Method and composition for improving assembly adeno-associated viruses (AAV)

Abstract

To provide a method for improving assembly of adeno-associated viruses (AAV) in a cell.SOLUTION: A method includes: a step of producing modified cells by increasing expression or activity of Hsc / Hsp70 or a cofactor thereof, vacuole-specific H+ATPase, and / or cyclin-dependent kinase 2 in a cell; a step of producing infected modified cells by infecting the modified cells with AAV vector; a step of culturing the infected modified cells; and a step of collecting assembled AAVs.SELECTED DRAWING: Figure 17

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority pursuant to 35 U.S.C. §119(e) to U.S. Application No. 62 / 833,595, filed April 12, 2019.

[0002] Technical Field The present disclosure relates generally to adeno-associated viruses and methods of producing adeno-associated viruses. [Background technology]

[0003] background Recombinant adeno-associated viruses (rAAV) for use in gene therapy have been mainly produced in mammalian cell lines, such as 293 cells, COS cells, HeLa cells, KB cells, and other mammalian cell lines (see, for example, U.S. Patent No. 6,156,303, U.S. Patent No. 5,387,484, U.S. Patent No. 5,741,683, U.S. Patent No. 5,691,176, U.S. Patent No. 5,688,676, US20020081721, WO00 / 47757, WO00 / 24916, and WO96 / 17947). However, in most of these mammalian cell culture systems, the number of rAAV particles produced per cell is only about 10E4, and larger-scale production of rAAV is required for clinical research. Summary of the Invention [Problem to be solved by the invention]

[0004] There is a need for methods to improve the production of rAAV. [Means for solving the problem]

[0005] overview The present disclosure provides multiple genes and proteins that can be modulated in cultured cells to potentially increase the production and / or assembly of recombinant adeno-associated viruses (rAAVs). Using the methods described herein, production titers can be maximized for use in clinical or experimental applications, including gene therapy. As described herein, modulation can include knockout, knockdown, or overexpression of a selected gene in a producer cell line, inhibition / activation of a selected gene by chemical compounds added to the culture medium, or a combination thereof.

[0006] In one embodiment, a method for improving the assembly of adeno-associated virus (AAV) in cells is provided. Such a method typically includes the steps of producing modified cells by increasing the expression or activity of Hsc / Hsp70 or its cofactors, vacuole-specific H+ ATPase, and / or cyclin-dependent kinase 2 in the cells, infecting the modified cells with an AAV vector to produce infected modified cells, culturing the infected modified cells, and collecting the assembled AAV.

[0007] In some embodiments, the amount of assembled AAV collected is greater than the amount of assembled AAV collected after AAV infection of cells without the modification, hi some embodiments, the method results in a significant increase in AAV titer.

[0008] In another aspect, a method for improving the assembly of adeno-associated virus (AAV) is provided. Such methods typically include infecting modified cells with AAV, wherein the cells have been modified to exhibit an increase or decrease in one or more genes or proteins encoded thereby selected from CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and TBCB, and collecting the assembled AAV.

[0009] In some embodiments, the modified cells are genetically engineered cells. Exemplary genetically engineered cells include knockout, knockdown, overexpression, or a combination thereof.

[0010] In some embodiments, the modified cells contain chemical compounds that increase or decrease one or more of the genes or proteins encoded thereby. Representative chemical compounds include, but are not limited to, Cdk1 inhibitor IV, Cdk2 inhibitor II, apoptazole, ML-792, BML282, NSC348884, FDNB, and bafilomycin A1.

[0011] In yet another embodiment, a cell-free culture system for assembling adeno-associated viruses is provided. Such a system typically includes a culture medium and at least two proteins selected from CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and TBCB, or nucleic acids encoding these at least two proteins. In some embodiments, the article comprises at least three (e.g., at least four, at least five, etc.) proteins or nucleic acids encoding these at least three (e.g., at least four, at least five, etc.) proteins.

[0012] In yet another embodiment, a cell line is provided, which typically comprises one or more mutations in Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2, and / or one or more exogenous constructs expressing Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2.

[0013] In some embodiments, the mutation comprises a knockout mutation. In some embodiments, the mutation comprises a knockdown mutation. In some embodiments, the one or more exogenous constructs comprise a recombinant construct.

[0014] In another aspect, a product for improving intracellular assembly of adeno-associated virus (AAV) is provided. Such a product typically includes at least one element selected from at least two of (a), (b), or (c): (a) Hsc / Hsp70 or a cofactor thereof, a nucleic acid encoding Hsc / Hsp70 or a cofactor thereof, or a compound that modulates Hsc / Hsp70 or a cofactor thereof; (b) vacuole-specific H+ ATPase, a nucleic acid encoding a vacuole-specific H+ ATPase, or a compound that modulates vacuole-specific H+ ATPase; and (c) cyclin-dependent kinase 2, a nucleic acid encoding cyclin-dependent kinase 2, or a compound that modulates cyclin-dependent kinase 2.

[0015] In some embodiments, the compound that modulates Hsc / Hsp70 or its cofactors is apoptazole. In some embodiments, the compound that modulates vacuolar-specific H+ ATPase is bafilomycin A1. In some embodiments, the compound that modulates cyclin-dependent kinase 2 is Cdk2 inhibitor II.

[0016] In one aspect, the disclosure features a method for improving the assembly of adeno-associated viruses (AAV). Such methods typically include infecting genetically engineered cells with AAV, wherein the cells have been engineered to exhibit an increase or decrease in one or more genes or proteins encoded thereby selected from CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and / or TBCB, and collecting the assembled AAV.

[0017] In another aspect, the disclosure features a method for improving the assembly of adeno-associated virus (AAV) in cells. Such a method typically includes infecting a cell with AAV in the presence of one or more factors selected from Hsc / Hsp70 or its cofactors, vacuole-specific H+ ATPase, cyclin-dependent kinase 2, and combinations thereof, and collecting the assembled AAV.

[0018] In yet another aspect, the disclosure features a method for improving the assembly of adeno-associated virus (AAV) in cells. Such methods typically include modulating one or more factors in cells, where the factors are selected from Hsc / Hsp70 or its cofactors, vacuole-specific H+ ATPase, and cyclin-dependent kinase 2, infecting the modulated cells with AAV, and collecting the assembled AAV.

[0019] Generally, after using the methods described herein, the amount of assembled AAV collected is greater than the amount of AAV collected after AAV infection of cells that have not been genetically engineered or regulated. Generally, the methods described herein result in increased AAV titers. Generally, the methods described herein result in increased quality and performance of AAV vector preparations.

[0020] In some embodiments, the genetic manipulation comprises one or more of knockout, knockdown, and / or overexpression, or a combination thereof. In some embodiments, the modulation is achieved using knockout, knockdown, overexpression, or a combination thereof. In some embodiments, the modulation is achieved using inhibition and / or activation of selected genes, for example, by adding chemical compounds to the culture medium.

[0021] In yet another aspect, the present disclosure provides a method for the accelerating of adeno-associated virus (AAV) in cells. The present invention provides a product for improving the assembly of Hsc / Hsp70. Such a product typically includes at least one element selected from at least two of (a), (b), or (c): (a) Hsc / Hsp70 or a cofactor thereof, a nucleic acid encoding Hsc / Hsp70 or a cofactor thereof, and / or a compound regulating Hsc / Hsp70 or a cofactor thereof; (b) vacuole-specific H+ ATPase, a nucleic acid encoding vacuole-specific H+ ATPase, and / or a compound regulating vacuole-specific H+ ATPase; and / or (c) cyclin-dependent kinase 2, a nucleic acid encoding cyclin-dependent kinase 2, and / or a compound regulating cyclin-dependent kinase 2.

[0022] In another aspect, cell lines are provided that comprise one or more mutations (e.g., knockout, knockdown) in Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2; and / or one or more exogenous (e.g., recombinant) constructs expressing Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2.

