Novel adeno-associated virus (AAV) vectors, AAV vectors with reduced capsid deamidation and uses thereof

Abstract

The present invention relates to novel adeno-associated virus (AAV) vectors, AAV vectors with reduced capsid deamidation and uses thereof. A recombinant adeno-associated virus (rAAV) vector includes an AAV capsid having a heterogeneous population of vp1 protein, a heterogeneous population of vp2 protein, and a heterogeneous population of vp3 protein. In particular, the present invention relates to a recombinant VP1 amino acid sequence comprising a capsid comprising a modified amino acid as compared to the encoded VP1 amino acid sequence, the capsid comprising a highly deamidated asparagine residue at an asparagine-glycine pair, and further comprising a plurality of other less deamidated asparagine residues, and optionally a glutamine residue. A method of reducing deamidation in an AAV capsid of an rAAV is provided.

Description

[0001] This application is a divisional application of the invention application with the application date of February 27, 2019, the Chinese national application number of 201980028806.2, and the invention name of “Novel adeno-associated virus (AAV) vector, AAV vector with reduced capsid deamidation and its use”.

[0002] Statement Regarding Federally Funded Research

[0003] This invention was made with government support under Grant No. P01HL059407 awarded by the National Heart, Lung, and Blood Institute of the National Institutes of Health. The government has certain rights in this invention. Background Art

[0004] The structure of the adeno-associated virus (AAV) capsid is icosahedral and contains 60 viral protein (VP) monomers (VP1, VP2 and VP3) in a ratio of 1:1:10 (Xie Q et al., Proc Natl Acad Sci USA 2002; 99(16): 10405-10). The entire VP3 protein sequence (519aa) is contained within the C-terminus of both VP1 and VP2, and the shared VP3 sequence is primarily responsible for the overall capsid structure. Due to the structural flexibility of the VP1 / VP2 unique region and the low presentation of VP1 and VP2 monomers relative to VP3 monomers in the assembled capsid, VP3 is the only capsid protein that can be resolved by x-ray crystallography (Nam HJ et al., J Virol 2007; 81(22): 12260-71). VP3 contains nine hypervariable regions (HVRs) that are the main source of sequence variation between AAV serotypes (Govindasamy L et al., Journal of Virology 2013; 87(20): 11187-99). Given its flexibility and location on the capsid surface, HVRs are primarily responsible for interacting with target cells and the immune system (Huang LY et al., Journal of Virology 2016; 90(11): 5219-30; Raupp C et al., Journal of Virology 2012; 86(17): 9396-408). Although the structures of various serotypes are published for structural entries for AAV2, AAVrh.8, AAV6, AAV9, AAV3B, AAV8 and AAV4, respectively (Protein Data Bank (PDB) ID 1LP3, 4RSO, 4V86, 3UX1, 3KIC, 2QA0, 2G8G from the Research Collaboration for Structural Bioinformatics (RCSB) database), there is little information in the literature about modifications on the surface of these capsids. Studies have shown that intracellular phosphorylation of the capsid occurs at specific tyrosine residues (Zhong L et al., Virology 2008; 381(2): 194-202). Despite the presence of putative glycosylation sites in the primary VP3 sequence, glycosylation events have not been identified in AAV2 (Murray S et al., Journal of Virology 2006; 80(12):6171-6; Jin X et al., Hum Gene Ther Methods 2017; 28(5):255-267); other AAV serotypes have not been evaluated for capsid glycosylation.

[0005] AAV gene therapy vectors have undergone less molecular-level scrutiny that typically accompanies the development and manufacture of recombinant protein therapeutics. AAV capsid post-translational modifications (PTMs) are largely unexplored, and thus, little is known about their potential to affect function or about strategies to control PTM levels in manufactured AAV therapeutics.

[0006] Variation in post-translational modifications of non-gene therapy protein therapeutics complicates their development as drugs. Jenkins, N, Murphy, L, and Tyther, R (2008), "Post-translational modifications of recombinant proteins: significance for biopharmaceuticals," Mol Biotechnol 39: 113-118; Houde, D, Peng, Y, Berkowitz, SA, and Engen, JR (2010), "Post-translational modifications differentially affect IgG1 conformation and receptor binding," Mol Cell Proteomics 9: 1716-1728. For example, deamidation of selected amino acids modulates the stability and immune response of anthrax vaccines based on recombinant protective antigens. (Powell BS et al., Proteins 2007; 68(2):45879; Verma A et al., Clin Vaccine Immunol 2016; 23(5):396-402). In some cases, this process is catalyzed by viral or bacterial deamidase to modulate host cell signaling pathways or innate immune responses (Zhao J et al., Journal of Virology 2016; 90(9):4262-8; Zhao J et al., Cell Host Microbe 2016; 20(6):770-84). More commonly, endogenous deamidation is an enzyme-independent, spontaneous process. Although the purpose of spontaneous deamidation has not been fully elucidated, previous studies have suggested that this event serves as a molecular clock to indicate the relative age of proteins and regulate their turnover (Robinson NE and Robinson AB, Proc Natl Acad Sci USA 2001;98(3):944-9).

[0007] Deamidation occurs when the amide group of asparagine or, less commonly, glutamine undergoes nucleophilic attack from an adjacent nitrogen atom and the amide group is lost. This process produces a succinimidyl intermediate (Yang H and Zubarev RA, Electrophoresis 2010; 31(11): 1764-72), which is decomposed by hydrolysis into a mixture of aspartic acid and isoaspartic acid (or glutamic acid and isoglutamic acid) (Catak S et al., J Phys Chem A 2009; 113(6): 1111-20). Studies of short synthetic peptides estimate that this hydrolysis produces a mixture of isoaspartic acid and aspartic acid in a ratio of 3: 1 (Geiger T. and Clarke S, J Biol Chem 1987; 262(2): 785-94).

[0008] There remains a need for compositions comprising AAV-based constructs to deliver heterologous molecules with stable receptor binding and / or stable capsids to avoid neutralizing antibodies and / or maintain purity upon storage. Summary of the invention

[0009] In one embodiment, a composition is provided, which includes a mixed population of recombinant adeno-associated viruses (rAAVs), each of which includes: (a) an AAV capsid, wherein the AAV capsid includes about 60 capsid vp1 proteins, vp2 proteins, and vp3 proteins, wherein the vp1 proteins, the vp2 proteins, and the vp3 proteins are: a heterogeneous population of vp1 proteins, wherein the vp1 proteins are selected from the AAV a heterogeneous population of vp2 proteins produced by a nucleic acid sequence encoding a selected AAVvp2 amino acid sequence; a heterogeneous population of vp3 proteins produced by a nucleic acid sequence encoding a selected AAVvp3 amino acid sequence, wherein the vp1 proteins, the vp2 proteins, and the vp3 proteins contain a subpopulation with amino acid modifications, the amino acid modifications comprising at least two highly deamidated asparagines (N) in the asparagine-glycine pairs in the AAV capsid, and optionally further comprising a subpopulation comprising other deamidated amino acids, wherein the deamidation causes an amino acid change; and (b) a vector genome in the AAV capsid, the vector genome comprising a nucleic acid molecule comprising an AAV inverted terminal repeat sequence and a non-AAV nucleic acid sequence encoding a product, the non-AAV nucleic acid sequence being operably linked to a sequence that directs expression of the product in a host cell. The mixed population of rAAV is produced by a production system using a single type of AAV capsid nucleic acid sequence that encodes the predicted AAVVP1 amino acid sequence of one AAV type. However, the production and manufacturing processes provide a heterogeneous population of capsid proteins as described above. In certain embodiments, the composition is as described in this paragraph, with the proviso that the rAAV is not AAVhu68. In certain embodiments, the composition is as described in this paragraph, with the proviso that the rAAV is not AAV2.

[0010] In certain embodiments, the deamidated asparagine is deamidated to aspartic acid, isoaspartic acid, an interconverted aspartic acid / isoaspartic acid pair, or a combination thereof. In certain embodiments, the capsid further comprises one or more deamidated glutamines, and the one or more deamidated glutamines are deamidated to (α)-glutamic acid, γ-glutamic acid, an interconverted (α)-glutamic acid / γ-glutamic acid pair, or a combination thereof.

[0011] In certain embodiments, a method for reducing deamidation of an AAV capsid is provided. The method comprises generating an AAV capsid from a nucleic acid sequence comprising a modified AAV vp codon, wherein the nucleic acid sequence comprises independently modified glycine codons at one to three asparagine-glycine pairs in an asparagine-glycine pair relative to a reference AAV vp1 sequence, such that the modified codon encodes an amino acid other than glycine.

[0012] In other embodiments, a method for reducing deamidation of an AAV capsid is provided. The method comprises generating an AAV capsid from a nucleic acid sequence comprising a modified AAV vp codon, wherein the nucleic acid sequence comprises at least one independently modified asparagine codon of an asparagine-glycine pair relative to a reference AAV vp1 sequence, such that the modified codon encodes an amino acid other than asparagine.

[0013] A method for increasing the titer, efficacy and / or transduction efficiency of AAV is provided. The method comprises producing an AAV capsid from a nucleic acid sequence containing at least one AAVvp codon, wherein the at least one AAVvp codon is modified to change the asparagine or glycine of at least one asparagine-glycine pair in the capsid to a different amino acid. In certain embodiments, the one or more modified codons are in the v2 and / or vp3 regions. In certain embodiments, the asparagine-glycine pair in the vp1 unique region is retained in the modified rAAV. In certain embodiments, nucleic acid molecule sequences encoding these mutant AAV capsids are provided.

[0014] In certain embodiments, the deamidation site (e.g., an asparagine-glycine pair or Gln) is modified at a position other than: (a) for AAV8 capsids, N57, N263, N385, N514, and / or N540 based on numbering of said AAV8 vp1 of SEQ ID NO:6 (encoded AAV8 vp1) with a starting M; (b) for AAV9 capsids, N57, N329, N452, and / or N512 based on numbering of SEQ ID NO:7 (encoded AAV9 vp1) with a starting M; or (c) for AAVrh10 capsids, N263, N385, and / or N514 based on numbering of SEQ ID NO:112 (encoded AAVrh10 vp1) with a starting M. In certain embodiments, the modified deamidation site is selected from the sites on Table F or Table G. In certain embodiments, the modified deamidation site is selected from the sites on Table F or Table G, excluding the positions in (a)-(c) above. In certain embodiments, the deamidation site (e.g., an asparagine-glycine pair or Gln (Q)) is modified at a position other than the following positions: (a) for AAV1 capsids, N57, N383, N512, and / or N718 based on numbering of the predicted vp1 amino acid sequence with a starting M, based on numbering of SEQ ID NO: 1; (b) for AAV3B capsids, N57, N382, N512, and / or N718 based on numbering of the predicted vp1 amino acid sequence with a starting M, with reference to numbering of SEQ ID NO: 2; (c) for AAV5 capsids, N56, N347, N347, and / or N509 based on numbering of the predicted vp1 amino acid sequence with a starting M, with reference to numbering of SEQ ID NO: 3; (d) for AAV7 capsids, N57, N383, N512, and / or N718 based on numbering of the predicted vp1 amino acid sequence with a starting M, with reference to numbering of SEQ ID NO: 1; (e) for AAVrh32.33 capsid, N57, N264, N292 and / or N318 based on the numbering of the predicted vp1 amino acid sequence with starting M, with reference to the numbering of SEQ ID NO:5; or (f) for AAV4 capsid, N56, N264, N318 and / or N546 based on the numbering of the predicted vp1 amino acid sequence with starting M, with reference to the numbering of SEQ ID NO:111. In certain embodiments, the modified deamidation site is selected from the sites on Table A, Table B, Table C, Table D, Table E, Table F or Table G. In certain embodiments, the modified deamidation site does not include the positions listed in (a)-(f) above.

[0015] In certain embodiments, the methods involve generating a recombinant AAV having a selected mutant AAV8 capsid having the following mutations based on numbering of AAV8 or in another AAV based on alignment of the selected sequence with AAV8: AAV8 G264A / G515A (SEQ ID NO:21); AAV8 G264A / G541A (SEQ ID NO:23); AAV8 G515A / G541A (SEQ ID NO:25) or AAV8 G264A / G515A / G541A (SEQ ID NO:27); AAV8 G264A / G541A / N499Q (SEQ ID NO:115); (c) AAV8 G264A / G541A / N459Q (SEQ ID NO:116); (d) AAV8 G264A / G541A / N305Q / N459Q(SEQ ID NO:117); (e)AAV8 G264A / G541A / N305Q / N499Q(SEQ ID NO:118);AAV8G264A / G541A / N459Q / N499Q(SEQ ID NO:119); or AAV8G264A / G541A / N305Q / N459Q / N499Q (SEQ ID NO:120). In certain embodiments, the method involves producing rAAV having a mutant AAV9 capsid selected from the group consisting of AAV9 G330 / G453A (SEQ ID NO:29), AAV9G330A / G513A (SEQ ID NO:31), AAV9G453A / G513A (SEQ ID NO:33) and / or AAV9 G330 / G453A / G513A (SEQ ID NO:35).

[0016] In certain embodiments, nucleic acid molecule sequences encoding these mutant AAV capsids are provided. In certain embodiments, nucleic acid sequences are provided in the following: for example, SEQ ID NO: 20 (AAV8 G264A / G515A), SEQ ID NO: 22 (AAV8 G264A / G541A), SEQ ID NO: 24 (AAV8 G515A / G541A), or SEQ ID NO: 26 (AAV8 G264A / G515A / G541A). In certain embodiments, nucleic acid sequences are provided in the following: for example, SEQ ID NO: 28 (9G330AG453A), SEQ ID NO: 30 (9G330AG513A), SEQ ID NO: 32 (9G453AG513A), SEQ ID NO: 34 (9G330AG453AG513A). In certain embodiments, other AAVs can be mutated to have such changes in these or corresponding NG pairs based on comparisons with AAV9.

[0017] A composition is provided, comprising a population of rAAVs with increased titer, efficacy or transduction. In certain embodiments, the composition comprises an rAAV having a capsid, which is modified to have reduced total deamidation compared to an rAAV having a deamidation pattern containing a capsid deamidation pattern according to any of the following: Table A (AAV1), Table B (AAV3B), Table C (AAV5), Table D (AAV7), Table E (AAVrh32.33), Table F (AAV8), Table G (AAV9) or Table H (AAVhu37). In certain embodiments, the rAAV is unmodified at the highly deamidated positions identified herein.

[0018] These and other aspects of the invention will become apparent from the following detailed description of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1A-1G : Electrophoretic analysis of AAV8 VP isoforms. ( Figure 1A ) The figure shows the mechanism by which an asparagine residue undergoes nucleophilic attack by an adjacent nitrogen atom to form a succinimidyl intermediate. This intermediate then undergoes hydrolysis to decompose into a mixture of aspartic and isoaspartic acids. The β carbon is labeled as such. The figure was generated in BIOVIADraw2018. ( Figure 1B ) 1 μg of AAV8 vector was run on denaturing one-dimensional SDS-PAGE. ( Figure 1C ) shows the isoelectric point of the carbonic anhydrase pI marker point. ( Figure 1D) 5 μg of AAV8 vector was analyzed by 2D gel electrophoresis and stained with Coomassie blue. Spots 1-20 are carbamylated carbonic anhydrase pI markers. Boxed areas are as follows: a = VP1, b = VP2, c = VP3, d = internal tropomyosin marker (arrow: tropomyosin spot MW = 33 kDa, pI = 5.2). Isoelectric focusing was performed in the pI range of 4-8. ( Figure 1E-1G ) Results of isoelectric focusing in the pI range of 4-8. Analysis of 1e11 GC of wtAAV8 vector by 2D gel electrophoresis ( Figure 1E ) or mutation vector ( Figure 1F and Figure 1G ) and stained with Sypro Ruby. Protein markers: A = VP1; B = VP2; C = VP3, D = hen egg white conalbumin marker, E = turbouclease marker. Isoelectric focusing was performed in the pI range of 6-10. The primary VP1 / 2 / 3 isoform spots are circled and the migration distance of the main spots of the markers is indicated by vertical lines (turbonuclease = dashed line, conalbumin = solid line).

[0020] Figure 2A-2E :Analysis of asparagine and glutamine deamidation in AAV8 capsid protein. Figure 2A-2B ) shows the electrospray ionization (ESI) mass spectrum and the Asn-94 ( Figure 2A ) and Asp-94( Figure 2B Theoretical and observed masses of the 3+ peptide (93-103) of . Figure 2C-Figure 2D ) shows the ESI mass spectrum and the Asn-254 ( Figure 2C ) and Asp-254( Figure 2D Theoretical and observed masses of the 3+ peptide (247-259) of ). The observed mass shifts of Asn-94 and Asn-254 are 0.982 Da and 0.986 Da, respectively, while the theoretical mass shift is 0.984 Da. Figure 2E ) shows the percentage of deamidation at specific asparagine and glutamine residues of interest for AAV8 tryptic peptides purified by different methods. Bars indicating deamidation at asparagine residues with N+1 glycines are marked with cross-hatching. Residues determined to be at least 2% deamidated in at least one of the preparations analyzed are included. Data are expressed as mean ± standard deviation.

[0021] Figure 3A-3E :Structural modeling of AAV8 VP3 monomer and analysis of deamidation sites. Figure 3A) shows the AAV8 VP3 monomer (PDB identifier: 3RA8) in a circle representation. The color of the band indicates the relative degree of flexibility (blue = highest rigidity / normal temperature coefficient, red = highest flexibility / high temperature coefficient). Spheres indicate residues of interest. The expanded view is a ball and stick representation of the residue of interest and its surrounding residues to show the local protein structure (blue = nitrogen, red = oxygen). Underlined residues are residues in the NG motif. Figure 3B-Figure 3E : shows the isoaspartate model of asparagine with N+1 deamidation of glycine. Figure 3C )N263、( Figure 3D )N514 and ( Figure 3E )N540 isoaspartate model compared with ( Figure 3B ) Asparagine model of N410 2FoFc electron density map (1σ level) generated from refinement of the AAV8 crystal structure (PDB ID: 3RA8). The electron density map is shown in magenta grid. The beta carbons are labeled as such. Arrows indicate electron density corresponding to the R group of the residue of interest.

[0022] Figure 4A-4D : Determination of factors affecting AAV8 capsid deamidation. Figure 4A ) were incubated at 70°C for three or seven days, ( Figure 4B ) were exposed to pH 2 or pH 10 for seven days, or prepared using D2O instead of H2O ( Figure 4C ) for mass spectrometry to identify possible sources of deamidation that are not intrinsic to AAV capsid formation. Figure 4D ) using B1 antibody (reacts with denatured capsid) and AAV8 conformation-specific antibody (reacts with intact capsid) as Figure 4A Dot blots of vectors were processed as such to assess capsid structural integrity.

[0023] Figure 5A-Figure 5B : Deamidation frequency in non-AAV proteins. For comparison with the AAV deamidation percentage, two non-AAV recombinant proteins containing the NG motif that may be deamidated are shown (human carbonic anhydrase ( Figure 5A ) and rat phenylalanine-hydroxylase ( Figure 5B ))'s deamidation percentage.

[0024] Figure 6 : Comparison of AAV8 deamidation percentages calculated using data analysis pipelines from two institutions. Shown are the deamidation percentages of AAV8 tryptic peptides evaluated at two different institutions at specific asparagine and glutamine residues of interest.

[0025] Figure 7A-7CDemonstrated functional asparagine substitutions at non-NG sites with high variability between batches. Fig. 7A ) The titers of wtAAV8 vector and mutant vectors were generated by small-scale triple transfection in 293 cells, as measured by quantitative PCR (qPCR). The titers relative to the wtAAV8 control are reported. Figure 8B Transduction efficiency was measured as described in . Titers and transduction efficiencies were normalized to the values ​​of the wtAAV8 control. Figure 7B ) shows representative luciferase images of mice receiving wtAAV8.CB7.ffluc and N499Q capsid mutant vectors at day 14 post-injection. ( Figure 7C ) Luciferase expression from C57BL / 6 mice injected intravenously with wtAAV8 vector or mutant vector (n=3 or 4) was measured by luciferase imaging on day 14 of the study period and reported in total flux units. All data are expressed as mean + standard deviation.

[0026] Fig. 8A and Figure 8B The results of in vitro analysis of the effect of gene deamidation on vector performance are shown. Fig. 8A ) Titers of wtAAV8 vectors and gene deamidation mutant vectors were generated by small-scale triple transfection in 293 cells, as measured by quantitative PCR (qPCR). Titers relative to wtAAV8 control are reported. Highly deamidated NG sites (patterned bars), lowly deamidated sites (white bars), and highly variable sites (black bars) are presented with wtAAV8 and negative controls. ( Figure 8B ) Transduction efficiency of mutant AAV8 vectors producing firefly luciferase reported relative to wtAAV8 control. Transduction efficiency is measured as luminescence units produced per GC added to HUH7 cells and determined by transduction with crude vector at multiple dilutions. Transduction efficiency data are normalized to wild-type (wt) reference. All data are expressed as mean ± standard deviation.

[0027] Figure 9A-9D demonstrated that loss of carrier activity over time was associated with progressive deamidation. Fig.9A ) Time course of vector production (DNAseI-resistant genomic copies, GC) of HEK 293 cells continuously transfected with triples to produce AAV8 vectors packaging a luciferase reporter gene. GC levels were normalized to the maximum observed value. ( Figure 8B ) Huh7 cells were transduced with purified time-course vectors. Multiple dilutions of purified time-course vector samples were used as Figure 8BTransduction efficiency was measured as in (luminescence units per GC added to target cells). Error bars represent standard deviations of at least 10 technical replicates for each sampling time. AAV8 NG sites of vectors collected 1, 2, and 5 days after transfection ( Fig. 9C ) and non-NG sites ( Fig.9D ) deamidation.

[0028] Figures 10A-10D The effect of stabilizing asparagine on the carrier performance is demonstrated. Fig. 10A Shown are the titers of wtAAV8 vector and +1 position mutant vector generated by small-scale triple transfection in 293 cells, as measured by quantitative PCR (qPCR). Titers relative to wtAAV8 control are reported. Fig. 10B The transduction efficiency of mutant AAV8 vectors producing firefly luciferase is reported relative to the wtAAV8 control. Figure 8B Measure transduction efficiency as in . Run a two-sample t test ( * p<0.005) to determine the significance between the transduction efficiency of wtAAV8 and the mutants G264A / G515A and G264A / G541A. Fig. 10C Luciferase expression in liver regions from C57BL / 6 mice injected intravenously with wtAAV8 vector or mutant vectors (n=3 to 5) at day 14 during the study, as measured by luciferase imaging, is shown and reported in total flux units. Fig. 10D Shown are the titers and transduction efficiencies of multi-site AAV8 mutant vectors producing firefly luciferase reported relative to wtAAV8 controls. All data are presented as mean ± standard deviation.

[0029] Figures 11A-11C :Analysis of asparagine and glutamine deamidation in AAV9 capsid protein. Fig.11A ) 1e11 GC of wtAAV9 was analyzed by 2D gel electrophoresis and stained with Sypro Ruby. Protein markers: A = VP1; B = VP2; C = VP3, D = hen egg white conalbumin marker, E = turbouclease marker. Isoelectric focusing was performed in the pI range of 6-10. ( Fig. 11B ) shows the percentage of deamidation at specific asparagine and glutamine residues of interest for AAV9 tryptic peptides purified by different methods. Bars indicating deamidation at asparagine residues with N+1 glycines are marked with cross-hatching. Residues determined to be at least 2% deamidated in at least one of the preparations analyzed are included. Data are expressed as mean ± standard deviation. ( Fig. 11C) The isoaspartate model of N512 is shown in the 2FoFc electron density map generated from unbiased refinement of the AAV9 crystal structure (PDB ID: 3UX1). The arrow indicates the electron density corresponding to the R group of residue N512.

[0030] Figure 11D-11F : Determination of factors affecting AAV9 capsid deamidation. Fig.11D ) Incubate the two AAV9 preparations at 70°C for three or seven days, or ( Fig.11F ) were exposed to pH 2 or pH 10 for seven days to identify possible sources of deamidation that are not intrinsic to AAV capsid formation. Data are presented as mean ± standard deviation. ( Fig.11F ) Using B1 antibody (reacts with denatured capsid) Fig.11D Dot blots of vectors were processed as such to assess capsid structural integrity.

[0031] Fig.11G and Fig.11H In vitro analysis of the effect of gene deamidation of AAV9 on vector performance is presented. Fig.11G ) Titers of wtAAV9 vectors and gene deamidation mutant vectors were generated by small-scale triple transfection in 293 cells, as measured by quantitative PCR (qPCR). Titers relative to wtAAV9 controls are reported. Highly deamidated NG sites (patterned bars), lowly deamidated sites (white bars), and highly variable sites (black bars) are presented with wtAAV8 and negative controls. ( Fig.11H ) Transduction efficiency of mutant AAV9 vectors producing firefly luciferase is reported relative to wtAAV9 control. All data are expressed as mean ± standard deviation.

[0032] Figure 11I-11K AAV9 vector in vitro efficacy over time is shown. Fig.11I ) Time course of vector production (DNAseI-resistant genomic copies, GC) of HEK 293 cells continuously transfected with triples to produce AAV9 vectors packaging the luciferase reporter gene. GC levels were normalized to the maximum observed value. ( Fig.11J )Huh7 cells were transduced with the crude time-course vector. Figure 11K ) shows the transduction efficiency of vector collected 1 day after transfection versus 5 days after transfection for crude vector samples and purified vector samples. Transduction efficiency is expressed as luciferase activity / GC, normalized to the value on day 1.

[0033] Figure 12A-12B : Characterization of PAV9.1 monoclonal antibodies and PAV9.1 epitope-based mutagenesis strategy. Fig. 12A: PAV9.1 recognition of various AAV serotypes based on capture ELISA with native or denatured capsid proteins. Fig. 12B : Alignment of AAVVP1 amino acid sequences (SEQ ID NO: 10-19, from top to bottom); the residues of interest in the epitope of PAV9.1 are in black boxes.

[0034] Figures 13A-13D : Cryo-EM reconstruction of AAV9 in complex with PAV9.1 Fab. Fig.13A : Describes the High-resolution reconstruction of the molecular surface of the AAV9 capsid (magenta) bound to PAV9.1 Fab (blue) at the three-fold axis protrusion. 3,022 particles were framed and reconstructed using electron microscopy using Auto3dEM. Fig. 13B : A cross section of the AAV9-PAV9.1 complex is depicted. Fig. 13C : Pseudo-atomic model of the AAV9-PAV9.1 trimer built into the density as obtained from cryo-reconstruction. VP3 monomers are shown in green, grey and cyan. Spheres indicate bound residues. A single PAV9.1 Fab has been shown with the heavy chain in indigo and the light chain in red. Fig.13D : A two-dimensional “road map” of residues involved in PAV9.1 binding.

[0035] Figure 14A-14E : Effect of epitope mutations on the EC50 of PAV9.1 mAb against AAV9. Capsid capture ELISA against AAV9 was used to analyze and generate binding curves for PAV9.1. Figure 14A-14E The following are shown: 586-590 exchange mutant ( Fig.14A ); 494-498 mutant ( Fig. 14B ); 586-590 point mutant ( Fig. 14C ); AAV9.TQAAA and AAV9.SAQAN single and combined mutants ( Fig.14D ); AAV9.TQAAA and AAV9.SAQAA single and combined mutants ( Fig.14E ). The absorbance was normalized to the maximum absorbance for each capsid. The best fit line and EC50 were determined using the dose response function in Prism.

[0036] Figure 15A-15K : Characterization of the effects of PAV9.1 epitope mutations on in vitro vector transduction and effective PAV9.1 mAb neutralization titers. Fig.15A: Transduction efficiency of PAV9.1 capsid mutants relative to AAV9.WT in HEK293 cells. Significance was determined by using a two-sided one-sample t-test and the transduction percentage of each mutant was compared to the transduction percentage of AAV9.WT (defined as 100%). P values ​​are indicated as follows: p * <0.05, p *** <0.001. Figure 15B-Figure 15K : Neutralization titers of PAV9.1 were determined when HEK293 cells were transduced with: AAV9.WT.CMV.LacZ ( Fig. 15B ); AAV9.AAQAA( Fig. 15C ); AAV9.QQNAA( Fig.15D ); AAV9.SSNTA( Fig.15E ); AAV9.RGNRQ( Fig.15F ); AAV9.RGHRE( Figure 15G ); AAV9.TQAAA( Fig.15H );AAV9.AANNN( Fig.15I ); AAV9.SAQAN( Fig.15J ); or AAV9.SAQAA( Figure 15K Neutralization titers were defined as the dilution prior to the time point at which 50% or more of the transduction level of the vector without mAb was achieved (levels measured in relative light units). All data are reported as mean ± SD.

[0037] Fig.16 : Correlation between PAV9.1 EC50 and neutralization titer of a panel of AAV9 mutants. The fold reduction of PAV9.1 neutralization titer for each mutant relative to the PAV9.1 neutralization titer for AAV9.WT was calculated. The data of the fold increase of PAV9.1 EC50 for each mutant relative to the PAV9.1 EC50 of AAV9.WT on a linear scale (semi-logarithmic plot) were plotted on a logarithmic scale. GraphPadPrism was used to determine the best fit semi-logarithmic line; R 2 =0.8474.

[0038] Figure 17A-17G : In vivo analysis of AAV9 PAV9.1 mutant vectors. C57BL / 6 mice were injected intravenously with 1e11 GC per mouse ( Figure 17A-17C ) or 1e12 GC per mouse ( Figure 17D-17F )AAV9.CMV.LacZ (WT or mutant; n=3). Mice were sacrificed on day 14 and tissues were collected for biodistribution analysis using Taqman qPCR ( Fig.17A and Fig.17D). Values ​​are reported as mean ± SD. Liver ( Fig. 17B and Fig.17E ),heart( Fig. 17C and Fig.17F ) and muscle ( Figure 17G ) for β-gal histochemistry to determine enzyme activity. Representative 10X images are shown; scale bar = 200 μm.

[0039] Figures 18A-18D : Effect of epitope mutations on EC50 of AAV9 injected mouse plasma. Using capsid capture ELISA, 7.5e8 GC / mouse (< 0.05) that received intravenous injection for AAV9.WT or AAV9 PAV9.1 mutant binding were analyzed. Fig.18A ) or 7.5e9 GC / mouse ( Fig.18B )wtAAV9.LSP.hFIX mice at day 56. The absorbance was normalized to the maximum absorbance achieved for each capsid. The best fit line and EC50 were determined using the dose response function in Prism. Each graph corresponds to a single animal. 7.5e8 GC / mouse ( Fig.18C ) or 7.5e9 GC / mouse ( Fig.18D ) to determine the mean for each mutant. A two-sided one-sample t-test was used to determine whether there was a significant difference between the EC50 of plasma for each mutant relative to the EC50 of plasma for AAV9.WT (defined as 1). The Bonferroni correction was applied to control for type 1 error. P values ​​are expressed as follows: ** =p<0.05, ** =p<0.01, *** =p<0.001. EC50 data are reported as mean±SD.

[0040] Figures 19A-19D : Effect of epitope mutations on the EC50 of NHP polyclonal sera against AAV9. Using capsid capture ELISA, ( Fig.19A ) NHP serum; or as AAV9 NAb (+) for AAV9.WT or AAV9 PAV9.1 mutant binding ( Fig.19B ) serum of untreated NHP. Absorbance was normalized to the maximum absorbance achieved for each capsid. The best fit line and EC50 were determined using the dose response function in Prism. Each graph corresponds to a single animal. Vehicle-treated NHP ( Fig.19C ); and the EC50 values ​​of untreated NAb(+)NHP to determine the average value for each mutant ( Fig.19DA two-sided one-sample t-test was used to determine whether the EC50 of plasma for each mutant was significantly different relative to the EC50 of plasma for AAV9.WT (defined as 1). A Bonferroni correction was applied to control for type 1 error. EC50 data are reported as mean ± SD.

[0041] Figure 20A-20B : Effect of epitope mutations on the EC50 of human donor polyclonal sera against AAV9. Fig. 20A : Sera from untreated human donors as AAV9NAbs (+) for AAV9.WT or AAV9 PAV9.1 mutant binding were analyzed using capsid capture ELISA. Best fit lines and EC50s were determined using the dose response function in Prism. Each graph corresponds to a single donor. Fig. 20B : EC50 values ​​of NAb(+) human donor sera were compiled to determine the mean for each mutant. Significance was determined using a two-sided one-sample t-test and the EC50 of plasma for each mutant was compared relative to the EC50 of plasma for AAV9.WT (defined as 1). The Bonferroni correction was applied to control for type 1 error. EC50 data are reported as mean ± SD.

[0042] Figure 21A-21B Shown are AAV8 in vitro titer and transduction data from 6-well plate scale experiments containing N57Q, N263Q, N385Q, N514Q, N540Q, N94Q, and N410Q mutants of AAV8.

[0043] Figure 22A-22B Shown are AAV9 in vitro titer and transduction data from 6-well plate scale experiments containing N57Q, N329Q, N452Q, N270Q, N409Q, N668Q, N94Q, N253Q, N663Q, and N704Q mutants of AAV9.

[0044] Figure 23A-23B In vivo transduction data for AAV8 and AAV9, respectively, in mice tested for mouse liver expression at day 14 are provided (luciferase imaging). Fig.23A AAV8 mutants N57Q, N263Q, and N385Q compared to wild-type AAV8 are shown. Fig. 23B AAV9 mutants N57Q, G58A, G330A compared to wild-type AAV9 are shown.

