Compositions and methods for producing recombinant aav

Abstract

To provide nucleic acid constructs, host insect cells, and methods for producing recombinant AAV capsids with high potency at high yield.SOLUTION: A nucleic acid construct comprises a nucleotide sequence encoding adeno-associated virus (AAV) VP1, VP2 and VP3 proteins, where the VP1 protein and the VP2 protein comprise two or more mutations at residues 157, 162, 164, 179, 188, 194, 196, 197, 200 and 201 in a specific sequence relative to wildtype VP1 and VP2 proteins.SELECTED DRAWING: None

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Application No. 62 / 806,317, filed February 15, 2019, which is hereby incorporated by reference in its entirety as part of this specification.

[0002] Sequence Listing This application contains a Sequence Listing that has been submitted electronically in ASCII format and is hereby incorporated by reference in its entirety as part of this specification. Said ASCII copy, created on February 13, 2020, is named 025297_WO004_SL.txt and is 47,056 bytes in size. [Background technology]

[0003] 2. Background of the Invention Adeno-associated viruses (AAVs) are small, non-enveloped viruses belonging to the Parvoviridae family, Dependoparvovirus genus. AAVs consist of a single-stranded DNA genome packaged into a capsid assembled with three capsid proteins (viral protein (VP) 1, VP2, and VP3) in a molar ratio of approximately 1:1:10 (Kondratov et al., Mol Ther. 25(12):2661-75 (2017)). The capsid proteins are encoded by a single-stranded capsid (cap) gene, which is produced by alternative splicing and differential codon usage (ibid.). VP1 is the largest capsid protein (81.6 kD). VP2 (66.6 kD) is an N-terminal truncated form of VP1. VP3 (59.9 kD) is an N-terminally truncated form of VP2 (see, e.g., Cecchini et al., Hum Gene Ther. 22:1021-30 (2011)).

[0004] Recombinant AAV (rAAV) has been actively studied as a vector for gene therapy and DNA vaccines in humans. The production of rAAV in mammalian cells requires a helper virus (e.g., adenovirus, vaccinia, or herpes virus). Over the years, several modifications have been made to facilitate the production of rAAV, including: (1) identification and cloning of a minimal set of helper proteins (adenovirus E1, E2A, E4, and VA), (2) provision of AAV Rep (Rep78, Rep68, Rep52, and Rep40) and capsid proteins that act in trans, and (3) development of a gene transfer system that allows packaging of DNA sequences flanked by the inverted repeats (ITRs) of AAV (Samulski et al., J Virol. 63(9):3822-8(1989)). By introducing these three components into mammalian cells, rAAV containing a transgene of interest can be easily purified. The recombinant AAV produced in this way has been used in clinical trials (Clement and Grieger, Mol Ther Methods Clin Dev. 3:16002 (2016)).

[0005] The need for scaled-up rAAV production for larger clinical trials and commercialization has led to the development of insect cell-based production systems that utilize baculovirus vectors for expression of Rep and capsid proteins and delivery of sequences encoding the AAV vector genome, including the transgene, to be packaged into the rAAV capsid. Such systems do not require adenovirus helper functions (see, e.g., Urabe et al. Hum Gene Ther. 13:1935-43 (2002); Urabe et al., J Vir. 80(4):1874-85 (2006); Chen et al., Mol Ther. 16(5):924-30 (2008); Smith et al., Mol Ther. 17(11):1888-96 (2009); and Mietzsch et al., Hum Gene Ther Methods 28(1):15-22 (2017)). Although the baculovirus-insect cell system has been successfully used to produce rAAV at various scales, it has been observed that rAAV produced in insect cells contains less VP1 (e.g., VP1:VP2:VP3 ratios of about 1:1:30 to 1:1:60) and results in lower titers compared to rAAV produced in mammalian cells (see, e.g., Urabe et al., 2002, supra; Kohlbrenner et al., Mol Ther. 12(6):1217-25 (2005); Urabe et al., 2006, supra; Aslanidi et al., Proc Natl Acad Sci USA 106(13):5059-64 (2009); Kondratov et al., supra; and Mietzsch et al., supra). Summary of the Invention [Problem to be solved by the invention]

[0006] Thus, there remains a need for improved baculovirus-insect cell systems for industrial-scale production of rAAV. [Means for solving the problem]

[0007] Summary of the Invention This disclosure describes modifications of the AAV cap gene in a baculovirus helper construct to improve capsid protein ratios, titers, and yields of rAAV produced in a baculovirus-insect cell system.

[0008] In one aspect, the disclosure provides a nucleic acid construct comprising a nucleotide sequence encoding AAV VP1, VP2, and VP3 proteins, wherein the VP1 protein and the VP2 protein comprise two or more mutations at residues 157, 162, 164, 179, 188, 194, 196, 197, 200, and 201 (numbering according to SEQ ID NO:1) compared to wild-type VP1 and VP2 proteins, respectively. As used herein, "numbering according to SEQ ID NO:1" refers to the position of an amino acid residue in SEQ ID NO:1, or the position of a corresponding amino acid residue in a different VP1 sequence (e.g., a VP1 sequence from a serotype other than AAV6). In some embodiments, the VP1 and VP2 proteins comprise two or more mutations selected from the group consisting of S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L (numbering according to SEQ ID NO: 1). In further embodiments, the VP1 and VP2 proteins comprise all ten of these mutations.

[0009] In some embodiments, the VP1 protein further comprises one or more mutations at residues 67, 81, 84, 85, and 92 (numbering according to SEQ ID NO: 1) compared to a wild-type VP1 protein. In certain embodiments, the VP1 protein comprises one or more mutations selected from the group consisting of A67E, Q81R, K84D, A85S, and R92K. In further embodiments, the VP1 protein comprises all five of these mutations.

[0010] In some embodiments, the VP1 protein and VP2 protein disclosed herein are identical to the VP1 protein and VP2 protein of AAV6, respectively, except for mutations.In certain embodiments, the VP1 protein and the VP2 protein are derived from AAV6 and comprise the following mutations compared to wild-type AAV6 VP1 and VP2 proteins, respectively: S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L.

[0011] In some embodiments of the nucleic acid construct, the nucleotide sequence also includes an open reading frame encoding an assembly activating protein (AAP), wherein the AAP comprises one or more mutations at residues 8, 10, 12, 17, 21, and 22 (numbering according to SEQ ID NO: 10) compared to wild-type AAP protein. As used herein, "numbering according to SEQ ID NO: 10" refers to the amino acid position in SEQ ID NO: 10 or the corresponding amino acid sequence position in a different AAP sequence (e.g., an AAP sequence from a serotype other than AAV6). In certain embodiments, the AAP comprises one or more (e.g., all) mutations selected from the group consisting of P8Q, H10L, L12Q, Q17P, L21Q, and L22V. In certain embodiments, the AAP is identical to the wild-type AAP of AAV6 except for the mutation of the AAP.

[0012] In certain embodiments, a nucleic acid construct of the disclosure comprises a sequence encoding SEQ ID NO:7, with or without the first amino acid. In further embodiments, the nucleic acid construct comprises nucleotides 18-151 of SEQ ID NO:14. In further embodiments, the nucleic acid construct comprises 10 or more (e.g., 20 or more, 30 or more, 40 or more, 50 or more, 75 or more, 100 or more, 125 or more, or 150 or more) contiguous nucleotides, or the entire nucleotide sequence, of SEQ ID NO:14.

