WO2025254596A1 - Modified adeno-associated virus 1 (aav1) - Google Patents

Abstract

The present disclosure provides a modified adeno-associated virus 1 (AAV1) comprising an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. The present disclosure also provides capsid proteins, nucleic acids, and compositions comprising said modified AAV1. The present disclosure also provides a method of treating a corneal disease to a subject, wherein the method comprises administering a therapeutically effective amount of the modified AAV1, or the composition of the disclosure to a subject, wherein the modified AAV1 comprises a payload for treating a corneal disease. Also provided by the present disclosure is a method of delivering a payload to a cornea of a subject, comprising administering the modified adeno-associated virus 1 (AAV1), or the composition of the disclosure, to the cornea of the subject.

Description

MODIFIED ADENO-ASSOCIATED VIRUS 1 (AAV1)

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] Tliis application claims the benefit of priority of Singapore provisional application no. 10202401616Q, filed 06 June 2024, the contents of it being hereby incorporated by reference in its entirety for all purposes.

TECHNICAL FIELD

[0002] The present invention relates generally to the field of biotechnology In particular, the present invention relates to modified adeno-associated virus 1 (AAV1) vectors and uses thereof for delivering a payload and treating comeal disease.

BACKGROUND

[0003] Comeal disorders and abnormalities affect about eight million people worldwide and are the third leading cause of blindness in the world. Comeal diseases and disorders have a broad range of pathologies leading to diverse outcomes and often result in reduced quality of life. The advancement of gene therapy in recent years has led to the treatment as well as prevention of some of these comeal diseases and disorders. However, the precision in the delivery of gene therapeutics, for example, for comeal dystrophies, a leading cause of comeal blindness, has been a major limitation for the therapeutic development. The lack of available tools to guide the improvement of the precision of adeno-associated vims (AAV) capsid delivery' to specific ocular cell types has impeded vector development for clinical applications. Similarly, beyond comeal disorders, gene therapy to other organs also requires efficient and targeted (tissue-specific) delivery vectors for meeting the required therapeutic profile.

[0004] Thus, there is an unmet need for delivery systems providing efficient and specific delivery' of gene therapeutics in corneal diseases.

SUMMARY

[0005] In one aspect, the present disclosure refers to a modified adeno-associated vims 1 (AAV1) comprising an AAV 1 VP3 capsid protein, wherein the AAV 1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[0006] In another aspect, the present disclosure refers to an AAV1 VPS capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[0007] In another aspect, the present disclosure refers to a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV 1 VP3 capsid protein comprises one or more mutations at aposition selected from the group consisting of: S507, and S587, relative to a wild-type AAV 1 capsid protein sequence of SEQ ID NO: 65. [0008] In another aspect, the present disclosure refers to a composition comprising the modified AAV1 of the disclosure and a pharmaceutically acceptable excipient.

[0009] In another aspect, the present disclosure refers to a method of treating a corneal disease, wherein the method comprises administering a therapeutically effective amount of the modified A AVI of the disclosure, or the composition of the disclosure to a subject, wherein the modified AAV1 comprises a payload for treating a corneal disease.

[00010] In another aspect, the present disclosure refers to a method of delivering a payload to a cornea of a subject, comprising administering the modified adeno-associated virus 1 (AAV1) of the disclosure, or the composition of the disclosure, to the cornea of the subject.

[00011] In another aspect, the present disclosure refers to a use of the modified AAV1 of the disclosure, or the composition of the disclosure, in the manufacture of a medicament for treating a comeal disease, wherein the modified AAV1 comprises a payload for treating a corneal disease.

BRIEF DESCRIPTION OF THE DRAWINGS

[00012] Fig. 1 shows the result of the assessment of transduction efficiency of different AAV serotypes in human comeal endothelial cells. Fig. 1A is a histogram graph of the percentage transduction efficiency of AAV serotypes in donor samples of human comeal endothelial cells.

[00013] Fig. IB is a schematic showing the intensity matrix format of the same data in Fig. 1A, where the percentage transduction efficiency of AAV serotypes in donor samples of human comeal endothelial cells is shown.

[00014] Fig. 2 shows the bioinfonnatic workflow for engineering AAV capsid vectors with intracellular stability and transduction efficiency. Fig. 2A is a schematic showing the bioinformatic analysis workflow for logical prediction of surface exposure for protein domains and amino acids in an AAV1 capsid sequence.

[00015] Fig 2B is a graph showing the Root Mean Square Distance (RMSF) scores of residues in AAV1 capsid VP3 region that are predicted to be surface-exposed. White arrows: sialic acid interaction domains. Black arrows: PKD2 interaction domains.

[00016] Fig. 2C is a schematic showing the capsid sequence of AAV1 where amino acid candidate residues identified for further experimental testing are underlined. Amino acid candidate residues identified for further experimental testing w^rere assessed fortheir role in altering transduction efficiency. Amino acid residues which correspond to surface exposed regions are in italics. * denotes the stop codon.

[00017] Fig. 2D is a schematic showing the PyMol analysis of domains to rule out amino acids that arc embedded in the protein structure. The sialic acid binding site and PKD2 binding sites are boxed up.

[00018] Fig. 2E is a schematic showing the PyMol structural illustration of candidate residues in AAV1, centered around the 3-fold region. Amino acids for candidate residues are indicated as follows, Serine(S): rectangle, Threonine(T): circle, Tyrosine(Y): hexagon, and Lysine(K): triangle. [00019] Fig. 3 shows the result of the transduction efficacy of AAV1 variants in ex vivo human comeal samples. Fig. 3A is a bar graph showing the transduction efficiency of pooled barcoded AAV1 variants transduced in human comeal samples and subjected to NGS sequencing and analysis. Results showed that three AAV1 variants (S455V, S507V and S587V) have increased transduction efficacy up to 3-fold, -fold, and 2-fold respectively relative to the wild-type. Results arc based on approximately 2 million reads per sample.

[00020] Fig. 3B is a schematic showing that residues in AAV 1 that alter transduction efficiencies are centered around the 3-fold region Rectangle: S455, Circle: S507, and Triangle: S587.

[00021] Fig. 4 shows the result of transduction efficiency of AAV1 variants when tested in ex vivo human comeal samples. Fig. 4A is a bar graph showing the transduction efficiency of pooled barcoded AAV1 variants with either single, double, ortriple mutation(s) which were transduced in human corneal samples and subjected to NGS sequencing and percentage read analysis. Results showed that the transduction efficiency of the AAV1 double mutation variant (S507V and S587V) is 22-fold higher compared to S455V single-mutation variant. Results arc based on approximately 2 million reads per sample.

[00022] Fig. 4B is a bar graph showing the transduction efficiency fold change of pooled barcoded AAV1 variants with either single, double, or triple mutation(s) transduced in human comeal samples and subjected to NGS sequencing. Results of NGS sequencing were normalized to S455V to derive the transduction fold change of AAV1 variants with either single, double, or triple mutation(s) in comparison to S455V single -mutant variant.

[00023] Fig. 5 shows the result of validating the AAV1-S507V, S587V engineered vector using an assay. In this assay, human cornea tissues are positioned in a manner that mimics the anterior chamber of the cornea. Fig. 5A is a bar graph showing the number of AAV copies in human cadaver cornea samples treated with different doses and variants of AAV 1. Three human cadaver samples were assigned to each treatment condition. Control samples were exposed to 1 x 10^A 10 vgs of AAV1 per cornea for 5mins and replaced with fresh media. Wild-type AAV1 treated group were exposed to 1 x 10^A10 vgs of AAV1 per cornea for 24h and replaced with fresh media for another 48h. Normal dose AAV1(S5O7V,S587V) treated group were exposed to 1 x 10^Al 0 vgs of AAV1(S5O7V,S587V) per cornea for 24h and replaced with fresh media for another 48h. Low dose AAV1(S5O7V,S587V) treated group were exposed to 1 x 10^A9 vgs of AAV1(S5O7V,S587V) per cornea for 24h and replaced with fresh media for another 48h.

[00024] Fig 5B is a bar graph showing the number of AAV copies in human cadaver cornea samples treated with different doses and variants of AAV 1. Three human cadaver samples were assigned to each treatment condition. Control samples were exposed to 1 x 10^A 10 vgs of AAV1 per cornea for 5mins and replaced with fresh media. Wild-type AAV 1 treated group were exposed to 1 x 10 ^A 10 vgs of AAV 1 per cornea for Ih and replaced with fresh media for another 48h. Normal dose AAV1(S5O7V,S587V) treated group were exposed to 1 x 10^A10 vgs of AAVl(S507V,S587V) per cornea for Ih and replaced with fresh media for another 48h. Low dose AAV1(S5O7V,S587V) treated group were exposed to 1 x 10^A9 vgs of AAV1 (S507V.S587V) per cornea for Ih and replaced with fresh media for another 48h

[00025] Fig. 6 shows the result of a human corneal endothelial primary cell wound healing assay using AAV1(S5O7V, S587V)-SOX2 and AAV1(S5O7V, S587V)-GFP. Fig. 6A is a schematic showing the scratch wound assay set-up using primary human corneal endothelial cells, for transduction with different AAV variants and creation of a scratch wound created after 24h transduction. Wound closure rate was measured for 5 days using the formulae shown in Fig. 6 A, in mm2/hr.