[0023] In another aspect, the disclosure features a cell-free culture system for assembling adeno-associated viruses. The system comprises a culture medium and at least two (e.g., at least three, at least four, at least five, etc.) proteins selected from CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and TBCB, or nucleic acids encoding these at least two proteins.

[0024] Unless otherwise defined, all scientific and technical terms used herein have the same meaning as commonly understood by those skilled in the art to which the method and composition of the present invention pertains.Methods and materials similar or equivalent to those described herein can be used in clinics or in testing the method and composition, and suitable methods and materials are described below.In addition, materials, methods, and examples are for illustrative purposes only and are not intended to be limiting.All publications, patent applications, patents, and other reference materials mentioned herein are incorporated by reference in their entirety.

[0025] DESCRIPTION OF THE DRAWINGS [Brief explanation of the drawings]

[0026] [Figure 1] FIG. 1 is a schematic diagram showing the procedural flow of the CRISPR screening used herein. [Figure 2A] 10 is a plot showing sgRNA abundance 18 hours after transfection. [Figure 2B] 10 is a plot showing sgRNA abundance 18 hours after transfection. [Figure 3A] 10 is a plot showing sgRNA abundance 24 hours after transfection. [Figure 3B] 10 is a plot showing sgRNA abundance 24 hours after transfection. [Figure 4] 1 is a graph showing the effect of transfection method on vector titer. The graph represents one experiment, and * indicates confirmed contamination. [Figure 5A] Graph showing the effect of siRNA pre-transfection and co-transfection on vector production as a ratio relative to parallel transfections using siRNA not targeting a sequence in the human genome. Averages of experiments performed in duplicate are represented. [Figure 5B] Graph showing the effect of siRNA pre-transfection and co-transfection on vector production as a percentage of parallel mock transfections. Means of experiments performed in duplicate are represented. [Figure 6A] FIG. 1 is a schematic diagram of HEK293 cells transfected with the expression constructs indicated in the numbered ovals. [Figure 6B]Figure 6B shows images of gels showing co-immunoprecipitations from lysates used for anti-HA (VP bait) and anti-FLAG (AAP bait) pulldowns (indicated by numbers in Figure 6A) from the same experiment performed twice (labeled A and B (above the lanes)). Figure 6B shows total protein stained with SYPRO Ruby, and Figure 6C was probed for VP and AAP by Western blot. [Figure 6C] Figure 6B shows images of gels showing co-immunoprecipitations from lysates used for anti-HA (VP bait) and anti-FLAG (AAP bait) pulldowns (indicated by numbers in Figure 6A) from the same experiment performed twice (labeled A and B (above the lanes)). Figure 6B shows total protein stained with SYPRO Ruby, and Figure 6C was probed for VP and AAP by Western blot. [Figure 6D] Figure 6 shows gels showing immunoprecipitations performed on equal volumes of lysates prepared as shown in Figure 6A. Equal volumes of eluate were loaded per lane, electrophoresed, and stained with SYPRO Ruby to show total protein (Figures 6D and 6F) or probed for VP and AAP by Western blot (Figures 6E and 6G). [Figure 6E] Figure 6 shows gels showing immunoprecipitations performed on equal volumes of lysates prepared as shown in Figure 6A. Equal volumes of eluate were loaded per lane, electrophoresed, and stained with SYPRO Ruby to show total protein (Figures 6D and 6F) or probed for VP and AAP by Western blot (Figures 6E and 6G). [Figure 6F] Figure 6 shows gels showing immunoprecipitations performed on equal volumes of lysates prepared as shown in Figure 6A. Equal volumes of eluate were loaded per lane, electrophoresed, and stained with SYPRO Ruby to show total protein (Figures 6D and 6F) or probed for VP and AAP by Western blot (Figures 6E and 6G). [Figure 6G]Figure 6 shows gels showing immunoprecipitations performed on equal volumes of lysates prepared as shown in Figure 6A. Equal volumes of eluate were loaded per lane, electrophoresed, and stained with SYPRO Ruby to show total protein (Figures 6D and 6F) or probed for VP and AAP by Western blot (Figures 6E and 6G). [Figure 7A] 1 is a heat map showing the mass spectrometry identification of coprecipitated proteins. [Figure 7B] 1 is a heat map showing the mass spectrometry identification of coprecipitated proteins. [Figure 8A] 1 is a bar graph showing the effect of pharmacological inhibition of selected host factors on vector production when added immediately prior to transfection. [Figure 8B] 1 is a bar graph showing the effect of pharmacological inhibition of selected host factors on vector production when added 4 hours post-transfection. [Figure 9] Bar graph (top) based on experimental data (bottom) showing the effect of cyclin-dependent kinase 1 inhibition 4 hours after transfection. [Figure 10] Graph (top) based on experimental data (bottom) showing the effect of cyclin-dependent kinase 2 inhibition 4 hours after transfection. [Figure 11] Graph (top) based on experimental data (bottom) showing the effect of heat shock protein 70 inhibition 4 hours after transfection. [Figure 12] Graph (top) based on experimental data (bottom) showing the effect of inhibition of sumo-activating enzyme 1 4 hours after transfection. [Figure 13] Graph (top) based on experimental data (bottom) showing the effect of inhibiting ubiquitin carboxy-terminal hydrolase isozyme L1 4 hours after transfection. [Figure 14] Graph (top) based on experimental data (bottom) showing the effect of nucleophosmin inhibition 4 hours after transfection. [Figure 15] Graph (top) based on experimental data (bottom) showing the effect of creatine kinase B inhibition 4 hours after transfection. [Figure 16] Graph (top) based on experimental data (bottom) showing the effect of inhibiting vacuolar H+ ATPase 4 hours after transfection. [Figure 17] FIG. 1 is a schematic diagram showing a hypothetical model of AAV assembly. DETAILED DESCRIPTION OF THE INVENTION

[0027] Detailed Description Using genomic and proteomic approaches, multiple host proteins involved in the production and / or assembly of adeno-associated virus (AAV) have been identified.As described herein, one or more of these proteins, or the nucleic acid encoding one or more of these proteins, can be modified to increase the production and / or assembly of AAV in cells.Similarly, one or more of these proteins, or the nucleic acid encoding one or more of these proteins, can be included in a cell-free system to increase the production and / or assembly of AAV.By using the methods described herein, the production titer of viral vectors for clinical or experimental use, including gene therapy, can be maximized.

[0028] As described herein, methods for improving the assembly of adeno-associated virus (AAV) in cells typically involve increasing the expression or activity of Hsc / Hsp70 or its cofactors, vacuole-specific H+ ATPase, and / or cyclin-dependent kinase 2 in the cells. Once the cells have been modified in this way, they can be infected with an AAV vector and subsequently cultured to collect the assembled AAV.