[0013] In one aspect, the disclosure provides a nucleic acid construct comprising a nucleotide sequence encoding AAV VP1, VP2, and VP3 proteins, wherein the VP1 protein comprises one or more mutations at residues 81, 84, 85, and 92 (numbering according to SEQ ID NO: 1) compared to wild-type VP1 protein. In some embodiments, the VP1 protein comprises one or more mutations selected from the group consisting of Q81R, K84D, A85S, and R92K. In further embodiments, the VP1 protein comprises all four of these mutations. In certain embodiments, the VP1 protein further comprises a mutation at residue 67, e.g., A67E. In some embodiments, the VP1 protein is identical to the VP1 protein of AAV6 except for the mutation. In certain embodiments, the VP1 protein comprises the following mutations compared to wild-type AAV6 VP1: A67E, Q81R, K84D, A85S, and R92K.

[0014] In one aspect, the disclosure provides a nucleic acid construct comprising a nucleotide sequence encoding an AAP, wherein the AAP comprises one or more mutations at residues 8, 10, 12, 17, 21, and 22 (numbered according to SEQ ID NO: 10) compared to a wild-type AAP protein. In some embodiments, the AAP comprises one or more (e.g., all) mutations selected from the group consisting of P8Q, H10L, L12Q, Q17P, L21Q, and L22V.

[0015] In some embodiments, a nucleic acid construct of the disclosure further comprises an AAP rep gene.

[0016] In some embodiments, the nucleic acid construct of the disclosure is a baculovirus vector, wherein the capsid nucleotide sequence is operably linked to a promoter active in an insect cell.

[0017] The present disclosure also provides insect cells containing the nucleic acid constructs described herein, and recombinant AAV virions produced in the insect cells.

[0018] The present disclosure also provides a recombinant AAV virion comprising (i) a genome having a transgene of interest flanked by a pair of AAV inverted repeats (ITRs), and (ii) a VP1 protein expressed from a nucleic acid construct described herein, or a capsid assembled from both VP1 and VP2 proteins.

[0019] In yet another aspect, the disclosure provides a method for producing a recombinant AAV virion, the method comprising: providing an insect cell as described herein, wherein the insect cell also expresses AAV Rep proteins and comprises a sequence encoding an AAV vector comprising a transgene of interest flanked by a pair of AAV ITRs; culturing the insect cell under conditions that allow expression of VP1, VP2, VP3 proteins, replication of the AAV vector, and packaging of the AAV vector into a capsid assembled from the expressed VP1, VP2, and VP3 proteins; and recovering the packaged capsid. In some embodiments, the genes encoding the AAV Rep, VP1, VP2, and VP3 proteins are present on one or more nucleic acid constructs. In some embodiments, the genes encoding the AAV Rep proteins are present on one or more nucleic acid constructs that are not the nucleic acid constructs expressing the VP1, VP2, and VP3 proteins. In some embodiments, the genes encoding the AAV VP1, VP2, and VP3 proteins are present on one or more nucleic acid constructs that are not the nucleic acid constructs expressing the Rep proteins. In some embodiments, the genes encoding the Rep proteins and the genes encoding the VP1, VP2, and VP3 proteins are integrated into the genome of an insect cell.

[0020] The present disclosure also provides a pharmaceutical composition comprising the rAAV and a pharma- ceutical acceptable carrier.

[0021] Other features, objects, and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that this detailed description, while illustrating embodiments and aspects of the present invention, is given by way of illustration only and not by way of limitation. Various changes and modifications within the scope of the present invention will become apparent to those skilled in the art from this detailed description. [Brief description of the drawings]

[0022] BRIEF DESCRIPTION OF THE DRAWINGS [Figure 1-1] FIG. 1A shows an alignment of the amino acid sequences of VP1 / VP2 from AAV serotypes 3B, 5, 6, 8, and 9 (SEQ ID NOs: 16-21, respectively, in order of appearance) between amino acid residues 155 and 207 (numbering based on consensus sequence). To identify the proteolytic cleavage site of AAV6 VP1 / VP2, N-terminal Edman degradation sequencing was performed on the proteolytic cleavage products excised from SDS-PAGE gels. The cleavage site was identified as residues G190E191. Multiple sequence alignments were generated using the Geneious software package with ClustalW alignment at default settings. The start site of VP3 at residue 205 is indicated by a triangle.

[0023] Figure 1B shows an alignment of the amino acid sequences of VP1 / VP2 from AAV serotypes 1, 2, 3B, 4, and 6-11 (SEQ ID NOs: 22-26, 23, and 27-31, respectively, in order of appearance) between amino acid residues 155 and 205 (numbering based on consensus sequence). Multiple sequence alignments were generated using the Geneious software package and ClustalW alignment with default settings.

[0024] [Figure 1-2]Figure 1C is a table comparing the corresponding amino acid positions in AAV serotypes 1, 4, 6, 7, 8, and 11 between amino acid residues 158 and 203 aligned in Figure 1B. The table shows amino acid mutations that may prevent proteolytic cleavage in these serotypes. The numbering of the amino acid residues is based on the consensus sequence shown in Figure 1B.

[0025] [Figure 2-1] Figures 2A-D show that replacement of 45 consecutive amino acids of AAV9 VP1 / VP2 (SEQ ID NO: 4) with the corresponding regions of the AAV6 VP1 / VP2 sequence (SEQ ID NO: 3) abolishes proteolytic cleavage in the resulting chimeric VP1 / VP2 protein (AAV6 / 9 VP1 / VP2). Figure 2A: Coomassie-stained SDS-PAGE gel analysis of purified rAAV6. Asterisks indicate proteolytic cleavage fragments. Figure 2B: Densitometry and capsid ratio analysis of standard helper and AAV9 grafted forms. Figure 2C: Titer comparison of rAAV6 produced in HEK293 cells ("293 AAV") or Sf9 cells ("standard helper" and "AAV9 grafted forms"). Figure 2D: Alignment of the nucleotide sequence of the grafted AAV9 cap with the nucleotide sequence of the AAV6 cap replaced in the AAV9 grafted form. Arrow "A" indicates the start codon of the AAP. The AAV6 contains a transgene for factor IX ("FIX"), and viral titers were assayed by transducing the virus into HepG2 cells and subsequently measuring FIX in the cells by ELISA. Titers were calculated relative to rAAV6 produced in HEK29 cells ("293 AAV"). Standard (Std) helper: a baculovirus helper vector in which the AAV cap gene does not contain heterologous sequences from other serotypes. AAV9 grafted: a baculovirus helper vector in which the AAV6 cap gene contains sequences from the AAV9 cap gene, as described above and further below. Figure 2D discloses SEQ ID NOs: 13 and 14 in order of appearance. [Figure 2-2] Same as above

[0026] [Figure 3-1] Figure 3A is a photograph of a Coomassie stained SDS-PAGE gel showing that the L188I single point mutation in AAV6 VP1 / VP2 does not prevent proteolytic cleavage, as indicated by the protein species highlighted by the arrow. The asterisk indicates an additional cleavage product. Reference: rAAV6 produced in HEK293 cells. L188I: Baculovirus AAV6 helper construct encoding AAV6 VP1 / VP2 proteins containing the L188I mutation (numbering in SEQ ID NO: 1).

[0027] [Figure 3-2] Figure 3B is a bar graph showing the relative titers of rAAV6 produced by baculovirus helpers carrying various AAV cap genes. Titers were calculated relative to 293 AAV.