[00026] Fig. 6B is a bar graph showing the wound closure rate for different AAV 1 variants. Differences between multiple groups were analyzed by one-way ANOVA. p-value was considered statistically significant as follows: *adj-p< 0.05.

[00027] Fig. 7 shows the result of a human corneal endothelial primary cell wound healing assay using AAV1(S5O7V, S587V)-SOX2 and AAV1(S5O7V, S587V)-GFP at a dosage of 10^A10vgs. Fig. 7A is a schematic showing the scratch-and-peel wound healing assay set-up. Human cadaver samples were transduced with different AAV 1 variants and scratch wounds were created after 24h transduction and 5 days of ROCKi treatment.

[00028] Fig. 7B are stereomicroscope images of wound closure rates taken at different time-points for two assays.

[00029] Fig. 7C is a graph showing the wound closure rate measured by the percentage of wounded area (%wounded area) for different transduction conditions at different time-points. Wound healing rate for AAV1(S507V,S587V)-SOX2 when compared to control AAV1(S5O7V,587V)-GFP at a dose of 10^A10vgs were found to be significantly improved. Differences between the two groups (GFP and SOX2) were analyzed by multiple paired studentt-test (two-tailed), p-value was considered statistically significant as follow: *p < 0.05, **p < 0.01. AAV1 variant with double mutations (S507V, S587V) delivering SOX2 gene results in a greater decrease of wounded area relative to when GFP gene is delivered.

[00030] Fig. 8 is a schematic which shows the design of AAV genomic cargo sequence for serotypes or variants barcoding. Fig. 8 is a schematic of the design of AAV genomic cargo for capture and analysis of serotype barcodes. A mammalian promoter is selected for expression of a non-host protein in the human organoid cells. An eGFP transgene with barcode is expressed and can be distinguished from host gene transcripts. A unique 8 base-pair barcodes with Hamming Distance of more than 5 betw een the pooled barcodes is included after the stop codon and before the polyadenylation tail. The list of barcodes representing each serotype or variant can be found in Tables 2 and 4 respectively. GFP: green fluorescent protein; Poly-A: polyadenine tail.

DETAILED DESCRIPTION

[00031] To address the unmet need for efficient and specific delivery systems for gene therapeutics in comeal diseases, the present disclosure provides a modified adeno-associated vims 1 (AAV1) comprising an AAV 1 VP3 capsid protein, wfterein the AAV 1 VP3 capsid protein comprises one or more mutations at positions including, but not limited to, S507, or S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. Such a delivery system allows for efficient and specific transduction of a payload, such as a therapeutic gene, a siRNA, a CRISPR-Cas9 system, a nucleic acid, or combinations thereof, to a human corneal cell

[00032] As used herein, the term “AAV” is used interchangeably with the term “adeno -associated virus”, and refers to a viral vector used in gene therapy, such as for the treatment or prevention of corneal diseases.

[00033] AAVs comprise a protein shell referred to as the “capsid”. As used herein, the term “capsid” refers to an icosahedral protein envelope which forms the general structure of the vims. The capsid is formed from overlapping viral proteins (VP) that overlap to form the capsid structure. Examples of viral proteins (VP) can be, but are not limited to, VP1, VP2, and VP3. The VP3 structure comprises a core eight-stranded p-barrel motif (PB-pI) and a small a-helix (aA), where variable regions of VP3 can be found on the capsid surface and are involved in receptor binding, transduction and antigenic specificity. [00034] The capsid of an AAV can further comprise a payload, which can be, but is not limited to, a nucleic acid, a CRISPR-Cas9 system, or combinations thereof. Examples of nucleic acids that can be encapsulated by the capsid can be, but are not limited to, a gene of interest, a therapeutic gene of interest, a siRNA, a wild-type viral genome, a mutated viral genome, or any combinations thereof.

[00035] The payload encapsulated by an AAV can be delivered upon infection of a host cell. As used herein, the term “host cell” refers to a cell to which an AAV can infect and deliver a payload which can be, but is not limited to, a nucleic acid, a CRISPR-Cas9 system, or combinations thereof. Examples of host cells can be, but are not limited to, comeal, comeal stroma, comeal epithelial, comeal endothelial, comeal tissue cells, or any combinations thereof.

[00036] The infection of a host cell by an AAV is mediated by binding of an AAV to a receptor on the host cell via the capsid protein of the AAV. Upon binding, the AAV is internalized into the host cell by cndocytosis where degradation of the capsid protein releases the payload contained in the capsid protein into the cytoplasm, thereby delivering the payload encapsulated by the AAV. Accordingly, in the context of gene therapy, AAVs can be engineered to encapsulate, for example, a therapeutic gene of interest, a siRNA, a CRISPR-Cas9 system, a nucleic acid, or any combinations thereof, for delivery' into a host cell.

[00037] The rate at which an AAV is able to successfully deliver a payload into a host cell is referred to as the “transduction efficiency”. As used herein, the term “transduction efficiency” refers to the proportion of infected cells within a given pool of cells treated with AAV, and which contains the payload delivered by the AAV. Since infection of a host cell is mediated by binding of an AAV to a receptor on the host cell via the capsid protein of the AAV, the transduction efficiency of an AAV can be dependent on factors such as, for example, the proteins on the capsid surface. Accordingly, AAVs can be grouped into different serotypes based on the type of proteins on their capsid surface . Examples of AAV serotypes can be, but are not limited to, AAV1, AAV2, AAV6, AAV7, AAV8, AAV9, AAV-rhlO, AAV-DJ, and AAV-Anc80. In one example, the AAV described herein is AAV1 .

[00038] Therefore, prior to beginning the bioinformatics analysis for identifying amino acids potentially involved in transduction efficiency, the AAV serotype with transduction efficiency was determined (example 1). It was demonstrated that of all the tested serotypes, AAV1 showed the highest transduction efficiency in ex vivo human comeal samples (Fig. I ) AAV1 was thus selected as a serotype for further study.

[00039] Within a single AAV serotype, the protein sequence of the capsid or the genomic sequence encoding the capsid protein can vary and also contribute to transduction efficiency. As used herein, the term “modified AAV” is used interchangeably with the term “AAV variant”, or “modified AAV variant”, and refers to a AAV that contains differences in the protein sequence of its capsid or part thereof relative to a reference AAV capsid sequence or part thereof. In one example, the modified AAV contains differences in the protein sequence of its capsid or part thereof relative to a wild-type AAV capsid or part thereof. Examples of modified AAVs can be, but arc not limited to, a modified AAV 1 , AAV AAV2, AAV6, AAV7, AAV8, AAV9, AAV-rhlO, AAV-DJ, or AAV-Anc80. In one example, the modified AAV is a modified AAV1. Examples of wild -type AAVs can be, but are not limited to, a wild-type AAV1, AAV AAV2, AAV6, AAV7, AAV8, AAV9, AAV-rhlO, AAV-DJ, or AAV-Anc80. In one example, the wild-type AAV is a wild-type AAV1. In another example, the modified AAV1 contains differences in the protein sequence of its capsid or part thereof relative to a wild-type AAV1 capsid sequence. In a specific example, the wild-type AAV1 comprises a capsid sequence defined by' SEQ ID NO: 65. In another example, the wild-type AAV1 comprises a VP3 sequence defined by SEQ ID NO: 1.

[00040] The differences in the protein sequence of the capsid or part thereof of a modified AAV relative to a reference AAV can be naturally occurring or engineered, and arc referred to herein as a “mutation”. As used herein, the term “mutation” refers to an engineered or naturally' occurring alteration in a genomic or protein sequence relative to a reference genomic or protein sequence. Examples of mutations can be, but are not limited to, a substitution, a deletion, an addition, or any combinations thereof. In one example, the mutation is a substitution. In another example, the mutation is in a protein sequence. In a further example, the mutation is in the protein sequence of the capsid of an AAV. In yet a further example, the mutation is in the VP3 sequence.