[0029] The following host proteins have been shown to be involved in AAV assembly, and therefore, modifying any one, or a combination of two or more, of these proteins, or the nucleic acids encoding them, as described herein, can improve AAV production and / or assembly: charged multivesicular body protein 7 (CHMP7), centrosomal protein 72 kDa (CEP72), CCR4-NOT transcription complex subunit 6 (CNOT6), solute carrier family 9 member 6 (SLC9A6), periphyllin 1 (PPHLN1), activating transcription factor 7 interacting protein (ATF7IP), set domain protein bifurcated 1 (SET 1), and / or ATP-dependent protein 1 (SET 2). 1;SETDB1), G protein-coupled receptor 89B (GPR89B), kinesin family member 16B (KIF16B), ATPase family AAA domain-containing protein 3A (ATAD3A), BAG family molecular chaperone regulator 2 (BAG2), E3 ubiquitin-protein ligase CHIP (STUB1), DnaJ homolog subfamily A member 1 (DNAJA1), DnaJ homolog subfamily C member 7 (DNAJC7), heat shock 70 kDa protein 8 (HSPA8), heat shock 70 kDa protein 1A (HSPA1A), cyclin-dependent kinase 1 (CDK1), cyclin-dependent kinase 2 (CDK2), coiled-coil-helix coiled-coil rix domain-containing protein 2 (CHCHD2), complement component 1Q subcomponent binding protein (mitochondrial) (C1QBP), dihydropyrimidinase-related protein 5 (DPYSL5), fragile X mental retardation syndrome-associated protein 1 (FXR1), fragile X mental retardation syndrome-associated protein 2 (FXR2), importin-5 (IPO5), lamin-B receptor (LBR), myotrophin (MTPN), nucleoplasmin-3 (NPM3), nucleophosmin (NPM1), periplakin (PPL), sorting nexin-3 (SNX3), E3-independent E2 ubiquitin-conjugating enzyme (UBE2O), SUMO-activating enzyme (SUMO-activating enzyme) subunit 1 (SAE1), coiled-coil domain-containing protein 124 (CCDC124), guanine nucleotide-binding protein G(I) / G(S) / G(T) subunit beta-1 (GNB1), Ras-associated protein Rab-1A (RAB1A), guanine nucleotide-binding protein subunit beta-4 (GNB4), 60S ribosomal protein L23 (RPL23), creatine kinase type B (CKB), signal recognition particle 9 kDa protein (SRP9), ubiquitin carboxy-terminal hydrolase isozyme L1 (UCHL1), and / or tubulin folding cofactor B (TBCB).

[0030] As described herein, AAV production and / or assembly can be improved (or increased) in a host cell by modifying the host cell to increase the expression or activity of Hsc / Hsp70, or a cofactor of the Hsc / Hsp70 pathway involved in VP folding (e.g., Bag2, STUB1, DnaJC7, or DnaJA1), vacuole-specific H+ ATPase, and / or cyclin-dependent kinase 2.

[0031] As used herein, the term "modification" or "modified" is understood to include any type of manipulation of a cell that results in an increase or decrease in the expression or activity of one or more of the host proteins described herein. Thus, as used herein, modification of a cell can include, but is not limited to, knocking out or knocking down the expression of an endogenous nucleic acid sequence (e.g., using mutagenesis) or overexpressing an exogenous nucleic acid sequence (e.g., a construct or vector containing a recombinant nucleic acid molecule). Modification of a cell can also include, but is not limited to, exposing the cell (e.g., in culture medium) to one or more chemical compounds to directly or indirectly increase or inhibit the activity of one or more of the endogenous host proteins described herein.

[0032] Representative chemical compounds that can be used to increase or decrease one or more genes or the proteins encoded thereby, including, but not limited to, Cdk1 inhibitor IV, Cdk2 inhibitor II, apoptazole, ML-792, BML282, NSC348884, FDNB, and bafilomycin A1.

[0033] Methods for increasing protein expression or activity are known.For example, the nucleic acid encoding the protein can be overexpressed in host cells.Nucleic acid constructs for overexpressing nucleic acids are known in the art and are commercially available.It is also understood that chemical compounds can be used to stimulate protein expression or activity, and methods for screening compounds that exhibit such activity are known in the art.

[0034] Methods for reducing the expression or activity of a protein are known. For example, the nucleic acid encoding the protein can be mutated to knock out or knock down the expression or activity of the protein. Mutagenesis methods are known in the art and are generally performed using polymerase chain reaction (PCR). It is also understood that chemical compounds can be used to inhibit the expression or activity of a protein, and such methods can be used to inhibit the expression or activity of a protein. Methods for screening compounds for activity are known in the art.

[0035] Also provided are cell lines comprising one or more mutations in Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2; and / or one or more exogenous constructs expressing Hsc / Hsp70 or its cofactors, vacuolar-specific H+ ATPase, and / or cyclin-dependent kinase 2.

[0036] Once cells are modified as described herein, these modified cells can be used to produce one or more AAV vectors under appropriate conditions.AAV vectors (for example, vectors encoding the minimum viral proteins required for the production and assembly of AAV particles) can be wild-type AAV or recombinant AAV (rAAV).Methods for producing AAV and suitable methods for culturing mammalian cells thereof are known in the art, and include transfecting required AAV components, using cell lines that stably express AAV production components, and / or delivering AAV components to cells by infecting cells with heterologous viral vector systems such as herpes simplex virus or adenovirus, as described by Sandoval et al. (2019, Viral Vectors for Gene Therapy).

[0037] AAV assembly can be improved by modifying host proteins or nucleic acids encoding such host proteins. Improved AAV assembly refers to an increase in the number of assembled AAVs, or an increase in the rate or efficiency of assembly, compared to AAVs assembled without modification. The methods described herein result in an increase (e.g., a significant increase) in the titer of the resulting AAV. In some cases, the methods described herein improve the quality and performance of AAV (e.g., increased yield, increased viability, and / or improved infectivity) compared to AAVs produced without modification.

[0038] The proteins identified herein, or nucleic acids encoding such proteins, can be used in the development of cell-free AAV production methods. For example, cell-free systems can be designed and used to produce AAV particles in the absence of actual cells. Such cell-free systems can, for example, contain nucleic acids encoding one or more of these proteins (e.g., at least one, at least two, at least three, at least four, at least five, at least six), or the proteins themselves. For example, one or more of the following proteins, or nucleic acids encoding one or more proteins, can be provided in a cell-free system: CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and / or TBCB.

[0039] The amount of assembled AAV collected after use of the methods described herein is typically greater (e.g., significantly greater) than the amount of assembled AAV collected after AAV infection of cells without the modification. In some cases, the methods described herein result in a significant increase in AAV titer.

[0040] Also provided are products for improving the assembly of adeno-associated viruses (AAV) in cells. Such products typically comprise at least one of (a), (b), or (c). It comprises at least one element selected from at least two types: (a) Hsc / Hsp70 or its cofactor, a nucleic acid encoding Hsc / Hsp70 or its cofactor, or a compound that regulates Hsc / Hsp70 or its cofactor; (b) vacuole-specific H+ ATPase, a nucleic acid encoding vacuole-specific H+ ATPase, or a compound that regulates vacuole-specific H+ ATPase; and (c) cyclin-dependent kinase 2, a nucleic acid encoding cyclin-dependent kinase 2, or a compound that regulates cyclin-dependent kinase 2.

[0041] A representative compound that regulates Hsc / Hsp70 or its cofactors is apoptazole, a representative compound that regulates vacuolar-specific H+ ATPase is bafilomycin A1, and a representative compound that regulates cyclin-dependent kinase 2 is Cdk2 inhibitor II.

[0042] The present invention may employ conventional molecular biology, microbiology, biochemistry, and recombinant DNA techniques within the ordinary skill in the art. Such techniques are fully explained in the literature. The present invention is further described in the following examples, which do not limit the scope of the methods and compositions described in the claims. [Example]

[0043] Example 1 - Screening for factors associated with optimal AAV assembly Briefly, two different screening approaches were performed to identify factors in HEK293 cells involved in AAV assembly and / or DNA packaging. These two properties combine to form production, and modulating one or more of these factors can increase AAV production yields. The two screening approaches were designed as follows:

[0044] Co-immunoprecipitation and mass spectrometry Various viral proteins known to be involved in assembly were tagged with proteins and co-immunoprecipitation was performed after transfection of HEK293 cells. The proteins pulled down in the complex were analyzed by mass spectrometry. These proteins were scored as the most promising hits based on their co-immunoprecipitation with various viral components, their biology, and their low likelihood of being background (e.g., based on control and control mass spectrometry datasets). This method primarily screened for proteins involved in some aspect of protein particle assembly (e.g., protein particle assembly) and cellular cofactors associated with these assembly proteins.

[0045] Sample preparation Previously (Maurer et al., 2019, J. Virol., 93:93(7): doi: 10.1128 / JVI.02013-18) Samples were prepared as described, but the blocking and washing steps were omitted due to the incompatibility of these reagents with LC-MS / MS and downstream analysis.

