Effective delivery of large genes by dual AAV vectors

RE050283 ยท 2025-01-28

Assignee

Inventors

Cpc classification

International classification

Abstract

The present invention relates to constructs, vectors, relative host cells and pharmaceutical compositions which allow an effective gene therapy, in particular of genes larger than 5 Kb.

Claims

1. A dual construct system to express the coding sequence of a gene of interest in a host cell, said coding sequence consisting of a 5end portion and of a 3end portion, said dual construct system comprising: a) a first plasmid comprising, in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; the 5 end portion of said coding sequence, said 5end portion being operably linked to and under control of said promoter; a nucleic acid sequence of a splicing donor signal; and a 3-inverted terminal repeat (3-ITR) sequence; and b) a second plasmid comprising, in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a splicing acceptor signal; the 3end of said coding sequence; a poly-adenylation signal nucleic acid sequence; a 3-inverted terminal repeat (3-ITR) sequence; wherein the nucleotide sequence of the respective ITRs is obtained from an adeno-associated virus (AAV) of the same AAV serotype or from an AAV of a different serotype; wherein said first plasmid further comprises a nucleic acid sequence of a recombinogenic region in 5 position of the 3ITR of said first plasmid, and wherein said second plasmid further comprises a nucleic acid sequence of a recombinogenic region in 3 position of the 5-ITR of said second plasmid; and wherein the recombinogenic region is an F1 phage recombinogenic region that consists of the sequence: TABLE-US-00007 (SEQIDNO:3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT.

2. The dual construct system according to claim 1, wherein upon introduction of said first plasmid and said second plasmid into the host cell, said coding sequence reconstitutes by means of the splicing donor and the splicing acceptor signals.

3. The dual construct system according to claim 1, wherein the 3-ITR of the first plasmid and the 5-ITR of the second plasmid are from the same AAV serotype.

4. The dual construct system according to claim 1, wherein the 5-ITR and 3-ITR of the first plasmid and the 5-ITR and 3-ITR of the second plasmid are respectively from different AAV serotypes.

5. The dual construct system according to claim 1, wherein the 5-ITR of the first plasmid and the 3-ITR of the second plasmid are from different AAV serotypes.

6. The dual construct system according to claim 1, wherein the coding sequence is split into the 5 end portion and the 3 end portion at a natural exon-exon junction.

7. The dual construct system according to claim 1, wherein the nucleic acid sequence of the splicing donor signal comprises the sequence: TABLE-US-00008 (SEQIDNO:1) GTAAGTATCAAGGTTACAAGACAGGTTTAAGGAGACCAATAGAAACTGGG CTTGTCGAGACAGAGAAGACTCTTGCGTTTCT.

8. The dual construct system according to claim 1, wherein the nucleic acid sequence of the splicing acceptor signal comprises the sequence TABLE-US-00009 (SEQIDNO:2) GATAGGCACCTATTGGTCTTACTGACATCCACTTTGCCTTTCTCTCCACA G.

9. The dual construct system according to claim 1, wherein the first plasmid further comprises at least one enhancer sequence, operably linked to the coding sequence.

10. The dual construct system according to claim 1, wherein the coding sequence is a nucleotide sequence encoding a protein able to correct an inherited retinal degeneration.

11. The dual construct system according to claim 10, wherein the coding sequence is selected from the group consisting of: ABCA4, MYO7A, CEP290, CDH23, EYS, USH2a, GPR98 and ALMS1.

12. A dual viral vector system comprising: a) a first viral vector containing the first plasmid, and b) a second viral vector containing the second plasmid, wherein said first and said second plasmids are as defined in claim 1, and wherein the vectors are adeno-associated virus (AAV) vectors.

13. The dual viral vector system according to claim 12, wherein the adeno-associated virus (AAV) vectors are the same or different AAV serotypes.

14. The dual viral vector system according to claim 12, wherein the AAV vectors have a serotype selected from the group consisting of serotype 2, serotype 8, serotype 5, serotype 7 and serotype 9.

15. An isolated host cell transformed with the dual viral vector system according to claim 12.

16. A pharmaceutical composition comprising the dual construct system according to claim 1, and a pharmaceutically acceptable vehicle.

17. A method for treating a subject having a disease characterized by a retinal degeneration comprising subretinally administering to said subject an effective amount of the dual viral vector system according to claim 12.

18. A pharmaceutical composition comprising the dual viral vector system according to claim 12 and a pharmaceutically acceptable vehicle.

19. A pharmaceutical composition comprising the isolated host cell according to claim 15 and a pharmaceutically acceptable vehicle.

.Iadd.20. A first polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; a 5 end portion of a coding sequence of a gene of interest, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00010 (SEQIDNO:3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.21. The first polynucleotide of claim 20, capable of undergoing homologous recombination with a second polynucleotide that comprises the Fl phage recombinogenic region. .Iaddend.

.Iadd.22. The first polynucleotide of claim 20, wherein: the nucleotide sequence of the 5-ITR and 3-ITR are obtained from an AAV of the same AAV serotype or a different AAV serotype; the nucleotide sequence of the 5-ITR and 3-ITR are obtained from AAV serotype 2; the coding sequence is split into the 5 end portion at a natural exon-exon junction; the nucleic acid sequence of the splicing donor signal comprises TABLE-US-00011 (SEQIDNO:1) GTAAGTATCAAGGTTACAAGACAGGTTTAAGGAGACCAATAGAAACTGGG CTTGTCGAGACAGAGAAGACTCTTGCGTTTCT; the 5 end portion of the coding sequence of the gene of interest is 4.5 Kb, 5 Kb, 5.5 Kb, 6 Kb, or a smaller size; and/or the coding sequence is a nucleotide sequence encoding a protein able to correct an inherited retinal degeneration, optionally wherein the coding sequence is selected from the group consisting of the gene coding sequences of: ABCA4, MYO7A, CEP290, CDH23, EYS, USH2a, GPR98, and ALMS1. .Iaddend.

.Iadd.23. The first polynucleotide of claim 20, wherein the promoter sequence is selected from the group consisting of a cytomegalovirus (CMV) promoter sequence, a chicken beta-actin (CBA) promoter sequence, a vitelliform macular dystrophy 2 (VMD2) promoter sequence, a interphotoreceptor retinoid binding protein promoter sequence, a rhodopsin (RHO) promoter sequence, and a rhodopsin kinase (RHOK) promoter sequence, optionally wherein the first polynucleotide further comprises at least one enhancer and/or intron sequence, operably linked to the coding sequence. .Iaddend.

.Iadd.24. A first viral vector comprising the first polynucleotide of claim 20, optionally wherein the vector is an AAV vector, optionally wherein: the adeno-associated virus is selected from serotype 2, serotype 8, serotype 5, serotype 7, and serotype 9; and/or the AAV vector is AAV2/8. .Iaddend.

.Iadd.25. An isolated host cell transformed with the first viral vector of claim 24. .Iaddend.

.Iadd.26. A pharmaceutical composition comprising the first viral vector of claim 24 and pharmaceutically acceptable vehicle, optionally wherein the pharmaceutical composition is administered to a subject via subretinal administration. .Iaddend.

.Iadd.27. A second polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of a coding sequence of a gene of interest; a poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00012 (SEQIDNO:3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.28. The second polynucleotide of claim 27, capable of undergoing homologous recombination with a first polynucleotide that comprises the Fl phage recombinogenic region. .Iaddend.

.Iadd.29. The second polynucleotide of claim 27, wherein: the nucleotide sequence of the 5-ITR and 3-ITR are obtained from an AAV of the same AAV serotype or a different AAV serotype; the nucleotide sequence of the 5-ITR and 3-ITR are obtained from AAV serotype 2; the coding sequence is split into the 3 end portion at a natural exon-exon junction; the nucleic acid sequence of the splicing acceptor signal comprises TABLE-US-00013 (SEQIDNO:2) GATAGGCACCTATTGGTCTTACTGACATCCACTTTGCCTTTCTCTCCACA G; the 3 end portion of the coding sequence of the gene of interest is 4.5 Kb, 5 Kb, 5.5 Kb, 6 Kb, or a smaller size; and/or the coding sequence is a nucleotide sequence encoding a protein able to correct an inherited retinal degeneration, optionally wherein the coding sequence is selected from the group consisting of the gene coding sequences of: ABCA4, MYO7A, CEP290, CDH23, EYS, USH2a, GPR98, and ALMS1. .Iaddend.

.Iadd.30. The second polynucleotide of claim 27, wherein the poly-adenylation signal nucleic acid sequence comprises a bovine growth hormone (BGH) poly-adenylation signal or a simian virus 40 (SV40) poly-adenylation signal. .Iaddend.

.Iadd.31. A second viral vector comprising the second polynucleotide of claim 27, optionally wherein the vector is an AAV vector, optionally wherein: the adeno-associated virus is selected from serotype 2, serotype 8, serotype 5, serotype 7, and serotype 9; and/or the AAV vector is AAV2/8. .Iaddend.

.Iadd.32. An isolated host cell transformed with the second viral vector of claim 31. .Iaddend.

.Iadd.33. A pharmaceutical composition comprising the second viral vector of claim 31 and pharmaceutically acceptable vehicle, optionally wherein the pharmaceutical composition is administered to a subject via subretinal administration. .Iaddend.

.Iadd.34. A dual construct system to express a coding sequence of a gene of interest in a host cell, the coding sequence having a 5 end portion and a 3 end portion, the dual construct system comprising: a) a first polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; a 5 end portion of the coding sequence of the gene of interest, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00014 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto; and b) a second polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of the coding sequence of the gene of interest; a poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are AAV ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00015 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.35. The dual construct system of claim 34, wherein upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of the splicing donor and the splicing acceptor signals. .Iaddend.

.Iadd.36. The dual construct system of claim 34, wherein: upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of homologous recombination between the Fl phage recombinogenic region of the first polynucleotide and the Fl phage recombinogenic region of the second polynucleotide; or upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of ITR-mediated concatemerization; or upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by one or both of: (i) ITR-mediated concatemerization; and (ii) homologous recombination between the Fl phage recombinogenic region of the first polynucleotide and the Fl phage recombinogenic region of the second polynucleotide, followed by splicing through the splicing donor and the splicing acceptor signals for the production of a mature mRNA. .Iaddend.

.Iadd.37. The dual construct system of claim 34, wherein: the 3-ITR of the first polynucleotide and the 5-ITR of the second polynucleotide are from the same AAV serotype; the 5-ITR and 3-ITR of the first polynucleotide and the 5-ITR and 3-ITR of the second polynucleotide are respectively from different AAV serotypes; the 5-ITR of the first polynucleotide and the 3-ITR of the second polynucleotide are from different AAV serotypes; the nucleotide sequence of the 5-ITR and 3-ITR of the first polynucleotide and the second polynucleotide are obtained from AAV serotype 2; the coding sequence is split into the 5 end portion and the 3 end portion at a natural exon-exon junction; the nucleic acid sequence of the splicing donor signal comprises the sequence: TABLE-US-00016 (SEQIDNO:1) GTAAGTATCAAGGTTACAAGACAGGTTTAAGGAGACCAATAGAAACTGGG CTTGTCGAGACAGAGAAGACTCTTGCGTTTCT; the nucleic acid sequence of the splicing acceptor signal comprises the sequence TABLE-US-00017 (SEQIDNO:2) GATAGGCACCTATTGGTCTTACTGACATCCACTTTGCCTTTCTCTCCACA G; the 5 end portion of the coding sequence of the gene of interest is 4.5 Kb, 5 Kb, 5.5 Kb, 6 Kb, or a smaller size; the 3 end portion of the coding sequence of the gene of interest is 4.5 Kb, 5 Kb, 5.5 Kb, 6 Kb, or a smaller size; the first polynucleotide further comprises at least one enhancer and/or intron sequence, operably linked to the coding sequence; and/or the coding sequence is a nucleotide sequence encoding a protein able to correct an inherited retinal degeneration, optionally wherein the coding sequence is selected from the group consisting of the gene coding sequences of: ABCA4, MYO7A, CEP290, CDH23, EYS, USH2a, GPR98, and ALMS1. .Iaddend.