[0045] Figure 24A-24BThe relative titers (GC) and transduction efficiencies of AAV9 double and triple mutants G330 / G453A, G330A / G513A, G453A / G513A, and G330 / G453A / G513A are shown. Fig.24A The relative titers of the mutants relative to AAV9wt were compared, and Fig. 24B The relative transduction efficiency (luciferase / GC) of the mutants relative to AAV9wt was compared. DETAILED DESCRIPTION

[0046] Provided herein is a recombinant adeno-associated virus (rAAV) and a composition containing the same, having sequence and charge heterogeneity in each of the three capsid protein populations VP1, VP2 and VP3 present in the capsid of the recombinant AAV. Provided herein is a novel rAAV and a method for reducing deamidation and optionally other capsid monomer modifications. Further provided herein is a modified rAAV with reduced modifications, which can be used to provide an rAAV with a capsid that maintains greater stability, efficacy and / or purity. In certain embodiments, the rAAV is not AAVhu68. In certain embodiments, the rAAV is not AAV2.

[0047] In one embodiment, a composition is provided, which includes a mixed population of recombinant adeno-associated viruses (rAAVs), each of which includes: (a) an AAV capsid, wherein the AAV capsid includes about 60 capsid vp1 proteins, vp2 proteins, and vp3 proteins, wherein the vp1 proteins, the vp2 proteins, and the vp3 proteins are: a heterogeneous population of vp1 proteins, wherein the vp1 proteins are selected from the AAV vp1 amino acid sequence; a heterogeneous population of vp2 proteins, the vp2 proteins are produced by a nucleic acid sequence encoding a selected AAVvp2 amino acid sequence; a heterogeneous population of vp3 proteins, the vp3 proteins are produced by a nucleic acid sequence encoding a selected AAVvp3 amino acid sequence, wherein the vp1 proteins, the vp2 proteins, and the vp3 proteins contain a subpopulation with amino acid modifications, the amino acid modifications include at least two highly deamidated asparagines (N) in the asparagine-glycine pairs in the AAV capsid, and optionally further include a subpopulation containing other deamidated amino acids, wherein the deamidation causes an amino acid change; and (b) a vector genome in the AAV capsid, the vector genome comprising a nucleic acid molecule, the nucleic acid molecule comprising an AAV inverted terminal repeat sequence and a non-AAV nucleic acid sequence encoding a product, the non-AAV nucleic acid sequence being operably linked to a sequence that directs the expression of the product in a host cell. In certain embodiments, the composition is as described in this paragraph, with the proviso that the rAAV is not AAVhu68. As used herein, AAVhu68 is as defined in WO 2018 / 160582. The predicted amino acid sequence of AAVhu68 VP1 is reproduced in SEQ ID NO: 114, and the native nucleic acid sequence is provided in SEQ ID NO: 113. In certain embodiments, the composition is as described in this paragraph, with the proviso that the rAAV is not AAV2.

[0048] In certain embodiments, the mixed rAAV population is produced by a production system using a single AAV capsid nucleic acid sequence that encodes the predicted AAVVP1 amino acid sequence of one AAV type. However, the production and manufacturing processes provide a heterogeneous population of capsid proteins as described above.

[0049] In certain embodiments, a recombinant AAV having a mutant AAV8 capsid is provided, wherein the mutant AAV8 capsid has one or more improved properties relative to an unmodified AAV8 capsid. Such improved properties may include, for example, increased titer and / or increased relative transduction efficiency compared to AAV8. In certain embodiments, the mutant may include AAV8G264A / G515A (SEQ ID NO: 21), AAV8G264A / G541A (SEQ ID NO: 23), AAV8G515A / G541A (SEQID NO: 25), or AAV8 G264A / G515A / G541A (SEQ ID NO: 27). In certain embodiments, nucleic acid sequences encoding these mutant AAV8 capsids are provided. In certain embodiments, the nucleic acid sequence is provided in the following: for example, SEQ ID NO: 20 (AAV8 G264A / G515A), SEQ ID NO: 22 (AAV8 G264A / G541A), SEQ ID NO: 24 (AAV8 G515A / G541A), or SEQ ID NO: 26 (AAV8 G264A / G515A / G541A). In certain embodiments, the AAV8 mutant can be N499Q, N459Q, N305Q / N459Q, N305QN499Q, N459Q, N305Q / N459Q, N305q / N499Q, or N205Q, N459Q, or N305Q / N459Q, N499Q. In certain embodiments, these mutations are combined with the G264A / G541A mutation. In certain embodiments, the mutation is AAV8 G264A / G541A / N499Q (SEQ ID NO: 115), AAV8 G264A / G541A / N459Q (SEQ ID NO: 116), AAV8 G264A / G541A / N305Q / N459Q (SEQ ID NO: 117), AAV8 G264A / G541A / N305Q / N499Q (SEQ ID NO: 118), G264A / G541A / N459Q / N499Q (SEQ ID NO: 119) or AAV8 G264A / G541A / N305Q / N459Q / N499Q (SEQ ID NO: 120). In other embodiments, a single mutant can be selected, such as AAV8N263A, AAV8N514A, AAV8N540A. In certain embodiments, other AAVs can be mutated to have changes in these or corresponding NG pairs based on comparison with AAV8. Such AAVs can be clade E AAVs. See, for example, the AAV8 mutants described in Example 2 (SEQ ID NO: 9).

[0050] In certain embodiments, the AAV8 mutants avoid altering the NG pair at positions N57, N94, N263, N305, G386, Q467, N479, and / or N653. In certain embodiments, the other AAVs avoid mutations at the corresponding N positions as determined based on alignment with AAV8 using AAV8 numbering as a reference.

[0051] In certain embodiments, there is provided a recombinant AAV having a mutant AAV9 capsid, which has one or more improved properties relative to an unmodified AAV9 capsid. Such improved properties may include, for example, increased titer and / or increased relative transduction efficiency compared to AAV9. In certain embodiments, the mutant AAV9 capsid may include, for example, AAV9 G330 / G453A (SEQ ID NO: 29), AAV9 G330A / G513A (SEQ ID NO: 31), AAV9 G453A / G513A (SEQ ID NO: 33) and / or AAV9 G330 / G453A / G513A (SEQ ID NO: 35). In certain embodiments, there is provided a nucleic acid sequence encoding these mutant AAV9 capsids. In certain embodiments, nucleic acid sequences are provided in the following: for example, SEQ ID NO: 28 (9G330AG453A), SEQ ID NO: 30 (9G330AG513A), SEQ ID NO: 32 (9G453AG513A), SEQ ID NO: 34 (9G330AG453AG513A). In certain embodiments, other AAVs can be mutated to have such changes in these or corresponding NG pairs based on alignment with AAV9. Such AAVs can be clade F AAVs.

[0052] In certain embodiments, rAAVs having mutant AAV capsids of clade A, clade B, clade C, or clade D can be engineered to have amino acid modifications of NG pairs corresponding to the NG pairs identified above for clades E and F. In certain embodiments, clade A (e.g., AAV) mutants can include mutations at positions N303, N497, or N303 / N497 with reference to the numbering of SEQ ID NO: 1 (AAV1). In certain embodiments, the mutant is N497Q. In certain embodiments, AAV3B mutants can include mutations at positions N302, N497, or N302 / N497 with reference to the numbering of SEQ ID NO: 2. In certain embodiments, the mutant is N497Q. In certain embodiments, AAV5 mutants can include mutations at positions N302, N497, or N302 / N497 with reference to the numbering of SEQ ID NO: 3. In certain embodiments, the mutant is N497Q.

[0053] Without wishing to be bound by theory, mass spectrometry revealed deamidation of asparagine at many positions on the capsid, which has not been previously described for AAV as an explanation for the presence of multiple VP isoforms. Additionally, the distribution and extent of deamidation were consistent in multiple vector purification methods, indicating that this phenomenon occurs independently of vector processing. The functional significance of these deamidations was explored by mutating some asparagines to aspartic acid, respectively. A subset of these mutations not only affects the efficiency of particle assembly, but also affects the ability of the vector to transduce target cells both in vitro and in vivo. Modeling these deamidated residues from scratch into the AAV8 structure also revealed structural evidence for the presence of these deamidation events, and provided a computational explanation for why the AAV8 capsid can tolerate these changes in amino acid identity and properties. Almost identical deamidation findings can be seen using AAV9 and various other AAVs. Therefore, rAAV is characterized by previously unknown AAV capsid structural heterogeneity.

[0054] In the studies reported here, 17 residues were found to be affected by extensive asparagine and occasional glutamine deamidation. The factors controlling AAV8 deamidation (primarily primary sequence and 3D structural constraints) are likely conserved throughout the AAV germline, as all serotypes analyzed to date exhibit strikingly similar modification patterns. Deamidation is therefore a potentially critical factor in the development of all future AAV therapeutics.

[0055] With this discovery, there is an impetus to explore the functional impact of AAV deamidation. The multimeric nature of AAV vector capsids, the extent and number of modified capsid residues, and the resulting mosaic diversity in vector particle composition present some special challenges for this analysis. The experimental repertoire that might adequately parameterize the impact of post-translational modifications (PTMs) in a simpler protein environment is not directly applicable to AAV capsid analysis. For example, it would not be possible to purify or even enrich a preparation of a vector species for a specific deamidation to directly and individually test its function.

[0056] Genetic substitution of aspartate is one approach that attempts to force an approximation of modification at a given site. In addition to the previously noted differences between the distribution of position-specific modifications on capsid assembly with endogenous (mosaic) deamidation versus genetic (intact) deamidation, the data also point to additional considerations for interpreting this data. For example, a >50-fold loss of transduction relative to wtAAV8 was observed for the N263D mutant ( Figure 8B). This is surprising, considering that the aspartate content at this position would change negligibly after gene conversion; N263 is 99% deamidated in wtAAV8. One explanation for this discrepancy is that the molecules encoded by the gene for aspartate and asparagine deamidation products are different (L-aspartate versus a putative 3:1 mixture of L / D-isoaspartate:L / D-aspartate). Therefore, gene approximation may be insufficient at some positions. Another residue, the highly conserved N57, was also intolerant to substitution with aspartate, despite being on average 80% and 97% deamidated in AAV8 and AAV9, respectively ( Figure 8B and 11). Here, the remaining intact amide could buffer the activity of the wt preparation via mosaicism, although the possibility of crosstalk with other asparagines confounding the analysis of N57 was also examined; while the position 57 amide was preserved by mutagenesis, adjacent N66 became significantly deamidated (N57Q, G58A, and G58S for AAV8; N57Q and G58A for AAV9; data not shown). This was the only case of significant crosstalk detected based on mass spectrometry analysis of the mutants, but it highlights another complication in interpreting loss-of-function mutagenesis data.

[0057] Given these caveats, evidence for the impact of deamidation was developed through time-course and gain-of-function mutagenesis experiments. The data are consistent with a role for a subset of NG sites in loss of function associated with very early time-point deamidation. To our knowledge, this phenomenon has not been reported previously. In fact, the specific experimental procedures used to identify this decay were informed by the novel observation of NG deamidation of very short half-life vectors; given the observed rate of spontaneous deamidation, storing early samples in the refrigerator for even a day would likely reduce their distinction from later time-point samples. Storage stability experiments comparing the activity of vector preparations within days or weeks of processing are routinely performed in laboratories and other manufacturing groups, but these comparisons are almost always performed using vector material that is at least 7 days old, when most or all of the activity decay (and NG site deamidation) is complete. The data highlight the opportunity for process intervention or N-stabilization mutagenesis methods to produce improved capsids. From a broader perspective, it is also of interest to consider the role of the "deamidation clock" in the natural ecology of AAV, in which case this phenomenon would likely favor the most recently translated viral particles from infected cells for the next round of infection.

[0058] The mechanistic basis for NG deamidation-induced loss of function has not been explored, although some prominent possibilities exist. All NG motifs in AAV8 and AAV9 VP3 are found in surface HVR loops. In AAV8, NG 514 and 540, known to play an important role in transduction due to interactions with cellular receptors, are localized in a region near the 3-fold axis. Although the AAV8 receptor binding site has not been fully interrogated, the LamR receptor has been implicated in AAV8 transduction. These studies identified aa491-557 as important for these interactions. Receptor binding for AAV9 is better characterized than for AAV8, as functional interrogation of the capsid identified residues in the galactose binding domain of AAV9. Of these residues, a single asparagine N515 was found to be deamidated at a low level (3%), while two other asparagines in this domain, N272 and N470, were found not to be deamidated. Therefore, although deamidation may affect galactose binding, it is likely to be only to a small extent.

[0059] In summary, AAV vector deamidation was identified as potentially affecting transduction efficiency and strategies to stabilize amides and improve vector performance were demonstrated. Key future goals are to extend these findings to appropriate animal model systems and begin to consider the impact of deamidation and the performance of stabilized variants in more complex functional contexts. Tissue tropism and capsid interactions with the immune system may be affected and must be carefully evaluated. Because these complex effects will likely be difficult to conclusively determine for every deamidated residue in the capsid, it would be prudent to target a limited number of residues with high batch-to-batch variability in deamidation for stabilization via mutagenesis, as was successfully demonstrated for variable AAV8 asparagines 459 and 499. Additionally, deamidation analysis of vector preparations using mass spectrometry workflows may prove beneficial in achieving functional consistency across manufactured AAV gene therapy batches.

[0060] "Recombinant AAV" or "rAAV" is a DNAse-resistant virus particle containing two elements, an AAV capsid and a vector genome containing at least a non-AAV coding sequence packaged in an AAV capsid. Unless otherwise indicated, this term can be used interchangeably with the phrase "rAAV vector". rAAV is a "replication-defective virus" or "viral vector" because it lacks any functional AAVrep gene or functional AAVcap gene and cannot produce offspring. In certain embodiments, the only AAV sequence is the AAV inverted terminal repeat (ITR), which is usually located at the 5' and 3' extreme ends of the vector genome to allow genes and regulatory sequences located between the ITRs to be packaged in the AAV capsid.

[0061] As used herein, "vector genome" refers to the nucleic acid sequence packaged inside the rAAV capsid that forms the virus particle. This nucleic acid sequence contains AAV reverse terminal repeats (ITR). In the examples herein, the vector genome contains at least AAV 5'ITR, one or more coding sequences and AAV 3'ITR from 5' to 3'. It is possible to select the ITR from AAV2, the AAV that is different from the capsid source or the AAV except the full-length ITR. In certain embodiments, ITR is from the AAV that provides the same AAV source of rep function during production or trans-supplementation of AAV. Further, other ITRs can be used. Further, the vector genome contains the regulatory sequence that guides the expression of the gene product. The suitable components of the vector genome are discussed in more detail herein.

[0062] rAAV is composed of AAV capsid and vector genome. AAV capsid is an assembly of a heterogeneous population of vp1, a heterogeneous population of vp2, and a heterogeneous population of vp3 proteins. As used herein, when used to refer to vp capsid proteins, the term "heterologous" or any grammatical variation thereof refers to a population composed of non-identical elements, such as vp1, vp2 or vp3 monomers (proteins) containing different modified amino acid sequences.

[0063] As used herein, the term "heterogeneous" used in conjunction with vp1, vp2, and vp3 proteins (alternatively referred to as isoforms) refers to differences in the amino acid sequences of the vp1, vp2, and vp3 proteins within the capsid. The AAV capsid contains subpopulations within the vp1 protein, within the vp2 protein, and within the vp3 protein that have modifications from predicted amino acid residues. These subpopulations contain at least some deamidated asparagine (N or Asn) residues. For example, certain subpopulations include at least one, two, three, or four highly deamidated asparagine (N) positions in asparagine-glycine pairs, and optionally further include other deamidated amino acids, wherein the deamidation causes amino acid changes and other optional modifications.

[0064] As used herein, unless otherwise specified, a "subpopulation" of vp proteins refers to a group of vp proteins that have at least one defined common property and consist of at least one group member to less than all members of a reference group. For example, a "subpopulation" of vp1 proteins is at least one (1) vp1 protein and less than all vp1 proteins in an assembled AAV capsid, unless otherwise specified. A "subpopulation" of vp3 proteins can be one (1) vp3 protein to less than all vp3 proteins in an assembled AAV capsid, unless otherwise specified. For example, vp1 proteins can be a subpopulation of vp proteins; vp2 proteins can be different individual subpopulations of vp proteins, and vp3 is yet another subpopulation of vp proteins in an assembled AAV capsid. In another example, vp1, vp2, and vp3 proteins can contain subpopulations with different modifications, such as at least one, two, three, or four highly deamidated asparagines, such as at asparagine-glycine pairs.

[0065] Unless otherwise indicated, highly deamidated refers to a reference amino acid position that is at least 45% deamidated, at least 50% deamidated, at least 60% deamidated, at least 65% deamidated, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 99%, or up to about 100% deamidated compared to the predicted amino acid sequence at the reference amino acid position (e.g., at amino acid 57 based on numbering of SEQ ID NO: 1 [AAV1], 2 [AAV3B], 4 [AAV7], 5 [AAVrh32.33], 6 [AAV8], 7 [AAV9], 9 [AAV8 triple], or 111 [AAVhu37] or at amino acid 58 based on numbering of SEQ ID NO: 1 [AAV1], 2 [AAV3B], 4 [AAV7], 5 [AAVrh32.33], 6 [AAV8], 7 [AAV9], 9 [AAV8 triple], or 111 [AAVhu37]). At least 80% of the asparagine at amino acid 56 of numbering NO:3 [AAV5] may be deamidated based on total vpl protein or may be deamidated based on total vpl, vp2 and vp3 proteins). Such percentages may be determined using 2D gels, mass spectrometry techniques or other suitable techniques.

[0066] As used herein, a "deamidated" AAV is an AAV in which one or more of the amino acid residues has been derivatized to a residue different from that encoded in the corresponding nucleic acid sequence.

[0067] Without wishing to be bound by theory, deamidation of at least highly deamidated residues in the vp protein in the AAV capsid is believed to be primarily non-enzymatic in nature, caused by functional groups within the capsid protein that deamidate selected asparagine and, to a lesser extent, glutamine residues. The efficient capsid assembly of the majority of deamidated vp1 proteins suggests that these events occur after capsid assembly, or that deamidation in individual monomers (vp1, vp2, or vp3) is structurally well tolerated and does not affect assembly kinetics to a large extent. Extensive deamidation in the VP1 unique (VP1-u) region (approximately aa 1-137) is generally believed to be localized internally prior to cell entry, suggesting that VP deamidation may occur prior to capsid assembly. Deamidation of N may occur by nucleophilic attack of the backbone nitrogen atom of its C-terminal residue on the side chain amide carbon atom of Asn. It is believed that an intermediate closed ring succinimide residue is formed. The succinimide residue is then rapidly hydrolyzed to produce the final product aspartic acid (Asp) or isoaspartic acid (IsoAsp). Therefore, in certain embodiments, the deamidation of asparagine (N or Asn) produces Asp or IsoAsp, which can be mutually converted through the succinimide intermediate, such as shown below.

[0068]

[0069] As provided herein, each deamidated N in VP1, VP2 or VP3 can independently be aspartic acid (Asp), isoaspartic acid (isoAsp), aspartic acid and / or an interconversion blend of Asp and isoAsp or a combination thereof. Any suitable ratio of α- and isoaspartic acid can be present. For example, in certain embodiments, the ratio can be 10:1 to 1:10 aspartic acid:isoaspartic acid, about 50:50 aspartic acid:isoaspartic acid, or about 1:3 aspartic acid:isoaspartic acid, or another selected ratio.

[0070] In certain embodiments, one or more glutamines (Q) can be derivatized (deamidated) to glutamic acid (Glu), i.e., α-glutamic acid, γ-glutamic acid (Glu), or a blend of α- and γ-glutamic acid, which can be interconverted through a common glutarimide intermediate. Any suitable ratio of α- and γ-glutamic acid can be present. For example, in certain embodiments, the ratio can be 10:1 to 1:10 α:γ, about 50:50 α:γ, or about 1:3 α:γ, or another selected ratio.

[0071]

[0072] Thus, the rAAV capsids of the rAAV comprising vp1, vp2 and / or vp3 proteins have a subpopulation of deamidated amino acids, including at least one subpopulation comprising at least one highly deamidated asparagine. Additionally, other modifications may include isomerization, particularly at selected aspartic acid (D or Asp) residue positions. In still other embodiments, the modification may include amidation at the Asp position.

[0073] In certain embodiments, the AAV capsid contains a subpopulation of vp1, vp2, and vp3 having at least 4 to at least about 25 deamidated amino acid residue positions, at least 1% to 10% of which are deamidated compared to the encoded amino acid sequence of the vp protein. The majority of these can be N residues. However, Q residues can also be deamidated.

[0074] In certain embodiments, rAAV has an AAV capsid containing vp1, vp2 and vp3 proteins, the proteins having a subpopulation of a combination of two, three, four or more deamidated residues at the positions listed in the table provided in the examples, and incorporated herein by reference. Deamidation in rAAV can be determined using 2D gel electrophoresis and / or mass spectrometry and / or protein modeling techniques. Online chromatography can be performed using an Acclaim PepMap column and a Thermo UltiMate3000RSLC system (Thermo Fisher Scientific) connected to a QExactive HF and NanoFlex source (Thermo Fisher Scientific). MS data are obtained using a data-dependent top 20 method for QExactive HF, which dynamically selects the most abundant precursor ions that have not yet been sequenced from a survey scan (200-2000m / z). Sequencing is performed by high-energy collision dissociation fragments, where the target value determined by predictive automatic gain control is 1e5 ions, and precursor separation is performed with a window of 4m / z. Survey scans were acquired at a resolution of 120,000 at m / z200. The resolution of the HCD spectrum can be set to 30,000 at m / z 200, with a maximum ion injection time of 50 milliseconds and a normalized collision energy of 30. The S-lens RF level can be set to 50 to achieve optimal transmission in the m / z region occupied by the digested peptides. Precursor ions with a single, unassigned, or six and higher charge states can be excluded from the fragment selection. BioPharmaFinder 1.0 software (Thermo Fisher Scientific) can be used to analyze the acquired data. For peptide mapping, a single entry protein FASTA database was used to search, with carbamidomethylation set as a fixed modification; and oxidation, deamidation, and phosphorylation set as variable modifications, 10ppm mass accuracy, high protease specificity, and a confidence level of 0.8 MS / MS spectra. Examples of suitable proteases can include, for example, trypsin or chymotrypsin. The mass spectrometry identification of deamidated peptides is relatively simple, because deamidation adds +0.984Da (-OH group and-NH2 group's mass difference) to the mass of the complete molecule. The deamidation percentage of a specific peptide is determined by dividing the mass area of ​​the deamidated peptide by the sum of the areas of the deamidated and native peptides. Considering the number of possible deamidation sites, isobaric species deamidated at different sites can co-migrate in a single peak. Therefore, fragment ions derived from peptides with multiple potential deamidation sites can be used to locate or distinguish multiple deamidation sites. In these cases, the relative intensity in the observed isotope pattern can be used to specifically determine the relative abundance of different deamidated peptide isomers.This method assumes that the fragmentation efficiency of all isomeric species is the same and is independent at the deamidation site. It will be appreciated by those skilled in the art that a variety of variations of these illustrative methods can be used. For example, a suitable mass spectrometer can include, for example, a quadrupole time-of-flight mass spectrometer (QTOF), such as Waters Xevo or Agilent 6530, or an Orbitrap instrument, such as Orbitrap Fusion or Orbitrap Velos (Thermo Fisher Scientific). Suitable liquid chromatography systems include, for example, AcquityUPLC systems or Agilent systems (1100 or 1200 series) from Waters (Waters). Suitable data analysis software can include, for example, MassLynx (Waters), Pinpoint and Petfinder (Thermo Fisher Scientific), Mascot (Matrix Science), Peaks DB (Bioinformatics Solutions). Still other techniques can be described in, for example, X. Jin et al., Human Gene Therapy Methods, Vol. 28, No. 5, pp. 255-267, published online on June 16, 2017.

[0075] In addition to deamidation, other modifications that do not result in the conversion of one amino acid into a different amino acid residue may occur. Such modifications may include acetylated residues, isomerization, phosphorylation or oxidation.

[0076] Regulation of deamidation: In certain embodiments, AAV is modified to change the glycine in the asparagine-glycine pair to reduce deamidation. In other embodiments, asparagine is changed to a different amino acid, such as glutamine that is deamidated at a slower rate; or changed to an amino acid lacking an amide group (e.g., glutamine and asparagine containing an amide group); and / or changed to an amino acid lacking an amide group (e.g., lysine, arginine, and histidine containing an amine group). As used herein, amino acids lacking amide or amine side groups refer to, for example, glycine, alanine, valine, leucine, isoleucine, serine, threonine, cystine, phenylalanine, tyrosine, or tryptophan and / or proline. Modifications such as described can be in one, two, or three asparagine-glycine pairs in the asparagine-glycine pairs present in the encoded AAV amino acid sequence. In certain embodiments, such modifications are not performed in all four asparagine-glycine pairs. A method for reducing the deamidation of AAV and / or engineered AAV variants with a lower deamidation rate is provided herein. Additionally or alternatively, one or more other amide amino acids may be changed to non-amide amino acids to reduce the deamidation of AAV. In certain embodiments, the mutant AAV capsid described herein contains a mutation in an asparagine-glycine pair, such that glycine becomes alanine or serine. The mutant AAV capsid may contain one, two or three mutants, wherein the reference AAV naturally contains four NG pairs. In certain embodiments, the AAV capsid may contain one, two, three or four such mutants, wherein the reference AAV naturally contains five NG pairs. In certain embodiments, the mutant AAV capsid contains only a single mutation in the NG pair. In certain embodiments, the mutant AAV capsid contains mutations in two different NG pairs. In certain embodiments, the mutant AAV capsid contains mutations in two different NG pairs located in a structurally separated position in the AAV capsid. In certain embodiments, the mutation is not in the VP1 unique region. In certain embodiments, one of the mutations is not in the VP1 unique region. Optionally, the mutant AAV capsid contains no modification in the NG pair, but contains mutations to minimize or eliminate deamidation in one or more asparagine or glutamine located outside of the NG pair.

[0077] In certain embodiments, a method for increasing the efficacy of an rAAV vector is provided, the method comprising engineering an AAV capsid, which eliminates one or more NGs in the NG in the wild-type AAV capsid. In certain embodiments, the coding sequence of the "G" of "NG" is engineered to encode another amino acid. In certain examples below, "S" or "A" is substituted. However, other suitable amino acid coding sequences can be selected. See, for example, the following table, in which the coding sequence for at least one of the following positions is modified based on the numbering of AAV8: N57+1, N263+1, N385+1, N514+1, N540+1. In certain embodiments, the AAV8 mutant avoids changing the NG pairs at positions N57, N94, N263, N305, Q467, N479 and / or N653. In certain embodiments, other AAVs avoid mutations at corresponding N positions as determined based on comparisons with AAV8 using AAV8 numbering as a reference.

[0078] These amino acid modifications can be carried out by conventional genetic engineering techniques.For example, a nucleic acid sequence containing a modified AAVvp codon can be produced, wherein one to three codons in the codon encoding glycine in the modified asparagine-glycine pair are modified to encode amino acids other than glycine. In certain embodiments, a nucleic acid sequence containing a modified asparagine codon can be engineered at one to three asparagine-glycine pairs in the asparagine-glycine pair so that the modified codon encodes amino acids other than asparagine. Each modified codon can encode different amino acids. Alternatively, one or more codons in the changed codon can encode the same amino acid. In certain embodiments, these modified AAV nucleic acid sequences can be used to produce mutant rAAVs with a capsid having a lower degree of deamidation than the natural capsid. Such mutant rAAVs can have reduced immunogenicity and / or increase stability during storage, particularly stability when stored in the form of a suspension.

[0079] Also provided herein is a nucleotide sequence encoding the AAV capsid with reduced deamidation. Designing the nucleotide sequence encoding this AAV capsid is within the technical scope of the art, including DNA (genome or cDNA) or RNA (e.g., mRNA). Such nucleotide sequences that can be codon-optimized to be expressed in a selected system (i.e., cell type) can be designed by various methods. This optimization can be performed using online available methods (e.g., GeneArt), disclosed methods, or companies (e.g., DNA2.0) (Menlo Park, California) that provide codon optimization services. For example, a codon optimization method is described in U.S. International Patent Publication No. WO 2015 / 012924, which is incorporated herein by reference in its entirety. Also referring to, for example, U.S. Patent Publication No. 2014 / 0032186 and U.S. Patent Publication No. 2006 / 0136184. Suitable is, the full length of the open reading frame (ORF) of the modified product. However, in some embodiments, only the fragment of ORF can be changed. By using a method in these methods, frequency can be applied to any given polypeptide sequence, and produce the nucleic acid fragment of the coding region through codon optimization that polypeptide is encoded.Many options can be used for codon to be actually changed or can be used for synthesizing the codon optimized coding region of design as described herein.Such change or synthesis can be carried out using standard and conventional molecular biology operations known to those of ordinary skill in the art.In one method, a series of complementary oligonucleotide pairs of the length of 80-90 nucleotide and the sequence of hope are synthesized by standard method.These oligonucleotide pairs are synthesized so that they form the double-stranded fragment of 80-90 base pair when annealing, and these double-stranded fragments contain sticky ends, for example, each oligonucleotide in the pair is synthesized to extend beyond 3,4,5,6,7,8,9,10 or more bases in the region complementary to another oligonucleotide in the pair.Every single-stranded end of oligonucleotide is designed to anneal with the single-stranded end of another pair of oligonucleotide. These oligonucleotide pairs are allowed to anneal, and then about five to six of these double-stranded fragments are allowed to anneal together via the sticky single-stranded ends, and then they are ligated together and cloned into a standard bacterial cloning vector, such as the PCR-based vector available from Invitrogen Corporation, Carlsbad, Calif. Vector. Then the construct is sequenced by standard methods. Several constructs in these constructs consisting of 5 to 6 fragments of 80 to 90 base pairs (i.e., fragments of about 500 base pairs) linked together are prepared so that the entire desired sequence is represented by a series of plasmid constructs. The inserts of these plasmids are then cut with appropriate restriction enzymes and linked together to form the final construct. The final construct is then cloned into a standard bacterial cloning vector and sequenced. Other methods will be immediately clear to the technician. In addition, gene synthesis is easily commercially available.

[0080] In certain embodiments, AAV capsids are provided having a heterogeneous population of AAV capsid isotypes (i.e., VP1, VP2, VP3) containing multiple highly deamidated "NG" positions. In certain embodiments, the highly deamidated positions are in the positions identified below with reference to the predicted full-length VP1 amino acid sequence. In other embodiments, the capsid gene is modified such that the reference "NG" is ablated and the mutant "NG" is engineered into another position.

[0081] In certain embodiments, AAV1 is characterized by a capsid composition of a heterogeneous population of VP isotypes that are deamidated as defined in the table below, based on the total amount of VP protein in the capsid as determined using mass spectrometry.

[0082] In certain embodiments, the AAV capsid is modified at one or more of the following positions, within the ranges provided below as determined using mass spectrometry. Suitable modifications include those described in the paragraph above labeled as deamidation regulation, which are incorporated herein.

[0083] In certain embodiments, one or more of the following positions or the glycine after the N is modified as described herein. In certain embodiments, AAV1 mutants are constructed in which the glycine after the N at positions 57, 383, 512, and / or 718 is preserved (i.e., remains unmodified). In certain embodiments, the NG at the four positions identified in the previous sentence is preserved with the native sequence. Residue numbering is based on the published AAV1 VP1 and is reproduced in SEQ ID NO: 1.

[0084] In certain embodiments, artificial NG is introduced into a location other than one of the locations identified below.

[0085] Residue numbering is based on the published AAV1 sequence and is reproduced in SEQ ID NO: 1.

[0086]

[0087] In certain embodiments, AAV3B is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the following table, based on the total amount of VP protein in the capsid as determined using mass spectrometry. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the range provided below as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraph above, which are incorporated herein. In certain embodiments, one or more of the following positions or the glycine after N are modified as described herein. In certain embodiments, AAV3 mutants are constructed in which the glycine after N at positions 57, 383, 512 and / or 718 is preserved (i.e., remains unmodified). In certain embodiments, the NG at the four positions identified in the previous sentence is preserved with the native sequence. The residue numbering is based on the published AAV3B VP1, reproduced in SEQ ID NO:2. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, the capsid is modified to remove an "N" or "Q" at a position other than an "NG" pair. Residue numbering is based on the published AAV3B sequence, reproduced in SEQ ID NO:2.

[0088]

[0089] In certain embodiments, AAV5 is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the table below, based on the total amount of VP protein in the capsid as determined using mass spectrometry. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the ranges provided below as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraphs above, which are incorporated herein. In certain embodiments, one or more of the following positions or a glycine after N is modified as described herein. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, the capsid is modified to reduce "N" or "Q" at positions other than the "NG" pair. The residue numbering is based on the published AAV5 sequence and is reproduced in SEQ ID NO:3.

[0090]

[0091] In certain embodiments, AAV7 is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the table below, based on the total amount of VP protein in the capsid as determined using mass spectrometry. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the ranges provided below as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraphs above, which are incorporated herein. In certain embodiments, one or more of the following positions or a glycine after N is modified as described herein. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, the capsid is modified to reduce "N" or "Q" at positions other than the "NG" pair. Residue numbering is based on the published AAV7 sequence and is reproduced in SEQ ID NO:4.

