Enhanced isolation method for fully loaded adeno-associated virus (AAV) capsids

Abstract

A method for enriching fully loaded adeno-associated virus (AAV) capsids from a mixture containing fully loaded adeno-associated virus (AAV) capsids, partially loaded and / or empty AAV capsids by chromatography, comprising the steps of: contacting the mixture with a strong or weak anion exchange material; eluting the loaded mixture with a neutral to alkaline buffer containing an organic modifier having a relative solvent polarity of 0.4 to 0.8; and recovering a fraction enriched with fully loaded AAV capsids.

Description

[Technical Field]

[0001] The present invention relates to a method for enriching fully loaded adeno-associated virus (AAV) capsids from a mixture containing fully loaded, partially loaded, and / or empty AAV capsids by chromatography. The method of the present invention is particularly useful for obtaining an enriched fully loaded AAV capsid fraction containing depleting contaminants such as empty, partially loaded, heavy, damaged AAV capsids, capsid aggregates, and others. [Background technology]

[0002] Gene therapy is a promising field of medicine that focuses on treating diseases by genetically modifying cells or repairing or reconstructing defective genetic material to achieve therapeutic effects. Genetic material is administered to patients suffering from disorders caused by a particular gene. One method of administering therapeutic genetic material is by using vectors for viral capsids or virus-like particles, particularly adeno-associated virus (AAV) capsids.

[0003] AAV is a widely used vector in gene therapy, primarily due to its safety profile and effective transduction into various target tissues. The production of AAV viral vectors is a complex process requiring innovative approaches to meet stringent safety and efficacy requirements, as well as demanding clinical and market needs.

[0004] Despite its widespread clinical applications, improved purification techniques are needed to remove product and process-related impurities and ensure high AAV product quality and potency throughout the entire manufacturing process.

[0005] Both empty capsids and capsids partially loaded with the target vector genome are product-related impurities and are technically difficult to separate from the target vector during downstream purification. Removal of empty capsids from AAV preparations and maximization of the ratio between fully loaded and empty capsids are among the purification goals, which address safety and regulatory recommendations for AAV-based gene therapies [1–3].

[0006] Depending on the precise loading sequence, partially loaded capsids may still contribute to target cell transduction [4]. However, AAV capsids containing host cell and / or helper DNA and manufacturing-related impurities may pose an immunological risk to patients [5].

[0007] This has led to the development and evaluation of many different materials and methods for purifying AAV. For example, metal affinity chromatography has been used to purify AAV capsids [6]. However, this does not distinguish between empty and fully loaded capsids and is serotype-dependent. A crude sample containing the desired AAV capsid and contaminants is brought into contact with a metal affinity material. Essentially, the AAV capsid binds to the affinity material, while the contaminants do not. The unbound contaminants are washed out, and the AAV capsid is recovered by chemically inhibiting the interaction between the affinity material and AAV.

[0008] In adeno-associated virus vector production, downstream processes, including purification, remain a major bottleneck. BIA Isolation, Sartorius offers a platform for the purification of adenovirus-associated vaccines using market-leading monolithic chromatography columns and an analytical toolbox for process monitoring of adenovirus-associated vaccines.

[0009] The simplified purification of AAV capsids consists of a standard downstream process including combined dissolution, clarification, tangential flow filtration (TFF), and enrichment of fully loaded AAV capsids by chromatographic capture on a pre-packed monolithic sulfonate (SO3) column and a pre-packed monolithic quaternary amine (QA) column.

[0010] This method provides pharmaceutical-grade adeno-associated virus, but it is desirable to further improve the purity of AAV for gene therapy, in particular to separate impurities or other contaminants carrying empty or partially loaded AAV capsids and / or DNA, etc., from the fully loaded AAV capsids required for treatment, which are loaded together with the genetic material. [Overview of the Initiative] [Problems that the invention aims to solve]

[0011] The object of the present invention is to provide a method for separating fully loaded adeno-associated virus (AAV) capsids from empty and partially loaded AAV capsids.

[0012] A further object of the present invention is to provide a method for obtaining an enriched fraction of fully loaded AAV capsid from a mixture containing fully loaded AAV capsid and empty, partially loaded, overloaded, damaged AAV capsid, capsid aggregates, and others.

[0013] Another objective is to provide a method for obtaining fully loaded AAV capsids with contaminants such as DNA, more specifically hcDNA and pDNA, removed as much as possible.

[0014] Another objective is to provide a method that can be used analytically in a similar manner on a preparative scale. [Means for solving the problem]

[0015] The subject of the present invention is a method for enriching a mixture containing full adeno-associated virus (AAV) capsids, partially filled AAV capsids, and / or empty AAV capsids by chromatography, This method is - The step of bringing the mixture into contact with a strong anion exchange material or a weak anion exchange material, - The loaded mixture is eluted with a neutral to alkaline buffer containing an organic modifier with a relative solvent polarity of 0.4 to 0.8. - A step in which a fully loaded AAV capsid recovers the enriched fraction, This method includes [something].

[0016] The relative measure of solvent polarity, RPM, is defined by the following formula: As described by Dukiie et al.[8], RPM=E T (n-hexane)-E T / E T (n-hexane) E T -Transition energy [kJ / mol]

[0017] Although the RPM values ​​in Table 1 are negative, it is still sufficient to use absolute RPM values ​​for correlation analysis [7].

[0018] The method of the present invention can be advantageously used for both analytical purposes and preparative manufacturing purposes for fully loaded AAV capsids.

[0019] According to one embodiment of the present invention, an organic modifier can be selected from the group consisting of methanol, 1,3-propanediol, 1,2-propanediol, N-methylformamide, diethylene glycol, triethylene glycol, 1,3-butanediol, 2-propyn-1-ol (propargyl alcohol), 2-methoxyethanol, 2-propen-1-ol (allyl alcohol), N-methylacetamide, ethanol, 2-aminoethanol, acetic acid, benzyl alcohol, 1-propanol, 1-butanol, 2-hydroxymethylfuran (furfuryl alcohol), 2-phenylethanol, 1-pentanol, 2-methyl-1-propanol (isobutyl alcohol), 1-hexanol, 2-propanol, 3-phenyl-1-propanol, 1-heptanol, 1-octanol, cyclopentanol, 1-decanol, 2,6-dimethylphenol (2,6-xylenol), 2-butanol, 3-methyl-1-butanol (isoamyl alcohol), cyclohexanol, 1-dodecanol, 1-phenylethanol, acrylonitrile, 4-methyl-1,3-dioxolan-2-one (propylene carbonate), 2-pentanol, nitromethane, acetonitrile, dimethyl sulfoxide, methyl acrylate, aniline, tetra-N-hexylammonium benzoate, tetrahydrothiophene 1,1-dioxide (sulfolane), 2-methyl-2-propanol (tert-butyl alcohol), acetic anhydride, N,N-dimethylformamide, N,N-dimethylacetamide, propionitrile, and nitroethane.

[0020] Preferred organic modifiers are acetonitrile, propylene carbonate, 2-propanol, 1-butanol, 1-propanol, t-butanol, ethanol, methanol, and mixtures thereof. Embodiments of the present invention are methods for enriching full-load adeno-associated virus (AAV) capsids from a mixture containing full-load adeno-associated virus (AAV) capsids, partially loaded AAV capsids, and / or empty AAV capsids by chromatography, wherein the method comprises - contacting the mixture with a strong anion exchange material or a weak anion exchange material; - eluting the loaded mixture with a neutral to alkaline buffer containing an organic modifier selected from the group consisting of methanol, 1,3-propanediol, 1,2-propanediol, N-methylformamide, diethylene glycol, triethylene glycol, 1,3-butanediol, 2-propyn-1-ol (propargyl alcohol), 2-methoxyethanol, 2-propen-1-ol (allyl alcohol), N-methylacetamide, ethanol, 2-aminoethanol, acetic acid, benzyl alcohol, 1-propanol, 1-butanol, 2-hydroxymethylfuran (furfuryl alcohol), 2-phenylethanol, 1-pentanol, 2-methyl-1-propanol (isobutyl alcohol), 1-hexanol, 2-propanol, 3-phenyl-1-propanol, 1-heptanol, 1-octanol, cyclopentanol, 1-decanol, 2,6-dimethylphenol (2,6-xylenol), 2-butanol, 3-methyl-1-butanol (isoamyl alcohol), cyclohexanol, 1-dodecanol, 1-phenylethanol, acrylonitrile, 4-methyl-1,3-dioxolan-2-one (propylene carbonate), 2-pentanol, nitromethane, acetonitrile, dimethyl sulfoxide, methyl acrylate, aniline, tetra-N-hexylammonium benzoate, tetrahydrothiophene 1,1-dioxide (sulfolane), 2-methyl-2-propanol (tert-butyl alcohol), acetic anhydride, N,N-dimethylformamide, N,N-dimethylacetamide, propionitrile, and nitroethane; - recovering the fraction enriched in full-length AAV capsids; and a method comprising the steps of.