.Iadd.38. The dual construct system of claim 34, wherein: the promoter sequence is selected from the group consisting of a cytomegalovirus (CMV) promoter sequence, a chicken beta-actin (CBA) promoter sequence, a vitelliform macular dystrophy 2 (VMD2) promoter sequence, a interphotoreceptor retinoid binding protein promoter sequence, a rhodopsin (RHO) promoter sequence, and a rhodopsin kinase (RHOK) promoter sequence; and/or the poly-adenylation signal nucleic acid sequence comprises a bovine growth hormone (BGH) poly-adenylation signal or a simian virus 40 (SV40) poly-adenylation signal. .Iaddend.

.Iadd.39. A dual viral vector system comprising: a) a first viral vector containing the first polynucleotide, and b) a second viral vector containing the second polynucleotide, wherein said first and said second polynucleotides are as defined in claim 34, and wherein the vectors are AAV vectors, optionally wherein: the AAV vectors are the same or different AAV serotypes; the AAV vectors have a serotype selected from the group consisting of serotype 2, serotype 8, serotype 5, serotype 7 and serotype 9; the AAV vector is AAV2/8; the dual viral vector system is capable of transducing one or both of retinal pigment epithelium and photoreceptors; and/or the dual viral vector system is capable of inducing stronger expression of the gene of interest compared to a dual AAV trans-splicing vector system. .Iaddend.

.Iadd.40. An isolated host cell transformed with the dual viral vector system according to claim 39. .Iaddend.

.Iadd.41. A pharmaceutical composition comprising the dual viral vector system of claim 39, and a pharmaceutically acceptable vehicle, optionally wherein the pharmaceutical composition is administered to a subject via subretinal administration. .Iaddend.

.Iadd.42. A method for treating a subject having a disease characterized by a retinal degeneration comprising subretinally administering to the subject an effective amount of the dual viral vector system of claim 39, optionally wherein the disease characterized by a retinal degeneration is Usher 1B and the coding sequence is of a MYO7A gene. .Iaddend.

.Iadd.43. A dual construct system to express a coding sequence of a gene of interest in an host cell, the coding sequence having a 5 end portion and a 3 end portion, the dual construct system comprising: a) a first polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a chicken beta-actin (CBA) promoter sequence; a 5 end portion of the coding sequence of a MYO7A gene, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00018 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto; and b) a second polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of the coding sequence of a MYO7A gene; a bovine growth hormone (BGH) poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are AAV ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00019 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.44. The dual construct system of claim 43, wherein: upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of the splicing donor and the splicing acceptor signals; upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of homologous recombination between the Fl phage recombinogenic region of the first polynucleotide and the Fl phage recombinogenic region of the second polynucleotide; the 3-ITR of the first polynucleotide and the 5-ITR of the second polynucleotide are from the same AAV serotype; the 5-ITR and 3-ITR of the first polynucleotide and the 5-ITR and 3-ITR of the second polynucleotide are respectively from different AAV serotypes; the 5-ITR of the first polynucleotide and the 3-ITR of the second polynucleotide are from different AAV serotypes; the nucleotide sequence of the 5-ITR and 3-ITR of the first polynucleotide and the second polynucleotide are obtained from AAV serotype 2; the coding sequence is split into the 5 end portion and the 3 end portion at a natural exon-exon junction; the MYO7A coding sequence is split between exons 24-25; the nucleic acid sequence of the splicing donor signal comprises the sequence: TABLE-US-00020 (SEQIDNO:1) GTAAGTATCAAGGTTACAAGACAGGTTTAAGGAGACCAATAGAAACTGGG CTTGTCGAGACAGAGAAGACTCTTGCGTTTCT; the nucleic acid sequence of the splicing acceptor signal comprises the sequence TABLE-US-00021 (SEQIDNO:2) GATAGGCACCTATTGGTCTTACTGACATCCACTTTGCCTTTCTCTCCACA G; and/or the first polynucleotide further comprises at least one enhancer and/or intron sequence, operably linked to the coding sequence. .Iaddend.

.Iadd.45. A dual viral vector system comprising: a) a first viral vector containing the first polynucleotide, and b) a second viral vector containing the second polynucleotide, wherein said first and said second polynucleotides are as defined in claim 43, and wherein the vectors are AAV vectors, optionally wherein: the AAV vectors are the same or different AAV serotypes; the AAV vectors have a serotype selected from the group consisting of serotype 2, serotype 8, serotype 5, serotype 7 and serotype 9; and/or the AAV vector is AAV2/8. .Iaddend.

.Iadd.46. The dual viral vector system of claim 45, capable of transducing one or both of retinal pigment epithelium and photoreceptors; capable of inducing stronger expression of MYO7A compared to a dual AAV trans-splicing vector system; and/or capable of transducing retina to a higher level compared to an oversized AAV vector system or a dual viral vector system comprising an alkaline phosphatase (AP) recombinogenic region. .Iaddend.

.Iadd.47. An isolated host cell transformed with the dual viral vector system according to claim 45, optionally wherein the host cell is a human cell. .Iaddend.

.Iadd.48. A pharmaceutical composition comprising the dual viral vector system of claim 45, and a pharmaceutically acceptable vehicle, optionally wherein the pharmaceutical composition is administered to a subject via subretinal administration. .Iaddend.

.Iadd.49. A method for treating a subject having Usher 1B, comprising subretinally administering to the subject an effective amount of the dual viral vector system of claim 45. .Iaddend.

.Iadd.50. A method for correctly localizing retinal pigment epithelium (RPE) melanosomes apically, comprising subretinally administering to the subject an effective amount of the dual viral vector system of claim 45. .Iaddend.

.Iadd.51. A method for reducing the accumulation of rhodopsin at the connecting cilium of photoreceptors, comprising subretinally administering to the subject an effective amount of the dual viral vector system of claim 45. .Iaddend.

.Iadd.52. A method to induce genetic recombination in a host cell, the method comprising introducing into the host cell a dual construct system, wherein the dual construct system comprises: a) a first polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; a 5 end portion of a coding sequence of a gene of interest, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00022 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto; and b) a second polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of a coding sequence of the gene of interest; a poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are AAV ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00023 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.53. The method of claim 52, wherein: upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of the splicing donor and the splicing acceptor signals; and/or upon introduction of the first polynucleotide and the second polynucleotide into the host cell, the coding sequence reconstitutes by means of homologous recombination between the Fl phage recombinogenic region of the first polynucleotide and the Fl phage recombinogenic region of the second polynucleotide. .Iaddend.

.Iadd.54. A method for reconstituting a coding sequence in a host cell, the method comprising introducing into the host cell a dual construct system, wherein the dual construct system comprises: a) a first polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; a 5 end portion of a coding sequence of a gene of interest, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00024 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto; and b) a second polynucleotide comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of a coding sequence of the gene of interest; a poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are AAV ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00025 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

.Iadd.55. A dual viral vector system comprising: a) a first viral vector comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a promoter sequence; a 5 end portion of a coding sequence of a gene of interest, the 5 end portion being operably linked to and under control of the promoter sequence; a nucleic acid sequence of a splicing donor signal; a nucleic acid sequence of a recombinogenic region; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are adeno-associated virus (AAV) ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00026 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto; and b) a second viral vector comprising in a 5-3 direction: a 5-inverted terminal repeat (5-ITR) sequence; a nucleic acid sequence of a recombinogenic region; a nucleic acid sequence of a splicing acceptor signal; a 3 end portion of a coding sequence of the gene of interest; a poly-adenylation signal nucleic acid sequence; and a 3-inverted terminal repeat (3-ITR) sequence, wherein the ITRs are AAV ITRs, and wherein the recombinogenic region is: (a) a Fl phage recombinogenic region that consists of the sequence: TABLE-US-00027 (SEQIDNO.3) GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACA AAAATTTAACGCGAATTTTAACAAAAT; or (b) a fragment of SEQ ID NO: 3 having at least 85% sequence identity thereto. .Iaddend.

Description

(1) The present invention will now be illustrated by means of non-limiting examples in reference to the following drawings.

(2) FIG. 1. Schematic representation of AAV-based strategies for large gene transduction.

(3) CDS: coding sequence; pA: poly-adenilation signal; SD: splicing donor signal; SA: splicing acceptor signal; AP: alkaline phosphatase recombinogenic region (39); AK: F1 phage recombinogenic region. Dotted lines show the splicing occurring between SD and SA, pointed lines show overlapping regions available for homologous recombination. The inventors found that dual trans-splicing and hybrid AK may be used to successfully reconstitute large gene expression. In particular dual trans-splicing and hybrid AK vectors, but not overlapping and hybrid AP vectors, transduce efficiently mouse and pig photoreceptors. Normal size and oversize AAV vector plasmids contained full length expression cassettes including the promoter, the full-length transgene CDS and the poly-adenilation signal (pA) (Table 1). The two separate AAV vector plasmids (5 and 3) required to generate dual AAV vectors contained either the promoter followed by the N-terminal portion of the transgene CDS (5 plasmid) or the C-terminal portion of the transgene CDS followed by the pA signal (3 plasmid, Table 1). The structure of all plasmids is indicated in the material and method section.

(4) FIG. 2. Dual AAV overlapping, trans-splicing and hybrid AK vectors efficiently transduce large genes in vitro.

(5) Western blot of HEK293 cells infected with AAV2/2 vectors encoding for EGFP (A and D), ABCA4 (B and E) and MYO7A (C and F). (A to C) The arrows indicate full-length proteins, the micrograms of proteins loaded are depicted under each lane, the molecular weight ladder is depicted on the left. (D to F) Quantification of EGFP (D), ABCA4 (E) and MYO7A (F) protein bands. The intensity of the EGFP, ABCA4 and MYO7A bands was divided by the intensity of the Tubulin (D) or Filamin A (E-F) bands. The histograms show the expression of proteins as a percentage relative to dual AAV trans-splicing (TS) vectors, the mean value is depicted above the corresponding bar. Error bars: means.e.m. (standard error of the mean). (A-C) The Western blot images are representative of and the quantifications are from n=4 (A-B) or n=3 (C) independent experiments. OZ: AAV oversize; OV: dual AAV overlapping; TS: dual AAV trans-splicing; AP: dual AAV hybrid AP; AK: dual AAV hybrid AK; 5+3: cells co-infected with 5- and 3-half vectors; 5: control cells infected with the 5-half vector only; 3: control cells infected with the 3-half only; -EGFP: anti-EGFP antibody; -3flag: anti-3flag antibody; -MYO7A: anti-MYO7A antibody; --Tubulin: anti--tubulin antibody; -Filamin A: anti-filamin A antibody. * ANOVA p value<0.05; ** ANOVA p value<0.001. (F) The asterisks depicted in the lower panel represent significant differences with both OZ and AP. In

(6) FIG. 3. Dual AAV overlapping vectors transduce RPE but not photoreceptors in the mouse and pig retina.

