagen. Plasmids expressing human FZ4, LRP6, and TSPAN12 (Chang et al. 2015) were provided by Professor Yvonne Jones (Oxford). M50 Super 8x TopFlash (12456) in pTA-Luc vector and M51 Super 8x FopFlash (12457) in pGL3 vector were obtained from Addgene. All plasmids were expanded using standard methods and purified using the Miraprep protocol (Pronobis et al. 2016). HEK293 cells were cultured in DMEM-high glucose medium (11965-084, Gibco) supplemented with 10% FBS (A38401, Gibco) at 37°C in a conventional 5% CO2 cell culture incubator (5% CO2, 37°C). Cells were passaged at a 1:5 ratio using 1x trypsin / EDTA (25300-054, Gibco).

[0125] TopFlash assay: EK293 cells were seeded at equal density in 96-well plates. The next day, cells were transfected for 24 hours using FuGENE® HD Reagent (2 μl FuGENE:1 μg DNA) (transfection mixture: 40 μg TopFlash, 10 μg mCherry, and 10 μg each of the NDP gene therapy construct and Norrin receptor plasmid combinations), then washed with and replaced with regular tissue culture medium. Treatment with 5 mM LiCl for 24 hours was used as a positive control for β-catenin activation. Cells were then assayed for β-catenin activity using the Dual Luciferase® Reporter Assay System (E1910, Promega) according to the manufacturer's instructions. Induced luminescence was measured using the recommended measurement parameters on a Fluostar Optima luminometer (BMG Labtech).

[0126] Western blotting: HEK293 cells were transfected with the NDP gene therapy construct as described above. After 48 hours, the transfection medium was removed, and cells were harvested in RIPA buffer containing the protease inhibitor cocktail cOmplete™ Mini (11836153001, Promega). Total protein was extracted and quantified by Bradford assay according to standard methods. Samples were then diluted and mixed with 4x Laemmli sample buffer (BioRad) with or without 5% β-mercaptoethanol, heat-inactivated at 75°C, or incubated at room temperature for 10 minutes and then kept on ice. 20–30 μg of protein per well was loaded onto a 1 mm 12% SDS-PAGE gel, separated by electrophoresis (Mini-PROTEAN, BioRad), and then transferred to a 0.2 μm pore-size nitrocellulose membrane (BioRad) using the TransBlot Semidrite Lasfer system. Membranes were washed in PBST, deblocked with 5% non-fat milk (Blotting-Grade Blocker, BioRad) in PBST, and probed with relevant antibodies: anti-GAPDH EMD Millipore MAB374, anti-actin, anti-FLAG eBioscience 14-6681-80, anti-GFP Abcam Ab6662, anti-desmin.

[0127] RNA extraction: Cochleae were isolated from surrounding tissue and vestibule and flash-frozen. Retinas were dissected from eyes and flash-frozen in cold PBS. Total RNA was extracted using a modified version of a published protocol (Vikhe Patil et al. 2015) (TRI Reagent 93289-25ML Sigma-Aldrich DirectZol kit). RNA was eluted in 40-50 μl of nuclease-free water and analyzed using a NanoDrop™ 2000 (Thermo Scientific) and Agilent Bioanalyzer platform.

[0128] Gene Expression Analysis: cDNA was synthesized from 100 ng of RNA using the RvertAid H Minus First Strand cDNA Synthesis Kit (K1631) with random hexamers according to the manufacturer's instructions. cDNA equivalent to 1 ng of RNA per reaction was used for gene expression analysis with PowerSYBR® Green PCR Master mix (436759) and relevant primers (Cldn5: F: 5'TTAAGGCACGGGTAGCACTCACG3' (SEQ ID NO: 5), R: 5'TTAGACATAGTTCTTCTTGTCGT3' (SEQ ID NO: 6); Plvap: F: 5'GTGGTTGGACTATCTGCCTC3' (SEQ ID NO: 7), R: 5'ATAGCGGCGATGAAGCGA3' (SEQ ID NO: 8); Actin-b: F: 5'TGTTACCAACTGGGACGACA3' (SEQ ID NO: 9), R: 5'CTGGGTCATCTTTCACGGT3' (SEQ ID NO: 10); Abcb1a, Slc7a1, Flt1). Whole transcriptome analysis was performed using strand-specific RNA sequencing with polyA selection on Illumina Nova-seq, with a targeted library size of 20 million paired-end 150-bp reads per sample.

