Characterization of impurities in adeno-associated virus (AAV) samples and formulation compositions to stabilize aav

Abstract

Methods for characterizing DNA impurities in adeno-associated virus (AAV) samples or biopharmaceuticals are provided including the use of size exclusion chromatography and spectrophotometry. Methods and compositions are also provided to minimize leakage of packed DNA from AAV vector including the use of excipients, such as a sugar, an amino acid, a surfactant, or polyols.

Description

[Technical Field]

[0001] Cross-reference to related applications

[0002] This application claims priority and benefit to U.S. Provisional Patent Application No. 63 / 108,480, filed November 2, 2020, which is incorporated herein by reference.

[0003] This invention generally relates to methods for characterizing nucleic acid impurities in samples containing adeno-associated virus (AAV) vectors using particle size chromatography and spectrophotometry. This application also provides methods and compositions for minimizing leakage of packaged DNA from AAV vectors. [Previous Technology]

[0004] Gene therapy mediates therapeutic effects by delivering appropriate amounts of therapeutic genes to target tissues without significant toxicity, and by integrating genetic material into the host genome to achieve long-term gene expression, for example. Gene therapy can include nucleic acids, plasmids, viruses, vectors, or genetically engineered microorganisms. Among all currently available viral vectors, including retroviruses, lentiviruses, adenoviruses, and AAV vectors, AAV vectors have been widely used for delivering genetic material in gene therapy.

[0005] The evaluation of gene therapy products includes assessing the manufacturing process to ensure product safety and quality, as well as evaluating the product during its manufacturing, storage, and distribution steps. Since gene therapy products are typically stored in cryosol form, it is important to verify the stability of gene therapy products through freeze-thaw cycles and agitation pressure. AAV vector-based gene therapy products can undergo degradation during freeze-thaw cycles, producing degradation products containing degraded viral proteins and DNA impurities. The packaged AAV genome within the vector may leak out during freeze-thaw cycles or agitation. Therefore, various formulation strategies for gene therapy products should be investigated to prepare formulations that increase product stability and minimize product degradation.

[0006] It will be understood that there is a need for methods for characterizing nucleic acid impurities in such gene therapy products during various storage conditions. Furthermore, there is a need to provide compositions or formulations comprising AAV vectors containing therapeutic genes, such compositions or formulations minimizing leakage of packaged DNA from the AAV vector. [Summary of the Invention]

[0007] This application provides a method for characterizing and identifying nucleic acid impurities in samples including AAV vectors or gene therapy products, comprising using particle size chromatography and spectrophotometry. This application also provides a method and composition for minimizing leakage of packaged DNA from the AAV vector under storage conditions, such as freeze-thaw cycles and agitation pressures that may occur during manufacturing, transportation, storage, and dispensing.

[0008] This disclosure provides a method for identifying nucleic acid impurities in a sample containing an AAV vector. In some exemplary embodiments, this disclosure provides a method comprising: contacting a sample containing an AAV vector with a particle size separation chromatography (SEC) column; washing the SEC column with a solution to provide at least one precipitate; and identifying nucleic acid impurities in the at least one precipitate using a spectrophotometer. In one aspect, the nucleic acid impurities are single-stranded DNA (ssDNA) genomes of AAV. In another aspect, the nucleic acid impurities are double-stranded DNA (dsDNA). In another aspect, the method of this application further comprises measuring the absorbance of the at least one precipitate at 280 nm using a spectrophotometer and measuring the absorbance ratio of the at least one precipitate at 260 / 280 nm using a spectrophotometer. In one aspect, the method of this application further comprises quantifying nucleic acid impurities based on a nucleic acid standard curve. In one aspect, the method of this application further comprises treating the nucleic acid impurities with a fluorescent dye that specifically binds to ssDNA. In another aspect, the method of this application further includes contacting the sample with a DNA nuclease before contacting the sample with the SEC column. In another aspect, the SEC system is a particle size sieving ultra-high performance liquid chromatography (SE-UPLC) system. In another aspect, the method of this application further includes performing next-generation sequencing (NGS) on at least one elution solution.

[0009] This disclosure provides at least in part a composition for protecting an AAV vector from leakage of packaged nucleic acids. In some exemplary embodiments, this disclosure provides a composition comprising at least one AAV vector and at least one excipient; wherein the at least one excipient is a sugar, amino acid, surfactant, or polyol; wherein the at least one AAV vector comprises packaged nucleic acids; and wherein the at least one AAV vector in the composition is protected from leakage of the packaged nucleic acids.

[0010] In one aspect, the AAV vector in the composition of this application is exposed to at least one freeze-thaw cycle. In one aspect, the packaged nucleic acids are the ssDNA genome of AAV, wherein the packaged nucleic acids are present within the AAV capsid. In one aspect, the excipients in the composition of this application are present at a concentration of about 0.001% to about 10%, wherein the composition further includes phosphate-buffered saline and a nonionic surfactant. In another aspect, the sugars in the composition of this application are sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose, or maltoheptaose. In one aspect, the amino acid in the composition of this application is proline. In one aspect, the surfactant in the composition of this application is poloxamer 188 (Pluronic® F68). In one aspect, the surfactant in the composition of this application is a nonionic surfactant, wherein the surfactant is present at a concentration of about 0.001% to about 0.2%. In another aspect, the composition includes sucrose and poloxamer 188. In a particular aspect, the composition includes sucrose at a concentration of 2.5% to 10% and poloxamer 188 at a concentration of 0.001% to 0.2%.

[0011] This disclosure provides at least in part a method for protecting an AAV vector from leakage of packaged nucleic acids. In some exemplary embodiments, the method includes: obtaining a sample containing at least one AAV vector, and adding a stabilizing component to the sample to form a protective formulation, wherein the protective formulation protects the at least one AAV vector from leakage of packaged nucleic acids, the stabilizing component comprising at least one excipient, and the excipient being a sugar, amino acid, surfactant, or polyol. In one aspect, the AAV vector is exposed to at least one freeze-thaw cycle. In another aspect, the packaged nucleic acids are ssDNA genomes of AAV, wherein the packaged nucleic acids are present within the AAV capsid.

[0012] In another aspect, the excipient is present in the protective formulation at a concentration of about 0.001% to about 10%, wherein the stabilizing component further comprises phosphate-buffered saline and a nonionic surfactant. In one aspect, the sugar is sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose, or maltoheptaose. In yet another aspect, the amino acid is proline. In one aspect, the surfactant is poloxamer 188 (Pluronic® F68). In one aspect, the surfactant in the composition of this application is a nonionic surfactant, wherein the surfactant is present in the protective formulation at a concentration of about 0.001% to about 0.2%. In another aspect, the stabilizing component comprises sucrose and poloxamer 188. In one particular aspect, the stabilizing component comprises sucrose at a final concentration of 2.5% to 10% in the protective formulation and poloxamer 188 at a final concentration of 0.001% to 0.2% in the protective formulation.

[0013] These and other aspects of the invention will be better understood and appreciated when considered in conjunction with the following description and accompanying drawings. While the following description indicates various embodiments and numerous specific details thereof, it is given by way of illustration only and not by way of limitation. Many substitutions, modifications, additions, or rearrangements may be made within the scope of the invention.

Implementation Method

[0027] Gene therapy has opened up broad prospects for altering, repairing, or replacing damaged or mutated genes to improve patients’ health. Adeno-associated virus (AAV) vector systems are ideal tools for gene delivery and have become excellent vectors for human gene therapy. AAVs can be engineered to perform specific functions to deliver nucleic acids in gene therapy applications. The main advantages of AAVs include low immunogenicity and pathogenicity, as well as high overall safety and stability (Naso et al., BioDrugs 31(4) (2017) 317-334; During, Adv Drug Deliv Rev 27(1) (1997) 83-94). AAV can infect numerous cells, regardless of whether those cells are actively dividing, and induces a mild immune response, thereby prolonging the duration of transgenic expression (During; Carter, Curr Opin Biotech 3(5) (1992) 533-539; Daya and Berns, Clin Microbiol Rev 21(4) (2008) 583-593). Furthermore, AAV is not associated with any known disease (Naso et al.; During; Carter; Daya and Berns; Xiao et al., Adv Drug Deliv Rev 12(3) (1993) 201-215; Muzyczka, Viral Expression Vectors, Springer 1992, pp. 97-129). Another advantage of AAV as a gene delivery agent is its low risk of random genomic integration, which may disrupt gene function and potentially trigger insertional mutations (Goswami et al., Front Oncol 9 (2019) 297; Nguyen et al., Blood 134(Supplement 1) (2019) 611-611; Lundstrom, Diseases 6(2) (2018) 42).

[0028] AAV is a non-enveloped virus and a non-pathogenic member of the Dependovirus genus within the Parvoviridae family. It requires an auxiliary agent, such as an adenovirus or herpesvirus, to achieve infection (Venkatakrishnan et al., Structure and Dynamics of Adeno-Associated Virus Serotype 1 VP1 - Unique N-Terminal Domain and Its Role in Capsid Trafficking, Journal of Virology, May 2013, Vol. 87, No. 9, pp. 4974-4984). AAV encapsulates an approximately 4.8 kb ssDNA genome in an icosahedral capsid, which is composed of a protein coat called capsid virus protein. Figure 1 shows the structure of AAV1, including the crystal structure of the AAV1 capsid VP3 monomer and the surface representation of the AAV1 capsid (Venkatakrishnan et al.). In addition to forming the outer capsid, capsid viral proteins actively participate in cell binding and internalization.