[0028] [Figure 4-1] Figures 4A-D show that AAV6 / 9 VP1 / VP2 variants prevent cleavage but do not improve AAV capsid titer. Figure 4A: Sequence alignment of AAV6 / 9 VP1 / VP2 variants 1, 2, 3, and 4 (numbering according to SEQ ID NO: 1). Figure 4A discloses SEQ ID NOs: 32-37, respectively, in order of appearance. [Figure 4-2] Figure 4B is a photograph of a Coomassie-stained SDS-PAGE gel. *: proteolytic cleavage products. **: additional cleavage products. [Figure 4-3] Figure 4C: Bar graph showing the relative titers of various AAV products. Reference: 293 AAV. Other definitions are as above. [Figure 4-4]Figure 4D: Partial alignment of cap / AAP nucleotide sequence and AAP amino acid sequence (rectangle) in AAV6, AAV9 graft, variant 1, variant 2, variant 3, and variant 4. In the indicated region, the AAP amino acid sequence of the AAV9 graft has six mutations (P8Q, H10L, L12Q, Q17P, L21Q, and L22V) compared to the AAP6 amino acid sequence, while the AAP amino acid sequences of variants 1-3 each have only two mutations (L21Q and L22V). Variant 4 has the native AAP6 sequence in the indicated region. Figure 4D discloses SEQ ID NOs: 38-44, 43, 45, 43, and 46-47, respectively, in order of appearance.

[0029] [Diagram 5] Figures 5A and 5B are comparisons of capsid protein ratios and titers of rAAV6 produced from standard helper and rAAV6 produced using a baculovirus AAV6 helper construct containing an AAV2 PLA2 (phospholipase A2 domain) graft ("AAV2 PLA2 graft" or "AAV6 / 2 VP1"), which contains a FIX transgene. Figure 5A: Capsid protein ratios of VP1:VP2:VP3 as determined by SDS-PAGE and Coomassie blue densitometry. Figure 5B: Titer comparisons of rAAV6 produced in (i) 293 cells, or (ii) Sf9 insect cells containing standard helper or an AAV6 helper delivering an engineered AAV6 VP1 with the AAV2 PLA2 domain. Titers were determined using an ELISA method to detect FIX cDNA transduction and expression of FIX, and were calculated relative to 293 AAV.

[0030] [Figure 6] Figure 6 is a graph comparing the activity of PLA2 in vitro. Vg: virus particles.

[0031] [Figure 7-1]Figure 7A shows an alignment of assembly activating protein (AAP) amino acid sequences from AAV serotypes 1, 2, 3B, 4, and 6-11 (SEQ ID NOs: 48-52, 8, 53-54, 9, and 55-56, respectively, in order of appearance) in the indicated regions. Amino acid residue numbering is based on consensus sequences. Multiple sequence alignments were generated using the Geneious software package and ClustalW alignment with default settings.

[0032] [Figure 7-2] Figure 7B shows the amino acid differences between (i) AAV1, AAV2, AAV4, AAV6, AAV7, AAV8, AAV10, or AAV11, and (ii) AAV9 or AAV3B in the indicated AAP regions (consensus numbering). The amino acid residue numbers are based on the consensus sequence shown in Figure 7A. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0033] Detailed Description of the Invention The present disclosure provides an improved baculovirus-insect cell system and related compositions for producing effective rAAV capsids in high yields. In this system, the cap gene in the baculovirus helper construct is mutated at two or more codons to make the VP1 and VP2 proteins encoded by this gene resistant to proteolysis during the production process. Because the cap gene also contains the open reading frame for AAP, alterations of these codons can also cause mutations in AAP. The inventors have made the unexpected discovery that mutations in VP1 / VP2 and / or AAP improve the infectivity, i.e., titer, of rAAV. Without being bound by theory, the inventors believe that this increased titer is due to improved integrity of the capsid proteins in insect cells and improved ability of the mutated AAP to promote capsid assembly.

[0034] Additionally or alternatively, the cap gene in the baculovirus helper construct is mutated at one or more codons in the region encoding the VP1 PLA2 domain, such that the engineered PLA2 domain acquires higher enzymatic activity. The inventors have made the unexpected discovery that higher PLA2 enzymatic activity results in higher yields of rAAV in insect cell production systems.

[0035] In this specification, when a specific AAV serotype is mentioned, it refers not only to the prototype of that serotype, but also to various isolates within the same clade as this prototype. For classification of AAV clades, see, for example, Gao et al., J Virol. (2004) 78(12):6381-8, the disclosure of which is hereby incorporated by reference as a part of this specification.

[0036] Removal of proteolytic sites in AAV VP1 / VP2 proteins In some embodiments, the improved baculovirus-insect cell system can be used to produce any AAV serotype that is susceptible to proteolysis in insect cells, including, for example, AAV1, AAV6, AAV8, or variants thereof, or any pseudotype or chimeric rAAV in which the VP1 / VP2 proteins are susceptible to proteolysis in insect cells.

[0037] "Pseudotype" or "cross-packaged" rAAV refers to a recombinant AAV whose capsid has been replaced with that of another AAV serotype, e.g., to alter the transduction capabilities or tropism profile of the virus (e.g., Balaji et al., J Surg Res. 184(1):691-8 (2013)). "Chimeric" or "hybrid" rAAV refers to a recombinant AAV whose capsid has been assembled from capsid proteins from different serotypes and / or whose capsid proteins are chimeric proteins with sequences from different serotypes (e.g., serotypes 1 and 2; see, e.g., Hauck et al., Mol Ther. 7(3):419-25 (2003)).

[0038] In the improved AAV production system of the present disclosure, two or more point mutations are introduced into the AAV VP1 / VP2 protein, e.g., from AAV1, AAV4, AAV6, AAV7, AAV8, or AAV11, to remove sites susceptible to proteolysis in insect cells. The point mutations introduced can be at residues identical to those at the corresponding positions in AAV2, AAV3, AAV5, AAV9, or AAV10.

[0039] A "corresponding" amino acid residue or region refers to an amino acid residue or region that matches (but is not necessarily identical to) the reference residue or region when the subject and reference sequences containing the residue or region are aligned for maximum homology (with gaps being recognized in the art). For example, amino acid residue 189(L) of AAV8 VP1 corresponds to amino acid residue 188 of VVP6 VP1.

[0040] In some embodiments, the baculovirus helper constructs of the present disclosure provide helper functions for the production of rAAV6, and the helper constructs contain a modified AAV6 cap gene (cap6) that encodes a mutated VP1 protein. The complete amino acid sequence of the AAV6 VP1 protein is shown below, with the start sites of VP2(T) and VP3(M) in bold and underlined: [ka] In the above sequence, the first X can be M (wild type; source: GenBank AAB95450.1), T, L, or V. In other embodiments, the start residue of VP1 can be another amino acid encoded by a non-canonical start codon, such as a suboptimal start codon. See, e.g., Kearse et al., Genes Dev. 31:1717-31 (2017).

[0041] In some embodiments, the AAV6 VP1 / VP2 protein comprises a mutation, relative to wild type, at two or more residues in a region corresponding to residues 138-208 (e.g., residues 151-201 or residues 157-201) of SEQ ID NO:1, where the mutated residues are selected from the group consisting of residues 157, 162, 164, 179, 188, 194, 196, 197, 200, and 201. In some embodiments, the AAV6 VP1 / VP2 protein comprises two or more mutations selected from the group consisting of S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L (numbering according to SEQ ID NO:1). For example, the AAV6 VP1 / VP2 protein may have the following mutations: (i) S179T, L188I, T194A, A196S, A197G, P200S, and T201L ("Variant 1"); (ii) S179T, L188I, T194A, A196S, and A197G ("Variant 2"); (iii) T194A, A196S, A197G, P200S, and T201L ("Variant 3"); (iv) S157A, T162S, and Q164A ("Variant 4"); or (v) P200S and T201L. For convenience, only the VP1 residue numbers are referred to herein. The corresponding residue numbers in VP2 can be readily discerned from SEQ ID NO:1 above. For example, residue S157 of VP1 is residue S20 of VP2.

[0042] In certain embodiments, the AAV6 VP1 / VP2 protein may include the L188I mutation and one or more other mutations in this group. For example, the AAV6 VP1 / VP2 protein may have the L188I, P200S, and T201L mutations.

