Analytical ultracentrifugation for characterizing recombinant virus particles

Abstract

To provide methods for characterizing a preparation of recombinant viral particles for monitoring drug product quality with regard to homogeneity, purity and consistency of manufacturing.SOLUTION: A method of characterizing a preparation of recombinant viral particles comprises the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant viral particles is monitored at time intervals; b) plotting differential sedimentation coefficient distribution value (C(s)) versus sedimentation coefficient in Svedberg units (S); and c) integrating areas under respective peaks in the C(s) distribution to determine a relative concentration of each peak, wherein each peak represents a species of recombinant viral recombinant viral particle.SELECTED DRAWING: None

Description

[Technical Field]

[0001] Cross-reference of related patents This application claims priority to U.S. Provisional Patent Application No. 62 / 105,714, filed January 20, 2015, which is incorporated herein by reference in its entirety for all purposes.

[0002] Field of Invention The present invention relates to a method for characterizing recombinant viral vectors; for example, recombinant adeno-associated virus (AAV) particles, recombinant adenovirus (rAd) particles, recombinant lentivirus particles, and recombinant herpes simplex virus (rHSV) particles using analytical ultracentrifugation. [Background technology]

[0003] Recombinant viruses have shown potential and utility as vehicles for delivering therapeutic nucleic acids for gene therapy applications. Numerous different recombinant viruses are used in these gene therapy applications based on a number of factors, including the size of the nucleic acid being delivered, the target cells or tissues to which the nucleic acid is delivered, the need for short-term or long-term expression of the therapeutic nucleic acid, and the integration of the therapeutic nucleic acid into the recipient's genome. Examples of viruses used in gene therapy applications include adeno-associated viruses (AAVs), adenoviruses, lentiviruses, and herpes simplex viruses (HSVs). [Overview of the Initiative] [Problems that the invention aims to solve]

[0004] The production of recombinant viral vectors for clinical use requires analytical methods to monitor drug product quality in terms of manufacturing uniformity, purity, and integrity, but to date, methods supporting such characterization have not been established. Typically, the DNA content of recombinant viral DNA vectors is measured by Southern blot analysis using sequence-specific probes. Viral capsids or envelopes are characterized by immunoassays using antibodies that specifically bind to the capsid or envelope protein of a particular recombinant virus. For example, Steinbach, S et al., (1997) J. Gen. Virol., 78: pp. 1453-1462, provides an immunoassay for rAAV serotypes. What is needed is a comprehensive assay for characterizing recombinant viral preparations, regardless of the nucleic acid sequence of the recombinant viral genome or the serotype of the capsid.

[0005] All references cited herein, including patent applications and publications, are incorporated by reference in their entirety. [Means for solving the problem]

[0006] In some embodiments, the present invention provides a method for characterizing a recombinant virus particle preparation, comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of the recombinant virus particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; and c) integrating the area under each peak in the C(s) distribution to determine the relative concentration of each peak, wherein each peak corresponds to the species of recombinant virus particle.

[0007] In some embodiments, the present invention relates to the vector of recombinant virus particles in a preparation of recombinant virus particles. A method for evaluating the integrity of a virus genome is provided, comprising the steps of: a) subjecting a preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution (C(s)) against the sedimentation coefficient (S) in Svedberg units; and c) identifying the species of recombinant virus particles in the preparation by the presence of a peak on the plot corresponding to the S value, wherein the genome size of the specific species of recombinant virus particles is calculated by comparing the S value of the species with a standard curve created by the S values ​​of recombinant virus particles containing capsidized viral genomes having known nucleotide sizes. In some embodiments, the method further comprises the step of integrating the area under each peak in the C(S) distribution to determine the relative concentrations of various recombinant virus particles of the recombinant virus.

[0008] In some embodiments, the present invention provides a method for determining the presence of empty capsids or capsid particles containing variant-sized recombinant viral genomes in a preparation of recombinant viral particles, comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant viral particles is monitored at time intervals; and b) plotting differential sedimentation coefficient distribution values ​​(C(s)) against sedimentation coefficient (S) in Svedberg units, wherein the presence of one or more peaks other than the peak for complete capsid particles containing intact recombinant viral genomes represents the presence of capsid particles containing variant-sized genomes and / or empty capsids.

[0009] In some embodiments, the present invention provides a method for measuring the relative amount of empty capsids in a preparation of recombinant virus particles, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant virus particles; and d) comparing the amount of recombinant virus particles having an S value corresponding to empty capsid particles with the amount of recombinant virus particles having an S value corresponding to recombinant virus particles containing intact viral genomes or the total amount of recombinant virus particles in the preparation.

[0010] In some embodiments, the present invention provides a method for measuring the relative amount of capsid particles containing variant recombinant viral genomes or empty viral capsid particles in a preparation of recombinant viral particles, comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant viral particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant viral particles; and d) comparing the amount of recombinant viral particles having an S value that does not correspond to recombinant viral particles containing intact viral genomes with the amount of recombinant viral particles having an S value that corresponds to recombinant viral particles containing intact viral genomes or the total amount of recombinant viral particles in the preparation.

[0011] In some embodiments, the present invention provides a method for measuring the relative amount of capsid particles containing variant recombinant viral genomes in a preparation of recombinant viral particles, comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant viral particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant viral particles; and d) comparing the amount of recombinant viral particles having an S value that does not correspond to recombinant viral particles containing intact viral genomes or empty capsid particles with the total amount of recombinant viral particles in the preparation. To provide.

[0012] In some embodiments, the present invention provides a method for measuring the relative amount of recombinant virus particles containing intact viral genomes in a preparation of recombinant virus particles, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals; b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant virus particles; and d) comparing the amount of recombinant virus particles having an S value corresponding to recombinant virus particles containing intact viral genomes with the amount of recombinant virus particles having an S value corresponding to empty capsid particles, capsid particles containing variant recombinant virus genomes, and / or the total amount of recombinant virus particles in the preparation.

[0013] In some embodiments, the present invention provides a method for monitoring the removal of empty capsids and / or capsid particles containing variant recombinant viral genomes during the purification of a recombinant viral particle preparation, comprising removing a recombinant viral particle sample from the preparation following one or more steps in the purification process, and analyzing the sample for the relative amount of empty capsids and / or capsid particles containing variant recombinant viral genomes by the method of any one of claims 5 to 8, wherein a decrease in the relative amount of capsids containing empty capsids and / or variant genomes to complete capsids indicates the removal of empty capsids from the recombinant viral particle preparation. In some embodiments, the presence of a peak corresponding to the S value of empty capsid particles indicates the presence of empty capsid particles. In some embodiments, the presence of one or more peaks other than the peak for complete capsid particles containing intact recombinant viral genomes or empty capsid particles indicates the presence of capsid particles containing variant-sized genomes. In some embodiments, the capsid particles containing a variant-sized genome include fragmented genomes, aggregates, recombinants, and / or DNA impurities.

[0014] In some embodiments, the present invention provides a method for determining heterogeneity of recombinant virus particles in a preparation of recombinant virus particles, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals; and b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units, wherein the presence of a peak in addition to the peak corresponding to a capsid containing intact viral genome indicates heterogeneity of recombinant particles in the preparation. In some embodiments, the presence of the additional peak indicates the presence of recombinant virus particles containing empty capsid particles and / or variant genomes. In some embodiments, the variant genome is a fragmented viral genome, aggregates, recombinants and / or DNA impurities. In some embodiments, the method further comprises the step of integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant virus particles.

[0015] In some embodiments, the present invention provides a method for monitoring the homogeneity of recombinant virus particles during the purification of a recombinant virus particle preparation, comprising removing a recombinant virus particle sample from the preparation following one or more steps in the purification process, and determining the heterogeneity of the recombinant virus particles by the method, wherein an increase in the relative amount of recombinant virus particles containing intact viral genomes indicates an increase in the homogeneity of the total virus particles in the recombinant virus particle preparation.

[0016] In some embodiments of the above-described model, the sedimentation of recombinant virus particles is monitored by absorbance. In some embodiments, the absorbance is at approximately 230 nm, 260 nm, or 280 nm. In some embodiments, the absorbance is at approximately 260 nm. In some embodiments, the sedimentation of recombinant virus particles is monitored by interference. In some embodiments, the interference is Rayleigh interference.

[0017] In some embodiments of the above-described model, the preparation is an aqueous solution. In further embodiments, the aqueous solution comprises a pharmaceutical formulation. In some embodiments, the aqueous solution comprises a buffer solution. In some embodiments, the buffer solution has a physiological pH. In some embodiments, the buffer solution has a physiological osmotic pressure. In some embodiments, the pharmaceutical formulation comprises phosphate-buffered saline (PBS). In some embodiments, the PBS has a pH of 7.2 and an osmotic pressure of about 300 mOsm / L. In some embodiments, monitoring further comprises comparison with a reference sample, the reference sample comprising an aqueous solution free of recombinant virus particles.

[0018] In some embodiments of the above configuration, the C(S) value is determined by an algorithm that includes a solution to the Lamb equation. In some embodiments, the algorithm is the SEDFIT algorithm. In some embodiments, sedimentation is monitored until recombinant virus particles with the lowest density settle to the bottom of a sector of the ultracentrifuge; for example, the sector may be part of the ultracentrifuge including a detection system. In some embodiments, the ultracentrifugation method utilizes an ultracentrifuge including an ultracentrifugation velocity cell. In some embodiments, sedimentation is monitored until recombinant virus particles settle to the bottom of the ultracentrifugation velocity cell. In some embodiments, sedimentation is monitored until recombinant virus particles with the lowest density settle and the optical window becomes clear.

[0019] In some embodiments, radial concentration is recorded for at least about 0.5 hours, 0.75 hours, 1.0 hour, 1.5 hours, 2.0 hours, 3.0 hours, 4.0 hours, or 5.0 hours. In some embodiments, radial concentration is recorded for about 1.0 hour. In some embodiments, radial concentration is recorded for about 1.2 hours. In some embodiments, radial concentration is recorded from about 0.5 hours to about 2.0 hours. In some embodiments, radial concentration is recorded from about 1.0 hour to about 2.0 hours.

[0020] In some embodiments of the above configuration, at least 30 scans are used to monitor the sedimentation of recombinant virus particles. In some embodiments, about 30 scans are used to monitor the sedimentation of recombinant virus particles. In other embodiments, about 30 to about 75 scans are used to monitor the sedimentation of recombinant virus particles. In other embodiments, about 30 to about 50 scans are used to monitor the sedimentation of recombinant virus particles. In other embodiments, about 50 to about 75 scans are used to monitor the sedimentation of recombinant virus particles.

[0021] In some embodiments of the above configuration, regularization is applied to the fitting level with an F-statistic confidence level of at least about 0.68. In some embodiments, the regularization is derivative regularization. In some embodiments, the regularization is maximum entropy regularization. In some embodiments, the regularization is applied to the fitting level with an F-statistic confidence level of about 0.68 to about 0.90. In some embodiments, the regularization is applied to the fitting level with an F-statistic confidence level of about 0.68 to about 0.99. In some embodiments, the regularization is applied to the fitting level with an F-statistic confidence level of about 0.68.