[00041] As shown herein, it was demonstrated that mutations on surface -exposed amino acid residues in VP3 altered transduction efficiencies of modified AAV1 into ex vivo human corneal samples (Fig. 3A). It has also been described herein how mutations that occur in the VP3 of the capsid of an AAV result in variations in the transduction efficiency of the modified AAV. The identification of residues involved in transduction efficiency of modified AAV1 was performed using a bioinformatics workflow (example 2). Following the workflow shown in Fig. 2A, surface-exposed amino acid residues were identified (Fig. 2B, 2D and 2E). Identified residues were mutated to obtain modified AAV1 and the transduction efficiency was measured (Fig. 2C, Fig. 8, examples 2 and 3). [00042] Accordingly, described herein is a modified AAV1 comprising a VP3 mutation relative to a wild-type A AVI capsid sequence defined by SEQ ID NO: 65.

[00043] In one example, the modified AAV 1 comprises a VP3 mutation, where a VP3 mutation refers to a mutation in the VP3 capsid protein. A mutation in the VP3 capsid protein at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, refers to a mutation at these positions relative to the capsid sequence of a wild -type AAV 1 , wherein these positions are further located in the region of the capsid sequence corresponding to the VP3. Thus, in one example, described herein is a modified AAV1 comprising aVP3 mutation at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, relative to the wild-type AAV1 capsid protein sequence defined by SEQ ID NO: 65. In a further example, VP3 mutations at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, will correspond to the 251^th position, 305^th position, 385^th position, and the 305^th and 385^th positions respectively in SEQ ID NO: 1 . The residues at the 455^th position, the 507^th position, and the 587^th position were identified to be serine and arc referred to as S455, S507, and S587 respectively. S455 and S587 were shown herein to be exposed on the exterior of capsid while S507 was positioned at the base of the spikes of the capsid protein (Fig. 3B).

[00044] Mutations at S455, S507, and S587 were demonstrated herein to alter transduction efficiency. Thus, in one aspect, disclosed herein is a modified adeno-associated virus 1 (AAV1) comprising an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at positions which can be, but are not limited to, S455, S507, or S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In another aspect, disclosed herein is a modified adeno-associated virus 1 (A AVI) comprising an AAV1 VP3 capsid protein, wherein the A AVI VP3 capsid protein comprises one or more mutations at positions which can be, but arc not limited to, S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[00045] As shown herein in example 3, it was shown that single mutations of two serine residues to valine at S455, S507, and S587 (referred to as S455V, S507V, and S587V respectively), resulted in increased gene transduction efficiency transduction of AAV1 in ex vivo human cornea samples relative to the wild-type (Fig. 3A). Fig. 3A shows that transduction of ex vivo human cornea samples was increased in AAV1 variants with mutations of serine residues in the VP3 region, specifically residues S455V, S507V and S587V.

[00046] Tims, in one example, described herein is a modified adeno-associated virus 1 (A AVI) comprising an AAV 1 VP3 capsid protein, wherein the AAV 1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to, S455V, S507V, or S587V, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In another example, described herein is a modified adeno- associated vims 1 (AAV1) comprising an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to, S507V, or S587V relative to a ivi Id-type AAV1 capsid protein sequence of SEQ ID NO: 65. In one specific example, disclosed herein is a modified AAV1 comprising a VP3 protein sequence as defined by SEQ ID NO: 3. In another specific example, disclosed herein is a modified AAV1 comprising a VP3 protein sequence as defined by SEQ ID NO: 4. In another specific example, disclosed herein is a modified AAV 1 comprising a VP3 protein sequence as defined by SEQ ID NO: 5

[00047] Also described herein arc other mutations with other amino acids that also comprise a hydrophobic side chain, which resulted in a similar improvement in gene transduction efficiency. In one example, disclosed herein is a modified AAV1 comprising an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to, S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587I, S507M, S587M, S507P, S587P, S507W, S587W, S507F, S587F, or any combinations thereof, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[00048] Dual or tnple mutations of senne residues at S455, S507, and S587 were next analysed for alterations in transduction efficiency relative to single mutations. As shown herein, dual or triple mutations at positions such as, for example, S455, S507, or S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65, resulted in a further increase in transduction efficiency relative to a single mutation (example 4, Fig. 4). Accordingly, in one example, disclosed herein is a modified AAV1 comprising mutations at S455 and S507, S455 and S587, or S507 and S587, relative to a wild-type AAV1 comprising a capsid protein sequence of SEQ ID NO: 65. In another example, disclosed herein is a modified AAV1 comprising mutations at S455, S507, and S587 relative to a wild-type AAV1 comprising a capsid protein sequence of SEQ ID NO: 65.

[00049] A modified AAV1 comprising dual mutations was demonstrated to have an increased gene transduction efficiency compared to wild-type AAV1 (example 5). The increase in gene transduction efficiency also enabled the delivery of the same number of gene copies with a lower dose compared to when wild-type AAV1 was used (Fig. 5A and 5B). Thus, in one example, disclosed herein is a modified adeno-associated virus 1 (AAV1) capable of delivering the same number of gene copies with a lower dose compared to wild-type AAV1 (e.g. a wild-type AAV1 comprising a capsid protein sequence of SEQ ID NO: 65, or a VP3 sequence of SEQ ID NO: 1). In one example, the dosage for delivery' of the modified AAV 1 is 5 to 10 -fold vector genomes (vgs) lower per cornea relative to the dosage of a wildtype AAV1 comprising a capsid protein sequence of SEQ ID NO: 65, or a VP3 capsid protein sequence of SEQ ID NO: 1.

[00050] Accordingly, also disclosed herein is a modified AAV1 comprising a S507V and a S587V mutation relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In one example, disclosed herein is a modified AAV1 comprising a VP3 capsid protein sequence as defined by SEQ ID NO: 2.

[00051] In another aspect, also disclosed herein is an AAV1 VP3 capsid protein. In one example, the AAV 1 VP3 capsid protein is that of a modified AAV 1. In another example, the AAV 1 VP3 capsid protein comprises a mutation. As used herein, a mutation in the AAV1 VP3 capsid protein at the 455¹¹' position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, refers to a mutation at these positions relative to the capsid protein sequence of a wild-type AAV1 , wherein these positions are further located in the region of the capsid protein sequence corresponding to the VP3. In one example, a mutation in the AAV1 VP3 capsid protein at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, refers to a mutation at these positions relative to the capsid protein of a wild -type AAV1 comprising a sequence defined by SEQ ID NO: 65. In a further example, mutations in the AAV1 VP3 capsid protein at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, will correspond to the 251^th position, 305^th position, 385^th position, and the 305^th and 385^th positions respectively in SEQ ID NO: 1.

[00052] In one example, described herein is an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein compnses one or more mutations at positions which can be, but are not limited to, S455, S507, or S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. Tn another aspect, described herein is an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at positions which can be, but are not limited to, S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[00053] In another example, disclosed herein is an AAV1 VP3 capsid protein comprising one or more mutations which can be, but are not limited to, S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587I, S507M, S587M, S507P, S587P, S507W, S587W, S507F, or S587F, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In a further example, disclosed herein is an AAV1 VP3 capsid protein comprising one or more mutations which can be, but are not limited to, S455V, S507V, or S587V, relative to a wild-type A AVI capsid protein sequence of SEQ ID NO: 65. Tn a further example, disclosed herein is an AAV 1 VP3 capsid protein comprising one or more mutations which can be, but are not limited to, S507V, or S587V, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[00054] In one example, disclosed herein is an AAV1 VP3 capsid protein comprising a sequence as defined by SEQ ID NO: 3. In another example, disclosed herein is an AAV1 VP3 capsid protein comprising a sequence as defined by SEQ ID NO: 4. In another example, disclosed herein is an AAV1 VP3 capsid protein comprising a sequence as defined by SEQ ID NO: 5.

[00055] In yet a further example, disclosed herein is an AAV 1 VP3 capsid protein comprising a S507V and a S587V mutation relative to a wild-type A AVI capsid protein sequence of SEQ ID NO: 65. Tn one example, disclosed herein is an AAV1 VP3 capsid protein comprising a sequence as defined by SEQ ID NO: 2.

[00056] Nucleic acids encoding the modified VP3 capsid protein of an AAV 1 are also described herein. As described herein, nucleic acids encoding the modified VP3 capsid protein of an AAV1 result in one or more mutations at positions corresponding to the 455^th, 507^th, or 587^th positions in tire encoded capsid protein sequence of the AAV1. In one example, the 455^th, 507^th, or 587^th positions are located in the region of the capsid sequence corresponding to the VP3. In another example, the 455^th, 507^th, or 587^th positions are relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. Tn a further example, mutations in the encoded capsid protein sequence of the AAV1 at the 455^th position, the 507^th position, the 587^th position, the 507^th and 587^th position, or any combinations thereof, will correspond to the 251^th position, 305^th position, 385^th position, and the 305^th and 385^th positions respectively in SEQ ID NO: 1

[00057] In one example, mutations in the nucleic acid which result in one or more mutations at the 455^th position, the 507^th position, or the 587^th position of the encoded VP3 capsid protein, occur at nucleotide positions 1363 to 1365, 1519 to 1521, or 1759 to 1761 respectively of SEQ ID NO: 66. In a further example, mutations in the nucleic acid which result in one or more mutations at the 455^th position, the 507^th position, or the 587^th position of the encoded VP3 capsid protein, occur at nucleotide positions 757 to 759, 913 to 915, or 1153 to 1155 respectively of SEQ ID NO: 6.