[0029] AAV is composed of three capsid proteins, namely VP1, VP2 and VP3, and a single-stranded DNA (ssDNA) is coated in the capsid (Agbandje-McKenna and Kleinschmidt, Adeno-Associated Virus, Springer 2012, pp. 47-92; Drouin and Agbandje-McKenna, Future virology 8(12) (2013) 1183-1199). After binding to the cell receptor, AAV is then endocytosed and released from the endosome, subsequently transported to the host cell nucleus and uncoated to release ssDNA from its capsid. Logically, transduction efficiency depends at least in part on the amount of coated ssDNA genome and the integrity of AAV before reaching the release site (Nonnenmacher and Weber, Gene Ther 19(6) (2012) 649-658; Hauck et al., J Virol 78(24) (2004) 13678-13686; Thomas et al., J Virol 78(6) (2004) 3110-3122).

[0030] To monitor the quality and efficiency of recombinant AAV as a gene delivery vector, it is important to monitor the purity, capsid identity, vector particle valence, and empty / total ratio of the AAV biopharmaceutical. Demonstrating the stability of the AAV biopharmaceutical under various storage conditions (such as stable temperature storage and freeze-thaw cycles), diluents (such as serum and solutes with different pH values), and dosing conditions is also important. Gene therapy products, such as AAV biopharmaceuticals, are typically stored in frozen solution form. Therefore, freeze-thaw cycles are unavoidable during manufacturing, product dosage preparation, and dosing. In rare cases, temperature deviations during transport or refrigeration unit malfunctions during storage can expose biopharmaceuticals to additional freeze-thaw cycles. AAV biopharmaceuticals may undergo degradation due to decreased stability during freeze-thaw cycles under storage conditions.

[0031] In academic research laboratories, after making numerous attempts to design and develop AAVs as gene delivery vectors, researchers generally store these vectors in common buffers, such as phosphate-buffered saline (PBS), with a relatively high concentration of glycerol added as a cryoprotectant, and store the AAV material at -80°C. Although this has become a standard procedure to some extent, these conditions may be detrimental to AAV function (Boyd, Gene Therapy Technologies, Applications and Regulations (1999) 383; Croyle et al., Gene Ther 8(17) (2001) 1281-1290). In some cases, it may also be necessary to dilute adequately before administration to reduce the toxicity of the cryoprotectant (Lee et al., Journal of Assisted Reproduction and Genetics 23(2) (2006) 87-91; Armitage et al., Cryobiology 50(1) (2005) 17-20). Studies have shown that the valence and transduction efficiency of AAV decrease after multiple freeze-thaw cycles; however, the reasons for this phenomenon and the mechanism of AAV degradation after freeze-thaw cycles are not fully understood (Croyle et al.; Rodrigues et al., Pharm Res, 2019, 36 (2):29; Howard and Harvey, Human Gene Therapy Methods 28(1) (2017) 39-48).

[0032] This disclosure describes a method for characterizing impurities in AAV samples. Using the method of this invention, AAV8 samples subjected to freeze-thaw cycles were characterized, and it was found that increasing the number of freeze-thaw cycles led to an increase in product-related impurities, which could be detected by particle size sieving ultra-high performance liquid chromatography (SE-UPLC). Further characterization of these AAV8 samples showed that the impurities mainly consisted of genomic DNA leaked from the AAV capsid. Similar phenomena were observed in other AAV serotypes.

[0033] When the AAV capsid is damaged during freeze-thaw cycles, the packaged ssDNA AAV genomic DNA within the capsid can leak out, as shown in Figure 2. The presence of free ssDNA observed in the AAV biopharmaceutical indicates decreased stability of the AAV biopharmaceutical during storage. Various excipients can be added to the gene delivery product to increase product stability and minimize product degradation. These excipients may include buffers, lyophilization protectants, tensioning agents, and surfactants (Rodrigues et al.). Additional formulations will be described in detail below.

[0034] A method is needed to characterize DNA impurities in AAV samples or biopharmaceuticals during various storage conditions. Furthermore, a stabilizing component or formulation of the AAV biopharmaceutical should also be provided to minimize leakage of packaged DNA from the AAV vector during storage conditions.

[0035] This application provides a method for identifying DNA impurities, such as the ssDNA genome of AAV, in AAV samples or biopharmaceuticals, comprising using an SEC to separate the DNA impurities from the AAV vector. Subsequently, the SEC fraction is monitored using a spectrophotometer. This application also provides a method and composition for minimizing leakage of packaged DNA from the AAV vector, comprising using an excipient in the presence of a phosphate buffer, salt, and a nonionic surfactant.

[0036] There is a need to develop rapid, reliable, sensitive, and high-throughput methods for identifying and monitoring impurities in AAV biopharmaceuticals. Any component in purified AAV biopharmaceuticals that is not the desired product, not a product-related substance, or not an intended excipient can be considered an impurity. Examples of vector product-related impurities include empty AAV capsid particles and AAV particles with capsids encapsulating unintended nucleic acid fragments. Additional DNA impurities may include ssDNA of the AAV genomic body that leaks from the AAV capsid during, after, or during storage. Impurities may include helper virus-dependent replication competent AAV particles, which may be unintentionally generated by recombination events in the biosynthetic environment of the vector generation system.

[0037] In addition, impurities in AAV biopharmaceuticals may also include residual proteins and nucleic acids derived from cell cultures used to produce AAV. Abundant residual protein impurities may include host cell proteins and bovine serum albumin. Abundant residual nucleic acids may include host cell DNA / RNA and DNA from accessory components, such as plastids or viruses. Residual host cell DNA may exist in two forms, including nuclease sensitivity program-related impurities and nuclease resistance product-related impurities. Nuclease sensitivity program-related impurities include nucleic acids non-specifically co-purified with the desired AAV vector product. Nuclease resistance product-related impurities include nucleic acids encapsulated within AAV particles. Minimizing these different forms of residual host cell DNA requires different manufacturing process optimization strategies. (JF Wright, Product-related impurities in clinical-grade recombinant AAV vectors: characterization and risk assessment, Biomedicines, 2014, 2, 80-97) Removal of DNA impurities from AAV biopharmaceuticals is quite complex, even when nuclease treatment is used during AAV vector purification to remove accessible nucleic acids. Undesirable DNA fragments can be packaged to make them resistant to nuclease treatment due to the integrity of the vector particles.

[0038] Since the AAV capsid encapsulates the ssDNA genome in an icosahedral capsid composed of a protein shell such as capsid viral proteins, the capsid must be very stable to protect the AAV genome until a suitable host cell appears, and the initiating cell enters to release the AAV genome for replication. The assembly of AAV viral capsid proteins can have a significant impact on viral infectivity and vector efficacy (Jin et al., Direct liquid chromatography / mass spectrometry analysis for complete characterization of recombinant adeno-associated virus capsid proteins, Human gene therapy methods, 2017, Vol. 28, No. 5, pp. 255-267). AAV capsid proteins play a role in the second-strand synthesis and transcription of the initiating introgression genome (Salganik et al., Adeno-associated virus capsid proteins may play a role in transcription and second-strand synthesis of recombinant genomes, Journal of Virology, January 2014, Vol. 88, No. 2, pp. 1071-1079). Degradation of capsid proteins can negatively impact AAV infectivity, potentially leading to leakage of the AAV ssDNA genome. The AAV ssDNA genome can also be ejected without damaging the capsid (Bernaud, Julien et al., Characterization of AAV vector particle stability at the single-capsid level. Journal of Biological Physics, 2018, Vol. 44, No. 2, pp. 181-194).

[0039] The AAV ssDNA genome contains three genes: rep (replication), cap (capsid), and aap (assembly). The rep gene is involved in viral genome replication and packaging. The cap gene encodes capsid viral proteins. The expression of the cap gene produces capsid viral proteins including VP1, VP2, and VP3, which are produced from alternating spliced ​​mRNAs with a common C-terminus. VP3 is approximately 61 kDa and accounts for approximately 85% of the capsid protein content. VP2 is approximately 73 kDa. VP1 is approximately 87 kDa. VP1 and VP2 contain N-terminal extensions (VP1u), which include the phospholipase A2 domain and nuclear localization signals (Rayaprolu et al., Comparative analysis of adeno-associated virus capsid stability and dynamics, Journal of Virology, December 2013, Vol. 87, No. 24, pp. 13150-13160). The AAV capsid is composed of 60 viral protein monomers, including VP1, VP2, and VP3. VP3 is the major capsid protein. There are approximately 50 VP3 replicas in the AAV capsid. There are also approximately 5 VP1 replicas and 5 VP2 replicas in the AAV capsid (Venkatakrishnan et al.).