[0043] In some embodiments, the AAV6 VP1 / VP2 protein contains all ten mutations, S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L, whose sequence in the region corresponding to residues 157-201 of SEQ ID NO:1 is as follows: AGIGKSGAQP AKKRLNFGQT GDTESVPDPQ PIGEPPAAPS GVGSL (SEQ ID NO: 4) This sequence is identical to that in the corresponding region of AAV9 VP1 / VP2.

[0044] Thus, using well-known molecular cloning techniques, the sequence encoding residues 157-201 of VP1 (i.e., residues 20-64 of VP2; SEQ ID NO:3) can be replaced with the sequence encoding the corresponding amino acid sequence of AAV9 (SEQ ID NO:4) to generate a modified cap6 gene in a baculovirus AAV6 helper construct, the resulting modified baculovirus AAV6 helper construct being referred to herein as an "AAV9 graft," and the resulting VP1 / VP2 proteins being referred to herein as "AAV6 / 9 VP1 / VP2." In some embodiments, the portion of the AAV6 cap gene sequence that is replaced comprises the underlined sequence shown below (see also FIG. 2D): AAGAGCCAGACTCCTCC TCGGGCATTGGCAAGACAGGCCAGCAGCCCGCTAAAAAGAGACTCAATTTTGGTCAGACTGGCGACTCAGAGTCAGTCCCCGACCCACAACCTCTCGGAGAACCTCCAGCAACCCCCGCTGCTGTGGGACCTAC TACAATGGCTTCAGGCGGTGGCGCACCAATGG (SEQ ID NO: 13)

[0045] In certain embodiments, the underlined portion of SEQ ID NO: 13 is replaced with the corresponding cap gene sequence from AAV9 to generate the AAV9 graft disclosed herein. In certain embodiments, the AAV9 cap sequence grafted into the AAV6 cap gene in the AAV9 graft comprises the underlined sequence shown below (see also FIG. 2D): AAGAGCCAGACTCCTCC GCGGGTATTGGCAAATCGGGTGCACAGCCCGCTAAAAAGAGACTCAATTTCGGTCAGACTGGCGACACAGAGTCAGTCCCAGACCCTCAACCAATCGGAGAACCTCCCGCAGCCCCCTCAGGTGTGGGATCTCT TACAATGGCTTCAGGCGGTGGCGCACCAATGG (SEQ ID NO: 14)

[0046] In other embodiments, a modified cap6 gene in a baculovirus AAV6 helper construct can be generated by replacing the sequence encoding VP1 residues 157-201 of SEQ ID NO:1 (i.e., residues 20-64 of VP2; SEQ ID NO:3) with a sequence encoding the corresponding amino acid sequence from a serotype whose VP1 / VP2 proteins are resistant to proteolytic cleavage in insect cells, e.g., AAV2, AAV3, AAV3B, AAV5, or AAV10.

[0047] In other embodiments, the VP1 / VP2 proteins of AAV serotypes, such as AAV1, AAV4, AAV7, AAV8, or AAV11, can be mutated such that they contain one or more of the mutations shown in Figure 1C. The resulting proteins are expected to be more resistant to proteolysis in insect cells. Non-limiting examples of engineered VP1 / VP2 proteins from various AAV serotypes are shown below: (1) The engineered AAV1 VP1 / VP2 proteins can contain one or more of the following mutations: S158T / A, I160T / V, T163A / S / K, Q165K / A, K169R, S180A / T, L189E / I, T195A, A197S / T, A198S / G, V199L, P201T / S, and T202N / L; (2) the engineered AAV4 VP1 / VP2 proteins can contain one or more of the following mutations: T158S / A, I160T / V, K163A / S, K165Q / A, K169R, K171R, V173N, E175G, D176Q / E, E177T, T178G, G179D / E, A180S / T, G181D / E, D182S, G183V, E189L / I, S191Q / E, T192P, S193P, G194A, M196P, S197T, D200G, D201T / S, S202N / L, and E203T; (3) the engineered AAV6 VP1 / VP2 proteins may contain one or more of the following mutations: S158T / A, I160T / V, T163A / S / K, Q165K / A, K169R, S180A / T, L189E / I, T195A, A197S / T, A198S / G, V199L, P201T / S, and T202N / L; (4) The engineered AAV7VP1 / VP2 proteins may contain one or more of the following mutations: T158S / A, I160T / V, K163A / S, Q165K / A, R169K, S180A / T, L189E / I, S197T, S198G, V199L, and G202N / L; (5) The engineered AAV8 VP1 / VP2 proteins can contain one or more of the following mutations: T158S / A, I160T / V, K163A / S, Q165K / A, R169K, S180A / T, L189E / I, S197T, G198S, V199L, P201T / S, and N202L; and (6) The engineered AAV11 VP1 / VP2 proteins can contain one or more of the following mutations: S158T / A, I160T / V, K163A / S, K165Q / A, R169K, E175G, E176Q, D177T, T178G, G179D / E, A180S / T, G181D / E, D182S, G183V, E189L / I, S191Q / E, D192P, T193P, S194A, M196P, S197T, S200G, D201T / S, I202N / L, and E203T (consensus numbering; see Figure 1C).

[0048] In some embodiments, VP1 / VP2 proteins from AAV serotypes that are susceptible to proteolysis in insect cells are mutated to contain one or more mutations at the following amino acid residues (consensus numbering in Figure 1C): 158, 163, 165, 180, 189, 195, 197, 198, 201, and 202. (These residues correspond to amino acid residues 157, 162, 164, 179, 188, 194, 196, 197, 200, and 201, respectively, according to the numbering of SEQ ID NO:1 (AAV6), or to corresponding sequences from other serotypes aligned for maximum homology to SEQ ID NO:1.) In further embodiments, the VP1 / VP2 proteins may further comprise one or more mutations at the following amino acid residues (consensus numbering in FIG. 1C): 160, 169, 171-179, 181-183, 191-194, 196, 199, and 203.

[0049] In some embodiments, VP1 / VP2 proteins from AAV serotypes that are susceptible to proteolysis in insect cells are mutated to contain one or more of the following mutations: S158T / A, I160T / V, K / T163S / A / T, Q165K / A, K169R, K171R, V173N, E175G, D176Q / E, D / E177T, T178G , G179D / E, S180A / T, G181D / E, D182S, G183V, L189E / I, Q / S191E, T / D192P, S / T193P, G / S194A, T / G195A, M196P, A197S / T, A198G / S, V199L, S200G, D / P201S / T, T / S202N / L, and E203T (consensus numbering in Figure 1C ).

[0050] In some embodiments, the engineered VP1 / VP2 proteins can contain one or more of the following mutations: S158A, T163S, Q165A, S180T, L189I, T195A, A197S, A198G, P201S, and T202L (consensus numbering in FIG. 1C). (These residues correspond to amino acid residues S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L, respectively, according to the numbering of SEQ ID NO:1 (AAV6), or to corresponding sequences from other serotypes aligned for maximum homology to SEQ ID NO:1.)

[0051] The mutations described herein eliminate a site in the AAV capsid protein that is susceptible to proteolytic cleavage in insect cells, and thus helper constructs containing modified cap genes not only produce improved capsid proteins but also produce rAAV products with greater purity and homogeneity.

[0052] Surprisingly, the inventors also found that point mutations introduced herein into the AAV VP1 / VP2-specific region (i.e., the region common to VP1 and VP2 but absent from VP3) can significantly improve rAAV titer through a mechanism that does not solely depend on prevention of proteolytic cleavage.

[0053] Increased rAAV production yield The disclosure also provides a baculovirus-insect system in which the cap gene in the baculovirus helper construct is altered to improve rAAV production yields in insect cells with respect to a VP1 unique region (i.e., a region present in VP1 but not in VP2 or VP3), which contains the PLA2 domain (corresponding to residues 1-137 of SEQ ID NO:1), and mutations in this region can increase the enzymatic activity of the PLA2 domain of the resulting VP1 protein.