[0129] Virus production and packaging: Gene therapy constructs were packaged into AAV capsids at the UCL AAV Facility using HEK293T / (AAV?) cell cultures as previously described. Viral titers were determined by RT-PCR using linearized construct plasmids as standards (ITR F: 5'GGAACCCCTAGTGATGGAGTT3' (SEQ ID NO:11), R: 5'CGGCCTCAGTGAGCGA3' (SEQ ID NO:12), EGFP F: 5'AGTCCGCCCTGAGCAAAGA3' (SEQ ID NO:13), R: 5'TCCAGCAGGACCATGTGATC3' (SEQ ID NO:14).

[0130] Animal testing Animal experiments were conducted under ethical review by University College London and King's College London, with permission from the UK Home Office, and in accordance with UK Home Office regulations and the Animals (Scientific Procedures) Act 1986. Mice were housed under a 12-hour light / 12-hour dark cycle and provided with food and water ad libitum.

[0131] Mice carrying the NdpTm1wbrg (Ndp-) allele were provided by Professor W. Berger (Berger et al. 1996), and this strain was maintained by mating heterozygous Ndp+ / - females with wild-type C57BL / 6 males from Charles River. Ndpy / - males and Ndp- / - females (Ndp-KO) and littermates or age-matched Ndpy / + males or Ndp+ / + females (WT) were used in the study. Ndp- / - females are known to be infertile (Luhmann et al. 2005b). Genomic DNA was isolated from ear or tail biopsies and genotyped by PCR (MyFi Mix (BIO-25050); F: 5'GTATTGCATCCATATTTCTTGG3' (SEQ ID NO: 15), R: 5'CTCTCCATCCCCTGACAAGGA3' (SEQ ID NO: 16), WT amplicon = 528 bp, KO amplicon ~1500 bp).

[0132] Treatment with AAV constructs: Eight microliters of AAV construct in PBS containing the ...vg ("low") dose was injected into the superficial temporal vein of newborn mice (P2) using a 100-microliter Hamilton syringe. P21 or P30 mice were maintained at 38°C for 10 minutes prior to injection to allow the vasculature to dilate. Forty or 50 microliters of ...vg (low dose) or ...vg (high dose) AAV construct was injected into the tail vein. Littlmate controls were injected with the same volume of PBS. Treated mice were followed and weighed three times weekly until P30 and once weekly thereafter.

[0133] Hearing tests (ABR, DPOAE, EP) were performed as previously described (JCI, inventors).

[0134] Electroretinograms (ERGs) were performed according to a published protocol by Ohlmann et al. (45). Mice were dark-adapted overnight for a minimum of 12 hours and then anesthetized by isoflurane inhalation, with pupils dilated (1% tropicamide eye drops) and anesthetized (proxymetacaine eye drops). Mice were then connected to the OcuScience® HMsERG system according to the manufacturer's instructions. Under dark-adapted conditions, single-flash recordings were obtained using stimulus intensities of 0.1 mcds / m², 1 mcds / m², 3 mcds / m², 10 mcds / m², 30 mcds / m², 0.1 cds / m², 0.3 cds / m², 1 cds / m², 10 cds / m², and 25 cds / m². Ten responses at each intensity were averaged.

[0135] TRITC-BSA permeability assay. Two-month-old mice were injected via the tail vein with 2.5% TRITC-BSA (Sigma) (w / v) in PBS. Mice were sacrificed by cervical dislocation 45 minutes later. Tissue samples were fixed and immunostained as described below.

[0136] Tissue Processing: Eyes were isolated and fixed in 4% paraformaldehyde (PFA) for 60-90 minutes, followed by multiple washes with PBS. The cochlea was isolated and dissected from the floor of the ear. The apex of the cochlea, the oval window, and the round window were opened, and 1 ml of 4% paraformaldehyde (PFA) was injected through the round window. Fixation continued in 4% PFA for 90-120 minutes, followed by decalcification in 4% EDTA in PBS (w / v), pH 7.4, for 72 hours and multiple washes with PBS. Brain: The right hemisphere was carefully removed from the skull and fixed in 4% PFA for 24 hours, followed by three 10-minute washes with PBS.