[00058] Thus, in one aspect, disclosed herein is a nucleic acid sequence encoding an A AVI VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations at positions which can be, but are not limited to, S455, S507, or S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In another aspect, also disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations at positions which can be, but are not limited to, S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

[00059] In one example, disclosed herein is a nucleic acid sequence encoding an AAV 1 VP3 capsid protein, wherein the encoded AAV 1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587T, S507M, S587M, S507P, S587P, S507W, S587W, S507F, or S587F relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65 In another example, disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV 1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to, S455V, S507V, or S587V, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO : 65. In yet another example, disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations which can be, but are not limited to, S507V, or S587V, relative to a wild-type AAV1 comprising a VP3 capsid protein sequence of SEQ ID NO: 65.

[00060] Tn a further example, the mutations are relative to a wild-type capsid protein encoded by a nucleic acid sequence defined by SEQ ID NO: 66. Thus, in one example, disclosed herein is a modified AAV1 comprising a VP3 encoded by a nucleic acid sequence as defined by SEQ ID NO: 8. In another example, disclosed herein is a modified AAV1 comprising a VP3 encoded by a nucleic acid sequence as defined by SEQ ID NO: 9. In another example, disclosed herein is a modified AAV1 comprising a VP3 encoded by a nucleic acid sequence as defined by SEQ ID NO: 10. [00061] In another further example, disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises a S507V and a S587V mutation relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65. In one example, disclosed herein is a nucleic acid sequence encoding an A AVI VP3 capsid protein comprising a sequence as defined by SEQ ID NO: 2. In another example, disclosed herein is a nucleic acid sequence as defined by SEQ ID NO: 7.

[00062] In a further example, disclosed herein is a nucleic acid sequence encoding an AAV 1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises a sequence as defined by SEQ ID NO: 3. In another example, disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises a sequence as defined by SEQ ID NO: 4. In another example, disclosed herein is a nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises a sequence as defined by SEQ ID NO: 5. Tn another example, disclosed herein is a nucleic acid sequence encoding an A AVI VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises a sequence as defined by SEQ ID NO: 2.

[00063] The modified AAV1 as described herein has been demonstrated to show improved gene transduction efficiency. Accordingly, the therapeutic applications of the modified AAV 1 as described herein was tested (examples 6 and 7).

[00064] Following the assay described in Fig. 6A, it was demonstrated herein that the modified AAV 1 comprising, for example, dual mutations S507V and S587V, successfully delivered a therapeutic gene SOX2 to human comeal endothelial primary cells and improved wound closure (Fig. 6B).

[00065] The improved wound healing capacity was also tested in human cornea endothelial tissue (example 7). Following the assay described in Fig. 7A, it was demonstrated herein that the modified AAV1 comprising dual mutations S507V and a S587V results in an increased wound healing capacity when delivering SOX2 relative to GFP in human cornea endothelial tissue (Fig. 7B and 7C).

[00066] Accordingly, the modified AAV1 of the disclosure can be formulated into compositions suitable for administration. The modified AAV1 may be administered with a pharmaceutically acceptable excipient. An "excipient" can include any pharmaceutically acceptable excipient as long as the excipient is compatible with other ingredients of the fomrulation and not injurious to the subject. The pharmaceutically acceptable excipient can comprise any suitable diluent, adjuvant, buffer, stabilizer, or preservative. It will be appreciated that the pharmaceutically acceptable excipient should be non-toxic and should not interfere with the efficacy of the AAV1 or composition described herein. Compositions can be produced, for instance, by means of conventional mixing, dissolving, encapsulating, or granulating, or lyophilising processes.

[00067] The composition can be administered to the subject in any suitable way, including: intracamerally, intrastromally, or subconjunctivally. [00068] The dosage forms of the composition described herein will typically vary depending on the intended use. For example, a dosage form used in the treatment of a disease or a related disease may contain larger amounts of the modified AAV 1 disclosed herein than a dosage form used in the prevention of the same disease. These and other ways in which specific dosage forms encompassed by this disclosure will vary from one another will be readily apparent to a a person skilled in the art. Examples of dosage forms include, but are not limited to: liquid dosage forms suitable for intracameral administration to a patient, including suspensions (e.g., aqueous or non-aqueous liquid suspensions), and solutions; liquid dosage forms particularly suitable for intracameral administration to a patient; and sterile solids (e.g., lyophilised or granulated solids) that can be reconstituted to provide liquid dosage forms suitable for intracameral administration to a patient. Thus, in one example, the modified AAV1 described herein is provided in a form that can be, but not limited to, solutions, suspensions, aqueous liquid suspensions, non-aqueous liquid suspensions, solutions, sterile solids, lyophilised solids, granulated solids, solids for reconstitution or combinations thereof. Suitable liquid compositions generally include a liquid earner such as water. Physiological saline solution may also be included.

[00069] Likewise, while the dosage of modified AAV1 to be formulated into a composition can be determined by a skilled person in the art, exemplary dosages can be between 10⁹ to 10¹⁰ vector genomes (vgs) for each cornea. In another example, the dosage is about 10¹⁰ vector genomes (vgs) for each cornea.

[00070] Thus, described herein is a composition comprising the modified AAV1 of the present disclosure and a phannaceutically acceptable excipient. In one example, the composition comprises one or more pharmaceutically acceptable vehicles or carriers. In another example, the composition further comprises a pharmaceutically acceptable carrier. Therefore, in one example, the composition may finthcr comprise a compound that can be, but not limited to, a pharmaceutically acceptable carrier, a liposomal carrier, an excipient, an adjuvant or combinations thereof.

[00071] In a fiirther example, the modified AAV1 formulated as a composition comprises a payload, which can be a therapeutic gene, a siRNA, a CRISPR-Cas9 system, a nucleic acid, or any combinations thereof. In one example, the CRISPR-Cas9 system targets a gene for treating a corneal disease. In another example, the CRISPR-Cas9 system comprises a guide RNA for targeting a gene for treating a comeal disease.

[00072] As used herein, the targeting of a gene by a CRISPR-Cas9 system refers to the introduction of a mutation, such as a breakage, in a strand of nucleic acid. A breakage in a strand of nucleic acid can be the cleavage of the phosphodicstcr bond between two nucleotides or nucleosides. The introduction of a breakage to a specific site in a strand of nucleic acid can be directed by a guide RNA, such as a guide RNA associated with the CRISPR-Cas9 system. In one example, the guide RNA comprises a sequence that enables its annealing to a target sequence on a target gene, such as the site where the introduction of a mutation is desired. Depending on the site in which the introduction of a mutation is desired, a skilled person in the art will be readily able to design a guide RNA for targeting of a gene. [00073] Upon the annealing of a guide RNA to a target sequence on a target gene, a breakage in the target gene can be introduced, A breakage in the target gene can subsequently lead to alterations in gene expression, such as an increase or decrease in gene expression relative to before the breakage was introduced. Thus, depending on the site in which the breakage is introduced, the expression of a gene can be increased or decreased. Consequently, a disease can be treated by the alteration of expression of a target gene.

[00074] Accordingly, to treat a corneal disease, a skilled person can refer to gene sequences known in the art for designing a guide RNA for a CRISPR-Cas9 system, where the guide RNA and the CRISPR- Cas9 system can be introduced as a payload in the modified AAV1 disclosed herein. In one example, the guide RNA targets the COL8A2 gene, such as the start codon of the COL8A2 gene. In a further example, targeting of the COL8A2 gene results in a decrease in gene expression for treating a comeal disease, such as Fuch’s dystrophy.

[00075] When the modified A AVI comprises a payload, which can be a therapeutic gene, a siRNA, a CR1SPR-Cas9 system, a nucleic acid, or any combinations thereof, the payload can be delivered to treat a disease as part of gene therapy. Thus, in one aspect, disclosed herein is a method of treating a corneal disease, wherein the method comprises administering a therapeutically effective amount of the modified AAV1, or the composition of the disclosure to a subject, wherein the modified AAV1 comprises a payload for treating a comeal disease . Examples of comeal disease can be, but are not limited to, Fuch’s dystrophy, keratitis, Congenital Hereditary' Endothelial Dystrophy (CHED), Posterior Polymorphous Corneal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) corneal dystrophy, kertoconus, a wound, comeal dystrophy, comeal ulcer, or corneal neovascularization. Tn one example, the comeal disease is a wound. In another example, the corneal disease is Fuch’s dystrophy.