[0054] In some embodiments, the baculovirus helper constructs of the present disclosure provide helper functions for the production of rAAV6 or rAAV9, and the helper constructs comprise a modified AAV6 or AAV9 cap gene (cap6 or cap9, respectively) with a mutated PLA2 domain. The mutated PLA2 domain may comprise a mutation at one or more positions in the region corresponding to residues 1-137 (e.g., 52-97 or 67-92) of SEQ ID NO:1, compared to wild type, where the positions are selected from the group consisting of residues 67, 81, 84, 85, and 92 (numbering according to SEQ ID NO:1). In some embodiments, the AAV6 VP1 protein comprises one or more mutations selected from the group consisting of A67E, Q81R, K84D, A85S, and R92K (numbering according to SEQ ID NO:1). In a specific embodiment, the AAV6 VP1 protein contains all five of the mutations and its sequence in the region corresponding to residues 52-97 of SEQ ID NO:1 (SEQ ID NO:5) is as follows: YLGPFNGLDK GEPVNEADAA ALEHDKAYDR QLDSGDNPYL KYNHAD (SEQ ID NO: 6) This sequence is identical to that in the corresponding region of AAV2 VP1, and therefore, to generate a modified cap6 gene in a baculovirus AAV6 helper construct, the sequence encoding residues 52-97, or 67-92 of VP1 can be replaced with the sequence encoding the corresponding amino acid sequence of AAV2, using well-known molecular cloning techniques.

[0055] Surprisingly, the inventors have found that the point mutations introduced herein into the AAV VP1 unique region significantly improve the enzymatic activity of the VP1 PLA2 domain and also result in a significant (e.g., 2-fold or more, 3-fold or more, 4-fold or more, 5-fold or more) improvement in the yield of rAAV in insect cells.

[0056] In certain embodiments, the engineered cap gene encodes an AAV6 VP1 protein that contains both the aforementioned mutations in the PLA2 domain and the aforementioned mutation that removes the proteolytic site. One exemplary modified AAV6 VP1 protein ("AAV6 / 2 / 9 VP1") has the following sequence: [ka] Other exemplary modified AAV6 VP1 proteins have the same sequence as SEQ ID NO:7, except that the initiation residue is absent or is another amino acid encoded by a non-canonical initiation codon, such as T, L, V, or a suboptimal initiation codon. Baculovirus helper constructs encoding one of these modified AAV6 VP1 proteins can be used to produce rAAV or pseudotyped or chimeric rAAV with improved titers and yields in insect cells.

[0057] Modification of AAP In other aspects, the disclosure provides AAV helper constructs that encode engineered AAPs with improved ability to stabilize capsid proteins and promote capsid assembly. The inventors have found that changing residues 1-30 of AAV6 AAP ("AAP6") to residues 1-30 of AAV9 AAP ("AAP9") significantly increases the titer of rAAV produced from this engineered helper construct. Residues 1-30 of AAP6 and AAP9 are shown below, with the six amino acid differences in wild-type AAV9 compared to wild-type AAP6 shown in bold and boxed: [ka] The complete AAV6 AAP wild type sequence is shown below: MATQSQSPTH NLSENLQQPP LLWDLLQWLQ AVAHQWQTIT KAPTEWVMPQ EIGIAIPHGW ATESSPPAPE HGPCPPITTT STSKSPVLQR GPATTTTTSA TAPPGGILIS TDSTAISHHV TGSDSSTTIG DSGPRDSTSS SSTSKSRRSR RMMASRPSLI TLPARFKSSR TRSTSCRTSS ALRTRAASLR SRRTCS (SEQ ID NO: 10)

[0058] In some embodiments, the engineered AAP protein contains one or more mutations at residues 8, 10, 12, 17, 21, and 22 (numbering based on the consensus sequence in Figure 7A). In further embodiments, the engineered AAP protein may contain the following amino acid residues: Q8, L10, Q12, P17, Q21, and V22. In some embodiments, the engineered AAP protein may contain one or more of the following mutations compared to the wild-type AAP sequence: P8Q, H10L, L12Q, Q17P, L21Q, and L22V. Non-limiting examples of engineered, improved AAP proteins are provided below. (1) Engineered AAP1 (serotype 1 AAP) contains one or more of the following mutations: P8Q, I9T, H10L, L12Q, L16H, Q17P, L21Q, L22V, and L26I; (2) engineered AAP2 (serotype 2 AAP) contains one or more of the following mutations: E2A, T5S, Y7S, L8Q, P10L, S11N, L12Q, D14E, S15N, H16L, Q17P, L21Q, E24D, I26L, and R27Q; (3) engineered AAP4 (serotype 4 AAP) contains one or more of the following mutations: E2A, Q3T, A4Q, T5S, D6Q, P7S, L12Q, R13S, D14E, Q15N, L16H, P17Q, E18Q, C22V, L23W, M24D, T25L, V26I / L, R27Q, and C28W; (4) engineered AAP6 (serotype 6 AAP) contains one or more of the following mutations: P8Q, H10L, L12Q, L16H, Q17P, L21Q, L22V, and L26I; (5) engineered AAP7 (serotype 7 AAP) contains one or more of the following mutations: P8Q, L12Q, L16H, Q17P, L21Q, L22V, and L26I; (6) engineered AAP8 (AAP of serotype 8) contains one or more of the following mutations: F7S, L12Q, L16H, Q17P, R19P, L21Q, and I26L; (7) engineered AAP10 (serotype 10 AAP) containing one or more of the following mutations: S3T, P8Q, Q12L, L16H, Q17P, A19P, L21Q, and V26I / L; and (8) Engineered AAP11 (serotype 11 AAP) contains one or more of the following mutations: E2A, P3T, E4Q, T5S, D6Q, P7S, L12Q, R13S, D14E, Q15N, I16L / H, P17Q, A18Q, C22V, L23W, Q24D, T25L, L26I, K27Q, and C28W (see also Figure 7B ).

[0059] To generate engineered AAP proteins, point mutations may be introduced. Also, because the open reading frame encoding AAP is embedded in the cap gene and is translated only by frameshifting, portions of the cap gene, including sequences encoding the N-terminal portion of AAP (e.g., residues 1-30, residues 1-28, residues 2-30, residues 2-28, etc.), may be transplanted from other serotypes. In some embodiments, portions of the AAV9 cap gene encoding AAV9 VP1 / VP2-specific regions (i.e., regions common to VP1 and VP2 but absent from VP3), e.g., the cap9 sequence including nucleotides 18-151 of SEQ ID NO:14, are replaced with the corresponding region of cap6 to generate engineered AAV6 VP1 / VP2 with portions of AAV9 VP1 / VP2 (e.g., to remove proteolytic sites as discussed above) and an engineered AAP6 in which the N-terminal 1-30 amino acid residues are identical to those of AAP9. In some embodiments, the AAP mutations are generated in a manner similar to the AAV9 cap gene transplantation described above, and further described in Example 1 below.

[0060] Production of rAAV in insect cells Production of rAAV in insect cells can be carried out as previously described, see, e.g., Urabe et al., 2002 and 2006, supra; Chen et al., supra; Smith et al., supra; Mietzsch et al., supra; WO2007 / 046703, WO2007 / 148971, WO2009 / 104964, WO2013 / 036118, and WO2008 / 024998, all of which are hereby incorporated by reference in their entireties.