[0137] Histology: After enucleation, the eyes were equilibrated sequentially in 15% and 30% sucrose, embedded in OCT medium (Thermofisher), flash-frozen in a dry ice-isopentane slurry, and stored at -80°C. Sections were cut at 12 μm thickness using a Leica cryostat and mounted on SuperFrost Plus slides (Thermofisher). The cochlea was embedded in 4% low-melting-point agarose in PBS. 150-200 μm-thick sections were cut using a vibratome and stored in PBS at 4°C. The sclera, choroid, and RPE were removed from the posterior segment of the eye, and retinal whole mounts were prepared by making five radial incisions in the retina, with the longest incision representing the ventral retina. Cochlear whole mounts were prepared by removing the otic capsule and dissecting below the spiral eminence / stria vascularis to separate the lateral wall from the middle ear.

[0138] Immunohistochemistry: Tissue samples were incubated in permeabilization / blocking solution (5% FBP, 1% BSA in PBS containing 0.1% (tissue sections) or 0.5% (whole mount) Triton X-100). Samples were incubated with primary antibodies diluted in permeabilization / blocking solution overnight at 4°C, washed with PBS, incubated with secondary antibodies for 2 hours at room temperature, washed with PBS, and mounted with Prolong Diamond (P36970, Invitrogen). Primary antibodies: Endomucin (SantaCruz Sc53941, 1:100), anti-desmin (Proteintech 16520-1-AP, 1:300), anti-Myo7a (Proteus 25-6790, 1:200), anti-claudin 5 (Invitrogen 34-1600, 1:500), anti-FLAG (Invitrogen 14-6681-80), anti-NDP (R&D systems AF3014), Alexa-fluor 594 anti-tubulin β (BioLegend, 801207 1:500), FITC anti-GFP (Abcam Ab6662), Alexa-fluor 488 anti-GFP (Invitrogen A21311). Secondary antibodies: anti-mouse IgG (H+L) Alexa Fluor 488 (Life Technologies A11001 1:500), anti-mouse IgG (H+L) Alexa Fluor 594 (Life Technologies A21203 1:500), anti-rat IgG (H+L) Alexa Fluor 647 (ThermoFisher A21247 1:250), anti-rabbit IgG (H+L) Alexa Fluor 488 (Life Technologies A21206 1:250), anti-rabbit IgG (H+L) Alexa Fluor 568 (ThermoFisher A11036 1:250).Markers: Alexa Fluor 647 phalloidin conjugate (Life Technologies A22287 1:200), Alexa Fluor™ Plus 750 phalloidin conjugate (Invitrogen A30105, ?), Alexa Fluor 594 isolectin GS-IB4 conjugate (Life Technologies I21413, 1:100), Alexa Fluor 647 isolectin GS-IB4 conjugate (Invitrogen I32450, ?). Images of samples were obtained using a Zeiss Observer, ... spinning disk, or Olympus microscope.

[0139] Branch point analysis: Branch points of stria vascularis capillaries were manually quantified using ImageJ from low-magnification images of lateral wall whole-mount preparations. A vascular "intersection" point was considered to be a branch point where three branches joined.

[0140] Hair cell quantification was performed on low-magnification images of whole-mount organ of Corti specimens. Each organ of Corti sample was mapped and divided into 24 equal sections using the Measure_line macro in ImageJ (REF, Liberman). A custom-written ImageJ macro was used to sample 200 μm-long rectangles from each section, and MyoVIIA-positive hair cells were counted using local maximum detection. Empty slots left by dead cells were counted manually. The percentage of viable cells was calculated as viable / (dead + viable) * 100%. Values ​​from three adjacent regions were averaged, for a total of eight regions per organ of Corti. Data from treatment and control groups were analyzed using two-way ANOVA with Dunnett's post-hoc test for multiple comparisons (GraphPad PRISM v7.0).

[0141] [Table 1] JPEG2025535106000003.jpg117170

[0142] Example 2 - Study to evaluate rescue of retinal pathology when codon-optimized NDP.AAV is administered intravenously and intraocularly in a Norrie disease model The Ndp-KO mouse model recapitulates human Tonorie's disease, and Ndp-KO mice were used in experiments in which codon-optimized NDP-AAV gene therapy was administered either intravenously or intraocularly.

[0143] Figure 16 shows the experimental design for intravenous codon-optimized NDP-AAV gene therapy. KO mice at p2 or p21 were intravenously injected with AAV2 / 9.Gfp / NDP (therapeutic dose) or eBSS (control dose). Genotyping was performed by PCR, and mice were examined for retinal vasculature and blood-retinal barrier function at P30 and electroretinograms at P45.