[0092]

[0093]

[0094] In certain embodiments, AAVrh32.33 is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the following table, based on the total amount of VP protein in the capsid as determined using mass spectrometry. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the range provided below as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraphs above, which are incorporated herein. In certain embodiments, one or more of the following positions or glycine after N are modified as described herein. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, the capsid is modified to reduce "N" or "Q" at positions other than the "NG" pair. Residue numbering is based on the published AAVrh32.33 sequence and is reproduced in SEQ ID NO:5.

[0095]

[0096] In certain embodiments, AAV8 is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the following table based on the total amount of VP protein in the capsid as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraph above, which are incorporated herein. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the ranges provided below as determined using mass spectrometry. In certain embodiments, one or more of the following positions or a glycine after N is modified as described herein. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, one or more of the following positions or a glycine after N is modified as described herein. For example, in certain embodiments, G can be modified to S or A, for example, at positions 58, 67, 95, 216, 264, 386, 411, 460, 500, 515, or 541. When NG57 / 58 is changed to NS57 / 58 or NA57 / 58, it is observed that deamidation is significantly reduced. However, in certain embodiments, when NG is changed to NS or NA, it is observed that deamidation increases. In certain embodiments, the N of the NG pair is modified to Q while retaining G. In certain embodiments, both amino acids of the NG pair are modified. In certain embodiments, N385Q causes the deamidation in the position to be significantly reduced. In certain embodiments, N499Q causes the deamidation in the position to be significantly increased. In certain embodiments, NG mutations are performed at the pair positioned at N263 (e.g., to N263A). In certain embodiments, NG mutations are performed at the pair positioned at N514 (e.g., to N514A). In certain embodiments, NG mutations are performed at the pair positioned at N540 (e.g., N540A). In certain embodiments, AAV mutants containing multiple mutations and at least one mutation in the mutations at these positions are engineered. In certain embodiments, mutations are not performed at position N57. In certain embodiments, no mutation is made at position N94. In certain embodiments, no mutation is made at position N305. In certain embodiments, no mutation is made at position G386. In certain embodiments, no mutation is made at position Q467. In certain embodiments, no mutation is made at position N479. In certain embodiments, no mutation is made at position N653. In certain embodiments, the capsid is modified to reduce "N" or "Q" at positions other than "NG" pairs. Residue numbering is based on the published AAV8 sequence and is reproduced in SEQ ID NO:6.

[0097]

[0098]

[0099] In certain embodiments, the mutants may comprise AAV8 G264A / G515A (SEQ ID NO: 21), AAV8 G264A / G541A (SEQ ID NO: 23), AAV8 G515A / G541A (SEQ ID NO: 25), or AAV8 G264A / G515A / G541A (SEQ ID NO: 27). In certain embodiments, nucleic acid sequences encoding these mutant AAV8 capsids are provided. In certain embodiments, nucleic acid sequences are provided in the following: for example, SEQ ID NO: 20 (AAV8 G264A / G515A), SEQ ID NO: 22 (AAV8 G264A / G541A), SEQ ID NO: 24 (AAV8 G515A / G541A), or SEQ ID NO: 26 (AAV8 G264A / G515A / G541A). In certain embodiments, the AAV8 mutant can be N499Q, N459Q, N305Q / N459Q, N305QN499Q, N459Q, N305Q / N459Q, N305q / N499Q or N205Q, N459Q or N305Q / N459Q, N499Q. In certain embodiments, these mutations are combined with G264A / G541A mutations. In certain embodiments, the mutation is AAV8 G264A / G541A / N499Q (SEQ ID NO: 115), AAV8 G264A / G541A / N459Q (SEQ ID NO: 116), AAV8 G264A / G541A / N305Q / N459Q (SEQ ID NO: 117), AAV8 G264A / G541A / N305Q / N499Q (SEQ ID NO: 118), G264A / G541A / N459Q / N499Q (SEQ ID NO: 119), or AAV8 G264A / G541A / N305Q / N459Q / N499Q (SEQ ID NO: 120). Nucleic acid sequences encoding these AAV8 mutants are also encompassed.

[0100] In certain embodiments, AAV9 is characterized by a capsid composition of a heterogeneous population of VP isotypes, which are deamidated as defined in the following table based on the total amount of VP protein in the capsid as determined using mass spectrometry. In certain embodiments, the AAV capsid is modified at one or more of the following positions within the range provided below as determined using mass spectrometry. Suitable modifications include modifications labeled as deamidation regulation described in the paragraph above, which are incorporated herein. In certain embodiments, one or more of the following positions or glycine after N are modified as described herein. In certain embodiments, AAV9 capsid coding position N214 / G215 is modified to N214Q, which is observed to have significantly increased deamidation. In certain embodiments, NG mutations are performed at the pair positioned at N452 (e.g., to N452A). In certain embodiments, mutations are not performed at position N57. In certain embodiments, AAV mutants containing multiple mutations and at least one of the mutations at these positions are engineered. In certain embodiments, an artificial NG is introduced into a position different from one of the positions identified below. In certain embodiments, the capsid is modified to reduce an "N" or "Q" at a position other than an "NG" pair. Residue numbering is based on the published AAV9 sequence and is reproduced in SEQ ID NO:7.

[0101]

[0102]

[0103] Additionally or alternatively, the AAVhu37 capsid comprises: a heterogeneous population of vp1 proteins as the product of a nucleic acid sequence encoding the amino acid sequence of SEQ ID NO:36; a heterogeneous population of vp2 proteins as the product of a nucleic acid sequence encoding at least about amino acids 138 to 738 of SEQ ID NO:36; and a heterogeneous population of vp3 proteins as the product of a nucleic acid sequence encoding at least amino acids 204 to 738 of SEQ ID NO:36, wherein: the vp1 proteins, the vp2 proteins, and the vp3 proteins contain a subpopulation having amino acid modifications, the amino acid modifications comprising at least two highly deamidated asparagines (N) in the asparagine-glycine pairs in SEQ ID NO:36, and optionally further comprising a subpopulation comprising other deamidated amino acids, wherein the deamidation results in an amino acid change. AAVhu37 is characterized by having highly deamidated residues at, for example, positions N57, N263, N385, and / or N514 based on the numbering of AAVhu37 VP1 (SEQ ID NO: 36).

[0104] As shown in the following table and examples, deamidation is also observed in other residues. In certain embodiments, the AAVhu37 capsid is modified in one or more of the following positions within the ranges provided below as determined using mass spectrometry using trypsin. In certain embodiments, the glycine after one or more of the following positions or N is modified as described herein. For example, in certain embodiments, G can be modified to S or A, for example, at positions 58, 264, 386, or 515. In one embodiment, the AAVhu37 capsid is modified to N57Q or G58A at position N57 / G58 to provide a capsid with reduced deamidation at this position. In another embodiment, N57 / G58 is changed to NS57 / 58 or NA57 / 58. However, in certain embodiments, when NG is changed to NS or NA, increased deamidation is observed. In certain embodiments, the N of the NG pair is modified to Q while retaining G. In certain embodiments, both amino acids of the NG pair are modified. In certain embodiments, N385Q causes a significant reduction in deamidation in the position. In certain embodiments, N499Q results in a significant increase in deamidation in that position.

[0105] In certain embodiments, AAVhu37 can have these or other deamidated residues, e.g., typically less than 10%, and / or can have other modifications including methylation (e.g., about R487) (typically less than 5%, more typically less than 1% at a given residue), isomerization (e.g., at D97) (typically less than 5%, more typically less than 1% at a given residue), phosphorylation (e.g., where present, ranging from about 10% to about 60%). %, or about 10% to about 30%, or about 20% to about 60%) (e.g., at one or more of S149, about S153, about S474, about T570, about S665) or oxidized (e.g., at one or more of W248, W307, W307, M405, M437, M473, W480, W480, W505, M526, M544, M561, W621, M637 and / or W697). Optionally, W can be oxidized to kynurenine.

[0106]

[0107] Other positions may have such or other modifications (e.g., acetylation or additional deamidation). In certain embodiments, a nucleic acid sequence encoding the AAVhu37 vpl capsid protein is provided in SEQ ID NO:37. In other embodiments, a nucleic acid sequence having 70% to 99.9% identity to SEQ ID NO:37 may be selected to express the AAVhu37 capsid protein. In certain other embodiments, the nucleic acid sequence is at least about 75% identical, at least 80% identical, at least 85% identical, at least 90% identical, at least 95% identical, at least 97% identical, or at least 99% to 99.9% identical to SEQ ID NO:37. However, other nucleic acid sequences encoding the amino acid sequence of SEQ ID NO:36 may be selected for use in generating rAAVhu37 capsids. In certain embodiments, the nucleic acid sequence has the nucleic acid sequence of SEQ ID NO: 37, or a sequence at least 70% to 99% identical, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 99% identical to SEQ ID NO: 37 encoding SEQ ID NO: 36. In certain embodiments, the nucleic acid sequence has the nucleic acid sequence of SEQ ID NO: 37, or a sequence at least 70% to 99%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 99% identical to about nt 412 to about nt 2214 of SEQ ID NO: 37 encoding the vp2 capsid protein (about aa 138 to 738) of SEQ ID NO: 36. In certain embodiments, the nucleic acid sequence has a nucleic acid sequence of about nt 610 to about nt 2214 of SEQ ID NO: 37 or a sequence at least 70% to 99%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 97%, at least 99% identical to nt SEQ ID NO: 37 encoding the vp3 capsid protein (about aa 204 to 738) of SEQ ID NO: 36. See EP2345731 B1 and SEQ ID NO: 88 therein, which are incorporated herein by reference.

[0108] As used herein, "encoded amino acid sequence" refers to the predicted amino acid based on the translation of known DNA codons of a reference nucleic acid sequence that is translated into amino acids. The following table shows DNA codons and twenty common amino acids, showing single letter codes (SLC) and three letter codes (3LC), respectively.

[0109]

[0110] rAAV vectors

[0111] As indicated above, novel AAV sequences and proteins can be used to produce rAAV, and can also be used for recombinant AAV vectors, which can be antisense delivery vectors, gene therapy vectors, or vaccine vectors. In addition, the engineered AAV capsids described herein can be used to engineer rAAV vectors to deliver a variety of suitable nucleic acid molecules to target cells and tissues.

[0112] The genomic sequence packaged into the AAV capsid and delivered to the host cell is usually composed of at least a transgene and its regulatory sequence and an AAV inverted terminal repeat (ITR). Both single-stranded AAV and self-complementary (sc) AAV are included in rAAV. Transgenes are nucleic acid coding sequences heterologous to the vector sequence, which encode polypeptides, proteins, functional RNA molecules (e.g., miRNA, miRNA inhibitors) or other gene products of interest. The nucleic acid coding sequence is operably connected to the regulatory component in a manner that allows the transgene to be transcribed, translated and / or expressed in the cells of the target tissue.

[0113] The AAV sequence of the vector generally includes cis-acting 5' and 3' inverted terminal repeats (see, e.g., BJ Carter, Handbook of Parvoviruses, edited by P. Tijsser, CRC Press, pp. 155 to 168 (1990)). The length of the ITR sequence is about 145 bp. Preferably, substantially the entire sequence encoding the ITR is used in the molecule, although minor modifications to these sequences are permitted to a certain extent. The ability to modify these ITR sequences is within the technical scope of the art. (See, e.g., texts such as Sambrook et al., Molecular Cloning. A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory, New York (1989); and K. Fisher et al., Journal of Virology, 70: 520-532 (1996)). An example of such a molecule employed in the present invention is a "cis-acting" plasmid containing a transgene in which the selected transgene sequence and associated regulatory elements flank the 5' and 3' AAV ITR sequences. In one embodiment, the ITR is from an AAV different from the AAV supplying the capsid. In one embodiment, the ITR is from AAV2. A shortened version of the 5' ITR referred to as ΔITR has been described in which the D sequence and the terminal resolution site (trs) are deleted. In other embodiments, full-length AAV 5' and 3' ITRs are used. However, ITRs from other AAV sources can be selected. In the case where the source of the ITR is from AAV2 and the AAV capsid is from another AAV source, the resulting vector can be referred to as a pseudotype. However, other configurations of these elements may be suitable.

[0114] In addition to the above-identified major elements of the recombinant AAV vector, the AAV vector also comprises necessary conventional control elements that are operably linked to the transgene in a manner that allows it to be transcribed, translated and / or expressed in cells transfected with the plasmid vector or infected with the virus produced by the invention. As used herein, "operably linked" sequences include expression control sequences that are contiguous with the gene of interest and expression control sequences that act in trans or at a distance to control the gene of interest.

[0115] Regulatory control elements typically contain a promoter sequence as part of an expression control sequence, for example positioned between a selected 5' ITR sequence and a coding sequence. Constitutive promoters, regulatable promoters [see, for example, WO 2011 / 126808 and WO 2013 / 04943], tissue-specific promoters or promoters responsive to physiological cues may be used in the vectors described herein. The one or more promoters can be selected from different sources, such as human cytomegalovirus (CMV) immediate early enhancer / promoter, SV40 early enhancer / promoter, JC polyoma virus promoter, myelin basic protein (MBP) or glial fibrillary acidic protein (GFAP) promoter, herpes simplex virus (HSV-1) latency-associated promoter (LAP), Rous sarcoma virus (RSV) long terminal repeat (LTR) promoter, neuron-specific promoter (NSE), platelet-derived growth factor (PDGF) promoter, hSYN, melanin concentrating hormone (MCH) promoter, CBA, matrix metalloprotein promoter (MPP) and chicken β-actin promoter. In addition to the promoter, the vector may also contain one or more other suitable transcription initiation, termination, enhancer sequences, effective RNA processing signals such as splicing and polyadenylation (polyA) signals; sequences that stabilize cytoplasmic mRNA, for example, WPRE; sequences that enhance translation efficiency (i.e., Kozak consensus sequences); sequences that enhance protein stability; and sequences that enhance secretion of the encoded product when desired. An example of a suitable enhancer is the CMV enhancer. Other suitable enhancers include enhancers suitable for the desired target tissue indications. In one embodiment, the expression cassette includes one or more expression enhancers. In one embodiment, the expression cassette contains two or more expression enhancers. These enhancers may be the same or different from each other. For example, an enhancer may include a CMV immediate early enhancer. Such an enhancer may be present in two copies positioned adjacent to each other. Alternatively, the double copies of the enhancer may be separated by one or more sequences. In still another embodiment, the expression cassette further contains an intron, for example, a chicken beta-actin intron. Other suitable introns include introns known in the art, for example, as described in WO 2011 / 126808. Examples of suitable polyA sequences include, for example, SV40, SV50, bovine growth hormone (bGH), human growth hormone, and synthetic polyA. Optionally, one or more sequences can be selected to stabilize mRNA. An example of such a sequence is a modified WPRE sequence, which can be engineered upstream of the polyA sequence and downstream of the coding sequence [see, for example, MA Zanta-Boussif et al., Gene Therapy (2009) 16: 605-619].

[0116] These rAAVs are particularly suitable for gene delivery for therapeutic purposes and for immunity, including inducing protective immunity. Further, the compositions of the present invention can also be used for in vitro production of desired gene products. For in vitro production, the desired product (e.g., protein) can be obtained from the desired culture after rAAV transfection host cells containing molecules encoding the desired product and culturing cell cultures under conditions that allow expression. The expressed product can then be purified and separated as needed. Suitable techniques for transfection, cell culture, purification and separation are known to those skilled in the art.

[0117] Therapeutic transgenic

[0118] Useful products encoded by transgenics include various gene products that replace defective or defective genes, inactivate or "knock out", or "knock down" or reduce the expression of genes that express or deliver gene products with desired therapeutic effects at undesirable high levels. In most embodiments, the therapy will be "somatic gene therapy", that is, the gene is transferred to human cells that do not produce sperm or eggs. In certain embodiments, the transgenic expressed protein has the sequence of a natural human sequence. However, in other embodiments, a synthetic protein is expressed. Such proteins can be used to treat humans, or in other embodiments, are designed to treat animals, including companion animals such as canine or feline groups, or for treating livestock or other animals in contact with human groups.

[0119] Examples of suitable gene products may include gene products associated with familial hypercholesterolemia, muscular dystrophy, cystic fibrosis, and rare or orphan diseases. Examples of such rare diseases may include spinal muscular atrophy (SMA), Huntington's disease, Rett Syndrome (e.g., methyl CpG binding protein 2 (MeCP2); UniProtKB-P51608), amyotrophic lateral sclerosis (ALS), Duchenne muscular dystrophy, Friedrichs Ataxia (e.g., ataxin), progranulin (PRGN) (associated with non-Alzheimer's brain degeneration, including frontotemporal dementia (FTD), progressive nonfluent aphasia (PNFA) and semantic dementia), etc. See, e.g., www.orpha.net / consor / cgi-bin / Disease_Search_List.php; rarediseases.info.nih.gov / diseases.

[0120] Examples of suitable genes can include, for example, hormones and growth and differentiation factors, including but not limited to insulin, glucagon, glucagon-like peptide-1 (GLP1), growth hormone (GH), parathyroid hormone (PTH), growth hormone releasing factor (GRF), follicle stimulating hormone (FSH), luteinizing hormone (LH), human chorionic gonadotropin (hCG), vascular endothelial growth factor (VEGF), angiogenin, angiostatin, granulocyte colony stimulating factor (GCSF), erythropoietin (EPO) (including, for example, human, canine or feline epo), connective tissue growth factor (CTGF), neurotrophic factors including, for example, basic fibroblast growth factor (bFGF), acidic fibroblast growth factor (aFGF), epidermal growth factor (EGF), platelet-derived growth factor (PDGF), insulin growth factor I and II (IGF-I and IGF-II), any of the transforming growth factor alpha superfamily (including TGFα, activin, inhibin), or bone morphogenetic protein (BMP) BMP Any of 1-15, the heregulin / neuregulin / ARIA / neu differentiation factor (NDF) family of growth factors, nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophins NT-3 and NT-4 / 5, ciliary neurotrophic factor (CNTF), glial cell line-derived neurotrophic factor (GDNF), neurotrophin, any of agrin, semaphorin / brain degeneration protein, spindle protein-1 and spindle protein-2, hepatocyte growth factor (HGF), ephrin, noggin, sonic hedgehog and any of the tyrosine hydroxylase family.

[0121] Other useful transgenic products include proteins that modulate the immune system, including but not limited to cytokines and lymphokines, such as thrombopoietin (TPO), interleukins (IL) IL-1 to IL-36 (including, for example, human interleukins IL-1, IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-6, IL-8, IL-12, IL-11, IL-12, IL-13, IL-18, IL-31, IL-35), monocyte chemoattractant protein, leukemia inhibitory factor, granulocyte-macrophage colony stimulating factor, Fas ligand, tumor necrosis factor α and β, interferon α, β and γ, stem cell factor, flk-2 / flt3 ligand. Gene products produced by the immune system can also be used in the present invention. These include, but are not limited to, immunoglobulins IgG, IgM, IgA, IgD and IgE, chimeric immunoglobulins, humanized antibodies, single-chain antibodies, T cell receptors, chimeric T cell receptors, single-chain T cell receptors, MHC class I and class II molecules, and engineered immunoglobulins and MHC molecules. For example, in certain embodiments, rAAV antibodies can be designed to deliver canine or feline antibodies, such as anti-IgE, anti-IL31, anti-CD20, anti-NGF, anti-GnRH. Useful gene products also include complement regulatory proteins, such as complement regulatory proteins, membrane cofactor proteins (MCPs), decay accelerating factors (DAFs), CR1, CF2, CD59, and C1 esterase inhibitors (C1-INH).

[0122] Still other useful gene products include any one of the receptors for hormones, growth factors, cytokines, lymphokines, regulatory proteins, and immune system proteins. The present invention encompasses receptors for cholesterol regulation and / or lipid regulation, including low-density lipoprotein (LDL) receptors, high-density lipoprotein (HDL) receptors, very low-density lipoprotein (VLDL) receptors, and scavenger receptors. The present invention also encompasses gene products such as members of the steroid hormone receptor superfamily, including glucocorticoid receptors and estrogen receptors, vitamin D receptors, and other nuclear receptors. In addition, useful gene products include transcription factors, such as jun, fos, max, mad, serum response factor (SRF), AP-1, AP2, myb, MyoD and myogenin, ETS-box containing proteins, TFE3, E2F, ATF1, ATF2, ATF3, ATF4, ZF5, NFAT, CREB, HNF-4, C / EBP, SP1, CCAAT box binding protein, interferon regulatory factor (IRF-1), Wilms tumor protein, ETS binding protein, STAT, GATA box binding protein (e.g., GATA-3) and the forkhead family of winged helix proteins.

[0123] Other useful gene products include carbamoyl synthetase I, ornithine transcarbamylase (OTC), argininosuccinate synthetase, argininosuccinate lyase (ASL) for the treatment of argininosuccinate lyase deficiency, arginase, fumarylacetoacetate hydrolase, phenylalanine hydroxylase, alpha-1 antitrypsin, rhesus alpha-fetoprotein (AFP), rhesus chorionic gonadotropin (CG), glucose-6-phosphatase, porphobilinogen deaminase, cystathionine beta synthase, branched-chain ketoacid decarboxylase, albumin, isovaleryl-CoA dehydrogenase, propionyl-CoA carboxylase, methylmalonyl-CoA mutase, glutaryl-CoA dehydrogenase, insulin, beta-glucosidase, pyruvate carboxylate, liver phosphorylase, phosphorylase kinase, glycine decarboxylase, H protein, T protein, cystic fibrosis transmembrane regulator (CFTR) sequence, and dystrophin gene products [e.g., mini- or micro-dystrophin]. Still other useful gene products also include enzymes such as those that can be used in enzyme replacement therapy, which can be used for a variety of conditions caused by insufficient enzyme activity. For example, an enzyme containing mannose-6-phosphate can be used in the therapy of lysosomal storage diseases (e.g., a suitable gene includes a gene encoding β-glucuronidase (GUSB)).

[0124] In certain embodiments, rAAV can be used in a gene editing system that can involve co-administration of one rAAV or multiple rAAV stocks. For example, rAAV can be engineered to deliver SpCas9, SaCas9, ARCUS, Cpf1, and other suitable gene editing constructs.

[0125] Still other useful gene products include gene products for treating hemophilia, including hemophilia B (comprising factor IX) and hemophilia A (comprising factor VIII and variants thereof, such as light and heavy chains of heterodimers and B-domain deletions; U.S. Pat. No. 6,200,560 and U.S. Pat. No. 6,221,349). In some embodiments, the minigene includes the first 57 base pairs of the factor VIII heavy chain, which encodes a 10 amino acid signal sequence and a human growth hormone (hGH) polyadenylation sequence. In alternative embodiments, the minigene further includes the A1 and A2 domains and 5 amino acids from the N-terminus of the B domain and / or 85 amino acids from the C-terminus of the B domain and the A3, C1, and C2 domains. In yet other embodiments, nucleic acids encoding the Factor VIII heavy and light chains are provided in a single minigene separated by 42 nucleic acids encoding the 14 amino acids of the B domain [U.S. Patent No. 6,200,560].

[0126] Other useful gene products include non-natural polypeptides, such as chimeric or hybrid polypeptides having non-natural amino acid sequences containing insertions, deletions or amino acid substitutions. For example, single-chain engineered immunoglobulins may be useful in certain immunocompromised patients. Other types of non-natural gene sequences include antisense molecules and catalytic nucleic acids, such as ribozymes, which can be used to reduce overexpression of targets.

[0127] Reducing and / or regulating gene expression is particularly desirable for treating the hyperproliferative condition characterized by cell hyperproliferation (such as cancer and psoriasis). Target polypeptides include those polypeptides produced specifically or at higher levels in hyperproliferative cells compared with normal cells. Target antigens include polypeptides encoded by oncogenes such as myb, myc, fyn and translocation genes bcr / abl, ras, src, P53, neu, trk and EGRF. In addition to the oncogene product as the target antigen, the target polypeptide for anticancer treatment and protection scheme includes the variable region of the antibody produced by B cell lymphoma and the variable region of the T cell receptor of T cell lymphoma, and in some embodiments, the variable region is also used as the target antigen of autoimmune disease. Other tumor-associated polypeptides can also be used as target polypeptides, such as polypeptides present at higher levels in tumor cells, including polypeptides identified by monoclonal antibody 17-1A and folic acid binding polypeptides.

[0128] Other suitable therapeutic polypeptides and proteins include those that can be used to treat individuals with autoimmune diseases and disorders by conferring a broad-based protective immune response against targets associated with autoimmunity, including cell receptors and cells that produce "self" directed antibodies. T cell-mediated autoimmune diseases include rheumatoid arthritis (RA), multiple sclerosis (MS), Sjögren's syndrome ( syndrome, sarcoidosis, insulin-dependent diabetes mellitus (IDDM), autoimmune thyroiditis, reactive arthritis, ankylosing spondylitis, scleroderma, polymyositis, dermatomyositis, psoriasis, Wegner's granulomatosis, Crohn's disease, and ulcerative colitis. Each of these diseases is characterized by a T cell receptor (TCR) that binds to endogenous antigens and initiates the inflammatory cascade associated with autoimmune diseases.

[0129] Additional illustrative genes that may be delivered via rAAV include, but are not limited to, glucose-6-phosphatase associated with glycogen storage disease or deficiency type 1A (GSD1); phosphoenolpyruvate carboxykinase (PEPCK) associated with PEPCK deficiency; cyclin-dependent kinase-like 5 (CDKL5), also known as serine / threonine kinase 9 (STK9), associated with seizures and severe neurodevelopmental disorders; galactose-1 phosphate uracil transferase associated with galactosemia; phenylalanine hydroxylase associated with phenylketonuria (PKU); and maple kinase-associated kinase. Branched-chain alpha-ketoacid dehydrogenase associated with diabetes mellitus; fumarylacetoacetate hydrolase associated with type 1 tyrosinemia; methylmalonyl-CoA mutase associated with methylmalonic acidemia; medium-chain acyl-CoA dehydrogenase associated with medium-chain acetyl-CoA deficiency; ornithine transcarbamylase (OTC) associated with ornithine transcarbamylase deficiency; argininosuccinate synthetase (ASS1) associated with citrullinemia; lecithin cholesterol acyltransferase (LCAT) deficiency; methylmalonic acidemia (MMA); Niemann-Pick disease (Niemann-Pick disease) disease) (type C1); propionic acidemia (PA); low-density lipoprotein receptor (LDLR) protein associated with familial hypercholesterolemia (FH); UDP-glucuronosyltransferase associated with Crigler-Najjar disease; adenosine deaminase associated with severe combined immunodeficiency disease; hypoxanthine guanine phosphoribosyltransferase associated with gout and Lesch-Nyhan syndrome; biotinidase associated with biotinidase deficiency; α-galactosidase A (α-GalA) associated with Fabrydisease; ATP7B associated with Wilson's Disease; β-glucocerebrosidase associated with Gaucher disease types 2 and 3; and Zellweger syndrome. syndrome); arylsulfatase A (ARSA) associated with metachromatic leukodystrophy; galactocerebrosidase (GALC) enzyme associated with Krabbe disease; alpha-glucosidase (GAA) associated with Pompe disease; sphingomyelinase (SMPD1) gene associated with Nieman-Pick disease type A; argininosuccinate synthase associated with adult-onset citrullinemia type II (CTLN2); carbamoyl phosphate synthase 1 (CPS1) associated with urea cycle disorders; survival motor neuron (SMN) protein associated with spinal muscular atrophy;ceramidase associated with Farber lipogranulomatosis; b-hexosaminidase associated with GM2 gangliosidosis and Tay-Sachs and Sandhoff diseases; aspartylglucosaminidase associated with aspartylglucosidase; fucosidase associated with fucosidosis; alpha-mannosidase associated with alpha-mannosidosis; porphobilinogen deaminase associated with acute intermittent porphyria (AIP); alpha-1 antitrypsin for treatment of alpha-1 antitrypsin deficiency (emphysema); erythropoietin for treatment of anemia due to thalassemia or renal failure; vascular endothelial growth factor, angiopoietin-1, and fibroblast growth factor for treatment of ischemic disease; thrombomodulin and tissue factor for treatment of blocked blood vessels as seen, for example, in atherosclerosis, thrombosis, or embolism pathway inhibitors; aromatic amino acid decarboxylase (AADC) and tyrosine hydroxylase (TH) for the treatment of Parkinson's disease; beta-adrenergic receptors in antisense or mutant form to phospholamban, sarcoplasmic (endoplasmic) reticulum adenosine triphosphatase 2 (SERCA2); cardiac adenylate cyclase for the treatment of congestive heart failure; tumor suppressor genes such as p53 for the treatment of various cancers; cytokines such as one of the various interleukins for the treatment of inflammatory and immune disorders and cancer; dystrophin or mini-dystrophin and dystrophin-related protein or mini-dystrophin for the treatment of muscular dystrophy; and insulin or GLP-1 for the treatment of diabetes.

[0130] In certain embodiments, the rAAV described herein can be used to treat a mucopolysaccharidosis (MPS) disorder. Such rAAV may contain nucleic acid sequences encoding α-L-iduronidase (IDUA) for the treatment of MPS I (Hurler, Hurler-Scheie and Scheie syndromes); nucleic acid sequences encoding iduronate-2-sulfatase (IDS) for the treatment of MPS II (Hunter syndrome); nucleic acid sequences encoding sulfamidase (SGSH) for the treatment of MPS IIIA, B, C and D (Sanfilippo syndrome); nucleic acid sequences encoding N-acetylgalactosamine-6-sulfate sulfatase (GALNS) for the treatment of MPS IVA and B (Morquio syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS ... VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome); nucleic acid sequences encoding sulphatase (SLS) for the treatment of MPS VI (Maroteaux-Lamy syndrome). syndrome); a nucleic acid sequence encoding a hyaluronidase for the treatment of MPS IX (hyaluronidase deficiency); and a nucleic acid sequence encoding a β-glucuronidase for the treatment of MPS VII (Sly syndrome).

[0131] Immunogenic transgene

[0132] In some embodiments, rAAV vectors comprising nucleic acids encoding gene products associated with cancer (e.g., tumor suppressors) can be used to treat cancer by administering rAAV containing rAAV vectors to subjects with cancer. In some embodiments, rAAV vectors comprising nucleic acids encoding small interfering nucleic acids (e.g., shRNA, miRNA) that inhibit the expression of gene products associated with cancer (e.g., oncogenes) can be used to treat cancer by administering rAAV containing rAAV vectors to subjects with cancer. In some embodiments, rAAV vectors comprising nucleic acids encoding gene products associated with cancer (or functional RNAs that inhibit the expression of genes associated with cancer) can be used for research purposes, such as studying cancer or identifying therapeutic agents for treating cancer. The following is a non-limiting list of exemplary genes (e.g., oncogenes and tumor suppressors) known to be associated with the development of cancer: AARS, ABCB1, ABCC4, ABI2, ABL1, ABL2, ACK1, ACP2, ACY1, ADSL, AK1, AKR1C2, AKT1, ALB, ANPEP, ANXA5, ANXA7, AP2M1, APC, ARHGAP5, ARHGEF5, ARID4A, ASNS, ATF4, ATM, AT P5B, ATP5O, AXL, BARD1, BAX, BCL2, BHLHB2, BLMH, BRAF, BRCA1, BRCA2, BTK, CANX, CAP1, CAPN1, CAPNS1, CAV1, CBFB, CBLB, CCL2, CCND1, CCND2, CCND3, CCNE1, CCT5, CCYR61, CD24, CD44, CD59, CDC20, CDC25, CDC25A, CDC25B , CDC2L5, CDK10, CDK4, CDK5, CDK9, CDKL1, CDKN1A, CDKN1B, CDKN1C, CDKN2A, CDKN2B, CDKN2D, CEBPG, CENPC1, CGRRF1, CHAF1A, CIB1, CKMT1, CLK1, CLK2, CLK3, CLNS1A, CLTC, COL1A1, COL6A3, COX6C, COX7A2, CRAT, CRHR1, C SF1R, CSK, CSNK1G2, CTNNA1, CTNNB1, CTPS, CTSC, CTSD, CUL1, CYR61, DCC, DCN, DDX10, DEK, DHCR7, DHRS2, DHX8 , DLG3, DVL1, DVL3, E2F1, E2F3, E2F5, EGFR, EGR1, EIF5, EPHA2, ERBB2, ERBB3, ERBB4, ERCC3, ETV1, ETV3, ETV6,<h2 style=";text-align:left;direction:ltr">F2R、FASTK、FBN1、FBN2、FES、FGFR1、FGR、FKBP8、FN1、FOS、FOSL1、FOSL2、FO XG1A, FOXO1A, FRAP1, FRZB, FTL, FZD2, FZD5, FZD9, G22P1, GAS6, GCN5L2, GD F15、GNA13、GNAS、GNB2、GNB2L1、GPR39、GRB2、GSK3A、GSPT1、GTF2I、HDAC1、 HDGF、HMMR、HPRT1、HRB、HSPA4、HSPA5、HSPA8、HSPB1、HSPH1、HYAL1、HYOU1、I CAM1, ID1, ID2, IDUA, IER3, IFITM1, IGF1R, IGF2R, IGFBP3, IGFBP4, IGFBP5, IL1B, ILK, ING1, IRF3, ITGA3, ITGA6, ITGB4, JAK1, JARID1A, JUN, JUNB, JU ND、K-α-1、KIT、KITLG、KLK10、KPNA2、KRAS2、KRT18、KRT2A、KRT9、LAMB1、LA MP2、LCK、LCN2、LEP、LITAF、LRPAP1、LTF、LYN、LZTR1、MADH1、MAP2K2、MAP3K8 、MAPK12、MAPK13、MAPKAPK3、MAPRE1、MARS、MAS1、MCC、MCM2、MCM4、MDM2、MD M4、MET、MGST1、MICB、MLLT3、MME、MMP1、MMP14、MMP17、MMP2、MNDA、MSH2、MS H6、MT3、MYB、MYBL1、MYBL2、MYC、MYCL1、MYCN、MYD88、MYL9、MYLK、NEO1、NF1 、NF2、NFKB1、NFKB2、NFSF7、NID、NINE、NMBR、NME1、NME2、NME3、NOTCH1、NOTC H2、NOTCH4、NPM1、NQO1、NR1D1、NR2F1、NR2F6、NRAS、NRG1、NSEP1、OSM、PA2G 4、PABPC1、PCNA、PCTK1、PCTK2、PCTK3、PDGFA、PDGFB、PDGFRA、PDPK1、PEA15、 PFDN4、PFDN5、PGAM1、PHB、PIK3CA、PIK3CB、PIK3CG、PIM1、PKM2、PKMYT1、PL K2、PPARD、PPARG、PPIH、PPP1CA、PPP2R5A、PRDX2、PRDX4、PRKAR1A、PRKCBP1、PRNP, PRSS15, PSMA1, PTCH, PTEN, PTGS1, PTMA, PTN, PTPRN, RAB5A, RAC1, RAD50, RAF1, RALBP1, RAP1A, RAR B. RASGRF1, RB1, RBBP4, RBL2, REA, REL, RELA, RELB, RET, RFC2, RGS19, RHOA, RHOB, RHOC, RHOD, RIPK1, RPN2, RPS6 KB1, RRM1, SARS, SELENBP1, SEMA3C, SEMA4D, SEPP1, SERPINH1, SFN, SFPQ, SFRS7, SHB, SHH, SIAH2, SIVA, SIVA TP53, SKI, SKIL, SLC16A1, SLC1A4, SLC20A1, SMO, Stimulus Protein Protein Dispersed Pain1(SMPD1), SNAI2, S.K ND1, SNRPB2, SOCS1, SOCS3, SOD1, SORT1, SPINT2, SPRY2, SRC, SRPX, STAT1, STA T2, STAT3, STAT5B, STC1, TAF1, TBL3, TBRG4, TCF1, TCF7L2, TFAP2C, TFDP1, TFD P2, TGFA, TGFB1, TGFBI, TGFBR2, TGFBR3, THBS1, TIE, TIMP1, TIMP3, TJP1, TK1, T LE1, TNF, TNFRSF10A, TNFRSF10B, TNFRSF1A, TNFRSF1B, TNFRSF6, TNFSF7, TNK1 TOB1, TP53, TP53BP2, TP5313, TP73, TPBG, TPT1, TRADD, TRAM1, TRRAP, TSG101 TUFM, TXNRD1, TYRO3, UBC, UBE2L6, UCHL1, USP7, VDAC1, VEGF, VHL, VIL2, WEE1. WNT1, WNT2, WNT2B, WNT3, WNT5A, WT1, XRCC1, YES1, YWHAB, YWHAZ, ZAP70, and ZNF9.