[0040] This application provides a method for characterizing, identifying, and / or quantifying DNA impurities in AAV samples or biopharmaceuticals, comprising separating DNA impurities from AAV vectors using a SEC. Subsequently, the SEC fraction is monitored using a spectrophotometer. This application also provides a method and composition for minimizing leakage of packaged DNA from AAV vectors, comprising using excipients in the presence of PBS and a nonionic surfactant. In some exemplary embodiments, this application provides a method for identifying DNA impurities in a sample containing an AAV vector, comprising: contacting the sample with an SEC column, washing the SEC column with a solution to provide a precipitate, and identifying nucleic acid impurities in the precipitate using a spectrophotometer. The sample containing the AAV vector is loaded onto the SEC column. The fraction passing through the SEC column is monitored using a spectrophotometer to measure the absorbance of the SEC fraction at 280 nm. A main peak corresponding to the AAV vector can be observed in the SEC fraction in the SEC precipitate curve. When detected using a spectrophotometer, the main peak exhibits a considerably high absorbance at 280 nm. A small peak corresponding to a DNA impurity is observed in the SEC fraction within the SEC dissolution curve. This small peak also exhibits absorbance at 280 nm when detected using a spectrophotometer.

[0041] In one aspect, the DNA impurities in the sample containing the AAV vector are the ssDNA genome of AAV. In another aspect, a spectrophotometer is used to analyze the SEC fraction to measure the absorbance ratio at UV 260 / 280 nm to assess nucleic acid purity. In another aspect, the method of this application includes using a spectrophotometer to measure the absorbance of the precipitate at 280 nm and to measure the absorbance ratio of the precipitate at 260 / 280 nm. In another aspect, the method of this application further includes quantifying DNA impurities based on a nucleic acid standard curve and treating the DNA impurities with a fluorescent dye.

[0042] In some exemplary embodiments, this application provides methods and stabilizing compositions for protecting AAV vectors from leakage of packaged nucleic acids during storage conditions, such as freeze-thaw cycles and agitation. In some aspects, the stabilizing composition comprising an AAV vector and an excipient includes sugars, amino acids, surfactants, or polyols. In some aspects, the method for protecting an AAV vector from leakage of packaged nucleic acids includes mixing the AAV vector with the stabilizing composition.

[0043] In one aspect, the excipient in the stabilizing composition of this application is present at a concentration of about 0.001% to about 10%, wherein the stabilizing composition further includes phosphate buffered saline and a nonionic surfactant. In one aspect, the sugar in the stabilizing composition of this application is sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose, or maltoheptaose. In another aspect, the amino acid in the stabilizing composition of this application is proline. In yet another aspect, the surfactant in the stabilizing composition of this application is Pluronic® F68. In one aspect, the surfactant in the stabilizing composition of this application is a nonionic surfactant, wherein the surfactant is present at a concentration of about 0.001% to about 0.2%.

[0044] The need to evaluate gene delivery products to ensure product safety and quality has led to an increasing demand for characterizing DNA impurities in AAV samples or biopharmaceuticals under various storage conditions. Furthermore, there is a need to provide compositions or formulations of AAV biopharmaceuticals that minimize leakage of packaged DNA from the AAV vector. The exemplary embodiments disclosed herein fulfill these needs by providing methods and compositions that meet long-standing requirements.

[0045] The term "a (kind)" should be understood to mean "at least one (kind)"; and those generally skilled in the art should understand that the terms "about" and "approximately" should be understood to mean permissible standard deviation; and when a range is provided, the endpoint is included. As used herein, the term "include / includes / including" is intended to be non-restrictive and should be understood to mean "comprise / comprises / comprising" respectively.

[0046] In some exemplary embodiments, this application provides a method for identifying nucleic acid impurities in a sample containing an AAV vector. The method includes: contacting the sample containing the AAV vector with a particle size sieving chromatography (SEC) column, washing the SEC column with a solution to provide at least one precipitate, and identifying nucleic acid impurities in the at least one precipitate using a spectrophotometer.

[0047] As used herein, the term “vector” refers to a recombinant plasmid or virus that carries nucleic acid in vitro or in vivo for delivery to a host cell.

[0048] Vectors derived from AAV are particularly attractive for delivering genetic material because (i) they can infect (transduce) a variety of non-dividing and dividing cell types, including myofibrils and neurons; (ii) they do not contain viral structural genes, thereby eliminating the natural host cell response to viral infection, such as interferon-mediated responses; (iii) wild-type viruses have never been associated with any human pathology; (iv) replication-defective AAV vectors generally maintain a free genomic form compared to wild-type AAVs that can integrate into the host cell genome, thereby limiting the risk of insertional mutations inducing or activating oncogenes; and (v) compared to other vector systems, AAV vectors do not trigger a significant immune response (see ii), thereby enabling long-term expression of therapeutically transfected genes (provided that their gene products are not rejected).

[0049] As used herein, the term "impurity" means any component in a purified AAV sample that is not the desired product, not a product-related substance, or not an intended excipient for formulation. DNA impurities may include ssDNA of the AAV genomic body that leaks from the AAV capsid during, after, or during storage. Impurities may include helper virus-dependent replication competent AAV particles that may be unintentionally generated by recombination events in the biosynthetic environment of the vector generation system. Impurities remaining after AAV purification may include residual amounts of proteins and nucleic acids derived from components used to generate AAV in cell cultures. Residual protein impurities may include host cell proteins and bovine serum albumin. Residual nucleic acids may include host cell DNA / RNA and DNA from helper components, such as plastids or viruses. Residual host cell DNA present may be in two forms: (1) nuclease sensitivity program-related impurities, such as impurities nonspecifically co-purified with the desired AAV vector product; and (2) nuclease resistance product-related impurities, such as impurities encased in the capsid of AAV particles (JF Wright).

[0050] As used herein, "nucleic acid" refers to a DNA or RNA molecule. In some exemplary embodiments, the term nucleic acid capture includes the sequence of any one of known DNA and RNA base analogs, such as, but not limited to, 4-acetylcytosine, 8-hydroxy-N6-methyladenosine, aziridinylcytosine, pseudoisocytosine, 5-(carboxyhydroxymethyl)uracil, 5-fluorouracil, 5-bromouracil, 5-carboxymethylaminomethyl-2-thiouracil, 5-carboxymethyl-aminomethyluracil, dihydrouracil, inosine, N6-isopentenyladenine, 1-methyladenine, 1-methylpseudouracil, 1-methylguanine, 1-methylinosine, 2,2-dimethylguanine, 2-methyladenine, 2-methylguanine, 3-methylcytosine, 5-methylcytosine. N6-methyladenine, 7-methylguanine, 5-methylaminomethyluracil, 5-methoxy-amino-methyl-2-thiouracil, β-D-mannosinolate queosine, 5'-methoxycarbonylmethyluracil, 5-methoxyuracil, 2-methylthio-N6-isopentenyladenine, methyl uracil-5-oxyacetate, uracil-5-oxyacetic acid, oxybutyroxyglycoside, pseudouracil, queosine, 2-thiocytosine, 5-methyl-2-thiouracil, 2-thiouracil, 4-thiouracil, 5-methyluracil, methyl uracil-5-oxyacetate, uracil-5-oxyacetic acid, pseudouracil, queosine, 2-thiocytosine, and 2,6-diaminopurine.

[0051] As used herein, the term "particle size sieving chromatography" or "SEC" refers to a chromatographic method, such as molecular sieve chromatography, that separates molecules in solution based on size (e.g., molecular weight) by filtering through a matrix. The matrix, such as a gel, consists of spherical beads containing pores of a specific size distribution. When molecules of different sizes are incorporated into or expelled from the pores within these beads, the molecules can be separated according to their size. Smaller molecules can diffuse into the pores, thereby increasing the time required for smaller molecules to pass through the SEC column. However, larger molecules do not enter the pores. When macromolecules do not enter the pores, they precipitate in the void volume of the SEC column. Generally, macromolecules pass through the SEC column for a shorter time than smaller molecules. Therefore, because molecules of different sizes pass through the SEC column in different times, molecules can be separated by size via the SEC column. SEC can be used to separate macromolecules or macromolecular complexes, such as protein complexes. Typically, the buffer solutions used in SEC are formulated to preserve the native structure and conformation of biomolecules in solution, thereby allowing the separation of biomolecules without disturbing their native structure and conformation.

[0052] As used herein, the term "spectrometer" includes instruments capable of quantitatively measuring the intensity of absorbed light according to wavelength, such as absorption spectrophotometers. A spectrophotometer includes a monochromator or prism for generating a light beam containing a desired wavelength, and components for measuring the ratio of the intensity of the light beam as it enters and exits a colorimetric tube or flow cell. In a UV-Vis spectrophotometer, a light beam from a suitable UV and / or visible light source is passed through a prism or diffraction grating monochromator. The light then passes through the sample to be analyzed before reaching the detector. A spectrophotometer may also be part of a SEC-HPLC / UPLC system and serve as the detector. Exemplary Examples

[0053] The embodiments disclosed herein provide methods for characterizing and identifying DNA impurities in AAV samples or biopharmaceuticals. The embodiments disclosed herein also provide methods and compositions for minimizing leakage of packaged DNA from AAV vectors.

[0054] In some exemplary embodiments, this application provides a method for identifying nucleic acid impurities in a sample containing an AAV vector. The method includes: contacting the sample containing the AAV vector with an SEC column; washing the SEC column with a solution to provide at least one elution solution; and identifying nucleic acid impurities in the at least one elution solution using a spectrophotometer. In one aspect, the nucleic acid impurities are single-stranded DNA genomes of AAV.