[0144] Figure 17 shows the experimental design for codon-optimized NDP-AAV gene therapy via intracochlear injection. KO mice were intracochlearly injected with AAV2 / SH10.Gfp / NDP (therapeutic dose) or eBSS (control dose) at p8 or p22. Genotyping was performed by PCR, and mice were examined for retinal vasculature and blood-retinal barrier function at p30 and electroretinograms at p45.

[0145] Figure 18A shows the preparation of a serial dilution of fluorescently labeled bovine serum albumin. The fluorescence of the mixture at each concentration was measured. The relationship between BSA concentration and fluorescence was found to be linear within the dynamic range required for subsequent experiments. Therefore, the total weight of BSA in each lysed retina can be inferred from the respective fluorescence measurements. Combined with BSA protein quantification, the number of ng of BSA-AF 594 per mg of retinal protein in each well can be read.

[0146] The graph in Figure 18B quantifies the integrity of the blood-retinal barrier in WT, Ndp-KO, and Ndp-KO mice receiving various intravitreal treatments. BSA (bovine serum albumin) normally does not cross the blood-retinal barrier. Therefore, low levels of fluorescently labeled BSA are observed in the lysed retina of WT mice (shown in blue). In Ndp-KO mice, significantly higher amounts of fluorescently labeled BSA are observed in the lysed retina (shown in red). The following four green bars represent mice treated with 1.76 × 10 vector genomes per eye of construct 0 (SEQ ID NO: 18, SEQ ID NO: 19), 3.47 × 10 vector genomes per eye of construct 6 (SEQ ID NO: 21, SEQ ID NO: 22), 1.74 × 10 vector genomes per eye of construct 6 (SEQ ID NO: 21, SEQ ID NO: 22), and 8.68 × 10 vector genomes per eye of construct 6 (SEQ ID NO: 21, SEQ ID NO: 22). In each case, the integrity of the blood-retinal barrier is rescued.

[0147] Figure 19 shows a cross-section of the retina from a p30 Ndp-KO mouse treated with intravitreal construct 6 (SEQ ID NO: 21, SEQ ID NO: 22) at p22 (3.47 x 10 vector genomes per eye) (Construct 6 (SEQ ID NO: 21, SEQ ID NO: 22) KO at p22). In particular, this figure shows the presence of claudin 5 within the wall of the retinal vasculature. Claudin 5 is an essential component of the blood-retinal barrier and is known to be absent in Ndp-KO mice. Thus, this figure provides histological evidence of rescue of the blood-retinal barrier. Blue: DAPI, Red: IB4, Yellow: Cld 5.

[0148] As shown in Figure 20, intravenous injection of AAV2 / 9.NDP at p2 and p21 restores the claudin+, PLVAP- state of the adult mouse retina. This figure shows cross sections of the retina from WT and Ndp-KO mice at p30. Two of the Ndp-KO mice were treated with systemic AAV2 / 9.NDP at p2 and p21, respectively. In WT mice, PLVAP is absent, but claudin 5 is expressed. Both the absence of PLVAP and the presence of claudin 5 are necessary for the integrity of the blood-retinal barrier. In KO mice, these changes are reversed, indicating that the blood-retinal barrier is permeable. In P2-treated mice, the pattern seen in WT mice is restored. In P-21-treated mice, the pattern seen in WT mice is partially restored (claudin 5 is present, but PLVAP is reduced, not absent).

[0149] Sections showing the retinal vasculature (three layers in WT vs. only one layer in Ndp-KO) are shown in Figure 21. Sections of WT, Ndp-KO, and Ndp-KO retinas treated with AAV2 / SH10.NDP (construct 0 (SEQ ID NO: 18, SEQ ID NO: 19)) at P30 are shown. Treated retinas were intravitreal injected at P8. In WT, three inner retinal vascular layers are present; in KO, only one layer is present. The normal three-layer retinal vasculature is restored in treated retinas.

[0150] As shown in Figure 22, intravitreal injection of AAV2 / ShH10.NDP at p8 rescues the middle and deep retinal capillary layers. Retinal whole mounts from mice culled at p30 are shown. The image on the left is from an eye treated with AAV2 / ShH10.NDP (construct 0 (SEQ ID NO: 18, SEQ ID NO: 19)) at p8. The image on the right is from an untreated eye. The treated retina shows denser angiogenesis, increased peripheral angiogenesis, and decreased neovascularization. The untreated retina shows poor peripheral angiogenesis, generally decreased retinal angiogenesis density, and an area of ​​severe neovascularization in the lower left section (white arrow).