[0133] The rAAV vector may include a nucleic acid encoding a protein or functional RNA that regulates apoptosis as a transgene. The following is a non-limiting list of genes associated with apoptosis, and nucleic acids encoding the products of these genes and their homologs and small interfering nucleic acids (e.g., shRNA, miRNA) that inhibit the expression of these genes and their homologs are used as transgenes in certain embodiments of the present invention: RPS27A, ABL1, AKT1, APAF1, BAD, BAG1, BAG3, BAG4, BAK1, BAX, BCL10, BCL2, BCL2A1, BCL2L1, BCL2L10, BCL2L11, BCL2L12, BCL2L13, BCL2L2, BCLAF1, BFAR, BID, BIK, NAIP, BIRC2, BIRC3, XIAP, BIRC5, BIRC6, BIRC7, BIRC8, BNIP1, BNIP2, BNIP3, BNI P3L, BOK, BRAF, CARD10, CARD11, NLRC4, CARD14, NOD2, NOD1, CARD6, CARDS, CARDS, CASP1, CASP10, CASP14, CASP2, CASP3, CAS P4, CASP5, CASP6, CASP7, CASP8, CASP9, CFLAR, CIDEA, CIDEB, CRADD, DAPK1, DAPK2, DFFA, DFFB, FADD, GADD45A, GDNF, HRK, I GF1R, LTA, LTBR, ​​MCL1, NOL3, PYCARD, RIPK1, RIPK2, TNF, TNFRSF10A, TNFRSF10B, TNFRSF10C, TNFRSF10D, TNFRSF11B, TNFRS F12A, TNFRSF14, TNFRSF19, TNFRSF1A, TNFRSF1B, TNFRSF21, TNFRSF25, CD40, FAS, TNFRSF6B, CD27, TNFRSF9, TNFSF10, TNFS F14, TNFSF18, CD40LG, FASLG, CD70, TNFSF8, TNFSF9, TP53, TP53BP2, TP73, TP63, TRADD, TRAF1, TRAF2, TRAF3, TRAF4 and TRAF5.

[0134] Useful transgenic products also include miRNA. MiRNA and other small interfering nucleic acids regulate gene expression by target RNA transcript cracking / degradation or translation inhibition of target messenger RNA (mRNA). MiRNA is naturally expressed, usually as final 19-25 kinds of non-translated RNA products. MiRNA shows its activity by interacting with the sequence specificity of the 3' non-translated region (UTR) of the target mRNA. These endogenously expressed miRNAs form hairpin precursors, which are subsequently processed into miRNA duplexes and further processed into "mature" single-stranded miRNA molecules. This mature miRNA guides the multiprotein complex miRISC, which identifies the target site of the target mRNA based on the complementarity with the mature miRNA, for example, in the 3'UTR region.

[0135] In certain embodiments of the methods, the following non-limiting list of miRNA genes and their homologs can be used as targets for the transgene or small interfering nucleic acids (e.g., miRNA sponges, antisense oligonucleotides, TuD RNA) encoded by the transgene: hsa-let-7a, hsa-let-7a * 、hsa-let-7b、hsa-let-7b * 、hsa-let-7c、hsa-let-7c * 、hsa-let-7d、hsa-let-7d * 、hsa-let-7e、hsa-let-7e * 、hsa-let-7f、hsa-let-7f-1 * 、hsa-let-7f-2 * 、hsa-let-7g、hsa-let-7g * 、hsa-let-71、hsa-let-71 * , hsa-miR-1, hsa-miR-100, hsa-miR-100 * 、hsa-miR-101、hsa-miR-101 * , hsa-miR-103, hsa-miR-105, hsa-miR-105 * , hsa-miR-106a, hsa-miR-106a * , hsa-miR-106b, hsa-miR-106b * , hsa-miR-107, hsa-miR-10a, hsa-miR-10a * 、hsa-miR-10b、hsa-miR-10b *hsa-miR-1178, hsa-miR-1179, hsa-miR-1180, hsa-miR-1181, hsa-miR-1182, hsa-miR-1183, hsa-miR-1184, hsa-miR-1185, hsa-miR-1197, hsa-miR-1200, hsa-miR -1201, hsa-miR-1202, hsa-miR-1203, hsa-miR-1204, hsa-miR-1205, hsa-miR-1206, hsa-miR-1207-3p, hsa-miR-1207-5p, hsa-miR-1208, hsa-miR-122, hsa-miR-122 * hsa-miR-1224-3p, hsa-miR-1224-5p, hsa-miR-1225-3p, hsa-miR-1225-5p, hsa-miR-1226, hsa-miR-1226 * hsa-miR-1227, hsa-miR-1228, hsa-miR-1228 * hsa-miR-1229, hsa-miR-1231, hsa-miR-1233, hsa-miR-1234, hsa-miR-1236, hsa-miR-1237, hsa-miR-1238, hsa-miR-124 * hsa-miR-1243, hsa-miR-1244, hsa-miR-1245, hsa-miR-1246, hsa-miR-1247, hsa-miR-1248, hsa-miR-1249, hsa-miR-1250, hsa-miR-1251, hsa-miR-1252, hsa-miR-1253, hsa- miR-1254、hsa-miR-1255a、hsa-miR-1255b、hsa-miR-1256、hsa-miR-1257、hsa-miR-125 8、hsa-miR-1259、hsa-miR-125a-3p、hsa-miR-125a-5p、hsa-miR-125b、hsa-miR-125b-1 * hsa-miR-125b-2 * hsa-miR-126 hsa-miR-126 *hsa-miR-1260, hsa-miR-1261, hsa-miR-1262, hsa-miR-1263, hsa-miR-1264, hsa-miR-1265, hsa-miR-1266, hsa-miR-1267, hsa-miR-1268, hsa-miR-1269, hsa-miR-1270, hsa-miR-1271, hsa-miR-1272, hsa-miR-1273, hsa-miR-127-3p, hsa-miR-1274a, hsa-miR-1274b, hsa -miR-1275, hsa-miR-127-5p, hsa-miR-1276, hsa-miR-1277, hsa-miR-1278, hsa-miR-1279, hsa-miR-128, hsa-miR-1280, hsa-miR-1281, hsa-miR-1282, hsa-miR-1283, hsa-miR-1284, hsa-miR-1285, hsa-miR-1286, hsa-miR-1287, hsa-miR-1288, hsa-miR-1289, hsa-miR-129 * hsa-miR-1290, hsa-miR-1291, hsa-miR-1292, hsa-miR-1293, hsa-miR-129-3p, hsa-miR-1294, hsa-miR-1295, hsa-miR-129-5p, hsa-miR-1296, hsa-miR-1297, hsa-miR-1298, hsa -miR-1299, hsa-miR-1300, hsa-miR-1301, hsa-miR-1302, hsa-miR-1303, hsa-miR-1304, hsa-miR-1305, hsa-miR-1306, hsa-miR-1307, hsa-miR-1308, hsa-miR-130a, hsa-miR-130a * 、hsa-miR-130b、hsa-miR-130b * hsa-miR-132 hsa-miR-132 * hsa-miR-1321, hsa-miR-1322, hsa-miR-1323, hsa-miR-1324, hsa-miR-133a, hsa-miR-133b, hsa-miR-134, hsa-miR-135a, and hsa-miR-135a *hsa-miR-135b hsa-miR-135b * hsa-miR-136 hsa-miR-136 * hsa-miR-137, hsa-miR-138, hsa-miR-138-1 * hsa-miR-138-2 * hsa-miR-139-3p, hsa-miR-139-5p, hsa-miR-140-3p, hsa-miR-140-5p, hsa-miR-141, hsa-miR-141 * hsa-miR-142-3p, hsa-miR-142-5p, hsa-miR-143, hsa-miR-143 * hsa-miR-144 hsa-miR-144 * hsa-miR-145 hsa-miR-145 * hsa-miR-146a hsa-miR-146a * hsa-miR-146b-3p, hsa-miR-146b-5p, hsa-miR-147, hsa-miR-147b, hsa-miR-148a, and hsa-miR-148a * hsa-miR-148b hsa-miR-148b * hsa-miR-149 hsa-miR-149 * hsa-miR-150 hsa-miR-150 * hsa-miR-151-3p, hsa-miR-151-5p, hsa-miR-152, hsa-miR-153, hsa-miR-154, hsa-miR-154 * hsa-miR-155 hsa-miR-155 * hsa-miR-15a hsa-miR-15a * hsa-miR-15b hsa-miR-15b * hsa-miR-16, hsa-miR-16-1 * hsa-miR-16-2 * hsa-miR-17 hsa-miR-17 * 、hsa-miR-181a、hsa-miR-181a * hsa-miR-181a-2* hsa-miR-181b, hsa-miR-181c, hsa-miR-181c * hsa-miR-181d, hsa-miR-182, hsa-miR-182 * hsa-miR-1825, hsa-miR-1826, hsa-miR-1827, hsa-miR-183, hsa-miR-183 * hsa-miR-184, hsa-miR-185, hsa-miR-185 * hsa-miR-186 hsa-miR-186 * hsa-miR-187 hsa-miR-187 * hsa-miR-188-3p, hsa-miR-188-5p, hsa-miR-18a, hsa-miR-18a * hsa-miR-18b hsa-miR-18b * hsa-miR-190, hsa-miR-190b, hsa-miR-191, hsa-miR-191 * hsa-miR-192 hsa-miR-192 * 、hsa-miR-193a-3p、hsa-miR-193a-5p、hsa-miR-193b、hsa-miR-193b * hsa-miR-194 hsa-miR-194 * hsa-miR-195 hsa-miR-195 * 、hsa-miR-196a、hsa-miR-196a * hsa-miR-196b, hsa-miR-197, hsa-miR-198, hsa-miR-199a-3p, hsa-miR-199a-5p, hsa-miR-199b-5p, hsa-miR-19a, hsa-miR-19a * hsa-miR-19b, hsa-miR-19b-1 * hsa-miR-19b-2 * hsa-miR-200a hsa-miR-200a * hsa-miR-200b hsa-miR-200b * 、hsa-miR-200c、hsa-miR-200c* hsa-miR-202 hsa-miR-202 * hsa-miR-203, hsa-miR-204, hsa-miR-205, hsa-miR-206, hsa-miR-208a, hsa-miR-208b, hsa-miR-20a, hsa-miR-20a * hsa-miR-20b hsa-miR-20b * hsa-miR-21 hsa-miR-21 * hsa-miR-210, hsa-miR-211, hsa-miR-212, hsa-miR-214, hsa-miR-214 * hsa-miR-215, hsa-miR-216a, hsa-miR-216b, hsa-miR-217, hsa-miR-218, and hsa-miR-218-1 * hsa-miR-218-2 * hsa-miR-219-1-3p, hsa-miR-219-2-3p, hsa-miR-219-5p, hsa-miR-22, hsa-miR-22 * hsa-miR-220a, hsa-miR-220b, hsa-miR-220c, hsa-miR-221, hsa-miR-221 * hsa-miR-222 hsa-miR-222 * hsa-miR-223 hsa-miR-223 * 、hsa-miR-224、hsa-miR-23a、hsa-miR-23a * hsa-miR-23b hsa-miR-23b * hsa-miR-24, hsa-miR-24-1 * hsa-miR-24-2 * hsa-miR-25 hsa-miR-25 * hsa-miR-26a, hsa-miR-26a-1 * hsa-miR-26a-2 * hsa-miR-26b hsa-miR-26b * hsa-miR-27a hsa-miR-27a *hsa-miR-27b hsa-miR-27b * hsa-miR-28-3p, hsa-miR-28-5p, hsa-miR-296-3p, hsa-miR-296-5p, hsa-miR-297, hsa-miR-298, hsa-miR-299-3p, hsa-miR-299-5p, hsa-miR-29a, hsa-miR-29a * hsa-miR-29b, hsa-miR-296-1 * hsa-miR-296-2 * hsa-miR-29c hsa-miR-29c * 、hsa-miR-300、hsa-miR-301a、hsa-miR-301b、hsa-miR-302a、hsa-miR-302a * 、hsa-miR-302b、hsa-miR-302b * 、hsa-miR-302c、hsa-miR-302c * hsa-miR-302d hsa-miR-302d * hsa-miR-302e, hsa-miR-302f, hsa-miR-30a, hsa-miR-30a * hsa-miR-30b hsa-miR-30b * hsa-miR-30c, hsa-miR-30c-1 * hsa-miR-30c-2 * hsa-miR-30d hsa-miR-30d * hsa-miR-30e hsa-miR-30e * hsa-miR-31 hsa-miR-31 * hsa-miR-32 hsa-miR-32 *hsa-miR-320a, hsa-miR-320b, hsa-miR-320c, hsa-miR-320d, hsa-miR-323-3p, hsa-miR-323-5p, hsa-miR-324-3p, hsa-miR-324-5p, hsa-miR-325, hsa-miR-326, hsa-miR-328, hsa-miR-329, hsa-miR-330-3p, hsa-miR-330-5p, hsa-miR-331-3p, hsa-miR-331-5p, hsa-miR-335, hsa-miR-335 * hsa-miR-337-3p, hsa-miR-337-5p, hsa-miR-338-3p, hsa-miR-338-5p, hsa-miR-339-3p, hsa-miR-339-5p, hsa-miR-33a, and hsa-miR-33a * hsa-miR-33b hsa-miR-33b * hsa-miR-340 hsa-miR-340 * hsa-miR-342-3p, hsa-miR-342-5p, hsa-miR-345, hsa-miR-346, hsa-miR-34a, hsa-miR-34a * hsa-miR-34b hsa-miR-34b * hsa-miR-34c-3p, hsa-miR-34c-5p, hsa-miR-361-3p, hsa-miR-361-5p, hsa-miR-362-3p, hsa-miR-362-5p, hsa-miR-363, hsa-miR-363 * hsa-miR-365, hsa-miR-367, hsa-miR-367 * hsa-miR-369-3p, hsa-miR-369-5p, hsa-miR-370, hsa-miR-371-3p, hsa-miR-371-5p, hsa-miR-372, hsa-miR-373, hsa-miR-373 * 、hsa-miR-374a、hsa-miR-374a * 、hsa-miR-374b、hsa-miR-374b * hsa-miR-375, hsa-miR-376a, hsa-miR-376a* hsa-miR-376b, hsa-miR-376c, hsa-miR-377, hsa-miR-377 * hsa-miR-378 hsa-miR-378 * hsa-miR-379 hsa-miR-379 * hsa-miR-380 hsa-miR-380 * hsa-miR-381, hsa-miR-382, hsa-miR-383, hsa-miR-384, hsa-miR-409-3p, hsa-miR-409-5p, hsa-miR-410, hsa-miR-411, hsa-miR-411 * hsa-miR-412, hsa-miR-421, hsa-miR-422a, hsa-miR-423-3p, hsa-miR-423-5p, hsa-miR-424, hsa-miR-424 * hsa-miR-425 hsa-miR-425 * hsa-miR-429, hsa-miR-431, hsa-miR-431 * hsa-miR-432 hsa-miR-432 * hsa-miR-433, hsa-miR-448, hsa-miR-449a, hsa-miR-449b, hsa-miR-450a, hsa-miR-450b-3p, hsa-miR-450b-5p, hsa-miR-451, hsa-miR-452, hsa-miR-452 * hsa-miR-453, hsa-miR-454, hsa-miR-454 * hsa-miR-455-3p, hsa-miR-455-5p, hsa-miR-483-3p, hsa-miR-483-5p, hsa-miR-484, hsa-miR-485-3p, hsa-miR-485-5p, hsa-miR-486-3p, hsa-miR-486-5p, hsa-miR-487a, hsa-miR-487b, hsa-miR-488, hsa-miR-488 *hsa-miR-489, hsa-miR-490-3p, hsa-miR-490-5p, hsa-miR-491-3p, hsa-miR-491-5p, hsa-miR-492, hsa-miR-493, hsa-miR-493 * hsa-miR-494, hsa-miR-495, hsa-miR-496, hsa-miR-497, hsa-miR-497 * hsa-miR-498, hsa-miR-499-3p, hsa-miR-499-5p, hsa-miR-500, hsa-miR-500 * hsa-miR-501-3p, hsa-miR-501-5p, hsa-miR-502-3p, hsa-miR-502-5p, hsa-miR-503, hsa-miR-504, hsa-miR-505, hsa-miR-505 * hsa-miR-506, hsa-miR-507, hsa-miR-508-3p, hsa-miR-508-5p, hsa-miR-509-3p, hsa-miR-509-5p, hsa-miR-510, hsa-miR-511, hsa-miR-512-3p, and hsa-miR-512-5p hsa-miR-513a-3p, hsa-miR-513a-5p, hsa-miR-513b, hsa-miR-513c, hsa-miR-514, hsa-miR-515-3p, hsa-miR-515-5p, hsa-miR-516a-3p, hsa-miR-516a-5p, hsa-miR-516b, and hsa-miR-517 * 、hsa-miR-517a、hsa-miR-517b、hsa-miR-517c、hsa-miR-518a-3p、hsa-miR-518a-5p、hsa-miR-518b、hsa-miR-518c、hsa-miR-518c * 、hsa-miR-518d-3p、hsa-miR-518d-5p、hsa-miR-518e、hsa-miR-518e * 、hsa-miR-518f、hsa-miR-518f *hsa-miR-519a, hsa-miR-519b-3p, hsa-miR-519c-3p, hsa-miR-519d, hsa-miR-519e, hsa-miR-519e * 、hsa-miR-520a-3p、hsa-miR-520a-5p、hsa-miR-520b、hsa-miR-520c-3p、hsa- miR-520d-3p、hsa-miR-520d-5p、hsa-miR-520e、hsa-miR-520f、hsa-miR-520g、 hsa-miR-520h, hsa-miR-521, hsa-miR-522, hsa-miR-523, hsa-miR-524-3p, hsa-miR-524-5p, hsa-miR-525-3p, hsa-miR-525-5p, hsa-miR-526b, and hsa-miR-526b * hsa-miR-532-3p, hsa-miR-532-5p, hsa-miR-539, hsa-miR-541, hsa-miR-541 * hsa-miR-542-3p, hsa-miR-542-5p, hsa-miR-543, hsa-miR-544, hsa-miR-545, hsa-miR-545 * 、hsa-miR-548a-3p、hsa-miR-548a-5p、hsa-miR-548b-3p、hsa-miR-5486-5p、hsa-miR-548c-3p 、hsa-miR-548c-5p、hsa-miR-548d-3p、hsa-miR-548d-5p、hsa-miR-548e、hsa-miR-548f、hsa-m iR-548g, hsa-miR-548h, hsa-miR-548i, hsa-miR-548j, hsa-miR-548k, hsa-miR-5481, hsa-miR-548m, hsa-miR-548n, hsa-miR-548o, hsa-miR-548p, hsa-miR-549, hsa-miR-550, hsa-miR-550 * 、hsa-miR-551a、hsa-miR-551b、hsa-miR-551b *hsa-miR-552, hsa-miR-553, hsa-miR-554, hsa-miR-555, hsa-miR-556-3p, hsa-miR-556-5p, hsa-miR-557, hsa-miR-558, hsa-miR-559, hsa-miR-561, hs a-miR-562, hsa-miR-563, hsa-miR-564, hsa-miR-566, hsa-miR-567, hsa-miR-568, hsa-miR-569, hsa-miR-570, hsa-miR-571, hsa-miR-572, and hsa-miR-573 hsa-miR-574-3p, hsa-miR-574-5p, hsa-miR-575, hsa-miR-576-3p, hsa-miR-576-5p, hsa-miR-577, hsa-miR-578, hsa-miR-579, hsa-miR-580, hsa-miR- 581, hsa-miR-582-3p, hsa-miR-582-5p, hsa-miR-583, hsa-miR-584, hsa-miR-585, hsa-miR-586, hsa-miR-587, hsa-miR-588, hsa-miR-589, hsa-miR-589 * hsa-miR-590-3p, hsa-miR-590-5p, hsa-miR-591, hsa-miR-592, hsa-miR-593, hsa-miR-593 * hsa-miR-595, hsa-miR-596, hsa-miR-597, hsa-miR-598, hsa-miR-599, hsa-miR-600, hsa-miR-601, hsa-miR-602, hsa-miR-603, hsa-miR-604, hsa-miR-605, hsa-miR-606, hs a-miR-607, hsa-miR-608, hsa-miR-609, hsa-miR-610, hsa-miR-611, hsa-miR-612, hsa-miR-613, hsa-miR-614, hsa-miR-615-3p, hsa-miR-615-5p, hsa-miR-616, hsa-miR-616 *hsa-miR-617, hsa-miR-618, hsa-miR-619, hsa-miR-620, hsa-miR-621, hsa-miR-622, hsa-miR-623, hsa-miR-624 * hsa-miR-625 hsa-miR-625 * hsa-miR-626, hsa-miR-627, hsa-miR-628-3p, hsa-miR-628-5p, hsa-miR-629, hsa-miR-629 * hsa-miR-630, hsa-miR-631, hsa-miR-632, hsa-miR-633, hsa-miR-634, hsa-miR-635, hsa-miR-636, hsa-miR-637, hsa-miR-638, hsa-miR-639, hsa-miR-640, hsa-miR-641, hsa-miR-642, hsa-miR-643, hsa-miR-644, hsa-miR-645, hsa-miR-646, hsa-miR-647, hsa-miR-648, hs a-miR-649, hsa-miR-650, hsa-miR-651, hsa-miR-652, hsa-miR-653, hsa-miR-654-3p, hsa-miR-654-5p, hsa-miR-655, hsa-miR-656, hsa-miR-657, hsa-miR-658, hsa-miR-659, hsa-miR-660, hsa-miR-661, hsa-miR-662, hsa-miR-663, hsa-miR-663b, hsa-miR-664, hsa-miR-664 * hsa-miR-665, hsa-miR-668, hsa-miR-671-3p, hsa-miR-671-5p, hsa-miR-675, hsa-miR-7, hsa-miR-708, hsa-miR-708 * hsa-miR-7-1 * hsa-miR-7-2 * hsa-miR-720, hsa-miR-744, hsa-miR-744 *, hsa-miR-758, hsa-miR-760, hsa-miR-765, hsa-miR-766, hsa-miR-767-3p, hsa-miR-767-5p, hsa-miR-768-3p, hsa-miR-768-5p, hsa-miR-769-3p, hsa-miR-769-5p, hsa-miR-770-5p, hsa-miR-802, hsa-miR-873, hsa-miR-874, hsa-miR-875-3p, hsa-miR-875-5p, hsa-miR-876-3p, hsa-miR-876-5p, hsa-miR-877, hsa-miR-877 * , hsa-miR-885-3p, hsa-miR-885-5p, hsa-miR-886-3p, hsa-miR-886-5p, hsa-miR-887, hsa-miR-888, hsa-miR-888 * , hsa-miR-889, hsa-miR-890, hsa-miR-891a, hsa-miR-891b, hsa-miR-892a, hsa-miR-892b, hsa-miR-9, hsa-miR-9 * , hsa-miR-920, hsa-miR-921, hsa-miR-922, hsa-miR-923, hsa-miR-924, hsa-miR-92a, hsa-miR-92a-1 * , hsa-miR-92a-2 * , hsa-miR-92b, hsa-miR-92b * , hsa-miR-93, hsa-miR-93 * , hsa-miR-933, hsa-miR-934, hsa-miR-935, hsa-miR-936, hsa-miR-937, hsa-miR-938, hsa-miR-939, hsa-miR-940, hsa-miR-941, hsa-miR-942, hsa-miR-943, hsa-miR-944, hsa-miR-95, hsa-miR-96, hsa-miR-96 * , hsa-miR-98, hsa-miR-99a, hsa-miR-99a * , hsa-miR-99b and hsa-miR-99b *For example, of interest may be a miRNA that targets chromosome 8 open reading frame 72 (C9orf72), which expresses superoxide dismutase (SOD1) that is associated with amyotrophic lateral sclerosis (ALS).

[0136] MiRNA suppresses the function of its targeted mRNA, and therefore suppresses the expression of the polypeptide encoded by mRNA.Therefore, (partially or entirely) blocking the activity of miRNA (for example, silencing miRNA) can effectively induce or restore the expression of the suppressed polypeptide (making the polypeptide de-suppress).In one embodiment, the de-suppression of the polypeptide encoded by the mRNA target of miRNA is achieved by suppressing the activity of miRNA in the cell by any of a variety of methods.For example, the activity of blocking miRNA can be achieved by hybridizing with small interfering nucleic acids (for example, antisense oligonucleotides, miRNA sponges, TuD RNA) complementary to miRNA or substantially complementary, thus blocking the interaction of miRNA with its target mRNA.As used herein, the small interfering nucleic acids substantially complementary to miRNA are small interfering nucleic acids that can hybridize with miRNA and block the activity of miRNA. In some embodiments, a small interfering nucleic acid that is substantially complementary to a miRNA is a small interfering nucleic acid that is completely complementary to a miRNA except for 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 or 18 bases. "MiRNA inhibitors" are agents that block miRNA function, expression and / or processing. For example, these molecules include, but are not limited to, microRNA-specific antisense molecules, microRNA sponges, strong decoy RNA (TuD RNA) and microRNA oligonucleotides (double-stranded, hairpins, short oligonucleotides) that inhibit the interaction of miRNA with the Drosha complex.

[0137] Still other useful transgenes can include transgenes that encode immunoglobulins that confer passive immunity to pathogens."Immunoglobulin molecules" are proteins containing immunologically active portions of immunoglobulin heavy chains and immunoglobulin light chains that are covalently coupled together and can be combined with antigen specificity.Immunoglobulin molecules can be any type (e.g., IgG, IgE, IgM, IgD, IgA and IgY), classification (e.g., IgG1, IgG2, IgG3, IgG4, IgA1 and IgA2) or subclass.Terms "antibody" and "immunoglobulin" can be used interchangeably in this article.

[0138] An "immunoglobulin heavy chain" is a polypeptide containing at least a portion of an immunoglobulin antigen binding domain and at least a portion of an immunoglobulin heavy chain variable region or at least a portion of an immunoglobulin heavy chain constant region. Therefore, an immunoglobulin-derived heavy chain has significant amino acid sequence homology regions with members of the immunoglobulin gene superfamily. For example, the heavy chain in a Fab fragment is an immunoglobulin-derived heavy chain.

[0139] An "immunoglobulin light chain" is a polypeptide that contains at least a portion of an immunoglobulin antigen binding domain and at least a portion of a variable region or at least a portion of a constant region of an immunoglobulin light chain. Thus, an immunoglobulin-derived light chain has significant regions of amino acid homology with members of the immunoglobulin gene superfamily.

[0140] "Immunoadhesins" are chimeric antibody-like molecules that combine the functional domains of a binding protein (usually a receptor, ligand or cell adhesion molecule) with immunoglobulin constant domains, which typically comprise a hinge and Fc region.

[0141] A "fragment antigen binding (Fab) fragment" is a region on an antibody that binds to an antigen. It is composed of one constant domain and one variable domain of each of the heavy and light chains.

[0142] The antipathogenic constructs are selected based on the causative agent / pathogen of the disease against which protection is sought. These pathogens may be of viral, bacterial or fungal origin and may be used to prevent humans from contracting human diseases, or to prevent veterinary diseases in non-human mammals or other animals.

[0143] rAAV may contain genes encoding antibodies and specifically neutralizing antibodies against viral pathogens. Such antiviral antibodies may include anti-influenza antibodies against one or more of influenza A, influenza B, and influenza C. Type A viruses are the most virulent human pathogens. Serotypes of influenza A associated with epidemics include: H1N1, which caused the Spanish flu in 1918 and the swine flu in 2009; H2N2, which caused the Asian flu in 1957; H3N2, which caused the flu in Hong Kong, China in 1968; H5N1, which caused the avian flu in 2004; H7N7; H1N2; H9N2; H7N2; H7N3; and H10N7. Other target pathogenic viruses include: arenaviruses (including funin, machupo and Lassa), filoviruses (including Marburg and Ebola), hantaviruses, picornoviridae (including rhinoviruses, echoviruses), coronaviruses, paramyxoviruses, measles virus, respiratory syncytial virus, enveloped virus, coxsackievirus, JC virus, parvovirus B19, paravirus Viral hemorrhagic fevers are caused by members of the arenavirus family (Lassa fever), which is also related to lymphocytic choriomeningitis virus (LCM), filoviruses (Ebola virus), and hantaviruses (puremala virus). Members of the picornaviruses (a subfamily of rhinoviruses) are associated with the common cold in humans. The coronavirus family includes a variety of non-human viruses, such as infectious bronchitis virus (poultry), transmissible gastroenteritis virus (swine), porcine hemagglutinin encephalomyelitis virus (swine), feline infectious peritonitis virus (cats), feline enteric coronavirus (cats), and canine coronavirus (dogs). Human respiratory coronaviruses have been postulated to be associated with the common cold, hepatitis non-A, B, or C, and severe acute respiratory syndrome (SARS). Paramyxovirus family Includes parainfluenza virus type 1, parainfluenza virus type 3, bovine parainfluenza virus type 3, mumps virus genus (mumps virus), parainfluenza virus type 2, parainfluenza virus type 4, Newcastle disease virus (chicken), rinderpest, measles virus (including measles and canine distemper) and pneumovirus (including respiratory syncytial virus (RSV). The parvovirus family includes feline parvovirus (feline enteritis), feline panleukopenia virus, canine parvovirus and porcine parvovirus. The adenovirus family includes viruses that cause respiratory diseases (EX, AD7, ARD, OB).Thus, in certain embodiments, the rAAV vectors described herein can be engineered to express anti-Ebola antibodies (e.g., 2G4, 4G7, 13C6), anti-influenza antibodies (e.g., FI6, CR8033), and anti-RSV antibodies (e.g., palivizumab, motavizumab).

[0144] Neutralizing antibody constructs directed against bacterial pathogens can also be selected for use in the present invention. In one embodiment, the neutralizing antibody constructs are directed against the bacteria themselves. In another embodiment, the neutralizing antibody constructs are directed against toxins produced by bacteria. Examples of airborne bacterial pathogens include, e.g., Neisseria meningitidis (meningitis), Klebsiella pneumoniae (pneumonia), Pseudomonas aeruginosa (pneumonia), Pseudomonas pseudomallei (pneumonia), Pseudomonas mallei (pneumonia), Acinetobacter (pneumonia), Moraxella catarrhalis, Moraxella lacunata, Alkaligenes, Cardiobacterium, Haemophilus influenzae (influenza), Haemophilus parainfluenzae, Bordetella pertussis (pertussis), Francisella tularensis (fever), tularensis (pneumonia / fever), Legionella pneumonia (Legionnella disease), Chlamydia psittaci (pneumonia), Chlamydia pneumoniae (pneumonia), Mycobacterium tuberculosis (tuberculosis (TB)), Mycobacterium kansasii (TB), Mycobacterium avium (pneumonia), Nocardia asteroides (pneumonia), Bacillus anthracis (anthrax), Staphylococcus aureus (pneumonia), Streptococcus pyogenes (scarlet fever), Streptococcus pneumoniae (pneumonia), Corynebacteria diphtheria (diphtheria), Mycoplasma pneumoniae (pneumonia).