(7) Western blot analysis of C57BL/6 (A) and Large White pig (B) retinal lysates one month following injection of AAV2/8 dual AAV overlapping vectors encoding for ABCA4-3flag (OV) or AAV2/8 vectors encoding for normal size EGFP (EGFP), under the control of the ubiquitous cytomegalovirus (CMV) promoter, the PR-specific Rhodopsin (RHO) and Rhodopsin kinase (RHOK) promoters, or the RPE-specific vitelliform macular dystrophy 2 (VMD2) promoter. (A-B) The arrows indicate full-length proteins, the molecular weight ladder is depicted on the left, 150 micrograms of proteins were loaded in each lane. The number (n) and percentage of ABCA4-positive retinas out of total retinas analyzed is depicted; -3flag: anti-3flag antibody; -Dysferlin: anti-Dysferlin antibody (C) Western blot analysis on C57/BL6 eyecups (left panel) and retinas (right panel) at 3 months following the injection of AAV2/8 overlapping vectors encoding for MYO7A-HA (OV) under the control of the ubiquitous chicken-beta-actin (CBA) promoter or the photoreceptor-specific rhodopsin (RHO) promoter. The arrow points at full-length proteins, the molecular weight ladder is depicted on the left, 100 micrograms of protein were loaded in each lane. The number (n) and percentage of MYO7A positive retinas out of total retinas analyzed is depicted. -HA: anti-hemagglutinin (HA) antibody.

(8) FIG. 4. Dual AAV trans-splicing and hybrid AK vectors efficiently transduce both RPE and photoreceptors.

(9) Fluorescence analysis of retinal cryosections from C57BL/6 mice one month following subretinal injection of AAV2/8 vectors encoding for EGFP under the control of the ubiquitous cytomegalovirus (CMV) promoter. The scale bar (20 m) is depicted in the figure. NS: AAV normal size; OZ: AAV oversize; TS: dual AAV trans-splicing; AP: dual AAV hybrid AP; AK: dual AAV hybrid AK; RPE: retinal pigmented epithelium; ONL: outer nuclear layer.

(10) FIG. 5. Dual AAV trans-splicing and hybrid AK efficiently transduce mouse and pig photoreceptors.

(11) (A) Fluorescence analysis of retinal cryosections from C57BL/6 mice one month following subretinal injection of AAV2/8 vectors encoding for EGFP under the control of the PR-specific Rhodopsin promoter (RHO). The scale bar (20 m) is depicted in the figure. (B) Fluorescence analysis of retinal cryosections from Large White pigs one month following subretinal injection of AAV2/8 vectors encoding for EGFP under the control of the PR-specific RHO promoter. The scale bar (50 m) is depicted in the figure. NS: AAV normal size; TS: dual AAV trans-splicing; AK: dual AAV hybrid AK; RPE: retinal pigmented epithelium; ONL: outer nuclear layer.

(12) FIG. 6. Subretinal administration of dual AAV trans-splicing and hybrid AK vectors results in robust yet variable levels of ABCA4 expression in mouse photoreceptors.

(13) (A) Western blot analysis of C57BL/6 retinal lysates one month following the injection of dual AAV trans-splicing (TS) and dual AAV hybrid AK (AK) vectors encoding for ABCA4 under the control of the PR-specific Rhodopsin promoter (RHO). The arrow points at full-length proteins, the molecular weight ladder is depicted on the left, 150 micrograms of protein were loaded in each lane. The number (n) and percentage of ABCA4-positive retinas out of total retinas analysed is depicted. 5+3: retinas co-injected with 5- and 3-half vectors; -3flag: anti-3flag antibody; -Dysferlin: anti-Dysferlin antibody. (B) Immuno-electron microscopy analysis with anti-HA antibody of retinal sections from wild-type Balb/C (WT; n=3 eyes) and Abca4/ mice injected with dual AAV hybrid AK vectors (AK-ABCA4; n=5 eyes) or with AAV normal size EGFP (EGFP, n=3 eyes) as control. The black dots represent the immuno-gold labelling of the ABCA4-HA protein. The scale bar (200 nm) is depicted in the figure.

(14) FIG. 7. Subretinal injection of dual AAV hybrid AK vectors reduces accumulation of lipofuscin granules in Abca4/ mice.

(15) (A) Transmission electron microscopy analysis of retinal sections from wild-type Balb/c (WT) and Abca4/ mice injected with either dual AAV hybrid AK vectors (Abca4/ AK-ABCA4) or with AAV normal size EGFP (Abca4/ EGFP) as control. The black arrows indicate lipofuscin granules. The scale bar (1.6 m) is depicted in the figure. (B) Quantification of the mean number of lipofuscin granules counted in at least 30 fields (25 m.sup.2) for each sample. WT: Balb/c mice; Abca4/ EGFP/5/3: Abca4/ mice injected with either AAV normal size EGFP or the 5 or 3 half vector of the dual AAV hybrid AK, as control; Abca4/ AK-ABCA4: mice injected with dual AAV hybrid AK vectors; Abca4/ TS-ABCA4: mice injected with dual AAV trans-splicing vectors. The number (n) of eyes analysed is depicted. The mean value is depicted above the corresponding bar. Error bars: means.e.m. (standard error of the mean). * p ANOVA<0.05

(16) FIG. 8. Subretinal injections of dual AAV hybrid AK vectors reduces the thickness of Abca4/ RPE.

(17) (A) Representative pictures of transmission electron microscopy analysis of retinal sections from wild-type Balb/c (WT) and Abca4/ mice injected with either dual AAV trans-splicing (TS-ABCA4) and hybrid AK vectors (AK-ABCA4) or with AAV normal size EGFP (EGFP) and 5 or 3 half of the dual hybrid AK vectors (5/3) as control. The dotted lines indicate the edges of RPE cells. The scale bar (3.8 m) is depicted in the figure. (B) Quantification of the mean RPE thickness counted in at least 30 fields for each sample. The number (n) of eyes analysed is depicted. The mean value is depicted above the corresponding bar. Error bars: means.e.m (standard error of the mean). s.d.m: WT: 716; TS-ABCA4: 698.

(18) FIG. 9. Subretinal administration of dual AAV trans-splicing and hybrid AK vectors results in robust MYO7A expression in mice.

(19) Western blot analysis of C57BL/6 eyecups one month following the injection of dual AAV trans-splicing (TS) and hybrid AK (AK) vectors encoding for MYO7A-HA under the control of the ubiquitous chicken beta-actin (CBA) promoter. The arrow indicates full-length proteins, the molecular weight ladder is depicted on the left, 100 micrograms of proteins were loaded in each lane. The number (n) and percentage of MYO7A-positive eyecups out of total retinas analyzed is depicted. 5+3: eyes co-injected with 5- and 3-half vectors; 5: eyes injected with 5-half vectors; 3: eyes injected with 3-half vectors; -HA: anti-hemagglutinin (HA) antibody; -Dysferlin: anti-Dysferlin antibody.

(20) FIG. 10. Subretinal administration of dual AAV trans-splicing and hybrid AK vectors rescues melanosome localization in sh1/ RPE.

(21) (A) Representative semi-thin retinal sections stained with Epoxy tissue stain of sh1+/+ and sh1+/ eyes injected with AAV normal size EGFP (EGFP, n=4 eyes), and of sh1/ eyes injected with dual AAV trans-splicing (TS-MYO7A, n=3 eyes), hybrid AK (AK-MYO7A; n=3 eyes) or 5-half vectors (5TS/5AK, n=4 eyes), as control. The scale bar (10 m) is depicted in the figure. (B) Quantification of melanosome localization in the RPE villi of sh1 mice two months following subretinal delivery of dual AAV vectors. The quantification is depicted as the mean number of apical melanosomes/field, the mean value is depicted above the corresponding bar. Error bars: means.e.m. (standard error of the mean). * p ANOVA<0.05, ** p ANOVA<0.001.

(22) FIG. 11. Subretinal administration of dual AAV trans-splicing and hybrid AK vectors reduces rhodopsin accumulation at sh1/ PR connecting cilia.

(23) Quantification of the number of rhodopsin gold particles at the PR connecting cilium of sh1 mice two months following subretinal delivery of dual AAV vectors. The quantification is depicted as the mean number of gold particles per length of connecting cilia (nm), the mean value is depicted above the corresponding bar. Error bars: means.e.m. (standard error of the mean).

(24) FIG. 12. Dual AAV trans-splicing and hybrid AK vectors efficiently transduce the large gene CEP290 in vitro.

(25) Western blot of HEK293 cells infected with AAV2/2 vectors encoding for CEP290 tagged at its C-terminus with the hemagglutinin (HA) tag (A-B). (A) The arrow indicate the full-length protein, 60 micrograms of proteins were loaded for each lane, the molecular weight ladder is depicted on the left. (B) Quantification of CEP290 protein bands. The intensity of the CEP290 bands was divided by the intensity of the Filamin A bands. The histogram shows the expression of proteins as a percentage relative to dual AAV trans-splicing (TS) vectors, the mean value is depicted above the corresponding bar. Error bars: means.e.m. (standard error of the mean). The Western blot image is representative of and the quantification is from n=5 independent experiments. OV: dual AAV overlapping; TS: dual AAV trans-splicing; AK: dual AAV hybrid AK; 5+3: cells co-infected with 5- and 3-half vectors; 3: control cells infected with the 3-half only; -HA: anti-HA antibody; -Filamin A: anti-filamin A antibody.

(26) FIG. 13. Improved recovery from light desensitization in 3 months old Abca4/ mice treated with dual AAV trans-splicing and hybrid AK vectors

(27) Recovery from light desensitization in Abca4/ and Balb/c mice at 6 weeks post-injection. The relative b-wave is the ratio between the post- and the pre-desensitization b-wave amplitudes (V) both evoked by 1 cd s/m.sup.2. The time (minutes) refers to the time post-desensitization. The mean recovery (%) at 60 minutes is depicted. p ANOVA Abca4/ AK-ABCA4 vs Abca4/ uninjected/5: 0.05; p ANOVA Abca4/ TS-ABCA4 vs Abca4/ uninjected/5: 0.009; p ANOVA Abca4/ AK-ABCA4 vs WT: 0.002; p ANOVA Abca4/ TS-ABCA4 vs WT: 0.02; p ANOVA WT vs Abca4/ uninjected/5: 0.00001. WT: Balb/c mice (n=4); Abca4/ TS-ABCA4: mice injected with dual AAV trans-splicing vectors (n=5); Abca4/ AK-ABCA4: mice injected with dual AAV hybrid AK vectors (n=5); Abca4/ uninjected/5: Abca4/ mice either not injected (n=2) or injected with the 5 half of the dual AAV TS or hybrid AK vectors (n=5). Data are depicted as means.e.m (standard error of the mean). * p ANOVA<0.05.