[0022] In some embodiments of the above configuration, the following C(S) parameters are kept constant: resolution is approximately 200S to approximately 5000S, Smin is approximately 1S to approximately 100S, Smax is approximately 100S to approximately 5000S, and the friction ratio is approximately 1.0 or transitions to a value determined by the centrifugal separation software. In some embodiments, the resolution is approximately 200S to approximately 1000S. In some embodiments, the resolution is approximately 200S. In some embodiments, Smin is approximately 1. In some embodiments, Smax is approximately 100S to approximately 1. In some embodiments, Smax is approximately 200S to 5000S. In other embodiments, Smax is approximately 200S. In some embodiments, the friction ratio is shifted to a value determined by centrifugal separation software. In some embodiments, the friction ratio is approximately 1.0. In some embodiments, radially invariant (RI) and time-invariant (TI) noise subtraction is applied.

[0023] In some embodiments of the above configuration, the sedimentation of recombinant virus particles is monitored approximately every 10 to 60 seconds. In some embodiments, the sedimentation of recombinant virus particles is monitored (e.g., scanned) approximately every 10 seconds. In other embodiments, the sedimentation of recombinant virus particles is monitored approximately every 60 seconds. In some embodiments, the sedimentation rate of recombinant virus during ultracentrifugation is determined by monitoring the sedimentation of recombinant virus particles once every approximately 15 seconds, 30 seconds, 45 seconds, 1 minute (60 seconds), 2 minutes, 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes, 15 minutes, 20 minutes, and 25 minutes.

[0024] In some embodiments of the above configuration, the boundary settlement velocity is performed at approximately 3,000 rpm to approximately 20,000 rpm. In some embodiments, the boundary settlement velocity is performed at approximately 3,000 rpm to approximately 10,000 rpm. In other embodiments, the boundary settlement velocity is performed at approximately 10,000 rpm to approximately 20,000 rpm. In other embodiments, the boundary settlement velocity is performed at approximately 15,000 rpm to approximately 20,000 rpm.

[0025] In some embodiments of the above-described model, the boundary settlement velocity is performed at approximately 4°C to approximately 20°C. In some embodiments, the boundary settlement velocity is performed at approximately 4°C.

[0026] In some embodiments of the above-described model, the recombinant virus particles are recombinant adeno-associated virus (AAV) particles, recombinant adenovirus particles, recombinant lentivirus particles, or recombinant herpes simplex virus (HSV) particles. In some embodiments, recombinant virus particles include AAV1 capsid, AAV2 capsid, AAV3 capsid, AAV4 capsid, AAV5 capsid, AAV6 capsid, AAV7 capsid, AAV8 capsid, AAVrh8 capsid, AAV9 capsid, AAV10 capsid, AAVrh10 capsid, AAV11 capsid, AAV12 capsid, AAV2R471A capsid, AAV2 / 2-7m8 capsid, AAV DJ capsid, AAV2 N587A capsid, AAV2 E548A capsid, AAV2 N708A capsid, and AAV The recombinant virus particles include V708K capsid, goat AAV capsid, AAV1 / AAV2 chimeric capsid, bovine AAV capsid, or mouse AAV capsid rAAV2 / HBoV1 (chimeric AAV / human bocavirus 1). In some embodiments, the recombinant virus particles include AAV1 ITR, AAV2 ITR, AAV3 ITR, AAV4 ITR, AAV5 ITR, AAV6 ITR, AAV7 ITR, AAV8 ITR, AAVrh8 ITR, AAV9 ITR, AAV10 ITR, AAVrh10 ITR, AAV11 ITR, or AAV12 ITR. In some embodiments, the AAV capsid includes tyrosine mutations or heparin-binding mutations. In other embodiments, the recombinant virus particles are recombinant adenovirus particles. In some embodiments, recombinant adenovirus particles contain capsids derived from adenovirus serotypes 2, 1, 5, 6, 19, 3, 11, 7, 14, 16, 21, 12, 18, 31, 8, 9, 10, 13, 15, 17, 19, 20, 22, 23, 24-30, 37, 40, 41, AdHu2, AdHu3, AdHu4, AdHu24, AdHu26, AdHu34, AdHu35, AdHu36, AdHu37, AdHu41, AdHu48, AdHu49, AdHu50, AdC6, AdC7, AdC69, bovine Ad3, canine Ad2, sheep Ad, or porcine Ad3. In some embodiments, recombinant adenovirus particles include a variant of adenovirus serotype 2 capsid or a variant of adenovirus serotype 5 capsid.In other embodiments, recombinant virus particles are recombinant lentivirus particles. In some embodiments, recombinant lentiviruses. The virus particles are pseudotyped with vesicular stomatitis virus (VSV), lymphocytic choriomeningitis virus (LCMV), Ross River virus (RRV), Ebola virus, Marburg virus, Mocola virus, rabies virus, RD114, or a variant thereof. In other embodiments, the recombinant virus particles are rHSV particles. In some embodiments, the HSV particles are HSV-1 particles or HSV-2 particles.

[0027] In some embodiments, the present invention provides a method for evaluating a process for producing recombinant viral particles comprising the method of any one of claims 1 to 67, wherein an increase in the relative amount of recombinant viral particles containing intact viral genomes compared to the relative amount of empty capsid particles, and / or recombinant viral capsid particles containing variant recombinant viral genomes compared to a reference preparation of recombinant viral particles, indicates an improvement in the production of recombinant viral particles. In some embodiments, the recombinant viral particles are recombinant adeno-associated virus (AAV) particles, recombinant adenovirus particles, recombinant lentivirus particles, or recombinant herpes simplex virus (HSV) particles. In some embodiments, rAAV particles are produced from a producing cell line. In other embodiments, rAAV particles are produced by triple transfection of i) nucleic acids encoding AAV rep and cap, ii) an rAAV vector sequence, and iii) nucleic acids encoding adenovirus helper function. In other embodiments, recombinant viral particles are produced by an AAV / HSV hybrid. In other embodiments, recombinant viral particles are produced from baculovirus cells. In some embodiments, recombinant viral particles are produced by transient transfection into a suitable host cell of a nucleic acid encoding an AAV vector sequence, AAV rep and cap coding regions, and AAV helper virus function. In some embodiments, recombinant viral particles are produced by introduction into a suitable host cell of one or more nucleic acids encoding an AAV vector sequence, AAV rep and cap coding regions, and AAV helper virus function, with one or more nucleic acids being introduced into the cell using a recombinant helper virus. In some embodiments, the recombinant helper virus is an adenovirus or herpes simplex virus. In some embodiments, the recombinant viral particles contain a self-complementary AAV (scAAV) genome. In some embodiments, the method is used to detect the presence of recombinant viral particles containing monomeric or dimeric forms of the scAAV genome.

[0028] In some embodiments of the above-described model, recombinant virus particles are produced by transient transfection of nucleic acids encoding an adenovirus vector sequence and an adenovirus replication and packaging sequence into a suitable host cell. In other embodiments, recombinant virus particles are produced by transient transfection of nucleic acids encoding a lentiviral vector sequence and / or a lentiviral replication and packaging sequence into a suitable host cell. In other embodiments, recombinant virus particles are produced by transient transfection of nucleic acids encoding an HSV vector sequence and / or an HSV replication and packaging sequence into a suitable host cell.

[0029] In some embodiments, the present invention provides a method for preparing recombinant virus particles containing recombinant virus particles and / or variant genomes with reduced empty capsids, comprising: a) culturing host cells under conditions suitable for recombinant virus production, wherein the cells comprise i) a nucleic acid encoding a heterologous transgene having at least one AAV ITR at its end, ii) a nucleic acid comprising AAV rep and cap coding regions and containing a p5 promoter, and iii) a nucleic acid encoding AAV helper virus function; b) lysing the host cells to release recombinant virus particles; c) isolating the recombinant virus particles produced by the host cells; and d) analyzing the recombinant virus particles by ultracentrifugation according to the above method for the presence of recombinant virus particles having empty capsids and / or variant genomes. In some embodiments, the present invention provides a method comprising: The present invention provides a method for preparing recombinant viral particles containing recombinant viral particles and / or variant genomes with reduced empty capsids, comprising: a) culturing host cells under conditions suitable for recombinant virus production, wherein the cells comprise: i) a nucleic acid encoding a heterologous transgene having at least one AAV ITR at its end; ii) a nucleic acid comprising AAV rep and cap coding regions, comprising a mutant p5 promoter, wherein rep expression from the p5 promoter is reduced compared to a wild-type p5 promoter; and iii) a nucleic acid encoding AAV helper virus function; b) lysing the host cells to release recombinant viral particles; c) isolating the recombinant viral particles produced by the host cells; and d) analyzing the recombinant viral particles by analytical ultracentrifugation according to the above method for the presence of recombinant viral particles having empty capsids and / or variant genomes. In some embodiments, the p5 promoter is located at 3' of the rep and / or cap coding regions. In some embodiments, the AAV helper virus function includes adenovirus E1A function, adenovirus E1B function, adenovirus E2A function, adenovirus VA function, and adenovirus E4 or f6 function.

[0030] In some embodiments of the preceding embodiments, recombinant virus particles are purified using one or more purification steps. [Brief explanation of the drawing]