[0145] rAAV can contain genes encoding antibodies and neutralizing antibodies specifically against bacterial pathogens, such as the pathogen of anthrax, i.e., the toxin produced by Bacillus anthracis. Neutralizing antibodies against protectant (PA), one of the three peptides that form the toxoid, have been described. The other two polypeptides consist of lethal factor (LF) and edema factor (EF). Anti-PA neutralizing antibodies have been described as effective for passive immunization against anthrax. See, e.g., U.S. Patent No. 7,442,373; R. Sawada-Hirai et al., Journal of Immunology-Based Vaccine Therapy (J Immune Based Ther Vaccines) 2004; 2:5. (Online May 12, 2004). Still other anti-anthrax toxin neutralizing antibodies have been described and / or can be produced. Similarly, neutralizing antibodies against other bacteria and / or bacterial toxins can be used to produce AAV-deliverable anti-pathogen constructs as described herein.

[0146] Antibodies against infectious diseases can be caused by parasites or fungi, including, for example, Aspergillus species, Absidia corymbifera, Rhixpus stolonifer, Mucorplumbeaus, Cryptococcus neoformans, Histoplasma capsulatum, Blastomyces dermatitidis, Coccidioides immitis, Penicillium species, Micropolyspora faeni, Thermoactinomyces vulgaris, Alternaria alternate, Cladosporium species, Helminthosporium, and Stachybotrys species.

[0147] rAAV can contain genes encoding antibodies and neutralizing antibodies specifically against pathogenic factors of the following diseases: such as Alzheimer's disease (AD), Parkinson's disease (PD), GBA-associated Parkinson's disease (GBA-PD), rheumatoid arthritis (RA), irritable bowel syndrome (IBS), chronic obstructive pulmonary disease (COPD), cancer, tumors, systemic sclerosis, asthma and other diseases. Such antibodies can be, but are not limited to, for example, α-synuclein, anti-vascular endothelial growth factor (VEGF) (anti-VEGF), anti-VEGFA, anti-PD-1, anti-PDL1, anti-CTLA-4, anti-TNF-α, anti-IL-17, anti-IL-23, anti-IL-21, anti-IL-6, anti-IL-6 receptor, anti-IL-5, anti-IL-7, anti-XII factor, anti-IL-2, anti-HIV, anti-IgE, anti-tumor necrosis factor receptor 1 (TNFR1), anti-notch 2 / 3, anti-notch 1, anti-OX40, anti-erb-b2 receptor tyrosine kinase 3 (ErbB3), anti-ErbB2, anti-β cell maturation antigen, anti-B lymphocyte stimulator, anti-CD20, anti-HER2, anti-granulocyte macrophage colony stimulating factor, anti-oncostatin M (OSM), anti-lymphocyte activation gene 3 (LAG3) protein, anti-CCL20, anti-serum amyloid protein P component (SAP), anti-prolyl hydroxylase inhibitors, anti-CD38, anti-glycoprotein IIb / IIIa, anti-CD52, anti-CD30, anti-IL-1β, anti-epidermal growth factor receptor, anti-CD25, anti-RANK ligand, anti-complement system protein C5, anti-CD11a, anti-CD3 receptor, anti-α-4 (α4) integrin, anti-RSV F protein and anti-integrin α4β7. Still other pathogens and diseases will be apparent to those skilled in the art. Other suitable antibodies may include antibodies useful for treating Alzheimer's disease, such as anti-β-amyloid (e.g., crenezumab, solanezumab, aducanumab), anti-β-amyloid fibrils, anti-β-amyloid plaques, anti-τ, bapineuzamab, and other antibodies. Other suitable antibodies for treating a variety of indications include antibodies described in, for example, PCT / US2016 / 058968, filed on October 27, 2016, disclosed as WO 2017 / 075119A1.

[0148] rAAV vector production

[0149] For the production of AAV viral vectors (e.g., recombinant (r) AAV), the expression cassette can be carried on any suitable vector (e.g., plasmid) for delivery to a packaging host cell. Plasmids useful in the present invention can be engineered to be suitable for in vitro replication and packaging in prokaryotic cells, insect cells, mammalian cells, and other cells. Suitable transfection techniques and packaging host cells are known and / or can be readily designed by one skilled in the art.

[0150] Methods for producing and isolating AAV suitable for use as a vector are known in the art. See, for example, Grieger and Samulski, 2005, "Adeno-associated virus as a gene therapy vector: Vector development, production and clinical applications", Adv. Biochem. Engin / Biotechnol 99: 119-145; Buning et al., 2008, "Recent developments in adeno-associated virus vector technology," J. Gene Med 10: 717-733; and the references cited below, each of which is incorporated herein by reference in its entirety. In order to package the transgene into the virion, ITR is the only AAV component in cis required in the construct identical to the nucleic acid molecule containing the expression cassette. The cap and rep genes can be supplied in trans.

[0151] In one embodiment, the expression cassette described herein is engineered to a gene element (e.g., a shuttle plasmid), and the immunoglobulin construct sequence carried thereon is transferred to a packaging host cell to produce a viral vector. In one embodiment, the selected gene element can be delivered to an AAV packaging cell by any suitable method, including transfection, electroporation, liposome delivery, membrane fusion technology, high-speed DNA-coated pellets, viral infection, and protoplast fusion. Stable AAV packaging cells can also be prepared. Alternatively, the expression cassette can be used to produce a viral vector other than AAV, or a mixture for producing antibodies in vitro. The method for preparing such a construct is known to nucleic acid manipulation technicians and includes genetic engineering, recombinant engineering, and synthetic techniques. See, for example, "Molecular Cloning: A Laboratory Manual," edited by Green and Sambrook, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York (2012).

[0152] The term "AAV intermediate" or "AAV vector intermediate" refers to an assembled rAAV capsid that lacks the desired genomic sequence packaged therein. These may also be referred to as "empty" capsids. Such capsids may contain no detectable genomic sequence of the expression cassette, or contain only partially packaged genomic sequence that is insufficient to achieve expression of the gene product. These empty capsids are non-functional to transfer the gene of interest to a host cell.

[0153] Recombinant adeno-associated viruses (AAV) described herein can be produced using known techniques. See, for example, WO2003 / 042397; WO 2005 / 033321, WO 2006 / 110689; US 7588772 B2. Such methods involve culturing host cells containing nucleic acid sequences encoding AAV capsid proteins; functional rep genes; expression cassettes consisting of at least AAV inverted terminal repeats (ITRs) and transgenes; and sufficient auxiliary functions to allow the expression cassettes to be packaged into AAV capsid proteins. Methods for producing capsids, their coding sequences, and methods for producing rAAV viral vectors have been described. See, for example, Gao et al., Proceedings of the National Academy of Sciences of the United States of America 100(10):6081-6086(2003) and US2013 / 0045186A1.

[0154] In one embodiment, a production cell culture that can be used to produce recombinant AAV is provided. This cell culture contains nucleic acids expressing AAV capsid proteins in host cells; nucleic acid molecules suitable for packaging into AAV capsids, such as vector genomes containing AAV ITR and non-AAV nucleic acid sequences encoding gene products, wherein the non-AAV nucleic acid sequences are operably connected to sequences for guiding product expression in host cells; and sufficient AAVrep functions and adenovirus helper functions to allow nucleic acid molecules to be packaged into recombinant AAV capsids. In one embodiment, the cell culture is composed of mammalian cells (e.g., human embryonic kidney 293 cells and other cells) or insect cells (e.g., baculovirus).

[0155] Optionally, the rep function is provided by an AAV other than the AAV that provides the capsid. For example, rep can be, but is not limited to, an AAV1 rep protein, an AAV2 rep protein, an AAV3 rep protein, an AAV4 rep protein, an AAV5 rep protein, an AAV6 rep protein, an AAV7 rep protein, an AAV8 rep protein; or rep 78, rep 68, rep 52, rep 40, rep68 / 78, and rep40 / 52; or a fragment thereof; or another source. Optionally, the rep and cap sequences are located on the same genetic element in cell culture. There may be a spacer between the rep sequence and the cap gene. Any of these AAV or mutant AAV capsid sequences may be under the control of an exogenous regulatory control sequence that directs its expression in a host cell.

[0156] In one embodiment, cells are manufactured in suitable cell culture (e.g., HEK 293) cells. The method for manufacturing the gene therapy vector described herein includes methods well known in the art, such as producing plasmid DNA, producing vectors and purifying vectors for producing gene therapy vectors. In certain embodiments, the gene therapy vector is an AAV vector, and the plasmid produced is an AAV cis plasmid encoding the AAV genome and the gene of interest, an AAV trans plasmid containing AAV rep and cap genes, and an adenovirus helper plasmid. The vector production process may include method steps, such as starting cell culture, performing cell passage, inoculating cells, transfecting cells with plasmid DNA, exchanging the culture medium after transfection with serum-free culture medium, and collecting cells and culture medium containing the vector. The collected cells and culture medium containing the vector are referred to as crude cell collections herein. In another system, the gene therapy vector is introduced into insect cells by infecting with a baculovirus-based vector. For a review of these production systems, see generally, e.g., Zhang et al., 2009, "Adenovirus-adeno-associated virus hybrid for large-scale recombinant adeno-associated virus production," Human Gene Therapy 20:922-929, the contents of each of which are incorporated herein by reference in their entirety. Methods for making and using these and other AAV production systems are also described in the following U.S. Patents, the contents of each of which are incorporated herein by reference in their entirety: 5,139,941; 5,741,683; 6,057,152; 6,204,059; 6,268,213; 6,491,907; 6,660,514; 6,951,753; 7,094,604; 7,172,893; 7,201,898; 7,229,823; and 7,439,065.

[0157] Thereafter, the crude cell collection can be subject to the method steps of the present invention, such as concentrating the vector collection, diafiltering the vector collection, microfluidizing the vector collection, nuclease digesting the vector collection, filtering the microfluidized intermediate, crude purification by chromatography, crude purification by ultracentrifugation, buffer exchange by tangential flow filtration, and / or formulation and filtration to prepare large quantities of vector.

[0158] Two-step affinity chromatography purification is performed at high salt concentrations, followed by anion exchange resin chromatography to purify the vector drug product and remove empty capsids. These methods are described in more detail in International Patent Application No. PCT / US2016 / 065970, filed December 9, 2016, and priority documents thereof, U.S. Patent Application No. 62 / 322,071, filed April 13, 2016, and No. 62 / 226,357, filed December 11, 2015, entitled "Scalable Purification Method for AAV9," which are incorporated herein by reference. For methods of purifying AAV8, see International Patent Application No. PCT / US2016 / 065976 filed on December 9, 2016 and its priority documents, U.S. Patent Application No. 62 / 322,098 filed on April 13, 2016, and No. 62 / 266,341 filed on December 11, 2015, and for methods of purifying rh10, see International Patent Application No. PCT / US16 / 66013 filed on December 9, 2016 and its priority documents, U.S. Patent Application No. 62 / 322,055 filed on April 13, 2016, and also filed on December 11, 2015, entitled “Scalable Purification Method for AAVrh10”. For methods of purifying AAV1, see International Patent Application No. PCT / US2016 / 065974 filed on December 9, 2016 and its priority documents, U.S. Patent Application No. 62 / 322,083 filed on April 13, 2016, and No. 62 / 26,351 filed on December 11, 2015, entitled “Scalable Purification Method for AAV1,” all of which are incorporated herein by reference.

[0159] To calculate the content of empty and intact particles, the VP3 band volume of the selected sample (e.g., in the examples herein, a preparation purified by an iodixanol gradient, where GC#=particle#) is plotted relative to the loaded GC particles. The resulting linear equation (y=mx+c) is used to calculate the number of particles in the band volume of the test article peak. The number of particles per 20 μL loaded (pt) is then multiplied by 50 to obtain particles (pt) / mL. Pt / mL is divided by GC / mL to obtain the ratio of particles to genome copies (pt / GC). Pt / mL-GC / mL gives empty pt / mL. Empty pt / mL is divided by pt / mL and ×100 to give the percentage of empty particles.

[0160] In general, methods for determining empty capsids and AAV vector particles with packaged genomes are known in the art. See, for example, Grimm et al., Gene Therapy (1999) 6: 1322-1330; Sommer et al., Molec. Ther. (2003) 7: 122-128. To test denatured capsids, the method comprises subjecting the treated AAV stock to SDS-polyacrylamide gel electrophoresis (composed of any gel capable of separating the three capsid proteins, such as a gradient gel containing 3-8% triacetate in a buffer), then running the gel until the sample material is separated, and blotting the gel onto a nylon or nitrocellulose membrane (preferably nylon). Then, an anti-AAV capsid antibody is used as a primary antibody that binds to the denatured capsid protein, preferably an anti-AAV capsid monoclonal antibody, most preferably a B1 anti-AAV2 monoclonal antibody (Wobus et al., Journal of Virology (2000) 74: 9281-9293). Then use secondary antibody, described secondary antibody is combined with primary antibody and contains a kind of device for detecting the combination with primary antibody, more preferably contains the anti-IgG antibody of the detection molecule covalently bound thereto, most preferably is the sheep anti-mouse IgG antibody covalently linked to horseradish peroxidase.A kind of method for detecting combination is used to semi-quantitatively determine the combination between primary antibody and secondary antibody, preferably can detect the detection method of radioisotope emission, electromagnetic radiation or colorimetric change, most preferably is chemiluminescent detection kit.For example, for SDS-PAGE, sample can be extracted from column fraction and heated in SDS-PAGE loading buffer containing reducing agent (for example, DTT), and capsid protein is resolved on prefabricated gradient polyacrylamide gel (for example, Novex).Silver staining can be carried out according to the manufacturer's instructions using SilverXpress (Invitrogen, California) or other suitable staining methods (i.e. SYPRO ruby ​​or Coomassie staining).In one embodiment, the concentration of AAV vector genome (vg) in column fraction can be measured by quantitative real-time PCR (Q-PCR). The sample is diluted and digested with DNase I (or another suitable nuclease) to remove exogenous DNA. After nuclease inactivation, primers and TaqMan TMThe fluorescent probe is further diluted and amplified. The number of cycles (threshold cycles, Ct) required for each sample to reach a defined fluorescence level is measured on an Applied Biosystems Prism 7700 sequence detection system. Plasmid DNA containing the same sequence as that contained in the AAV vector is used to generate a standard curve in a Q-PCR reaction. The value of the cycle threshold (Ct) obtained from the sample is used to determine the vector genome titer by normalizing it relative to the Ct value of the plasmid standard curve. Endpoint determination based on digital PCR can also be used.

[0161] On the one hand, an optimized q-PCR method is used, which utilizes a broad spectrum serine protease, such as proteinase K (such as commercially available from Qiagen). More specifically, the optimized qPCR genomic titer assay is similar to the standard assay, except that after DNase I digestion, the sample is diluted with proteinase K buffer and treated with proteinase K, and then heat inactivated. Suitably, the sample is diluted with proteinase K buffer in an amount equal to the sample size. The proteinase K buffer can be concentrated 2 times or more. Typically, proteinase K treatment is about 0.2 mg / mL, but can vary between 0.1 g / mL and about 1 mg / mL. The treatment step is usually carried out at about 55°C for about 15 minutes, but can be carried out at a lower temperature (e.g., about 37°C to about 50°C) for a longer period of time (e.g., about 20 minutes to about 30 minutes), or at a higher temperature (e.g., up to about 60°C) for a shorter period of time (e.g., about 5 to 10 minutes). Similarly, heat inactivation is typically performed at about 95°C for about 15 minutes, but the temperature can be lowered (e.g., from about 70°C to about 90°C) and the time can be extended (e.g., from about 20 minutes to about 30 minutes). The sample is then diluted (e.g., 1000-fold) and subjected to TaqMan analysis as described in the standard assay.

[0162] Additionally or alternatively, droplet digital PCR (ddPCR) can be used. For example, methods for determining the titer of single-stranded and self-complementary AAV vector genomes by ddPCR have been described. See, for example, M. Lock et al., Human Gene Therapy Methods 2014 Apr; 25(2): 115-25. doi: 10.1089 / hgtb.2013.131. Epub 2014 Feb 14.

[0163] In short, the method for separating rAAV particles with packaged genomic sequences from genome-defective AAV intermediates involves subjecting a suspension comprising recombinant AAV virus particles and AAV capsid protein intermediates to high performance liquid chromatography, wherein the AAV virus particles and AAV intermediates are bound to a strong anion exchange resin balanced at high pH and subjected to a salt gradient while monitoring the UV absorbance of the eluent at about 260 and about 280. The pH can be adjusted based on the selected AAV. See, for example, WO2017 / 160360 (AAV9), WO2017 / 100704 (AAVrh10), WO 2017 / 100676 (e.g., AAV8) and WO2017 / 100674 (AAV1), which are incorporated herein by reference. In this method, when the ratio of A260 / A280 reaches an inflection point, the AAV complete capsid is collected from the eluted fraction. In one example, for the affinity chromatography step, the diafiltered product can be applied to a Capture Select PCR system that efficiently captures AAV2 serotypes. TM Poros-AAV2 / 9 affinity resin (Life Technologies). Under these ionic conditions, a significant percentage of residual cellular DNA and proteins flowed through the column, while AAV particles were efficiently captured.

[0164] Composition and use

[0165] Provided herein is a composition comprising at least one rAAV stock solution (e.g., rAAV stock solution or mutant rAAV stock solution) and optional carriers, excipients and / or preservatives. The rAAV stock solution refers to a plurality of rAAV vectors, the amount of the plurality of rAAV vectors being the same as described, for example, in the discussion below about concentrations and dosage units.

[0166] As used herein, "carrier" includes any and all solvents, dispersion media, vehicles, coatings, diluents, antibacterial and antifungal agents, isotonic and absorption delaying agents, buffers, carrier solutions, suspensions, colloids, etc. The use of such media and agents for pharmaceutically active substances is well known in the art. Supplementary active ingredients can also be incorporated into the composition. The phrase "pharmaceutically acceptable" refers to molecular entities and compositions that do not produce allergic or similar adverse reactions when administered to a host. Delivery vehicles (such as liposomes, nanocapsules, microparticles, microspheres, lipid particles, vesicles, etc.) can be used to introduce the compositions of the present invention into suitable host cells. In particular, the transgene delivered by the rAAV vector can be formulated for delivery, and the transgene is encapsulated in lipid particles, liposomes, vesicles, nanospheres or nanoparticles, etc.

[0167] In one embodiment, the composition comprises a final formulation suitable for delivery to a subject, such as an aqueous liquid suspension buffered to a physiologically compatible pH and salt concentration. Optionally, one or more surfactants are present in the formulation. In another embodiment, the composition can be transported as a concentrate that is diluted to be administered to a subject. In other embodiments, the composition can be lyophilized and reconstituted at the time of administration.

[0168] A suitable surfactant or combination of surfactants can be selected from non-toxic nonionic surfactants. In one embodiment, a difunctional block copolymer surfactant terminated with a primary hydroxyl group is selected, such as F68 [BASF], also known as Poloxamer 188, has a neutral pH and an average molecular weight of 8400. Other surfactants and other poloxamers may be selected, i.e., nonionic triblock copolymers consisting of a central hydrophobic chain of polyoxypropylene (poly(propylene oxide)) flanked by two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)), SOLUTOL HS15 (polyethylene glycol-15 hydroxystearate), LABRASOL (polyoxycaprylylglyceride), polyoxyl 10 oleyl ether, TWEEN (polyoxyethylene sorbitan fatty acid esters), ethanol, and polyethylene glycol. In one embodiment, the formulation contains a poloxamer. These copolymers are usually named with the letter "P" (for poloxamer), followed by three numbers: the first two digits x 100 give the approximate molecular weight of the polyoxypropylene core, and the last digit x 10 gives the percentage of the polyoxyethylene content. In one embodiment, poloxamer 188 is selected. The surfactant may be present in an amount up to about 0.0005% to about 0.001% of the suspension.

[0169] The vector is administered in sufficient amounts to transfect cells and provide sufficient levels of gene transfer and expression to provide therapeutic benefit without excessive side effects or with medically acceptable physiological effects, which can be determined by a technician in the medical field. Conventional and pharmaceutically acceptable routes of administration include but are not limited to direct delivery to a desired organ (e.g., liver (optionally via the hepatic artery), lung, heart, eye, kidney), oral, inhaled, intranasal, intrathecal, intratracheal, intraarterial, intraocular, intravenous, intramuscular, subcutaneous, intradermal, and other parent routes of administration. If desired, routes of administration can be combined.

[0170] The dosage of the viral vector will depend primarily on factors such as the condition being treated, the age, weight and health of the patient, and therefore may vary between patients. For example, a therapeutically effective human dose of a viral vector is typically in the range of about 25 to about 1000 microliters to about 100 mL of a solution containing a concentration of about 1×10 9 to 1×1016 Genomic viral vectors. The dosage will be adjusted to balance the therapeutic benefit with any side effects, and such dosage may vary depending on the therapeutic application for which the recombinant vector is employed. The expression level of the transgene may be monitored to determine the dosage frequency of the resulting viral vector, preferably an AAV vector containing a minigene. Optionally, a dosage regimen similar to that described for therapeutic purposes may be used for immunization using the compositions of the invention.

[0171] The replication-defective virus composition can be formulated in dosage units to contain about 1.0×10 9 GC to about 1.0×10 16 GC is in the range of 1.0×10 GC (to treat subjects with an average weight of 70 kg), including all integers and fractions within the range, and for human patients, is preferably 1.0×10 12 GC to 1.0×10 14 GC. In one embodiment, the composition is formulated to contain at least 1×10 9 , 2×10 9 , 3×10 9 , 4×10 9 , 5×10 9 , 6×10 9 ,7×10 9 , 8×10 9 or 9×10 9 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×10 10 , 2×10 10 , 3×10 10 , 4×10 10 , 5×10 10 , 6×10 10 ,7×10 10 , 8×10 10 or 9×10 10 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×10 11 , 2×10 11 , 3×10 11 , 4×10 11 , 5×10 11 , 6×10 11 ,7×10 11 , 8×10 11 or 9×10 11 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×1012 , 2×10 12 , 3×10 12 , 4×10 12 , 5×10 12 , 6×10 12 ,7×10 12 , 8×10 12 or 9×10 12 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×10 13 , 2×10 13 , 3×10 13 , 4×10 13 , 5×10 13 , 6×10 13 ,7×10 13 , 8×10 13 or 9×10 13 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×10 14 , 2×10 14 , 3×10 14 , 4×10 14 , 5×10 14 , 6×10 14 ,7×10 14 , 8×10 14 or 9×10 14 GC, including all integers and fractions within the range. In another embodiment, the composition is formulated to contain at least 1×10 15 , 2×10 15 , 3×10 15 , 4×10 15 , 5×10 15 , 6×10 15 ,7×10 15 , 8×10 15 or 9×10 15 GC, including all integers and fractions within the range. In one embodiment, for human use, the dosage range may be 1×10 10 To about 1×10 12 GC, including all integer or fractional quantities within the stated range.

[0172] These above-mentioned doses can be administered with various volumes of carrier, excipient or buffer formulations, ranging from about 25 to about 1000 microliters or more, including all numbers within the range, depending on the size of the area to be treated, the viral titer used, the route of administration, and the desired effect of the method. In one embodiment, the volume of the carrier, excipient or buffer is at least about 25 μL. In one embodiment, the volume is about 50 μL. In another embodiment, the volume is about 75 μL. In another embodiment, the volume is about 100 μL. In another embodiment, the volume is about 125 μL. In another embodiment, the volume is about 150 μL. In another embodiment, the volume is about 175 μL. In another embodiment, the volume is about 200 μL. In another embodiment, the volume is about 225 μL. In another embodiment, the volume is about 250 μL. In another embodiment, the volume is about 275 μL. In another embodiment, the volume is about 300 μL. In another embodiment, the volume is about 325 μL. In another embodiment, the volume is about 350 μL. In another embodiment, the volume is about 375 μL. In another embodiment, the volume is about 400 μL. In another embodiment, the volume is about 450 μL. In another embodiment, the volume is about 500 μL. In another embodiment, the volume is about 550 μL. In another embodiment, the volume is about 600 μL. In another embodiment, the volume is about 650 μL. In another embodiment, the volume is about 700 μL. In another embodiment, the volume is between about 700 and about 1000 μL.

[0173] In certain embodiments, the dose may be about 1×10 9 GC / g brain mass to about 1×10 12 GC / g brain mass. In certain embodiments, the dose may be in the range of about 3×10 10 GC / g brain mass to about 3×10 11 GC / g brain mass. In certain embodiments, the dose may be in the range of about 5×10 10 GC / g brain mass to about 1.85×10 11 GC / g brain mass.

[0174] In one embodiment, the viral construct can be expressed in an amount of at least about 1×10 9 GC to about 1×10 15 or about 1×10 11 to 5×10 13Dose delivery of GC. Suitable volumes for delivering these dosages and concentrations can be determined by those skilled in the art. For example, a volume of about 1 μ L to 150 mL can be selected, wherein for adults, a larger volume is selected. Typically, for newborn infants, a suitable volume is about 0.5 mL to about 10 mL, and for older infants, a volume of about 0.5 mL to about 15 mL can be selected. For young children, a volume of about 0.5 mL to about 20 mL can be selected. For children, a volume of about 30 mL can be selected. For prepubertal teenagers and adolescents, a volume of about 50 mL can be selected. In still other embodiments, the volume that the patient can accept for intrathecal administration is selected to be about 5 mL to about 15 mL or about 7.5 mL to about 10 mL. Other suitable volumes and dosages can be determined. The dosage will be adjusted to balance the therapeutic benefit with any side effects, and this dosage can vary according to the therapeutic application using the recombinant vector.

[0175] The recombinant vector described above can be delivered to a host cell according to the disclosed method. The rAAV preferably suspended in a physiologically compatible carrier can be administered to a human or non-human mammal patient. In certain embodiments, for administration to a human patient, the rAAV is appropriately suspended in an aqueous solution containing saline, a surfactant, and a physiologically compatible salt or a mixture of salts. Suitably, the formulation is adjusted to a physiologically acceptable pH, for example, in the range of pH 6 to 9, or pH 6.5 to 7.5, pH 7.0 to 7.7, or pH 7.2 to 7.8. Since the pH of cerebrospinal fluid is about 7.28 to about 7.32, a pH within this range may be desired for intrathecal delivery; and for intravenous delivery, a pH of about 6.8 to about 7.2 may be desired. However, other pHs within the widest range and these sub-ranges may be selected for other delivery routes.

[0176] In another embodiment, the composition comprises a carrier, a diluent, an excipient and / or an adjuvant. In view of the indication for which the transfer virus is intended, a person skilled in the art can easily select a suitable carrier. For example, a suitable carrier comprises saline, which can be formulated with a variety of buffer solutions (e.g., phosphate-buffered saline). Other exemplary carriers include sterile saline, lactose, sucrose, calcium phosphate, gelatin, dextran, agar, pectin, peanut oil, sesame oil and water. The buffer / carrier should include components that prevent rAAV from adhering to the infusion tube but do not interfere with the in vivo binding activity of rAAV. Suitable surfactants or combinations of surfactants can be selected from non-toxic nonionic surfactants. In one embodiment, a bifunctional block copolymer surfactant terminated in a primary hydroxyl group is selected, such as F68 [BASF], also known as Poloxamer 188, has a neutral pH and an average molecular weight of 8400. Other surfactants and other poloxamers may be selected, i.e., nonionic triblock copolymers consisting of a central hydrophobic chain of polyoxypropylene (poly(propylene oxide)) flanked by two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)), SOLUTOL HS15 (polyethylene glycol-15 hydroxystearate), LABRASOL (polyoxycaprylyl glyceride), polyoxyl oil ethers, TWEEN (polyoxyethylene sorbitan fatty acid esters), ethanol, and polyethylene glycol. In one embodiment, the formulation contains a poloxamer. These copolymers are usually named with the letter "P" (for poloxamer), followed by three numbers: the first two digits x 100 give the approximate molecular weight of the polyoxypropylene core, and the last digit x 10 gives the percentage of the polyoxyethylene content. In one embodiment, poloxamer 188 is selected. The surfactant may be present in an amount of up to about 0.0005% to about 0.001% of the suspension. In one example, the formulation may contain, for example, a buffered saline solution comprising one or more of sodium chloride, sodium bicarbonate, dextran, magnesium sulfate (e.g., magnesium sulfate·7H2O), potassium chloride, calcium chloride (e.g., calcium chloride·2H2O), disodium hydrogen phosphate, and mixtures thereof in water. Suitably, for intrathecal delivery, the osmolarity is within a range compatible with cerebrospinal fluid (e.g., about 275 to about 290); see, e.g., emedicine.medscape.com / article / 2093316-overview. Optionally, for intrathecal delivery, a commercially available diluent may be used as a suspending agent, or in combination with another suspending agent and other optional excipients. See, e.g., Elliotts Solution [Lukare Medical]. In other embodiments, the formulation may contain one or more penetration enhancers. Examples of suitable penetration enhancers may include, for example, mannitol, sodium glycocholate, sodium taurocholate, sodium deoxycholate, sodium salicylate, sodium caprylate, sodium caprate, sodium lauryl sulfate, polyoxyethylene-9-lauryl ether, or EDTA.

[0177] Optionally, the compositions of the invention may contain other conventional pharmaceutical ingredients in addition to rAAV and one or more carriers, such as preservatives or chemical stabilizers. Suitable exemplary preservatives include chlorobutanol, potassium sorbate, sorbic acid, sulfur dioxide, propyl gallate, methyl paraben, ethyl vanillin, glycerol, phenol, and p-chlorophenol. Suitable chemical stabilizers include gelatin and albumin.

[0178] The composition according to the invention may include a pharmaceutically acceptable carrier, as defined above. Suitably, the composition described herein includes an effective amount of one or more AAVs suspended in a pharmaceutically suitable carrier and / or mixed with a suitable excipient designed to be delivered to a subject by injection, an osmotic pump, an intrathecal catheter, or by another device or route. In one example, the composition is formulated for intrathecal delivery.

[0179] As used herein, the term "intrathecal delivery" or "intrathecal administration" refers to a route of drug administration by injection into the spinal canal, more specifically into the subarachnoid space so that it reaches the cerebrospinal fluid (CSF). Intrathecal delivery can include lumbar puncture, intraventricular (including intracerebroventricular (ICV)), suboccipital / intracisternal and / or C1-2 puncture. For example, material can be introduced by lumbar puncture to diffuse throughout the subarachnoid space. In another example, injection can be made into the cisterna magna.

[0180] As used herein, the term "intracisternal delivery" or "intracisternal administration" refers to administration of a drug directly into the cerebrospinal fluid of the cisterna magna cerebellomedularis, more specifically by suboccipital puncture or by injection directly into the cisterna magna or by a permanently placed cannula.

[0181] On the one hand, the carrier provided herein can be administered intrathecally by the method and / or device. See, for example, WO2017 / 181113, which is incorporated herein by reference. Alternatively, other devices and methods can be selected. The method comprises the following steps: advancing the spinal needle into the patient's cisterna magna; a certain length of flexible tube is connected to the proximal hub of the spinal needle, and the output port of the valve is connected to the proximal end of the flexible tube, and after the advancement and connection steps are performed, and after allowing the tube to use the patient's cerebrospinal fluid for self-priming, a first container containing a certain amount of isotonic solution is connected to the flushing inlet of the valve, and then a second container containing a certain amount of pharmaceutical composition is connected to the carrier inlet of the valve. After the first blood vessel and the second blood vessel are connected to the valve, a fluid flow path is opened between the carrier inlet and outlet of the valve, and the pharmaceutical composition is injected into the patient by the spinal needle, and after the injection of the pharmaceutical composition, a fluid flow path is opened by the flushing inlet and outlet of the valve, and an isotonic solution is injected into the spinal needle to flush the pharmaceutical composition into the patient.

[0182] This method and this device can each optionally be used for intrathecal delivery of the compositions provided herein. Alternatively, other methods and devices can be used for such intrathecal delivery.

[0183] It should be noted that the term "a" or "an" refers to one or more. Thus, the terms "a or an", "one or more" and "at least one" are used interchangeably herein.

[0184] The words "comprise", "comprises" and "comprising" are to be interpreted as inclusive rather than exclusive. The words "consist", "consisting" and variations thereof are to be interpreted as exclusive rather than inclusive. Although various embodiments in the specification are presented using "comprising" language, in other cases, it is also intended to use the language of "consisting of" or "consisting essentially of" to explain and describe the relevant embodiments.

[0185] As used herein, unless otherwise indicated, the term "about" means a variability of 10% (± 10%) relative to a given reference.

[0186] As used herein, "disease," "disorder," and "condition" are used interchangeably to refer to an abnormal state of a subject.

[0187] Unless otherwise defined in this specification, technical and scientific terms used herein have the same meanings as commonly understood by one of ordinary skill in the art and with reference to the public texts, which provide those skilled in the art with a general guide to many of the terms used in this application.

[0188] The term "expression" is used in the broadest sense herein and includes the production of RNA or RNA and protein. With respect to RNA, the term "expression" or "translation" particularly relates to the production of peptides or proteins. Expression may be temporary or may be stable.