[00076] To treat the corneal disease, a person skilled in the art can select a suitable payload based on the available literature. In one example, the payload is a therapeutic gene. Examples of therapeutic genes can be, but are not limited to, SOX2, siRNA, SLC4A11, MDPZ, or any combinations thereof. In one example, the therapeutic gene is SOX2. In another example, the therapeutic gene is SLC4A11, or MPDZ. In yet another example, the payload is siRNA. In another example, the payload is a CRISPR- Cas9 system.

[00077] In a specific example, when the therapeutic gene is SOX2, the disease to be treated is a wound. In another specific example, when the payload is siRNA, the disease to be treated is Fuch’s dystrophy. [00078] Accordingly, disclosed herein is a method of treating a corneal wound, wherein the method comprises administering a therapeutically effective amount of the modified AAV1, or the composition of the disclosure to a subject, wherein the modified AAV1 comprises SOX2.

[00079] In one example, the subject is a human. In another example, the subject is a mouse, rat, monkey, dog, cat, horse, rabbit, or any other animal which has a cornea. [00080] In another example, disclosed herein is the modified AAV1, or the composition of the disclosure, wherein the modified AAV1 comprises a payload, for use in therapy.

[00081] In yet another example, disclosed herein is a use of the modified AAV1, or the composition of the disclosure, wherein the modified AAV1 comprises a payload, in the manufacture of a medicament for treating a corneal disease. Also disclosed herein is the modified AAV1, or the composition of the disclosure, wherein the modified AAV1 comprises a payload, for use in treating a comeal disease.

[00082] In another example, disclosed herein is use of the modified AAV1 in the manufacture of a medicament for treating a disease that is to be treated by gene therapy. In one example, the disease that is to be treated by gene therapy is a comeal disease. Examples of comeal disease that is to be treated by gene therapy can be, but are not limited to, Fuch’s dystrophy, keratitis, Congenital Hereditary Endothelial Dystrophy (CHED), Posterior Polymorphous Comeal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) corneal dystrophy, kertoconus, a wound, comeal dystrophy, comeal ulcer, or comeal neovascularization. Tn one example, the corneal disease is a wound. In another example, the corneal disease is Fuch’s dystrophy.

[00083] The treatment of a corneal disease as described herein can be mediated by the delivery of the payload by the modified AAV 1. Thus, in one aspect, disclosed herein is a method of delivering a payload to a cornea of a subject, comprising administering the modified adeno-associated virus 1 (AAV1) of, or the composition of the disclosure, to the cornea of the subject. In one example, the payload can be, but is not limited to, a therapeutic gene, a siRNA, a CRISPR-Cas9 system, a nucleic acid, or any combinations thereof.

[00084] Depending on the desired clinical outcome, the dosage of modified AAV1 containing the payload can be decided by a a person skilled in the art. Tn one example, the dosage for delivery of the modified AAV 1 is between 10⁹ to 10¹⁰ vector genomes (vgs) per cornea. In another example, the dosage for delivery' of the modified AAV1 is about 10¹⁰ vector genomes (vgs) per cornea.

[00085] The results as shown herein demonstrate the improved gene transduction efficiency of the modified AAV 1 as described herein. Consequently, lower doses of modified AAV 1 can be used for gene delivery' in therapy, such as for preventing or treating human corneal endothelial diseases and disorders. Thus, in yet another example, the dosage for delivery' of the modified AAV1 is 5 to 10 -fold vector genomes (vgs) lower per cornea relative to the dosage of a wild-type AAV1 comprising a VP3 capsid protein sequence of SEQ ID NO: 1.

[00086] The invention as described herein thus provides a modified A AVI which demonstrates transduction efficiency and specificity in human corneal cells and tissues. The disclosure has also demonstrated that the modified AAV1 as described herein is capable of delivering a payload, which can be, therapeutic genes, siRNAs, CRISPR-Cas9 system, nucleic acids, or combinations thereof, for treating corneal diseases. Accordingly, the invention as described herein can be applied for human comeal gene therapeutic applications. [00087] The invention illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, for example, the terms "comprising", "including", "containing", etc. shall be read expansively and without limitation. Additionally, the terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications arc possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the inventions embodied therein herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention.

[00088] As used in this application, the singular form “a,” ‘’an,” and ‘‘the” include plural references unless the context clearly dictates otherwise. For example, the term “a therapeutic gene” includes one or more therapeutic genes, including combinations thereof.

[00089] As used herein, the term “about”, in the context of dosages for delivery' of the modified AAV 1 , typically means +/- 5% of the stated value, more typically +/- 4% of the stated value, more typically +/- 3% of the stated value, more typically, +/- 2% of the stated value, even more typically +/- 1% of the stated value, and even more typically +/- 0.5% of the stated value.

[00090] Throughout this disclosure, certain embodiments may be disclosed in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosed ranges. Accordingly, the description of a range should be considered to have specifically disclosed all the possible sub-ranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed sub-ranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

[00091] Certain embodiments may also be described broadly and generically herein. Each of the narrow er species and sub-generic groupings falling w ithin the generic disclosure also form part of the disclosure. This includes the generic description of the embodiments with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or notthe excised material is specifically recited herein.

[00092] The invention has been described broadly and generically herein. Each of the narrower species and sub-generic groupings falling within the generic disclosure also form part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.

EXAMPLES [00093] Example 1: Identification of the AAV serotype vector for gene delivery into human corneal endothelial cells.

[00094] AAV transduction is initiated through capsid-host receptor binding. Different tissues within the body or even the same tissue type between different species express different surface receptors that result in different transduction efficacy by an AAV vector. Discrepancies between tissues and species result in many conventional approaches conducted in cell culture or mouse models being unable to be successfully translated to human applications. Thus, ex vivo human corneal samples were used as described herein. To identify AAV vector serotypes for gene delivery into the human corneal endothelial cells, ex vivo human corneal samples with barcoded pooled AAV serotypes (Table 2) were used, where Descemet membrane (DM) peel was performed to separate the endothelial cells from the stroma and epithelial layers. NGS libraries were then constructed, and the barcodes were counted to detennine the comparative transduction levels within the pooled AAVs, as described in the experimental section.

[00095] Tn Fig. 1 , human corneas were infected with pooled AAVs for 7 days and the stroma layer is separated from the endothelial layer before processing for individual Amp-scq libraries and scqucnccd on a MiSeq sequencer. Fastq files were analysed using an in-house Python script to count for seroty pe barcodes. The data is based on more than 1 million reads per sample and uses an in-house script for barcode counting. From Fig. 1, AAV1 was identified for delivery of genes into the human cornea endothelial cells.

[00096] The results showed that the percentage of reads representing AAV1 and AAVDJ were higher than the other serotypes in transducing the human comeal endothelial cells (Fig. 1A). By comparison across the 6 donor samples, the result showed that AAV 1 is more consistent than AAVDJ in transducing the comeal endothelial cells across all 6 donor samples (Fig. IB). Hence, AAV1 was selected as the serotype for gene delivery’ into human corneal endothelial cells.

[00097] Example 2: Bioinformatics structural analysis for identification of surface exposed amino acid on AAV1 capsid predicted to be involved in the transduction process.

[00098] To identify surface exposed residues in the AAV1 capsid sequence for mutagenesis study, an in-house bioinformatics workflow was used (Fig. 2A), as described in the experimental section. First, the surface exposure of amino acid residues on the AAV capsid sequence was evaluated via the use of Gromacs, RMSF and RMSD analysis. Domains with high RMSF scores in AAV1 capsid VP3 region were selected for mutagenesis study (Fig. 2B). Within the predicted surface exposed regions, the lysine (K), serine (S), tyrosine (Y) and threonine (T) residues were mutated to glutamate (E), valine (V), phenylalanine (F) and valine (V) respectively (Fig. 2C). Using PyMol software, candidate amino acid residues that were short-listed for mutagenesis were confirmed to not be embedded within the protein domain stmeture (Fig. 2D) and are exposed on the surface of the viral particle using a 3 -fold modeling (Fig. 2E). Site-directed mutagenesis was then carried out to generate single-mutant or combinatorial- mutant variants. Each AAV serotype variant is then individually used to package a unique barcoded GFP transgene driven by a CMV promoter (Fig. 8, Table 4). This design enables a multiplex setup to measure transduction efficiency of each AAV variant through DNA-seq of transduced cells, where relative counts of the uniquely barcoded GFP sequences correlate to transduction efficiencies. The barcodes were designed using a Hamming distance of more than 5 between barcodes of other variants to unambiguously differentiate the barcodes even with background sequencing errors. Out of the thirty- six single-mutants and five combinatorial -mutants that were constructed, twenty-nine single-mutants and three combinatorial-mutants generated viral yields for further analyses (Table 4).

[00099] Example 3: Single mutation variants exhibit transduction efficacy to corneal endothelial cells in ex vivo human corneal samples.