[0055] In some exemplary embodiments, this application provides a composition for protecting AAV vectors from leakage of packaged nucleic acids, wherein the composition includes at least one excipient comprising sugars, amino acids, surfactants, or polyols. In one aspect, the concentration of the excipient in the composition of this application is about 0.001% to about 10%, about 0.001% to about 8%, about 0.001% to about 6%, about 0.05% to about 5%, about 0.1% to about 10%, about 0.1% to about 8%, about 0.1% to about 6%, about 0.1% to about 5%, about 5%, about 4%, about 3%, about 2.5%, about 2%, about 1.5%, about 1%, about 0.1%, about 0.05%, or about 0.001%.

[0056] In one aspect, the composition further includes phosphate-buffered saline and a nonionic surfactant. In another aspect, the sugars in the composition of this application are sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose, or maltoheptaose. In another aspect, the amino acid in the composition of this application is proline. In yet another aspect, the surfactant in the composition of this application is Pluronic® F68. In one aspect, the surfactant in the composition of this application is a nonionic surfactant, wherein the surfactant is present at a concentration of about 0.001% to about 0.2%, about 0.1%, about 0.05%, or about 0.001%.

[0057] It should be understood that the system is not limited to any of the aforementioned adeno-associated virus, particle size sieve chromatography, spectrophotometer, ssDNA, AAV vector and excipient.

[0058] The sequential designation of the method steps provided herein by numbers and / or letters is not intended to limit the method or any embodiment thereof to the particular order indicated. This specification references numerous disclosures throughout, including patents, patent applications, published patent applications, registration numbers, technical papers, and academic papers. Each of these referenced references is incorporated herein by full text and is used for all purposes. Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. This disclosure will be more fully understood with reference to the following examples, which are provided to describe this disclosure in more detail. These examples are intended to illustrate the scope of the invention and should not be construed as limiting the scope of the invention. Examples

[0059] Spectrophotometer: A spectrophotometer is used to detect and monitor proteins and nucleic acids in fractions obtained from particle size separation chromatography columns. Protein detection is performed using absorbance at 280 nm. DNA or RNA detection is performed using absorbance at 260 nm. Proteins in solution absorb ultraviolet light and have maximum absorbance at 280 nm and 200 nm. Amino acids with aromatic rings have an absorbance peak at 280 nm. Nucleic acids can be quantified by measuring the amount of ultraviolet radiation absorbed by the bases at 260 nm using a spectrophotometer. DNA nucleotides contain a sugar backbone, bases, and phosphate groups. Bases are nitrogen-rich and absorb light at 260 nm. For an optical path length of 1 cm, the optical density (OD260, i.e., absorbance at 260 nm) of the following solutions at 260 nm is equal to 1.0: 50 μg / mL double-stranded DNA solution, 33 μg / mL single-stranded DNA solution, 20-30 μg / mL oligonucleotide solution, or 40 μg / mL RNA solution.

[0060] DNA samples may contain impurities that can affect absorbance at 260 nm. The absorbance ratio at 260 / 280 nm is analyzed to assess nucleic acid purity. Pure DNA has an absorbance ratio of approximately 1.8–2.0 at 260 / 280 nm. Pure RNA has an absorbance ratio of approximately 2.0 at 260 / 280 nm. High-quality DNA or RNA samples should have an absorbance ratio higher than approximately 1.7 at 260 / 280 nm. A low absorbance ratio at 260 / 280 nm indicates potential contamination by proteins or phenols. A maximum absorbance at 280 nm indicates potential protein contamination. A maximum absorbance at 230 nm indicates potential phenolic or thiocyanate contamination. A maximum absorbance at 340 nm indicates light scattering caused by particles.

[0061] Sample preparation. For freeze-thaw studies, 100 µL of AAV solution in a 1 mL glass vial is frozen at -80°C for at least one hour and then thawed at room temperature for at least 30 minutes to complete one cycle.

[0062] To characterize the effect of benzonase on SE-UPLC peak 1, benzonase® (Millipore Sigma, Burlington, MA) was added to the AAV8.GFP solution at a volume ratio of 1:50 and mixed thoroughly before SE-UPLC analysis was performed.

[0063] For formulation screening studies, different AAV serotypes were adjusted to a gene titer of 3.7 × 10¹² vg / mL. Then, a 2-fold concentrated formulation buffer was mixed with samples of different AAV serotypes to obtain the target formulation, and these samples were subjected to freeze-thaw cycles. Analysis was performed at t=0 and after 1, 2, 4, or 10 freeze-thaw cycles.

[0064] SE-UPLC. AAV8.GFP samples were analyzed using a Sepax SEC 5 µm, 4.6 × 300 nm column with a 500 Å pore size, operated on a Waters ACQUITY UPLC system. Calibration curves were constructed by injecting 1 µL, 5 µL, 10 µL, 15 µL, and 20 µL of 1.83 × 10¹² vg / mL standards. Subsequently, 10 µL of AAV8 sample was injected and dissolved at a flow rate of 0.5 mL / min using 2 × DPBS as the mobile phase. Dissolution was monitored using a photodiode array detector, and spectra in the 210 nm to 400 nm range were recorded for further analysis.

[0065] To partially separate different species in the SE-UPLC dissolution profile, the Waters Fraction Manager-Analytical was connected to the ACQUITY UPLC system and used to collect the fractions dissolved at the corresponding residence times. Multiple repeated injections were performed, and the fractions obtained from the multiple injections were pooled together and then concentrated 5-fold (volume reduction) using an Amicon Ultra-0.5 centrifugal filtration unit. Next-generation sequencing (NGS) was performed on the collected fractions to determine their identity.

[0066] Next-generation sequencing. Sample preparation was performed using a slightly modified protocol described by Lecomte et al. (Lecomte et al., Molecular Therapy-Nucleic Acids 4 (2015) e260). DNA isolation was performed without any deoxyribonuclease treatment to identify impurity peaks (peak 1 in the SE-UPLC curve preceding the main precipitation peak). Illumina libraries were prepared using the Nextera XT DNA library preparation kit, following the manufacturer's instructions. The final pool was sequenced using a mid-output kit with 150 cycles in the NextSeq 550. Sequencing reads were analyzed using an internal data analysis pipeline.

[0067] Fluorescence-based cell-free DNA analysis. The Quant-iT™ OliGreen® ssDNA analysis kit (Invitrogen™) was purchased from Thermo Fisher Scientific. Before use, warm the 200× stock Quant-iT OliGreen ssDNA reagent and 100 µg / mL oligonucleotide standard to room temperature. Dilute the 20× stock TE buffer 20-fold with Milli-Q water to the working concentration (10 mM Tris-HCl, 1 mM EDTA, pH 7.5). Then, according to the user guide, dilute the reagent and the oligonucleotide standard to the working concentration using TE buffer. Using the ratio in the user guide, mix the AAV and oligonucleotide standard samples with the reagent, and use a standard linear curve generated from five samples containing different concentrations (0 ng / mL, 10 ng / mL, 100 ng / mL, 500 ng / mL, and 1000 ng / mL) of oligonucleotide standards to extrapolate the ssDNA in the AAV samples. The samples were placed in a 384-well plate and analyzed using a Synergy™ Neo2 multimode reader (BioTek). The excitation and emission wavelengths were set to 480 nm and 520 nm, respectively. Example 1. Characterization of DNA impurities in AAV samples.

[0068] AAV serotype 8 (AAV8) is generated in HEK-293 cells and has been reported to contain at least 5 × 10¹² genome copies / mL in PBS containing 0.001% Pluronic® F68. DNA impurities were analyzed using AAV8 loaded with GFP (green fluorescent protein), such as AAV8-GFP. Different amounts of AAV8-GFP samples, such as 1 µL, 5 µL, 10 µL, 15 µL, or 20 µL, were analyzed using SEC columns. The AAV8-GFP samples were loaded onto particle size separation chromatography (SEC) columns. The absorbance of the SEC flow through the SEC column was measured at 280 nm using a spectrophotometer, as shown in Figure 3. In Figure 3, the Y-axis represents the absorbance at 280 nm in absorbance units (AU), and the X-axis represents the residence time of the flow through the column. As shown in Figure 3, in the SEC fraction of the SEC dissolution curve, a main peak corresponding to the AAV8-GFP viral particles was observed at a residence time between 12 and 15 minutes. This main peak exhibited considerable absorbance at 280 nm when detected using a spectrophotometer. In the SEC dissolution curve, a smaller peak with absorbance at 280 nm was observed at a residence time between 8 and 11 minutes, such as peak 1 in Figure 3.

[0069] The fluorescence intensity of the SEC flow fraction under 280 nm excitation and 350 nm emission was also monitored, as shown in Figure 4. In Figure 4, the Y-axis represents the normalized fluorescence intensity and the X-axis represents the residence time of the flow fraction through the SEC column. The flow fraction corresponding to the main peak in Figure 3 exhibits a similar peak with considerably high fluorescence intensity at the same residence time, i.e., a residence time between 12 minutes and 15 minutes, as shown in Figure 4. The flow fraction corresponding to peak 1 in Figure 3 did not show detectable fluorescence intensity at the corresponding residence time, i.e., a residence time between 8 minutes and 11 minutes (under 280 nm excitation and 350 nm emission), as shown in Figure 4.