[0021] In another embodiment of the invention, the buffer may contain an alkaline earth metal salt, particularly a salt of magnesium or calcium and / or mixtures thereof.

[0022] In yet another embodiment of the present invention, the alkaline earth metal salt may be magnesium acetate or magnesium formate, or calcium acetate or calcium formate, or mixtures thereof and / or more cosmotropic alternatives thereof. Preferred more cosmotropic alternatives are inorganic acids, organic acids or organic hydroxy acids, amino acids, or polycarboxylic acids having 10 or fewer carbon atoms, such as magnesium and calcium salts of oxalic acid or citric acid.

[0023] In yet another embodiment of the present invention, the buffer solution may have a pH value of about 7.0 to about 10.5, and particularly about 7.5 to about 9.50.

[0024] In further embodiments of the present invention, the buffer may contain an isotonic substance selected from the group consisting of sucrose, sorbitol, mannitol, and xylitol.

[0025] According to another embodiment of the present invention, the strong or weak anion exchange material may be a strong or weak anion exchange material, a monolithic anion exchange or multimodal material, a particulate anion exchange or multimodal material, and / or an anion exchange or multimodal material arranged in a film, and / or a particle-packed anion exchange or multimodal column, and / or a fiber chromatography anion exchange or multimodal column, as a multimodal material having hydrogen bonding properties and together with or without a charged metal affinity ligand.

[0026] Multimodal chromatography, also known as mixed-mode chromatography (MMC), is a chromatographic method that utilizes two or more forms of interaction between the stationary phase and the electrolyte to perform these separations [16, 17, 18]. MMC can be classified into physical MMC and chemical MMC. In the former method, the stationary phase consists of two or more types of packing material. In the chemical method, one type of packing material containing two or more functional groups is used. One technique is to connect two commercially available columns in series, called a "tandem column." Another technique is a "two-phase column," which involves packing two stationary phases separately at both ends of the same column. A further technique is to homogenize two or more different types of stationary phases in a single column, called a "hybrid column" or "mixed-bed column."

[0027] According to yet another embodiment of the present invention, AAV may be selected from different serotypes, such as serotypes and hybrid serotypes selected from the group consisting of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.10, AAV11, and AAV12. The AAV serotypes analyzed by the method of the present invention may be recombinant hybrid serotypes such as AAV2 / 8, other hybrid serotypes, chimeras, surface-modified AAVs, and synthetically derived AAV-like particles.

[0028] A chimeric virus is a virus that contains genetic material derived from two or more different viruses.

[0029] Surface-modified viruses are known and are described, for example, in

[20] .

[0030] Synthetic AAV-like particles are known and are described, for example, in

[21] .

[0031] The subject of the present invention is an aqueous solution having a neutral to alkaline pH, further comprising a buffering agent and an organic modifier having a relative polarity scale (RPM) of 0.4 to 0.8 of solvent equivalents.

[0032] According to another embodiment of the aqueous solution of the present invention, the organic modifier is methanol, 1,3-propanediol, 1,2-propanediol, N-methylformamide, diethylene glycol, triethylene glycol, 1,3-butanediol, 2-propyne-1-ol (propargyl alcohol), 2-methoxyethanol, 2-propen-1-ol (allyl alcohol), N-methylacetamide, ethanol, 2-aminoethanol, acetic acid, benzyl alcohol, 1-propanol, 1-butanol, 2-hydroxymethylfuran (furfuryl alcohol), 2-phenylethanol, 1-pentanol, 2-methyl-1-propanol (isobutyl alcohol), 1-hexanol, 2-propanol, 3-phenyl-1-propanol, 1-heptanol, 1-oc The following can be selected from the group consisting of tanol, cyclopentanol, 1-decanol, 2,6-dimethylphenol (2,6-xylenol), 2-butanol, 3-methyl-1-butanol (isoamyl alcohol), cyclohexanol, 1-dodecanol, 1-phenylethanol, acrylonitrile, 4-methyl-1,3-dioxolan-2-one (propylene carbonate), 2-pentanol, nitromethane, acetonitrile, dimethyl sulfoxide, methyl acrylate, aniline, tetra-N-hexylammonium benzoate, tetrahydrothiophene 1,1-dioxide (sulfolane), 2-methyl-2-propanol (tert-butyl alcohol), anhydride acetate, N,N-dimethylformamide, N,N-dimethylacetamide, propionitrile, and nitroethane.

[0033] Preferred organic modifiers are acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol, and propanol.

[0034] According to yet another embodiment of the aqueous solution of the present invention, the solution may contain an alkaline earth metal salt.

[0035] According to yet another embodiment of the aqueous solution of the present invention, the alkaline earth metal salt may be a salt of magnesium or calcium and mixtures thereof, in particular magnesium acetate or magnesium formate, and / or calcium acetate or calcium formate, and / or more cosmotropic alternatives thereof. Preferred more cosmotropic alternatives are inorganic acids, organic acids or organic hydroxy acids or amino acids, or polycarboxylic acids having 10 or fewer carbon atoms, such as magnesium and calcium salts of oxalic acid or citric acid.

[0036] According to a further embodiment of the aqueous solution of the present invention, a buffering substance for buffering an aqueous solution in the pH range of pH 6 to pH 12 is used, in particular, the buffering substance is 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol (Bis-Tris), 2,2′,2′′-nitrilotriacetate (ADA), 2-[(2-amino-2-oxoethyl)amino]ethane-1-sulfonic acid (ACES), 2,2′-(piperazine-1,4-diyl)di(ethane-1-sulfonic acid) (PIPES), 2-hydroxy-3-(morpholine-4-yl)propane Pan-1-sulfonic acid (MOPSO), 2,2′-[propane-1,3-diylbis(azandiyl)]bis[2-(hydroxymethyl)propane-1,3-diol](BTP), N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES), 3-(morpholine-4-yl)propane-1-sulfonic acid (MOPS), 2-{[1,3-dihydroxy-2-(hydroxymethyl)propane-2-yl]amino}ethane-1-sulfonic acid (TES), 2-[4-(2-hydroxyethyl)piperazine-1-yl]ethane -1-Sulfonic acid (HEPES), 3-[N,N-bis(2-hydroxyethylamino)-2-hydroxy-1-propanesulfonic acid (DIPSO), 4-(4-morpholinyl)butanesulfonic acid (MOBS), 2-hydroxy-3-[tris(hydroxymethyl)methylamino]-1-propanesulfonic acid (TAPSO), 2-amino-2-(hydroxymethyl)-1,3-propanediol (Trizma), 4-(2-hydroxyethyl)piperazine-1-(2-hydroxypropane-3-sulfonic acid) (HEPPSO), piperadi N-N,N′-bis(2-hydroxypropanesulfonic acid), (POPSO), triethylamine (TEA), 4-(2-hydroxyethyl)-1-piperazine-propanesulfonic acid (EPPS), N-tris(hydroxymethyl)methylglycine (tricine), N,N-bis(2-hydroxyethyl)glycine (bicine), N-(2-hydroxyethyl)piperazine-N'-(4-butanesulfonic acid) (HEPBS), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TAPS), 2-amino-2-methyl-1,You may select from the group consisting of 3-propanediol (AMPD), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TABS), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPSO), 2-(N-cyclohexylamino)ethanesulfonic acid (CHES), sodium 3-cyclohexylamino-2-hydroxypropanesulfonate (CAPSO), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPS), 3-(cyclohexylamino)-1-propanesulfonic acid (CAPS), and 4-[cyclohexylamino]-1-butanesulfonic acid (CABS).

[0037] According to another embodiment of the aqueous solution of the present invention, the buffer solution may contain an isotonic additive, particularly one selected from the group consisting of sucrose, sorbitol, mannitol, and xylitol.

[0038] The buffer may also contain a nonionic surfactant, which is useful for suppressing undesirable effects, such as interactions between components in a mixture containing AAV capsid. In particular, so-called poloxamers such as poloxamer 188 can be used.

[0039] A further subject of the present invention is the use of an aqueous solution of the present invention for separating a fully loaded AAV capsid from an empty AAV capsid by the method of the present invention.