[0031] [Figure 1A] This figure shows that analytical ultracentrifugation (AUC) can be used to characterize recombinant viral vector particles. (Figure 1A) Representative scan profile of boundary sedimentation velocity showing absorbance (260 nm) of AAV2 mixture against radius (cm) at a time interval (T) of 1.2 hours. The AAV2 mixture contained empty capsids ("empty Cap") and whole genome capsids ("intact vector"). [Figure 1B] This figure shows that analytical ultracentrifugation (AUC) can be used to characterize recombinant viral vector particles. (Figure 1B) A plot of concentration in detection units, C(S), against the sedimentation coefficient (Svedberg units, S), showing that AUC can be used to measure the concentrations of empty capsids and whole genome capsids from an 80% / 20% mixture. Each peak shows the particle type as well as its corresponding sedimentation coefficient (S) and relative presence (%). [Figure 2] Figures 2A and 2B are graphs showing the AUC of pure populations of empty AAV2 capsids (Figure 2A) and genome-containing AAV2-transgene-1 capsids (Figure 2B). Each peak is indicated by the capsid species and its sedimentation coefficient (S). [Figure 3] Figures 3A and 3B are graphs comparing interference by AUC and absorbance detection. (Figure 3A) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients created using interference optical detection for a 1:1 mixture of empty capsids and genome-containing capsids. The sedimentation coefficient and relative presence (%) for each type are shown. (Figure 3B) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients created using absorbance optical detection (260 nm) for a 1:1 mixture of empty capsids and genome-containing capsids. The sedimentation coefficient and relative presence (%) for each type are shown. [Figure 4] This figure illustrates a triple transfection method for AAV vector production. Three vectors containing the target gene ("pVector"), the AAV Rep and Cap genes ("pHLP"), and adenovirus components ("pIAdeno") are shown. Note that both genome-containing vectors (indicated as "ITR transgene ITR" in the figure) and empty capsids (blank spaces) are produced. [Figure 5]This figure illustrates a cell line method for AAV vector production. As shown, the HeLa S3 cell line contains the integrated Rep, Cap, and puromycin resistance genes along with the target ITR flanking transgene. This cell line is infected with adenovirus ("Ad5") to stimulate recombinant virus production. Note that both genome-containing (indicated as "recombinant viral vector") and empty capsids are produced in addition to adenovirus particles. [Figure 6] Figures 6A, 6B, and 6C illustrate that vector production by the production cell line and the triple transfection method yields different vector preparations, as is evident from AUC analysis. (Figure 6A) Schematic diagram of the AAV2-transgene 2 vector and its 3.4kb genome. (Figure 6B) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for vector preparations produced by the production cell line method. The sedimentation coefficient and relative presence (%) are shown for each. (Figure 6C) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for vector preparations produced by the triple transfection method. The sedimentation coefficient and relative presence (%) are shown for each. [Figure 7-1] Figures 7A, 7B, and 7C are graphs showing that the AUC method can be used to monitor the quality and effectiveness of vector purification. (Figure 7A) A plot showing the purification of a whole-genome AAV2-transgene 1 capsid from an empty capsid using anion exchange chromatography. The corresponding peak fractions are shown. (Figure 7B) A plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for the vector preparation after elution from the anion exchange column. The sedimentation coefficient and relative presence (%) for each are shown. (Figure 7C) A plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for the vector preparation before chromatography. The sedimentation coefficient and relative presence (%) for each are shown. [Figure 7-2] Continuation of Figure 7-1. [Figure 8] This graph shows the linear relationship between the sedimentation coefficient and the vector genome size. The standard curve plotting the sedimentation coefficient (S) against genome size is shown along with the best-fit line, its equation, and its correlation R² value. [Figure 9-1] Figures 9A, 9B, and 9C show that the assessment of capsid genome size using AUC data correlates with the assessment of genome size by Southern blotting. (Figure 9A) A plot of the differential sedimentation coefficient distribution values, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for scAAV9 EGFP vector preparations. Single-stranded monomer (82S) and double-stranded dimer (101S) species are shown with their corresponding sedimentation coefficients and relative abundance values ​​(%). A schematic diagram of the vector is also provided. (Figure 9B) Alkaline Southern blotting analysis of DNA from scAAV9 EGFP (lane 1) and single-stranded AAV9 EGFP (lane 2) vector capsids. Corresponding bands are shown as described in the blot legend. 4.2 and 2.4 kb size standards are provided as labels. (Figure 9C) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for single-stranded AAV9 EGFP vector preparations. The 82S and 99S (whole genome) peaks are shown with their corresponding sedimentation coefficients and relative abundance values ​​(%). [Figure 9-2] Continuation of Figure 9-1. [Figure 10]Figures 10A, 10B, and 10C illustrate how the Rep / Cap promoter position affects genome packaging in recombinant viral vectors produced by triple transfection. (Figure 10A) Schematic diagram of a self-complementary scAAV2 EGFP vector with estimated sedimentation coefficients for dimeric and monomeric genome species. (Figure 10B) A plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for scAAV2 EGFP vector preparations produced using a "wild-type" helper plasmid with an endogenous p5 promoter ("WT Rep") driving Rep78 / 6 expression. Peaks for single-stranded monomer (80S) and double-stranded dimer (100S) species are shown in their corresponding relative abundance values ​​(%). (Figure 10C) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units yields the distribution of sedimentation coefficients for scAAV2 EGFP vector preparations produced using a “wild-type” helper plasmid with a p5 promoter driving Rep78 / 68 expression migrated downstream of the cap2 sequence (“pHLP Rep”). Peaks for single-stranded monomer (82S) and double-stranded dimer (100S) species are shown in their corresponding relative abundance values ​​(%). [Figure 11]Figures 11A, 11B, 11C, and 11D illustrate how the Rep / Cap promoter positions affect genomic packaging in two additional AAV vectors. (Figures 11A and 11B) Plots of differential sedimentation coefficient distribution values, c(s), against sedimentation coefficient (S) in Svedberg units show the distribution of sedimentation coefficients for single-stranded AAV5 factor IX vector (AAV5 hFIX16) containing a cap5 sequence produced by a helper plasmid with an endogenous p5 promoter ("WT Rep" Figure 11B) or a p5 promoter downstream of the cap5 sequence ("pHLP Rep" Figure 11A). (Figures 11C-11D) The plot of the differential sedimentation coefficient distribution value, c(s), against the sedimentation coefficient (S) in Svedberg units shows the distribution of sedimentation coefficients for single-stranded AAV5hSMN vectors (AAV5SMN) containing a cap5 sequence produced by a helper plasmid having an endogenous p5 promoter ("WT Rep" Figure 11D) or a p5 promoter downstream of the cap5 sequence ("pHLP Rep" Figure 11C). [Figure 12] Figures 12A and 12B illustrate that Southern blot analysis correlates with AUC analysis, but misses some fragmented genomes detectable by AUC. (Figure 12A) Southern blot analysis of vector DNA from AAV5SMN preparations made with pHLP helper plasmid (lane 2) or WT Rep plasmid (lane 1). 4.6 and 2.4 kb size standards are provided as labels. (Figure 12B) Southern blot analysis of vector DNA from AAV5FIX preparations made with pHLP helper plasmid (lane 1) or WT Rep plasmid (lane 2). 4.3, 3.0 and 1.9 kb size standards are provided as labels. [Figure 13] This figure shows a map of the AAV5 factor IX vector, indicating the locations of the hFIX transgene, ITR, Rep origin, and AmpR marker gene, along with other characteristics. Note that the AmpR marker is located upstream of the ITR, enhancer, and promoter regions. [Figure 14]Figures 14A and 14B show that the WT Rep vector genome packages the sequence upstream of the 5'ITR in the AAV5 factor IX vector, unlike the pHLP Rep vector genome. (Figure 14A) Southern blot analysis using an hFIX transgene-specific probe comparing the pHLP Rep (lane 1) and WT Rep (lane 2) vector genomes. (Figure 14B) Southern blot analysis using a Rep ori / AmpR-specific probe comparing the pHLP Rep (lane 1) and WT Rep (lane 2) vector genomes. [Figure 15] Figures 15A and 15B show the fragmentation of large AAV vector genomes as demonstrated by AUC analysis. (Figure 15A) Plot of concentration, C(S) against sedimentation coefficient (S) created by AUC for AAV vectors with large genomes. This genome contains a full-length chicken β-actin (CBA) promoter that drives the expression of β-phosphate diesterase (ssAAV2 / 5CBA-βPDE). Peaks for detected species are indicated by the observed sedimentation coefficient (S) and relative abundance (%). (Figure 15B) Plot of concentration, C(S) against sedimentation coefficient (S) created by AUC for AAV vectors containing fragmented genomes. This genome contains a CBA promoter with reduced-size introns that drives the expression of β-phosphate diesterase (AAV5 minCBAPDE6B). Peaks for detected species are indicated by the observed sedimentation coefficient (S) and relative abundance (%). [Figure 16] This figure shows the AUC profile of a pure preparation of adenovirus capsid. The sedimentation coefficient (S) and interference values ​​are given for each peak. [Modes for carrying out the invention]

[0032] The present invention provides a method for characterizing a preparation of viral particles using analytical ultracentrifugation. By subjecting the preparation to analytical ultracentrifugation (AUC) under boundary sedimentation velocity conditions, the sedimentation of viral particles is monitored at time intervals (e.g., once or more). The differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units is then plotted, and the area under each peak in the C(S) distribution is integrated to determine the relative concentration of each peak. Each peak corresponds to the species of viral particle reflecting its molecular weight. Species that can be detected by these methods include, but are not limited to, recombinant adeno-associated virus (rAAV) particles, recombinant adenovirus (rAd) particles, recombinant lentivirus particles, and recombinant herpes simplex virus (rHSV) particles. Using rAAV particles as an illustrative example, these methods enable the detection of rAAV species, including rAAV capsid particles containing intact rAAV genomes (e.g., complete capsids), empty viral capsids where the rAAV genome is not capsidized, and rAAV particle variants where a variant rAAV genome is capsidized (e.g., particles containing AAV capsidized DNA impurities, fragmented viral genomes, aggregates, etc.). These methods can be applied to the preparation of viral particles regardless of the nucleotide sequence of the viral genome or, in the case of recombinant viral particles, the serotype of the recombinant viral capsid. These methods can be applied to rAAV, rAd, recombinant lentivirus, and rHSV viral particles.

[0033] In some embodiments, the present invention provides a method for evaluating the integrity of the vector genome of recombinant virus particles in a recombinant virus particle preparation by subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of the recombinant virus particles is monitored at time intervals (e.g., once or more times). The species of recombinant virus particles in the preparation can be identified by plotting the differential sedimentation coefficient distribution value C(S) against the sedimentation coefficient (S) in Svedberg units, by the presence of a peak on the plot corresponding to the S value. The genome size of a specific species of recombinant virus particle can be calculated, for example, by comparing the S value of the species with a standard curve created by the S values ​​of recombinant virus particles containing capsidized viral genomes of various known sizes. The vector genomes that can be evaluated by these methods include, but are not limited to, recombinant viral capsid particles containing intact recombinant viral genomes (e.g., complete capsids), empty viral capsids in which recombinant viral genomes have not been capsidized, and recombinant viral particle variants in which variant recombinant viral genomes (e.g., particles containing AAV capsidized DNA impurities, fragmented viral genomes, aggregates, etc.) have been capsidized. In some embodiments, the viral particles are rAAV, rAd, recombinant lentivirus, or rHSV viral particles.

[0034] In some embodiments, the present invention provides a method for determining heterogeneity of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) in a preparation of recombinant virus particles by AUC under boundary sedimentation velocity conditions, wherein the presence of a peak in addition to the peak corresponding to the capsid containing the intact viral genome in a C(S) vs. S plot indicates heterogeneity of recombinant virus particles in the preparation. In some embodiments, the relative amount of each recombinant virus species in the preparation is calculated by integrating the area of ​​each peak in the plot.

[0035] In some embodiments of the present invention, AUC is used to determine the presence of empty capsids and / or recombinant virus particle variants in a preparation of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), and the presence of a peak in the C(S) vs. S plot corresponding to the S value of empty capsid particles and / or recombinant virus particle variants indicates the presence of empty capsid particles and / or recombinant virus particle variants. In some embodiments, the relative amounts of empty capsids and / or recombinant virus particle variants in a recombinant virus particle preparation are determined by integrating the area under each peak in a plot of C(S) against S, and by comparing the amount of recombinant virus particles with S values ​​corresponding to empty capsid particles and / or recombinant virus particle variants with the amount of recombinant virus particles with S values ​​corresponding to recombinant virus particles containing intact viral genomes. In some embodiments, the amount of recombinant virus particles with S values ​​corresponding to empty capsid particles and / or recombinant virus particle variants is compared with the amount of all recombinant virus particles in the preparation by integrating and summing the areas under all peaks in the plot.

[0036] In some embodiments, the present invention provides a method for monitoring the removal of empty capsids and / or recombinant virus particle variants during the purification of recombinant virus particle preparations (e.g., rAAV, rAd, lentivirus, or rHSV particles) by using AUC. A recombinant virus particle sample derived from the preparation after one or more steps of the purification process is analyzed for the relative amount of empty capsids and / or recombinant virus particle variants, and a decrease in the relative amount of empty capsids and / or recombinant virus particle variants relative to complete capsid particles indicates the removal of empty capsids and / or recombinant virus particle variants from the recombinant virus particle preparation.