[0151] Intraretinal angiogenesis is beneficial to retinal function, whereas neovascularization (which occurs uncontrolled on the retinal surface) is detrimental and predisposes to vitreous hemorrhage and retinal detachment in humans.

[0152] A macro was devised to measure the density of peripheral retinal angiogenesis from retinal whole-mount images (see Figure 23). The analysis steps include 1) subtracting the background, 2) extracting the vessels and binarizing the image, 3) placing a circle of a set size around the center of the retina, centered on the disk, and 4) measuring the mean gray value outside the circle (peripheral density).

[0153] Figure 24 shows peripheral blood vessel density assessed at p30 in BSS-treated WT mice, BSS-treated Ndp-KO mice, and AAV2 / SH10-treated Ndp-KO mice at p8 or p21. The density of retinal peripheral blood vessel formation was increased by intravitreal treatment with construct 0 (SEQ ID NO: 18, SEQ ID NO: 19) at P8 but not at P21.

[0154] Figure 25 shows quantification of peripheral vessel radius. The analysis steps include: 1) measuring the total retinal radius; 2) measuring the retinal vessel radius; and 3) dividing the vascularized retinal radius by the total retinal radius to determine the ratio of vascularized retinal radius to total retinal radius. The radius of retinal peripheral vascularization was assessed by dividing the radius of the maximal peripheral extent of the vasculature by the total retinal radius.

[0155] As shown in FIG. 26, the radius of retinal peripheral blood vessel formation was increased by intravitreal treatment with Construct 0 (SEQ ID NO: 18, SEQ ID NO: 19) at P8, but not at P21.

[0156] Intravenous delivery of AAV2 / 9.NDP can partially rescue b-wave height at p2, but not at p21 (see Figure 27). ERG was performed at approximately p45. KO Ndp mice exhibited reduced b-wave amplitude, suggesting insufficient signaling from photoreceptors to bipolar cells. Mice treated intravenously with construct 0 (SEQ ID NO: 18, SEQ ID NO: 19) at p2 (early treatment) showed a statistically significant increase in b-wave height. However, treatment at p21 did not rescue the b-wave.

[0157] As shown in Figure 28, intravenous delivery of AAV2 / 9.NDP (construct 0 (SEQ ID NO: 18, SEQ ID NO: 19)) at p2, but not p21, rescues b-wave height. The statistically significant increase in b-wave height with early treatment at p2 remains at least until 10 months of age.

[0158] Method used in Example 2 Blood-retinal barrier assay: To measure transcellular (PLVAP-associated) permeability of the blood-retinal barrier, mice were injected via the tail vein with fluorescently labeled bovine serum albumin (BSA-AF 594, 0.5% in PBS). Three hours later, under terminal anesthesia, the circulation was flushed by transcardial perfusion with 40 ml of sterile PBS. If the blood-retinal barrier is not functional during the three hours between injection and perfusion, BSA-AF 594 will accumulate in the retinal tissue. Afterward, the eyes were enucleated and either fixed in PFA to prepare flat mounts or the retinas were immediately isolated. In early experiments, retinal isolation was performed using the same method as used for qPCR and Western blot analysis. However, in subsequent experiments, careful dissection under a surgical microscope was performed to avoid inadvertently including iris or choroidal tissue in the sample. Samples were snap-frozen in 1.5 ml Eppendorf tubes on dry ice. The retina was then thawed on ice, lysed in buffer containing cOmplete™ protease inhibitor (Roche 11836153001), spun at 10,000 g for 5 minutes, and the supernatant was saved. 150 μl of supernatant per retina was divided into three technical replicates and pipetted into a black 96-well flat-bottom plate. The fluorescence of the red band (from BSA-AF 594) and green band (from eGFP [enhanced green fluorescent protein] in treated animals) in the supernatant was measured according to the following excitation and barrier wavelengths: green: excitation wavelength λ485 nm, barrier wavelength λ535 nm; red: excitation wavelength λ580 nm, barrier wavelength λ632 nm. To normalize the results to the amount of retinal protein present in each sample, the protein concentration of the samples was measured using the BCA assay according to the manufacturer's instructions. The relationship between known BSA-AF 594 concentrations and fluorescence was also examined within the dynamic range of the experiment to aid in the interpretation of the results.