[0070] In addition, the absorbance ratio of peak 1 and the main peak of the SEC fraction was measured at UV 260 / 280 nm using a spectrophotometer to assess nucleic acid purity. As shown in Table 1, peak 1 of the SEC fraction showed a UV 260 / 280 nm reading of 1.7, indicating that the DNA had good purity and was free of protein contamination. The main peak of the SEC fraction showed a UV 260 / 280 nm reading of 1.3, indicating the presence of protein. The UV-vis of the whole DS (active drug containing AAV biopharmaceutical) showed a UV 260 / 280 nm reading of 1.3, corresponding to the purified AAV vector. When the AAV8-GFP sample was treated with nuclease, the intensity of peak 1 decreased significantly. The results indicate that the SEC fraction corresponding to peak 1 contains DNA. The results further indicate that peak 1 of the SEC fraction may contain ssDNA of the AAV genome. Table 1. Absorbance ratio of SEC fractions at 260 / 280 nm sample UV 260 / 280 nm SEC Peak 1 1.7 SEC Summit 1.3 Overall DS UV-vis 1.3

[0071] Since the results indicate that peak 1 contains high-purity DNA and is free of protein contamination, the combination of SEC and absorbance measurement can effectively quantify ssDNA impurities in AAV samples. AAV samples can be loaded onto an SEC column to separate DNA impurities from the AAV vector. The absorbance of the SEC fraction at 280 nm can be monitored to identify DNA impurities and AAV vectors in the SEC precipitation curve. Subsequently, the SEC fraction containing DNA impurities can be quantified by measuring the absorbance at 260 nm. The purity of DNA can be assessed by measuring the absorbance ratio at 260 / 280 nm. DNA impurities in purified AAV samples may include ssDNA that has leaked from the AAV capsid during non-standard storage conditions such as freeze-thaw cycles and agitation. Example 2. Characterization of ssDNA leakage after freeze-thaw cycles.

[0072] Currently, AAV gene therapy drugs are stored in cryogenic liquid form to maintain product quality and stability; therefore, AAV particles undergo multiple freeze-thaw cycles during manufacturing, transportation, storage, and administration. Therefore, multiple freeze-thaw cycles were selected as a stress condition in the characterization and stability studies to understand the stability of AAV8 formulations and identify potential degradation pathways.

[0073] AAV8 containing a GFP payload was used as the raw material to characterize ssDNA leakage after freeze-thaw cycles. This analysis was performed using a raw material containing 8.34 × 10¹² vg / mL AAV8-GFP in PBS at pH 7.4 containing 0.001% pluronic F68. Samples containing different AAV valences were prepared using this raw material. The raw material was concentrated using an Amicon centrifuge filter to achieve the desired target concentration (higher concentration), such as valences at 2× (2-fold concentration), 4×, and 10×. Freeze-thaw cycles were performed by freezing the samples containing the AAV vector at -80°C for one hour or longer, followed by thawing at room temperature for 0.5 hours or longer. Unless otherwise described, eight freeze-thaw cycles were performed for sample processing.

[0074] AAV8-GFP samples processed with eight freeze-thaw cycles were loaded onto an SEC column containing raw material (AAV8.GFP in Figure 5) and samples with different valences, such as at 2×, 4×, and 10×, as shown in the SEC chromatography diagram in Figure 5. The absorbance of the SEC stream at 280 nm was measured by monitoring the stream through the SEC column using a spectrophotometer. In Figure 5, the Y-axis represents the normalized absorbance at 280 nm (UV), and the X-axis represents the residence time of the stream through the SEC column. As shown in Figure 5, a main peak corresponding to the AAV8-GFP viral particles was observed in the SEC stream in the SEC chromatography curve at a residence time between 12 and 15 minutes. This main peak had a relatively high absorbance at 280 nm, as detected by a spectrophotometer. In the SEC dissolution curve, a small peak with absorbance at 280 nm was observed at a residence time between 8 and 11 minutes, which is marked as peak 1 in Figure 5.

[0075] After eight freeze-thaw cycles, the size of peak 1 in the SEC dissolution curve of the AAV8-GFP sample significantly increased. The increase in peak 1 size depends on the concentration of AAV8-GFP, such as the valence of AAV. Table 2 shows the changes in the size of peak 1 in the SEC dissolution curve before and after eight freeze-thaw cycles. The results show that the size of peak 1 significantly increased after freeze-thaw treatment, indicating that the ssDNA of the AAV genome was released from the viral capsid after freeze-thaw treatment. In addition, the peak 1 and main peak of the SEC fraction were analyzed using a spectrophotometer to measure the absorbance ratio at UV 260 / 280 nm to assess nucleic acid purity. As shown in Table 3, peak 1 of the SEC fraction showed a UV 260 / 280 nm reading of 2.1, indicating that the DNA had good purity and contained very little protein. The main peak of the SEC fraction showed a UV 260 / 280 nm reading of 1.4, indicating the presence of protein. The overall DS UV-vis display shows a UV 260 / 280 nm reading of 1.3, corresponding to viral particles.

[0076] In addition, dynamic light scattering (DLS) was used to determine the size distribution curve of the AAV carrier spheres in the sample. Table 4 shows the hydrodynamic radius (in nm) of the AAV carrier before and after eight freeze-thaw (F / T) cycles detected by DLS. Table 2. Changes in the size of peak 1 before and after freeze-thaw treatment. Changes in the size of peak 1 in SEC flow Power Price Peak percentage before treatment 1% Peak percentage after treatment 1% Peak 1 concentration (mg) % raw materials 7.8% 27.9% 0.008 2× 6.8% 11.0% 0.011 4× 6.8% 10.7% 0.013 10× 4.8% 8.5% 0.049 Table 3. Absorbance ratio of SEC fractions at UV 260 / 280 nm. sample UV 260 / 280 nm SEC Peak 1 2.1 SEC Summit 1.4 Overall DS UV-Vis 1.3 Table 4. Hydrodynamic radius of AAV carrier Hydrodynamic radius (nm) measured by DLS Power Price 8× F / T before treatment After treatment 8× F / T raw materials 13.1 13.4 2× 13.1 12.6 4× 13.0 12.3 10× 13.2 12.7

[0077] The release of ssDNA from the AAV viral capsid after freeze-thaw treatment was characterized and quantified using a fluorescent dye that specifically binds to ssDNA. Following freeze-thaw (F / T) treatment, the fluorescent dye was added to the AAV8-GFP sample. Samples containing different concentrations of synthetic oligonucleotides were prepared as standards to generate standard curves. Standard curves were prepared by mixing 25 µL of ssDNA standard solution with 25 µL of 1× dye solution to obtain ssDNA concentrations ranging from 10 ng / mL to 1000 ng / mL. The standards were analyzed to detect the fluorescence of 50 µL samples in a 384-well dish at 480 nm excitation and 520 nm emission.

[0078] The concentration of ssDNA in AAV8-GFP samples was quantified by multiple freeze-thaw cycles, such as five freeze-thaw cycles (F / T 5×) or ten freeze-thaw cycles (F / T 10×), or by horizontal shaking for 48 hours. Figure 6 shows the quantification of ssDNA concentration in AAV8-GFP samples after freeze-thaw treatment and 48 hours of shaking. As shown in Figure 6, the amount of ssDNA in AAV8-GFP samples increased after freeze-thaw treatment and shaking. The Y-axis in Figure 6 represents the ssDNA concentration in ng / mL.

[0079] To further evaluate the effect of freeze-thaw cycles on ssDNA leakage, SE-UPLC was used to measure the changes in the content of size variants in AAV8.GFP samples caused by freeze-thaw pressure. The SE-UPLC dissolution curve of the untreated AAV8.GFP sample at t=0 showed a main peak at a residence time of approximately 7.9 minutes, and a pre-dissolution peak at approximately 6.3 minutes, referred to as peak 1 in this paper, as shown in Figure 7. Notably, after ten freeze-thaw cycles, the peak area and peak height of peak 1 increased. Since peak 1 dissolves earlier than the main peak in the SE-UPLC curve, a possible explanation for this is that the peak at approximately 6.3 minutes represents high molecular weight species, such as aggregates of the AAV capsid.