(28) FIG. 14. Dual AAV hybrid AK vectors induce stronger MYO7A expression than dual AAV trans-splicing vectors in sh1/ photoreceptors.

(29) Quantification of MYO7A levels from dual AAV vectors in sh1/ eyes relative to endogenous Myo7a expressed in sh1+/+ eyes. Sh1/ eyes were injected with dual AAV TS and hybrid AK vectors encoding MYO7A under the control of either the CBA (left panel) or RHO (right panel) promoters. The histograms show the expression of MYO7A protein as percentage relative to sh1+/+Myo7a; the mean value is depicted above the corresponding bar. The quantification was performed by Western blot analysis using the anti-MYO7A antibody and measurements of MYO7A and Myo7a band intensities normalized to Dysferlin (data not shown). Error bars: means.d.m. (standard deviation of the mean). The quantification is representative of: i. left panel: n=2 sh1+/+ eyecups, and n=5 or n=1 sh1/ eyecups treated with either TS-MYO7A or AK-MYO7A, respectively; ii. right panel: n=2 sh1+/+ retinas, and n=1 or n=3 sh1/ retinas treated with either TS-MYO7A or AK-MYO7A, respectively. ** p Student's t-test<0.001.

(30) FIG. 15. AAV normal size, dual AAV trans-splicing and hybrid AK vectors provide the most robust transduction following subretinal delivery in mice.

(31) Live-imaging fundus fluorescence of C57BL/6 eyes one month following subretinal injection of AAV2/8 vectors encoding for EGFP. NZ: Normal Size; OZ: AAV oversize; TS: dual AAV trans-splicing; AP: dual AAV hybrid AP; AK: dual AAV hybrid AK. Each panel shows a different eye.

(32) FIG. 16. Robust ABCA4 and MYO7A expression following delivery of dual AAV trans-splicing and hybrid AK vectors to the pig retina. (a) Western blot analysis of large white pig retinal lysates 1 month following injection of dual AAV2/8 trans-splicing (TS; n=2) and hybrid AK (AK; n=3) vectors encoding for ABCA4-3flag or AAV2/8 vectors encoding for NS EGFP (neg), as negative control, under the control of the photoreceptor-specific rhodopsin (RHO) promoter. (b) Western blot analysis of large white pig retinal lysates one month following injection of dual AAV2/8 trans-splicing (TS: n=5 RPE; n=3 retina) and hybrid AK (AK: n=5 RPE, n=5 retina) vectors encoding for MYO7A-HA under the control of the ubiquitous chicken beta actin (CBA) promoter or single 3-half of dual AAV-MYO7A-HA (neg), as negative control. (a-b) The arrows indicate full-length proteins, the molecular weight ladder is depicted on the left, 150-180 g of proteins were loaded in each lane. -3flag, anti-3flag antibody; -HA, anti-hemagglutinin antibody; -dysferlin, anti-dysferlin antibody.

(33) FIG. 17. Dual AAV hybrid AK vectors with heterologous ITRs transduce large genes in vitro. (a) Design of dual AAV hybrid AK vectors with heterologous ITR2 and ITR5. (b) Western blot analysis of HEK293 cells infected with dual AAV hybrid AK vectors with heterologous ITRs encoding for ABCA4 (left panel) and MYO7A (right panel). The arrows indicate full-length proteins, 50 micrograms of proteins were loaded, the molecular weight ladder is depicted on the left. 5+3: cells co-infected with 5- and 3-half vectors; 5: control cells infected with the 5-half vector only; 3: control cells infected with the 3-half vector only; neg: cells infected with AAV2/8 vectors encoding for EGFP. -3flag: anti-3flag antibody; -MYO7A: anti-MYO7A antibody; -Filamin: anti-filamin A antibody. (a) Prom: promoter; CDS: coding sequence; pA: poly-adenylation signal; SD: splicing donor signal; SA: splicing acceptor signal; Pointed lines show overlapping regions available for homologous recombination, dotted lines show the splicing occurring between SD and SA. The position of the heterologous ITR2 and ITR5 is depicted.

DETAILED DESCRIPTION OF THE INVENTION

(34) Materials and Methods

(35) Generation of AAV Vector Plasmids

(36) The plasmids used for AAV vector production were derived from either the pZac2.1 (52) or pAAV2.1 (53) plasmids that contain the inverted terminal repeats (ITRs) of AAV serotype 2 (Table 1).

(37) TABLE-US-00005 TABLE 1 Plasmids for AAV vector production. Size AAV serotype Plasmid ITR- ITR (bp) 2/2 2/8 Normal Size pZac2.1-CMV-EGFP-SV40 3006 X X (NS) pZac2.1-RHO-EGFP-SV40 2900 X Oversize pAAV2.1-CMV-EGFP-9.9-BGH 9951 X X (OZ) pZac2.1-CMV-ABCA4_3xflag-SV40 8619 X pAAV2.1-CBA-MYO7A_HA-BGH 8220 X Overlapping pZac2.1-CMV-ABCA4_5 4900 X X (OV) pZac2.1-RHO-ABCA4_5 4805 X pZac2.1-RHOK-ABCA4_5 4169 X pZac2.1-VMD2-ABCA4_5 4658 X pAAV2.1-CBA-MYO7A_5 4708 X X pAAV2.1-RHO-MYO7A_5 4699 X pZac2.1-ABCA4_3_3xflag_SV40 4740 X X pAAV2.1-MYO7A_3_HA_BGH 4655 X X Trans-splicing pZac2.1-CMV-ABCA4 _5TS 4431 X (TS) pZac2.1-RHO-ABCA4_ 5TS 4321 X pZac2.1-ABCA4 _3TS_3xflag_SV40 4587 X X pAAV2.1-CBA-MYO7A _5TS 4468 X X pAAV2.1-RHO-MYO7A _5TS 4459 X pAAV2.1-MYO7A_3TS_HA_BGH 4298 X X pZac2.1-CMV-EGFP_5TS 1906 X X pZac2.1-RHO-EGFP_5TS 1802 X pZac2.1-EGFP_3TS_SV40 1510 X X Hybrid AP pZac2.1-CMV-ABCA4_5AP 4708 X (AP) pZac2.1-ABCA4 _3AP_3xflag_SV40 4871 X pAAV2.1-CBA-MYO7A_5AP 4746 X pAAV2.1-MYO7A_3AP_HA_BGH 4576 X pZac2.1-CMV-EGFP_5AP 2183 X X pZac2.1-EGFP_3AP_SV40 1783 X X Hybrid AK pZac2.1-CMV-ABCA4 _5AK 4540 X (AK) pZac2.1 (ITR5:2)-CMV-ABCA4_5AK 4604 X pZac2.1-RHO-ABCA4_ 5AK 4436 X pZac2.1-ABCA4 _3AK_3xflag_SV40 4702 X X pZac2.1 (ITR2:5)- ABCA4 _3AK_3xflag_SV40 5192 X pZac2.1-ABCA4_ 3AK_HA_SV40 4663 X pAAV2.1-CBA-MYO7A_5AK 4577 X X pAAV2.1 (ITR5:2)-CBA-MYO7A_5AK 4503 X pAAV2.1-RHO-MYO7A _5AK 4568 X pAAV2.1-MYO7A_3AK_HA_BGH 4421 X X pAAV2.1 (ITR2:5)-MYO7A_3AK_HA_BGH 4386 X pZac2.1-CMV-EGFP_5AK 2015 X X pZac2.1-RHO-EGFP_5AK 1911 X pZac2.1-EGFP_3AK_SV40 1614 X X N.B. CMV: cytomegalovirus promoter; CBA: chicken beta-actin; RHO: human Rhodopsin promoter; RHOK: human Rhodopsin kinase promoter; Vmd2: vitelliform macular dystrophy 2 promoter; EGFP: enhanced green fluorescent protein; ABCA4: human ATP-binding cassette, sub-family A, member 4; MYO7A: human MYOSIN VIIA; SV40: simian virus 40 poly-adenilation signal; BGH: bovine growth hormone poly-adenilation signal; 3xflag: 3xflag tag; HA: hemagglutinin tag; AP: alkaline phosphatase recombinogenic region; AK: F1 phage recombinogenic region; TS: trans-splicing; ITR5:2: plasmid with the left ITR from AAV serotype 5 and the right ITR from AAV serotype 2; ITR2:5: plasmid with the left ITR from AAV serotype 2 and the right ITR from AAV serotype 5. When not specified the left and right ITR are from AAV serotype 2.

(38) Normal size and oversize AAV vector plasmids contained full length expression cassettes including the promoter, the full-length transgene CDS and the poly-adenilation signal (pA) (Table 1). The two separate AAV vector plasmids (5 and 3) required to generate dual AAV vectors contained either the promoter followed by the N-terminal portion of the transgene CDS (5 plasmid) or the C-terminal portion of the transgene CDS followed by the pA signal (3 plasmid, Table 1). Normal size EGFP plasmids were generated by cloning the EGFP CDS of pAAV2.1-CMV-EGFP plasmid (720 bp) (53) in pZac2.1 (52); oversize EGFP was generated from pAAV2.1-CMV-EGFP (53) by inserting a DNA stuffer sequence of 3632 bp from human ABCA4 (NM_000350.2, bp 1960-5591) upstream of the CMV promoter and a second DNA stuffer sequence of 3621 bp, composed of: murine ABCA4 (NM 007378.1, 1066-1 and 7124-6046 bp; 2145 total bp) and human Harmonin (NM153676.3 131-1606 bp; 1476 total bp), downstream of the pA signal (This construct was used in the experiments of FIG. 1a, d, FIG. 4 and FIG. 15). To generate dual AAV vector plasmids, the EGFP CDS (720 bp) was split into two constructs: one containing the N-terminal CDS (PMID: 9759496, bp 1-393) and the other containing the C-terminal CDS (PMID: 9759496, bp 394-720).

(39) The oversize ABCA4 plasmids contained the full-length human ABCA4 CDS (GeneNM_000350.2, bp 105-6926), while the oversize MYO7A plasmids contained the full-length human MYO7A CDS from isoform 1 (NM_000260.3, bp 273-6920). To generate plasmids for dual AAV OV vectors the ABCA4 and MYO7A CDS were split into two constructs, one containing N-terminal CDS (ABCA4: NM_000350.2, bp 105-3588; MYO7A: NM_000350.2, bp 273-3782) and the other containing C-terminal CDS (ABCA4: NM_000350.2, bp 2819-6926; MYO7A: NM_000350.2, bp 2913-6920). Therefore, the region of homology shared by overlapping vector plasmids was 770 bp for ABCA4 and 870 bp for MYO7A. To generate plasmids for dual AAV OV vectors the human CEP290 CDS was split into two constructs, one containing N-terminal CDS (CEP290: NM_025114, bp 345-4076) and the other containing C-terminal CDS (CEP290: NM_025114, bp 3575-7784). Therefore, the region of homology shared by overlapping vector plasmids was 502 bp.