[0159] Intravitreal injection: Eye injections were performed as follows: In p8 mice, the eyelids were first carefully divided along the suture line with a scalpel. To avoid corneal damage, the lateral canthus was lifted from the globe with toothed forceps and cut from posterior to anterior. This was not necessary in p22 mice, as the eyelids open at p12. Tropicamide 1% was instilled to dilate the pupil, followed by a topical anesthetic, proxymetacaine. Viscotea was instilled to prevent ocular dryness and reversible cataract formation. A clear cover glass was placed over the cornea to allow direct observation of the posterior segment under an operating microscope. The upper and lower eyelids were then gently spread apart to minimize eyeball detachment and expose the limbus. The 34-gauge needle of a Hamilton syringe was inserted into the limbus at a posterior angle to avoid damaging the lens, and 0.5 μl (p8) or 1 μl (p22) of AAV2.NDP (construct 0 (SEQ ID NO: 18, SEQ ID NO: 19) or 6) / ShH10 or eBSS [Earl's Balanced Salt Solution] was injected.

[0160] Example 3 - Research and development of gene therapy for Norrie disease Norrie disease (ND) is a rare X-linked recessive disorder. As shown in Figure 29, Norrie disease manifests as congenital blindness followed by progressive hearing loss. The disease is caused by mutations in the NDP gene, which encodes the secreted soluble WNT analog Norrin.

[0161] Blindness in Norrie disease is caused by underdevelopment of the deep retinal vasculature. Patients with Norrie disease and adult Norrie disease mice are known to have cochlear vascular pathology. The schematic diagram below shows the vasculature of the cochlea (A) and retina (B, C).

[0162] Ndp-KO of a previously generated Norrie disease model mouse (Berger et al., 1996) recapitulated human Norrie disease. We investigated the early phenotype to identify a therapeutic window.

[0163] Figure 31 shows that retinal pathology begins with the sprouting of the retinal vasculature. The deep vascular plexus fails to develop.

[0164] We have demonstrated that cochlear vascular pathology also begins earlier than previously thought. As shown in Figure 32, as early as P10, there is significant morphological pathology in the spiral ligament vasculature, and the barrier is not established (Bryant et al. 2022), followed by degeneration of cochlear sensory hair cells.

[0165] It can be concluded that optimal salvage can be achieved by early treatment if vascular disease begins early. Figure 33 shows the therapeutic window.

[0166] This raises the following hypotheses: 1) it is possible to generate fully functional tagged NDP gene therapy constructs for proof-of-concept treatment, and 2) the retina and cochlea of ​​Norrie disease mice can be rescued with neonatal treatment.

[0167] Method used in Example 3 1) This gene therapy construct encodes a tagged NDP that allows for tag-based alternative detection and GFP that labels transduced cells (see Figure 34). 2) The constructs were tested for functionality in vitro. 3) The construct was packaged into an AAV vector using a triple transfection protocol (see Figure 35). 4) To establish functionality in vivo, newborn mice were treated at the onset of early vascular development in the retina and inner ear. 5) Mice were followed up for adverse events three times a week. 6) The vascular phenotype of the retina and cochlea was analyzed at 2 months of age.

[0168] result In vitro tests demonstrate the functionality of the construct. 1) The construct labeled transduced HEK293 cells. Anti-flag tag immunostaining colocalized with endogenous GFP (see Figure 36). 2) Western blot analysis revealed that the NDP protein monomer and GFP were expressed as separate proteins, as evidenced by the different molecular weights of the NDP and GFP bands (see Figure 37). 3) Norrin is a secreted signaling protein that induces canonical WNT / β-catenin signaling through the FZ4 complex of LRP5 or LRP6 and TSPAN12 (see Figure 38A). Activation of this receptor complex inhibits the β-catenin destruction complex.

[0169] The ability of the constructs to induce β-catenin signaling through these receptors was demonstrated in vitro using the TopFlash assay, as shown in Figure 38B. Lithium was used as a positive control.

[0170] In vivo administration to mice with Norrie disease was safe and rescued vascular morphology in the eyes and ears.

[0171] As shown in Figure 39, neonatally treated Ndp-KO mice grew and developed normally. The body weight of treated mice was similar to that of littermate controls. The retinal vascular morphology of treated Ndp-KO mice was comparable to that of WT mice, with the formation of three complete layers of the retinal vasculature. As seen in Ndp-KO mice, neovascularization was also prevented by treatment.