[0037] In some embodiments, the present invention provides a method for evaluating a process for the production of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) by AUC. Recombinant virus particle preparations are analyzed for the presence of intact whole viral capsid particles, empty particles, and / or recombinant virus particle variants. An increase in the relative amount of recombinant virus particles containing intact viral genomes compared to the relative amount of empty capsid particles, and / or an increase in recombinant virus particle variants (e.g., particles containing AAV capsidized DNA impurities, fragmented viral genomes, aggregates, etc.) compared to a reference preparation of recombinant virus particles (e.g., a standard recombinant virus preparation process), indicates an improvement in recombinant virus particle production.

[0038] I. General techniques For example, Molecular Cloning: A Laboratory Manual (Sambrook et al., 4th edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 2012); Current Protocols in Molecular Biology (FMAusubel et al., eds., 2003); the series Methods in Enzymology (Academic Press, Inc.); PCR 2: A Practical Approach (edited by MJMacPherson, BD Hames and GRTaylor, 1995); Antibodies, A Laboratory Manual (edited by Harlow and Lane, 1988); Culture of Animal Cells: A Manual of Basic Technique and Specialized Applications(R.I.Freshney, 6th Edition, J.Wiley and Sons, 2010); Oligonucleotide Synthesis(M.J.Gait ed., 1984); Methods in Molecular Biology, Humana Press; Cell Biology: A Laboratory Notebook(J.E.Cellis ed., Academic Press, 1998); Introduction to Cell and Tissue Culture(J.P.Mather and P.E.Roberts, Plenum Press, 1998); Cell and Tissue Culture: Laboratory Procedures(A.Doyle, J.B.Griffiths, and D.G.Newell eds., J.Wiley and Sons, 1993 - 8); Handboo k of Experimental Immunology(D.M.Weir and C.C.Blackwell eds., 1996); Gene Transfer Vectors for Mammalian Cells(J.M.Miller and M.P.Calos eds., 1987); PCR: The Polymerase Chain Reaction, (Mullis et al. eds., 1994); Current Protocols in Immunology(J.E.Coligan et al. eds., 1991); Short Protocols in Molecular Biology(Ausubel et al. eds., J.Wiley and Sons, 2002); Immunobiology(C.A.Janeway et al., 2004); Antibodies(P.Finch, 1997); Antibodies: A Practical Approach(D.Catty. ed., IRL Press, 1988 - 1989); Monoclonal The techniques and procedures described and referenced herein, including the widely used methods described in Antibodies: A Practical Approach (edited by P. Shepherd and C. Dean, Oxford University Press, 2000); Using Antibodies: A Laboratory Manual (E. Harlow and D. Lane, Cold Spring Harbor Laboratory Press, 1999); The Antibodies (edited by M. Zanetti and JDCapra, Harwood Academic Publishers, 1995) and Cancer: Principles and Practice of Oncology (edited by V. DeVita et al., JBLippincott Company, 2011), are generally well understood and commonly used by those skilled in the art using conventional methods.

[0039] II. Definition As used herein, “vector” refers to a recombinant plasmid or virus containing nucleic acid that is delivered to a host cell either in vitro or in vivo.

[0040] As used herein, the terms “polynucleotide” or “nucleic acid” refer to polymeric forms of nucleotides of any length, whether ribonucleotides or deoxyribonucleic acids. Thus, the term includes, but is not limited to, single, double, or multi-stranded DNA or RNA, genomic DNA, cDNA, DNA-RNA hybrids, or polymers containing purine and pyrimidine bases, or other naturally occurring, chemically or biochemically modified, unnatural, or derived nucleotide bases. The polynucleotide backbone contains sugars and phosphate groups (typically as found in RNA or DNA) or modified or substituted sugars or phosphate groups. Alternatively, the polynucleotide backbone may contain polymers of synthetic subunits such as phosphoramidates, which may be oligodeoxynucleoside phosphoramidates (P-NH2) or mixed phosphoramidate-phosphate diester oligomers. Additionally, double-stranded polynucleotides can be obtained from chemosynthetic single-stranded polynucleotide products by either synthesizing a complementary strand and annealing the strands under appropriate conditions, or by synthesizing the complementary strand in situ using DNA polymerase with appropriate primers.

[0041] The terms “polypeptide” and “protein” are used interchangeably to refer to polymers of amino acid residues and are not limited to the shortest length. Such polymers of amino acid residues may contain, but are not limited to, native amino acid residues, peptides, oligopeptides, dimers, trimers, and polymers of amino acid residues. Both full-length proteins and their fragments are included in the definition. The term also includes post-expression modifications of polypeptides, such as glycosylation, sialylation, acetylation, and phosphorylation. Furthermore, for the purposes of this invention, “polypeptide” refers to proteins that include modifications such as deletions, additions, and substitutions (generally conserved in nature) to the native sequence, as long as the protein maintains the desired activity. Modification may be intentional through site-directed mutagenesis, or accidental through mutations in the protein-producing host or errors during PCR amplification.

[0042] A "recombinant viral vector" refers to a recombinant polynucleotide vector containing one or more heterogeneous sequences (i.e., non-viral nucleic acid sequences). In the case of a recombinant AAV vector, the recombinant nucleic acid has at least one terminal inversion sequence (ITR) at one end. In some embodiments, the recombinant nucleic acid has two terminal inversion sequences (ITRs) at one end.

[0043] A "recombinant AAV vector (recombinant adeno-associated virus vector)" refers to a polynucleotide vector containing one or more heterogeneous sequences (i.e., nucleic acid sequences not derived from AAV) having at least one AAV terminal inversion sequence (ITR) at its end. In some embodiments, the recombinant nucleic acid has two terminal inversion sequences (ITRs) at its end. Such a recombinant viral vector, when present in a host cell infected with a suitable helper virus (or expressing a suitable helper function) and expressing AAV rep and cap gene products (i.e., AAV Rep and Cap proteins), is replicated and packaged into infectious viral particles. When a recombinant viral vector is incorporated into a large polynucleotide (e.g., in a chromosome or another vector such as a plasmid used for cloning or transfection), the recombinant viral vector may be referred to as a "provector" that is "rescued" by replication and capsidation in the presence of AAV packaging function and suitable helper function. Recombinant viral vectors may be any number of forms, including, but are not limited to, plasmids, linear artificial chromosomes, lipid-containing complexes, capsidized within liposomes, and capsidized within viral particles, such as AAV particles. Recombinant viral vectors are packaged into AAV viral capsids that produce "recombinant adeno-associated virus particles (recombinant viral particles)."

[0044] "rAAV virus" or "rAAV virus particle" refers to a viral particle consisting of at least one AAV capsid protein and a capsidized rAAV vector genome.

[0045] A "recombinant adenovirus vector" refers to a polynucleotide vector containing one or more heterogeneous sequences (i.e., non-adenovirus-derived nucleic acid sequences) having at least one adenovirus terminal inversion sequence (ITR) at its ends. In some embodiments, the recombinant nucleic acid has two terminal inversion sequences (ITRs) at its ends. Such a recombinant viral vector, when present in a host cell expressing key adenovirus genes deleted from the recombinant viral genome (e.g., E1, E2, E4 genes, etc.), is replicated and packaged into infectious viral particles. When a recombinant viral vector is incorporated into a large polynucleotide (e.g., in a chromosome or another vector such as a plasmid used for cloning or transfection), the recombinant viral vector may be referred to as a "provector" that is "rescued" by replication and capsidation in the presence of adenovirus packaging capabilities. Recombinant viral vectors may be any number of forms, including, but are not limited to, plasmids, linear artificial chromosomes, lipid-containing complexes, capsidized within liposomes, and capsidized within viral particles, such as adenovirus particles. Recombinant viral vectors are packaged into adenovirus viral capsids that produce "recombinant adenovirus particles."

[0046] A "recombinant lentiviral vector" refers to a polynucleotide vector containing one or more heterogeneous sequences (i.e., non-lentiviral nucleic acid sequences) having at least one lentiviral terminal repeat sequence (LTR) at its end. In some embodiments, recombinant The nucleic acid has two lentiviral terminal repeat sequences (LTRs) at its ends. Such recombinant viral vectors replicate and package into infectious viral particles when present in infected host cells with suitable helper functions. The recombinant lentiviral vector is packaged into a lentiviral capsid that produces a "recombinant lentiviral particle".

[0047] A "recombinant herpes simplex vector (recombinant HSV vector)" refers to a polynucleotide vector containing one or more heterogeneous sequences (i.e., nucleic acid sequences not derived from HSV) having an HSV terminal repeat sequence at its end. Such recombinant viral vectors are replicated and packaged into infectious viral particles if suitable helper functions are present in infected host cells. When a recombinant viral vector is incorporated into a large polynucleotide (e.g., in a chromosome or another vector such as a plasmid used for cloning or transfection), the recombinant viral vector may be referred to as a "provector" that is "rescued" by replication and capsidation in the presence of HSV packaging functions. Recombinant viral vectors may be any number of forms, including, but are not limited to, plasmids, linear artificial chromosomes, lipid-containing complexes, capsidated within liposomes, and capsidated within viral particles, such as HSV particles. Recombinant viral vectors are packaged into HSV capsids that produce "recombinant herpes simplex virus particles."

[0048] "Heterogeneity" means that it originates from an entity whose genotype differs from other parts of the entity being compared to, or into which it is introduced or incorporated. For example, a polynucleotide introduced into a different cell type by genetic engineering is a heterogeneous polynucleotide (and can encode a heterogeneous polypeptide if expressed). Similarly, a cellular sequence (e.g., a gene or a portion thereof) incorporated into a viral vector is a heterogeneous nucleotide sequence with respect to the vector.

[0049] The term "transgene" refers to a polynucleotide that is introduced into a cell, transcribed into RNA, and, if applicable, translated and / or expressed under favorable conditions. In one embodiment, it confers a desirable characteristic to the cell into which it is introduced, or leads to a desired therapeutic or diagnostic outcome. In another embodiment, it is transcribed into a molecule that mediates RNA interference, such as siRNA.

[0050] The terms “genome particle (gp),” “genome equivalent,” or “genome copy,” used in reference to viral titer, refer to the number of virions containing recombinant viral DNA or RNA genomes, regardless of infectivity or functionality. The number of genome particles in a specific vector preparation is given in the examples herein or, for example, Clark et al., (1999) Hum. Gene It can be measured by procedures such as those described in Ther., 10: pp. 1031-1039; Veldwijk et al., (2002) Mol.Ther., 6: pp. 272-278.

[0051] As used herein with reference to viral titer, the terms “infectious unit (iu),” “infectious particle,” or “replication unit” refer to the number of infectious and replication-eligible recombinant viral vector particles, measured by an infectious center assay, also known as a replication center assay, as described by McLaughlin et al., (1988) J. Virol., pp. 62:1963–1973, for example, AAV.

[0052] The term “introduced unit (tu)” used in reference to viral titer refers to the number of infectious recombinant viral vector particles that produce a functional transgene product, as measured by functional assays such as those described in the examples herein or, for example, Xiao et al., (1997) Exp. Neurobiol., pp. 144:113-124 or Fisher et al., (1996) J. Virol., pp. 70:520-532 (LFU assay) for AAV. It refers to.

[0053] A "terminal inversion sequence" or "ITR" sequence is a well-understood term in the art, referring to a relatively short sequence found at the end of a viral genome that is oriented in the opposite direction.

[0054] A well-understood term in this art, the “AAV terminal inversion sequence (ITR)” is a sequence of approximately 145 nucleotides present at both ends of a natural single-stranded AAV genome. The outermost 125 nucleotides of the ITR may exist in two different orientations, resulting in heterogeneity between different AAV genomes and between two ends of a single AAV genome. The outermost 125 nucleotides also contain several self-complementary short regions (denoted as regions A, A', B, B', C, C', and D), which enable intra-strand base pairing to occur within this portion of the ITR.