[0189] As used herein, the term "NAb titer" is a measure of how many neutralizing antibodies (e.g., anti-AAV Nabs) are produced that neutralize the physiological effects of the epitope (e.g., AAV) to which they are targeted. Anti-AAV NAb titers can be measured as described in Calcedo, R et al., "Worldwide Epidemiology of Neutralizing Antibodies to Adeno-Associated Viruses," Journal of Infectious Diseases, 2009. 199(3): pp. 381-390, which is incorporated herein by reference.

[0190] As used herein, "expression cassette" refers to a nucleic acid molecule comprising a coding sequence, a promoter and other regulatory sequences thereof, which can be delivered to a packaging host cell via a genetic element (e.g., a plasmid) and packaged into the capsid of a viral vector (e.g., a viral particle). Typically, such an expression cassette for producing a viral vector contains the coding sequence of a gene product described herein, flanked by a packaging signal and other expression control sequences of the viral genome, such as the sequences described herein.

[0191] The abbreviation "sc" refers to self-complementary. "Self-complementary AAV" refers to a construct in which the coding region carried by the recombinant AAV nucleic acid sequence has been designed to form an intramolecular double-stranded DNA template. After infection, instead of waiting for the cell-mediated second strand synthesis, two complementary half-scAAVs will associate to form a double-stranded DNA (dsDNA) that is easy to replicate and transcribe immediately. See, for example, DM McCarty et al., "Self-complementary recombinant adeno-associated virus (scAAV) vectors promote efficient transduction independently of DNA synthesis", "Gene Therapy", (August 2001), Vol. 8, No. 16, pp. 1248-1254. Self-complementary AAVs are described, e.g., in U.S. Pat. Nos. 6,596,535; 7,125,717; 7,456,683, each of which is incorporated herein by reference in its entirety.

[0192] As used herein, the term "operably linked" refers to both expression control sequences that are contiguous to a gene of interest and expression control sequences that function in trans or at a distance to control the gene of interest.

[0193] The term "heterologous" when used in conjunction with a protein or nucleic acid indicates that the protein or nucleic acid comprises two or more sequences or subsequences that are not found in the same relationship to each other in nature. For example, nucleic acids are typically produced recombinantly, having two or more sequences from unrelated genes arranged to produce a new functional nucleic acid. For example, in one embodiment, a nucleic acid has a promoter from one gene arranged to direct expression of a coding sequence from a different gene. Thus, the promoter is heterologous with respect to the coding sequence.

[0194] "Replication-defective virus" or "viral vector" refers to a synthetic or artificial virus particle in which an expression cassette containing a gene of interest is packaged in a viral capsid or envelope, wherein any viral genomic sequence also packaged in the viral capsid or envelope is replication-defective; that is, it cannot produce progeny virions, but retains the ability to infect target cells. In one embodiment, the genome of the viral vector does not contain genes encoding enzymes required for replication (the genome can be engineered to be "gutless"-containing only the transgene of interest, which is flanked by the signals required for amplification and packaging of the artificial genome), but these genes can be supplied during production. Therefore, this is considered safe for gene therapy because replication and infection by progeny virions will not occur unless the viral enzymes required for replication are present.

[0195] In many cases, rAAV particles are referred to as DNase resistant. However, in addition to this endonuclease (DNase), other endonucleases and exonucleases can also be used in the purification steps described herein to remove contaminating nucleic acids. Such nucleases can be selected to degrade single-stranded DNA and / or double-stranded DNA and RNA. Such steps can contain a single nuclease or a mixture of nucleases for different targets, and can be an endonuclease or an exonuclease.

[0196] The term "nuclease-resistant" indicates that the AAV capsid has been fully assembled around an expression cassette designed to deliver the transgene to the host cell and protect these packaged genomic sequences from degradation (digestion) during nuclease incubation steps designed to remove contaminating nucleic acids that may be present during the production process.

[0197] In the context of the present invention, the term "translation" relates to the ribosome process in which an mRNA chain controls the assembly of an amino acid sequence to produce a protein or peptide.

[0198] As used throughout this specification and claims, the terms "comprises" and "comprising" include other components, elements, integers, steps, etc. In contrast, the term "consisting of" and its variations do not include other components, elements, integers, steps, etc.

[0199] As described above, unless otherwise stated, the term "about" when used to modify a numerical value means a variation of ±10%.

[0200] The following examples are illustrative only and are not intended to limit the invention.

[0201] Examples

[0202] The following examples report extensive deamidation of AAV8 and 7 additional different AAV serotypes with supporting evidence from structural, biochemical, and mass spectrometric approaches. The extent of deamidation at each site depends on the age of the vector as well as multiple primary sequence and 3D structural factors, but is largely independent of the conditions of vector recovery and purification. It was demonstrated that deamidation may affect vector transduction activity and that early time point loss of vector activity was associated with rapidly progressive spontaneous deamidation at several AAV8 asparagines. Mutation strategies to stabilize side chain amides were explored, thereby improving vector transduction and reducing batch-to-batch molecular variability, which is a critical issue in bioproduction. This study demonstrates previously unknown aspects of AAV capsid heterogeneity and highlights its importance in the development of these vectors for gene therapy.

[0203] Example 1 below provides a characterization of post-translational modifications of the AAV8 vector capsid by one-dimensional and two-dimensional gel electrophoresis, mass spectrometry, and de novo structural modeling. After identifying multiple putative deamidation sites on the capsid surface, their effects on capsid structure and function were evaluated both in vivo and in vitro. Example 1 further extends this analysis to AAV9 to determine whether this phenomenon applies to serotypes other than AAV8, thereby confirming that deamidation of the AAV capsid is not serotype specific. Examples 2 and 3 demonstrate deamidation in additional AAVs.

[0204] Example 4 relates to novel epitopes mapped on the AAV9 capsid.

[0205] Example 1: Deamidation of amino acids on the surface of adeno-associated virus capsid

[0206] A. Materials and Methods

[0207] 1.1D and 2D gel electrophoresis

[0208] For 1D SDS polyacrylamide gel electrophoresis (SDS-PAGE) analysis, the AAV vectors were first denatured at 80°C for 20 minutes in the presence of lithium dodecyl sulfate and a reducing agent. They were then run on a 4-12% Bis-Tris gel at 200 V for 90 minutes and stained with Coomassie blue. Figure 1A-Figure 1D 2D gel electrophoresis was performed by Kendrick Laboratories, Inc. (Madison, WI) for the data in the literature. For subsequent experiments, 2D SDS-PAGE was performed in-house. For this purpose, 3 × 10 11The AAV vector of GC was combined with 500U turbonuclease marker (Accelagen, San Diego, CA) in 150 μL of phosphate buffered saline (PBS) containing 35 mM NaCl and 1 mM MgCl2, and incubated at 37°C for ten minutes. Next, nine volumes of absolute ethanol were added, the sample was vortexed, and then incubated at -80°C for at least two hours, then incubated on ice for five minutes, and then centrifuged at maximum speed for 30 minutes at 15°C. The supernatant was poured out and the precipitate was air-dried, and then the precipitate was resuspended in resuspension buffer #1 [ddH2O containing 0.15% SDS, 50 mM dithiothreitol (DTT), 10 mM Tris pH 7.5 and 1 μL pH 6-9 ampholytes (ZM0023 of Thermo Fisher Scientific), added once a day] and incubated undisturbed at room temperature. After 30 minutes, the sample tubes were flicked to mix, 1 μg of chicken conalbumin marker (Sigma Aldrich, St. Louis, MO) was added, and the samples were incubated at 37°C for 30 minutes, flicked to mix for 15 minutes. The samples were then transferred to 50°C for 15-20 minutes, vortexed, incubated at 95°C for 2.5 minutes, and allowed to cool before centrifugation at maximum speed for one minute, and vortexed briefly. 10 μL of each sample was then mixed with 140 μL of resuspension buffer #2 (ddH2O containing 9.7 M urea, 2% CHAPS, 0.002% bromophenol blue, and 0.05% ampholytes, as described above, added once a day) and incubated at room temperature for ten minutes. The mixture was then applied to pH gradient (IPG) strips fixed at pH 6-10 (Thermo Fisher Scientific, Waltham, MA) and run on a ZOOM IPG Runner system according to the manufacturer's instructions. The following isoelectric focusing parameters were used: 100-1,000 V for 120 min, 1,000-2,000 V for 120 min, and 2,000 V for 120 min, with limits of 0.1 W and 0.05 mA for each strip run. The IPG strips were then reduced and loaded in a single-well 4-12% Bis-Tris gel and run for 1D as described above. The relative migration of AAV VPs was determined by comparing the internal control protein turbonuclease (Anolon, 27 kDa) with hen egg white conalbumin (Sigma-Aldrich, 76 kDa, pI 6.0-6.6).

[0209] 2. Vector Production

[0210] Recombinant AAV vectors for 1D and 2D gel electrophoresis and mass spectrometry experiments were produced by Vector Core at the University of Pennsylvania and purified by cesium chloride or iodixanol gradients as previously described. (Lock M et al., Human Gene Therapy 2010; 21(10): 1259-71; Gao GP et al., Proc. Natl. Acad. Sci. USA 2002; 99(18): 11854-9). Affinity purified vectors were generated as follows: HEK293 cells were grown in ten 36-layer super stack containers (Corning) and co-transfected with a mixture of vector genome plasmids (pAAV-LSP-IVS2.hFIXco-WPRE-bGH), trans plasmids containing AAV2 rep and AAV8 cap genes, and adenovirus helper plasmids. PEIpro (PolyPlus) was used as the transfection reagent. Five days after transfection, the supernatant was collected, clarified by Sartoguard PES Midicap filter (Sartorius Stedim), and treated with benzonase (Millipore), and then salt was added to reach 0.6 M. The clarified bulk collection material was concentrated ten times by tangential flow filtration (TFF), and then diafiltration was performed for four volumes of affinity column loading buffer. The carrier was captured on POROS Capture Select (Thermo Fisher Scientific) affinity column, and the carrier peak was directly eluted into neutralization buffer at low pH. The neutralized eluate was diluted into a high pH binding buffer, and loaded onto an anion exchange refining column (Cimultus QA-8; Bia separation), wherein the preparation was enriched with (complete) particles containing genomes. The complete carrier particles were eluted with a shallow salt elution gradient, and neutralized immediately. Finally, the vector was subjected to a second round of TFF for final concentration and buffer exchange into formulation buffer (PBS + 0.001% pluronic F-68).

[0211] Mutation vectors for in vitro assays were generated by small-scale triple transfection of HEK293 cells in six-well plates. 5.6 μL 1 mg / mL polyethyleneimine solution was mixed with plasmid DNA (0.091 μg cis plasmid, 0.91 μg trans plasmid, 1.82 μg δF6 Ad-helper plasmid in 90 μL serum-free medium), incubated at room temperature for 15 minutes, and added to cells and another 0.8 mL fresh serum-free medium. The next day, 0.5 mL of top culture medium was replaced with complete serum medium. Three days after transfection, the vector was collected by three freeze / thaw cycles, and then centrifuged to remove cell debris and collect supernatant. The cis plasmid contains a transgenic cassette that encodes the firefly luciferase transgene under the control of the chicken β-actin (CB7) promoter and the Promega chimeric intron and rabbit β-globin (RBG) polyadenylation signal. The trans plasmid encodes the wtAAV8 cap gene; to generate mutant AAV8 cap variants, the Quikchange Lightning Mutagenesis Kit (Agilent Technologies, Wilmington, DE) was used. The vector was titrated as previously described. (Lock M et al., Human Gene Ther. 2010; 21(10): 1259-71).

[0212] For time course vector production experiment, vector was produced by medium-scale triple transfection of HEK293 cells in 15cm tissue culture dishes. For each plate, 36 μ L 1mg / mL polyethyleneimine solution was mixed with plasmid DNA (0.6 μ g cis plasmid, 5.8 μ g trans plasmid, 11.6 μ g δF6 Ad-helper plasmid) in 2mL serum-free culture medium, incubated for 15 minutes at room temperature, and added to cells with about 60% confluence on a plate updated with 14ml serum-free culture medium. The next day, 8ml of top culture medium was replaced with fresh full serum culture medium. By collecting all top culture mediums, cells were scraped from culture dishes and frozen at -80 ℃ to collect vectors. Crude vector was recovered from supernatant / cell mixture by applying 3 freeze / thaw cycles and by centrifugation to clarify lysate. The vector for mass spectrometry analysis was purified and concentrated by adding benzonase, 1M Tris pH 7.5 and 5M NaCl to the clarified lysate to a final concentration of 20mM Tris and 360mM NaCl. The vector was captured on a 1ml POROS CaptureSelect affinity column and the vector peak was eluted directly into neutralization buffer at low pH. The fractions were analyzed by absorbance at 280nm and the most concentrated fractions were subjected to mass spectrometry analysis.

[0213] For in vivo experiments, vectors with wtAAV8 capsid or with one of six deamidation mutants were generated as previously described; the transgene cassette contained the CB7 promoter, the PI intron, the firefly luciferase transgene and the RBG polyadenylation signal (Lock M et al., Human Gene Ther. 2010; 21(10): 1259-71).

[0214] 3. Mass Spectrometry Run / Digestion / Analysis

[0215] Materials: Ammonium bicarbonate, DTT, iodoacetamide (IAM), and 18O-enriched water (97.1% purity) were purchased from Sigma (St. Louis, MO); and acetonitrile, formic acid, trifluoroacetic acid (TFA), 8 M guanidine hydrochloride (GndHCl), and trypsin were purchased from Thermo Fisher Scientific (Rockford, IL).

[0216] Trypsin digestion: A stock solution of 1M DTT and 1.0M iodoacetamide was prepared. The capsid protein was denatured and reduced at 90°C for ten minutes in the presence of 10mM DTT and 2M GndHCl. The sample was allowed to cool to room temperature and then alkylated with 30mM IAM for 30 minutes at room temperature in the dark. The alkylation reaction was quenched by adding 1mL DTT. 20mM ammonium bicarbonate (pH 7.5-8) was added to the denatured protein solution in a volume that diluted the final GndHCl concentration to 200mM. Trypsin solution was added to a trypsin to protein ratio of 1:20 and incubated overnight at 37°C. After digestion, TFA was added to a final concentration of 0.5% to quench the digestion reaction.

[0217] For the 18O water experiments, the capsid samples were first buffer exchanged into 100 mM ammonium bicarbonate prepared in 18O water using a Zeba spin desalting column (Thermo Fisher Scientific, Rockford, IL). To ensure complete removal of water from the samples, two buffer exchanges were performed. Stock solutions of 1 M DTT and 1 M IAM were prepared in 18O water. The same denaturation, alkylation, and digestion steps were performed as above using 18O water reagents and buffers.

[0218] Liquid chromatography tandem mass spectrometry: Online chromatography was performed using an Acclaim PepMap column (15 cm length, 300 μm inner diameter) and a Thermo UltiMate 3000RSLC system (Thermo Fisher Scientific) coupled to a Q Exactive HF (Thermo Fisher Scientific) with a NanoFlex source. The column temperature was maintained at 35°C during online analysis. The peptides were separated using a gradient of mobile phase A (MilliQ water with 0.1% formic acid) and mobile phase B (acetonitrile with 0.1% formic acid). The gradient was run from 4% B to 6% B in 15 minutes, to 10% B in 25 minutes (40 minutes total), and then to 30% B in 46 minutes (86 minutes total). The sample was loaded directly onto the column. The column dimensions were 75 cm x 15 um ID and were loaded with 2 microns of C18 medium (Acclaim PepMap). The total time for each LC-MS was approximately two hours due to loading, introduction, and washing steps.

[0219] Mass spectrometry data were acquired using a data-dependent top 20 method on a Q Exactive HF mass spectrometer that dynamically selected the most abundant precursor ions that had not yet been sequenced from the survey scan (200-2000 m / z). Sequencing was performed by high-energy collisional dissociation fragmentation with a target value of 1e5 ions determined by predictive automatic gain control; and precursor separation was performed with a window of 4 m / z. Survey scans were acquired at a resolution of 120,000 at 200 m / z. The resolution of the HCD spectra was set to 30,000 at m / z 200 with a maximum ion injection time of 50 milliseconds and a normalized collision energy of 30. The S-lens RF level was set to 50 to achieve optimal transmission of the m / z region occupied by peptides from digestion. Precursor ions with single, unassigned, or six and higher charge states were excluded from the fragmentation selection.

[0220] Data processing: use BioPharmaFinder 1.0 software (Thermo Fisher Scientific) to analyze all data acquired.For peptide mapping, use single entry protein FASTA database to search, wherein carbamidomethylation is set to fixed modification, and oxidation, deamidation and phosphorylation are set to variable modification.For tandem mass spectrometry, mass accuracy of 10ppm, high protease specificity and confidence level of 0.8 were used.The mass spectrum identification of deamidated peptide is relatively simple, because deamidation adds +0.984Da (-OH group and-NH2 mass difference between group) to the mass of complete molecule.The deamidation percentage of specific peptide is determined by the mass area of ​​deamidated peptide divided by the area sum of deamidated and natural peptide.Considering the quantity of possible deamidation sites, isobaric species deamidated at different sites can co-migrate at single peak.Therefore, the fragment ion derived from the peptide with multiple potential deamidation sites can be used for locating or distinguishing multiple deamidation sites. In these cases, the relative intensities within the observed isotopic pattern can be used to specifically determine the relative abundance of the different deamidated peptide isomers. This approach assumes that the fragmentation efficiency of all isomeric species is the same and that the deamidation sites are independent. This approach allows the definition of specific sites involved in deamidation as well as potential combinations involved in deamidation.

[0221] Secondary Data Processing: Secondary analysis of raw mass spectra was performed at the University of Maryland, Baltimore, using the following methods. All mass spectrometric analyses were performed using Peaks Studio v5.3 software (Bioinformatics Solutions, Inc.). Data refinement was performed on raw data files using the following parameters: m / z tolerance for precursors ≤ 10 ppm, and precursor charge states were a minimum of 2 and a maximum of 4. Input spectra were sequenced de novo using the peak algorithm with a precursor ion error tolerance of 10 ppm and a product ion error tolerance of 0.1 Da. The digestion enzyme was set to trypsin, the variable modifications were set to oxidation, phosphorylation, and deamidation, and the fixed modifications were set to carbamidomethylation of cysteine.

[0222] 4. Structural Analysis of AAV Capsid

[0223] AAV8 atomic coordinates, structure factors and related capsid models were obtained from the RCSB protein database (PDB ID: 3RA8). Structural refinement was performed, and electron density independent of the primary amino acid sequence of AAV8 VP3 was generated for three-dimensional (3D) structural analysis of the capsid. This analysis was performed to observe the isoaspartic acid electron density in the AAV8 capsid that was not biased by the primary sequence of the expected AAV8 VP3. Using the resulting structure, four asparagines in the primary sequence of AAV8 VP3 were modeled with N+1 glycine as isoaspartic acid, and then the AAV8 capsid structure was refined using Crystallography and NMR System (CNS) software by strictly applying an icosahedral non-crystallization matrix using a standard refinement scheme (Brunger AT et al., Acta Crystallogr D Biol Crystallogr 1998; 54 (Pt5): 905-21). The structural model of isoaspartate was obtained from the HIC-UP database, and a molecular dictionary was generated in PRODRG for structural refinement (Kleywegt GJ, Acta Crystallographica D Region - Biological Crystallography 2007; 63(Pt 1): 94-100). The average electron density map of the AAV8 capsid (also in CNS) was then calculated and visualized using COOT software, and the resulting model was then slightly adjusted to fit the modeled isoaspartate residue into the electron density map (Emsley P and Cowtan K, Acta Crystallographica D Region - Biological Crystallography 2004; 60(Pt 12Pt 1): 2126-32). This protocol was repeated to additionally model N512 in the AAV9 VP3 primary sequence using N+1 glycine (PDB ID: 3UX1). All figures were generated using COOT, PyMol, and UCSF Chimera (Emsley P and Cowtan K et al., Acta Crystallographica Section D - Biological Crystallography 2004; 60(Pt 12Pt 1):2126-32; DeLano WL PyMOL: An Open-Source Molecular Graphics Tool vol. 40, 2002:82-92; Pettersen EF et al., J Comput Chem 2004; 25(13):1605-12).The structures of several previously identified deamidated proteins (PDB IDs: 1DY5, 4E7G, 1RTU, 1W9V, 4E7D, and 1C9D) were obtained for comparison of the electron density maps of their deamidated isoaspartate residues with the modeled isoaspartate residues from AAV8 and AAV9 (Rao FV et al., Chem Biol 2005; 12(1):65-76; Noguchi S et al., Biochem 1995; 34(47):15583-91; Elliott et al., J Mol Biol 2000; 297(3):713-32).

[0224] The temperature coefficients of the deamidated residues were determined by averaging the temperature coefficients for each atom of each asparagine residue reported in the atomic coordinates of the AAV8 or AAV9 crystal structure (PDB ID: 3RA8, 3UX1).

[0225] 5. Animal studies

[0226] All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania. To evaluate vector performance, 3e10 GC wtAAV8 or capsid mutant vectors were injected intravenously into eight-week-old C57BL / 6 mice by tail vein injection in a volume of 100 μL. All mice were killed on the 14th day. In order to evaluate luciferase expression in vivo, mice (about 20 g) were anesthetized and injected intraperitoneally with 200 μL or 15 mg / mL luciferin substrate (Perkin Elmer, Waltham, MA, Waltham, Massachusetts). Five minutes after luciferin was applied, mice were imaged and imaged by the IVISXenogen in vivo imaging system. The signals in the described areas of interest were quantified using Living Image 3.0 software. Measurements were performed on the 7th and 14th days.

[0227] 6. Evaluation of Mutation Vector Titer and In Vitro Transduction Efficiency

[0228] The vector titer was determined by qPCR against DNAseI genome. The qPCR primers were bonded to the polyadenylation sequence of the packaged transgene. In order to evaluate the vector transduction efficiency in vitro by luciferase expression, 0.9e5 Huh7 cells / well were seeded in a black wall 96-well plate in complete DMEM (10% fetal bovine serum, 1% penicillin / streptomycin). The next day, the culture medium was removed and replaced with 50 μL of crude or purified vectors diluted in complete culture medium. For each crude vector sample, 4 dilutions were tested in a 3-fold dilution series. After 48 hours, luciferin (Madison, Wisconsin) was prepared in complete culture medium at 0.3 μg / μL and added to the transduced cells with a volume of 50 μL. The results were read on a BiotekClarity luminometer. It was found that the luciferase activity / GC added to the target cells was constant in a wide GC range, but may become saturated at high MOI. Therefore, the dilution series data (luminescent units to GC) for linearity were checked, and if saturation was obvious, the highest point was excluded and the average luciferase / GC was calculated for the values ​​in the linear range of each assay for each variant. This produces a transduction efficiency value. The data were normalized to simplify comparison by setting the value of the wt control to 1.

[0229] 7. Biodistribution

[0230] DNA was extracted from liver samples using the QIAamp DNA mini kit (Qiagen, Hilden, Germany), and then analyzed by real-time PCR for vector GC as previously described using primer / probe sets designed against the RBG polyadenylation signal of the transgene cassette (Chen SJ et al., Hum Gene Ther Clin Dev 2013; 24(4): 154-60).

[0231]

[0232]

[0233]

[0234]

[0235] B. Results

[0236] AAV8 shows substantial charge heterogeneity in its capsid protein

[0237] To quantitatively assess the presence of post-translational modifications on the AAV8 vector capsid that could affect vector performance, AAV8 total capsid protein purified by iodixanol gradient was analyzed by both 1D and 2D gel electrophoresis. In a 1D reducing sodium dodecyl sulfate SDS gel, VP1, VP2, and VP3 resolved as single bands at the appropriate molecular weight ( Figure 1B )(Rose JA et al., Journal of Virology 1971;8(5):766-70). When proteins are separated by charge ( Figure 1C When further evaluated by 2D gel electrophoresis of VP, each of the capsid proteins was additionally resolved into a series of distinct spots with different isoelectric points (pI) ranging from pH 6.3 to >7.0 ( Figure 1D ). Individual spots on each VP are separated by discrete intervals of 0.1 pI units as measured as migration relative to an internal isoelectric point standard for carbonic anhydrase isoforms, indicating that single residue charge changes occur. The presence of these isoforms suggests that each VP has the potential to undergo a number of modifications that cause it to migrate differently under isoelectric focusing.

[0238] Deamidation in which a portion of the side chain amide group (usually asparagine) is converted to a carboxylic acid Figure 1A ) is a common source of charge heterogeneity in protein preparations. To determine whether deamidation might be responsible for the different populations of VP charge isoforms, two AAV8 asparagine residues were individually mutated to aspartic acid. These capsid mutations should shift the charge by an amount equal to that required to fully deamidate a single additional asparagine residue. 2D gel analysis of the mutants showed that the major spots for VP1, VP2, and VP3 were shifted to one spot position (0.1 pH unit) more acidic compared to the equivalent spots in wild-type (wt) AAV8 ( Figure 1E-1G ). The magnitude of this shift is equal to the spacing observed between wtVP charge isoforms. Thus, 2D gel patterning of AAV capsid proteins is consistent with multisite deamidation.

[0239] Spontaneous deamidation occurs on the AAV8 vector capsid

[0240] To identify the modifications responsible for the discrete spot patterns of each capsid protein, a panel of AAV8 vectors was analyzed by mass spectrometry. The average coverage of AAV8 capsid proteins was >95% of the total VP1 sequence (data not shown). Extensive deamidation of a subset of asparagine and glutamine residues was detected by mass spectrometry, showing an observed increase in the mass of individual peptides of approximately 1 Da compared to the predicted values ​​based on the DNA coding sequence; this pattern of deamidation was observed in all preparations of AAV8 vectors ( Figure 2A-2D ).

[0241] To assess the overall heterogeneity of deamidation between commonly used purification methods and to examine deamidation in unique regions of VP1 and VP2, nine batches of AAV8 produced in 293 cells by triple transfection were selected and purified by cesium chloride gradients, iodixanol gradients, or affinity chromatography. The vectors also differed in promoter and transgene cassette. To determine whether the presence of the vector genome had an effect on deamidation, AAV8 preparations produced in 293 cells by triple transfection in the absence of the cis plasmid (producing only empty capsids) were also evaluated and purified by iodixanol gradients.

[0242] There is a wide range of deamidation across asparagine and glutamine residues in the AAV8 capsid, ranging from no detectable to more than 99% of individual amino acids being deamidated ( Figure 2E ). The highest levels of deamidation (>75%) occurred at asparagine residues where the N+1 residue was glycine (i.e., NG pairs) (Table 1). Lower levels of deamidation (i.e., up to 17%) were detected at other asparagine residues where N+1 was not glycine. The average deamidation of asparagine was largely consistent between preparations. Deamidation at glutamine residues was also detected, but at a lower frequency than at asparagine; the highest percentage observed at Q467 was <2% (Figure 7). This observation was not consistent across all preparations (data not shown). The greatest inter-preparation variation was observed at residue N499 (where the N+1 residue was asparagine), with values ​​ranging from <1% to more than 50% deamidation. Regardless, the variations observed in deamidation between vector preparations did not appear to be associated with the purification method, transgene identity, or the presence of a vector genome, suggesting that these factors do not affect the deamidation rate.

[0243] Table 1: Identity of residues of interest for AAV8 deamidation. Asterisks indicate residues selected for further analysis.

[0244]

[0245]

[0246] Next, a series of experiments were performed to determine whether sample treatments contributed to the observed deamidation levels in AAV8. Extreme temperature (70°C for 7 days) or pH (pH 2 or pH 10 for 7 days) did not significantly induce additional deamidation in the AAV8 capsid ( Figure 4A and Figure 4BGiven this resistance, it is considered unlikely that the observed deamidation occurs only during the relatively short and mild purification phase. Attempts to perform mass spectrometry analysis of unpurified vector to determine the extent of deamidation before and after purification were unsuccessful. Similarly, heavy water controls suggest that treatments specific to the mass spectrometry workflow do not contribute to additional deamidation events ( Figure 4C ).

[0247] To validate the mass spectrometry workflow, two recombinant proteins that had been previously evaluated for deamidation were examined; the findings ( Figure 5A and Figure 5B ) is consistent with published results [Henderson, LE, Henriksson, D and Nyman, PO (1976), "Primary structure of human carbonic anhydrase C" Journal of Biological Chemistry 251: 5457-5463 and Carvalho, RN, Solstad, T, Bjorgo, E, Barroso, JF and Flatmark, T (2003), Deamidations in recombinant human phenylalanine hydroxylase) Identification of labile asparagine residues and functional characterization of Asn-->Asp mutant forms Journal of Biological Chemistry 278: 15142-1515]. In addition, a secondary institution was hired to evaluate the original data from AAV8. This independent analysis identified sites identical to those that were deamidated, with minimal variation in the extent of modification at each site, which was attributed to software-to-software variation in peak detection and area calculations ( Figure 6 ).

[0248] Structural topology, temperature coefficients, and identity of N+1 amino acids contribute to deamidation frequency

[0249] Since the structure of AAV8 has been solved and published (PDB identifier: 2QA0) (Nam HJ et al., J. Virol. 2011; 85(22): 11791-99), the AAV8 capsid structure was examined next to provide evidence for favorable conditions for non-enzymatic deamidation and to correlate the deamidation percentage with established structural features (Nam HJ et al., J. Virol. 2007; 81(22): 12260-71). We focused only on asparagine residues because the factors that affect asparagine deamidation are better characterized in the literature and asparagine deamidation events are far more common than glutamine deamidation events (Robinson, NE and Robinson, AB (2001), "Molecular clocks" Proc. Natl. Acad. Sci. USA 98: 944-949). The temperature (or B) coefficient for each of these residues from the AAV8 crystal structure was also determined; the temperature coefficient is a measure of the displacement of an atom from its average position, with larger values ​​indicating larger displacements, larger thermal vibrations, and therefore greater flexibility (Parthasarathy S and Murphy MR, Protein Science: A Publication of the Protein Society 1997; 6: 2561-7). Most of the asparagines of interest are located in or near surface-exposed HVRs (Table 1), which structurally favor deamidation and provide an environment for solvent exposure (Govindasamy L et al., Journal of Virology 2013; 87(20): 11187-99). It was found that on average residues located in these flexible loop regions were more frequently deamidated than residues in less flexible regions such as β strands and α helices. For example, the NG residue at position N263 is part of HVR I, has a high temperature coefficient, and is deamidated by >98% on average ( Fig. 7A and Figure 6 , Table 1). About 85% of N514 was deamidated within the time (Figures 3 and Figure 6, Table 1) is also present in the HVR with an N+1 glycine (HVR V); however, the local temperature coefficient is relatively low compared to that of N263 due to its interactions with residues on other VP monomers at the three-fold axis. Unfavorable +1 residues and lower local temperature coefficients are associated with lower deamidation, even for HVR residues. For example, N517 averages only 4% deamidation (Table 1); this residue has a temperature coefficient comparable to the highly deamidated N514, but its N+1 residue is a serine, which reduces the likelihood of deamidation events due to steric hindrance. This suggests that many factors cumulatively determine the extent of deamidation at a given capsid position, although the identity of the +1 residue is clearly the most influential factor.

[0250] To test the role of the +1 residue in asparagine deamidation, mutant vectors were generated in which the AAV8 NG site was mutated to alanine or serine at the +1 position, respectively. Model peptide studies have shown that the deamidation half-life of NG peptides is as short as 1 day, while the deamidation of NA peptides or NS peptides is generally 25-fold or 16-fold slower, respectively (Robinson NE and Robinson AB, Proceedings of the National Academy of Sciences of the United States of America 2001; 98 (8): 4367-72). Mass spectrometry analysis of vector mutants confirmed that the +1 site plays a central role in determining the degree of vector deamidation. When the +1 site is changed to alanine (<5% deamidation) or serine (<14% deamidation), the NG site in this collection (>80% deamidation, in wt) shows selective stabilization of the adjacent asparagine (Table 2).

[0251] Table 2: Degree of deamidation (%) at the five AAV8 NG sites and six +1 site mutants in wt.

[0252]

[0253] In the intact fully assembled AAV8 capsid, residues that are at least partially buried and not easily exposed to solvent and / or located in regions of low local flexibility are deamidated less frequently than residues located in more favorable environments (Table 1). Nevertheless, a few residues under unfavorable conditions are still deamidated. For example, N630 is at least partially buried but still has a detectable degree of deamidation. For this residue, the presence of phenylalanine as the N+1 residue suggests that this region may be a novel site for non-enzymatic autoproteolytic cleavage within the AAV8 VP3 protein.

[0254] Structural modeling of AAV8 VP3 confirms the deamidation event

[0255] To provide direct evidence for deamidation in the context of assembled capsids, the crystal structure of AAV8 was evaluated (Nam HJ et al., J Virol 2011; 85(22): 11791-9). The available crystal structures of this serotype (i.e., ) is not sufficient to identify the terminal atom in the R group and, therefore, is insufficient to distinguish directly between asparagine, aspartic acid and isoaspartic acid residues. Other aspects of the structure of the aspartic acid isomers formed under these conditions provide The structure determines the chance of deamidation. This analysis is based on two assumptions: 1) the main product of spontaneous deamidation of asparagine is isoaspartic acid rather than aspartic acid, which is produced in a ratio of 3: 1 (Geiger T and Clarke S, Journal of Biological Chemistry 1987; 262 (2): 785-94); and 2) asparagine or aspartic acid can be distinguished from isoaspartic acid due to the shorter length of the electron density map corresponding to the R group of isoaspartic acid. After the succinimidyl intermediate is resolved during the deamidation reaction, when the β carbon of the R group from isoaspartic acid is incorporated into the main chain of the AAV8VP3 capsid protein backbone and lost, this will produce a shorter R group.