(40) To generate trans-splicing and hybrid vector plasmids the ABCA4 and MYO7A CDS were split at a natural exon-exon junction. ABCA4 was split between exons 19-20 (5 half: NM_000350.2, 105-3022 bp; 3 half: NM_000350.2, bp 3023-6926) and MYO7A was split between exons 24-25 (5 half: NM_000350.2, bp 273-3380; 3 half: NM_000350.2, bp 3381-6926). The ABCA4 and MYO7A proteins were both tagged at their C-terminus: ABCA4 with either the 3flag (gactacaaagaccatgacggtgattataaagatcatgacatcgactacaaggatgacgatgacaag) .Iadd.(SEQ ID NO: 30) .Iaddend.or hemagglutinin (HA) tag (tatccgtatgatgtgccggattatgcg) .Iadd.(SEQ ID NO:32) .Iaddend.; MYO7A with the HA tag only. To generate trans-splicing and hybrid vector plasmids the CEP290 CDS was split at a natural exon-exon junction: between exons 29-30 (5 half: NM_025114, 345-3805; 3 half: NM_025114, 3806-7784). The CEP290 protein was tagged at its C-terminus with the hemagglutinin (HA) tag. The splice donor (SD) and splice acceptor (SA) signals contained in trans-splicing and hybrid dual AAV vector plasmids are as follows:

(41) TABLE-US-00006 5GTAAGTATCAAGGTTACAAGACAGGTTTAAGGAGACCAATAGAAACTG GGCTTGTCGAGACAGAGAAGACTCTTGCGTTTCT-3(SD)SEQID No.1; 5GATAGGCACCTATTGGTCTTACTGACATCCACTTTGCCTTTCTCTCCA CAG-3(SA),SEQIDNo.2.
The recombinogenic sequence contained in hybrid AP vector plasmids (present in both first and second plasmids) were derived from alkaline phosphate (AP) genes (NM_001632, bp 823-1100), as previously described (39). The recombinogenic sequence contained in hybrid AK vector plasmids (present in both first and second plasmids) were derived from the phage F1 genome (Gene Bank accession number: J02448.1; bp 5850-5926). The AK sequence is:
5GGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTAACGCGAATTTTAACAAAAT-3, SEQ ID No. 3.

(42) The ubiquitous CMV promoter is that contained in pZac2.1 (52) or pAAV2.1-CMV-EGFP (53); the ubiquitous CBA promoter was derived from pAAV2.1-CBA-EGFP (11), the PR-specific human RHO and RHOK promoters were derived from pAAV2.1-RHO-EGFP and pAAV2.1RHOK-EGFP, respectively (10); the RPE-specific Vmd2 promoter (NG_009033.1, 4870-5470 bp) corresponds to the previously described EcoRI-XcmI promoter fragment (41) and was amplified by human genomic DNA.

(43) To generate dual AAV hybrid AK vectors with heterologous ITRs from AAV serotype 2 and 5 we exchanged the left ITR2 of the 5-half plasmid and the right ITR2 of the 3-half plasmid with the ITR5 (as depicted in FIG. 17a). The plasmids for the production of AAV2 vectors with heterologous ITRs are the following: pZac5:2-CMV-5ABCA4-SD-AK, pZac2:5-AK-SD-3ABCA4-3flag, pAAV5:2-CBA-5MYO7A-SD-AK and pAAV2:5-AK-SD-3MYO7A-HA (Table 1).

(44) For the purposes of this invention, a coding sequence of ABCA4, MYO7A and CEP290 which are preferably respectively selected from the sequences herein enclosed, or sequences encoding the same amino acid sequence due to the degeneracy of the genetic code, is functionally linked to a promoter sequence able to regulate the expression thereof in a mammalian retinal cell, particularly in photoreceptor cells. Suitable promoters that can be used according to the invention include the cytomegalovirus promoter, Rhodopsin promoter, Rhodopsin kinase promoter, Interphotoreceptor retinoid binding protein promoter, vitelliform macular dystrophy 2 promoter, fragments and variants thereof retaining a transcription promoter activity.

(45) Viral delivery systems include but are not limited to adenoviral vectors, adeno-associated viral (AAV) vectors, pseudotyped AAV vectors, herpes viral vectors, retroviral vectors, lentiviral vectors, baculoviral vectors. Pseudotyped AAV vectors are those which contain the genome of one AAV serotype in the capsid of a second AAV serotype; for example an AAV2/8 vector contains the AAV8 capsid and the AAV 2 genome (Auricchio et al. (2001) Hum. Mol. Genet. 10(26):3075-81). Such vectors are also known as chimeric vectors. Other examples of delivery systems include ex vivo delivery systems, which include but are not limited to DNA transfection methods such as electroporation, DNA biolistics, lipid-mediated transfection, compacted DNA-mediated transfection.

(46) The construction of an AAV vector can be carried out following procedures and using techniques which are known to a person skilled in the art. The theory and practice for adeno-associated viral vector construction and use in therapy are illustrated in several scientific and patent publications (the following bibliography is herein incorporated by reference: Flotte T R. Adeno-associated virus-based gene therapy for inherited disorders. Pediatr Res. 2005 December; 58(6):1143-7; Goncalves M A. Adeno-associated virus: from defective virus to effective vector, Virol J. 2005 May 6; 2:43; Surace E M, Auricchio A. Adeno-associated viral vectors for retinal gene transfer. Prog Retin Eye Res. 2003 November; 22(6):705-19; Mandel R J, Manfredsson F P, Foust K D, Rising A, Reimsnider S, Nash K, Burger C. Recombinant adeno-associated viral vectors as therapeutic agents to treat neurological disorders. Mol Ther. 2006 March; 13(3):463-83).

(47) Suitable administration forms of a pharmaceutical composition containing AAV vectors include, but are not limited to, injectable solutions or suspensions, eye lotions and ophthalmic ointment. In a preferred embodiment, the AAV vector is administered by subretinal injection, e.g. by injection in the subretinal space, in the anterior chamber or in the retrobulbar space. Preferably the viral vectors are delivered via subretinal approach (as described in Bennicelli J, et al Mol Ther. 2008 Jan. 22; Reversal of Blindness in Animal Models of Leber Congenital Amaurosis Using Optimized AAV2-mediated Gene Transfer).

(48) The doses of virus for use in therapy shall be determined on a case by case basis, depending on the administration route, the severity of the disease, the general conditions of the patients, and other clinical parameters. In general, suitable dosages will vary from 10.sup.8 to 10.sup.13 vg (vector genomes)/eye.

(49) AAV Vector Production

(50) AAV vectors were produced by the TIGEM AAV Vector Core by triple transfection of HEK293 cells followed by two rounds of CsCl2 purification (54). For each viral preparation, physical titers [genome copies (GC)/ml] were determined by averaging the titer achieved by dot-blot analysis (55) and by PCR quantification using TaqMan (54) (Applied Biosystems, Carlsbad, Calif.).

(51) AAV Infection of HEK293 Cells

(52) HEK293 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum and 2 mM L-glutamine (GIBCO, Invitrogen S.R.L., Milan, Italy). Cells were plated in six-well plates at a density of 210.sup.6 cells/well and transfected 16 hours later with 1.3 g of pDeltaF6 helper plasmid which contains the Ad helper genes (56) using the calcium phosphate method. After 5 hours, cells were washed once with DMEM and incubated with AAV2/2 vectors (m.o.i: 10.sup.5 GC/cell of each vector; 1:1 co-infection with dual AAV vectors resulted in of 210.sup.5 total GC/cell) in a final volume of 700 L serum-free DMEM. Two hours later 2 ml of complete DMEM was added to the cells. Cells were harvested 72 hours following infection for Western blot analysis.

(53) Animal Models

(54) This study was carried out in accordance with the NIH Guide for the Care and Use of Laboratory Animals, the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research, and the Italian Ministry of Health regulation for animal procedures. Mice were housed at the Institute of Genetics and Biophysics animal house (Naples, Italy) and maintained under a 12-hour light/dark cycle (10-50 lux exposure during the light phase). C57BL/6 and BALB/c mice were purchased from Harlan Italy SRL (Udine, Italy). Albino Abca4/ mice were generated through successive crosses and backcrosses with BALB/c mice (homozygous for Rpe65 Leu450) (57) and maintained inbred. Breeding was performed crossing homozygous mice. Pigmented sh14626SB/4626SB (referred to as sh1/) mice were imported from the Wellcome Trust Sanger Institute (Cambridge, UK, a kind gift of Dr. Karen Steel) and back-crossed twice with CBA/Ca mice purchased from Harlan Italy SRL (Udine, Italy) to obtain heterozygous sh1+/4626SB (referred to as sh1+/) mice to expand the colony. The mice were maintained intercrossed; breeding was performed crossing heterozygous females with heterozygous males. The pigmented sh1 mice used in this study were either Usher 1B affected (sh1/) or unaffected (sh1+/ and sh1+/+). The genotype for the MYO7A.sup.4626SB allele was performed by PCR analysis of genomic DNA (extracted from the mouse tail tip) followed by DNA sequencing. The primers used for the PCR amplification are as follows: Fw1 (GTGGAGCTTGACATCTACTTGACC) .Iadd.(SEQ ID NO: 33) .Iaddend.and Rev3 (AGCTGACCCTCATGACTCTGC) .Iadd.(SEQ ID NO: 34).Iaddend., which generate a product of 712 bp that was sequenced with the Fw1 primer. The Large White Female pigs used in this study were registered as purebred in the LWHerd Book of the Italian National Pig Breeders' Association (Azienda Agricola Pasotti, Imola, Italy).

(55) Subretinal Injection of AAV Vectors in Mice and Pigs

(56) Mice (4-5 weeks-old) were anesthetized with an intraperitoneal injection of 2 ml/100 g body weight of avertin [1.25% w/v of 2,2,2-tribromoethanol and 2.5% v/v of 2-methyl-2-butanol (Sigma-Aldrich, Milan, Italy)] (58), then AAV2/8 vectors were delivered subretinally via a trans-scleral transchoroidal approach as described by Liang et al (59). All eyes were treated with 1 L of vector solution. The AAV2/8 doses (GC/eye) delivered vary across the different mouse experiments as it is described in the RESULTS section. AAV2/1-CMV-human Tyrosinase (60) (dose: 210.sup.8 GC/eye) or AAV2/5-CMV-EGFP (encoding normal size EGFP, dose: 410.sup.8 GC/eye) was added to the AAV2/8 vector solution that was subretinally delivered to albino (Abca4/ and BALB/c) (FIG. 6B, 7-8) or pigmented sh1 mice (FIG. 10-11), respectively. This allowed us to mark the RPE within the transduced part of the eyecup, which was subsequently dissected and analyzed. (FIG. 6B, 7-8, 10-11). Subretinal delivery of AAV vectors to the pig retina was performed as previously described (11). All eyes were treated with 100 L of AAV2/8 vector solution. The AAV2/8 dose was 110.sup.10 (FIG. 3B) or 110.sup.11 GC of each vector/eye (FIGS. 5B and 16) and co-injection of dual AAV vectors resulted in a total dose of 210.sup.10 GC/eye or 210.sup.11 GC/eye, respectively.