[0061] The insect cells in the production method of the present disclosure contain a baculovirus helper construct carrying an expression cassette encoding a modified cap gene as described herein. The insect cells include, but are not limited to, BTI-TN-5B1-4 (High Five 商標The helper construct may be a cultured cell line such as Sf9 or Sf21 cells (both from Spodoptera frugiperda), Sf9 or TN368 cells with a mammalian glycan profile (GlycoBac), or Sf-RVN cells (MilliporeSigma). The insect cell further comprises an expression cassette containing the AAV rep gene on the same or additional helper construct. The helper construct in the insect cell comprises transcriptional control elements that are associated with and control the expression of the Rep and capsid proteins. These elements include, but are not limited to, constitutive or inducible promoters active in insect cells (e.g., p5 promoter, p10 promoter, p19 promoter, p40 promoter, polh promoter, E1 promoter, and ΔE1 promoter); Kozak sequences; transcription start and stop sites (either modified or unmodified); mRNA splice sites (either modified or unmodified, internal or adjacent to the polypeptide-encoding sequence); and viral, eukaryotic, or prokaryotic RNA elements that control mRNA splicing, nuclear export, localization, stabilization, or translation (e.g., the woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) (Zufferey et al., J Virol. 73(4):2886-92 (1999)), the MMLV / MPMV eukaryotic constitutive transport element (CTE) (Li et al., Nature 443(7108):234-7 (2006)), RNA zipcodes (Jambhekar and DeRisi, RNA 13(5):625-42 (2007)), and omega or other 5'UTR RNA elements to increase translation efficiency. For example, the rep and / or cap genes on the helper constructs may be codon-optimized to increase or decrease the amount of expression of the polypeptide product, remove potential undesired transcription / translation start sites, remove cryptic promoter activity, and / or remove destabilizing elements (e.g., Smith et al., supra).

[0062] rAAV may comprise an AAV vector that contains within its capsid a transgene of interest that may encode a reporter protein for detection using biochemical (luciferase, SEAP) or imaging (GFP, Venus, dTomato) techniques. The transgene may encode a therapeutic protein, including, but not limited to, a chimeric antigen receptor (CAR), C-peptide or insulin, collagen XII, IGF-I, lipoprotein lipase, fibrinogen, prothrombin, factor V, factor VIII, factor IX, factor XI, factor XII, factor XIII, von Willebrand factor, prekallikrein, high molecular weight kininogen (Fitzgerald factor), fibronectin, antithrombin III, heparin coenzyme II, protein C, protein S, protein Z, protein Z-related protease inhibitor, plasminogen, alpha 2-antiplasmin, tissue plasminogen activator, urokinase, plasminogen activator inhibitor 1, plasminogen activator inhibitor 2. The transgene may encode a sequence-specific nuclease (ZFN, TALEN, or Cas9), or a sequence-specific binding protein (ZFP, TALE, or dCas9). The transgene may carry sequences that can be integrated into a specific site in the host genome (donor) by homologous recombination to express a therapeutic protein (as described above). The transgene may also encode an immunogenic protein (e.g., a tumor antigen) for vaccination.

[0063] AAV vector may contain transcriptional regulatory elements (such as promoters and enhancers) that can regulate and express transgenes in human cells.The AAV vector may also contain AAV complete or partial inverted repeats (ITRs) at one or both ends of transgene expression cassette.ITRs are necessary for packaging and integration of virus into host (human) genome.

[0064] Pharmaceutical Compositions of rAAV The rAAV capsid produced by the method of the present application can be formulated as a pharmaceutical composition together with a pharma- ceutically acceptable carrier. Formulations include, but are not limited to, suspension in a liquid or emulsion. Pharmaceutically acceptable carriers include, for example, water, saline, dextrose, glycerol, sucrose, and the like, and combinations thereof. In addition, the composition may contain auxiliary substances, such as wetting or emulsifying agents, pH buffering agents, stabilizers, or other agents that enhance the effectiveness of the rAAV pharmaceutical composition.

[0065] The rAAV pharmaceutical composition can be delivered in vivo to a patient, for example, by systemic administration (e.g., intravenous, intraperitoneal, intramuscular, subcutaneous, intrathecal, or intracranial injection) or by local injection. Alternatively, the rAAV can be delivered ex vivo to cells, for example, cells removed from a patient (e.g., lymphocytes, bone marrow aspirates, or tissue biopsies) or to allogeneic cells (e.g., universal donor cells such as universal CAR T cells), and then the treated cells can be introduced into the patient, usually after selection of cells that have incorporated the rAAV vector.

[0066] Unless otherwise defined herein, scientific and technical terms related to the present invention shall have the meanings commonly understood by those of ordinary skill in the art. Exemplary methods and materials are described below, however methods and materials similar or equivalent to those described herein may also be used in practicing or testing the present invention. In case of conflict, the present specification, including definitions, will control. In general, the nomenclature used in connection with, and techniques of, cell and tissue culture, molecular biology, virology, immunology, microbiology, genetics, analytical chemistry, organic synthetic chemistry, medicinal and pharmaceutical chemistry, and protein and nucleic acid chemistry, and hybridization, described herein, are those well known and commonly used in the art. Enzymatic reactions, and purification techniques are performed according to manufacturer's specifications, as commonly done in the art, or as described herein. Furthermore, unless otherwise required by context, singular terms shall include the plural, and plural terms shall include the singular. Throughout this specification and the embodiments, the terms "has" and "includes", or variations such as "has", "having", "including", or "comprises", are understood to imply the inclusion of a stated integer or group of integers, but not the exclusion of other integers or groups of integers. All publications and other references mentioned herein are incorporated by reference in their entirety as part of this specification. Although a number of documents are cited in this specification, this citation is not an admission that any of these documents form part of the common general knowledge in the art.

[0067] In order that the specification may be further understood, the following examples are set forth, which are for illustrative purposes only and should not be construed as limiting the scope of the invention in any manner. EXAMPLES

[0068] Working Example For the following examples, AAV carrying a transgene containing a factor IX (FIX) expression cassette was produced in HEK293 (mammalian) cells using a triple transfection method, or in Sf9 cells (insect) by infecting naive Sf9 cells with baculovirus-infected insect cells (BIICs) expressing Rep / Cap proteins or donating the AAV genome, and the produced AAV was purified using a discontinuous cesium chloride density gradient (reviewed in Clement and Grieger, supra). After purification, AAV samples were analyzed by Taqman qPCR using transgene-specific primer probes to quantify viral particle (vg) amounts. AAV samples (total 2×10 11 (within vg) was mixed with loading dye and reducing agent, incubated at 95°C for 5 min, and then subjected to 4-12% NuPAGE. 商標 The gels were loaded and electrophoresed at 150V for 80 minutes (ThermoFisher Scientific). The gels were stained with SimplyBlue SafeStain (ThermoFisher Scientific) and destained according to the manufacturer's instructions. The destained gels were scanned using Li-Cor Odyssey CLx (Li-Cor Biosciences) and bands were quantified using Li-Cor Odyssey software. Capsid ratios were calculated relative to VP1 and normalized to the molecular weight of the protein. To test the titer of AAV, 1 × 10 6 The cells were transfected with 1000 cells / cell at an MOI (multiplicity of infection) of 1000 vg / cell and incubated for 5 days. After 5 days, tissue culture supernatants were collected and the expression level of FIX protein was quantified using a FIX ELISA kit (Affinity Biologicals, Inc.). The amount of FIX in each sample was quantified based on a standard curve and compared to 293 AAV (AAV produced in HEK293 cells) to calculate the relative titer.