[0256] The AAV8 structure itself was first refined to generate AAV8 capsid electron density that was not biased by the known AAV8 VP3 sequence. The refined AAV8 crystal structure was then examined to demonstrate deamidation based on the presence of shorter R groups relative to isoaspartate ( Figure 3A-3E The electron density map confirmed that no deamidated asparagine was detected by mass spectrometry at 410 ( Figure 3B ), at position 263 ( Figure 3C )、385 places (not shown), 514 places ( Figure 3D ) and 540 ( Figure 3E ) The R group of the highly deamidated N+1 glycine residue is shorter. Therefore, the deamidation indicated by the electron density map is consistent with the data of >75% deamidation at these sites by mass spectrometry. The resulting isoaspartate model is comparable to the isoaspartate residues observed in the crystal structures of other known deamidated proteins, which supports the validity of the analysis of AAV8 (Rao FV et al., Chemical Biology 2005; 12(1): 65-76; Noguchi S et al., Biochemistry 1995; 34(47): 15583-91; Elliott et al., Journal of Molecular Medicine 2000; 297(3): 713-32). This structural analysis serves as an independent confirmation of the deamidation phenomenon observed when analyzing the AAV8 capsid by mass spectrometry.

[0257] Deamidation of the AAV capsid is not serotype-specific

[0258] Serotypes other than AAV8 were investigated to demonstrate capsid deamidation. Using 2D gel electrophoresis ( Fig.11A ) and mass spectrometry ( Fig. 11B ) examined AAV9 vector preparations, including controls for potential vector treatment effects ( Figure 11D-11F ). The pattern and extent of AAV9 deamidation were similar to those of AAV8. All four AAV9 NG sites were >85% deamidated; 13 non-NG sites were deamidated to a lesser extent, with a few showing large batch-to-batch variability in % deamidation. Next, the structural analysis workflow was applied and the existing AAV9 crystallographic data were refitted ( Fig. 11C , Table 3). As with AAV8, isoaspartic acid fits better to the electron density of several NG sites in the AAV9 crystal structure. 2D gel analysis (data not shown) and mass spectrometry (summarized in Table 4) were extended to five additional evolutionarily distinct serotypes (rh32.33, AAV7, AAV5, AAV4, AAV3B, and AAV1). All capsids examined contained similar patterns and extents of deamidation, suggesting that this modification is widespread in clinically relevant AAV vectors and is determined by similar underlying primary sequence and structural factors.

[0259]

[0260]

[0261]

[0262] Deamidation events may affect capsid assembly and transduction efficiency

[0263] One way to test the functional impact of deamidation is to replace asparagine with aspartic acid by genetic mutation. Aspartate mutant vectors encoding a luciferase reporter gene were generated by small-scale triple transfection of 293 cells for each deamidated AAV8 asparagine and the vectors were titrated by qPCR against DNAseI genomic copies ( Fig. 8A ). Relative to wtAAV8, the mutations minimally affected capsid assembly and were limited to mostly buried non-NG sites with low deamidation overall in the wt vector. Next, the mutant group was evaluated for in vitro transduction efficiency in human liver-derived Huh7 cells ( Figure 8B Several mutants showed impaired transduction efficiency, with positions N57, N94, N263, N305, Q467, N479, and N653 exhibiting >10-fold loss of transduction. A similar number of AAV9 sensitive sites ( Fig.11G and Fig.11H ). Typically, only a subset of residues are endogenously deamidated at a given position, so this approach may overestimate loss of function in proteins such as capsids, where functional units are assembled homomonomeric; endogenous modifications at one capsid site may be compensated by adjacent subunits with intact residues. Nonetheless, it is thought that the approach may help prioritize deamidated residues for further monitoring during preparation or mutational stabilization. Functional data from a population of endogenously deamidated vectors will be needed to place such loss-of-function mutagenesis data in proper context.

[0264] Loss of carrier activity over time correlates with progressive deamidation

[0265] Given the short half-life of NG deamidation, it is believed that vector samples with an age difference of only 1 day may show different deamidation spectra, providing an opportunity to correlate endogenous deamidation with function. The large-scale vector preparation protocol requires triple transfection of 293 cells, followed by incubation for 5 days to produce the vector, and 1-2 days for vector purification. To approximate this process, a medium-scale triple transfection of 293 cells (10×15 cm per cell culture dish) was prepared with wtAAV8. The vectors were collected at intervals of 1 day (2×15 cm cell culture dishes / day) for 5 days, and the vectors were stored by freezing at -80°C until the end of the 5-day time point. Next, the crude vector titer and in vitro transduction efficiency were evaluated as described above. As expected, the number of assembled anti-DNAseI genomic copies increased over time ( Fig.9A ). Crude vectors at early (days 1 and 2) and late (day 5) time points were then rapidly processed by affinity purification and the in vitro transduction efficiency of Huh7 cells was measured. The relative transduction efficiency of the vectors gradually decreased over time ( Fig. 9B ). The efficiency of the day 5 vector in terms of transgene expression per GC added to the target cells was only 40% of the efficiency of the day 1 material. A decrease in this activity was also observed for the crude material, suggesting a change in the molecular composition prior to purification (Figure). A similar trend of loss of activity was observed for AAV9 over 5 days, with a decrease in vector potency of approximately 40% ( Figures 11I-11K ).

[0266] Next, deamidation of the time course samples was measured by mass spectrometry. NG site deamidation was substantially complete in each interval, with an average of 25% deamidation on day 1 and >60% of sites converted by day 5 ( Fig. 9C ). Deamidation of non-NG sites generally proceeded within 5 days, although levels were much lower and less consistent between days 2 and 5 ( Fig.9D). The data correlate endogenous vector deamidation with an early time point decay in specific activity and highlight potential opportunities to capture more active vectors by shortening production cycles or finding capsid mutations that stabilize asparagine.

[0267] It should be noted that Figure 2A-2E The material used for mass spectrometry in the was at least 7 days post-transfection, due to the additional 2 days used for purification. The higher NG site deamidation in these samples (>80%) suggests that deamidation may continue at about the same rate after the expression period and during the recovery and purification process until the NG site is completely deamidated or the vector sample is frozen. Therefore, deamidation is largely determined by the age of the vector and is not a process unique to or caused by the recovery and purification process. This is illustrated by the much lower deamidation values ​​of the day 1 material relative to the day 5 material (both affinity purified).

[0268] Stabilizing NG-asparagine can improve carrier performance

[0269] Given the correlation between vector NG deamidation and loss of transduction efficiency, we thought that stabilizing NG amide by +1 site mutagenesis could improve vector function. Vectors were generated on a small scale for AAV8 NG site mutants in which each +1 residue was converted to alanine or serine, respectively. Fig. 10A ) and transduction efficiency ( Fig. 10B ), single + 1 mutants are well tolerated. The G386 substitution (Aydemir F et al., Journal of Virology, July 2016; 90(16): 7196-204), which is located near the previously defined "dead zone" on the capsid surface, has an in vitro transduction defect. The loss of function of the G386 mutant may indicate a preference for the deamidated asparagine at N385. Alternatively, the additional side chain entity at the + 1 position may have a negative impact on the function independent of amide group stabilization. Although its adjacent asparagine has significant stability, single site mutants do not significantly improve in vitro transduction (Table 2). Because the transduction activity in vitro and in vivo may be inconsistent, a subset of single site + 1 mutants for liver transduction was tested in C57BL / 6 mice. Intravenous tail vein injections (n ​​= 3 to 5) were performed, and luciferase expression was examined by weekly imaging for 2 weeks ( Fig. 10C ). The in vivo and in vitro transduction data were consistent with (i.e., within) the error range associated with each assay. The G386 substitution was defective in transduction, whereas +1 site mutations at other positions were largely tolerated, transducing the liver at levels equal to, but not exceeding, wtAAV8.

[0270] Since stabilizing amides at any one NG site may be necessary but not sufficient to restore function, vector variants with combinations of +1 site alanine substitutions were evaluated next. All 3 AAV8 NG sites (N263, N514, and N540) were functional with recombinant +1 alanines. Some combinations containing the triple mutant G264A / G515A / G541A assembled poorly and were dysfunctional for transduction. However, two pairwise combinations involving N263 (G246A / G515A and G264A / G541A) both improved in vitro transduction efficiency (2.0-fold and 2.6-fold that of wtAAV8, respectively) without loss of titer ( Fig. 10D ). Because these mutations introduce at least two changes (N-amide stabilization and +1 residue side chain substitution), these data cannot conclusively link NG deamidation to loss of function. However, the data are consistent with a model established in the time course studies in which NG site deamidation may affect in vitro transduction efficiency.

[0271] Functional asparagine substitution improves batch-to-batch reproducibility in vector preparation

[0272] Another potentially problematic aspect of the reported deamidation profiles of the vectors is the high batch-to-batch variability in deamidation at some positions. For wtAAV8, this variability was most pronounced for N459 (observed deamidation ranged from 0% to 31%) and N499 (observed deamidation ranged from 0% to 53%). Variability in post-translational modifications is often practically avoided during biologic development by either completely avoiding clones that exhibit this variability, carefully monitoring and controlling production strains and conditions, or performing protein engineering of affected candidates.

[0273] Since it is not possible to identify production or processing factors that cause variability in deamidation of N459 and N499 ( Figure 2E ), so functional amino acid substitutions were sought at these positions. The possibility of conservative substitutions of glutamine at each position in a small-scale vector preparation was first evaluated. Both N459Q and N499Q were efficiently assembled into the vector and were comparable to the in vitro transduction efficiency reference of wtAAV8 ( Fig. 7A Next, the mutants were produced on a large scale and subjected to mass spectrometry. Consistent with the observation of extremely rare glutamine deamidation, selective and complete stability of glutamine at position 459 or 499 was observed in these mutants (data not shown). Liver transduction of these mutant batches was evaluated in vivo after tail vein injection in C57BL / 6 mice as described above ( Figure 7B and Figure 7C). The wtAAV8 vector batch used as a control in this experiment was deamidated by 16.8% at N499, but no deamidation was detected at N459 (data not shown). Liver transduction of both mutants at day 14 was equivalent to that of wtAAV8. This data demonstrates the potential of protein engineering approaches to address molecular variability associated with deamidation in manufactured AAV vectors.

[0274] C. Discussion

[0275] Non-enzymatic deamidation of asparagine and glutamine residues on the AAV8 capsid was identified and evaluated by 2D gel electrophoresis, mass spectrometry, de novo protein modeling, and functional studies both in vitro and in vivo. Deamidation has been shown to occur in a wide variety of proteins and significantly affect the activity of biologics, including antibody-based therapeutics (Nebija D et al., Int J Mol Sci 2014; 15(4): 6399-411) and peptide-based vaccines (Verma A et al., Clinical and Vaccine Immunology 2016; 23(5): 396-402). Other viral proteins, such as the VP6 protein of rotavirus, have been shown by mass spectrometry to undergo deamidation events (Emslie KR et al., Funct Integr Genomics 2000; 1(1): 12-24).

[0276] The occurrence of these deamidations in AAV8 suggests that they are the result of spontaneous non-enzymatic events. Asparagine residues are known to be more extensively deamidated than glutamine residues; the amino acids downstream of asparagine substantially affect the rate of deamidation, with N+1 glycine (ie, NG) being most efficiently deamidated. The role of the N+1 amino acid in the deamidation of the AAV capsid was observed to be clearly confirmed, as each NG present in VP1 was deamidated at a level of >75%, while the deamidation of any of the other asparagine or glutamine in the capsid was never >20%. In fact, all NG motifs in AAV8 and AAV9 capsids (ie, 7 / 9) are also present on the surface of the capsid contained in the HVR region associated with high rates of conformational flexibility and thermal vibration. This is consistent with previous reports of NG motifs in other proteins that are located in regions where flexibility may be required for proper protein function, rather than in more ordered structures such as alpha helices or beta sheets (Yan BX and Sun YQ, Journal of Biochemistry 1997; 272(6): 3190-4). The preference for NG motifs in surface-exposed HVRs further enhances the deamidation rate by providing solvent accessibility and conformational flexibility, thereby promoting the formation of succinimidyl intermediates. As expected, the unfavorable environment results in a much lower deamidation rate.

[0277] An important question about the biology of AAV and its use as a vector is the functional consequences of these deamidations. Mutagenesis of the capsid DNA converting asparagine to aspartic acid allows the evaluation of capsids in which all amino acids at specific sites are represented as aspartic acid. However, there is no easy strategy to prevent deamidation using mutagenesis, except for potentially mutating the N+1 residue, which is confused with the direct consequences of the second site mutation. A limited number of variants have been studied in which asparagine residues are converted to aspartic acid by mutagenesis. Functional analysis includes capsid assembly and in vitro and in vivo transduction. The most basic effect of mutagenesis on vector function is the effect of asparagine that is not completely deamidated at the baseline and is not exposed on the surface. However, surprisingly, mutagenesis of highly deamidated asparagine to aspartic acid at 514 does have some effect on function. This result indicates that the presence of the residual amount of the corresponding amide may affect function. This may be due in part to the presence of a hydrogen bonding interaction between N514 and D531 of another three-fold related VP3 monomer (identified in the wtAAV8 crystal structure) that is lost upon conversion of this residue to aspartic acid following deamidation.

[0278] In assessing the impact of these deamidations on the development of novel therapeutics, it is important to better understand the factors that influence the extent of deamidation in AAV vectors. Incubating the vector under extreme conditions (known to significantly accelerate deamidation kinetics) had little effect. Combined with isotope incorporation studies, this result suggests that deamidation occurs during capsid assembly and is not an artifact of vector processing or mass spectrometry analysis. Deamidation at the NG site is unlikely to have a substantial effect on vector performance because the reaction was virtually complete in every sample evaluated. However, initial functional studies suggest that residual amounts of undeamidated asparagine can contribute to function. More attention is paid to sites where deamidation is incomplete, which in most cases is also associated with sample-to-sample variation. An example is the deamidation range of 0% to 53% for asparagine at position 499, with an average of 17%. Slight differences in vector production conditions may contribute to this heterogeneity. The striking similarity of deamidation in AAV8 and AAV9 suggests that this is a property of this entire virus family.

[0279] In summary, a great deal of heterogeneity was found in the primary amino acid structure of AAV8 and AAV9 capsid proteins. These studies may impact the development of AAV as a vector in several ways. First, the actual amino acid sequence of the VP protein is not predicted by the corresponding DNA sequence. Second, various aspects of the production process may lead to differences in deamidation and corresponding changes in vector function. Until the factors that affect the rate of deamidation at non-NG sites are mastered and their functional consequences are better understood, it may be necessary to include deamidation in the characterization of clinical-grade AAV vectors. 2D gel electrophoresis can provide an overall assessment of net deamidation, although mass spectrometry is essential for assessing deamidation at specific residues.

[0280] Example 2: Deamidated AAV8 triple mutant (clade E)

[0281] The AAV8 triple mutant capsid is used to generate rAAV vectors. The predicted amino acid sequence of the VP1 protein of this capsid is provided in SEQ ID NO:9 herein, and the nucleic acid sequence encoding the capsid is provided in SEQ ID NO:8. See also PCT application PCT / US17 / 27392, published as WO 2017 / 180854.

[0282] Deamidation of the AAV8 triple mutant vector was assessed as described for AAV8 in Example 1. Highly deamidated residues were seen at N57, N384, N498, N513, N539. Deamidation of 10% to 40% was observed at N94, N254, N255, N304, N409, N516.

[0283]

[0284]

[0285] Example 3: Additional deamidation studies

[0286] Illustrative vectors were evaluated for deamidation as described for AAV8 and AAV9 in Example 1. AAV1 fell within clade A, AAV7 fell within clade D, and AAV3B, AAV5, AAVrh32 / 33, and AAV4 were outside of any of clades AF.

[0287] A. Deamidation of AAV1

[0288] The AAV1 vector was evaluated for deamidation as described for AAV8 and AAV9 in Example 1. The results showed that the vector contained four amino acids (N57, N383, N512, and N718) that were highly deamidated, based on the numbering of the primary sequence of AAV1 VP1 reproduced in SEQ ID NO: 1.

[0289]

[0290]

[0291] B. AAV3B deamidation

[0292] Deamidation of the AAV3B vector was assessed as described for AAV8 and AAV9 in Example 1. High levels of deamidation were observed at four asparagine residues, N57, N382, N512, and N718, numbered with reference to AAV3B. These numbers are based on the AAV3B VP1 reproduced in SEQ ID NO:2.

[0293] AAV3B modification Enzymes Trypsin Trypsin Trypsin Trypsin Trypsin Trypsin cover% N+1 89.8 83.3 89.8 88.3 92.6 86.4 About N57+ deamidation G 94.6 89.4 91.6 89.9 88.9 88.2 About N94+ deamidation H 7.6 7.9 5.9 5.9 5.4 About N214+ deamidation N 1.3 About N223+ deamidation S 12.1 3.4 N227+ deamidation W 1.6 About N253+ deamidation N 6.8 2.6 About N268+deamidation D 1.4 3.6 5.9 2.4 About N270+ deamidation H 5.7 0.6 3.3 2.8 About Q259+deamidation I 1.5 About N302+ deamidation NNN 23.8 18.4 11.7 19.8 20.9 9.5 About N382+ deamidation G 96.2 87.6 84.4 100.0 85.9 83.0 About Q465+ deamidation A 0.8 N477+deamidation W 1.2 1.3 1.5 About N512+ deamidation G 97.2 100.0 91.5 88.2 92.9 93.2 About N582+ deamidation N 11.5 3.6 6.4 About Q599+deamidation G 2.3 1.0 N691+deamidation S 13.5 1.4 N718+deamidation G 100.0 98.9 97.8 90.8 89.5 97.3

[0294] C. AAV5 deamidation

[0295] Deamidation of the AAV5 vector was assessed as described for AAV8 and AAV9 in Example 1. High levels of deamidation were observed at residues N56, N347, N347, and N509. Deamidation of about 1% to about 35% was observed for the following positions: N34, N112, N213, N243, N292, N325, N400, Q421, N442, N459, and N691. These numbers are based on the AAV5 VP1 reproduced in SEQ ID NO: 3.

[0296] AAV5 modification Enzymes Trypsin Trypsin Trypsin Trypsin Trypsin Trypsin cover% N+1 88.7 89.2 81.4 88.8 91.7 82.9 N34+ deamidation Q 7.6 N56+ deamidation G 99.9 87.3 84.9 88.3 82.8 87.9 About N79+ deamidation E 0.3 About N93+ deamidation H 6.3 5.8 5.5 7.7 2.9 N112+ deamidation L 2.3 About N213+ deamidation A 16.5 About N243+ deamidation N 32.8 24.8 About N259+ deamidation A 2.7 About N292+ deamidation N 27.6 27.0 N325+ deamidation N 9.9 N347+deamidation G 81.1 94.2 91.7 87.2 88.1 85.7 About N400+ deamidation N 5.4 3.3 2.4 4.8 3.3 3.3 About Q421+deamidation N 7.0 About N442+ deamidation N 24.1 About N459+deamidation T 12.5 About N509+deamidation G 85.1 92.6 89.9 98.0 92.0 94.0 About N572+ deamidation N 0.9 4.1 2.5 0.1 2.3 About N691+deamidation N 23.1 13.3 4.2 0.4 0.5

[0297] D. AAV7 deamidation

[0298] Deamidation of AAV7 vectors was assessed as described for AAV8 and AAV9 in Example 1. High levels of deamidation were observed at N41, N57, N384, and N514. Deamidation was observed at rates of 1% to 25% at N66, N224, N228, N304, N499, N517, N705, and N736. These numbers are based on the AAV7 VP1 reproduced in SEQ ID NO:4.

[0299]

[0300]

[0301] E. Deamidation of AAVrh32.33

[0302] Deamidation of the AAVrh32.33 vector was assessed as described for AAV8 and AAV9 in Example 1. High levels of deamidation were observed at positions N57, N264, N292, N318. Deamidation ranging from 1% to 45% was observed at positions N14, N113, Q210, N247, Q310, N383, N400, N470, N510, and N701.

[0303] These numberings are based on the rh32.33 AAV VP1 reproduced in SEQ ID NO:5.

[0304] AAVrh32.33 modification WL1408S Enzymes Trypsin cover% 100 N14+ deamidation 3.0 N57+ deamidation 100.0 N113+ deamidation 1.1 Q210+deamidation 14.8 N247+ deamidation 31.1 About N264+ deamidation 100.0 About N292+ deamidation 50.2 Q310+deamidation 5.4 About N318+ deamidation 92.2 N383+deamidation 2.1 About N400+ deamidation 39.7 About Q449+deamidation 2.9 N470+deamidation 2.6 N498+deamidation 0.6 About N510+deamidation 27.3 About N701+deamidation 6.2 N731+deamidation 40.2

[0305] F. AAV4 deamidation

[0306] AAV4 was evaluated as previously described. High levels of deamidation were observed at positions 56 and 264. Other positions with high levels of deamidation may include positions 318 and 546.

[0307]

[0308]

[0309] 1 Not covered by trypsin

[0310] 2 Not covered by chymotrypsin

[0311] * If a residue was observed in both formulations, the average was taken. If a residue was in one formulation, only this formulation was used.

[0312] Trypsin and chymotrypsin preparations are reported separately. However, based on the sequences and peptides obtained, some residues will be lost by trypsin or chymotrypsin. In cases where residues are observed in both preparations, the deamidation is consistent, so the averages should not differ too much.

[0313] Example 4: Mapping AAV-9-specific neutralizing epitopes

[0314] In this study, an attempt was made to identify neutralizing epitopes on AAV9 that had not been evaluated by this epitope mapping approach. Importantly, AAV9 is currently being used clinically by intravenous administration for a variety of cardiac, musculoskeletal, and central nervous system indications (Bish LT et al., Human Gene Therapy 2008; 19(12): 1359-68; Foust KD et al., Nature Biotechnology 2009; 27(1): 59-65; Kornegay JN et al., Mol Therapy 2010; 18(8): 1501-8), most notably spinal muscular atrophy (Mendell JR et al., NEngl J Med 2017; 377(18): 1713-22). Here, the highest resolution AAV-Ab complex reconstructed to date is reported: The structure was complexed with the effective NAb PAV9.1. The epitope of PAV9.1 was verified by using serotype exchange, alanine substitution and additional point mutants, and the resulting mutants were shown to significantly interfere with the ability of PAV9.1 to bind and neutralize. However, when the mutants were tested for a group of polyclonal samples from various sources, this effect on both PAV9.1 binding and neutralization ability was significantly reduced or not observed. This result shows that although this epitope can play a role in the neutralization of AAV transduction in some cases, a wider range of targeted mutants of neutralizing epitopes are still needed to engineer novel capsids that can circumvent the NAb library responsible for blocking AAV transduction.

[0315] A. Materials and Methods

[0316] 1. Hybridoma Generation

[0317] Balb / c mice received up to five immunizations with AAV9 vectors. Splenocytes were harvested and fused. Clonal supernatants were generated by ProMab Biotechnology, Inc. (Richmond, CA) according to the company's standard custom mouse monoclonal antibody hybridoma development protocol. Thirty supernatants were screened for AAV9 reactivity by ELISA and for their ability to neutralize AAV9 by NAb assay. After screening, purified PAV9.1 mAb was obtained at a concentration of 3 mg / mL.

[0318] 2. AAV capsid ELISA

[0319] Corning polystyrene high-binding microplates were coated with 1e9 GC / well AAV diluted in phosphate-buffered saline (PBS) and left overnight at 4°C. After discarding the coating solution, the plates were blocked with 3% bovine serum albumin (BSA) in PBS at room temperature for 2 hours and then washed three times with 300 μL PBS+0.05% Tween. Hybridoma supernatants, purified mAbs, serum or plasma (diluted in PBS with 0.75% BSA) were then incubated at 37°C for 1 hour and then washed three times with 300 μL PBS+0.05% Tween. Next, mouse samples were detected using 1:10,000 goat anti-mouse IgG HRP (diluted in PBS with 0.75% BSA; Catalog No. 31430; Thermo Fisher Scientific, Waltham, MA) for 1 hour at 37°C and then washed three times with 300 μL PBS+0.05% Tween. Human and non-human primate samples were then probed with 1:10,000 (diluted in PBS) goat anti-human IgG biotin-SP (Cat. No. 109-065-098, Jackson ImmunoResearch Inc., West Grove, PA) for 1 hour at room temperature, followed by three washes with 300 μL PBS+0.05% Tween and 1:30,000 (diluted in PBS) unconjugated streptavidin (Cat. No. 016-000-084, Jackson ImmunoResearch Inc., West Grove, PA) for 1 hour at room temperature (followed by three washes with 300 μL PBS+0.05% Tween). All ELISAs were developed with tetramethylbenzidine.

[0320] 3. Neutralizing Antibody Assay

[0321] NAb assays were performed as previously described with minor modifications (Calcedo R et al., Journal of Infectious Diseases 2009; 199(3): 381-90). HEK293 cells were seeded at a density of 1e5 cells / well on black-walled, clear-bottomed poly-lysine-coated plates (Catalog No. 08-774-256, Fisher Scientific Company, Hampton, NH). A multiplicity of infection of 90 wtAd5 / cell was used, and a working solution of 4e10 GC / mL AAV9.CMV.LacZ vector was used to reach a final concentration of 2e9 GC / well. Bioluminescence was measured using a SpectraMaxM3 (Molecular Devices, Sunnyvale, CA) according to the manufacturer's protocol. For any given sample, the NAb titer was defined as the last dilution at which AAV transduction was reduced by >50% in the presence of the sample compared to WT.AAV transduction in the presence of an untreated control. HEK293 transduction experiments were performed as described above, but with the neutralizing serum retained.

[0322] 4. Fab Production and AAV-Fab Complexation

[0323] PAV9.1 Fab (0.211 mg / mL) was produced using the Pierce Fab preparation kit (Thermo Fisher Scientific, Waltham, Mass.) according to the manufacturer's instructions. Next, PAV9.1 Fab was complexed with AAV9 vector at a ratio of 600 Fab:1 AAV9 capsid (or 10 Fab:1 potential binding site) for 30 minutes at room temperature.

[0324] 5. Cryo-EM Sample Preparation, Data Acquisition, and Complex Reconstruction

[0325] Sample preparation: 3 μL of PAV9.1-AAV9 complex was applied to a freshly washed and glow-discharged holey carbon grid. After blotting with Whatman #1 filter paper for 3 to 4 seconds at 22°C and 95% relative humidity, the grid was quickly frozen in liquid ethane slurry using a Vitrobot Mark IV (FEI). Next, a single blot was performed with Whatman filter paper at 22°C at 95% relative humidity for 3 to 4 seconds. After freezing, the grid was stored in liquid nitrogen. The grid was then transferred to a FEI Talos Arctica electron microscope operating at 200 kV and equipped with a Gatan K2 Summit direct electron detection camera (Gatan, Pleasanton, USA).

[0326] Data acquisition: Data were acquired using SerialEM software (Mastronarde DN, J Struct Biol 2005; 152(1):36-51). Images were captured with a calibrated pixel size of 2.21 electrons / square angstrom / second and a defocus range of 1.0-2.0 μm (Rohou A. and Grigorieff N, Journal of Structural Biology 2015; 192(2):216-21). For each exposure, a 60-frame dose-graded movie stack was recorded in super-resolution mode for a total of 12 seconds. Movie frames were aligned using the "alignframes" program within the IMOD software package (Kremer JR et al., Journal of Structural Biology 1996; 116(1):71-6).

[0327] Data collection and processing: All particle images were manually extracted from each micrograph in the micrographs and processed using the e2boxer program available in the EMAN2 suite (Tang G et al., Journal of Structural Biology 2007; 157(1): 38-46). The framed particles were then transferred to the AUTO3DEM program for cryo-reconstruction to generate a starting low-resolution model based on 150 particle images. (Yan X et al., Journal of Structural Biology 2007; 157(1): 73-82). The program uses a random model generation procedure and applies a strict 60 non-crystallographic symmetry axes. This low-resolution reconstructed model map can be used to determine the particle starting point, complete orientation, and improve the contrast transfer function of all images using AUTO3DEM. In order to improve the quality of the reconstructed map, temperature coefficient correction was applied and the map was visualized in the graphics programs Coot and Chimera (Pettersen EF et al., Journal of Computational Chemistry 2004; 25(13): 1605-12; Emsley P and Cowtan K, Acta Crystallographica D - Biological Crystallography 2004; 60(Pt 12Pt 1): 2126-32). Model docking and interpretation were performed using a correction map with a temperature coefficient of 150. A total of 3,022 framed particles were extracted from 1,100 micrographs to finally generate The reconstructed image with a resolution of 0.15, where the Fourier spherical shell correlation is 0.15. The VIPER database is used to generate the AAV9-60mer model, while applying a strict icosahedral symmetry axis (T=1) (Carrillo-Tripp M et al., "Nucleic Acids Res" 2009; 37 (database issue number): D436-42). Using the FIT function in the Chimera program, the 60-mer copy of the AAV9 capsid is docked into the frozen reconstructed electron density map. This produces a correlation coefficient of 0.9. The docked models in Coot and Chimera are visualized and adjusted for accuracy. The antibody model is generated using ABodyBuilder, and then docked and manually adjusted to the frozen reconstructed density using Chimera (Leem J et al., MAbs 2016; 8 (7): 1259-1268). The model is then visualized to explain the AAV9 and antibody binding regions. All figures are made using Chimera and PyMOL programs. The RIVEM program was used to create a two-dimensional description of the roadmap (DeLano WL, PyMOL: An Open-Source Molecular Graphics Tool 2002; Vol. 40: 82-92). The RIVEM program was used to create a two-dimensional description of the roadmap (Xiao C and Rossmann MG, J. Struct. Biol. 2007. 158(2): 182-7).

[0328] 6. Construction of AAV9-PAV9.1 mutant trans plasmid

[0329] AAV9 capsid mutagenesis was performed using the in-house trans plasmid construct pAAV2 / 9 (AAV2 rep / AAV9 cap). All capsid mutants were constructed using the QuikChange Lightning Mutagenesis Kit (Agilent, Santa Clara, CA) according to the manufacturer’s instructions.

[0330] 7. Vector Production

[0331] AAV9.CMV.LacZ.bGH and AAV9 mutant vectors were generated in HEK293 cells by triple transfection and then purified by iodixanol gradient as previously described (Lock M et al., Human Gene Therapy 2010; 21(10): 1259-71). Vectors were titrated using quantitative PCR (qPCR) against bGH polyA at the University of Pennsylvania Vector Core as previously described (Lock M et al., Human Gene Therapy 2010; 21(10): 1259-71).

[0332] 8. Determination of EC50 of PAV9.1 mAb and polyclonal serum / plasma

[0333] As described above, capsid capture ELISA was performed with AAV9.WT or AAV9 mutant vectors. EC50 values ​​were calculated using GraphPadPrism. In short, PAV9.1 mAb concentrations were logarithmically transformed in the form of mg / mL and plotted on the x-axis. The IgG concentration in mouse plasma was defined as 5 mg / mL (Mink JG, "Serum immunoglobulin levels and immunoglobulin heterogeneity in the mouse" "Diss. Erasmus MC. 1980), and the IgG concentration in non-human primates and human serum was defined as 10 mg / mL (Gonzalez-Quintela A et al., "Clinical and Experimental Immunology" 2008; 151 (1): 42-50). Plasma / serum concentrations (in μg / mL) were logarithmically transformed and plotted on the x-axis. The maximum absorbance achieved with each mutant was defined, the absorbance was normalized to 100%, and plotted on the y-axis. Then, a dose-response curve (antibody binding) was generated using the "log (agonist) vs. normalized response - variable slope" function of GraphPadPrism. Finally, the EC50 of PAV9.1 mAb, polyclonal serum, or polyclonal plasma was calculated.

[0334] 9. Animal research

[0335] Animal protocols were approved and performed in accordance with the standards of the Institutional Animal Care and Use Committee of the University of Pennsylvania. Male C57BL / 6 mice (n=3) received intravenous injections of 1e11 GC / mouse AAV9.CMV.LacZ.bGH or AAV9 mutant vectors with the same transgenic cassette in the tail vein. Animals were sacrificed 14 days after receiving the vectors. Organs from each animal were separated and snap-frozen on dry ice for biodistribution or embedded in optimal cutting temperature compound and frozen for subsequent sectioning and β-gal activity staining.

[0336] 10. Biodistribution Analysis

[0337] DNA was extracted from tissues of interest using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Tissues of carrier GCs were analyzed by qPCR for bGH polyadenylation signals as previously described (Chen SJ et al., Human Gene Therapy Clin Dev 2013; 24(4): 154-60).

[0338] 11.β-gal activity staining

[0339] The frozen sections were fixed with PBS containing 0.5% glutaraldehyde at 4°C for 10 minutes, and then stained for β-gal activity. After washing in PBS, the sections in 1 mg / ml X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) were incubated in PBS (pH about 7.3) containing 20 mM potassium ferrocyanide, 20 mM potassium ferrocyanide, 2 mM MgCl2, and the tissues were placed overnight at 37°C. After counterstaining the sections with nuclear fast red (Vector Laboratories), they were dehydrated using ethanol and xylene and then mounted on slides.