[0055] The "terminal resolution sequence" or "trs" is a sequence within the D region of the AAV ITR that is cleaved by the AAV rep protein during viral DNA replication. Mutant terminal resolution sequences are resistant to cleavage by the AAV rep protein.

[0056] "AAV helper function" refers to a function that enables host cells to replicate and package AAV. AAV helper functions are provided in any number of forms, including, but not limited to, helper viruses or helper virus genes that assist in AAV replication and packaging. Other AAV helper functions are well known in the art, such as genotoxic agents.

[0057] A “helper virus” for AAV refers to a virus that enables AAV (deficient parvovirus) to replicate and package by a host cell. Helper viruses provide a “helper function” that allows AAV replication. Numerous such helper viruses have been identified and include adenoviruses, herpesviruses, poxviruses such as vaccinia, and baculoviruses. Adenoviruses include numerous different subgroups, but adenovirus type 5 (Ad5) of subgroup C is the most commonly used. Numerous adenoviruses of human, non-human mammalian, and avian origin are well known and available from contractors such as ATCC. Viruses of the herpes family, also available from contractors such as ATCC, include, for example, herpes simplex virus (HSV), Epstein-Barr virus (EBV), cytomegalovirus (CMV), and pseudorabies virus (PRV). Examples of adenovirus helper functions for AAV replication include E1A, E1B, E2A, VA, and E4orf6 functions. Baculoviruses available from the contractor include Autographa californica nuclear polyhedrosis. (Contains virus)

[0058] The rAAV preparation has a ratio of infectious AAV particles to infectious helper virus particles of at least approximately 10 2 :1, at least about 10 4 :1, at least about 10 6 :1 or at least about 10 8If the value is 1 or greater, the preparation is considered "substantially free" of helper viruses. In some embodiments, the preparation also does not contain an equal amount of helper virus protein (i.e., the protein present as a result of such a level of helper virus if the helper virus particle impurities described above were present in a disrupted form). Viral and / or cellular protein contamination is generally observed as the presence of Coomassie staining bands on the SDS gel (e.g., the appearance of bands other than those corresponding to AAV capsid proteins VP1, VP2, and VP3).

[0059] As used herein, "differential coefficient distribution value" or "C(S)" refers to a variable in the distribution of the Lamb equation solution used to describe the distribution of settled particles, for example, during ultracentrifugation.

[0060] As used herein, the “Svedberg unit” refers to a unit of sedimentation velocity. The sedimentation velocity of a particle of a given size and shape measures how quickly the particle settles. One Svedberg unit is equal to 10⁻¹⁰ -13 Equivalent to seconds. For example, the Svedberg unit is often used to reflect the speed at which molecules move under the centrifugal force of centrifugation.

[0061] As used herein, “sedimentation velocity conditions” or “boundary sedimentation velocity conditions” refer to any experimental conditions under which a sample solution is subjected to sedimentation velocity analysis. Sedimentation velocity allows for the study of particles over a wide range of pH and ionic strength conditions and temperatures from 4 to 40°C. The speed at which the sedimentation boundary moves is a measure of the sedimentation coefficient of the sedimenting species. The sedimentation coefficient also depends on molecular weight (larger particles sediment faster) and molecular shape. The minimum width of the sedimentation boundary is related to the diffusion coefficient of the molecules; the presence of multiple species with similar sedimentation coefficients results in a boundary wider than would be predicted based on diffusion alone. Sedimentation velocity conditions may include, but are not limited to, any conditions related to rotor speed, distance between the sample and the rotor center, temperature, solvent, sample, buffer, ultracentrifugation time, detection time interval, sector and optical window characteristics, AUC measurement means (including ultracentrifuge and detector), equilibrium dialysis of the reference solvent, and data analysis algorithms.

[0062] As used herein, the term “analytical density gradient sedimentation equilibrium” refers to methods for measuring the suspended density of particles or using differences in suspended density to separate different species of particles. These methods can, for example, utilize AUC sedimentation equilibrium techniques. In these methods, a particle solution (e.g., a solution of polypeptides, polynucleotides, or viral capsids, not limited to) is subjected to ultracentrifugation in a solvating compound gradient, such as a cesium chloride or cesium sulfate gradient, until equilibrium with the solvating compound is achieved. At equilibrium, the particle solution collects at a position in the gradient where the density of the particles is equal to that of the solvating compound, or forms a band. The position of the band can be used to calculate the particle density, and the band can be extracted to isolate a single species of particle.

[0063] The “SEDFIT algorithm” used herein is an algorithm that enables the analysis of hydraulic data such as sedimentation velocity (Schuck (2000) Biophys.J., 78: pp. 1606-19). In the SEDFIT algorithm, a grid of sedimentation coefficients is created over the predicted range. The sedimentation boundary is simulated for each sedimentation coefficient using a solution to the Lamb equation, and a constant particle shape and solvent friction ratio are estimated.

[0064] In this specification, the terms "F-statistic" or "F-ratio" refer to the confidence level. This parameter controls the amount of regularization used. It has different meanings within different ranges: no regularization is used between 0 and 0.5. Values ​​between 0.5 and 0.999 correspond to probability P (confidence level). The desired chi-squared increase from these P values ​​is calculated using the F-statistic, which is the parsimony constraint on regularization. A value of 0.51 results in almost no regularization; values ​​between 0.68 and 0.90 correspond to commonly used confidence levels (typically, a chi-squared increase corresponding to a probability of 0.7 after 50 scans is around 0.1%), while values ​​close to 0.99 result in very high regularization. The relationship between these values ​​and probability is examined using the F-statistic calculator. When a number > 1 is entered, they are considered directly as the chi-squared ratio (since there is no probability > 1). For example, a value of 1.1 results in regularization with a 10% chi-squared increase.

[0065] "To reduce" means to decrease, diminish, or suppress the activity, function, and / or quantity compared to a reference. In certain embodiments, "to reduce" means the ability to produce a reduction of 20% or more overall. In another embodiment, "to reduce" means the ability to produce a reduction of 50% or more overall. In yet another embodiment, "to reduce" means the ability to produce a reduction of 75%, 85%, 90%, 95% or more.

[0066] As used herein, “reference” refers to any sample, standard, or level used for comparative purposes. For example, when measuring the absorbance or refraction of AAV in an aqueous solution, the absorbance or refraction of the solution is compared to the absorbance or refraction of an aqueous solution without AAV (i.e., a reference solution). In other examples, a reference may refer to a standard procedure well known in the art. For example, when analyzing a procedure for improving the quality (e.g., uniformity) of AAV production, the AAV produced by a candidate procedure is compared to a procedure well known in the art (i.e., a reference procedure).

[0067] The term "isolated" means that a molecule (e.g., nucleic acid or protein) or cell has been identified, separated, and / or recovered from its natural environment. Therefore, for example, isolated rAAV particles are produced using purification techniques to concentrate them from a source mixture such as a culture lysate or a productive culture supernatant. Concentration is measured in various ways, such as by the proportion of DNase-resistant particles (DRPs) present in the solution or by infectivity, or it is measured in relation to second potential interfering substances present in the source mixture, such as productive culture contaminants or process contaminants including helper viruses, culture medium components, etc.

[0068] In this specification, a reference to “about” a value or parameter includes (and describes) embodiments that apply to the value or parameter itself. For example, a statement referring to “about X” includes a statement of “X”.

[0069] As used herein, the singular forms of items, "a," "an," and "the," include plural references unless otherwise indicated. For example, the phrase "a rAAV particle" includes one or more rAAV particles.

[0070] It is understood that the aspects and embodiments of the present invention described herein include "including," "consisting of," and / or "essentially consisting of" the aspects and embodiments.

[0071] III. Analytical Ultracentrifugation Method Analytical ultracentrifugation is a method for evaluating the molecular weight, as well as the hydrodynamic and thermodynamic properties, of proteins or other macromolecules. It reveals heterogeneity of proteins or macromolecules due to sedimentation rates across a wide range of conditions, including concentration, temperature, ionic strength, and pH. For example, proteins are analyzed in clinically relevant formulations. The use of analytical ultracentrifugation to characterize adenovirus preparations is provided by Berkowitz, SA & Philo JS, (2007) Anal. Biochem., 362: pp. 16-37.

[0072] In certain embodiments, the present invention provides a method for characterizing preparations of viral particles using analytical ultracentrifugation (AUC). For example, in some embodiments, the present invention provides a method for evaluating the integrity of the vector genome of recombinant adeno-associated virus (rAAV) particles in preparations of rAAV particles using AUC to identify viral particles having a complete, intact genome, empty viral capsids, and variant (e.g., fragmented, aggregated, impurity-containing, etc.) viral genomes. In other embodiments, these methods are used for adeno Similar methods are applied to analyze viral, lentivirus, and herpes simplex virus (HSV) particles. AUC analysis refers to a quantitative method for characterizing the biophysical properties of particles (e.g., polypeptides, polynucleotides, and viral capsids) by measuring their migration through a solvent in a centrifugal field. AUC analysis has been well-characterized for decades and is highly versatile. Because AUC analysis relies on first-principles hydraulic and thermodynamic information, AUC can be applied to determine the biophysical properties of many types of particles over a wide range of particle concentrations and sizes. AUC analysis typically encompasses two basic types of experiments: sedimentation velocity and sedimentation equilibrium. Sedimentation equilibrium analysis yields thermodynamic properties of particles that can be used to measure features such as stoichiometry and association constants. Sedimentation velocity yields hydrodynamic properties of particles that can be used to measure features such as size, shape, and concentration. A characteristic of AUC analysis of viral preparations is that the same assay conditions can be used to analyze various preparations of viral particles, regardless of the nucleotide sequence of the viral genome or the serotype of the capsid.

[0073] Certain aspects of this disclosure relate to the use of sedimentation velocity analysis to characterize viral capsid properties. In some embodiments, sedimentation velocity analysis uses an ultracentrifugal cell having two sectors in dialysis equilibrium (one for the experimental sample and the other for a reference sample of solvent only), each having two optical windows that allow light to pass through the compartment. Ultracentrifugation applies angular velocity to the cell, leading to the rapid sedimentation of solute particles to the bottom of the sector. As sedimentation occurs, the solute is depleted near the meniscus at the top of the cell, creating a sedimentation boundary between the depleted region and the settled solute. The rate of movement or transition of the sedimentation boundary is measured by taking measurements comparing the properties of the sample and the reference sector at specific time intervals (for sedimentation velocity, these intervals are typically several minutes). If multiple types of solute are present, this results in the formation of multiple sedimentation boundaries, each corresponding to a degradable species.

[0074] Several methods for optically detecting sedimentation boundaries and measuring their migration or transfer rate are well known in the art (see, for reference, Cole et al., (2008) Methods Cell Biol., 84: pp. 143-79). In some embodiments, the reference and sample sectors are assayed using absorbance detection. In this detection method, absorbance at specific wavelengths is measured for the sample and reference sectors at different radial positions within each sector. Alternatively, the change in absorbance over time at a single radial position is measured. Beer's law provides a mathematical relationship between absorbance and the extinction coefficient of the solute.