[0080] To further characterize the species in peak 1, the UV signals collected from the SE-UPLC chromatography of AAV8.GFP samples at 260 nm and 280 nm were analyzed. Generally, samples with a UV260 / 280 ratio of 0.6 are considered "pure" proteins, and samples with a UV260 / 280 ratio of 1.8 are considered "pure" DNA (Wilfinger et al., Biotechniques 22(3) (1997) 474-481; Porterfield and Zlotnick, Virology 407(2) (2010) 281-288). For AAV8 samples analyzed by SE-UPLC, the UV260 / 280 ratio of peak 1 was 1.8, and the UV260 / 280 ratio of the main peak was 1.2 (Table 5). The observed UV260 / 280 ratio of the main peak indicates the contribution of both protein and DNA to UV absorbance, consistent with the main peak representing the AAV capsid encapsulating the ssDNA genome. The observed UV260 / 280 ratio of peak 1 indicates DNA impurities rather than AAV aggregates, attributed to the contribution of protein absorbance at 280 nm; these AAV aggregates actually have a lower UV260 / 280 ratio. Table 5. UV260 / 280 ratios of the leading and main peaks in the SE-UPLC chromatography of AAV8.GFP. UV260 / 280 front peak UV260 / 280 main peak AAV8 t=0 1.8 1.2 AAV8 10× Freeze-Thaw 1.9 1.1

[0081] To confirm the identity of this impurity peak, additional characterization was performed using SE-HPLC (SE-HPLC-FLR) connected to a fluorescence detector, nuclease treatment, and next-generation sequencing (NGS). First, the fluorescence detector was connected to the liquid chromatography system, and fluorescence signals were collected at 280 nm excitation and 350 nm emission. Using the fluorescence detector, only the main peak was observed, and peak 1 was not observed in the UV chromatography. This observation further confirmed that the impurity peak was composed of DNA, rather than a protein-containing species. Second, the AAV8.GFP sample was treated with benzonase. Benzonase is a well-known DNA nuclease and has been widely used to digest DNA in samples for various purposes (Sastry et al., Human Gene Therapy 15(2) (2004) 221-226; Antonioli et al., Journal of Chromatography A 1216(17) (2009) 3606-3612; Konz et al., Biotechnology Progress 21(2) (2005) 466-472). As shown in Figure 8, the intensity of peak 1 at approximately 6.3 minutes decreased significantly (to approximately baseline level), and the height of the main peak also decreased. This data indicates that the impurity peaks precipitated before the main species in SE-UPLC are actually DNA.

[0082] To identify peak 1 at the molecular level, the Illumina platform was used to selectively enrich the fractions collected from elution at 5.8 min to 7.2 min by concentrating them 5-fold for NGS analysis (Table 6). Following the protocol of Lecomte et al. for single-stranded DNA virus sequencing (SSV-Seq) (Lecomte et al., Adeno-Associated Virus Vectors, Springer 2019, pp. 85-106), an NGS library was prepared using a concentration pool. The SSV-Seq protocol consists of four experimental steps: (Step 1) DNA extraction from the AAV pool; (Step 2) synthesis of a second DNA strand; (Step 3) preparation of the Illumina sequencing library; and (Step 4) high-throughput sequencing based on the Illumina platform. Data analysis was performed using an internal dedicated bioinformatics workflow. SSV-Seq is a sequence-agnostic technique (as opposed to qPCR) because it can detect and identify DNA species without any prior knowledge (such as the knowledge required to generate primers). Table 6. NGS results of the peaks before enrichment in the SE-UPLC chromatogram of AAV8.GFP. copy# Total original reading passage Overall filtering of the mapping segment Reference Name Mapping to reference reading segment Percentage of mapped read segments Copy 1 18410428 17716518 pAdHelper_VVT 69904 0.61 Rep-Cap8_VVT 129735 1.13 CAG-eGFP 11158267 96.98 CAG-eGFP main chain 148288 1.29 Ad5_E1AB 0 Copy 2 18526989 17876959 pAdHelper_VVT 69957 0.59 Rep-Cap8_VVT 126958 1.07 CAG-eGFP 11534278 97.12 CAG-eGFP main chain 145542 1.23 Ad5_E1AB 0 Copy 3 18835821 18179263 pAdHelper_VVT 71006 0.59 Rep-Cap8_VVT 130130 1.08 CAG-eGFP 11672300 97.09 CAG-eGFP main chain 148659 1.24 Ad5_E1AB 0

[0083] Given that benzonase treatment was observed to eliminate impurity peaks, DNA isolation was performed without any deoxyribonuclease treatment. Reads were assigned to each of the following reference sequences: (i) the CAG-eGFP bGHpA genotype containing the ITR; (ii) the bacterial backbone of the vector plasmid; (iii) the complete Rep-Cap8 plasmid; (iv) the complete helper plasmid; (v) a fragment of the Ad5 genotype integrated into the HEK293 packaging cell line genotype; and (vi) the human genotype major assembly GRCh38.

[0084] The vast majority of reads obtained by NGS mapped to the CAG-eGFP genome (approximately 97%), followed by a smaller number matching the vector plasmid backbone (approximately 1.2%), Rep-Cap8 (approximately 1-1.1%), and accessory plasmids (0.6%). No reads mapped to the human genome. These observations definitively identify the DNA impurity as an AAV vector genome.

[0085] To quantify the absolute amount of cell-free DNA in AAV samples, the well-established Quant-iT ssDNA analysis was used. In this analysis, an external fluorescent dye was added to the test sample to amplify the fluorescence signal of ssDNA after excitation. Standard curves with various concentrations of oligonucleotides were established and used to calculate the DNA concentration in unknown samples. As shown in Figure 9, the Quant-iT ssDNA detection analysis demonstrated that after ten freeze-thaw cycles, the absolute amount of cell-free ssDNA in the AAV8.GFP sample in the basal formulation (1×DPBS, 0.001% Pluronic F68) increased 3.6-fold, from 364 ng / mL to 1303 ng / mL. The control sample of AAV8.GFP in the basal formulation contained 34,000 ng / mL ssDNA after heating at 95°C for 10 minutes. Assuming heat treatment at 95°C releases 100% of the coated ssDNA, these data indicate that after ten freeze-thaw cycles, the concentration of free ssDNA increases from approximately 1% to 4% of the total coated ssDNA in the AAV8.GFP sample. Recently, Bee et al. reported an increase in free DNA in their AAV8-X samples after freeze-thaw cycles using a similar dye-based method (Bee et al., Journal of Pharmaceutical Sciences, 2021). An increase of 0.38% ± 0.08% of free DNA was observed for each freeze-thaw cycle in DPBS buffer with poloxamer 188. This is consistent with the data described herein, as they observed nearly 1% free DNA in the initial sample and nearly 3% after five freeze-thaw cycles. Overall, the product-related impurities, represented by peak 1 in the SE-UPLC curve, are ssDNA leaking from the AAV capsid. Example 3. Leakage of ssDNA in other AAV serotypes

[0086] To determine whether ssDNA leakage was specific to AAV8.GFP, the aforementioned analysis was performed on a group of different AAV serotypes: AAV2, AAV3b, AAV5, AAV7, and AAVDJ. All samples were adjusted to the same valence as the AAV8 sample in the same basal formulation. These samples were then subjected to ten freeze-thaw cycles and analyzed by Quant-iT and SE-UPLC. As shown in Figure 10, all tested AAV serotypes initially showed very low levels of free DNA. After ten freeze-thaw cycles, a significant increase in UV signal (SE-UPLC) and fluorescence signal (Quant-iT analysis) of peak 1 impurity was observed in all samples except AAV2. Although the absolute amount of total ssDNA leaked due to freeze-thaw varied among these samples, ssDNA leakage was detected for all AAV serotypes except AAV2. This result indicates that the mechanism of ssDNA release from the AAV8 capsid due to freeze-thaw stress is applicable to other serotypes.

[0087] The main differences in capsid protein sequences among different AAV serotypes lie in the hypervariable surface region, thereby enabling them to bind to different cell types with varying efficiencies (Snyder and Moullier, Adeno-associated virus: methods and protocols, Springer 2011). The replication and structural genes that produce the capsid protein are generally highly conserved across serotypes, which may make them similarly susceptible to freeze-thaw stress. Therefore, the effect of excipients on the stability of AAV during freeze-thaw stress can be applied to each of these AAV serotypes. Example 4. Excipient screening for reducing DNA leakage.

[0088] Since the integrity and successful delivery of the encapsulated ssDNA are crucial for AAV-based gene therapy, minimizing ssDNA leakage from the AAV capsid under various conditions, including freeze-thaw stress, is critical. Adding protein stabilizing excipients to AAV formulations can enhance AAV stability against freeze-thaw stress and also help understand the ssDNA leakage mechanism. Therefore, formulations containing protein stabilizing excipients were designed to test whether capsid protein stabilization could reduce ssDNA leakage during freeze-thaw cycles. DPBS buffer was chosen as the base formulation because it is commonly used for AAVs due to its physiological adaptability. Several commonly used classes of excipients were selected to evaluate their ability to protect proteins against freeze-thaw stress (e.g., AAV capsid integrity) and thereby potentially reduce ssDNA leakage. These excipients included polyols (sucrose, trehalose, mannitol, glycerol, propylene glycol, and polyethylene glycol), amino acids (proline), and surfactants (Pluronic™ F68 (poloxam 188)).