[0069] Example 1: Elimination of proteolytic cleavage sites in rAAV6 We produced several AAV serotypes in the insect cell system described above and found that an extra protein fragment co-purified on a cesium chloride density gradient with the expected VP1, VP2, and VP3 proteins of serotypes 6 and 8. This extra protein fragment was seen as an extra band slightly above the VP3 band on Coomassie blue stained SDS-PAGE gels, indicating that it was a VP1 / VP2 proteolytic cleavage product. This cleavage product was not observed in serotypes 3B (Rutledge et al., J Virol. 72(1):309-19 (1998)), 5, or 9 produced in insect cells. This cleavage product was also not observed in serotype 6 produced in HEK293 cells. These data suggest that the VP1 / VP2 proteins of certain AAV serotypes, e.g., serotypes 6 and 8, are susceptible to proteolytic cleavage in insect cells.

[0070] We performed Edman degradation sequencing on the extra protein fragment from AAV6 and determined its N-terminal sequence to be EPPAT (SEQ ID NO: 15), which is consistent with a proteolytic cleavage site of the sequence PQPLG↓EPPAT (SEQ ID NO: 2; "↓" indicates the cleavage point). Sequence alignment of the VP1 / VP2 leader sequences of multiple AAV serotypes, immediately prior to the start of the VP3 protein, revealed that this sequence in AAV6 is distinct from serotypes 3B, 5, 8, and 9 (Figures 1A and 1B).

[0071] To test whether altering the sequence surrounding the AAV6 cleavage site could prevent proteolytic cleavage, we grafted a nucleotide sequence from AAV9 corresponding to the 45 amino acids preceding the VP3 start codon into the AAV6 cap gene (cap6) of a baculo-helper vector for AAV6 (FIG. 2D). The wild-type AAV6 and AAV9 sequences in this region (residues 157-201 of SEQ ID NO:1) are shown below, with the amino acid changes compared to wild-type AAV6 after grafting of the AAV9 sequence shown in bold and boxed. [ka]

[0072] This grafting resulted in a total of 10 amino acid changes in the resulting AAV6 VP1 / VP2 unique region. We found that when rAAV6 was produced using baculovirus helper constructs encoding engineered VP1 / VP2 proteins ("AAV9 grafted" or "AAV6 / 9 VP1 / VP2"), proteolytic cleavage was abolished (Figure 2A), the VP1:VP2:VP3 capsid protein ratio was improved compared to standard helper (Figure 2B), and the titer of rAAV produced with the AAV9 grafted helper was nearly two-fold increased compared to standard helper (Figure 2C).

[0073] To define the elements in the grafted AAV9 sequence that contribute to this improvement, we engineered a single L188I mutation into the AAV6 VP1 / VP2 protein. 189 and E 190 Although the proteolytic cleavage between the two bands was reduced (Fig. 3A, arrow), a novel band was observed above it (Fig. 3A, *), suggesting that the single L188I mutation did not solve the proteolytic problem.

[0074] To investigate whether the L188I mutation affects the titer of rAAV6 produced from the helper, the present inventors produced rAAV containing FIX cDNA in 293 cells (293 AAV), or in Sf9 cells using standard helper, AAV9 grafted helper (expressing VP1 / VP2 containing the 10 amino acid changes compared to standard helper), or L188I mutant helper, and compared the expression level of FIX from cells transfected with the produced rAAV. The relative titer was calculated based on 293 AAV. As shown in Figure 3B, AAV produced with AAV9 grafted helper showed significantly higher titer than AAV produced with L188I mutant helper (0.52 vs. 0.31).

[0075] Previous studies have suggested that increasing the VP1 content of the AAV capsid could increase AAV titer, but most studies attempted to achieve this by manipulating the start codon or the Kozak sequence near the start codon (Kohlbrenner et al., supra; Urabe et al., 2006, supra; and Kondratov et al., supra). Our work suggests that preventing proteolytic cleavage of VP1 is another viable approach to improving rAAV titer.

[0076] We generated AAV9 graft variants (variants 1-4) that contained a subset of the 10 amino acid changes in the AAV9 graft (Figure 4A). As shown in Figure 4B, some mutations abolished proteolytic cleavage (variants 1 and 3), while others did not abolish proteolytic cleavage (variant 4) or resulted in additional proteolytic cleavage (variant 2). Furthermore, when comparing the titers of these variants to 293 AAV, standard helper, or AAV9 grafts, none of the variants achieved titers comparable to the AAV9 grafts (Figure 4C). Thus, although variants 1 and 3 were not detectably cleaved in insect cells, these two variants did not achieve improved titers compared to rAAV produced with the native AAV6 helper construct.

[0077] We noted that the AAV9 grafted constructs not only had mutated VP1 / VP2 residues compared to native AAV6 VP1 / VP2, but the engineered constructs also had changes in the AAP open reading frame, resulting in the following point mutations compared to native AAP6: P8Q, H10L, L12Q, Q17P, L21Q, and L22V (Figure 4D). Variants 1-3 only had two mutations, L21Q and L22V. It is likely that one or more of these mutations, e.g., P8Q, H10L, L12Q, and Q17P, contributed to the observed increased titers of rAAV produced with the AAV9 grafted helper constructs.

[0078] Example 2: Improving the yield of recombinant AAV Groups working with the production of AAV5 and AAV8 in baculovirus-insect cell systems have reported that rAAV produced in insect cells exhibits poorer transduction properties compared to rAAV produced in 293 cells (Kohlbrenner et al., supra; and Urabe et al., supra). These groups grafted 134 or 142 amino acids of the N-terminal portion of the VP1 sequence from AAV2 into the AAV5 or AAV8 VP1 leader sequence and demonstrated improved capsid protein ratios and titers for rAAV produced from these chimeric helpers compared to rAAV produced in 293 cells (Id.). Unfortunately, these studies did not identify the elements in the grafted sequences that contribute to this improvement.

[0079] The AAV VP1 leader sequence contains a phospholipase A2-like (PLA2) enzymatic domain that is required for infection (see, e.g., Zadori et al., Developmental cell, 1:291-302 (2001); and Girod et al., J of General Virology, 83:973-8 (2002)). To examine the role of the PLA2 domain in rAAV capsid protein ratios, titers, and yields, we grafted the amino acid sequence from the PLA2 domain of AAV2 VP1 (SEQ ID NO:6) into a baculovirus AAV6 helper, replacing the native AAV6 sequence (SEQ ID NO:5). The wild-type AAV6 and AAV2 sequences in this region (residues 52-97 of SEQ ID NO:1) are shown below, with the five amino acid changes compared to wild-type AAV6 after grafting of the AAV2 sequence shown in bold and boxed: [ka] We observed no significant change in the ratio of capsid proteins between the standard helper, which contains the cap gene sequence entirely derived from AAV6, and the chimeric baculovirus helper containing the AAV2 PLA2 graft ("AAV6 / 2 VP1") (Figure 5A). There was also no significant difference in titer between rAAV6 produced with the standard helper and the chimeric AAV6 / 2 helper (Figure 5B).

[0080] However, we observed a significant and reproducible 4-fold increase in yield of rAAV6 produced with the chimeric AAV6 / 2 helper compared to the standard helper, as shown in Table 1 below, where the average yield was calculated based on the yield of the standard helper and the N number indicates the number of independent production samples. We found that this result was independent of the sequence packaged into the rAAV. [Table 1]

[0081] Unexpectedly, rAAV6 production using the chimeric AAV6 / 2 helper resulted in a significant increase in yield compared to the standard helper.