[0040] Table 1 shows several organic solvents and their respective RPM values. The table is taken from reference [7]. [Brief explanation of the drawing]

[0041] [Figure 1] Figure 1 illustrates the separation capabilities of fully loaded / empty AAV capsids for different organic modifiers. [Figure 2A]Figures 2A and 2B illustrate the effect of organic modifiers on the separation efficiency of fully loaded / empty AAV capsids. Figure 2C illustrates the effect of the presence of the organic modifiers magnesium chloride (MgCl2), magnesium acetate (MgAc2), and magnesium formate (MgFor2) on the separation efficiency of fully loaded / empty AAV capsids. [Figure 2B] Figures 2A and 2B illustrate the effect of organic modifiers on the separation efficiency of fully loaded / empty AAV capsids. Figure 2C illustrates the effect of the presence of the organic modifiers magnesium chloride (MgCl2), magnesium acetate (MgAc2), and magnesium formate (MgFor2) on the separation efficiency of fully loaded / empty AAV capsids. [Figure 2C] Figures 2A and 2B illustrate the effect of organic modifiers on the separation efficiency of fully loaded / empty AAV capsids. Figure 2C illustrates the effect of the presence of the organic modifiers magnesium chloride (MgCl2), magnesium acetate (MgAc2), and magnesium formate (MgFor2) on the separation efficiency of fully loaded / empty AAV capsids. [Figure 3] Figure 3 illustrates the effect of preferred organic modifiers on the separation ability of fully loaded / empty AAV capsids compared to poloxamer 188. [Figure 4] Figure 4 illustrates the effect of the percentage of organic modifiers on the separation of empty / fully loaded AAV capsids. [Figure 5] Figure 5 illustrates the effect of Mg2+ ion concentration on the separation performance of the elution buffer. [Figure 6] Figure 6 illustrates the pH range of the elution buffer used for separating empty / full AAV8 capsids. [Figure 7] Figure 7 illustrates the effect of pH on the separation of AAV8 and AAV9 serotype capsids. [Figure 8A] Figure 8A illustrates the sorting (purification) procedure for samples containing different populations of AAV capsids. Analysis is performed on the recovered fractions, and multiple detector results are shown in Figures 8B, 8C, 8D, and 8E. [Figure 8B]Figure 8A illustrates the sorting (purification) procedure for samples containing different populations of AAV capsids. Analysis is performed on the recovered fractions, and multiple detector results are shown in Figures 8B, 8C, 8D, and 8E. [Figure 8C] Figure 8A illustrates the sorting (purification) procedure for samples containing different populations of AAV capsids. Analysis is performed on the recovered fractions, and multiple detector results are shown in Figures 8B, 8C, 8D, and 8E. [Figure 8D] Figure 8A illustrates the sorting (purification) procedure for samples containing different populations of AAV capsids. Analysis is performed on the recovered fractions, and multiple detector results are shown in Figures 8B, 8C, 8D, and 8E. [Figure 8E] Figure 8A illustrates the sorting (purification) procedure for samples containing different populations of AAV capsids. Analysis is performed on the recovered fractions, and multiple detector results are shown in Figures 8B, 8C, 8D, and 8E. [Figure 9] Figure 9 illustrates the effect of preferred organic modifiers on the separation ability of fully loaded / empty AAV capsids on a weak anion exchanger monolith. [Figure 10] Figure 10 illustrates the effect of preferred organic modifiers on the separation capability of fully loaded / empty AAV capsids on a multimodal exchanger monolith or membrane. [Figure 11A] Figures 11A and 11B illustrate the effect of higher percentages of organic modifiers on AAV capsid separation, particularly on baseline separation of empty and partially loaded capsids from fully loaded capsids. [Figure 11B] Figures 11A and 11B illustrate the effect of higher percentages of organic modifiers on AAV capsid separation, particularly on baseline separation of empty and partially loaded capsids from fully loaded capsids. [Figure 12A] Figure 12 illustrates the eluted fractions from the preparative operation analyzed by vertical density gradient ultracentrifugation connected to a PATfix® (Sartorius BIA Separations) multi-detector setup. [Figure 12B]Figure 12 illustrates elution fractions from a preparative operation analyzed by vertical density gradient ultracentrifugation coupled with a PATfix (trademark) (Sartorius BIA Separations) multiple detector setup. [Figure 13] Figure 13 illustrates a comparison of AAV capsid separation using a QA column and an anion exchange membrane adsorbent. [Figure 14] Figure 14 illustrates the separation ability of overloaded / empty AAV capsids with different loading and elution strategies. [Figure 15] Figure 15 illustrates a chromatogram showing the tryptophan fluorescence and light scattering separation ability of harvest AAV8 and overloaded / empty AAV8 capsids from the same batch using a two-dimensional chromatography system.

Mode for Carrying Out the Invention

[0042] The term "separation ability" is known to those skilled in the art. The separation ability is calculated by dividing the peak retention time difference between different chromatographic peaks by the peak width at half the height of each peak using the following formula [9]. R = 1.18(t R2 - tr R1 ) / (W h1 + W h2 ) where t R2 > t R1 t R2 、t R1 - Peak retention time W h1 、W h2 - Peak width at half the height

[0043] The term "overloaded AAV capsid" means that the capsid encapsulates a sufficient amount of vector genome to provide therapeutic efficacy.

[0044] The term "empty AAV capsid" means that the capsid lacks a sufficient vector genome and thus cannot provide a therapeutic effect.

[0045] If temperature is mentioned or not mentioned, the temperature is room temperature (23°C).

[0046] If volume is mentioned and temperature is not specified, it is assumed to be room temperature.

[0047] According to the method of the present invention, a fully loaded adeno-associated virus (AAV) capsid is separated from an empty AAV capsid by chromatography using a strongly anion-exchange material in contact with the mixture to be separated. After contact with the mixture, the mixture is eluted from the strongly anion-exchange material under appropriate conditions known to those skilled in the art using a neutral to alkaline buffer containing an organic modifier with a relative polarity scale of 0.4 to 0.8. In particular, during the elution step, the empty AAV capsid can be separated from the fully loaded AAV capsid and recovered in a fraction separated from the other components of the mixture, especially the empty AAV capsid.

[0048] Table 1 lists the RPM values ​​of various organic compounds. Table 1 enumerates typical organic modifiers used in the methods of the present invention, such as acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol or propanol, or mixtures thereof.

[0049] Figure 1 summarizes the results obtained when using a standard organic modifier. In a series of experiments that formed the basis for the results illustrated in Figures 2 and 3, samples of a mixture containing empty and fully loaded AAV capsids, excluding other contaminants, were subjected to a monolithic QA anion exchanger in buffers containing different organic modifiers. These buffers contained TRIS as a buffer, sorbitol for capsid stabilization, magnesium acetate as an eluting salt, and different organic modifiers. In the standard buffers instead of organic modifiers, poloxamer 188 was used as a nonionic surfactant.

[0050] The separation of empty AAV capsids from fully loaded AAV capsids is significantly improved when using the organic modifier according to the present invention compared to the elution buffer without the organic modifier. Poloxamer 188, which is an organic compound but is outside the RPM range of 0.4-0.8, shows a separation efficiency of only 1.89 as illustrated in Figure 3, whereas the separation efficiency obtained using the organic modifier according to the present invention is in the range of 2.00-2.40.

[0051] The experimental results shown in Figure 3 confirm the effect of the improved separation capability of fully loaded / empty AAV capsids in the presence of at least one organic modifier according to the present invention. Using acetonitrile alone as an organic modifier in the presence of magnesium acetate in the elution buffer yields a separation capability of 2.40, while a buffer containing the organic component poloxamer 188 yields a significantly lower separation capability of only 1.89.

[0052] It is advantageous for the elution buffer to contain not only organic modifiers but also alkaline earth metal salts, particularly magnesium or calcium salts and mixtures thereof. Typically, alkaline earth metal salts are magnesium acetate and / or calcium acetate and / or magnesium formate or calcium formate. The experimental results illustrated in Figures 2A and 3 show that the presence of alkaline earth metal salts, such as magnesium acetate, in the elution buffer improves the separation efficiency of fully loaded / empty AAV capsids at conventional concentrations compared to the absence of alkaline earth metal salts. In the experiment in Figure 2A, potassium acetate was used instead of magnesium acetate (Figure 3). The separation efficiency was 1.57 in Figure 2A and 2.40 in Figure 3. In the experiment in Figure 2C, the separation efficiency increased from 1.75 to 2.06 when the inorganic anion magnesium chloride was changed to organic magnesium formate. Organic anions with two carbon atoms (or more if soluble in water) improved the separation efficiency. In both cases, acetonitrile was added as an organic modifier, and the separation efficiency of magnesium formate (Figure 2C) improved compared to magnesium chloride (Figure 2C).