[0089] Results from fluorescence-based cell-free DNA analysis and SE-UPLC analysis showed that after ten freeze-thaw cycles, the designed formulation exhibited reduced ssDNA leakage compared to the base formulation (base), as shown in Figure 9. For example, as detected by SE-UPLC, the addition of 5% sucrose or 5% trehalose reduced ssDNA leakage from a 4-fold increase to less than a 2-fold increase, as shown in Figure 9A and Table 7. As detected by fluorescence-based cell-free DNA analysis, the addition of 5% sucrose or 5% trehalose reduced ssDNA leakage from a 3.6-fold increase to less than a 50% increase, as shown in Figure 9B and Table 8. However, 5% mannitol was less effective than sucrose and trehalose in reducing ssDNA leakage. Similar to sucrose and trehalose, 5% glycerol, 5% propylene glycol, or 5% polyethylene glycol 400 (PEG 400) reduced ssDNA leakage from a 4-fold increase to less than 50% as detected by SE-UPLC, and reduced ssDNA leakage from a 3.6-fold increase to approximately 50% as detected by fluorescent cell-free DNA analysis. 0.1% Pluronic F68, a surfactant commonly used in AAV product formulations, demonstrated comparable ability to protect AAV from ssDNA leakage to sucrose and trehalose. Based on fluorescent cell-free ssDNA analysis and SE-UPLC analysis, proline, as an amino acid cryoprotectant, also showed a significant protective effect against ssDNA leakage. Furthermore, the hydrodynamic radius of viral particles in the samples was analyzed using DLS, as shown in Figure 11 and Table 9. Table 7. Characterization of DNA impurities in AAV8-GFP samples using SEC. SEC Results t0 (%) 10× F / T (%) F1 Basics 6.05 24.48 F2 5% sucrose 6.61 10.16 F3 5% Trehalose 5.76 10.27 F4 5% Mannitol 6.44 19.81 F5 2.5% Proline 5.36 13.16 F6 5% Glycerin 6.66 6.91 F7 0.1% F68 2.72 9.21 F8 5% PG 4.19 5.94 Table 8. Quantification of ssDNA impurities in AAV8-GFP samples using fluorescent dye labeling ssDNA (fluorescence) t0 (ng / mL) 10× F / T (ng / mL) F1 Basics 363.9 1302.9 F2 5% sucrose 363.2 520.8 F3 5% Trehalose 395.9 565.8 F4 5% Mannitol 399.8 917.0 F5 2.5% Proline 392.1 490.6 F6 5% Glycerin 407.2 356.2 F7 0.1% F68 383.3 451.8 F8 5% PG 408.9 401.1 Transgender std n / a 10246.6 Table 9. Hydrodynamic radius of viral particles in AAV8-GFP samples analyzed using DLS DLS t0 10× F / T F1 Basics 13.1 12.4 F2 5% sucrose 9.1 7.7 F3 5% Trehalose 7.4 8.5 F4 5% Mannitol 11.3 10 F5 2.5% Proline 12.5 11.7 F6 5% Glycerin 13.4 12.8 F7 0.1% F68 11.2 11.1 F8 5% PG 14.7 14

[0090] Several excipients have been found to effectively minimize ssDNA leakage from the AAV capsid after freeze-thaw cycles. Excipients that effectively reduce ssDNA leakage during testing include sucrose, trehalose, mannitol, proline, glycerol, Pluronic F68, and propylene glycol. Based on these results, sucrose, propylene glycol, and Pluronic F68 were selected for further analysis. Formulations containing different concentrations of sucrose (2.5%, 5%, and 10%), propylene glycol (2.5%, 5%, and 10%), and Pluronic F68 (0.001%, 0.01%, and 0.1%) were analyzed after 1, 2, or 4 freeze-thaw cycles, the number of which reflects the representative number of freeze-thaw cycles in the development, manufacture, and administration of one or more AAV products. As shown in Figure 12, an increase in ssDNA leakage was observed with increasing freeze-thaw cycles applied to the AAV in F1. After four freeze-thaw cycles, the three tested concentrations (2.5%, 5%, and 10%) of sucrose and propylene glycol showed significant and comparable protective effects against ssDNA leakage. A slight reduction in ssDNA leakage was observed with higher concentrations of sucrose and propylene glycol. Pluronic F68 was less effective than sucrose and propylene glycol in preventing ssDNA leakage. Nevertheless, increasing the concentration of Pluronic F68 gradually reduced ssDNA leakage.

[0091] Cryoprotectants and surfactants are often used together in bioformulations to provide protection against various types of stress. To test the effect of combined excipients against freeze-thaw stress and to test whether protein stability excipients can reduce ssDNA leakage in other serotypes, formulations containing both sucrose and Pluronic F68 were designed and evaluated using various AAV serotypes.

[0092] In Figure 13, the basic formulation (BF) is DPBS, while the optimized formulation (OF) contains 10% sucrose and 0.1% Pluronic F68 in DPBS. After 10 freeze-thaw cycles, compared with the basic formulation which does not contain sucrose and Pluronic F68, the optimized formulation significantly inhibited ssDNA leakage of AAV3b, AAV5, AAV7, AAV8, and AAVDJ. By fluorescence-based cell-free DNA analysis, the reduction in ssDNA leakage was between 10 and 20-fold; by SE-UPLC assay, the reduction in ssDNA impurity peak area was between 3 and 10-fold, with a 40-fold reduction observed for AAVDJ. Notably, the protective effect on AAV2 was minimal, possibly because, compared to other serotypes, its cell-free DNA leakage is negligible in the absence of cryoprotectants and surfactants. Based on these findings, optimized formulations containing cryoprotectants and surfactants can protect various AAV serotypes from the effects of freeze-thaw stress.

[0093] Excipients are selected based on their properties and established protein stabilization mechanisms. As subtypes of polyols, sucrose and trehalose are the most commonly used sugar-based excipients for stabilizing proteins (Singh, Challenges in Protein Product Development, Springer 2018, pp. 63-95). Glycerol, propylene glycol, and polyethylene glycol (PEG 400 in these experiments) are also known polyol cryoprotectants for protein molecules. Cryoprotectants function through various mechanisms, including preferential removal from the protein surface, formation of a glassy matrix around the protein molecule, or formation of hydrogen bonds with the protein (Timasheff, Annual Review of Biophysics and Biomolecular Structure 22(1) (1993) 67-97; Corradini et al., Scientific Reports 3(1) (2013) 1-10; Markarian et al., Cryobiology 49(1) (2004) 1-9). Pluronic F68 is added as a surfactant to AAV formulations to resist mechanical stress during manufacturing and transportation, as well as interfacial stress during freeze-thaw cycles. For these purposes, it is often used in protein pharmaceuticals (Nail and Akers, Development and manufacture of protein pharmaceuticals, Springer Science & Business Media 2012). As a nonionic surfactant, Pluronic F68 reduces surface tension at the interface caused by mechanical disturbance and freeze-thaw cycles, thereby inhibiting protein adsorption at the interface due to competition with surfactants (Kasimbeg et al., Journal of Pharmaceutical Sciences 108(1) (2019) 155-161; Khan et al., European Journal of Pharmaceutics and Biopharmaceutics 97 (2015) 60-67; Dixit et al., Pharmaceutical Research 30(7) (2013) 1848-1859). Given the role of these excipients in stabilizing and maintaining protein structure, these findings also contribute to elucidating the mechanism of ssDNA leakage from the AAV capsid.

[0094] In these stability studies, an increase in the content of free ssDNA in the AAV solution was observed after freeze-thaw cycles. The free ssDNA in the bulk drug substance prior to freeze-thaw treatment could be residual DNA from cell culture or purification. The ssDNA that increased after multiple freeze-thaw cycles was genosome-derived ssDNA, which was shown to have leaked from the AAV capsid. This finding is supported by NGS characterization results and is consistent with observations of ssDNA leakage after freeze-thaw stress was inhibited by protein stabilizing excipients. Whether ssDNA leakage occurs via genosome ejection (i.e., with the capsid remaining intact) or via AAV capsid decomposition will be discussed further below.

[0095] AAV8 and AAV9 have been studied at the single viral capsid level to explore their degradation mechanisms under thermal stress, such as at 50 to 80°C (Bernaud et al., Journal of Biological Physics 44(2) (2018) 181-194). As observed by atomic force microscopy (AFM), the encapsulated DNA gene system ejects from the intact or partially unfolded capsid. Assuming that ejection is the main mechanism of ssDNA leakage, the valence of the intact capsid should decrease and be accompanied by an increase in the valence of the empty capsid, and the sum of the intact and empty capsids should remain constant before and after stress. Alternatively, if ssDNA leakage occurs due to AAV decomposition (i.e., capsid rupture), the total capsid valence should decrease after stress due to the decrease in the valence of the intact capsid (the valence of the empty capsid should remain unaffected). Since free ssDNA accounts for approximately 1% of the total ssDNA in untreated AAV samples and increases to approximately 5% after 10 freeze-thaw cycles, the change in intact capsid valence / empty capsid valence should be at a similar level.

[0096] Unfortunately, since the coefficient of variation (CV%) in enzyme-linked immunosorbent assays (ELISA) concerning capsid valence is typically 10-20%, it is difficult to determine whether a 1% to 5% change in capsid valence is accompanied by a 1% to 5% change in free ssDNA. Similarly, SE-UPLC analyses performed at 260 nm and 280 nm are not convincing in determining the degradation mechanism due to methodological variations and the limits of approximations versus estimates (e.g., extinction coefficients and molecular weights of ssDNA and capsid proteins, and the path length of the flow channel in the UV detector). Dynamic light scattering (DLS) was used as an alternative to SE-UPLC to monitor size variants, but no obvious AAV aggregation or fragmentation was observed. Careful examination of the SE-UPLC dissolution curves also did not reveal any obvious VP protein peaks following the AAV capsid. Overall, the variability of DLS and SE-UPLC (and possibly ELISA) methods hinders a conclusive elucidation of the leakage mechanism of ssDNA in AAV. Further research using more accurate and precise valence methods may be needed to better understand the ssDNA leakage mechanism. As previously mentioned, observational tools such as AFM can also provide observations of capsid integrity after freeze-thaw cycles and contribute to understanding the mechanism.