[0082] We applied a similar strategy to the baculovirus AAV9 helper to produce rAAV9 capsids. We generated AAV9 / 2 VP1 by replacing the AAV9 VP1 PLA2 domain with the PLA2 domain from AAV2 VP1. The wild-type AAV9 VP1 sequence (UniProtKB-Q6JC40(Q6JC40_9VIRU)) is shown below, where the five amino acid residues that differ from the corresponding amino acid residues in the PLA2 domain of AAV9 / 2 VP1 are shown in bold and boxed: [ka]

[0083] The wild-type AAV9 and AAV2 sequences in this region (residues 52-97 of SEQ ID NO:11) are shown below, with the five amino acid changes compared to wild-type AAV9 after AAV2 sequence grafting shown in bold and boxed: [ka]

[0084] We found that the chimeric helper construct expressing AAV9 / 2 VP1 resulted in a 3-fold increase in rAAV9 yield compared to the original AAV9 helper containing the native PLA2 domain (Table 2). [Table 2]

[0085] The present inventors also evaluated the PLA2 enzyme activity of AAV6 / 2 VP1. 登録商標 A phospholipase A2 assay kit (E10217) was used. AAV capsids were dissociated by incubating the virus preparation in 50 mM NaOH for 10 min at room temperature, and samples were neutralized with 100 mM HCL before assaying PLA2 activity according to the manufacturer's instructions. The fluorescence emission of the enzyme substrate at 515 nm was recorded. The data show that rAAV6 produced with AAV6 / 2 helper (helper containing the AAV2 PLA2 domain) has significantly higher PLA2 activity compared to rAAV6 produced with standard helper (with and without dissociation) (Figure 6).

[0086] These results suggest that the PLA2 domain has additional and previously unknown functions in AAV production that directly affect the yield of rAAV production. Helper viruses that provide more enzymatically active PLA2 may be useful for improving rAAV yields in insect cells. [Table 3-1]

Table 3-2

Claims

1. A nucleic acid construct comprising a nucleotide sequence encoding the VP1, VP2, and VP3 proteins of an adeno-associated virus (AAV), wherein the VP1 protein and the VP2 protein comprise two or more mutations at residues 157, 162, 164, 179, 188, 194, 196, 197, 200, and 201 (numbered according to SEQ ID NO:1) compared to wild-type VP1 and VP2 proteins.

2. 2. The nucleic acid construct of claim 1, wherein the VP1 protein and the VP2 protein comprise two or more mutations selected from the first group consisting of S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L.

3. The nucleic acid construct of claim 2, wherein said VP1 protein and said VP2 protein comprise all of the mutations in said first group.

4. 2. The nucleic acid construct of any one of the preceding claims, wherein the VP1 protein further comprises one or more mutations at residues 67, 81, 84, 85, and 92 (numbered according to SEQ ID NO: 1) compared to a wild-type VP1 protein.

5. 5. The nucleic acid construct of claim 4, wherein the VP1 protein comprises one or more mutations selected from the second group consisting of A67E, Q81R, K84D, A85S, and R92K.

6. The nucleic acid construct of claim 5 , wherein the VP1 protein comprises all of the mutations in the second group.

7. 2. The nucleic acid construct of any one of the preceding claims, wherein the VP1 and VP2 proteins are identical to wild-type AAV6 VP1 and VP2 proteins, respectively, except for the mutation.

8. 2. The nucleic acid construct of any one of the preceding claims, wherein the nucleotide sequence also comprises an open reading frame encoding an assembly activating protein (AAP), wherein the AAP comprises one or more mutations at residues 8, 10, 12, 17, 21, and 22 (numbering according to SEQ ID NO: 10) compared to a wild-type AAP protein.

9. 9. The nucleic acid construct of claim 8, wherein the AAP comprises one or more AAP mutations selected from a third group consisting of P8Q, H10L, L12Q, Q17P, L21Q, and L22V.

10. 10. The nucleic acid construct of claim 9, wherein the AAP is identical to the wild-type AAP of AAV6 except for the AAP mutation.

11. 11. The nucleic acid construct of claim 10, wherein the VP1 and VP2 proteins are derived from AAV6 and contain the following mutations compared to the AAV6 VP1 and VP2 proteins: S157A, T162S, Q164A, S179T, L188I, T194A, A196S, A197G, P200S, and T201L.

12. 11. The nucleic acid construct of claim 10, wherein the construct comprises a sequence encoding SEQ ID NO:7, with or without the first amino acid.

13. A nucleic acid construct comprising a nucleotide sequence encoding the VP1, VP2, and VP3 proteins of an adeno-associated virus (AAV), wherein the VP1 protein comprises one or more mutations at residues 81, 84, 85, and 92 (numbered according to SEQ ID NO:1) compared to a wild-type VP1 protein.

14. 14. The nucleic acid construct of claim 13, wherein the VP1 protein comprises one or more mutations selected from the group consisting of Q81R, K84D, A85S, and R92K.

15. 15. The nucleic acid construct of claim 14, wherein said VP1 protein comprises all of the above groups of mutations.

16. 16. The nucleic acid construct of any one of claims 13-15, wherein the VP1 protein further comprises a mutation at residue 67.

17. 17. The nucleic acid construct of claim 16, wherein the VP1 protein comprises an A67E mutation.

18. 18. The nucleic acid construct of any one of claims 13-17, wherein the VP1 protein is identical to the wild-type AAV6 VP1 protein except for the mutation.

19. 20. The nucleic acid construct of claim 18, wherein the VP1 protein comprises the following mutations compared to AAV6 VP1: A67E, Q81R, K84D, A85S, and R92K.

20. A nucleic acid construct comprising a nucleotide sequence encoding an assembly activating protein (AAP), wherein the AAP comprises one or more mutations at residues 8, 10, 12, 17, 21, 22 (numbered according to SEQ ID NO: 10) compared to a wild-type AAP protein.

21. 21. The nucleic acid construct of claim 20, wherein the AAP comprises one or more mutations selected from the group consisting of P8Q, H10L, L12Q, Q17P, L21Q, and L22V.

22. 2. The nucleic acid construct of any one of the preceding claims, wherein the construct further comprises an AAV rep gene.

23. 10. A nucleic acid construct according to any one of the preceding claims, wherein the construct is a baculovirus vector and the nucleotide sequence is operably associated with a promoter active in an insect cell.

24. 24. An insect cell comprising the nucleic acid construct of claim 23.

25. A recombinant AAV virion comprising: (i) a genome having a transgene of interest flanked by a pair of AAV inverted repeats (ITRs); and (ii) a VP1 protein expressed from a nucleic acid construct according to any one of claims 1-23, or a capsid assembled from both VP1 and VP2 proteins.

26. 25. A recombinant AAV virion produced in the insect cell of claim 24.

27. 1. A method for producing recombinant AAV virions, comprising: providing an insect cell of claim 24, wherein said insect cell expresses AAV Rep proteins and contains sequences encoding an AAV vector containing a transgene of interest flanked by a pair of AAV inverted repeats (ITRs); culturing said insect cells under conditions that permit expression of VP1, VP2, VP3 proteins, replication of the AAV vector, and packaging of the AAV vector into capsids assembled from the expressed VP1, VP2, and VP3 proteins; recovering the packaged capsids; A method comprising:

28. 28. The method of claim 27, wherein the genes encoding the AAV Rep, VP1, VP2 and VP3 proteins are present on one or more nucleic acid constructs.

29. 29. The method of claim 27 or 28, wherein the genes encoding the AAV Rep proteins are present on one or more nucleic acid constructs other than the nucleic acid constructs expressing the VP1, VP2, and VP3 proteins.

30. 30. The method of any one of claims 27-29, wherein the genes encoding the AAV VP1, VP2 and VP3 proteins are present on one or more nucleic acid constructs that are not the nucleic acid construct expressing the Rep proteins.

31. The method according to any one of claims 27 to 30, wherein the genes encoding the Rep proteins and the genes encoding the VP1, VP2 and VP3 proteins are integrated into the genome of the insect cell.

32. A recombinant AAV virion produced by the method of any one of claims 27-31.

33. 33. A pharmaceutical composition comprising the recombinant AAV virion of any one of claims 25, 26, and 32, and a pharma- ceutically acceptable carrier.

Images