[0075] In some embodiments, the reference and sample sectors are assayed using interference detection (e.g., Rayleigh interference detection). In Rayleigh interference detection, the interference optical system has two parallel slits. A single coherent beam of light is split to pass through both windows, and then the two beams are remixed. When these two light waves are mixed, they form an interference pattern of alternating light and dark fringes. If the sample and reference sample have the same refractive index, the resulting interference fringes will be perfectly linear. As the concentration of the solute increases, the refractive index of the solution increases, thereby delaying the sample light beam and causing a vertical fringe shift. By measuring this fringe shift, the concentration of the solute in the sample can be measured. Unlike absorbance detection, which measures absolute values ​​for the sample and reference, interference detection measures the relative difference between the sample and the reference. However, interference detection is used for types of samples that do not absorb significantly and produce an integrated peak directly proportional to the concentration. For reference on using Rayleigh interferometry in AUC, see Furst (1997) Eur. Biophys. J. 35: pp. 307-3010.

[0076] Measuring the velocity at which the sedimentation boundary moves is used to derive numerous physical properties of solute particles. The velocity of boundary movement determines the sedimentation coefficient, which is based on the particle's mass and shape (coefficient of friction). The settling coefficient s of a particle refers to the ratio of its velocity to the acceleration applied to it by the centrifugal force field. The settling coefficient is expressed in Svedberg units, S (1 Svedberg unit = 10⁻¹⁰ -13 It is expressed as (equal to seconds). The settling coefficient of a particle or a solution of particles depends on its properties, such as molecular weight (corrected for buoyancy) and the properties of the solvent.

[0077] The change at the solute concentration boundary over time during ultracentrifugation can be determined using the Lamb equation (Schuck (2000) Biophys.J., 78: pp. 1606-19). Briefly, the Lamb equation calculates the change at the solute concentration boundary over time in response to competing forces of sedimentation (concentrating the solute) and diffusion (dispersing the solute), taking into account the sector-shaped cellular and centrifugal fields generated by the rotor. The Lamb equation is: Equation 1: ∂c / ∂t=D[(∂^2c / ∂r^2)+1 / r(∂c / ∂r)]-sω^2[r(∂c / ∂r)+2c] It can be expressed as follows: In the formula, c represents the solute concentration, D represents the solute diffusion constant, s represents the settling coefficient, ω represents the rotor angular velocity, r is the radius, and t is time.

[0078] By fitting raw AUC data to a solution of the Lamb equation, solution characteristics such as the settling coefficient and changes in concentration distribution can be determined. For example, experimentally determined values ​​for the rate of change of the settling boundary can be modeled using the Lamb equation to derive the settling coefficient, molecular weight, or concentration of the solute forming the boundary. Several programs, such as SEDFIT (Schuck (2000) Biophys.J., 78: pp. 1606-19), are well known in the art and are used to model the Lamb equation onto AUC data. These programs also allow the application of the Lamb equation to solutions containing multiple solutes or multiple settling boundaries.

[0079] An example of a suitable program for determining the characteristics of a solute is the SEDFIT algorithm. In some embodiments, the SEDFIT algorithm is used to calculate the differential coefficient distribution value or C(S) using AUC data from a solution containing a mixture of particle species (see Schuck (2000) Biophys. J., 78:1606-19 for reference). In the SEDFIT algorithm, a grid of sedimentation coefficients is created over a predicted range. The sedimentation boundary is simulated for each sedimentation coefficient using the solution to the Lamm equation, and a constant particle shape and solvent friction ratio are estimated. The actual AUC data is then fit to these Lamm solutions to derive the differential coefficient distribution value or C(S). A number of other programs useful for analyzing AUC data can be found in Cole and Hansen (1999) J. Biomol. Tech. 10:163-76.

[0080] In some embodiments of the invention, the recombinant virus particles are highly purified, appropriately buffered, and concentrated. In some embodiments, the virus particles are at least about 1x10 7 vg / mL, 2x10 7 vg / mL, 3x10 7 vg / mL, 4x10 7 vg / mL, 5x10 7 vg / mL, 6x10 7 vg / mL, 7x10 7 vg / mL, 8x10 7 vg / mL, 9x10 7 vg / mL, 1x10 8 vg / mL, 2x10 8 vg / mL, 3x10 8 vg / mL, 4x10 8 vg / mL, 5x10 8 vg / mL, 6x10 8 vg / mL, 7x10 8 vg / mL, 8x10 8 vg / mL, 9x10 8 vg / mL, 1x10 9 vg / mL, 2x10 9 vg / mL, 3x10 9 vg / mL, 4x109 vg / mL、5x10 9 vg / mL、6x10 9 vg / mL、7x10 9 vg / mL、8x10 9 vg / mL、9x10 9 vg / mL、1x10 10 vg / mL、2x10 10 vg / mL、3x10 10 vg / mL、4x10 10 vg / mL、5x10 10 vg / mL、6x10 10 vg / mL、7x10 10 vg / mL、8x10 10 vg / mL、9x10 10 vg / mL、1x10 11 vg / mL、2x10 11 vg / mL、3x10 11 vg / mL、4x10 11 vg / mL、5x10 11 vg / mL、6x10 11 vg / mL、7x10 11 vg / mL、8 x10 11 vg / mL、9x10 11 vg / mL、1x10 12 vg / mL、2x10 12 vg / mL、3x10 12 vg / mL、4x10 12 vg / mL、5x10 12 vg / mL、6x10 12 vg / mL、7x10 12 vg / mL、8x10 12 vg / mL、9x10 12 vg / mL、1x10 13 vg / mL、2x10 13 vg / mL、3x10 13 vg / mL、4x10 13 vg / mL、5x10 13 vg / mL、6x10 13 vg / mL、7x10 13 vg / mL、8x10 13 vg / mL、9x10 13It is concentrated to any of vg / mL. In some embodiments, the virus particles are from about 1x10 7 vg / mL to about 1x10 13 vg / mL, from about 1x10 8 vg / mL to about 1x10 13 vg / mL, from about 1x10 9 vg / mL to about 1x10 13 vg / mL, from about 1x10 10 vg / mL to about 1x10 13 vg / mL, from about 1x10 11 vg / mL to about 1x10 13 vg / mL, from about 1x10 12 vg / mL to about 1x10 13 vg / mL, from about 1x10 7 vg / mL to about 1x10 12 vg / mL, from about 1x10 8 vg / mL to about 1x10 12 vg / mL, from about 1x10 9 vg / mL to about 1x10 12 vg / mL, from about 1x10 10 vg / mL to about 1x10 12 vg / mL, from about 1x10 11 vg / mL to about 1x10 12 vg / mL, from about 1x10 7 vg / mL to about 1x10 11 vg / mL, from about 1x10 8 vg / mL to about 1x10 11 vg / mL, from about 1x10 9 vg / mL to about 1x10 11 vg / mL, from about 1x10 10 vg / mL to about 1x10 11 vg / mL, from about 1x10 7 vg / mL to about 1x10 10 vg / mL, from about 1x10 8 vg / mL to about 1x10 10 vg / mL, from about 1x10 9 vg / mL to about 1x10 10 vg / mL, from about 1x10 7 vg / mL to about 1x10 9 vg / mL, from about 1x10 8 vg / mL to about 1x109 vg / mL or approximately 1x10 7 vg / mL to approximately 1x10 8 It is concentrated to vg / mL.

[0081] In some embodiments, viral particles are generated and purified in suitable host cells. In some embodiments, viral particles are purified by affinity chromatography. Methods for purifying viral particles (e.g., AAV particles, adenovirus particles, lentivirus particles, HSV particles) are well known in the art. For example, by using antibodies or binding ligands of viral capsid proteins immobilized on a chromatographic medium. Examples of viral capsid affinity chromatography include, but are not limited to, AVB affinity chromatography for AAV (GE Healthcare), metal affinity chromatography for adenoviruses and HSV, and heparin affinity chromatography for AAV and lentiviruses. Methods for purifying adenovirus particles can be found, for example, in Bo, H et al., (2014) Eur. J. Pharm. Sci. 67C: pp. 119-125. Methods for purifying lentivirus particles can be found, for example, in Segura MM, et al., (2013) Expert Methods for purifying HSV particles can be found, for example, in Goins, WF et al., (2014) Herpes Simplex Virus Methods in Molecular Biology 1144: pp. 63-79.

[0082] In some embodiments, recombinant virus particles are formulated in a pharmaceutical composition. In relevant embodiments, the pharmaceutical composition contains a buffer having a physiological pH and / or physiological osmotic pressure. A non-limiting example of the pharmaceutical formulation is phosphate-buffered saline (PBS), and in some embodiments, the PBS has a physiological osmotic pressure (e.g., about pH 7.2 and about 300 mOsm / L). In some embodiments, sample preparation is performed to a target concentration of 0.1 to 1.0 by optical density measurement at 260 nm. In some embodiments, this concentration yields reproducible and consistent AUC data. In some embodiments, the concentration of virus particles is adjusted either by direct dilution in PBS or by further concentration; for example, by using a centrifugal filter device.

[0083] In some embodiments of the present invention, sedimentation velocity analysis by ultracentrifugation (SV-AUC) is performed. The analysis is carried out using an analytical ultracentrifuge capable of characterizing the sample in its natural state under physically appropriate solution conditions (e.g., ProteomeLab® XL-I (Beckman Coulter)). When using ProteomeLab® XL-1, the sample is loaded into the sample sector of two sector velocity cells, and a vehicle control (e.g., recombinant virus-free PBS) is loaded into the corresponding reference sector. The sample is placed in a four-well rotor and equilibrated in the instrument to a temperature of approximately 20°C, with a high vacuum maintained for approximately 1 hour. In exemplary embodiments, sedimentation velocity centrifugation is carried out without delay and repetition at approximately 20,000 rpm, approximately 20°C, and a process setting of approximately 0.003 cm radius. Various parameters can be used for centrifugation, as described below. In some embodiments, absorbance (260 nm) and / or interference optics (e.g., Rayleigh interference optics) are used to simultaneously record the radial concentration as a function of time until the minimum settling component clears the optical window. In some embodiments, radial concentration is recorded until the lowest density settling species clears the sector. In some embodiments, sedimentation is monitored until the lowest density recombinant virus particles settle to the bottom of the ultracentrifuge sector. The sector is a part of the ultracentrifuge; for example, an ultracentrifuge velocity cell. In some embodiments, the sector may be a part of the ultracentrifuge from which the sample is detected. In some embodiments, the ultracentrifugation method utilizes an ultracentrifuge that includes an ultracentrifuge velocity cell. In some embodiments, it is monitored until recombinant virus particles settle to the bottom of the ultracentrifuge velocity cell. In some embodiments, sedimentation is monitored until the lowest density recombinant virus particles settle and clear the optical window. In some embodiments, radial concentration is recorded for at least about 0.5 hours, 0.75 hours, 1.0 hour, 1.5 hours, 2.0 hours, 3.0 hours, 4.0 hours, or 5.0 hours. In some embodiments, radial concentrations are recorded over one of the following time periods: approximately 0.5 hours to approximately 0.75 hours, approximately 0.75 hours to approximately 1.0 hours, approximately 1.0 hours to approximately 1.5 hours, approximately 1.5 hours to approximately 2.0 hours, approximately 2 hours to approximately 3 hours, approximately 3 hours to approximately 4 hours, or approximately 4 hours to approximately 5 hours.In some embodiments, radial concentration is recorded for approximately 1.2 hours. Optimizing the running conditions includes, for example, maintaining a constant temperature of 20°C and continuing the run at a speed of 18,000 to 20,000 rpm until all settling species have completely settled to the bottom of the sector. Other temperatures and speeds may also be used, as described below.