[0053] In another embodiment, the effect of acetonitrile as a standalone organic modifier is illustrated in Figure 2B. The separation efficiency between fully loaded and empty AAV capsids is increased from 1.51 in the presence of 188 poloxamer to 1.75 in the presence of acetonitrile.

[0054] The results illustrated in Figures 2A, 2B, and 2C demonstrate that using acetonitrile as an organic modifier in the elution buffer improves the separation between empty and fully loaded AAV capsids, regardless of the presence of different types of alkaline earth metal salts such as potassium acetate, magnesium acetate, or magnesium chloride (different cations or anions in the elution buffer). The addition of acetonitrile as an organic modifier in the elution buffer provides a particularly advantageous effect on the separation of empty / fully loaded AAV capsids.

[0055] The concentration of the organic modifier in the elution buffer should be as high as necessary and as low as possible. The upper limit of the amount of organic modifier depends, for example, on the miscibility of the organic modifier with water and its compatibility with other components in the elution buffer. Those skilled in the art can easily estimate the range of organic modifier concentrations. Typically, organic modifiers are present in the range of 1% [volume / volume] to 5% [volume / volume], and at organic modifier concentrations higher than 5%, the separation efficiency between empty and fully loaded capsids decreases with increasing percentage of organic modifier, as shown in Figure 4. While the use of high concentrations of organic modifier may not seem very important, those skilled in the art will avoid the use of unnecessarily high concentrations of organic modifier.

[0056] In principle, those skilled in the art can easily adjust the appropriate concentration of alkaline earth metal salts in the elution buffer used in the preparation of AAV capsids. Typically, the concentration range is about 0.5 mM [weight / volume] to 10 mM [weight / volume].

[0057] Figure 5 shows the results obtained when the magnesium acetate concentration in the elution buffer is changed. While the use of high concentrations of magnesium acetate may not be significant, those skilled in the art will avoid the use of unnecessarily high concentrations. A magnesium acetate concentration of 5 mM appears preferable, but the resolution changes only slightly with different magnesium acetate concentrations. The pH of the elution buffer is typically in the range of approximately pH 7.5 to approximately pH 9.25, as shown in Figure 6. The optimal pH of the buffer depends on the serotype of each AAV capsid. The optimal pH can be evaluated by a simple experiment.

[0058] The AAV capsid is selected from, for example, different serotypes such as serotypes AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.10, AAV11, AAV12, and hybrid serotypes, as well as chimeras, surface-modified AAVs, and any synthetically derived AAV-like particles. The AAV serotype analyzed by the method of the present invention may be a recombinant hybrid serotype such as AAV2 / 8 or another hybrid serotype.

[0059] For example, while the separation of fully loaded and empty AAV8 serotype capsids is typically performed at a pH of approximately 8.5, the separation of fully loaded and empty AAV9 capsids requires a higher pH value for optimal separation, as shown in Figure 7.

[0060] Buffers used to adjust the pH value to the optimal level for separation can provide buffering capacity for aqueous solutions in the pH range of pH 7 to pH 12. Typically, these include 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol (Bis-Tris), 2,2′,2′′-nitrilotriacetate (ADA), 2-[(2-amino-2-oxoethyl)amino]ethane-1-sulfonic acid (ACES), 2,2′-(piperazine-1,4-diyl)di(ethane-1-sulfonic acid) (PIPES), 2-hydroxy-3-(morpholine-4-yl)propane-1-sulfonic acid (MOPSO), and 2,2′-[propane-1,3-diyl Rubis(azandiyl)bis[2-(hydroxymethyl)propane-1,3-diol](BTP), N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES), 3-(morpholine-4-yl)propane-1-sulfonic acid (MOPS), 2-{[1,3-dihydroxy-2-(hydroxymethyl)propane-2-yl]amino}ethane-1-sulfonic acid (TES), 2-[4-(2-hydroxyethyl)piperazine-1-yl]ethane-1-sulfonic acid (HEPES), 3-[N,N- Bis(2-hydroxyethylamino)-2-hydroxy-1-propanesulfonic acid (DIPSO), 4-(4-morpholinyl)butanesulfonic acid (MOBS), 2-hydroxy-3-[tris(hydroxymethyl)methylamino]-1-propanesulfonic acid (TAPSO), 2-amino-2-(hydroxymethyl)-1,3-propanediol (Trizma), 4-(2-hydroxyethyl)piperazine-1-(2-hydroxypropane-3-sulfonic acid) (HEPPSO), piperazine-N,N′-bis(2 -Hydroxypropanesulfonic acid), (POPSO), triethylamine (TEA), 4-(2-hydroxyethyl)-1-piperazine-propanesulfonic acid (EPPS), N-tris(hydroxymethyl)methylglycine (tricine), N,N-bis(2-hydroxyethyl)glycine (bicine), N-(2-hydroxyethyl)piperazine-N'-(4-butanesulfonic acid) (HEPBS), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TAPS), 2-amino-2-methyl-1,Selected from the group consisting of 3-propanediol (AMPD), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TABS), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPSO), 2-(N-cyclohexylamino)ethanesulfonic acid (CHES), sodium 3-cyclohexylamino-2-hydroxypropanesulfonate (CAPSO), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPS), 3-(cyclohexylamino)-1-propanesulfonic acid (CAPS), and 4-[cyclohexylamino]-1-butanesulfonic acid (CABS).

[0061] Isotonic substances may be present in the elution buffer. Typically, these are selected from the group consisting of sucrose, sorbitol, mannitol, and xylitol.

[0062] The buffer solution may contain a nonionic surfactant. In particular, so-called poloxamers such as poloxamer 188 can be used. A poloxamer is a nonionic triblock copolymer consisting of a central hydrophobic chain of polyoxypropylene (poly(propylene oxide)) sandwiched between two hydrophilic chains of polyoxyethylene (poly(ethylene oxide)) [U.S. Patent No. 3,740,421]. Poloxamers are also known by the trade name Pluronic®.

[0063] The strong anion exchange material comprises quaternary amine ligands trade names such as Q, QA, QAE, QAM, TEAE, TMAM, or TMAE, which maintain a constant charge over a range of approximately pH 2 to pH 13. The quaternary amine anion exchange material is disclosed in relation to the separation of empty capsids and fully loaded capsids in

[10] .

[0064] The weak anion exchanger DEAE (diethylaminoethyl) material has a pK of approximately 11.5. aThe tertiary amine ligand contains a tertiary amine ligand, which enables the elution of empty and fully loaded capsids at a moderate pH value, as shown in Figure 9.

[11]

[0065] The material may be a monolithic anion exchange material, a particulate anion exchange material, and / or anion exchange material arranged in a film, and / or a particle-packed anion exchange column, and / or a fiber chromatography anion exchange or multimodal fiber column.

[0066] The material may be a multimodal metal affinity exchanger material. This material impairs the properties of charged metal affinity ligands and hydrogen bonding weak anion exchangers [6, 12]. As shown in Figure 10, a subgroup of empty capsids is first separated, and then the fully loaded capsids are separated in a linear magnesium chloride gradient and subsequently in a high-salt step in which almost all empty capsids are eluted.

[0067] The effects of higher percentages of organic modifiers on AAV capsid separation, particularly the separation of empty and partially loaded capsids using strong anion exchangers, are shown in Figures 11A and 11B.

[0068] In addition, the method of the present invention was tested using several loading and elution combinations, for example, loading with different salts as the eluting salt (e.g., loading with more cosmotropic / less cosmotropic salts) and loading with different organic modifiers as the eluting organic modifiers. According to both options, the mixed salt or organic modifier gradient was applied to the elution gradient (Figure 14).

[0069] In addition, different sample pretreatments (samples pretreated with different salts or ultimately with different organic modifiers) and organic modifier reduction gradients (2.5% to 0.0% or 20.0% to 0.0%) were also tested (Figure 14).