[000100] The transduction potential of the twenty -nine single-mutants and three combinatorial -mutants of AAV1 vector were assessed in ex vivo human comeal samples. The S455V, S507V and S587V single mutants showed higher transduction efficacy compared with the AAV1-WT vector as assessed by NGS read counting of barcoded vector genome (Fig. 3 A). The improved efficacy was observed across 2 donor samples out of 4 donor samples tested for S455V single-mutant and across all four donor samples for S507V single-mutant and S587V single-mutant. The increase in efficacy is between 2.71 to 3.71-fold across the two samples for S455V single -mutant, between 2.43 to 5.83-fold across the four samples for S507V single-mutant and 1.87 to 2.61-fold across the four samples for S587V single-mutant. The location of the three amino acid residues on the surface of the AAV1 capsid protein were then visualized using the PyMol software (Fig. 3B). Amino acid residues S455 and S587 were found to be exposed on the exterior of the spikes on the capsid while amino acid residues S507 were positioned in the groove at the base of the spikes of the capsid protein.

[000101] Example 4: Combinatorial variants identified for improved transduction efficacy to corneal endothelial cells in ex vivo human corneal samples.

[000102] Next, combinatorial variants from the selected mutations S455V, S507V and S587V were created and their transduction efficacy was compared in ex vivo human corneal samples. Using amplicon sequencing, the percentage read depth of each ofthe pooled barcoded AAV1 combinatorial variants was analysed (Fig. 4A). After normalization of the data to S455V for relative comparison, transduction fold change of each of the barcoded AAV1 combinatorial variants in comparison to S455V single-mutant variant was derived (Fig. 4B). It was shown that S507V single mutation has up to 8.55-fold improvement while S587V has up to 4.36-fold improvement over the single-mutant S455V. Among the double-mutants, the results demonstrated that the double mutation variant (S507V and S587V) has up to 24.54-fold improvement, S455V and S507V double mutant has up to 3.56-fold improvement while S455V and S587V double mutant has up to 2.19-fold improvement.

[000103] Example 5: Validation of engineered AAV1 double mutant (S507V and S587V) using an assay and demonstration of lower dosage treatment to achieve similar transduction performance as wild-type AAV1 vector.

[000104] To validate the perfonnance of the vectors identified from the high throughput screen, an assay for a high-re solution comparison with the wild-type AAV 1 vector performance was set-up. In this assay, human cadaver samples were positioned in a trephine base to prevent tipping of the cornea and to mimic the anterior chamber of the cornea during intracameral injection treatment. A PBS moat was included to avoid drying out of the cornea. AAVs were introduced to the anterior chamber in a 200pl volume for 24h and media was changed and incubated for another 48h. For the control, AAVs were removed after 5mins of exposure and incubated with fresh media. After 72h, the cornea DM was peeled and analysed by qPCR for AAV copy number and cell diploid number. The assay mimics the corneal endothelial cells’ exposure to AAVs during an intracameral injection treatment. After 24h treatment followed by 48h incubation, the three corneas were harvested from each treatment group and Descemet membrane peeling was performed for analysis of the AAV transduction efficiency into the human comeal endothelial layer. The results showed that the control sample’s background has no AAV copy detected while the wild-type AAV1 yields a mean value of 36 AAV copies/diploid (Fig. 5A). Tire engineered AAV1(S5O7V, S587V) vector at the same dosage as wild-type AAV1 of l x 10' l Ovgs per cornea yielded a mean value of 546 AAV copies/diploid with a transduction improvement of up to 15- fold in the human cadaver samples (Fig. 5A). At a lower dosage of IxlOMOvgs, the engineered AAV1(S5O7V, S587V) vector yielded a mean value of 15 AAV copies/diploid (Fig. 5A). The ~15-fold enhanced improvement by the engineered vector indicates that a more efficient gene transduction efficiency can be obtained at the same dosage. This will also allow for decreased costs in manufacturing, and can help to increase clinical trial sizes and avoiding or minimizing adverse immune reactions against the therapeutics in patients. Similar trends were also observed with Ih treatment protocol (Fig. 5B)

[000105] Example 6: AAV1(S5O7V, S587V)-SOX2 improves the wound closure rate compared to AAV1-SOX2 using a human corneal endothelial primary cell wound healing functional assay.

[000106] To investigate if the improved transduction efficacy provided by the engineered AAV1(S5O7V,S587V) vector can improve the wound closure rate, a wound healing assay using human corneal endothelial primary cells was set up (Fig. 6A). The results showed that when the primary cells were transduced with 10^A10vgs of AAV1(S507V,S587V)-SOX2 and scratched, there is significant improvement to the wound closure compared to control AAV1(S5O7VS587V)-GFP (Fig. 6B). The improvement in wound closure is also significant when compared to AAV1-GFP conditions using Oneway ANNOVA testing .

[000107] Example 7: AAV1(S5O7V, S587V)-SOX2 improves wound healing at a dose of 10^A10vgs as demonstrated in a scratch-and-peel human cadaver assay.

[000108] To confirm that the same wound healing capacity is recapitulated in the human cornea, a scratch-and-peel human cadaver assay was set up to assess wound healing (Fig. 7A). At the dose of 10^A10vgs, significant improvement (p < 0.05) in wound closure rate for the SOX2 condition was observed compared to the GFP control using the AAV1(S5O7V,587V) vector for transduction (Fig. 7B and 7C). The result demonstrates that the engineered AAV1(S5O7VS587V) vector is able to significantly improve the wound healing when applied to human comeal endothelial tissue. EXPERIMENTAL SECTION

[000109] Bioinformatics pipeline for capsid surface protein analysis

[000110] Molecular Dynamics (MD) simulations of AAVs VP3 capsid protein. AAVs VP3 capsid proteins were subjected to MD simulations using Gromacs (version 2018.1) package. The respective crystal structures for AAV1 (PDB ID: 5cgc) were taken from RCSB Protein Data Bank. The CHARMM27 force field was applied. All systems were solvated in cubic water box using the minimum with Simple Point Charge (SPC) water model. 3 sodium ions were added to neutralize the entire system ensuring overall charge neutrality. 500 steps were specified to remove low van der Waals contact via a steepest descent algorithm energy minimization. Finally, the equilibrated system was put through a production run at 300K and 1 bar pressure for lOOOOps (10ns). Dynamic behavior and stability of each residue in the VP3 capsid protein were subsequently analyzed including Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) using Gromacs in-built analysis software. RMSD is used for measuring the difference between the backbones of a protein from its initial structural conformation to its final position. RMSD ensures that protein is consistently stable after a certain time point. RMSF measures the fluctuations about the residue/element equilibrium above its average position. RMSF score is averaged over time and usually corresponds to the crystallographic B-factors. Possible regions involved in the secondary and potential surface residues can be identified from RMSF’s values. The higher the RMSF score, the more flexible the regions are, indicating a potential surface residue. RSMF interpretations were cross-checked against the current literature and visualization of the capsid protein. Pymol (The PyMOL Molecular Graphics System, Schrodinger LLC, http://www.pymol.org/) was next used to render and generate depictions of amino acid positions and cartoon models and cross-checked against current literature indications of 5-fold, 3-fold axis, capsid protein residues interactions from vipcrdb3.0 and surface interaction domains (such as Sialic Acid and PKD-2). Amino acid residues within domains that are known to affect the proteasome degradation pathway were identified. Additionally, AAV2, AAV3 and AAV6 PDBs (ID: 6ih9, 3kic, 3oah respectively) were collected to run a MD simulation and generate depictions of amino acid positions/cartoon models similarly for each AAV. Their VP3 sequences were also aligned using Clustal Omega (vl.2.4) (https://www.ebi.ac.uk/Tools/msa/clustalo/).

[000111] Hamming Distances between Barcodes

[000112] Barcodes consisting of 8 bases were generated randomly for tagging the AAV serotypes and variants. To ensure that there is minimal spilling-over of barcodes due to error in sequencing, each barcode has a minimum hamming distance of 5 to the other barcodes within the same pool via an inhouse python script. The selected barcode is then used to represent the respective serotypes (Table 1) or variants (Table 4).

[000113] Plasmids cloning

[000114] For the wildtype AAV barcoded panel: The barcoded eGFP plasmids were constructed by introducing a short sequence TAATAAATCGATCGNNNNININNN (SEQ ID NO: 67) after the eGFP transgene stop codon in the plasmid backbone pZac2.l-CMV-eGFP.igb, a gift from Luk Vandenbeighe. Primers with overhanging barcode were designed for first round PCR to generate barcoded eGFP fragments that terminate at ITR sequences. A second round of nested PCR amplifies shorter fragments of barcoded eGFP which are digested with restriction enzyme Nbel and Bam HL Digested fragments arc ligated with the vector backbone which is digested using the same restriction enzymes. The sequences of the clones were checked by Sanger sequencing. The representing barcodes for each AAV serotype are shown in Table 2, The serotype-specific pAAV-RepCap plasmids were constructed by cloning tlie Cap genes from the different serotypes into the pAAV-RepCap backbone using Gibson assembly. The different serotypes Cap genes were ordered as gene blocks (IDT) and cloned into HindlH/Pmel-digested pAAV-RepCap backbone via Gibson assembly to construct the pAAV-RepCap with the different serotypes Cap genes.