[0084] The percentage of complete capsids is determined by analyzing multiple scans (e.g., 75) from each detection method using the SEDFIT continuous-size C(S) distribution model. nd Differential regularization is applied to the fitting. In some embodiments, the confidence level of the F-statistic is approximately 0.68. In some embodiments, the confidence level of the F-statistic is higher than approximately 0.68, 0.70, 0.75, 0.80, 0.85, 0.90, 0.95, or 0.99. In some embodiments, the confidence level of the F-statistic is from approximately 0.68 to approximately 0.90. In some embodiments, the confidence level of the F-statistic is from approximately 0.68 to approximately 0.99. In some embodiments, the following C(S) parameters are kept constant: resolution is approximately 200S to approximately 5000S, Smin is approximately 1S to approximately 100S, Smax is approximately 100S to approximately 5000S, and the friction ratio is approximately 1.0, or transitions to a value determined by the centrifugal separation software. In some embodiments, the resolution is one of approximately 200S, 300S, 400S, 500S, 600S, 700S, 800S, 900S, or 1000S. In some embodiments, the resolution is approximately 200S to approximately 1000S, 200S to approximately 900S, 200S to approximately 800S, 200S to approximately 700S, 200S to approximately 600S, 200S to approximately 500S, 200S to approximately 400S, 200S to approximately 300S, 300S to approximately 1000S, 300S to approximately 900S, 300S to approximately 800S, 300S to approximately 700S. S, 300S to approximately 600S, 300S to approximately 500S, 300S to approximately 400S, 400S to approximately 1000S, 400S to approximately 900S, 400S to approximately 800S, 400S to approximately 700S, 400S to approximately 600S, 400S to approximately 500S, 500S to approximately 1000S, 500S to approximately 900S, 500S to approximately 800S, 500S to approximately 700S,The resolution is between 500S and approximately 600S, 600S and approximately 1000S, 600S and approximately 900S, 600S and approximately 800S, 600S and approximately 700S, 700S and approximately 1000S, 700S and approximately 900S, 700S and approximately 800S, 800S and approximately 1000S, 800S and approximately 900S, or 900S and approximately 1000S. In some embodiments, the resolution is approximately 200S. In some embodiments, Smax is one of approximately 100S, 200S, 300S, 400S, 500S, 600S, 700S, 800S, 900S, or 1000S. In some embodiments, Smax is approximately 100S to 1000S, 100S to 900S, 100S to 800S, 100S to 700S, 100S to 600S, 100S to 500S, 100S to 400S, 100S to 300S, 100S to 200S, 200S to 1000S, and from 200S. Approximately 900S, 200S to approximately 800S, 200S to approximately 700S, 200S to approximately 600S, 200S to approximately 500S, 200S to approximately 400S, 200S to approximately 300S, 300S to approximately 1000S, 300S to approximately 900S, 300S to approximately 800S, 300S to approximately 700S, 300S to approximately 600S, 300S or From approximately 500S, 300S to approximately 400S, 400S to approximately 1000S, 400S to approximately 900S, 400S to approximately 800S, 400S to approximately 700S, 400S to approximately 600S, 400S to approximately 500S, 500S to approximately 1000S, 500S to approximately 900S, 500S to approximately 800S, 500S to approximately 700S, 500 Smax is between approximately 600S, 600S to approximately 1000S, 600S to approximately 900S, 600S to approximately 800S, 600S to approximately 700S, 700S to approximately 1000S, 700S to approximately 900S, 700S to approximately 800S, 800S to approximately 1000S, 800S to approximately 900S, or 900S to approximately 1000S. In some embodiments, Smax is approximately 200S to approximately 5000S. In some embodiments, Smax is approximately 200S. In some embodiments, the friction ratio is shifted to a value determined by centrifugal separation software. In some embodiments, the friction ratio is approximately 1.0. In some embodiments, radially invariant (RI) and time-invariant (TI) noise subtraction is applied.In some embodiments, the meniscus position is shifted, and the software selects the optimal position. In some embodiments, the friction ratio is shifted, and the software selects the optimal position. The model fits the data to the Lamb equation, and the resulting size distribution is in units of fringe or OD. 260 This is a "distribution of the settling coefficient," like a chromatogram, with the area under each peak proportional to the concentration in units. The settling coefficient (in Svedberg units) and relative concentration (in OD units) are determined for each component in the distribution. In some embodiments, multiple AUC runs are independent assays, and each analysis is monitored to ensure the quality of the results for the following attributions: goodness of fit (rmsd), and OD for each peak. 260nm / Fringe interference signal ratio (A260 / IF ratio), agreement of sedimentation coefficients for various aspects between runs, and overall scan quality.

[0085] In some embodiments of the present invention, the extinction coefficient is used to calculate the molar concentration and the actual percentage value of the intact vector peak from absorbance data. Empty capsid (E 260 / カプシド =3.72e6) and intact vector (E 260 The molar extinction coefficients for both the vector and the intact vector (3.00e7) can be calculated based on the published formula (Sommer et al., (2003) Mol Ther., 7:122-128). The extinction coefficients are available for both empty capsids and intact vector peaks. The C(S) value can be determined using the SEDFIT algorithm described in Schuck (2000) Biophys.J., 78:1606-19. The molar concentrations of both intact vectors and empty capsids are calculated using Baer's law, and the percentage of complete capsids is derived from these values. In some embodiments, the values ​​are reported as percentages of complete capsids.

[0086] In some embodiments, it is impossible to experimentally determine the extinction coefficient of a specific species of recombinant virus particle (e.g., a virus particle with a fragmented genome of unknown size and sequence). The relationship between the S value and genome size is established by analyzing recombinant viral vector preparations containing capsidized viral genomes with known nucleotide sizes, and the corresponding S values ​​are determined as described herein. The calculated S values ​​are plotted to generate a standard curve on which recombinant viral species of unknown molecular weight or genome size are compared to determine the molecular weight of the unknown species.

[0087] In some embodiments, the present invention provides a method for characterizing preparations of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of the recombinant virus particles is monitored at time intervals (e.g., once or more); b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; and c) integrating the area under each peak in the C(S) distribution to determine the relative concentration of each peak, wherein each peak corresponds to a species of recombinant virus particle. In some embodiments, the species of recombinant virus particles identified by the method of the present invention include, but are not limited to, fully recombinant virus particles containing an intact recombinant virus genome, empty recombinant virus capsid particles, and recombinant virus particles containing a variant recombinant virus genome. In some embodiments, the variant genome is smaller than the intact recombinant virus genome (e.g., a truncated genome). In some embodiments, the variant genome is larger than the intact recombinant virus genome (e.g., aggregates, recombinants, etc.). In some embodiments, the present invention provides a method for evaluating the integrity of the vector genome of recombinant virus particles in a preparation of recombinant virus particles, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals (e.g., once or more); b) plotting the differential sedimentation coefficient distribution value C(s) against the sedimentation coefficient (S) in Svedberg units; and c) identifying the species of recombinant virus particles in the preparation by the presence of a peak on the plot corresponding to the S value, wherein the genome size of the specific species of recombinant virus particle is calculated by comparing the S value of the species with a standard curve created by the S values ​​of recombinant virus particles containing capsidized virus genomes of various known sizes. In some embodiments, the method further comprises the step of determining the relative concentrations of various recombinant virus particles by integrating the area under each peak in the C(S) distribution.In some embodiments, the sedimentation of recombinant virus particles is monitored once per time interval. In some embodiments, the sedimentation of recombinant virus particles is monitored more than once per time interval.

[0088] In some embodiments of the present invention, the sedimentation of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) is monitored by measuring optical density or absorbance at approximately 260 nm. Means for measuring absorbance are well known in the art. In some embodiments, the ultracentrifuge used for AUC is equipped with means for measuring absorbance. In other embodiments, the sedimentation of recombinant virus particles is monitored by interference. In some embodiments, the sedimentation of recombinant virus particles is monitored by Rayleigh interference. Means for measuring interference are well known in the art (Furst (1997) Eur. Biophys. J. 35: pp. 307-310). In some embodiments, the ultracentrifuge used for AUC is equipped with means for measuring interference. In some embodiments, the sedimentation of recombinant virus particles is monitored by both absorbance and interference. In some embodiments, absorbance and / or interference are measured using a reference standard. In some embodiments, the reference standard is identical to a solution of the recombinant virus preparation, except that the recombinant virus is absent. For example, the recombinant virus preparation contains the recombinant virus in a buffer such as phosphate-buffered saline. In this example, the reference standard is phosphate-buffered saline that does not contain recombinant virus particles.

[0089] In some embodiments of the present invention, preparations of virus particles are contained in pharmaceutical formulations. Such formulations are well known in the art (see, for example, Remington's Pharmaceutical Sciences, 15th edition, pp. 1035-1038 and 1570-1580). Such pharmaceutical formulations may be sterile liquids such as water and oils, including petroleum, animal, peanut oil, soybean oil, mineral oil, and other plant or synthetic sources. Saline solutions and glucose aqueous solutions, polyethylene glycol (PEG), and glycerin solutions can also be used as liquid carriers, particularly in injectable solutions. Pharmaceutical formulations may further contain additional components, such as preservatives, buffers, isotonic agents, antioxidants and stabilizers, nonionic wetting or clarifying agents, thickeners, and the like. In some embodiments of the present invention, the pharmaceutical formulation contains phosphate-buffered saline.