[0172] Figure 40 shows that the retinal vascular morphology of treated Ndp-KO mice was comparable to that of WT mice, with three complete layers of retinal vasculature formed. As seen in Ndp-KO mice, neovascularization was also prevented by treatment.

[0173] Figure 41 shows that the cochlear vascular morphology of Ndp-KO mice (C) was comparable to that of WT mice (A). Note the dilated, less branched vessels in Ndp-KO mice (C, arrowheads), and the uniform vessel distribution and comparable diameters in WT and Ndp-KO treated mice (C, arrows) (scale bar represents 100 μm).

[0174] The following conclusions can be made from this study: 1) the tagged NDP gene therapy construct produced functional NDP protein, 2) the vascular morphology of the retina and cochlea of ​​Ndp-KO mice was rescued by neonatal gene therapy, and 3) this construct has potential for clinical application.

[0175] Figure 42 shows the expression of NDP constructs 0, 1, 2, and 3 in HEK293 cells using the EGFP tag.

[0176] References [Table 2] JPEG2025535106000005.jpg209170 JPEG2025535106000006.jpg220170 JPEG2025535106000007.jpg201170 JPEG2025535106000008.jpg199170 JPEG2025535106000009.jpg216170 JPEG2025535106000010.jpg146170 JPEG2025535106000011.jpg208170

[0177] Sequence Listing [ka]

[0178] [ka]

[0179] [ka]

[0180] [ka]

[0181]

change

[0182]

change

[0183]

change

[0184]

change

[0185]

change

[0186]

change

[0187]

change

[0188]

change

[0189]

change

[0190]

change

[0191]

change

[0192]

change

[0193]

change

[0194]

change

[0195]

change

[0196]

change

[0197]

change

[0198]

change

[0199]

change

[0200]

change

Claims

1. A construct comprising a wild-type NDP nucleic acid sequence or a codon-optimized NDP nucleic acid sequence.

2. The construct of claim 1 , wherein the wild-type or codon-optimized NDP nucleic acid sequence is a human NDP open reading frame.

3. 3. The construct of claim 1, wherein the wild-type NDP nucleic acid sequence comprises the nucleotide sequence of SEQ ID NO:

1.

4. 3. The construct of claim 1, wherein the codon-optimized NDP nucleic acid sequence comprises the nucleotide sequence of SEQ ID NO:

3.

5. 3. The construct of claim 1 or claim 2, wherein the NDP nucleic acid sequence comprises a nucleotide sequence having at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:

1.

6. 3. The construct of claim 1 or claim 2, wherein the NDP nucleic acid sequence comprises a nucleotide sequence having at least about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to SEQ ID NO:

3.

7. The construct of any one of claims 1 to 6, wherein the wild-type or codon-optimized NDP nucleic acid sequence is under the control of a CAG promoter.

8. The construct of any one of claims 1 to 6, wherein the wild-type or codon-optimized NDP nucleic acid sequence is under the control of a CBA promoter.

9. The construct of any one of claims 1 to 8, wherein downstream of the wild-type or codon-optimized NDP nucleic acid sequence is a WPRE element.

10. 10. The construct of any one of claims 1 to 9, further comprising one or more of the following elements: 5' and 3' inverted terminal repeats, a self-cleaving P2A linker, a FLAG epitope sequence tagging the C-terminus, a simian virus 40 polyA (SV40 late polyA) sequence, a bovine growth hormone polyadenylation signal (BGHpA), an EGFP open reading frame sequence, and a pUC ori.

11. 11. The construct of any one of claims 1 to 10, wherein the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into one or more of the following vectors: AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV10, AAV-rh8, AAV-rhlO, AAV-rh39, AAV-rh43, AAVAnc80, AAV 2 / ShH10, AAV-S vector.

12. The construct of any one of claims 1 to 10, wherein the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV 2 / ShH10 vector.

13. The construct of any one of claims 1 to 10, wherein the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV 2 / 9 vector.

14. The construct of any one of claims 1 to 10, wherein the wild-type or codon-optimized NDP nucleic acid sequence is incorporated into an AAV-S vector.

15. The construct of any one of claims 1 to 14, formulated for intravenous administration.

16. The construct of any one of claims 1 to 15, formulated for intraocular administration, preferably for intravitreal injection.

17. A construct according to any one of claims 1 to 16, formulated for intracochlear administration, preferably for intracochlear injection.

18. The construct of any one of claims 1 to 17, formulated for at least two of intravenous, intraocular, and intracochlear administration.