[0046] Peptide digestion After immunoprecipitation, the proteins bound to the beads were frozen at -80°C in 50 μL of 50 mM Tris. The samples were thawed, and a solution of urea, dithiothreitol (DTT, Thermo Scientific, 20291), trypsin (Promega, V511X), and Tris was added to suspend the beads in 80 μL of 2 M urea, 50 mM Tris HCl (pH 8), 5 μg / mL trypsin, and 5 mM DTT for each of 20 samples. The samples were incubated at 25°C with shaking for 1 hour. The supernatant was transferred to a new tube, and the beads were resuspended in 60 μL of 5.33% Tris HCl. The supernatant was then combined with the wash buffer for each sample. The solution was centrifuged at 5000 rcf for 2 minutes and then transferred to a new tube. Proteins were reduced with 4 mM DTT for 30 minutes at 25°C and then alkylated with 10 mM iodoacetamide (Sigma, A3221) for 45 minutes at 25°C in the dark. Then, 0.5 μg of trypsin was added to each sample, and the samples were digested overnight at 25°C with shaking. Digestion was stopped with 1% formic acid (FA; Fluka, 56302). Samples were desalted using a stage tip containing two Empore C18 punches (3M, 2315). All spins were performed at 1,500 rcf for 2 minutes. The stage chip was conditioned with 1 x 50 μL of 50% acetonitrile (ACN) / 0.1% FA and 2 x 50 μL of 0.1% FA. The sample was then loaded onto the stage chip and spun. The sample was then washed with 2 x 50 μL of 0.1% FA and eluted from the stage chip with 1 x 50 μL of 50% ACN / 0.1% FA and dried.

[0047] TMT labeling and strong cation exchange fractionation For TMT labeling, each sample was reconstituted in 100 μL of 50 mM HEPES. Then, 0.8 mg of TMT10 Isobaric Mass Tag (Thermo Fisher) in 41 μL of 100% ACN was added to each sample. Samples were labeled for 1 hour at 25°C. Labeling efficiency was checked to ensure proper and complete labeling. A mixing control was also performed for each TMT 10-plex sample to ensure that all samples were mixed 1:1. The labeling reaction was stopped with 8 μL of 5% hydroxylamine for 15 minutes at 25°C. The 10-plex samples were then mixed in a volume that provided a 1:1 ratio for all samples and dried. Each 10-plex sample was resuspended in 1 mL of 0.1% FA and desalted using a Sep-Pak C18 column (Waters, 100 mg WAT023590). The column was conditioned with 1 x 1 mL of 100% ACN, 1 x 1 mL of 50% ACN / 0.1% FA, and 4 x 1 mL of 0.1% trifluoroacetic acid (TFA). Each sample was loaded onto the column and washed with 3 x 1 mL of 0.1% TFA and 1 x 1 mL of 1% FA. Peptides were eluted from the column with 2 x 0.6 mL of 50% ACN / 0.1% FA and dried.

[0048] Samples were reconstituted in 200 μL of 3% ACN / 0.1% FA. Half of each sample was dried and analyzed using HPLC-HCD-MS / MS. The other half was dried, resuspended in 250 μL of 0.5% acetic acid (AcOH), and fractionated using strong cation exchange. Samples were fractionated using a stage tip with three SCX punches (3M, 2251) underneath and two C18 punches on top. All spins were performed at 3,500 rcf for 2 minutes. Stage tips were conditioned with 1x 100 μL of methanol (MeOH), 1x 100 μL of 80% ACN / 0.5% AcOH, 1x 100 μL of 0.5% AcOH, 1x 100 μL of 20% ACN / 0.5% AcOH / 500 mM NH4AcO, and 1x 0.5% AcOH. Samples were loaded and spun. The stage tip was washed with 2 × 100 μL of 0.5% AcOH. Peptides were cross-eluted from the C18 punch to the SCX punch using 1 × 100 μL of 80% ACN / 0.5% AcOH. Peptides were then eluted from the stage tip into three fractions using 1 × 50 μL of 20% ACN / 50 mM NH4AcO (pH 5.15), 1 × 50 μL of 20% ACN / 50 mM NH4HCO3 (pH 8.25), and 1 × 50 μL of 20% ACN / 0.1% NH4AcO (pH 10.3). Each eluted fraction was then diluted with 200 μL of 0.5% AcOH and desalted using a two-punch C18 stage tip. The stage chip was conditioned with 1x 100 μL MeOH, 1x 100 μL 80% ACN / 0.5% AcOH, and 2x 100 μL 0.5% AcOH. The sample was loaded and spun. The stage chip was then washed with 2x 100 μL 0.5% AcOH. The samples were eluted from the stage tip with 0 μL of 80% ACN / 0.5% AcOH. Each fraction was dried and resuspended in 9 μL of 3% ACN / 0.1% FA for LC-MS-MS analysis. The unfractionated half of the sample was also resuspended in 9 μL of 3% ACN / 0.1% FA for LC-MS-MS analysis.

[0049] LC-MS / MS and spectral analysis All samples were analyzed using nanoflow HPLC-HCD-MS / MS equipped with an Orbitrap Fusion Lumos mass spectrometer and an Easy-nLC 1200 system. 4 μL of each sample was injected onto a Picofrit column heated to 50°C and self-packed with 1.9 μm C-18 beads at a flow rate of 500 nl / min. LC-MS / MS gradients and flow rates were as previously described (Mertins et al., 2016, Nature, 534:55-62). Spectra were acquired over a 110-minute period. S1 scans were acquired at 60k resolution, a scan range of 350-1800 m / z, and a maximum injection time of 50 ms. Ions were fragmented at a collision energy of 38%. MS2 scans were acquired at 50k resolution, a maximum injection time of 105 ms, and a 0.7 isolation window.

[0050] Data were searched in Spectrum Mill (Agilent) using the Uniprot human database, followed by viral proteins. Fixed modifications (cysteine ​​carbamidomethylation) and variable modifications (N-terminal protein acetylation, methionine oxidation, and TMT 10plex labeling) were searched for. Enzyme specificity was set to LysC / trypsin, and a maximum of three missed cleavages were searched for. The maximum precursor ion charge state was set to 6. The mass error range for precursor and product ions was set to 20 ppm. The calculated false-identification rates for peptides and proteins were less than 1%. Nonhuman and human proteins identified by only a single peptide spectral match were excluded from downstream analysis. A moderated T-test (see software.broadinstitute.org / cancer / software / genepattern / on the World Wide Web) was used to identify statistically enriched proteins in each individual experiment. After correction for multiple comparisons (Benjamini-Hochberg procedure), any protein with a corrected p-value of less than 0.05 was considered statistically enriched. Enriched proteins were normalized to the amount of prey protein in each individual experiment by subtracting the log2 fold enrichment value of the prey from each enriched protein value. Data were compiled and average relative enrichment across the four experimental conditions was visualized using software.broadinstitute.org / morpheus / on the World Wide Web.

[0051] Use of CRISPR / Cas9 AAV was produced in HEK293 cells expressing CRISPR / Cas9 and transfected with a gRNA library containing gRNAs for all known human genes. AAV-producing cells were selected by FACS sorting using an anti-AAV antibody that detects only assembled capsid proteins (i.e., not monomers). Enriched gRNAs in selected cells indicate that the target of this gRNA improved AAV production upon knockout. (a) The significance of enrichment relative to the control, (b) the number of gRNAs against the same gene that exhibited this effect (these libraries overlapped in that they often had six gRNAs per target gene), and (c) the number of gRNAs against the same gene that exhibited this effect were significant. and (c) scoring the most statistically significant hits based on the biology of the gene / protein hits that may be relevant to AAV assembly / packaging. This method examines both assembly and packaging (e.g., entry of viral DNA into preformed particles).