(57) Western Blot Analysis

(58) Samples (HEK293 cells, retinas or eyecups) for Western blot analysis were lysed in RIPA buffer (50 mM Tris-Hcl pH 8.0, 150 mM NaCl, 1% NP40, 0.5% Na-Deoxycholate, 1 mM EDTA pH 8.0, 0.1% SDS) to extract EGFP and MYO7A proteins, or in SIE buffer (250 mM sucrose, 3 mM imidazole pH 7.4, 1% ethanol, and 1% NP-40) to extract ABCA4 protein.

(59) Pig samples (the treated areas of the retina as well as whole RPE sheets) were lysed in RIPA buffer to extract MYO7A from RPE sheets, and in SIE buffer to extract MYO7A and ABCA4 from retinas.

(60) Lysis buffers were supplemented with protease inhibitors (Complete Protease inhibitor cocktail tablets, Roche, Milan, Italy) and 1 mM phenylmethylsulfonyl. After lysis EGFP and MYO7A samples were denatured at 99 C. for 5 minutes in 1 Laemli Sample buffer; ABCA4 samples were denatured at 37 C. for 15 minutes in 1 Laemli sample buffer supplemented with 4M urea. Lysates were separated by 7% (ABCA4 and MYO7A samples) or 12% (EGFP samples) SDS-polyacrylamide gel electrophoresis. The antibodies used for immuno-blotting are as follows: anti EGFP (sc-8334, Santa Cruz, Dallas, Tex., USA, 1:500); anti-3flag (A8592, Sigma-Aldrich, 1:1000); anti-Myo7a (polyclonal, Primm Srl, Milan, Italy, 1:500) generated using a peptide corresponding to aminoacids 941-1070 of the human MYO7A protein; anti-HA antibody (PRB-101P-200, HA.11, Covance, Princeton, N.J., USA, 1:2000); anti- Tubulin (T5201, Sigma Aldrich, 1:10000); anti-Filamin A (catalog#4762, Cell Signaling Technology, Danvers, Mass., USA, 1:1000); anti-Dysferlin (Dysferlin, clone Ham1/7B6, MONX10795, Tebu-bio, Le Perray-en-Yveline, France, 1:500). The quantification of EGFP, ABCA4 and MYO7A bands detected by Western blot was performed using ImageJ software (free download is available at http://rsbweb.nih.gov/ij/). ABCA4 and MYO7A expression was normalized to Filamin A or Dysferlin for the in vitro and in vivo experiments, respectively. EGFP expression was normalized to -Tubulin or g of proteins for in vitro and in vivo experiments, respectively. Different proteins were used for normalization based on the similarity of their molecular weight to those of the different transgene products.

(61) Fundus Photography

(62) The fundus live-imaging was performed by dilating the eye of C57BL/6 with a drop of tropicamide 1% (Visufarma, Rome, Italy) and subsequent eye stimulation with a 300 W flash. Fundus photographs were taken using a Topcon TRC-50IX retinal camera connected to a charge-coupled-device Nikon D1H digital camera (Topcon Medical System, Oakland, N.J., USA).

(63) Histology, Light and Fluorescence Microscopy

(64) To evaluate EGFP expression in histological sections, eyes from C57BL/6 mice or Large White pigs (11) were enucleated one month after AAV2/8 injection. Mouse eyes were fixed in 4% paraformaldehyde over-night and infiltrated with 30% sucrose over-night; the cornea and the lens were then dissected and the eyecups were embedded in optimal cutting temperature compound (O.C.T. matrix, Kaltek, Padua, Italy). Pig eyes were fixed in 4% paraformaldehyde for 48 hours, infiltrated with 10% sucrose for 4 hours, 20% sucrose for 4 hours and finally 30% sucrose overnight. Then, the cornea, the lens, and the vitreous body were dissected and the EGFP-positive portions of the eyecups were embedded in optimal cutting temperature compound (O.C.T. matrix, Kaltek). Serial cryosections (10 m thick) were cut along the horizontal meridian and progressively distributed on slides. Retinal histology pictures were captured using a Zeiss Axiocam (Carl Zeiss, Oberkochen, Germany). To analyze melanosome localization in the RPE of pigmented sh1 mice, eyes were enucleated 2 months following the AAV injection, fixed in 2% glutaraldehyde-2% paraformaldehyde in 0.1M phosphate buffer over-night, rinsed in 0.1M phosphate buffer, and dissected under a florescence microscope. The EGFP-positive portions of the eyecups were embedded in Araldite 502/EMbed 812 (catalog #13940, Araldite 502/EMbed 812 KIT, Electron Microscopy Sciences, Hatfield, Pa., USA). Semi-thin (0.5-m) sections were transversally cut on a Leica Ultratome RM2235 (Leica Microsystems, Bannockburn, Ill., USA), mounted on slides, and stained with Epoxy tissue stain (catalog #14950, Electron Microscopy Sciences). Melanosomes were counted by a masked operator analyzing 10 different fields/eye under a light microscope at 100 magnification. Retinal pictures were captured using a Zeiss Axiocam (Carl Zeiss).

(65) Electron Microscopy and Immuno-Gold Labelling

(66) For electron microscopy analyses eyes were harvested from Abca4/ or sh1 mice at 3 and 2 months after AAV injection, respectively. Eyes were fixed in 0.2% glutaraldehyde-2% paraformaldehyde in 0.1M PHEM buffer pH 6.9 (240 mM PIPES, 100 mM HEPES, 8 mM MgCl.sub.2, 40 mM EGTA) for 2 hours and then rinsed in 0.1 M PHEM buffer. Eyes were then dissected under light or fluorescence microscope to select the Tyrosinase- or EGFP-positive portions of the eyecups of albino (Abca4/ and BALB/c) and pigmented sh1 mice, respectively. The transduced portion of the eyecups were subsequently embedded in 12% gelatin, infused with 2.3M sucrose and frozen in liquid nitrogen. Cryosections (50 nm) were cut using a Leica Ultramicrotome EM FC7 (Leica Microsystems) and extreme care was taken to align PR connecting cilia longitudinally. Measurements of RPE thickness and counts of lipofuscin granules in Abca4/ eyes were performed by a masked operator (Roman Polishchuk) using the iTEM software (Olympus SYS, Hamburg, Germany). Briefly, RPE thickness was measured in at least 30 different areas along the specimen length using the Arbitrary Line tool of iTEM software. The Touch count module of the iTEM software was utilized to count the number of lipofuscin granules in the 25 m.sup.2 areas distributed randomly across the RPE layer. The granule density was expressed as number of granules per 25 m.sup.2. The immuno-gold analysis aimed at testing the expression of ABCA4-HA in Abca4/ samples after AAV vector delivery was performed by incubating cryosections successively with monoclonal anti-HA antibody (MMS-101P-50, Covance, 1:50), rabbit anti-mouse IgG, and 10-nm gold particle-conjugated protein A. To quantify rhodopsin localization to the connecting cilium of sh1 PR, cryosections of sh1 mice were successively incubated with anti-rhodopsin antibody (1D4, ab5417, Abcam, Cambridge, UK, 1:100), rabbit anti-mouse IgG, and 10-nm gold particle-conjugated protein A. The quantification of gold density of rhodospin in the connecting cilia was performed by a masked operator using iTEM software (Olympus SYS). Briefly, the Touch count module of the iTEM software was used to count the number of gold particles per cilium that were normalized to the cilium perimeter (nm) that was measured using the Closed polygon tool. Gold density was expressed as gold particles/nm. Immunogold labelled cryosections were analyzed under FEI Tecnai-12 (FEI, Eindhoven, The Netherlands) electron microscope equipped with a Veletta CCD camera for digital image acquisition.

(67) Electrophysiological Analyses

(68) To assess the recovery from light desensitization eyes were stimulated with 3 light flashes of 1 cd s/m2 and then desensitized by exposure to constant light (300 cd/m2) for 3 minutes. Then, eyes were stimulated over time using the pre-desensitization flash (1 cd s/m2) at 0, 5, 15, 30, 45 and 60 minutes post-desensitization. The recovery of rod activity was evaluated by performing the ratio between the b-wave generated post-desensitization (at the different time points) and that generated pre-desensitization. The recovery from light desensitization was evaluated in 2-month-old Abca4/ mice at 6 weeks post treatment (FIG. 13).

(69) Statistical Analysis

(70) Data are presented as meanstandard error of the mean (s.e.m.). Statistical p values<0.05 were considered significant. One-way ANOVA with post-hoc Multiple Comparison Procedure was used to compare data depicted in: FIG. 2 (p ANOVA: A. 0.0002; B. 0.0015; C. 210.sup.7); FIG. 8B (p ANOVA: 0.076); FIG. 11B (p ANOVA: 0.5). As lipofuscin granules (FIG. 7B) and melanosomes (FIG. 10B) were counted, counts were analyzed by deviance from a Negative Binomial generalized linear models (61) (FIG. 7B: p value analysis of deviance 0.03794; FIG. 10B: p value analysis of deviance<<210.sup.10). The statistically significant differences between groups determined with the post-hoc Multiple Comparison Procedure are marked by asterisks in the Figures.

(71) Results

(72) Generation of Normal Size, Oversize and Dual AAV Vectors.

(73) The inventors generated oversize (OZ), dual AAV trans-splicing (TS), and hybrid vectors that included either the reporter EGFP, the therapeutic ABCA4-3flag or the MYO7A-HA coding sequences. The inventors also generated dual AAV trans-splicing (TS), and hybrid vectors that included the therapeutic CEP290 tagged at its C-terminus with HA tag. The recombinogenic sequences included in the dual AAV hybrid vectors were based on either a previously reported region of the alkaline phosphatase transgene (AP, dual AAV hybrid AP) (39) or a 77 bp sequence from the F1 phage genome (AK, dual AAV hybrid AK) that the inventors found to be recombinogenic in previous experiments (Colella and Auricchio, unpublished data). The inventors also generated dual AAV overlapping (OV) vectors for ABCA4, MYO7A and CEP290. The inventors did not generate dual AAV OV vectors for EGFP because the efficiency of this approach relies on transgene-specific overlaps for reconstitution (38) and therefore cannot be extrapolated from one gene to another. Instead, for EGFP the inventors generated single AAV vectors of normal size (NS) to compare levels of transgene expression from the various strategies. The constructs generated for production of all AAV vectors used in this study are listed in Table 1 and a schematic representation of the various approaches is depicted in FIG. 1.

(74) The inventors used AAV2/2 vectors for the in vitro experiments, with the ubiquitous cytomegalovirus (CMV) or chicken beta-actin (CBA) promoters, which efficiently transduce HEK293 cells (40). In addition, since the use of heterologous ITRs from AAV serotypes 2 and 5 can increase the productive reassembly of dual AAV vectors (51), the inventors also generated dual AAV AK vectors with heterologous ITRs (FIG. 17a) encoding ABCA4 and MYO7A. AAV vectors with heterologous ITRs were packaged in AAV capsids from serotype 2 and tested in vitro.

(75) In the experiments performed in vivo in the retina, The inventors used AAV2/8 vectors, which efficiently transduce RPE and PR (10-12) but poorly infect HEK293 cells, and either the ubiquitous CBA and CMV promoters (11), or the RPE-specific vitelliform macular dystrophy 2 (VMD2) (41) or the PR-specific Rhodopsin (RHO) and Rhodopsin kinase (RHOK) promoters (10) (Table 1).