[0070] The method of the present invention is also applicable to the analysis of different AAV serotyped samples or lysate samples in one or more two-dimensional chromatography systems. The two-dimensional chromatography system, PATfix® AAV Switcher (Sartorius BIA Separations, Aidovščina, Slovenia), enables the analysis of complex crude samples upstream in process development. The first column serves to pre-purify the sample and is a strong cation exchange (CEX) column, and the second column is an anion exchange (AEX) column, performing empty, partially loaded, fully loaded or other capsid separation as shown in Figure 15. [Examples]

[0071] Example 1. Preparation of AAV capsids rAAV2 / 8 was prepared by triple transfection of the chemically defined HEK293 cell line in culture medium suspension. Rep2-Cap8 and helper plasmids were used with a cis construct containing a GFP expression cassette sandwiched between reverse terminal repeats (ITRs) from AAV2. The plasmids were mixed in a molar ratio of 1:1:1 and transfected into cells using PEI MAX transfection reagent (Polysciences). Transfection was performed in a 5 L agitated biostat B-DCU bioreactor (Sartorius) using a fed-batch method. Cell lysis was performed for 72 hours post-transfection by directly adding Tween20 (Sigma-Aldrich) surfactant to the bioreactor. The material was collected and frozen at -80°C until use. The lysed AAV 2 / 8 serotype samples were clarified and then treated with a TFF pre-capture step combined with DNase treatment. The samples were captured and purified using a cation exchange chromatography column - CIMmultus® SO3 (Sartorius BIA Separations). The capture step eluate was concentrated, and the fully loaded AAV capsid was enriched using an anion exchange CIMmultus® QA column. The separately recovered empty and fully loaded AAV capsids, including the fractions, were then buffer-exchanged to formulation buffer using a Vivaspin Turbo 100kDa PES concentrator. Pools of empty and fully loaded fractions at ratios of 1:2 and 2:1, and the prepared samples, were used for all analytical and preparative operations, respectively.

[0072] Example 2. 2a. Effects of different organic modifiers on the separation capability of empty / fully loaded AAV capsids Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL strong anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated at pH 8.5 with 2 mM magnesium acetate, 2.5% organic modifier, 20 mM TRIS, and 1% sorbitol, and eluted at pH 8.5 with a linear salt gradient to 80 mM magnesium acetate, 2.5% organic modifier, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% organic modifier, 20 mM TRIS, and 1% sorbitol were used at pH 8.5.

[0073] A screening of several readily available organic modifiers is shown in Figure 1, and the highest separation quotients of approximately 2.28 for empty and fully loaded AAV capsids are provided by isopropanol, acetonitrile, then ethanol, tert-butanol, and 1-butanol, as shown in Figure 1. Other organic solvents tested provide lower separation quotients and are therefore less preferred, such as propylene carbonate and methanol.

[0074] In addition to empty and fully loaded AAV capsid separation, all tested organic modifiers exhibit sub-population separation.

[0075] 2b-Effect of replacing conventionally used poloxamer 188 organic modifiers with 2b-empty / full AAV capsids, and the effect of different loading and elution salts. Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL+ anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated and eluted using the following: Regarding Figure 2A: The solution was equilibrated at pH 9.3 using 2 mM magnesium chloride, 20 mM BTP, 1% sorbitol, and 2.5% acetonitrile or poloxamer 188. Elution was performed at pH 9.3 using a linear salt gradient to 2 mM magnesium chloride, 400 mM potassium acetate, 20 mM BTP, 1% sorbitol, and 2.5% acetonitrile or poloxamer 188. The volumetric flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile or 0.1% poloxamer 188, and 20 mM BTP were used at pH 9.3. Regarding Figure 2B: The solution was equilibrated at pH 8.5 using 2 mM magnesium acetate, 2.5% acetonitrile or 0.1% poloxamer 188, 20 mM TRIS, and 1% sorbitol. Elution was performed at pH 8.5 using a linear salt gradient to 50 mM magnesium acetate, 2.5% acetonitrile or 0.1% poloxamer 188, 20 mM TRIS, and 1% sorbitol. The volumetric flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile or 0.1% poloxamer 188, and 20 mM TRIS were used at pH 8.5.

[0076] Figures 2A and 2B show a comparison of conventionally used organic modifiers, poloxamer 188 and acetonitrile. In both experiments, regardless of the salt used, the separation performance was improved by approximately 20–40% only by replacing poloxamer 188 with acetonitrile. Furthermore, the percentage of AAV capsid in high-salt concentration stripping was lower with acetonitrile in both cases (Figures 2A and 2B). Regarding Figure 2C: The solution was equilibrated at pH 8.5 using 2 mM magnesium acetate / magnesium formate / magnesium hydrochloride, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. Elution was performed at pH 8.5 using a linear salt gradient to 50 mM magnesium acetate / magnesium hydrochloride (80 mM in the case of magnesium formate), 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. The volumetric flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile, and 20 mM TRIS were used at pH 8.5.

[0077] From Figure 2C, it is clear that magnesium acetate exhibits the best separation ability compared to formic acid, and especially compared to the inorganic chlorides tested.

[0078] Effect of individual organic modifiers compared to poloxamer 188 on the separation ability of 2c-empty / fully loaded AAV capsids. Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL strong anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated at pH 8.5 with 2 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol, and eluted at pH 8.5 with a linear salt gradient to 80 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol were used at pH 8.5.

[0079] The corresponding buffer combination provides a high resolution of 2.40 for empty and fully loaded AAV capsids, as shown in Figure 3, compared to a resolution of 1.89 when using poloxamer 188. Combinations of poloxamer 188 with an organic modifier, such as acetonitrile, would be lower compared to acetonitrile alone. Elution was performed using a linear salt gradient of magnesium acetate from 2 mM to 80 mM using a 160 column volume (CV). In addition to the separation of empty and fully loaded AAV capsids, the buffer combination exhibits subgroup separation.

[0080] Effect of different percentages of organic modifiers on the separation capability of 2d-empty / fully loaded AAV capsids Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL strong anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated at pH 8.5 with 2 mM magnesium acetate, 1% acetonitrile, 20 mM TRIS, and 1% sorbitol, and eluted at pH 8.5 with a linear salt gradient to 50 mM magnesium acetate, 1% acetonitrile, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 1% acetonitrile, 20 mM TRIS, and 1% sorbitol were used at pH 8.5. For comparison, 1% poloxamer 188 was used instead of acetonitrile. X: 1%, 2.5%, 5%, 10%, 20% acetonitrile

[0081] The best results, as shown in Figure 4, were found to be with 5% and 2.5% acetonitrile (highest separation efficiency between empty and fully loaded AAV8 and the lowest percentage high-salt concentration stripping). However, the lowest possible percentage of organic modifiers, such as acetonitrile, is preferred for reasons of the possibility of scaling up the method to a preparative scale and for safety reasons. With higher or lower percentages of acetonitrile, the separation efficiency decreases. With 1% poloxamer 188, the separation efficiency was lowest, and the percentage of AAV was highest in high-salt concentration stripping (Figure 4).

[0082] 2e - Effect of different magnesium acetate or calcium acetate loading concentrations on the separation efficiency of empty / full AAV capsids Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL strong anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated at pH 8.5 with X mM magnesium acetate or calcium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. Elution was performed at pH 8.5 with a linear salt gradient to Y mM magnesium acetate or calcium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol were used at pH 8.5. X: 0 mM, 0.5 mM, 5 mM, 10 mM, 20 mM magnesium acetate or calcium acetate Y: The elution concentration of magnesium acetate or calcium acetate was adjusted to reach the same gradient rate as the standard 2-80 mM magnesium acetate method.

[0083] In general, calcium acetate is less preferred than magnesium acetate (Figure 5), which provides significantly better separation performance. Furthermore, the preferred concentration of loading magnesium acetate is about 5 mM. Magnesium acetate at concentrations of 0.5 mM and 10 mM yields slightly worse results.

[0084] 2f - Effect of different pH values ​​on the separation efficiency of empty / full AAV capsids Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL strong anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated at pH 7.50–9.25 with 2 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol. Elution was performed at pH 7.50–9.25 using a linear salt gradient to 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol were used at pH 7.50–9.25.

[0085] The highest separation efficiency between empty and fully loaded AAV8 capsids, as well as the lowest percentage of high-salt concentration stripping, is achieved using a buffer solution at pH 8.50, as illustrated in Figure 6. Other pH values ​​(particularly below 8.00 and above 8.75) yield worse results.

[0086] 2g - Extension of the method to other AAV serotypes Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL-plus anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated and eluted at pH 8.50 using 2 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. Elution was then performed at pH 8.50 using a linear salt gradient to 80 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol. The volume flow rate was 1 mL / min. As the stripping buffer, 2000 mM potassium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol were used at pH 8.50.

[0087] The method of the present invention was also tested for AAV2 and AAV9 serotypes (Figure 7). However, it is recommended to modify the method according to each serotype to improve separation accuracy.