[0052] Figure 1 shows a schematic diagram of the CRISPR screening pipeline. HEK293 cells were transduced at an MOI of <1 with the Brunello library, a lentiviral Cas9+sgRNA library containing an average of four guides targeting each protein-encoding gene. Untransduced cells were removed by antibiotic selection, and the resulting knockout cell library was propagated in culture. Cells were transfected with an adenovirus helper (dF6) and the rep2-cap8 AAV production plasmid. Approximately 800 million cells were collected at each of three time points (12, 18, and 24 hours). Cells were gently fixed and permeabilized (to allow the capsid antibody to reach the nucleus) and then stained with the ADK8 monoclonal antibody, which recognizes a structural epitope present only in assembled capsids, followed by the addition of a fluorescent secondary antibody. Stained cells were subjected to FACS sorting to separate producer cells (POS, containing capsids) from non-producer cells (NEG). Genomic DNA was then extracted, and the integrated guide sequences were amplified by PCR and sequenced by Illumina.

[0053] Next-generation sequencing data analysis Differentially enriched sgRNAs were confirmed using the MAGeCK software package, version 0.5.7 (Li et al., 2014, Genome Biology, 15:554). , after sequencing, mapping, and quantification of sgRNAs (Sanson et al., 2018, Nat. Commun., 9:5416), standardized based on the number of reads per million according to the following formula: Calculated: number of reads per sgRNA / total number of reads in each condition * 10 6To control for the effect of sorting on sgRNA abundance, sgRNA abundance from positive cells was compared with that from negative cells from the same sort. Differential enrichment of sgRNAs at the gene level was then confirmed by Robust Ranking Aggregation (RRA) constructed in MAGeCK.

[0054] Visualization of NGS analysis The results of the statistical analysis of differential enrichment of sgRNAs were visualized using the Matplotlib visualization library (Hunter, 2007, Comp. Science & Eng., 9:90-95), which is written in Python. To visualize the relationships between genes, the genes were plotted in rows in Table 1. Genes were classified by the indicated Gene Ontology terms (Ashburner et al., 2000, Nat. Genet., 25:25-29; The Gene Ontology C., 2017, Nuc. Acids Res., 45:D331-8; Carbon et al., 2009, Bioinform., 25:288-9). The complete list of human genes associated with the above Gene Ontology terms was accessed and downloaded using Amigo 2.5.2 on September 13, 2018.

[0055] The top 1000 genes (ranked by significance of enrichment) were then plotted with dots on the y-axis according to their uncorrected p-values ​​from these multiple tests, and randomly distributed within that category on the x-axis. The size of the dots in the plot varied according to the formula 5 x Ne3, where N is the number of sgRNAs determined to be differentially enriched by upstream analysis.

[0056] [Table 1]

[0057] Example 2 - Identification of assembly and restriction factors by genome-wide screening We selected producer cells from a pooled library of cells exhibiting genome-wide perturbations, with the ultimate goal of creating cell lines or culture conditions that maximize the number of vector-producing cells, the amount of vector produced by each cell, or both. For the initial screen, we used the CRISPR knockout Brunello library (Doench et al., 2014), with the intention of repeating the screen with the CRISPR library. 2016, Nat. Biotechnol., 34:184-91) was selected. AAV2 is the most well-studied serotype and the only one that exhibits strict nucleolar assembly, among other unique phenotypes. AAV8 was selected to maximize the opportunity for new insights and, furthermore, because AAV8 is more phylogenetically close to many natural variants, potentially broadly apply these insights to the production of multiple serotypes. The experimental procedure is shown in Figure 1, where guide sequences serve as readouts for gene knockouts associated with the production cells.

[0058] In fact, no cells staining positive with capsid-specific antibodies were observed prior to 12 hours post-transfection. At 12 hours, less than 1% of the population stained positive for capsid. Within the positive population at 12 hours, no genes were statistically enriched compared to the unsorted population. The 18-hour and 24-hour samples were 6.25% and 6.29% positively stained, respectively, and analysis of the guide sequences enriched or depleted in these samples is shown in Figures 2 and 3.

[0059] The top 1000 genes ranked by significance of enrichment or depletion in the positively sorted populations 18 hours (Figure 2) and 24 hours (Figure 3) after transfection relative to the negative populations were plotted as dots on the y-axis according to their uncorrected p-values ​​from multiple testing, and randomly distributed on the x-axis within their Gene Ontology categories (listed by color below the plot). Dot size in the plots varied according to the formula 5 x Ne3, where N is the number of sgRNAs determined to be differentially enriched by upstream analysis.

[0060] Although several statistically significant guides were unique to only one time point (18 or 24 hours), nine statistically significant guides overlapped both sets (bold in Figures 2 and 3). Two genes, ATF7IP and SETDB1, showed particularly high significance in both samples.

[0061] To further investigate these genes, siRNA knockdown was chosen because it allows for easy transfection of HEK293 cells and offers time, labor, and cost advantages over generating stable cell lines with genetic perturbations. A subset of the top 9 genes identified was selected because they were among the top 65 hits in both samples, suggesting high potential for functional relevance (bold in Figures 2 and 3).

[0062] In preparation for siRNA transfection, transfection reagents and media other than PEI and serum-free DMEM were examined to determine their effect on vector production. Lipofectamine and OptiMEM, known for their high transfection efficiency and commonly used for transfection of small nucleic acids, were tested in parallel and with all permutations of PEI and DMEM (Figure 4). Cells were grown in DMEM + 10% FBS in 6-well plates to 90% confluence before transfection. The same amount of plasmid was used in each condition, and either AAV2 / 2 or AAV2 / 8 rep / cap plasmid or empty plasmid (NEG) was used, as indicated on the x-axis.

[0063] Across all conditions, the ratio of μg of transfection reagent to μg of DNA was maintained at 1.375:1 (x-axis, PEI Max and Lipofectamine). Transfection mixtures were prepared in 100 μL of serum-free DMEM (dark bars) or 100 μL of OptiMEM (light bars). After incubation, the transfection mixture was added to the cells in one of two ways: (a) aspirating the medium from the wells, adding 1.9 mL of the appropriate medium to the transfection mixture, mixing, and adding to the wells (solid bars), or (b) aspirating the medium from the wells and replacing it with 1.9 mL of the appropriate medium, then adding the transfection mixture dropwise onto the medium in each well (patterned bars). In the NEG condition, PEI Max was used with the mixing method. Crude vector preparations were harvested after 48 hours, and DRP was quantified by qPCR. One of the samples was confirmed to be contaminated (indicated by an asterisk).

[0064] Although initial transfection efficiency was very good using Lipofectamine (based on visual assessment of the number and intensity of GFP-positive cells), the titers of AAV2 and AAV8 vectors were significantly lower using Lipofectamine, in some cases more than 100-fold lower. PEI was preferred for use in vector production due to its very low cost, but the findings described herein suggest that Lipofectamine-based transfection has an inhibitory effect on production. Taking these adverse effects into account, PEI, which has a low transfection efficiency, was used in siRNA experiments.

[0065] siRNAs targeting overlapping genes were transfected 24 hours prior to transfection with the vector production plasmid to induce knockdown before the start of production. The production plasmid and siRNA were then co-transfected, and crude preparations were harvested and titrated 48 hours after the second transfection (Figure 5). Briefly, SmartPOOL siRNAs targeting the genes indicated on the x-axis were transfected into HEK293 cells using the drip method with DMEM + 5% FBS. After 24 hours, cells were retransfected with siRNA, and the helper, rep-cap, and ITR.cmv.EGFP.T2A.luciferase.ITR plasmids were added. Crude vector preparations were harvested 48 hours later.

[0066] DRP was quantified by qPCR and reported in two ways: as a ratio relative to a parallel transfection using siRNA not targeting a sequence in the human genome (Figure 5A) and as a ratio relative to a parallel mock transfection (Figure 5B). As a control for the effect of siRNA, one well was transfected in parallel with a non-targeting siRNA, and the other well was not transfected with siRNA (ie, production plasmid only).