19. 16. A construct comprising the wild-type or codon-optimized NDP nucleic acid sequence of any one of claims 1 to 15 for use in treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases.

20. An AAV particle (NDP.AAV) comprising the construct of any one of claims 1 to 19.

21. The AAV particles of claim 20 for use in treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases.

22. The AAV particle of claim 20, which is administered intravenously.

23. 23. The AAV particle of claim 22, which is an AAV2 / 9 particle.

24. The AAV particles of claim 20, which are administered intraocularly.

25. 25. The AAV particle of claim 24, which is an AAV 2 / ShH10 particle.

26. The AAV particle of claim 20, which is administered intracochlearly.

27. 21. The AAV particle of claim 20, which is administered by at least two of intravenous administration, intraocular administration, and intracochlear administration.

28. The AAV particles of claim 20 for use in treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases, wherein the initial dose is administered intraocularly during the neonatal period.

29. 29. The AAV particles of claim 28, wherein the first dose is administered at about 14-17 post-conception weeks (pcws), more preferably at about 15-16 pcws.

30. 30. The AAV particle of any one of claims 28 and 29, wherein the second dose is administered intraocularly during the neonatal period.

31. 31. The AAV particles of claim 30, wherein the second dose is administered at about 22-26 post-conception weeks (pcws), more preferably at about 24 pcws.

32. 30. The AAV particles of any one of claims 28 and 29, wherein the second dose is administered intraocularly at any time after birth.

33. The AAV particles of any one of claims 30 to 31, wherein the third dose is administered intraocularly at any time after birth.

34. The AAV particles of claim 20 for use in treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases, wherein the initial dose is administered intracochlearly during the neonatal period.

35. 35. The AAV particles of claim 34, wherein the initial dose is administered at about 13-20 post-conception weeks (pcws), more preferably at about 15-18 pcws.

36. 36. The AAV particle of any one of claims 34 and 35, wherein the second dose is administered intracochlearly during adolescence, more preferably at or below about 12 years of age.

37. The AAV particles of any one of claims 34 to 35, wherein the second dose is administered intracochlearly in adulthood, more preferably at age 12 or older.

38. 37. The AAV particles of claim 36, wherein the third dose is administered intracochlearly in adulthood, more preferably at age 12 or older.

39. The AAV particles of claim 20 for use in treating one or more of Norrie's disease, age-related hearing loss, diabetic maculopathy and retinopathy, retinal neovascularization, retinal exudation, retinopathy of prematurity (ROP), familial exudative vitreoretinopathy (FEVR), Coats' disease, and other NDP-associated diseases, or other beta-catenin signaling-associated retinal diseases, wherein the initial dose is administered intravenously.

40. 40. The AAV particle of claim 39, wherein the initial dose is administered neonatally or postnatally.

41. 41. The AAV particle of claim 40, wherein the second dose is administered neonatally or postnatally.

42. 42. The AAV particle of claim 41, wherein the third dose is administered neonatally or postnatally.

43. 21. The AAV particle of claim 20, adapted to target fibrocytes of the spiral ligament (SL), basal and marginal cells of the stria vascularis and lateral wall, glial cells of the modiolus; retinal Müller cells, RGCs, RPE cells and PRS, with evidence of rescue after transduction of RGCs alone and / or Müller cells alone.

44. A method for rescuing the vasculature of the ear and eye, comprising administering NDP.AAV particles during early development, preferably prenatal development, or during the first year of life, preferably neonatally, by at least one of intravenous, intraocular, and intracochlear administration.

45. A method for rescuing blood-barrier function in the ear and eye, comprising administering NDP.AAV particles during early development, preferably prenatal development, or during the first year of life, preferably neonatally, by at least one of intravenous, intraocular, and intracochlear administration.

46. A method for preventing or treating ocular neovascularization, comprising administering NDP.AAV particles during early development, preferably prenatal development, or during the first year of life, preferably the neonatal period, by at least one of intravenous and intraocular administration.

47. A method for rescuing the blood-retinal barrier and / or the cochlea and / or endocochlear potential barrier, comprising administering NDP.AAV particles at a later stage of development, preferably postnatal development, by at least one of intravenous, intraocular, and intracochlear administration.

48. A method for treating retinal neovascularization in the retina, comprising administering NDP.AAV particles at a later stage of development, preferably at a postnatal developmental stage, by at least one of intravenous and intraocular administration.

Images