[000115] For the engineered AAV 1 barcoded panel: Primers were designed bearing the single mutation using the software on NEB website (http:/7nebasechanger.neb.conf). Using the pRepCap l plasmid as backbone, a full-length plasmid amplification by PGR using Q5 polymerase (NEB) and the following condition was set. 98C Imin, 25 cycles of 98C 15s, 60C - 30s (use annealing temperature suggested by NEB software), 72. C - 4mins. 72C -- lOmins. lire PCR products were separated using a gel and extracted using a PCR clean-up kit (Promega). Tire extracted products were subjected to Dpnl (NEB) digestion and T4 PNF (NEB) phosphorylation in T4 ligase buffer for an hour at 37°C. Following that, ligation using T4 ligase (NEB) were performed at RT for 2 hours and transformed using DH5a cells (Invitrogen) and plated on to the ampicillin resistance plates for growing the clones. The minipreps of the cloned plasmid bearing the mutation(s) were sent for Sanger sequencing for sequence verification. [000116] AAV production

[000117] AAV viruses from different serotypes or bearing different mutations each bearing its own barcode were produced as per standard protocol. Briefly, AAV were packaged via a triple transfection of 293 AAV cell line (Cell Biolabs AAV- 100) that were plated in a HYPERFlask ‘M’ (Coming) in growth media consisting of DMEM+glutaMax+pyruvate+10%FBS (Thermo Fisher), supplemented with IX MEM non-essential amino acids (Gibco). Confluency at transfection was between 70-90%. Media was replaced with fresh pre-wanned growth media before transfection. For each HYPERFlask *M’, 200 pg of pHelper (Cell Biolabs), 100 pg of pRepCap [encoding capsid proteins for different serotypes or variants], and 100 pg of pZac-CASI-GFP (barcoded) were mixed in 5 ml ofDMEM, and 2 mg of PEI “MAX’’ (Polysciences) (40 kDa, 1 ing/ml in HzO, pH 7 1) added for PEI: DNA mass ratio of 5: 1 The mixture was incubated for 15 min, and transferred drop-wise to the cell media. The day after transfection, the media was changed to DMEM+glutamax+pyruvate+2%FBS. Cells were harvested 48- 72 hrs after transfection by scrapping or dissociation with 1*PBS (pH7.2) + 5 mM EDTAand pelleted at 1500 g for 12 min. Cell pellets were resuspended in 1-5 ml of lysis buffer (Tris HC1 pH 7.5 + 2 mM MgCl + 150 mM NaCl), and freeze-thawed 3 - between dry-ice-etihanol bath and 37 °C water bath. Cell debris was clarified via 4000 g for 5 min, and the supernatant collected. The collected supernatant was treated with 50 UAn! of Benzonase (Sigma-Aldrich) and 1 U/ml of RNase cocktail (Invitrogen) for 30 min at 37 °C to remove unpackaged nucleic acids. After incubation, the lysate was loaded on top of a discontinuous density gradient consisting of 6 ml each of 15%, 25%, 40%, 60% Optiprep (Sigma- Aldrich) in a 29.9 ml Optiseal polypropylene tube (Beckman-Coulter). The tubes were ultra-centrifuged at 54000 rpm, at 18 °C, for 1 .5 hr, on a Type 70 Ti rotor. The 40% fraction was extracted and dialyzed with IxPBS (pH 7.2) supplemented with 35 mM NaCl, using Amicon Ultra- 15 (100 kDa MWCO) (Millipore). The titer of the purified AAV vector stocks were determined using real-time qPCR with ITR-sequence-specific primers and probe, referenced against the ATCC reference standard material 8 (ATCC).

[000118] Research-Grade Human Corneoscleral Tissue

[000119] Research-grade human cadaveric corneoscleral tissues were obtained from Lions World Vision Institute (Tampa, FL, USA) or Saving Sight, Missouri Lions Eye Research Foundation (Kansas City, MO, USA) and transported to the Singapore Eye Research Institute laboratory' via air freight. All specimens were preserved in Optisol solution (Bausch & Lomb, Rochester, NY, USA) within a corneal viewing chamber en-route. All corneas were incubated in a wash buffer solution containing gentamicin (25pg/mL), amphotericin B (Ipg/mL) and IX Penicillin/Streptomycin (all components from Life Technologies, Thermo Fisher Scientific Corporation, Carlsbad, CA, USA) for three cycles of 5 minutes each before use.

[000120] Transduction of ex vivo human cornea samples

[000121] A pool of AAV serotypes was created by pooling each AAV serotype at 1 x IO¹⁰ vg, giving a final viral copy of 9 x IO¹⁰ that is used for the transduction of ex vivo human cornea sample in each well of a 12-well plate. A AVI, 2, 6, 7, 8, 9, rhlO, DJ and Anc80 serotypes were used for the pooling. Human cornea samples were transduced with pooled AAV s for 24 hrs with media changed every 48 hrs and samples were incubated for 7 days before harvesting for NGS library preparation. For assessment of AAV1 variant transduction performance, AAV variants were pooled to a final total viral copy of 1 x 10¹⁰ vg and are used for the transduction of ex vivo human cornea sample in each well of a 12-well plate. Thirty -three AAV1 and variants were used for the pooling (Table 4). Human cornea samples were transduced with pooled AAVs for 24 hrs with media changed every 48 hrs and samples were incubated for 7 days before harvesting for NGS library preparation.

[000122] NGS amplicon library preparation, sequencing and analysis

[000123] Descemet membrane peel was performed on the transduced cornea samples and separated into the DM sample and stroma/cpithchal sample. The samples were excised to the size of Icmxlcmxlcm, and total genomic material was extracted by following the QuickExtract protocol (Lucigen. Following extraction, a first round PCR reaction was set up to amplify eGFP barcoded region using specific primers (Table 3). Reaction mix was prepared as follows: Extracted DNA 1 pl, Amplicon PCR Forward Primer 10 pM Ipl, Amplicon PCR Reverse Primer 10 pM Ipl, Water 9.5pl, 2x Q5 Hotstart mix 12.5pl Total 25 pl. PCR was performed in a thermal cycler using the following program: • 95°C for 2 minutes • 20 cycles of: — 95°C for 30 seconds — 55°C for 30 seconds — 72°C for 30 seconds • 72°C for 5 minutes • Hold at 4°C. Total PCR product were ran on 1 % agarose gel to check and excise the correct band size (150bp). Excised DNA was purified using a gel purification kit.

[000124] Using 2^nd round PCR primers with P5 and P7 adapter region (Table 3), 1 Ong of template was used to set up a second PCR reaction as follows: DNA ( 1 Ong/pl) 1 pl, Amplicon PCR P7 Forward Primer 10 LIM 1 pl, Amplicon PCRP5 Reverse Primer 10 pM 1 pl. Water 9.5pl, 2x Q5 Hotstartmix 12.5 pl Total 25 pl. PCR was performed on a thermal cycler using the following program: • 95°C for 2 minutes • 20 cycles of: — 95°C for 30 seconds — 55°C for 30 seconds — 72°C for 30 seconds • 72°C for 5 minutes • Hold at 4°C. Total PCR products were ran on 1% agarose gel to check and excise the correct band size (~210bp). Excised DNA was purified using the gel purification kit. Amplicon libraries were quantitated by Qubit and purity tested on Nanodrop. Amplicon libraries are sequenced on an iSeq and the barcodes were then counted using an in-house python script.

[000125] AAV1 Incubation and Ex Vivo Culture for Scratch Wound Assay

[000126] After incubation in wash buffer, corneas were placed in a corneal trephine holder to maintain the comeal cup structure during AAV1 incubation. Approximately 200pL of M5 culture medium was added to the comeal cup, followed by either AAV 1-GFP orAAVl-SOX2, and incubated for 24 hours at 37°C and 5% CO2. To prevent drying up of the media during incubation, the entire holder with the cornea was placed in a 100mm tissue culture dish surrounded by distilled water. The formulation of the M5 culture medium comprises human endothelial serum-free media (Life Technologies) and 5% Fetal Bovine Serum (Atlas Biologicals) with IX antibiotics/antimycotics. Corneas were maintained ex vivo in M5 culture medium after AAV 1 -incubation for the remainder of the experiment, with full media change conducted eveiy two to three days. Rho-kinase inhibitor, Y27632 (Miltenyi Biotec) was supplemented in the culture medium for five days post AAV 1 incubation and removed after wound creation.