[0067] When titers of crude AAV2 and AAV8 preparations were recorded as a percentage of the titer of a non-targeting siRNA control (Figure 5A), no significant increase in vector titer was observed upon knockdown of any of the targeted genes. However, when recorded as a percentage of the no-siRNA control (Figure 5B), AAV2, but not AAV8, showed a significant increase in titer, including a 6.5-fold increase with CNOT6 knockdown. Perhaps the most intriguing observation is that even non-targeting siRNAs modestly increased vector titer, suggesting that activation of the RNAi pathway somehow enhanced production.

[0068] Example 3 - Protein interactions between virus and host proteins in the context of assembly To understand the assembly process and the mechanisms directly involved, we identified host factors that physically interact with AAP and capsid monomers before and during assembly into icosahedrons. Previously, immunoprecipitation of VP-VP-AAP complexes using HA-tagged VP1 as bait did not precipitate fully assembled capsids (Maurer et al., 2018, Cell Resp., 23:1817-30). Furthermore, AAP is a They are not thought to be part of the assembled capsid; however, if so, they may be internal and therefore not accessible to and coprecipitated by antibodies. Therefore, we speculated that proteins coprecipitated by FLAG-AAP or HA-VP1 represent interactions that occur before and during capsid assembly. To maximize insight into this potential mechanism, we designed pull-downs to probe many different interactions (Figure 6).

[0069] Briefly, HEK293 cells were transfected with the expression constructs shown in Figure 6A. Cell lysates were prepared, and total protein was measured by BCA and then diluted to ensure equal total protein concentrations in all samples. Boxes indicate the lysates used for anti-HA (VP bait) pulldown (Figure 6B) and anti-FLAG (AAP bait) pulldown (Figure 6C). Two percent of the input lysates (numbered in Figure 6A) from duplicate experiments (indicated by "A" and "B" above the lanes) were electrophoresed and stained with SYPRO Ruby to reveal total protein (Figure 6B) or probed for VP and AAP by Western blot (Figure 6C). Equal volumes of lysates were then subjected to immunoprecipitation (as described in Figure 6A). Equal volumes of the eluate were loaded into each lane, electrophoresed, and stained with SYPRO Ruby to reveal total protein (Figures 6D and 6F) or probed for VP and AAP by Western blot (Figures 6E and 6G).

[0070] These experiments allowed us to examine cellular proteins that bind to VP monomers (sample 2), AAP (sample 6), and VP-AAP complexes or VP oligomers (sample 3), and also to examine how the interaction partners of VP change in the presence or absence of AAP, and vice versa (comparing sample 2 or 6 with sample 3).

[0071] The N-terminal third of AAP (AAPN) contains a hydrophobic region and a conserved core and has been shown to be important for the joint function of AAP and VP (Maurer et al., 2018, supra; Tse et al., 2018, J. Virol., 92(14):doi:10.,1128 / JVI.00393-18), and the C-terminal part of AAP (AAPC) is involved in nucleolar signaling and nuclear transport. The signal contained a signal (Earley et al., 2015, J. Virol., 89:3038-48). Therefore, we speculated that cellular proteins responsible for the nuclear translocation of AAP and / or VP might bind to AAPC, with AAPN being primarily responsible for direct binding to VP. To investigate this, FLAG-tagged truncated AAP (AAPN and AAPC) were included in transfections and pulldowns (Figure 6, Samples 4 and 5, respectively). Co-immunoprecipitations were performed in both directions using HA-VP1 or FLAG-AAP as bait. Nonspecific binding proteins were also assessed by anti-HA and anti-FLAG pulldowns using control samples expressing untagged VP and AAP (Samples 1 and 7), which were used as background and subtracted from the other samples.

[0072] In these experiments, AAV8 VP1 was used because we used AAV8 in the CRISPR screen, allowing us to compare potential hits in both methods (e.g., comparing two different serotypes would not be appropriate). AAP2 was used because it robustly trans-complements AAV8 production (Maurer et al., 2014). et al., 2018, supra; Earley et al., 2017, J. Virol., 91(3):doi: 10.1128 / JVI.01980-16; Grosse et al., 2017, J. Virol., 91(20):doi: 10.1128 / JVI.01198-17; Sonntag et al., 2011, J. Virol., 85:12686-97), and in the past, no AAP cells were identified from any serotype. Because we have studied AAP2 in great detail, given that no cytosolic binding partners have been identified, we thought it appropriate to report our conclusions and hypotheses about the interactions of AAP with cellular proteins.

[0073] Biological duplicate transfections and immunoprecipitations (Fig. 6) were performed, and differential banding was observed between pulldowns (Fig. 6D and 6F), indicating the presence of unique cellular proteins in each condition. LC-MS / MS analysis was performed on the eluates, and statistically significantly enriched proteins under each condition were normalized to the bait protein and summarized as heat maps in Fig. 7A (showing proteins pulled down for HA-VP1, and thus binding partners of VP1) and Fig. 7B (showing proteins pulled down for FLAG-AAP2, and thus binding partners of AAP2). Row names in Fig. 7A and Fig. 7B include the Uniprot accession number, gene symbol, protein name, and the number of unique peptides identified for that gene product. Column designations in Fig. 7A and Fig. 7B include the tagged bait protein and other viral proteins expressed in each condition (as described in Fig. 6).

[0074] The identified binding partners are unique among proteins previously identified in vector preparations. Furthermore, the proteins coprecipitated by VP are distinct from those in AAP pulldowns, and this set varies depending on the coexpression of viral proteins. For example, VP1 coprecipitates with Bag2, STUB1, and DnaJC7 except when full-length AAP or AAPN is coexpressed. Conversely, AAP coprecipitates with DnaJA1 only when VP1 is coexpressed, indicating that the binding of DnaJA1 to VP1 is dependent on AAP. These four proteins are cofactors for the ubiquitously expressed heat shock cognate 70 and heat shock protein 70 (Hsc / Hsp70), essential chaperone proteins that directly catalyze protein folding. Although Hsp70 was detected in the pull-downs, true enrichment in specific pull-downs was difficult to determine because the expression level of Hsp70 in HEK293 cells was very high, driving up background levels and causing more frequent nonspecific binding events.

[0075] In addition, the interaction of the protein with VP or AAP is not significantly affected by the presence of each other. Some proteins are also involved. Nucleophosmin, a nucleolus-enriched protein shown to interact with intact capsids (Dong et al., 2014, PLoS One, 9:e86453) and to be required for nuclear / nucleolar import during infection (Bevington et al., 2007, Virology, 357:102-13; Johnson and Samulski, 2009, J. Virol., 83:2632-44), was coprecipitated by full-length AAP2, and more specifically, by AAPC but not by AAPN, in the presence and absence of VP1. Importin-5, a member of the importin beta family that mediates import through the nuclear pore complex, showed a similar pattern to nucleophosmin, binding to AAP (and AAPC, but not AAPN) even in the presence of VP. Importin-beta has been shown to colocalize with incoming AAV2 particles, but specific members of this family have not been investigated (Nicolson and Samulski, 2014, J. Virol., 88:4132-44). These results suggest that the C-terminal sequence of AAP mediates nuclear / nucleolar transport of AAP and co-transport of VP proteins for assembly in the nucleolus. Indeed, the nuclear and nucleolar localization signals described for AAP2 are located in the C-terminus (Earley et al., 2014, J. Virol., doi:10.1128 / jvi.03125-14). Other classes of proteins that are coprecipitated by AAP, such as creatine kinase B and myotrophin, were not pulled down except when VP1 was coexpressed. These and many other proteins identified by the MS approach described herein have not previously been linked to the AAV replication cycle. Some of the identified binding partners are poorly characterized genes / proteins, such as FXR1 and FXR2.