[0088] Example 3. Preparative separation using QA Empty fractions from fully loaded AAV capsids were separated using a CIMmultus® QA-1 mL (2 μm) column. The buffer and elution conditions of Example 2 were used, except that ethanol was used instead of acetonitrile as the organic modifier. Fractions E1-E4 from the fractionation operation were recovered as shown in Figure 8A. The fractions were analyzed individually on an analytical scale using a multi-detector setup with UV 260 nm and 280 nm, light scattering, intrinsic protein fluorescence mainly induced by tryptophan, and exogenous fluorescence using an intercalation dye. Tryptophan is abundant in the AAV capsid protein, accounting for 2.2% of the capsid [13, 14], and its fluorescence is measured at excitation 280 nm and emission 348 nm. Exogenous fluorescence is measured at excitation 485 nm and emission 520 nm using the intercalation dye picogreen to amplify the sensitivity of nucleic acid impurities

[15] . Figure 8B shows the UV results, where fraction E1 is empty capsid with a 260 / 280 wavelength ratio of 0.64. Fraction E2 was recovered between empty and fully loaded peaks (valley fraction) and shows two populations with 260 / 280 wavelength ratios of 0.82 and 0.73. This fraction is very likely to be partially loaded capsid or empty capsid-related DNA obtained from exogenous picogreen fluorescence. Fraction E3 is filled with fully loaded capsid with a 260 / 280 wavelength ratio of 1.37. Fraction E4 (tail fraction) is very likely to be damaged AAV capsid and / or aggregates. The eluted fractions monitored by the light scattering detector are shown in Figure 8C. The eluted fractions monitored by tryptophan fluorescence are shown in Figure 8D, and the eluted fractions monitored by picogreen fluorescence are shown in Figure 8E. The high picogreen fluorescence signal in the E2 fraction (Figure 8E) indicates a significant amount of DNA-related impurities compared to other eluted fractions. Note: The UV260 / 280 ratio is abbreviated as Rat in Figure 8B.

[0089] Example 4. Separation of empty and fully loaded AAV capsids using a weak anion exchanger Empty and fully loaded AAV capsids were separated using a 100 μL CIMac® DEAE column.

[0090] The column was equilibrated at pH 9.0 with 20 mM TRIS, 0.5% acetonitrile or 1% poloxamer 188, and 1% sorbitol, and eluted at pH 9.0 with a linear salt gradient to 50 mM magnesium acetate, 20 mM TRIS, 0.5% acetonitrile or 1% poloxamer 188, and 1% sorbitol. The volumetric flow rate was 1 mL / min. Due to weak sample binding, no salt was added to the equilibrium buffer. Figure 9 illustrates the results showing that the resolution was slightly improved when an organic modifier was introduced into the buffer. As the stripping buffer, 2000 mM potassium acetate, 0.5% acetonitrile or 0.1% poloxamer, and 20 mM TRIS were used at pH 9.0.

[0091] Example 5. Separation of empty and fully loaded AAV capsids using multimodal materials Empty and fully loaded AAV capsids were separated using a 100 μL CIMac® PrimaT column, which contains a multimodal metal affinity ligand that binds hydrogen to anion exchange chromatography.

[0092] Sample binding on the column was achieved at neutral pH by predominantly hydrogen bonding using 25 mM HEPES, 1% sucrose, 0.1% poloxamer 188, or 2.5% acetonitrile buffer; pH transition was performed at pH 9.0 using 50 mM Tris, 13.6 mM borate, 1% sucrose, 0.1% poloxamer 188, or 2.5% acetonitrile, to influence electrostatic interactions more than hydrogen bonding. Most of the fully loaded AAV capsids and some of the empty AAV capsids were eluted at pH 9.0 using a linear salt gradient to 50 mM magnesium chloride, 50 mM Tris, 9.6 mM borate, 1% sucrose, 0.1% poloxamer 188, or 2.5% acetonitrile. Subsequently, the remaining empty AAV capsids were eluted at pH 9.0 using a high-salt linear gradient to 2 M NaCl, 50 mM Tris, 12 mM borate, 1% sucrose, 0.1% poloxamer 188, or 2.5% acetonitrile. The volumetric flow rate was 1 mL / min.

[0093] When poloxamer 188 was replaced with acetonitrile, separation efficiency was improved, and fewer empty AAV capsids were observed on high-salt linear gradients [Figure 10].

[0094] Example 6. Effect of organic modifier concentration The effect of higher percentages of organic modifiers combined with salt gradients on baseline separation of AAV capsids was investigated using a CIMac® QA column. The column was equilibrated at pH 8.5 (Buffer A) with 10 mM magnesium acetate, 50 mM TRIS, 2% acetonitrile, and 1% sorbitol. Elution was first performed at pH 8.5 using a linear acetonitrile gradient to 30%, followed by 10 mM magnesium acetate, 50 mM TRIS, and 1% sorbitol. After using the acetonitrile gradient, the column was stopped in Buffer A to reduce the percentage of acetonitrile, and elution was performed at pH 8.5 using a salt gradient to 50 mM magnesium acetate, followed by 50 mM TRIS, and 1% sorbitol. The volumetric flow rate was 1 mL / min.

[0095] Using higher concentrations of organic modifiers, followed by elution conditions employing a salt gradient, enabled baseline separation of empty and fully loaded AAV capsids from other species. In Figure 11A, UV 260nm and 280nm chromatograms show the separation of AAV capsids achieved by a combination of reversed-phase conditions (elution of empty AAV capsids with a 260 / 280 wavelength ratio of 0.64 and elution of partially loaded AAV capsids with a 260 / 280 wavelength ratio of 1.08), followed by anion exchange elution conditions with a dominant peak of fully loaded capsids with a 260 / 280 wavelength ratio of 1.32 and a tail peak with a 260 / 280 wavelength ratio of 1.25. Higher temperatures, 40°C, and the combination of reversed phase and anion exchange conditions allowed for baseline separation of empty and partially loaded AAV capsids from fully loaded AAV capsids, particularly fully loaded and heavily loaded AAV capsids, or other subpopulations of aggregates. The same trend was observed in the tryptophan fluorescence chromatogram in Figure 11B.

[0096] Example 7. Orthogonal analysis of preparative fractions using density gradient ultracentrifugation was combined with a PATfix® multiple detector array

[19] . During density gradient ultracentrifugation, the capsid population was separated based on their densities; under the CsCl gradient, fully loaded capsids separated to the bottom and empty capsids to the top. After ultracentrifugation, the fractions were pumped for analysis of UV 260nm and 280nm signals (Figure 12A) and tryptophan-specific fluorescence (Figure 12B). The results of the UV 260nm and 280nm signals for the E1 fraction (Figure 12A), eluted at 6.85 min, were consistent with empty AAV capsids. In the E2 fraction, partially loaded AAV capsids were observed at 6.78 min. A distinct fronting peak in the E2 fraction at approximately 5.5–6.5 min indicated the presence of slightly overloaded (heavy) capsids compared to empty capsids. At 5.17 min, the E3 fraction mainly showed fully loaded capsids. The E4 fraction eluted faster than the E3 fraction and showed a fronting peak at 3-4 min indicating the presence of heavier AAV capsids or aggregates.

[0097] The obtained density ultracentrifugation results support the analytical and preparative results.

[0098] Example 8. Comparison of AAV capsid separation using QA column and anion exchange membrane adsorbent. The sample material described in Example 1 was loaded onto a monolithic anion exchange column and a Sartobind® Q-3 mL (Sartorius) membrane adsorbent column for preparative extraction. The buffer and elution conditions of Example 3 were used. As shown in Figure 13, similar separation profiles were obtained with both the QA column and the membrane adsorbent.

[0099] Example 9. Preparation of AAV capsids rAAV2 / 8 was prepared by triple transfection of the chemically defined HEK293 cell line in a culture medium suspension. Rep2-Cap8 and helper plasmids were used with a cis construct containing a GFP expression cassette sandwiched between reverse terminal repeats (ITRs) from AAV2. The plasmids were mixed in a molar ratio of 1:1:1 and transfected into cells using PEI MAX transfection reagent (Polysciences). Transfection was performed in a 5 L agitated biostat B-DCU bioreactor (Sartorius) using a fed-batch method. Cell lysis was performed for 72 hours post-transfection by directly adding Tween20 (Sigma-Aldrich) surfactant to the bioreactor. The material was collected and frozen at -80°C until use. The lysed collected material of the AAV 2 / 8 serotype (used for our own experiment in Example 2) was clarified and then treated with a TFF pre-capture step combined with DNase treatment. The sample was captured and purified using a cation exchange chromatography column - CIMmultus™ SO3 (Sartorius BIA Separations). The capture step eluate was concentrated, and the buffer was replaced with formulation buffer using a Vivaspin Turbo 100kDa PES concentrator for experimental use. This sample was used only for the experiments in Examples 10 and 11.