(76) Dual AAV Vectors Allow High Levels of Transduction In Vitro.

(77) The inventors initially compared the efficiency of the various OZ, dual AAV OV, TS and hybrid AP and AK strategies for AAV-mediated large gene transduction in vitro by infecting HEK293 cells with the AAV2/2 vectors [multiplicity of infection, m.o.i.: 10.sup.5 genome copies (GC)/cell of each vector] with ubiquitous promoters (CMV for EGFP, ABCA4-3flag, and CEP290-HA, and CBA for MYO7A-HA).

(78) Cell lysates were analyzed by Western blot with anti-EGFP (FIG. 2A), -3flag (to detect ABCA4-3flag, FIG. 2B), -MYO7A (FIG. 2C) and -HA (to detect CEP290-HA) (FIG. 12A) antibodies. Representative Western blots are shown in FIGS. 2A-C and 12A. All strategies resulted in the expression of proteins of the expected size. As predicted, no bands of the expected size were observed when only one of the dual AAV vectors was used for infection (FIGS. 2A-C and 12A). Quantification of transgene expression (FIG. 2D-F) showed that the dual AAV hybrid AP approach resulted in the lowest levels of transgene expression, while the dual AAV OV, TS and hybrid AK approaches were more efficient than the AAV OZ approach. Dual AAV TS and hybrid AK approaches confirmed their ability to efficiently express large genes also in the case of CEP290 (FIG. 12B). In addition, the use of dual AAV AK vectors with heterologous ITRs resulted in expression of full-length ABCA4 and MYO7A proteins in vitro (FIG. 17).

(79) Dual AAV TS and Hybrid AK but not OV Vectors Transduce Mouse and Pig Photoreceptors.

(80) The inventors then evaluated each of the AAV-based systems for large gene transduction in the mouse retina. To test the dual AAV OV, which was transgene-specific, The inventors used the therapeutic ABCA4 and MYO7A genes (FIG. 3). The inventors used EGFP to evaluate the AAV OZ and the dual AAV TS, hybrid AP and AK approaches (FIG. 4). Western blot analysis on retinal lysates, one month after subretinal delivery in C57BL/6 mice of the dual AAV OV vectors (dose of each vector/eye: 1.310.sup.9 GC), encoding ABCA4-3flag from the ubiquitous CMV promoter, revealed robust protein expression (FIG. 3A). To determine which cell type in the retina expressed ABCA4, The inventors used dual AAV OV vectors that contained either the PR-specific RHO and RHOK, or the RPE-specific VMD2 (dose of each vector/eye: 110.sup.9 GC) promoters. The inventors detected ABCA4 protein expression in retinas injected with the VMD2 but not in those containing the RHO and RHOK promoters (FIG. 3A). These results were also confirmed in the Large White pig retina. The pig retina is an excellent model to evaluate vector efficiency because of its size, which is similar to the human retina, and because it is enriched with cones that are concentrated in a streak-like region whose cone density is comparable to that of the primate macula (11). The inventors injected Large White pig subretinally with dual AAV OV vectors encoding ABCA4-3flag (dose of each vector/eye: 110.sup.10 GC), and observed ABCA4 protein expression with the CMV but not the RHO promoter (FIG. 3B). Similarly, subretinal administration of dual AAV OV vectors encoding MYO7A-HA resulted in weak MYO7A protein expression in the mouse retina with the ubiquitous CBA (dose of each vector/eye: 2.510.sup.9 GC) and no detectable expression with the RHO (dose of each vector/eye: 3.210.sup.9 GC) promoter (FIG. 3C). Overall, these data suggested that the dual AAV OV approach was more efficient for large gene transfer to RPE than to PR, which are a major target of gene therapy for IRDs, such as STGD and USH1B.

(81) To find an AAV-based strategy that efficiently transduces large genes in PR, the inventors evaluated the retinal transduction properties of the AAV OZ and dual AAV TS, hybrid AP, and AK approaches. The inventors initially used EGFP, which allowed us to easily localize transgene expression in the various retinal cell types including PR as well as to properly compare the levels of AAV-based large transgene transduction to those of a single AAV NS vector. C57BL/6 mice were subretinally injected with AAV NS, OZ and dual AAV TS, and hybrid AP and AK vectors (dose of each vector/eye: 1.710.sup.9 GC), all encoding EGFP under the transcriptional control of the CMV promoter. One month later, fundus photographs showed that the highest levels of fluorescence were obtained with the AAV NS, and dual AAV TS and hybrid AK approaches (FIG. 15). Fluorescence microscope analysis of retinal cryosections showed that detectable levels of RPE or PR transduction could be observed in: 77% (10/13) retinas injected with AAV NS and OZ vectors; 92% (12/13) retinas injected with dual AAV TS, hybrid AP and AK vectors. FIG. 4 shows the best transduced retinas from each of these groups. The most robust levels of PR transduction were obtained with the AAV NS and dual AAV TS and hybrid AK approaches.

(82) The inventors then assessed PR-specific transduction levels in C57BL/6 mice following subretinal administration of dual AAV TS and hybrid AK vectors, which appears the most promising for large gene reconstitution in PR, as well as AAV NS vectors for comparison (dose of each vector/eye: 2.410.sup.9 GC). All vectors encoded EGFP under the transcriptional control of the PR-specific RHO promoter. One month after vector administration retinas were cryosectioned and analyzed under a fluorescence microscope (FIG. 5A). All approaches resulted in high levels of PR transduction, which seemed more consistent with the single AAV NS vector. The inventors found PR transduction in: 100% (6/6) of the retinas injected with AAV NS; 60% (9/15) of the retinas injected with dual AAV TS; 71% (10/14) of the retinas injected with dual AAV hybrid AK. FIG. 5A shows the best transduced retinas from each of these groups. Thus, the inventors conclude that dual AAV TS and hybrid AK strategies allow efficient mouse PR transduction although at levels which are lower than those obtained with a NS AAV. The inventors then confirmed that subretinal administration of dual AAV TS and hybrid AK vectors (dose of each vector/eye: 110.sup.11 GC; EGFP-positive retinas out of total injected: 2/2 dual AAV TS; 2/2 dual AAV hybrid AK) transduced PR of White Large pigs (FIG. 5B).

(83) In addition, subretinal delivery to the pig retina of dual AAV TS and hybrid AK vectors (dose of each vector/eye: 110.sup.11) resulted in efficient expression of both full-length ABCA4-3flag specifically in PRs (FIG. 16a) and full-length MYO7A-HA in RPE and PRs (FIG. 16b) Interestingly, dual AAV hybrid AK vectors resulted in more consistent expression of the large ABCA4 and MYO7A proteins in PRs, compared with dual AAV TS vectors (FIG. 16).

(84) Dual AAV Vectors Improve the Retinal Phenotype of STGD and USH1B Mouse Models.

(85) To understand whether the levels of PR transduction obtained with the dual AAV TS and hybrid AK approaches may be therapeutically relevant, the inventors investigated them in the retina of two mouse models of IRDs, STGD and USH1B caused by mutations in the large ABCA4 and MYO7A genes, respectively.

(86) Although the Abca4/ mouse model does not undergo severe PR degeneration (42), the absence of the ABCA4-encoded all-trans retinal transporter in PR outer segments (43-44) causes an accumulation of lipofuscin in PR as well as in RPE, as result of PR phagocytosis by RPE (45). As a consequence, both the number of lipofuscin granules in the RPE and the thickness of RPE cells are greater in Abca4/ mice than in control mice (45). Moreover the Abca4/ mouse model is characterized by delayed dark adaptation (57, 62). Since ABCA4 is expressed specifically in PR, the inventors generated dual AAV TS and hybrid AK vectors encoding ABCA4-3flag under the transcriptional control of the RHO promoter. These vectors were subretinally injected in wild-type C57BL/6 mice (dose of each vector/eye: 3-510.sup.9 GC) and one month later retinas were lysed and analyzed by Western blot with anti-3flag antibodies. Both approaches resulted in robust yet variable levels of ABCA4-3flag expression. ABCA4-3flag expression levels were more consistent in retina treated with the dual AAV hybrid AK vectors (FIG. 6A). These results were confirmed in Large White pigs (data not shown). In addition, one month-old albino Abca4/ mice were injected subretinally with the dual AAV hybrid AK RHO-ABCA4-HA vectors (dose of each vector/eye: 1-310.sup.9 GC). Three months later, eyes were harvested and immuno-electron microscopy analysis with anti-hemagglutinin (HA) antibodies of retinal sections confirmed that immunogold particles were correctly localized in PR outer segments only in animals that were injected with the combination of 5 and 3 dual AAV hybrid AK vectors (FIG. 6B). To assess the functionality of the ABCA4 protein expressed by the dual vectors, the inventors also performed transmission electron microscopy to assess the presence and number of RPE lipofuscin granules (FIG. 7) and RPE thickness (FIG. 8). Both were greater in the retina of Abca4/ mice injected with control vectors than in the retina of wild-type, age-matched Balb/C controls, and were reduced or normalized in the eyes injected with the therapeutic dual AAV TS or hybrid AK vectors (FIGS. 7B and 8B). In addition, the ability of Abca4/ photoreceptors to recover from light desensitization was significantly improved in the retinas treated with the therapeutic vectors when compared to control retinas (FIG. 13).

(87) The inventors then tested PR transduction levels and efficacy of dual AAV-mediated MYO7A gene transfer in the retina of sh1 mice, the most commonly used model of USH1B (23-24, 46-48). In sh1 mice, a deficiency in the motor Myo7a causes the mis-localization of RPE melanosomes (47), which do not enter into the RPE microvilli, and the accumulation of rhodopsin at the PR connecting cilium (48). Since MYO7A is expressed in both RPE and PR (22-23), the inventors then used dual AAV TS and hybrid AK vectors expressing MYO7A-HA under the transcriptional control of the ubiquitous CBA promoter. One month-old wild-type C57BL/6 mice were injected with the dual AAV vectors (dose of each vector/eye: 1.710.sup.9 GC) and eyecup lysates were evaluated one month later using Western blot analysis with anti-HA antibodies. Results showed similarly robust and consistent levels of MYO7A expression in retinas treated with both approaches (FIG. 9). Taking advantage of our anti-MYO7A antibody able to recognize both murine and human MYO7A, we compared the levels of MYO7A achieved following delivery of dual AAV vectors to the sh1/ eye to those expressed endogenously in the sh1+/+ eye (FIG. 14). We used both the CBA (FIG. 14, left panel, dose of each vector/eye: 1-610.sup.9GC) and the RHO promoters (FIG. 14, right panel, dose of each vector/eye: 210.sup.9GC) to distinguish MYO7A expression achieved in both PR and RPE from that in PR alone: the former is about 20% (FIG. 14, left panel) and the latter up to about 50% of endogenous Myo7a (FIG. 14, right panel). Our analysis additionally shows that the levels of MYO7A expression achieved in PR by dual AAV hybrid AK are higher than those obtained with the dual AAV TS vectors despite the number of transduced retinas is similar (TS-MYO7A: 3 retinas positive out of 8 injected; AK-MYO7A: 4 retinas positive out of 8 treated; FIG. 14, right panel).