[0076] Example 4 - Functional evaluation of selected hits by chemical inhibition As a first step to demonstrate the effect of binding partners on vector production, we tested all commercially available hits against their corresponding chemical inhibitors. We also tested bafilomycin A1, which inhibits the vacuolar-specific H+ ATPase, the five subunits of which were significantly depleted in the CRISPR screen (Figures 2 and 3). Bafilomycin A1 has also been shown to rescue VP proteins in the absence of AAP (Maurer et al., 2018, ). (mentioned above).

[0077] Inhibitors were serially diluted in medium at a range of concentrations centered around the dosage recommended by the manufacturer for use in cultured cells. Medium containing inhibitors was added to wells, and transfection mixtures containing either AAV2 (light bars) or AAV8 (dark bars) production plasmids were added dropwise on top. Vectors were harvested 48 hours later (Fig. 8A). Inhibition of Hsc / Hsp70, nucleophosmin, and creatine kinase B at the highest dosage, as well as bafilomycin A1 at all dosages, reduced vector production to background levels. To examine whether any of these effects resulted from poor transfection efficiency rather than direct inhibition of capsid assembly, we repeated the experiment by adding inhibitors of upstream factors critical for vector production 4 hours after the transfection step was initiated by the addition of the transfection mixture (Fig. 8B).

[0078] The x-axis shows the concentration of inhibitor (μM) in DMEM + 10% FBS. To assess the assay background (NEG), positive control transfections received only DMSO instead of the cap gene (DMSO), and negative controls received DMSO and empty plasmid. Crude vector preparations were harvested 48 hours post-transfection and DRP quantified by qPCR. Viable cells were assayed at 48 hours using the Cell Titer Glow kit, and values ​​were plotted using a standard curve of cells plated at 2-fold dilutions in the same plate. Total transfection efficiency was assessed visually by imaging EGFP-positive cells (EGFP indicates the transgene). (The gene is present in a plasmid sandwiched between ITRs.) Cell health was assessed visually by brightfield imaging and quantitatively by ATP levels in drug-treated wells (lightly shaded, Figure 8B). Furthermore, some inhibitor concentrations were adjusted because they were ineffective in the first experiment, and some of these concentrations may have been higher than would have guaranteed specificity for the target enzyme.

[0079] Results from the second of two experiments, normalized to viable cell counts for each inhibitor, are shown in Figures 9-16 along with corresponding microscopy images. In these experiments, production plasmids were transfected into HEK293 cells, and serial dilutions of the appropriate inhibitor were added 4 hours later at the concentrations indicated on the x-axis. Inhibitors were added at the same concentrations to parallel plates, and cell viability after 48 hours of incubation was assessed by quantifying ATP levels using Cell Titer Glo®. Immediately prior to vector harvest from the wells, total transfection efficiency (EGFP; present in the plasmid with the transgene sandwiched between the ITRs) and cell morphology (bright field) were assessed by microscopy at the concentrations indicated on the x-axis at 48 hours. AAV2 (light bars) and AAV8 (dark bars) vector titers were quantified by qPCR against DRP. The genome copy number (GC) in each preparation was normalized to the number of viable cells examined for each condition, and the GC per viable cell was reported as a ratio to the positive control transfection (DMSO) without inhibitor. The unnormalized data for these experiments are shown in Figure 9B.

[0080] Inhibition of Cdk1 with Cdk1 inhibitor IV affected vector titers, but this was accompanied by a decrease in both cell health and transfection efficiency (Figure 9). Thus, these experiments cannot confirm or refute the possibility that Cdk1 may be involved in vector production. A significant decrease in AAV8 titer without an effect on cell viability was observed at 6.25 μM for AAV8 only when the inhibitor was added 4 h posttransfection, but this concentration had no effect on AAV2 production. Considering that Cdk1 was identified as a potential binding partner when AAV8 VP1 was used as bait, this may suggest a serotype-specific preferential interaction with AAV8 over AAV2.

[0081] Inhibition of Cdk2 with Cdk2 inhibitor II had a dose-dependent effect on both AAV2 and AAV8 production at high concentrations, particularly at 90–270 μM, without affecting cell viability (assayed by ATP) or total transfection efficiency (by EGFP visualization) (Figure 10). Cell morphology was affected at 270 μM, with apparently live GFP+ cells clumping toward the center of each well. The nearly 10-fold decrease in AAV8 titer suggests a role for Cdk2 in production, although further experiments are required to fully rule out nonspecific effects at these high drug concentrations.

[0082] Although there are no commercially available inhibitors for Bag2, STUB1, DnaJC7, or DnaJA1, Apoptazole inhibits the chaperones Hsc70 and Hsp70, thereby documenting the function of all four co-chaperones. Inhibition of Hsc / Hsp70 with Apoptazole resulted in a highly dose-dependent decrease in vector production (down to 5% for AAV8) without any apparent decrease in transfection efficiency, but dramatically altered cell morphology (Figure 11). At 300 μM, cells clustered toward the center of the well but still displayed robust EGFP expression and no significant cell death as assayed by ATP abundance. Obviously, inhibition of this chaperone, which is critical for the folding of the majority of cellular proteins, may indirectly affect other mechanisms required for production. Further studies are needed to confirm the direct involvement of Hsc / Hsp70 in VP folding. These results reveal that Hsc / Hsp70-related functions are important for production.

[0083] Inhibition of SAE1 with ML-792 (Figure 12) or UCHL1 with BML282 (Figure 13) had a significant effect on vector titer, which was accompanied by a dramatic decrease in EGFP expression, suggesting that this effect on titer may have resulted from poor transfection.

[0084] Nucleophosmin (NPM1) has a role as a chaperone for folding and nuclear / nucleolar transport, particularly in the context of nascent ribosome biogenesis. NPM1 has previously been shown to interact with Rep proteins and assembled capsids, although the experimental design of these studies likely did not address AAP-NPM1 interactions (Dong et al., 2014). et al., 2014, PLoS One, 9:e86453; Bevington et al., 2007, Virology, 357:102-13 Pharmacological inhibition of NPM1 with NSC348884 immediately before transfection significantly affected vector titers (Fig. 8A), but this effect was not seen when it was added 4 h after transfection (Fig. 8B and Fig. S14), suggesting a very early role for NPM1.

[0085] Creatine kinase B (CKB) plays a key role in metabolism, particularly in regenerating cellular ATP in situ at sites of rapid consumption. Inhibition of CKB by FDNB appears to increase with cellular vector production (Fig. 15), which may be related to the indirect quantification of viable cells by measuring ATP levels; inhibition of ATP regeneration may artificially reduce viable cell numbers and therefore increase genome copy numbers per viable cell.

[0086] The vacuole-specific H+ ATPase (V-ATPase) is a multisubunit enzyme complex that catalyzes the translocation of protons across membranes of intracellular compartments, affecting their internal pH. V-ATPase is important for the acidification of lysosomes, endosomes, and autophagosomes, and its inhibition by bafilomycin A1 inhibits AAV infection (Bartlett et al., 2000, J. Virol., 74:2777-85; Sonntag et al., 2006, J. Virol., 80:11040-54). Bafilomycin A1 also inhibits PEI-mediated transfection of plasmid DNA (You and Auguste, 2010, Biomater., 31:6859-66), explaining the abolition of vector production observed when V-ATPase was inhibited immediately prior to transfection (Figure 8A) and the depletion of subunits from genome-wide CRISPR knockout screens (Figures 2 and 3). A 10-fold decrease in vector titer was observed when a small amount of bafilomycin A1 was added 4 hours post-transfection, without any significant visual effect on transfection efficiency (Figure 16). Treatment with bafilomycin A1 rescues degradation of VP proteins of some serotypes in the absence of AAP (Maurer et al., 2018, supra), suggesting that acidification of the digestive tract has an antagonistic effect on AAV production. However, in the presence of AAP, such as in (Sonntag et al., 2011, J. Virol., 85:12686-97), this acidification also appears to be important for production. These results suggest that endosomal, lysosomal, and / or autophagosome compartments have complex roles in all aspects of the AAV replication cycle and as sites of recombinant vector production.