[000127] Wound Creation

[000128] To study the effects of over-expression of SOX2 in comeal endothelial cells, a scratch-and- peel method, as described previously was utilized. Briefly, a 2.5mm trephine was used to mark two circular outlines on the cornea endothelial surface. Minimal pressure was applied with the trephine to prevent any injury to the Descemet’s membrane (DM) or underlying comeal stroma. Scratched wounds were created by using a silicon-tipped cannula to gently denude cells without damaging the DM and underlying stroma. Peeled wounds involved an initial creation of a DM tear at the edge of the marked circle, followed by an extension of the tear inward until a complete circle of the DM had been peeled off. All manipulations were performed under direct visualization with a Nikon SMZ1270 stereomicroscope (Nikon Instruments). Trypan blue was used to assist the visualization of DM, endothelial cells, and underlying stroma.

[000129] Imaging and Image Processing [000130] Corneas were imaged immediately following manipulation, and subsequently every three to four days until day 32. A 1 :3 mixture of 0.4% Trypan blue solution (Sigma Aldrich) to culture media was used to visualize the wounds that were created. After 45 seconds of immersion in Trypan blue, wounded sections stained blue while intact comeal endothelium remained clear, images were obtained using the SMZ1270 stcrcomicroscopc with the Nikon DSFi-3 color camera (Nikon Instruments). Automatic white balancing was performed in-camera prior to acquisition of each image. All images were analyzed with the ImageJ software (National Institute of Health, Bethesda, MD, USA). Wound margins were denoted manually, and endothelial recovery was given by the area of wound measured as a percentage of the total initial wound area.

[000131] In Vitro Scratch Wound Assay

[000132] Primary human comeal endothelial cells from a single donor were seeded in triplicates in a 24 well plate at a density' of 500cells/mm². Cells were allowed to adhere and form a flat, confluent monolayer before AAV1 was added. Cells were incubated with AAV1 for 24h at 37°C and 5% CO2 before full media was replaced. Cells were maintained for 5 days in culture before the scratch assay was carried out. On the day of the assay, culture medium was aspirated, and the confluent monolayer was scratched using a 1 OOitL pipette tip perpendicular to the bottom of the well and dragged across the well in a straight line unidirectionally. M5 culture medium was then added back to the wells. Whole-well images were obtained every 2 hours using the Incucyte S3 instrument (Sartorius) and analyzed using ImageJ. Wound closure rate was given by the area of the wound divided by total time taken for the cells to achieve the recovery.

[000133] Ethics Statement

[000134] The protocols conformed to the tenets of the Declaration of Helsinki, and written consent was acquired from the next of kin of all deceased donors regarding eye donation for research.

[000135] Table 1: amino acid and nucleotide sequences of AAV1 VP3, where sites of mutations are underlined, while mutations are both underlined and bolded. [000136] Table 2: Plasmids and its respective barcodes for tagging AAV serotypes

[000137] Table 3: primers for amplifying eGFP barcoded region

[000138] Table 4: viral production of AAV1 variants

[000139] Variants with better production yield (++ or +++) were selected for the comparative transduction study;

[000140] Table 5: Capsid sequence of AAV1, where sites of mutations are underlined.

[000141 [ Table 6: Sequence for constructing barcoded eGFP plasmids | 67 | IAATAAATCGATCGNNANNNNN

Claims

1 . A modified adeno-associated virus 1 (AAV1) comprising an AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

2. The modified AAV1 of claim 1, wherein the AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587I, S507M, S587M, S507P, S587P, S507W, S587W, S507F, and S587F, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

3. The modified AAV1 of claim 1, wherein the AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, and S587V, relative to a wild -type A AVI capsid protein sequence of SEQ ID NO: 65.

4. The modified AAV1 of claim 1, wherein the AAV1 VP3 capsid protein comprises a S507V and a S587V mutation relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

5. An AAV1 VP3 capsid protein, wherein the AAV1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

6. The AAV1 VP3 capsid protein of claim 5, wherein the AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587I, S507M, S587M, S507P, S587P, S507W, S587W, S507F, and S587F, relative to a wildtype AAV1 capsid protein sequence of SEQ ID NO: 65.

7. The AAV1 VP3 capsid protein of claim 5, wherein the AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, and S587V. relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

8. The AAV1 VP3 capsid protein of claim 5, wherein the AAV1 VP3 capsid protein comprises a S507V and a S587V mutation relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

9. A nucleic acid sequence encoding an AAV1 VP3 capsid protein, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations at a position selected from the group consisting of: S507, and S587, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

10. The nucleic acid sequence of claim 9, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, S587V, S507A, S587A, S507L, S587L, S507I, S587I, S507M, S587M, S507P, S587P, S507W, S587W, S507F, and S587F, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

11. The nucleic acid sequence of claim 9, wherein the encoded AAV1 VP3 capsid protein comprises one or more mutations selected from the group consisting of: S507V, and S587V, relative to a wild-type AAV1 capsid protein sequence of SEQ ID NO: 65.

12. The nucleic acid sequence of claim 9, wherein the encoded AAV1 VP3 capsid protein comprises a S507V and a S587V mutation relative to a wild-type AAV 1 capsid protein sequence of SEQ ID NO: 65.

13. A composition comprising the modified AAV1 of any one of claims 1 to 4 and a pharmaceutically acceptable excipient.

14. A method of treating a corneal disease, wherein the method comprises administering a therapeutically effective amount of the modified AAV1 of any one of claims 1 to 4, or the composition of claim 13 to a subject, wherein the modified AAV1 comprises a payload for treating a corneal disease.

15. The method of claim 14, wherein the payload is selected from SOX2 gene, siRNA, SLC4A11 gene, CRISPR-Cas system, or MDPZ gene.

16. The method of claim 14, wherein the comeal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy, keratitis, Congenital Hereditary' Endothelial Dystrophy (CHED), Posterior Polymorphous Comeal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy, or kertoconus.

17. The method of claim 15, wherein if the payload is SOX2 gene, the comeal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy', keratitis, Congenital Hereditary' Endothelial Dystrophy (CHED), Posterior Polymorphous Corneal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy', or kertoconus.

18. The method of claim 15, wherein if the payload is siRNA, the comeal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy, keratitis, Congenital Hereditary' Endothelial Dystrophy (CHED), Posterior Polymorphous Corneal Dystrophy (PPCD), Fuchs Endothelial Corneal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy, or kertoconus.

19. The method of claim 15, wherein if the payload is SLC4A11 gene, the corneal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy, keratitis, Congenital Hereditary Endothelial Dystrophy (CHED), Posterior Polymorphous Comeal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy, or kertoconus.

20. The method of claim 15, wherein if the payload is a CRISPR-Cas system, the comeal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy, keratitis, Congenital Hereditary Endothelial Dystrophy (CHED), Posterior Polymorphous Corneal Dystrophy (PPCD), Fuchs Endothelial Corneal Dystrophy' (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy, or kertoconus.

21. The method of claim 15, w herein if the pay load is MPDZ gene, the comeal disease is selected from the group consisting of: a wound, comeal dystrophy, comeal ulcer, comeal neovascularization, Fuch’s dystrophy', keratitis, Congenital Hereditary' Endothelial Dystrophy (CHED), Posterior Polymorphous Corneal Dystrophy (PPCD), Fuchs Endothelial Comeal Dystrophy (FECD), Transforming growth factor beta-induced (TGFBI) comeal dystrophy', or kertoconus.

22. The method of claim 15, wherein if the payload is SOX2 gene, the comcal disease is a wound.

23. The method of claim 15, wherein if the payload is siRNA, the comeal disease is Fuch’s dystrophy.

24. A method of delivering a paydoad to a cornea of a subject, comprising administering the modified adeno-associated vims 1 (AAV1) of any one of claims 1 to 4, or the composition of claim 13, to the cornea of the subject.

25. The method of any one of claims 14 to 24, wherein the dosage for delivery' of the modified AAV1 is 5 to 10 -fold vector genomes (vgs) low'cr per comca relative to the dosage of a wild -type AAV1 comprising a capsid protein sequence of SEQ ID NO: 65 or a VP3 capsid protein sequence of SEQ ID NO: 1.

26. The method of any one of claims 14 to 24, wherein tire dosage for delivery' of the modified AAV1 is between 10⁹ to IO¹⁰ vector genomes (vgs) per cornea.

27. The method of any one of claims 14 to 24, wherein the dosage for delivery' of the modified AAV1 is about IO¹⁰ vector genomes (vgs) per cornea.

28. Use of the modified AAV1 of any one of claims 1 to 4, or the composition of claim 13, in the manufacture of a medicament for treating a corneal disease, wherein the modified A AVI comprises a payload for treating a corneal disease.