[0100] Example 10. Effects of different loading and elution variables on the separation capability of empty / fully loaded AAV capsids Analytical separation of empty and fully loaded AAV capsid samples was performed using a 100 μL+ anion exchanger and a CIMac® AAV fully loaded / empty column. The column was equilibrated with a loading buffer (Buffer A) containing different magnesium salts or different organic solvents or percentages of organic solvents acting as organic modifiers. In two cases, the percentage of organic solvent was 0.0% (decrease gradient of organic modifier). Buffer A (Loading Buffer): X mM magnesium salt, Y% organic modifier, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B (elution buffer): 50 mM magnesium acetate, Y% organic modifier, 20 mM TRIS, and 1% sorbitol, pH 8.5. The sample was loaded into buffer A and eluted using a linear salt gradient to buffer B. The volume flow rate was 1 mL / min.

[0101] After elution with buffer B, buffer C was applied to elute any remaining, more electronegative compounds. Buffer C (high-salt wash): 2000 mM potassium acetate, 2.5% ethanol, 20 mM TRIS, pH 8.5

[0102] 1. Experiment: Buffer A: 5 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 2. Experiment: Buffer A: 5 mM magnesium lactate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 3. Experiment: Buffer A: 5 mM magnesium formate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 4. Experiment: Buffer A: 5 mM magnesium hydrochloride, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 5. Experiment: Buffer A: 5 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5

[0103] The sample was prepared with 5 mM magnesium hydrochloride, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, at pH 8.5.

[0104] 6. Experiment: Buffer A: 5 mM magnesium acetate, 2.5% acetonitrile, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B: 50 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 7. Experiment: Buffer A: 5 mM magnesium acetate, 2.5% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B (no organic modifier): 50 mM magnesium acetate, 20 mM TRIS, and 1% sorbitol, pH 8.5 8. Experiment: Buffer A: 5 mM magnesium acetate, 20.0% ethanol, 20 mM TRIS, and 1% sorbitol, pH 8.5 Buffer B (no organic modifier): 50 mM magnesium acetate, 20 mM TRIS, and 1% sorbitol, pH 8.5

[0105] The above results suggest that all options provide comparable results with no significant differences between experiments regarding separation efficiency between empty and fully loaded AAV capsids, or regarding the percentage of fully loaded AAV8s. The lowest separation efficiency and estimated percentage for fully loaded AAVs was achieved when loading with high-percentage ethanol and eluting with 0-percentage ethanol, as shown in Figure 14, Sample 8. Furthermore, loading with less cosmotropic salt (magnesium chloride) resulted in lower separation efficiency, particularly in Figure 14 (Sample 4). On the other hand, as expected, loading with more cosmotropic salt (magnesium lactate) resulted in slightly higher separation efficiency compared to magnesium acetate (Sample 1) in Figure 14.

[0106] When combining different dissolution strategies, having robust binding and loading conditions to obtain reproducible results is beneficial.

[0107] Example 11. PATfix AAV Switcher for lysate samples (Sartorius BIA Separations, Aidovština, Slovenia) The following buffers were used for AEX separation: Buffer A: 20 Tris + 5 mM magnesium acetate + 1% sorbitol + 2.5% EtOH; pH 8.50 Buffer B: 20 ​​Tris + 65 mM magnesium acetate + 1% sorbitol + 2.5% EtOH; pH 8.50 Buffer solution C: 500 mM sodium acetate; pH 5.50 Buffer solution D: 100 mM sodium hydroxide + 2000 mM sodium chloride

[0108] Empty, partially loaded, fully loaded, and / or other capsids in complex lysate samples were separated using a CIMac® QA column under similar conditions to those described above. Lysate samples were first purified using a cation exchange column with a pH gradient from pH 4.50 to pH 9.50. Next, the pH gradient elution rates were inline redirected on a second AEX column where multiple subgroup separations were performed. In addition to empty and fully loaded capsids, further partially loaded or other impurities were observed (Figure 15). The fronting peak at approximately 8–10 minutes represents several protein impurities present in the collected sample (observable by tryptophan fluorescence but not by light scattering). The peak at approximately 13.7 minutes corresponds to the most likely other high molecular weight protein-related impurities present in the collected sample (peaks observed by fluorescence and light scattering). Other contaminants eluted during the AEX column washing step (from approximately 18 minutes). One-dimensional analysis would yield similar results, but it would increase the likelihood of more impurities from the sample eluting co-eluting with the AAV capsid, and further shorten the column lifetime due to highly negatively charged impurities (which can be removed by in-line purification of the cation exchange (CEX) column) remaining bound to and remaining on the AEX column.

[0109] Table 1 [Table 1-1] [Table 1-2] [Table 1-3]

[0110] References [1]JFWright,AAV empty capsids:for better or for worse? Mol.Ther.22(2014)1-2. [2]C. Ling, Y. Wang, Y. Lu, L. Wang, G. R. Jayandharan, G. V. Aslanidi, B. Li, B. Cheng, W. Ma, T. Lentz, C. Ling, X. Xiao, R. J. Samulski, N. Muzyczka, A. Srivastava, The adeno-associated virus genome packaging puzzle, J. Mol. Genet. Med. 9 (2015) 175 [3]K. Gao, M. Li, L. Zhong, Q. Su, J. Li, S. Li, R. He, Y. Zhang, G. Hendricks, J. Wang, G. Gao, Empty virions in AAV8 vector preparations reduce transduction effi-ciency and may cause total viral particle dose-limiting side effects, Mol. Ther. Methods Clin. Dev. 1 (2014) 20139. [4]Sihn CR, Handyside B, Liu S et al. Molecular analysis of AAV5-hFVIII-SQ vector-genome-processing kinetics in transduced mouse and nonhuman primate livers. Mol Ther Methods Clin Dev. 2021 Dec 21; 24:142-153. [5]J. F. Wright, AAV vector manufacturing process design and scalability-Bending the trajectory to address vector-associated immunotoxicities. Mol Ther. 2022 Jun 1;30(6):2119-2121. [6]International Publication No. 2022 / 038164 (A1) [7] C. Reichardt, Empirical Parameters of Solvent Polarity as Linear Free-Energy Relationships. 1979 Jan Angewandte Chemie International Edition 18:98-110. [8] S. Dukiie, F. Shoh, K: D. Nulre, R. Radeglia, Ukr. Khim. Zh (Russ. Ed.) 41, 1170 (1975); Chem. Abstr. 84, 430861 (1976). [9] United States Pharmacopeia 621 Chromatography, 2022 Dec.

[10] Lock M, Vandenberghe LH, Wilson JM. Scalable production method for AAV (U.S. Patent Application Publication No. 2016 / 0040137(A1)), 2016.

[11] P. Gagnon, B. Goricar, S. D. Prebil, H. Jug, M. Leskovec, A. Strancar, Separation of Empty and Full Adeno-Associated Virus Capsids from a Weak Anion Exchanger by Elution with an Ascending pH Gradient at Low Ionic Strength, 2021. https: / / bioprocessingjournal.com / afp / J20OA-Gagnon.pdf (accessed May 25, 2022).

[12] P. Gagnon, M. Leskovec, S. D. Prebil, Rok Zigon, M. Stokelj, A. Raspor, S. Peljhan, A. Strancar. J Chromatogr A. Removal of empty capsids from adeno-associated virus preparations by multimodal metal affinity chromatography 2021 Jul 19; 1649:462210.

[13] Chen,R.F.Fluorescence quantum yield of tryptophan and tyrosine.Analyt.Lett.1967,1,35-42.

[14] Ghisaidoobe,A.B.T.; Chung,S.A.Intrinsic tryptophan fluorescence in the detection and analysis of proteins: A focus on the Forster resonance energy transfer techniques.Intl.J.Mol.Sci.2014,15,22518-22538.

[15] Singer,V.; Jones,L.; Yue,S.T.; Hauglund,R.P.Characterization of Picogreen reagent and development of a fluorescence-based solution assay for double-stranded DNA quantitation.Anal.Biochem.1997,249,228-238.

[16] Yang,Yun; Geng,Xindu(2011).“Mixed-mode chromatography and its applications to biopolymers”.Journal of Chromatography A.1218(49): 8813-8825.doi:10.1016 / j.chroma.2011.10.009.ISSN 0021-9673.PMID 22033107.

[17] Zhao,Guofeng; Dong,Xiao-Yan; Sun,Yan(2009).“Ligands for mixed-mode protein chromatography: Principles,characteristics and design”.Journal of Biotechnology.144(1): 3-11.doi:10.1016 / j.jbiotec.2009.04.009.ISSN 0168-1656.PMID 19409941.

[18] McLaughlin,Larry W.(1989).“Mixed-mode chromatography of nucleic acids”.Chemical Reviews.89(2): 309-319.doi:10.1021 / cr00092a003.ISSN 0009-2665.