[0087] Example 5 - Proposed Model Figure 17 is a schematic diagram showing a hypothetical model of AAV assembly incorporating multiple proteins of interest. Nascent VP proteins (far left) are translated in the cytoplasm and degraded by the proteasome in the absence of AAP. In the presence of AAP, VP oligomerizes into assembly intermediates that are transported into the nucleus for full capsid assembly. It is not known whether AAP co-transports VP as monomers or oligomers. All other proteins shown in the figure were identified in a series of pulldowns where they were assembled in hypothetical mechanisms based on their presence or absence in the samples (e.g., Figure 7) and their known functions from the literature.

[0088] Genome-wide CRISPR knockout screening yielded at least two hits (ATF7IP and SETDB1) in both the 18-hour and 24-hour samples, with highly statistically significant results. These proteins coexist within the HUSH complex, a repressive chromatin regulator that transcriptionally silences active loci by trimethylating H3K9; these proteins are functionally codependent (24). One possible model is that the HUSH complex silences the expression of viral proteins from transfected plasmids, or HUSH could silence the expression of other factors that promote capsid assembly.

[0089] These proteomics studies yielded more hits (with statistical significance) than CRISPR screening, and several hits that differentially bind to viral proteins have roles in feeding different branches of the same pathway, such as DnaJA1, DnaJC7, BAG2, and STUB1. Hsc / Hsp70 activity depends on cochaperones, such as J-proteins, which stimulate ATP hydrolysis by Hsc / Hsp70 and increase substrate binding affinity, and BAG family proteins, which function as nucleotide exchange factors. J-proteins also directly bind to Hsc / Hsp70 substrates, indicating that DnaJ proteins are involved in Hsp70 substrate selectivity. These cochaperones and Hsc / Hsp70 together can drive substrate folding but may also be involved in substrate degradation. STUB1 is an E3 ligase that acts as a negative regulator of protein stability by ubiquitinating incompletely or improperly folded Hsc / Hsp70 substrates and targeting them for proteasomal degradation. BAG2, together with DnaJ proteins, stimulates the ATPase activity of Hsp70 and inhibits the ubiquitination activity of STUB1. DNAJC7 has been suggested to play a role in stabilizing substrates by inhibiting their degradation by the proteasome, and it has been proposed to act as a recycling chaperone by returning improperly folded substrates to earlier stages of the folding pathway. The differential binding of VP1 to DNAJA1, BAG2, and STUB1 in the presence and absence of AAP suggests a direct involvement of AAP in the selectivity, interaction, and functionality of co-chaperones with Hsc / Hsp70. Furthermore, this suggests a functional network in which AAP acts upstream to inhibit proteasomal degradation of VP proteins.

[0090] Other embodiments While the methods and compositions are described herein with several different embodiments, it is understood that the above descriptions of these various embodiments are intended to be illustrative and not limiting of the scope of the methods and compositions. Other embodiments, advantages, and modifications are within the scope of the following claims.

[0091] Disclosed are methods and compositions that can be used to prepare for, or can be used in conjunction with, or can be used in the preparation of the products of the disclosed methods and compositions, or such products. These and others are disclosed herein, and it is understood that combinations, subsets, interactions, groups, etc. of these methods and compositions are disclosed. That is, although specific reference to these compositions and methods individually, or to their various collective combinations and permutations, may not be expressly disclosed, each of these is specifically disclosed. are contemplated and described herein. For example, where a particular composition or method is disclosed and discussed and multiple compositions or methods are discussed, each and every combination and permutation of those compositions and methods is specifically contemplated unless specifically indicated otherwise. Likewise, any subset or combination of these is also specifically contemplated and disclosed.

Claims

1. A method for improving the assembly of adeno-associated virus (AAV) within cells, A step of producing modified cells by increasing the expression or activity of vacuole-specific H+ ATPase, cyclin-dependent kinase 2, and / or Hsc / Hsp70 or its cofactors within the cell, The process involves producing infected modified cells by infecting the modified cells with an AAV vector, The process of culturing the aforementioned infected modified cells, A method comprising the step of assembling an AAV.

2. The method according to claim 1, wherein the amount of assembled AAV collected is greater than the amount of assembled AAV collected after infecting cells without the modification with AAV.

3. The method according to claim 1 or 2, wherein as a result of the method, the titer of AAV is significantly increased.

4. A method for improving the assembly of adeno-associated virus (AAV), A step of producing AAV from modified cells having AAV, wherein the cells are CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, A process in which one or more genes selected from FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and TBCB are modified to exhibit an increase or decrease in the protein encoded by them, and A method comprising the step of assembling the AAV.

5. The method according to claim 4, wherein the modified cells are genetically modified cells.

6. The method according to claim 5, wherein the genetically modified cells include knockout, knockdown, overexpression, or a combination thereof.

7. The method according to claim 4, wherein the modified cells include a chemical compound that increases or decreases one or more of the genes or proteins encoded by them.

8. The method according to claim 7, wherein the chemical compound is selected from the group consisting of Cdk1 inhibitor IV, Cdk2 inhibitor II, apoptasol, ML-792, BML282, NSC348884, FDNB, and bafilomycin A1.

9. A cell-free culture system for assembling adeno-associated viruses, Culture medium and A cell-free culture system comprising at least two proteins selected from CHMP7, CEP72, CNOT6, SLC9A6, PPHLN1, ATF7IP, SETDB1, GPR89B, KIF16B, ATAD3A, BAG2, STUB1, DNAJA1, DNAJC7, HSPA8, HSPA1A, CDK1, CDK2, CHCHD2, C1QBP, DPYSL5, FXR1, FXR2, IPO5, LBR, MTPN, NPM3, NPM1, PPL, SNX3, UBE2O, SAE1, CCDC124, GNB1, RAB1A, GNB4, RPL23, CKB, SRP9, UCHL1, and TBCB, or nucleic acids encoding the at least two proteins.

10. The cell-free culture system according to claim 9, comprising at least three types (for example, at least four types, at least five types, etc.) of proteins, or nucleic acids encoding the at least three types of proteins.

11. One or more mutations in vacuole-specific H+ ATPase, cyclin-dependent kinase 2, and / or Hsc / Hsp70 or its cofactors, and / or A cell line comprising one or more exogenous constructs expressing vacuole-specific H+ ATPase, cyclin-dependent kinase 2, and / or Hsc / Hsp70 or its cofactors.

12. The cell line according to claim 11, wherein the mutation includes a knockout mutation.

13. The cell line according to claim 11, wherein the mutation includes a knockdown mutation.

14. The cell line according to claim 11, wherein the one or more exogenous constructs are recombinant constructs.

15. A product for improving the assembly of adeno-associated virus (AAV) in cells, comprising at least one element from at least two of (a), (b), or (c): (a) vacuole-specific H+ ATPase, nucleic acids encoding vacuole-specific H+ ATPase, or compounds that modulate vacuole-specific H+ ATPase; (b) Cyclin-dependent kinase 2, nucleic acids encoding cyclin-dependent kinase 2, or compounds that modulate cyclin-dependent kinase 2; and (c) Hsc / Hsp70 or its cofactor, nucleic acids encoding Hsc / Hsp70 or its cofactor, or compounds that modulate Hsc / Hsp70 or its cofactor.

16. The product according to claim 15, wherein the compound that modulates Hsc / Hsp70 or its cofactor is apoptazole.

17. The product according to claim 15, wherein the compound that modulates vacuole-specific H+ ATPase is bafilomycin A1.

18. The product according to claim 15, wherein the compound that modulates cyclin-dependent kinase 2 is a Cdk2 inhibitor II.

Images