[0097] AAV-based gene therapy drugs typically undergo one or more freeze-thaw cycles during manufacturing, transportation, storage, and administration. Characterization of AAV8 in DPBS buffer has revealed that even after one freeze-thaw cycle, a portion of the coated ssDNA leaks from AAV8. The leaked ssDNA was confirmed to be the original genomic DNA initially coated in the capsid. The degree of gene leakage increases with the number of freeze-thaw cycles. Varying degrees of gene leakage have also been observed in other test serotypes, including AAV2, AAV3b, AAV5, AAV7, and AAVDJ.

[0098] Formulation screening studies identified excipients, such as polyols, amino acids, and surfactants, that effectively inhibit ssDNA leakage in AAV. Further studies using sucrose, propylene glycol, and Pluronic F68 provided a better understanding of their roles in the stability of AAV in terms of ssDNA leakage. These results provide insight into ssDNA leakage in AAV when degradation pathways are utilized from freezing and thawing stresses. Furthermore, these findings are highly valuable for the future development of stable AAV active pharmaceutical ingredients and pharmaceutical formulations. [Simplified Explanation of the Diagram]

[0014] Figure 1 shows the structure of the AAV1 carrier according to Venkatakrishnan et al., including the crystal structure of the AAV1 capsid VP3 monomer and the surface representation of the AAV1 capsid.

[0015] Figure 2 illustrates the effect of freezing-thawing an AAV carrier in the presence or absence of a protective formulation, according to an exemplary embodiment.

[0016] Figure 3 shows the absorbance of an SEC fraction of an AAV8-GFP sample at 280 nm according to an exemplary embodiment. According to an exemplary embodiment, different amounts of AAV8-GFP sample were loaded into an SEC column, such as 1 µL, 5 µL, 10 µL, 15 µL or 20 µL of sample.

[0017] Figure 4 shows the fluorescence intensity of the SEC fraction of an AAV8-GFP sample monitored at 280 nm excitation and 350 nm emission according to an exemplary embodiment.

[0018] Figure 5 shows the absorbance of an SEC fraction of an AAV8-GFP sample at 280 nm according to an exemplary embodiment. According to an exemplary embodiment, AAV8-GFP samples of different concentrations (valences) were loaded into an SEC column.

[0019] Figure 6 shows the quantification of ssDNA concentration in AAV8-GFP samples after different numbers of freeze-thaw cycles and agitation treatments, according to an exemplary embodiment.

[0020] Figure 7 shows SE-UPLC chromatograms of AAV8.GFP at t=0 (thick line) and after ten freeze-thaw (F / T) cycles (dashed line) according to an exemplary embodiment.

[0021] Figure 8 shows SE-UPLC chromatograms of AAV8.GFP with and without benzonase treatment according to an exemplary embodiment.

[0022] Figure 9A shows the percentage of impurities in the DNA peaks of AAV8.GFP in various formulations after ten freeze-thaw (F / T) cycles, according to an exemplary embodiment. Figure 9B shows the amount of ssDNA leaked from AAV8.GFP in various formulations after ten freeze-thaw (F / T) cycles, according to an exemplary embodiment.

[0023] Figure 10A shows the amount of free ssDNA leaked from AAV2, AAV3b, AAV5, AAV7, and AAVDJ after ten freeze-thaw (F / T) cycles, as measured by the Quant-iT ssDNA kit according to an exemplary embodiment. Figure 10B shows the percentage of impurities in the DNA peaks of AAV2, AAV3b, AAV5, AAV7, and AAVDJ in SE-UPLC chromatography after ten freeze-thaw (F / T) cycles, according to an exemplary embodiment.

[0024] Figure 11 shows, according to one embodiment, the hydrodynamic radius of viral particles in an AAV8-GFP sample was analyzed using DLS, with different formulations used to screen various excipients to reduce DNA leakage from the AAV capsid. According to an exemplary embodiment, DNA impurities in an AAV8-GFP sample were characterized before and after treatment with ten freeze-thaw (F / T) cycles.

[0025] Figure 12A shows the amount of free ssDNA leaked from formulations containing different concentrations of sucrose, propylene glycol, or Pluronic F68 after 1, 2, or 4 freeze-thaw (F / T) cycles, according to an exemplary embodiment. Figure 12B shows the percentage of impurities in the DNA peaks of AAV8.GFP in formulations containing different concentrations of sucrose, propylene glycol, or Pluronic F68 in SE-UPLC chromatography, according to an exemplary embodiment, after 1, 2, or 4 freeze-thaw (F / T) cycles.

[0026] Figure 13A shows the amount of free ssDNA leaked from the base formulation (BF) or optimized formulation (OF) after ten freeze-thaw (F / T) cycles, according to an exemplary embodiment. Figure 13B shows the percentage of impurities in the DNA peaks of AAV2, AAV3b, AAV5, AAV7, AAV8, and AAVDJ in the base formulation (BF) or optimized formulation (OF) after ten freeze-thaw (F / T) cycles, according to an exemplary embodiment.

Claims

1. A method for identifying nucleic acid impurities in a sample containing adeno-associated virus (AAV) vector, the method comprising: The sample containing the adeno-associated virus (AAV) vector is brought into contact with a particle size sieving chromatography (SEC) column; The SEC column was washed with a solution to provide at least one precipitate; and nucleic acid impurities in the at least one precipitate were identified using a spectrophotometer.

2. The method of claim 1, wherein the nucleic acid impurities are single-stranded DNA genomes of AAV.

3. The method of claim 1, further comprising treating the nucleic acid impurities with a fluorescent dye.

4. The method of claim 1, further comprising using the spectrophotometer to measure the absorbance of the at least one precipitate at 280 nm.

5. The method of claim 1, further comprising using the spectrophotometer to measure the absorbance ratio of the at least one precipitate at 260 / 280 nm.

6. The method of claim 1, further comprising quantifying such nucleic acid impurities based on a nucleic acid standard curve.

7. The method of claim 1, further comprising contacting the sample with a DNA nuclease before contacting the sample with the SEC column.

8. The method of claim 1, wherein the SEC system is a particle size sieving ultra-high performance liquid chromatography (SE-UPLC) system.

9. The method of claim 1, further comprising performing next-generation sequencing (NGS) on the at least one precipitate.

10. A composition comprising at least one adeno-associated virus (AAV) vector and at least one excipient, wherein the at least one excipient is a sugar, an amino acid, a surfactant or a polyol, wherein the at least one AAV vector comprises packaged nucleic acids; and wherein the at least one AAV vector in the composition is protected against leakage of the packaged nucleic acids.

11. The composition of claim 10, wherein the at least one AAV carrier is exposed to at least one freeze-thaw cycle.

12. The composition of claim 10, wherein the packaged nucleic acids are single-stranded DNA derived from the AAV genotype, wherein the packaged nucleic acids are present within the AAV capsid.

13. The composition of claim 10, wherein the excipient is present at a concentration of about 0.001% to about 10%, wherein the composition further comprises phosphate-buffered saline and a nonionic surfactant.

14. The composition of claim 10, wherein the sugar is sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose or maltoheptaose.

15. A composition of claim 10, wherein the amino acid is proline.

16. The composition of claim 10, wherein the surfactant is poloxamer 188.

17. The composition of claim 10, wherein the surfactant is a nonionic surfactant and wherein the surfactant is present at a concentration of about 0.001% to about 0.2%.

18. The composition of claim 10, wherein the composition comprises sucrose and poloxamer 188.

19. The composition of claim 18, wherein the sucrose is present at a concentration of 2.5% to 10% and the poloxamer 188 is present at a concentration of 0.001% to 0.2%.

20. A method for protecting an adeno-associated virus (AAV) vector from leakage of packaged nucleic acids, comprising: Obtain a sample containing at least one AAV carrier; The stabilizing component is added to the sample to form a protective formulation, wherein the protective formulation protects the at least one AAV vector from leakage of packaged nucleic acids, and the stabilizing component includes at least one excipient, which is a sugar, amino acid, surfactant or polyol.

21. The method of claim 20, wherein the AAV carrier is exposed to at least one freeze-thaw cycle.

22. The method of claim 20, wherein the packaged nucleic acids are single-stranded DNA genomes of AAV, wherein the packaged nucleic acids are present within the AAV capsid.

23. The method of claim 20, wherein the excipient is present in the protective formulation at a concentration of about 0.001% to about 10%, wherein the stabilizing component further comprises phosphate-buffered saline and a nonionic surfactant.

24. The method of claim 20, wherein the sugar is sucrose, trehalose, mannitol, raffinose, lactose, glucose, maltose, maltotriose, maltotetraose, maltopentose, or maltoheptaose.

25. The method of claim 20, wherein the amino acid is proline.

26. The method of claim 20, wherein the surfactant is poloxamer 188.

27. The method of claim 20, wherein the surfactant is a nonionic surfactant, and wherein the surfactant is present in the protective formulation at a concentration of about 0.001% to about 0.2%.

28. The method of claim 20, wherein the stabilizing component comprises sucrose and poloxamer 188.

29. The method of claim 28, wherein the sucrose is present in the protective formulation at a concentration of 2.5% to 10%, and the poloxamer 188 is present in the protective formulation at a concentration of 0.001% to 0.2%.

Images