[0340] B. Results

[0341] 1. NAb PAV9.1 is effective and specific for AAV9

[0342] The first aim was to identify novel and potent anti-AAV9 NAbs for epitope mapping. A panel of 30 hybridoma clones was screened for AAV reactivity by enzyme-linked immunosorbent assay (ELISA) against multiple serotypes and for AAV9 neutralization by NAb assay. From this panel, the monoclonal antibody PAV9.1 ( Fig. 12A ). PAV9.1 recognizes only intact capsids by ELISA ( Fig. 12A ), and did not recognize AAV by western blot (data not shown), indicating that PAV9.1 identifies a conformational epitope on the capsid surface. This is in contrast to the remaining clones, which more broadly bound to this panel of AAVs included in the screen and also recognized AAV by western blot (data not shown). In the NAb assay, the purified PAV9.1 mAb showed an effective NAb titer of 1:163,840, indicating that this novel anti-AAV9 antibody is an effective neutralizer of AAV9. Again, this is in contrast to other clones screened by the NAb assay that were unable to neutralize AAV transduction.

[0343] 2. Cryoconstitution of AAV9 and complexing with PAV9.1

[0344] After AAV9 was complexed with the PAV9.1 antigen-binding fragment (Fab), 1,100 images were captured, 3,022 particles were framed, and AUTO3DEM was used to generate a map of the complex. The Fab density was observed to extend from the three-fold axis formed by HVRs IV, V, and VIII, and the inner surface of the vertically centered three-fold protrusion was decorated with Fab electron density ( Fig.13A and Fig. 13B This region contains mainly charged residues that facilitate strong electrostatic interactions between the three-fold related VP monomers and with the receptor and mAb. A single Fab molecule binds at each three-fold axis and extends across two of the three protrusions, thereby blocking the binding of additional Fab molecules at these sites due to steric hindrance ( Fig. 13C The regions of the PAV9.1 Fab complementarity determining regions (CDRs) that contact the triplet protrusions had an average density of 2.5σ, which is comparable to the density reported for other AAV-Fab reconstructions. A density of approximately 0.8σ was observed for the PAV9.1 Fab constant region, or approximately one-third the density observed for the contact regions of the PAV9.1 CDRs, corresponding to a single Fab occupancy for each triplet axis. The PAV9.1 Fab CDRs directly interact with residues 496-NNN-498 (HVR V) and 588-QAQAQT-593 (HVR VIII). Fig. 13C and Fig.13D ). PAV9.1 binding additionally blocks residues G455 and Q456 (HVR IV), T494, Q495 and E500 (HVR V), and N583, H584, ​​S586 and A587 (HVR VIII), which are not involved in electrostatic interactions with PAV9.1 but may provide structural stability to this region of the capsid after Fab binding (Table 3). The CDRs of the heavy chain interact with HVR V, while the CDRs of the light chain interact with HVR VIII of the same VP3 monomer ( Fig. 13C ).

[0345] Table 3: PAV9.1 Fab epitope residues

[0346]

[0347] Based on PAV9.1 footprint ( Fig.13D , Table 3), two groups of five residues were selected for focused mutagenesis for epitope validation and circumvention mutant design: 586-SAQAQ-590 and 494-TQNNN-498. Residue 586-SAQAQ-590 was selected because this site contains a high degree of sequence diversity ( Fig. 12B). The selected motif contains residues identified by reconstruction as directly interacting with PAV9.1, as well as residues identified as blocking, allowing querying the connection between binding residues and blocking residues. These residues are also related to neutralizing epitopes of AAV1, AAV2, and AAV8, allowing comparison of AAV9 epitope residues with previously disclosed residues (Tseng YS and Agbandje-McKenna M, Front Immunol 2014; 5:9). Finally, limiting HVR VIII mutagenesis to these five residues increases the likelihood that the capsid will tolerate larger mutations because this motif has more limited interactions with regions that contribute to the integrity of the capsid structure. Although PAV9.1 is specific for AAV9, the HVRV motif 496-NNN-498 identified as interacting with PAV9.1 is highly conserved between serotypes ( Fig. 12B ). However, unpublished phage display work (data not shown) suggests that the asparagine-rich motif is associated with the epitope of PAV9.1; therefore, this motif was selected for mutagenesis. Residues 494-TQ-495 were also added to re-interrogate the connection between the binding and blocking residues, as they have been previously implicated in AAV-Ab interactions (Tseng YS and Agbandje-McKenna M, Frontiers in Immunology 2014; 5:9).

[0348] 3. Epitope-based mutations significantly reduce AAV9-PAV9.1 binding

[0349] First, site-directed mutagenesis was used to generate the 586-SAQAQ-590 serotype exchange mutant. Based on the knowledge that PAV9.1 specifically recognizes AAV9 and that the amino acid sequence and structural conformation at this position vary greatly between AAV serotypes, a complete exchange with the corresponding residues of representative serotypes from clade B (AAV2), clade C (AAV3B), and clade D / E (AAV8 / rh10) was selected (Table 4).

[0350] Table 4: Mutagenesis strategy for PAV9.1 HVR VIII epitope residues

[0351]

[0352] In doing so, it was desired to maximize the likelihood of efficient capsid assembly while also maximizing natural variation at this position. Two additional mutants, AAV9.AAQAA (more convergent than AAV9.QQNAA) and AAV9.RGHRE (more divergent than AAV9.RGNRQ), were generated to determine (1) the minimal mutations required to disrupt the PAV9.1 interaction and (2) the maximal disruption that could be introduced. The titers of vectors produced by the AAV9.AAQAA, AAV9.QQNAA, and AAV9.SSNTA mutants were equivalent to those of AAV9.WT vectors; however, the titers of AAV9.RGNRQ and AAV9.RGHRE were reduced two- to three-fold relative to that of AAV9.WT (data not shown). Binding of the PAV9.1 mAb to each mutant capsid was determined by capture ELISA compared to AAV9.WT ( Fig.14A ). The EC50 or concentration of PAV9.1 mAb required to achieve half-maximal binding to PAV9.1 was significantly increased for each exchange mutant relative to the EC50 of AAV9.WT (indicating reduced capsid binding). This result confirms the results of epitope mapping, indicating that residues 586-SAQAQ-590 are involved in the AAV9-PAV9.1 interaction. The increase in EC50 ranged from 45-fold (AAV9.AAQAA) to almost 300-fold (AAV9.RGHRE) (Table 5); the increase in EC50 is directly related to the degree of sequence divergence from AAV9 at this position. One exception is AAV9.RGNRQ, which shares Q590 with AAV9, which may contribute to stronger PAV9.1 binding than expected by sequence analysis.

[0353] Table 5: Summary of AAV9 capsid mutant properties after in vitro evaluation

[0354] Nab titer reduction fold EC50 fold increase WT transduction percentage WT 1 1 100 AAQAA 16 45 27 QQNAA 128 124 53 SSNTA 512 264 58 RQ 8 96 233 RGHRE 2048 294 60 QUR 16 15 50 SAQAN 16 40 76 SAQAA 4 20 54

[0355] Since the S586A and Q590A mutations in AAV9.AAQAA were sufficient to disrupt PAV9.1 binding to AAV9, we next determined the minimal changes required to induce this disruption. Point mutations were introduced at one of these positions, either by alanine substitution or a more conservative substitution (S->T or Q->N). Mutations at S586 to alanine or threonine did not significantly reduce PAV9.1 binding, whereas single mutations at Q590 to alanine or asparagine were sufficient to disrupt capsid recognition by PAV9.1 ( Fig. 14C ). This result indicates that position 590 is critical for PAV9.1 to recognize AAV9 capsid.

[0356] Next, the 494-TQNNN-498 motif of HVRV was interrogated for inclusion in the PAV9.1 epitope using the same mutagenesis strategy as follows: groups of residues were mutated to only evolutionarily conserved amino acids or alanine. Since 496-NNN-498 was conserved in all serotypes tested, only alanine substitutions were used for this stretch of residues; for 494-TQ-495, mutations were to AA as well as GQ and TD to represent the naturally occurring diversity at this site. Despite PAV9.1's specificity for AAV9 and diversity at this position, AAV9.GQNNN, AAV9.TDNNN, and AAV9.AANNN did not increase the EC50 of PAV9.1 against AAV ( Fig. 14B ). This confirms the conclusion from the cryo-reconstruction map that the 494-TQ-495 site is not involved in the PAV9.1 epitope. However, the AAV9.TQAAA mutation increased the PAV9.1 EC50 by 15-fold, suggesting that although 496-NNN-498 is a conserved motif, it still plays an important role in AAV9-specific binding of PAV9.1. Finally, combinatorial mutants were generated from HVRV and minimal HVRVIII mutations (AAV9.TQAAA / SAQAN, AAV9.TQAAA / SAQAA); the PAV9.1 EC50 values ​​of these combinatorial mutants suggest that the effects of the altered motifs in the PAV9.1 epitope are cumulative ( Fig.14D and Fig.14E ).

[0357] 4. Epitope-based mutation regulation of AAV9 transduction

[0358] To evaluate the ability of novel AAV9 mutants to circumvent NAbs while maintaining the properties of AAV9.WT, we first assessed transduction in vitro and in vivo. Most mutations that caused reduced PAV9.1 binding also reduced transduction efficiency in HEK293 cells, with the notable exception of AAV9.RGNRQ, which increased vector transduction by 2.3-fold ( Fig.15A ). This improvement may be due to the introduction of R586 and R589 (R585 and R588 by AAV2 VP1 numbering), two residues responsible for heparin recognition by AAV2, which has significantly better in vitro performance than AAV9 in most cell lines due to the inclusion of these heparin binding motifs ((Ellis BL et al., J Virol 2013;10(1):74). However, AAV9.RGHRE, which shares R586 and R589 with AAV9.RGNRQ, did not exhibit AAV2-like transduction efficiency, suggesting the involvement of other factors. AAV9.AAQAA exhibited the greatest reduction in transduction efficiency, suggesting that S586 and / or Q590 are essential residues for AAV9 transduction in vitro.

[0359] 5. Epitope-based Mutation Ablation of PAV9.1 Neutralization

[0360] Next, the effects of the mutations on PAV9.1 neutralization titers were examined. The mutant AAV9.AANNN, which did not affect PAV9.1 binding, did not affect neutralization titers ( Fig. 15B and Fig.15I ). However, all mutant vectors that increased PAV9.1 EC50 reduced the effective neutralizing titer of PAV9.1. AAV9.RGHRE, which had the most significant increase in EC50 by nearly 300-fold, reduced the NAb titer of PAV9.1 by at least 2,048-fold (from 1:163,840 to <1:80, the lowest dilution tested) ( Figure 15C-15K ). Mutant vectors that more modestly increased EC50 (such as AAV9.SAQAN) reduced the effective NAb titer of PAV9.1 to a lesser extent. In summary, a strong correlation was observed between reduced PAV9.1 binding as measured by EC50 and reduced effective NAb titer ( Fig.16 The notable exception was again AAV9.RGNRQ, which, despite being the fourth most potent mutant in reducing PAV9.1 binding, reduced NAb titers by only eightfold (the second lowest reduction).

[0361] 6. The PAV9.1 epitope is important for AAV9 liver tropism

[0362] To evaluate the feasibility of these mutants as AAV9-like gene therapy vectors, C57BL / 6 mice were intravenously injected with 1e11 genome copies (GC) / mouse of AAV9.WT.CMV.LacZ or AAV9 mutant vectors with reduced PAV9.1 activity (n=3 per group). Biodistribution of tissue samples on day 14 showed reduced liver transduction for all mutants. AAV9.QQNAA performed most similarly to AAV9.WT, with a 17-fold reduction in GC / μg DNA, while AAV9.RGHRE had the lowest liver transduction efficiency, with a 1,110-fold reduction in GC / μg DNA ( Fig.17A). However, in other organs (such as heart and brain), most mutants maintained transduction levels near AAV9.WT, with the exception of AAV2-like mutants, AAV9.RGNRQ and AAV9.RGHRE. ​​Although these differences in tissue GCs were not statistically significant, the observed trends suggest that these residues are important for AAV9 liver tropism but play a smaller role in transduction of other tissues, as most mutants exhibited a “liver off-targeting” phenotype. These results were further reflected in the expression of β-galactosidase (β-gal) in the liver and heart; liver β-gal activity was highest in animals receiving AAV9.WT, while cardiac β-gal activity was similar between AAV9.WT and most mutants (except AAV2-like mutants) ( Fig. 17B and Fig. 17C ).

[0363] These experiments were repeated at a ten-fold higher dose (1e12 GC / mouse) for a representative subset of AAV9 mutant vectors. Although differences in transduction did not reach significance at this dose, trends in tissue tropism were consistent with those observed at lower doses, particularly for heart and muscle samples ( Fig.17D ). These results were also reflected in the β-gal activity in histological sections of liver, heart and muscle ( Figure 17E-17G ).

[0364] 7. Epitope-based mutations in AAV9 do not significantly affect binding or neutralization by polyclonal plasma or serum

[0365] Next, the ability of mutant vectors based on the PAV9.1 epitope to circumvent binding and neutralization by polyclonal plasma or serum was assessed. Plasma from C57BL / 6 mice previously injected intravenously with AAV9.WT was first used (7.5e8 or 7.5e9 GC / mouse, n=6 per group). The plasma dilution required to achieve half-maximal binding was determined. Plasma from low-dose mice had almost no difference in binding to the mutant vectors and to AAV9.WT ( Figures 18A-18C ). In contrast, significant differences in the EC50s from plasma from high-dose mice were observed for a subset of mutants relative to the EC50 for AAV9.WT, most notably AAV9.RGNRQ ( Figure 18B-Figure 18D Although the EC50 of high-dose mouse plasma increased by an average of two-fold for AAV9.RGNRQ, no reduction in effective NAb titers in plasma was observed in this mutant (data not shown).

[0366] To determine if this trend of increasing EC50 was also true for nonhuman primate samples, sera were obtained from a group of six macaques that received either AAV9 vectors or novel vectors closely related to AAV9 with the same VP3 sequence (2 amino acid differences in the nonstructural VP1 region). The macaques were confirmed to have NAb titers <1:5 against AAV9 prior to administration (defined as NAb negative). Although some shifts in the EC50 of each animal's sera against the mutant vectors were indeed observed compared to the EC50 of AAV9.WT, no clear trend toward increased or decreased binding based on mutant identity was observed ( Fig.19A and Fig.19C When sera from macaques were tested with pre-existing NAb titers against AAV9 (due to previous AAV infection), little change in the EC50 of sera from the AAV9 mutant groups was observed ( Fig.19B and Fig.19D This is in stark contrast to the changes seen in the EC50 of injected sera, suggesting that there are fundamental differences between the relevant anti-AAV epitope repertoires of sera generated in response to AAV infection and AAV vector administration. In addition, the increase in the EC50 of injected non-human primate sera against AAV9.RGNRQ did not reduce the effective NAb titers of sera against AAV9.RGNRQ (data not shown).

[0367] Finally, NAb-positive serum samples from four normal human donors were evaluated for binding to AAV9.WT and mutant vectors. As was the case with the uninjected NAb-positive nonhuman primate serum samples, all four NAb-positive normal human donor samples showed minimal EC50 changes for the AAV9 mutant vs. WT vector ( Figure 20A-20B As expected, the lack of change in EC50 of the mutant vectors translated into no reduction in NAb titers in sera from AAV9 mutant vectors (data not shown).

[0368] C. Discussion

[0369] Here, cryo-reconstitution of AAV9 complexed with the highly potent and specific mAb PAV9.1 is reported. The epitopes identified for PAV9.1 overlap to a large extent with the epitope regions of other AAV NAbs isolated from mouse hybridomas, namely, ADK8 (AAV8; 586-LQQQNT-591), E4E (AAV1; 492-TKTDNNN-498), 5H7 (AAV1; 496-NNNS-499, 588-STDPATGD-595), and C37 (AAV2; 492-SADNNNS-498, 585-RGNRQ-589) (Gurda BL et al., J Virol 2012;86(15):7739-51; Gurda BL et al., J Virol 2013;87(16):9111-24; Tseng YS et al., J Virol 2015;89(3):1794-1808). Thus, despite a large degree of sequence and structural divergence between HVR V and VIII serotypes, this finding suggests that the triple protrusion may be an important site for AAV9 neutralization, as it is for other serotypes. Thus, previous findings on the repertoire of NAbs against other AAV capsids may apply to AAV9. Although the various mapped neutralizing epitopes showed overlap, the binding angle and orientation of the NAbs varied significantly. When bound to AAV9, PAV9.1 extends toward the center of the three-fold symmetry axis, spatially constrained to occupancy of 20 Fab particles; in contrast, mAbs generated against other serotypes bind at the top or facing outward from the three-fold axis, allowing for higher occupancy. Studies have identified HVR V and VIII as shared antigenic regions across serotypes, including AAV2 (complexed with C37B, ), AAV8 (complexed with ADK8, ) and AAV1 (complexed with 5H7, ), which has the greatest similarity to the binding footprint of PAV9.1 to AAV9 (Gurda BL et al., Journal of Virology 2012; 86(15):7739-51; Gurda BL et al., Journal of Virology 2013; 87(16):9111-24; Tseng YS et al., Journal of Virology 2015; 89(3):1794-1808). Therefore, the structure reported here is similar to the low-resolution structures previously reported for other AAV serotypes.

[0370] HVR VIII serotype swapping confers varying degrees of binding and neutralization circumvention to its corresponding mutant vectors. Swapping this region with an AAV2-based RGHRE motif, the most dispersed mutant from the WT.AAV9 sequence, ablated PAV9.1 neutralization at all dilutions tested. Thus, engineering only five amino acids in the capsid can circumvent monoclonal Nabs. In fact, the minimum change required to significantly reduce PAV9.1 activity is a single amino acid substitution, where even the conservative amino acid results in ablation of both binding and neutralization. Mutations in the NNN motif in HVRV reduce the ability of PAV9.1 to bind and neutralize AAV9, despite being highly conserved between serotypes, suggesting that it is also an integral part of the PAV9.1 epitope.

[0371] It was observed that there was a strong correlation between the reduced binding of PAV9.1 to a given AAV9 mutant and its ability to block transduction of this mutant in vitro, indicating that the relative strength of NAb to AAV is related to the neutralizing ability of NAb. However, data from the laboratory and other laboratories show that the binding antibody titer for AAV is not always a good prediction of the NAb titer of an individual, because some individuals have moderate binding titers for AAV, but NAb negative (Falese L et al., Gene Therapy 2017; 24 (12): 768-78; Huttner NA et al., Gene Therapy 2003; 10 (26): 2139-47) (unpublished data). Despite these findings, the exclusion criteria for some clinical trials include not only NAb titers, but also binding titers (George LA et al., Blood 2017; 130 (Suppl 1): 604; Mendell JR et al., New England Journal of Medicine 2017; 377 (18): 1713-22). Epitope mapping studies are therefore essential to identify the characteristics of the binding epitope and to determine whether the binding epitope shares any commonality with the neutralizing epitope. The shared motifs suggest that binding strength, rather than interactions with specific residues, plays an important role in AAV neutralization, thus allowing researchers to simply focus on reducing NAb binding. However, the distinct motifs suggest that neutralization is more a function of binding location rather than binding strength, and suggest that researchers should focus on ablating NAb binding to these unique regions.

[0372] Although mutations in the AAV9 vectors significantly reduced binding and neutralization by purified monoclonal PAV9.1 antibodies, these mutations did not significantly circumvent binding or neutralization by polyclonal antibodies from serum or plasma of mice, macaques, or human donors previously exposed to AAV. Most notably, plasma from mice that received higher intravenous doses of AAV9 vectors bound the RGNRQ mutants twofold less efficiently than the WT.AAV9 vectors; this change was much smaller than the 50-fold reduction observed with the PAV9.1 mAb. Although the QQNAA, SSNTA, and RGHRE mutations had a greater effect on PAV9.1 binding and neutralization than the RGNRQ mutations, polyclonal plasma bound these mutants in the same manner as WT.AAV9. This result suggests that although the 586-SAQAQ-590 motif is a potent neutralizing epitope and mutations in this region can block PAV9.1 activity, in vitro activity against mAbs does not predict activity against polyclonal antibodies. Perhaps surprisingly, the RGNRQ mutant effectively blocked the binding of AAV9 antibodies by using the triple overhang. This result clearly demonstrates that not all mutations behave equally against polyclonal responses and that a larger repertoire of antibodies utilizes this region for binding.

[0373] Despite the reduction in polyclonal binding, the RGNRQ mutant vector did not circumvent the polyclonal NAb response generated by these mice in response to vector administration. As expected, mutants that did not reduce binding to polyclonal plasma did not circumvent neutralization. Given that the nearly 100-fold increase in the EC50 of PAV9.1 of RGNRQ relative to WT.AAV9 resulted in only an eight-fold decrease in PAV9.1 neutralization titer, it is not surprising that a two-fold increase in the EC50 of polyclonal plasma of RGNRQ did not reduce neutralization titer. Although studies have shown that most mapped AAV epitopes are located on the three-fold axis and HVR VIII is associated with the mapped epitopes of most serotype-specific NAbs, it was surprising to find that none of the mutations tested in this region significantly affected polyclonal activity (it should be noted that the mapped epitopes may not represent the complete library because the total number of mapped epitopes is small and the exact screening and selection methods of some studies are unknown).

[0374] Tse and colleagues recently used a library approach to combine epitopes of three different NAbs identified against AAV1 and generate novel AAV1-based capsids with more than 20 amino acid changes from the parental AAV1. This capsid can not only evade anti-AAV1 monoclonal NAbs, but also evade polyclonal samples from mice and non-human primates injected with AAV vectors in addition to polyclonal samples from normal human donors exposed to AAV (Tse LV et al., Proceedings of the National Academy of Sciences of the United States of America 2017; 114(24): E4812-21). This suggests that neutralizing epitopes can overlap after vector exposure and viral infection, but the library is slightly different. In other words, the total number of residues that need to be modified to confer evasion of binding and neutralization to AAV is more extensive than previously thought. Engineered novel capsids that can address both of these situations may require combinatorial and high-throughput approaches.

[0375] This study explored whether vectors engineered to circumvent pre-existing NAb responses from prior AAV infection would also function in the setting of re-administration. Polyclonal samples that exhibited even minimal circumvention with PAV9.1-based AAV9 mutant vectors were obtained from sources that had already received AAV vectors and were not obtained from sources that had been previously infected with AAV. However, the injected samples showed moderately variable binding curves for the AAV9 mutant panel, and the binding curve generated by vectors from untreated but virus-exposed sources was similar to that of WT.AAV9. These differences highlight fundamental differences between the AAV antibody repertoires generated in response to vector administration or infection.

[0376] Historically, untreated subjects injected with AAV vectors have generated NAb responses that are specific to the administered vector or limited to closely related serotypes (Flotte TR et al., Human Gene Therapy 2011; 22(10): 1239-47) (unpublished data). Most macaque studies and gene therapy clinical trials have shown similar results (Greig JA et al., Vaccine 2016; 34(50): 6323-29; Greig JA et al., Human Gene Therapy Clinical Development 2017; 28(1): 39-50) (unpublished data). In stark contrast, subjects with pre-existing antibodies to one AAV serotype are almost always seropositive and have NAbs to most other serotypes, even distantly related ones (Calcedo R and Wilson JM, Human Gene Therapy Clin Dev 2016;27(2):79-82; Flotte TR et al., Human Gene Therapy 2011;22(10):1239-47; Harrington EA et al., Human Gene Therapy 2016;27(5):345-53) (unpublished data). To date, all novel mapped AAV mAbs are specific for a single serotype and cross-react only with closely related serotypes (e.g., 5H7, which binds to both AAV1 and AAV6); none of the previously isolated neutralizing AAV mAbs recapitulates the broader responses commonly seen following AAV infection (Gurda BL et al., J Virol 2013;87(16):9111-24). Therefore, further studies are necessary to identify motifs that encompass broadly neutralizing epitopes associated with preexisting immunity, determine whether the epitopes overlap with serotype-specific epitopes, and evaluate how overlapping motifs confer a broadly neutralizing phenotype on NAbs.

[0377] The magnitude of the NAb response varies widely across exposure regimens; individuals with natural immunity rarely have NAb titers exceeding 1:80 (human) or 1:320 (macaque); in contrast, NAb titers >1:1,000 can be readily achieved in response to delivery of moderate doses of vector (Greig JA et al., Vaccine 2016;34(50):6323-29; Greig JA et al., Human Gene Therapy Clinical Dev 2017;28(1):39-50; Greig JA et al., PLoS One 2014;9(11):e112268). In this study, mice that received the highest vector dose, resulting in the highest NAb titers, had measurable changes in mutant vector binding; this suggests that the magnitude of the NAb response affects mutant efficiency. Typically, studies aim to reduce the NAb titer of an individual to below a threshold that interferes with gene transfer (1:10 for intravenous administration) (Chicoine LG et al., Mol Therapy 2014; 22(2):338-47; Wang L et al., Human Gene Therapy 2011; 22(11):1389-1401). Conferring mutant capsids engineered based on a single neutralizing epitope that circumvents high titer sera will not significantly increase the number of individuals eligible for AAV gene therapy, as lower titers are still above the threshold for significant inhibition of transduction.

[0378] Reducing the minimum mutation required for PAV9.1 binding at Q590 in HVR VIII, even after conservatively replacing amino acids with asparagine, also conferred a liver detargeting phenotype for the resulting mutant. Mutations in the HVRV portion of the epitope also reduced liver transduction. These results are consistent with previous observations that these residues in HVR V and VIII play an indispensable role in liver transduction, and are consistent with previous reports of neutralizing AAV epitopes that overlap with the mapping of regions necessary for gene transfer (Adachi K et al., Nature Communications 2014; 5: 3075; Tseng TS et al., Journal of Virology 2015; 89 (3): 1794-808). This suggests that while maintaining the parental transduction spectrum, it would be difficult to engineer mutants that can circumvent NAbs. For some indications in the heart and muscle, where liver transduction may be less indirect, this tropism modification may be acceptable. Notably, most mutants maintained WT.AAV9 transduction levels in peripheral organs at both doses.

[0379] However, the RGNRQ mutant showed modest binding modifications in the presence of polyclonal antibodies, showing an AAV2-like transduction profile: poor transduction not only of the liver but of all peripheral organs. Taken together, these data demonstrate the importance of integrating knowledge about mapped neutralizing epitopes with available information about AAV functional domains. It is not sufficient to generate capsids that can circumvent NAbs, since the capsid is only useful if it can still perform its primary function of target tissue transduction. Recent studies have used this strategy to incorporate multiple epitopes of AAV1 to generate AAV1-based vectors that can circumvent NAbs while maintaining an AAV1-like transduction profile (Tse LV et al., Proc. Natl. Acad. Sci. USA 2017; 114(24): E4812-21).

[0380] In conclusion, this study provides critical information regarding the design of AAV9-based vectors that are able to circumvent humoral immune responses. Further studies are needed to further understand the complexity of NAb responses to AAV9 vectors to inform the design of next-generation capsids.

[0381] (Sequence Listing Free Text)

[0382] For digital identifiers <223> Below the sequence containing free text, the following information is provided.

[0383]

[0384]

[0385] All documents cited in this specification are incorporated herein by reference. U.S. Provisional Patent Application Nos. 62 / 722,388 and 62 / 722,382, both filed on August 24, 2018, 62 / 703,670 and 62 / 703,673, both filed on July 26, 2018, 62 / 677,471 and 62 / 677,474, both filed on May 29, 2018, 62 / 667,585, filed on May 29, 2018, and 62 / 635,964, filed on February 27, 2018, are incorporated herein by reference. U.S. Provisional Patent Application No. 63 / 667,881 filed on May 7, 2018, U.S. Provisional Patent Application No. 62 / 667,888 filed on May 7, 2018, U.S. Provisional Patent Application No. 62 / 667,587 filed on May 6, 2018, U.S. Provisional Patent Application No. 62 / 663,797 filed on April 27, 2018, U.S. Provisional Patent Application No. 62 / 663,788 filed on April 27, 2018, and U.S. Provisional Patent Application No. 62 / 635,968 filed on February 27, 2018 are incorporated by reference. The SEQ ID NOs cited herein and appearing in the attached sequence listing are incorporated by reference. Although the present invention has been described with reference to specific embodiments, it should be understood that modifications can be made without departing from the spirit of the present invention. Such modifications are intended to fall within the scope of the appended claims.

Claims

1. A composition comprising a mixed population of recombinant adeno-associated viruses (rAAVs), each of which comprises: (a) an AAV capsid, wherein the AAV capsid comprises about 60 capsid vp1 proteins, vp2 proteins, and vp3 proteins, wherein the vp1 proteins, the vp2 proteins, and the vp3 proteins are: a heterogeneous population of vp1 proteins produced by a nucleic acid sequence encoding a selected AAV vp1 amino acid sequence, a heterogeneous population of vp2 proteins produced by a nucleic acid sequence encoding a selected AAV vp2 amino acid sequence, a heterogeneous population of vp3 proteins produced by a nucleic acid sequence encoding a selected AAV vp3 amino acid sequence, wherein: the vp1 protein, the vp2 protein, and the vp3 protein contain a subpopulation with amino acid modifications, the amino acid modifications comprising at least two highly deamidated asparagines (N) in the asparagine-glycine pairs in the AAV capsid, and optionally further comprising a subpopulation comprising other deamidated amino acids, wherein the deamidation results in an amino acid change, provided that the rAAV is not AAVhu68; and (b) a vector genome in the AAV capsid, the vector genome comprising a nucleic acid molecule comprising an AAV inverted terminal repeat sequence and a non-AAV nucleic acid sequence encoding a product, the non-AAV nucleic acid sequence being operably linked to a sequence that directs expression of the product in a host cell.

2. The composition of claim 1, wherein the deamidated asparagine is deamidated to aspartic acid, isoaspartic acid, an interconverted aspartic acid / isoaspartic acid pair, or a combination thereof.

3. The composition of claim 1, wherein the capsid further comprises one or more deamidated glutamines, wherein the one or more deamidated glutamines are deamidated to (α)-glutamic acid, γ-glutamic acid, an interconverted (α)-glutamic acid / γ-glutamic acid pair, or a combination thereof.

4. The composition of any one of claims 1 to 4, wherein the capsid comprises four to five highly deamidated asparagines in asparagine-glycine pairs.

5. The composition of any one of claims 1 to 5, wherein the capsid comprises an asparagine that is 65% to 100% deamidated at position 57 relative to the numbering of AAV8 or AAV9, as determined using mass spectrometry.

6. A composition according to any one of claims 1 to 5, comprising: (a) rAAV having an AAV8 capsid, the composition further comprising a subpopulation in which at least 70% to 100% of the N in the capsid are deamidated at the following positions based on numbering of the AAV8 vp1 of SEQ ID NO:6 (encoded AAV8 vp1) with an initial M: N57, N263, N385, N514 and / or N540; (b) an rAAV having an AAV9 capsid, the rAAV further comprising a subpopulation in which at least 65% to 100% of the Ns in the capsid are deamidated at the following positions based on numbering of SEQ ID NO: 7 (encoded AAV9 vp1) with a starting M: N57, N329, N452, and / or N512; (c) an rAAV having an AAVrh10 capsid (AAVrh10), the rAAV further comprising a subpopulation of vp1, vp2, and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of SEQ ID NO: 112 (encoded AAVrh10 vp1) with a starting M: N263, N385, and / or N514, or (d) an rAAV having an AAVhu37 capsid (AAVhu37), further comprising a subpopulation of vp1, vp2 and / or vp3, wherein vp1, vp2 and / or vp3 have at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of SEQ ID NO:36 (encoded AAVhu37 vp1) with a starting M: N263, N385 and / or N514.

7. The composition according to any one of claims 1 to 5, wherein the composition comprises: (a) an rAAV having an AAV1 capsid, the rAAV comprising a subpopulation of vp1, vp2 and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a starting M, based on numbering of SEQ ID NO: 1: N57, N383, N512, N718; (b) an rAAV having an AAV3B capsid, the rAAV comprising a subpopulation of vp1, vp2 and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a start M, with reference to numbering of SEQ ID NO:2: N57, N382, N512, N718; (c) an rAAV having an AAV5 capsid, the rAAV comprising a subpopulation of vp1, vp2, and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a start M, with reference to numbering of SEQ ID NO:3: N56, N347, N347, N509; (d) an rAAV having an AAV7 capsid, the rAAV comprising a subpopulation of vp1, vp2 and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a start M, with reference to numbering of SEQ ID NO:4: N41, N57, N384, N514; (e) an rAAV having an AAVrh32.33 capsid, the rAAV comprising a subpopulation of vp1, vp2, and / or vp3 having at least 70% to 100% N deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a start M, with reference to numbering of SEQ ID NO:5: N57, N264, N292, N318; or (f) an rAAV4 vector comprising a subpopulation of vp1, vp2 and / or vp3 having at least 70% to 100% N-deamidation at the NG pair at one or more of the following positions based on numbering of the predicted vp1 amino acid sequence with a starting M and with reference to numbering of SEQ ID NO: 111: N56, N264, N318, N546.

8. The composition of any one of claims 1 to 7, wherein the capsid comprises an asparagine that is 80% to 100% deamidated at position 57 relative to the numbering of AAV8 or AAV9.

9. The composition of any one of claims 1 to 8, wherein all or a subpopulation of the AAVvp1 proteins and / or vp3 proteins are truncated at their N-termini by about 1 to about 5 amino acids.

10. The composition of any one of claims 1 to 9, wherein all or a subpopulation of the AAVvp1 proteins and / or vp3 proteins are truncated at their C-termini by about 1 to about 5 amino acids.

Images