(88) To test the ability of MYO7A expressed from dual AAV vectors to rescue the defects of the sh1/ retina, the inventors then subretinally injected the CBA sets of dual AAV TS and hybrid AK vectors (dose of each vector/eye: 2.510.sup.9 GC) in one month-old sh1 mice. The inventors assessed RPE melanosome (FIG. 10) and rhodopsin localization (FIG. 11) by analysis of semi-thin retinal section and by immuno-electron microscopy, respectively. Unlike unaffected sh1+/, the sh1/ melanosomes do not enter the RPE microvilli after delivery of control vectors (each single 5 half of the dual-AAV strategies, FIG. 10). The number of RPE melanosomes correctly localized apically was significantly improved after the delivery of either dual AAV TS or hybrid AK vectors encoding MYO7A (FIG. 10B). Remarkably, the inventors also found that the MYO7A expression mediated by dual AAV TS and hybrid AK vectors reduced the accumulation of rhodopsin at the connecting cilium of sh1/ PR (FIG. 11).

(89) Discussion

(90) While AAV-mediated gene therapy is effective in animal models and in patients with inherited blinding conditions (5-9, 49), its application to diseases affecting the retina and requiring a transfer of genes larger than 5 kb (referred to as large genes) is inhibited by AAV limited cargo capacity. To overcome this, the inventors compared the efficiency of various AAV-based strategies for large gene transduction including: AAV OZ and dual AAV OV, TS and hybrid approaches in vitro and in mouse and pig retina. In previous experiments, inventors selected a 77 bp sequence from the F1 phage genome that the inventors identified for its recombinogenic properties and used in the dual hybrid approach (AK, dual AAV hybrid AK).

(91) The inventors' in vitro and in vivo results show that the dual AAV hybrid AK surprisingly outperforms the dual AAV hybrid AP and that all dual AAV strategies the inventors tested (with the exception of the dual AAV hybrid AP) outperform AAV OZ vectors in terms of transduction levels. This may be explained by the homogenous size of the dual AAV genome population when compared to OZ genomes, which may favor the generation of transcriptionally active large transgene expression cassettes.

(92) The dual AAV OV approach seems particularly interesting when compared to the TS or hybrid AK approaches as dual AAV OV vectors only contain sequences belonging to the therapeutic transgene expression cassette. However, when the inventors administered dual AAV OV vectors to the subretinal space of adult mice and pigs, the inventors were only able to detect expression of the large ABCA4 protein when the ubiquitous or the RPE-specific promoters, but not the PR-specific promoters, were used. This may suggest that the homologous recombination required for dual AAV OV reconstitution is more efficient in RPE than PR. This is consistent with the low levels of homologous recombination reported in post-mitotic neurons (50) and may partially explain the lack of dual AAV OV-mediated MYO7A transduction recently reported by other groups (30). The inventors conclude that subretinal administration of dual AAV OV vectors should not be used for large gene transfer to PR, although the inventors cannot exclude that sequences that are more recombinogenic than those included in the inventors' dual AAV OV ABCA4 and MYO7A vectors may allow efficient homologous recombination in PR.

(93) Dual AAV TS and hybrid AK approaches efficiently transduce mouse and pig PR, differently from what the inventors observed with dual AAV OV. This is consistent with the knowledge that the mechanism of large gene reconstitution mediated by dual AAV TS and hybrid AK approaches may be via ITR-mediated head-to-tail rejoining (32, 35, 51) rather than homologous recombination.

(94) The levels of mouse PR transduction the inventors achieved with dual AAV TS and hybrid AK is lower and less consistent than with single NS vectors. However, dual AAV may be effective for treating inherited blinding conditions that require relatively low levels of transgene expression, i.e. diseases inherited as autosomal recessive. Indeed, the inventors show that subretinal delivery of dual AAV TS and hybrid AK improves and even normalizes the retinal defects of two animal models of inherited retinal diseases, STGD and USH1B, which are due to mutations in large genes and are attractive targets of gene therapy.

(95) The genome size of dual AAV vectors is homogenous, which means identity and safety issues related to their use should be less considerable than those related to AAV OZ vectors, which have heterogeneous genome sizes. In contrast, the inventors detected neither ERG or retinal histological abnormalities in the mice that the inventors followed up to 1-2 months after dual AAV vector delivery (data not shown).

(96) In conclusion, the inventors identified a new recombinogenic sequence (AK) that strikingly improves the performance of the AAV dual hybrid vector system. In fact they found that dual AAV vectors are efficient both in vitro and in the retina in vivo. While dual AAV OV vectors efficiently transduce RPE, they do not transduce PR, whereas dual AAV TS and hybrid AK approaches drive efficient large gene reconstitution in both cell types. Administration of dual AAV TS and hybrid AK approaches improved the retinal phenotype of mouse models of STGD and USH1B, providing evidence of the efficacy of these strategies for gene therapy for these and other blinding conditions, which require large gene transfer to retinal PR as well as RPE. These findings will greatly broaden the application of AAV vectors for gene therapies not only to eyes, but also to muscle as well as to other organs and tissues. Diseases other than IRD caused by defective genes larger than 5 kb include non-limiting examples of muscular dystrophies, dysferlin deficiencies (limb-girdle muscular dystrophy type 2B and Miyoshi myopathy), Cystic Fibrosis, Hemophilia.

REFERENCES

(97) 1. M. M. Sohocki, et al. Hum. Mutat. 17, 42-51 (2001). 2. T. Dryja, in The Online Metabolic & Molecular Bases of Inherited Diseases C. Scriver, A. Beaudet, W. Sly, D. Valle, Eds. (McGraw-Hill, New York, N.Y., 2001), vol 4, pp. 5903-5933. 3. P. Colella, et al. Trends Mol. Med. 15, 23-31 (2009). 4. L. H. Vandenberghe, A. Auricchio, Gene Ther. 19, 162-168 (2012). 5. J. W. Bainbridge, et al. N. Engl. J. Med. 358, 2231-2239 (2008). 6. A. V. Cideciyan, et al. N. Engl. J. Med. 361, 725-727 (2009). 7. A. M. Maguire, et al. N. Engl. J. Med. 358, 2240-2248 (2008). 8. A. M. Maguire, et al. Lancet 374, 1597-1605 (2009). 9. F. Simonelli, et al. Mol. Ther. 18, 643-650 (2010). 10. M. Allocca, et al. J. Virol. 81, 11372-11380 (2007). 11. C. Mussolino, et al. Gene Ther. 18, 637-645 (2011). 12. L. H. Vandenberghe, et al. Sci. Transl. Med. 3, 88ra54 (2011). 13. A. Auricchio, Hum. Gene Ther. 22, 1169-1170 (2011). 14. M. Natkunarajah, et al. Gene Ther. 15, 463-467 (2008). 15. B. Dong, et al. Mol. Ther. 18, 87-92 (2010). 16. Y. Lai, et al. Mol. Ther. 18, 75-79 (2010). 17. Z. Wu, et al. Mol. Ther. 18, 80-86 (2010). 18. Y. Wang, et al. Hum. Gene Ther. Methods 23, 225-233 (2012). 19. P. L. Hermonat, et al. FEBS Lett. 407, 78-84 (1997). 20. R. Allikmets, Nat. Genet. 17, 122 (1997). 21. J. M. Millan, et al. J. Ophthalmol. 2011, 417217 (2011). 22. T. Hasson, et al. Proc. Natl. Acad. Sci. USA 92, 9815-9819 (1995). 23. X. Liu, et al. Cell. Motif. Cytoskeleton 37, 240-252 (1997). 24. D. Gibbs, et al. Invest. Ophthalmol. Vis. Sci. 51, 1130-1135 (2010). 25. J. C. Grieger, et al. J. Virol. 79, 9933-9944 (2005). 26. J. Wu, et al. Hum. Gene Ther. 18, 171-182 (2007). 27. M. Allocca, et al. J. Clin. Invest. 118, 1955-1964 (2008). 28. P. E. Monahan, et al. Mol. Ther. 18, 1907-1916 (2010). 29. W. E. Grose, et al. PLoS One 7, e39233 (2012). 30. V. S. Lopes, et al. Gene Ther. (2013). 31. M. L. Hirsch, et al. Mol. Ther. 18, 6-8 (2010). 32. D. Duan, et al. J. Virol. 72, 8568-8577 (1998). 33. Z. Yan, et al. Proc. Natl. Acad. Sci. USA 97, 6716-6721 (2000). 34. D. Duan, et al. Mol. Ther. 4, 383-391 (2001). 35. A. Ghosh, et al. Mol. Ther. 16, 124-130 (2008). 36. Y. Lai, et al. Nat. Biotechnol. 23, 1435-1439 (2005). 37. S. J. Reich, et al. Hum. Gene. Ther. 14, 37-44 (2003). 38. A. Ghosh, et al. J. Gene Med. 8, 298-305 (2006). 39. A. Ghosh, et al. Hum. Gene Ther. 22, 77-83 (2011). 40. X. Dong, et al. PLoS One 5, e13479 (2010). 41. N. Esumi, et al. J. Biol. Chem. 279, 19064-19073 (2004). 42. L. Wu, et al. Adv. Exp. Med. Biol. 664, 533-539 (2010). 43. M. Illing, et al. J. Biol. Chem. 272, 10303-10310 (1997). 44. H. Sun, et al. Nat. Genet. 17, 15-16 (1997). 45. N. L. Mata, et al. Invest. Ophthalmol. Vis. Sci. 42, 1685-1690 (2001). 46. C. Lillo, et al. Adv. Exp. Med. Biol. 533, 143-150 (2003). 47. X. Liu, et al. Nat. Genet. 19, 117-118 (1998). 48. X. Liu, et al. J. Neurosci. 19, 6267-6274 (1999). 49. S. G. Jacobson, et al. Mol. Ther. 13, 1074-1084 (2006). 50. M. L. Fishel, et al. Mutat. Res. 614, 24-36 (2007). 51. Z. Yan, et al. J. Virol. 79, 364-379 (2005). 52. G. Gao, et al. Hum. Gene Ther. 11, 2079-2091 (2000). 53. A. Auricchio, et al. Hum. Gene Ther. 12, 71-76 (2001). 54. C. Mueller, et al. Curr. Protoc. Microbiol. Chapter 14, Unit 14D 11 (2012). 55. L. Drittanti, et al. Gene Ther. 7, 924-929 (2000). 56. Y. Zhang, et al. J. Virol. 74, 8003-8010 (2000). 57. R. A. Radu, et al. Proc. Natl. Acad. Sci. USA 101, 5928-5933 (2004). 58. V. E. Papaioannou, J. G. Fox, Lab. Anim. Sci. 43, 189-192 (1993). 59. F. Q. Liang, et al. in Methods in Molecular Medicine: Vision Research Protocols, P. E. Rakoczy, Ed. (Humana Press Inc, Totowa, N.J., 2000), vol. 47, pp. 125-139. 60. A. Gargiulo, et al. Mol. Ther. 17, 1347-1354 (2009). 61. V. N. Venables, B. D. Ripley, Modern Applied Statistics with S, S. J. Chambers, W. Eddy, W. Hardle, Sheater S, L. Tierney, Eds., (Springer Science+Business Media, New York, USA, 2002). [fourth edition]. Weng, J., et al. I Cell 98, 13-23 (1999).