[19] Peljhan S,Stokelj M,Prebil SD,Gagnon P,Strancar A.Multiple-parameter profiling of density gradient ultracentrifugation for characterization of empty and full capsid distribution in AAV preparations.Cell Gene Ther.Insights 2021; 7(2),161-169.

[20] Hildegard Buning,Arun Srivastava Capsid Modifications for Targeting and Improving the Efficacy of AAV Vectors.Mol Ther Methods Clin Dev.2019 12: 248-265.

[21] Pierce J.Ogden,Heric D.Kelsic,Sam Sinai,George M.Church Comprehensive AAV capsid fitness landscape reveals a viral gene and enables machine-guided design.Science 2019 Vol 366,Issue 6469 pp.1139-1143.

Claims

1. A method for enriching a mixture containing full adeno-associated virus (AAV) capsids, partially filled AAV capsids, and / or empty AAV capsids by chromatography, The aforementioned method, - The step of bringing the mixture into contact with a strong anion exchange material or a weak anion exchange material, - The loaded mixture is eluted with a neutral to alkaline buffer containing an organic modifier having a relative solvent polarity of 0.4 to 0.

8. - A step in which the fully loaded AAV capsid recovers the enriched fraction, The method, including the method described above.

2. The method according to claim 1, wherein the organic modifier is selected from the group consisting of acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol, propanol, and mixtures thereof.

3. The method according to claim 1 or 2, wherein the buffer solution comprises an alkaline earth metal salt, and in particular the alkaline earth metal salt is magnesium acetate, magnesium formate, calcium acetate, or calcium formate, or a mixture thereof and / or a more cosmotropic alternative thereof.

4. The method according to any one of claims 1 to 3, wherein the buffer solution has a pH value of pH 7.0 to pH 10.

50.

5. The aforementioned buffer solution is - An isotonic substance selected from the group consisting of sucrose, sorbitol, mannitol, xylitol and mixtures thereof, and / or - Nonionic surfactants such as poloxamer 188 The method according to any one of claims 1 to 4, including the method described in any one of claims 1 to 4.

6. The strong anion exchange material or the weak anion exchange material is (i) Multimodal materials having or not having hydrogen bonding properties, and strong anion exchange materials or weak anion exchange materials with or without charged metal affinity ligands, (ii) Monolithic anion exchanger, (iii) Monolithic multimodal material, (iv) Particulate anion exchanger, (v) Particulate multimodal materials, (vi) anion exchanger or multimodal material arranged in the film, and / or (vii) Particle-packed anion exchanger or multimodal column, and / or (viiii) Fiber chromatography anion exchanger or multimodal fiber column, The method according to any one of claims 1 to 5.

7. The method according to any one of claims 1 to 6, wherein the AAV is selected from serotypes AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh. 10, AAV11, AAV12, and further serotypes such as hybrid serotypes (particularly recombinant hybrid serotypes such as AAV2 / 8, chimeras, surface-modified AAVs, and synthetically derived AAV-like particles).

8. An aqueous solution having a pH ranging from neutral to alkaline, comprising a buffering substance and an organic modifier having a relative scale of solvent polarity equivalent of 0.4 to 0.

8.

9. The aqueous solution according to claim 8, wherein the organic modifier is selected from the group consisting of acetonitrile, 1-butanol, t-butanol, propylene carbonate, isopropanol, ethanol, methanol, propanol, and mixtures thereof.

10. An aqueous solution according to claim 8 or claim 9, comprising an alkaline earth metal salt.

11. The aqueous solution according to claim 10, wherein the alkaline earth metal salt is a salt of magnesium or calcium and a mixture thereof, and in particular the alkaline earth metal salt is magnesium acetate, magnesium formate, calcium acetate or calcium formate or a mixture thereof and / or a more cosmotropic alternative thereof.

12. The buffering substance buffers aqueous solutions in the pH range of 7 to 12, and is particularly composed of 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol (Bis-Tris), 2,2',2''-nitrilotriacetate (ADA), 2-[(2-amino-2-oxoethyl)amino]ethane-1-sulfonic acid (ACES), 2,2'-(piperazine-1,4-diyl)di(ethane-1-sulfonic acid) (PIPES), and 2-hydroxy-3-(morpholine-4-yl)propane-1-sulfonic acid (MOPSO). , 2,2′-[propane-1,3-diylbis(azandiyl)]bis[2-(hydroxymethyl)propane-1,3-diol](BTP), N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES), 3-(morpholine-4-yl)propane-1-sulfonic acid (MOPS), 2-{[1,3-dihydroxy-2-(hydroxymethyl)propane-2-yl]amino}ethane-1-sulfonic acid (TES), 2-[4-(2-hydroxyethyl)piperazine-1-yl]ethane-1-sulfonic acid (HE PES), 3-[N,N-bis(2-hydroxyethylamino)-2-hydroxy-1-propanesulfonic acid (DIPSO), 4-(4-morpholinyl)butanesulfonic acid (MOBS), 2-hydroxy-3-[tris(hydroxymethyl)methylamino]-1-propanesulfonic acid (TAPSO), 2-amino-2-(hydroxymethyl)-1,3-propanediol (Trizma), 4-(2-hydroxyethyl)piperazine-1-(2-hydroxypropane-3-sulfonic acid) (HEPPSO), piperazine-N,N '-Bis(2-hydroxypropanesulfonic acid), (POPSO), triethylamine (TEA), 4-(2-hydroxyethyl)-1-piperazine-propanesulfonic acid (EPPS), N-tris(hydroxymethyl)methylglycine (Tricine), N,N-bis(2-hydroxyethyl)glycine (Bicin), N-(2-hydroxyethyl)piperazine-N'-(4-butanesulfonic acid) (HEPBS), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TAPS), 2-amino-2-methyl-1,An aqueous solution according to any one of claims 8 to 11, wherein the buffering substance is selected from the group consisting of 3-propanediol (AMPD), N-tris(hydroxymethyl)methyl-4-aminobutanesulfonic acid (TABS), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPSO), 2-(N-cyclohexylamino)ethanesulfonic acid (CHES), sodium salt of 3-cyclohexylamino-2-hydroxypropanesulfonate (CAPSO), N-(1,1-dimethyl-2-hydroxyethyl)-3-amino-2-hydroxypropanesulfonic acid (AMPS), 3-(cyclohexylamino)-1-propanesulfonic acid (CAPS), and 4-[cyclohexylamino]-1-butanesulfonic acid (CABS).

13. The aforementioned buffer solution is - Isotonic additives (in particular, isotonic additives selected from the group consisting of sucrose, sorbitol, mannitol, xylitol and mixtures thereof), and / or - Nonionic surfactants such as poloxamer 188, An aqueous solution according to any one of claims 8 to 12, comprising:

14. Use of an aqueous solution according to any one of claims 8 to 13 for separating a fully loaded adeno-associated virus (AAV) capsid from an empty AAV capsid, particularly for separation by the method according to any one of claims 1 to 7.

15. A method for enriching a mixture containing fully loaded adeno-associated virus (AAV) capsids, partially loaded AAV capsids, and / or empty AAV capsids by chromatography, - The step of bringing the mixture into contact with a strong anion exchange material or a weak anion exchange material, - Methanol, 1,3-propanediol, 1,2-propanediol, N-methylformamide, diethylene glycol, triethylene glycol, 1,3-butanediol, 2-propyne-1-ol (propargyl alcohol), 2-methoxyethanol, 2-propen-1-ol (allyl alcohol), N-methylacetamide, ethanol, 2-aminoethanol, acetic acid, benzyl alcohol, 1-propanol, 1-butanol, 2-hydroxymethylfuran (furfuryl alcohol), 2-phenylethanol, 1-pentanol, 2-methyl-1-propanol (isobutyl alcohol), 1-hexanol, 2-propanol, 3-phenyl-1-propanol, 1-heptanol, 1-octanol, cyclopentanol, 1-decanol, 2,6-dimethylphenol (2, The steps include: eluting the loaded mixture with a neutral to alkaline buffer containing an organic modifier selected from the group consisting of 6-xylenol, 2-butanol, 3-methyl-1-butanol (isoamyl alcohol), cyclohexanol, 1-dodecanol, 1-phenylethanol, acrylonitrile, 4-methyl-1,3-dioxolan-2-one (propylene carbonate), 2-pentanol, nitromethane, acetonitrile, dimethyl sulfoxide, methyl acrylate, aniline, tetra-N-hexylammonium benzoate, tetrahydrothiophene 1,1-dioxide (sulfolane), 2-methyl-2-propanol (tert-butyl alcohol), acetic anhydride, N,N-dimethylformamide, N,N-dimethylacetamide, propionitrile, and nitroethane; - A step in which the fully loaded AAV capsid recovers the enriched fraction, Methods that include...

Images