[0090] In some embodiments of the present invention, the sedimentation rate of virus particles during ultracentrifugation is determined by continuously monitoring the sedimentation of virus particles during ultracentrifugation. Optimizing the AUC parameter for various types of virus particles is within the control of those skilled in the art. Not intended to be bound by theory, the range of AUC settings that allows analysis of both AAV and adenovirus particles should also allow analysis of other virus particles, including lentiviruses and HSV, since the size of HSV and lentivirus particles is between that of AAV and adenovirus particles. In some embodiments, data acquisition for rAAV, rHSV, lentivirus, and / or rAd particles is performed at AUC rates between approximately 3,000 and approximately 20,000 rpm. In some embodiments, data analysis for rAAV, HSV, lentivirus, and / or adenovirus particles is performed at an AUC rate of approximately 1 s. min and about 1000S of S maxThis is carried out by the following embodiment: In some embodiments, data analysis for rAAV, rHSV, lentiviruses and / or rAd particles is performed at a resolution of about 200S to about 1,000S. In some embodiments, the resolution is one of about 200S, 300S, 400S, 500S, 600S, 700S, 800S, 900S or 1000S. In some embodiments, the resolution is approximately 200S to 1000S, 200S to 900S, 200S to 800S, 200S to 700S, 200S to 600S, 200S to 500S, 200S to 400S, 200S to 300S, 300S to 1000S, 300S to 900S, 300S to 800S, 300S to 700S, 300S to 600S, 300S to 500S, 300S to 400S, 400S to 1000S, 400S to 900S, and 400S to 800S. The resolution is between 400S and approximately 700S, 400S and approximately 600S, 400S and approximately 500S, 500S and approximately 1000S, 500S and approximately 900S, 500S and approximately 800S, 500S and approximately 700S, 500S and approximately 600S, 600S and approximately 1000S, 600S and approximately 900S, 600S and approximately 800S, 600S and approximately 700S, 700S and approximately 1000S, 700S and approximately 900S, 700S and approximately 800S, 800S and approximately 1000S, 800S and approximately 900S, or 900S and approximately 1000S. In some embodiments, the resolution is approximately 200S. Data analysis for rAAV, rHSV, lentiviruses, and / or rAd particles is performed at one of the following Smax values: approximately 100S, 200S, 300S, 400S, 500S, 600S, 700S, 800S, 900S, or 1000S.In some embodiments, Smax is approximately 100S to 1000S, 100S to 900S, 100S to 800S, 100S to 700S, 100S to 600S, 100S to 500S, 100S to 400S, 100S to 300S, 100S to 200S, 200S to 1000S, 200S to 900S, 200S to 800S, 200S to 700S, 200S to 600S, and 200S. From approximately 500S, from 200S to approximately 400S, from 200S to approximately 300S, from 300S to approximately 1000S, from 300S to approximately 900S, from 300S to approximately 800S, from 300S to approximately 700S, from 300S to approximately 600S, from 300S to approximately 500S, from 300S to approximately 400S, from 400S to approximately 1000S, from 400S to approximately 900S, from 400S to approximately 800S, from 400S to approximately 700S, from 400S to approximately 600S, from 400S to approximately 500S. The time range is between 500S and approximately 1000S, 500S and approximately 900S, 500S and approximately 800S, 500S and approximately 700S, 500S and approximately 600S, 600S and approximately 1000S, 600S and approximately 900S, 600S and approximately 800S, 600S and approximately 700S, 700S and approximately 1000S, 700S and approximately 900S, 700S and approximately 800S, 800S and approximately 1000S, 800S and approximately 900S, or 900S and approximately 1000S. In some embodiments, Smax is between approximately 200S and approximately 5000S. In some embodiments, Smax is approximately 200S. In some embodiments, radially invariant (RI) and time-invariant (TI) noise subtraction is applied. In some embodiments, the meniscus position is shifted and the software selects the optimal position. In some embodiments, the friction ratio is shifted and the software selects the optimal position. In some embodiments, the data analysis for rAAV and / or adenovirus particles is kept constant at 1. In some embodiments, the data analysis for rAAV, HSV, lentivirus, and / or adenovirus particles is shifted using the FIT command with values ​​optimized using nonlinear regression.

[0091] With respect to recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), in some embodiments, the sedimentation rate of recombinant viruses during ultracentrifugation is determined by monitoring (e.g., scanning) the sedimentation of recombinant virus particles once every approximately 15 seconds, 30 seconds, 45 seconds, 1 minute (60 seconds), 2 minutes, 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes, 8 minutes, 9 minutes, 10 minutes, 15 minutes, 20 minutes, and 25 minutes. Scans are acquired continuously and without delay as quickly as possible with the optical system. Interferometric scanning is rapid, with one scan completed in approximately 10-15 seconds, while absorbance scanning requires approximately 60 seconds. If dual detection is used, the scan acquisition rate for both is determined by the absorbance system. In some embodiments of the present invention, during ultracentrifugation, more than approximately 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, or 100 scans are used to monitor the sedimentation of recombinant virus particles. In some embodiments, a minimum of 30 scans is required for analysis, and the scans are recovered before the sedimentation process is complete. In some embodiments, the sedimentation process is typically described as being between 40 and 75 scans. In some embodiments, the sedimentation rate of recombinant virus particles is determined based on approximately 75 scans. In some embodiments, the sedimentation rate of recombinant virus particles is determined based on approximately 55 to approximately 75 scans. In some embodiments, the sedimentation rate of recombinant virus particles is determined based on approximately 55 to approximately 60 scans. In some embodiments, the sedimentation rate of recombinant virus particles is determined based on approximately 60 to approximately 75 scans. In some embodiments, the sedimentation rate of recombinant virus particles is determined based on approximately 60 to approximately 70 scans. In some embodiments, the sedimentation velocity of recombinant virus particles is determined based on multiple ultracentrifugation (executions). In some embodiments, the sedimentation velocity of recombinant virus particles is determined based on one, two, three, four, five, six, seven, eight, nine, or ten or more ultracentrifugation executions. In some embodiments, the sedimentation velocity is used to determine the C(S) value using the SEDFIT algorithm.In some embodiments, second-order differential regularization is applied to a fitting level with an F-statistic confidence level of approximately 0.68. In some embodiments, the following C(S) parameters are kept constant: resolution is 100S to approximately 200S, Smin is approximately 1, Smax is approximately 200S to 300S, and friction ratio is approximately 1.0 to 1.2S. In some embodiments, radially invariant (RI) and time-invariant (TI) noise subtraction is applied.

[0092] In some embodiments of the present invention, recombinant virus particles are obtained by ultracentrifugation of a preparation of recombinant virus particles at any of the following speeds: approximately 5,000 rpm; 10,000 rpm; 15,000 rpm; 20,000 rpm; 25,000 rpm; 30,000 rpm; 35,000 rpm; 40,000 rpm; 45,000 rpm; or 50,000 rpm. Boundary sedimentation velocity of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) in the preparation. In some embodiments, ultracentrifugation is performed at approximately 5,000 rpm to approximately 50,000 rpm; approximately 10,000 rpm to approximately 50,000 rpm; approximately 15,000 rpm to approximately 50,000 rpm; approximately 20,000 rpm to approximately 50,000 rpm; approximately 25,000 rpm to approximately 50,000 rpm; approximately 30,000 rpm to approximately 50,000 rpm; approximately 35,000 rpm to approximately 50,000 rpm; approximately 40,000 rpm to approximately 50,000 rpm; approximately 45,000 rpm to approximately 50,000 rpm; approximately 5,000 rpm From approximately 45,000 rpm; from approximately 10,000 rpm to approximately 45,000 rpm; from approximately 15,000 rpm to approximately 45,000 rpm; from approximately 20,000 rpm to approximately 45,000 rpm; from approximately 25,000 rpm to approximately 45,000 rpm; from approximately 30,000 rpm to approximately 45,000 rpm; from approximately 40,000 rpm to approximately 45,000 rpm; from approximately 5,000 rpm to approximately 40,000 rpm; from approximately 10,000 rpm to approximately 40,000 rpm; from approximately 15,000 rpm to approximately 40,000 rpm; from approximately 20,000 rpm to approximately 4 0,000rpm; approximately 25,000rpm to approximately 40,000rpm; approximately 30,000rpm to approximately 40,000rpm; approximately 35,000rpm to approximately 40,000rpm; approximately 5,000rpm to approximately 35,000rpm; approximately 10,000rpm to approximately 35,000rpm; approximately 15,000rpm to approximately 35,000rpm; approximately 20,000rpm to approximately 35,000rpm; approximately 25,000rpm to approximately 35,000rpm; approximately 30,000rpm to approximately 35,000rpm; approximately 5,000rpm to approximately 30,000rpm 0 rpm; approximately 10,000 rpm to approximately 30,000 rpm; approximately 15,000 rpm to approximately 30,000 rpm; approximately 20,000 rpm to approximately 30,000 rpm; approximately 25,000 rpm to approximately 30,000 rpm; approximately 5,000 rpm to approximately 25,000 rpm; approximately 10,000 rpm to approximately 25,000 rpm; approximately 20,000 rpm to approximately 25,000 rpm; approximately 5,000 rpm to approximately 20,000 rpm; approximately 10,000 rpm to approximately 20,000 rpm; approximately 15,000 rpm to approximately 20,000 rpm;The process is performed at approximately 5,000 rpm to approximately 15,000 rpm; approximately 10,000 rpm to approximately 15,000 rpm or between approximately 5,000 rpm and approximately 10,000 rpm. In some embodiments of the present invention, the boundary sedimentation velocity of recombinant virus particles in a recombinant virus particle preparation is obtained by ultracentrifugation of the recombinant virus particle preparation at approximately 20,000 rpm. In some embodiments of the present invention, the boundary sedimentation velocity of recombinant virus particles in a recombinant virus particle preparation is obtained by ultracentrifugation of the recombinant virus particle preparation at approximately 15,000 rpm to approximately 20,000 rpm.

[0093] In some embodiments of the present invention, the boundary sedimentation velocity of recombinant virus particles in a preparation of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) by ultracentrifugation of the preparation of recombinant virus particles at about 4°C, 10°C, 15°C, 20°C, 25°C, or 30°C or higher. In some embodiments, the ultracentrifugation method is performed between about 4°C and about 30°C, about 4°C and about 25°C, about 4°C and about 20°C, about 4°C and about 15°C, about 4°C and about 10°C, about 10°C and about 30°C, about 10°C and about 25°C, about 10°C and about 20°C, about 10°C and about 15°C, about 15°C and about 30°C, about 15°C and about 25°C, about 15°C and about 20°C and about 30°C, or about 20°C and about 25°C. In some embodiments, the boundary sedimentation velocity of recombinant virus particles in a recombinant virus particle preparation obtained by ultracentrifugation of the recombinant virus particle preparation at approximately 20°C. In some embodiments, the boundary sedimentation velocity of recombinant virus particles in a recombinant virus particle preparation obtained by ultracentrifugation of the recombinant virus particle preparation at approximately 15°C to approximately 20°C.

[0094] Numerous types of recombinant virus particles, as disclosed herein, are analyzed by the methods disclosed herein (e.g., AAV, adenovirus, lentivirus and / or HSV particles). Suitable ultracentrifugation conditions, analytical algorithms and other parameters can be experimentally determined through methods well known in the art. Exemplary parameters for Russ and HSV particles are provided non-limitingly in Table 1 below, along with guidance on selecting specific parameter options.

[0095] [Table 1]

[0096] [Table 2]

[0097] In some embodiments, the present invention provides a method for determining the presence of empty capsids in a preparation of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of the recombinant virus particles is monitored at time intervals (e.g., once or more); and b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units, wherein the presence of a peak corresponding to the S value of an empty capsid particle indicates the presence of an empty capsid particle. In some embodiments, the present invention provides a method for measuring the relative amount of empty capsids in a preparation of recombinant virus particles, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals (e.g., once or more); b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) integrating the area under each peak in the C(S) distribution to determine various relative concentrations of recombinant virus particles; and d) comparing the amount of recombinant virus particles having an S value corresponding to empty capsid particles with the amount of recombinant virus particles having an S value corresponding to recombinant virus particles containing intact viral genomes. In some embodiments, the amount of recombinant virus particles having an S value corresponding to empty capsid particles is compared with the total amount of all recombinant virus particles in the preparation by integrating all peaks in the C(S) vs. S plot.

[0098] In some embodiments, the present invention provides a method for determining the presence of recombinant virus particle variants in a preparation of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals (e.g., once or more); and b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units, wherein the presence of a peak corresponding to an S value different from that of recombinant virus capsid particles containing an intact fully recombinant virus genome indicates the presence of recombinant virus particle variants. In some embodiments, the present invention provides a method for measuring the relative amount of recombinant virus particle variants in a recombinant virus particle preparation, comprising the steps of: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals (e.g., once or more); b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units; c) determining various relative concentrations of recombinant virus particles by integrating the area under each peak in the C(S) distribution; and d) comparing the amount of recombinant virus particles having an S value corresponding to empty capsid particles with the amount of recombinant virus particles having an S value corresponding to recombinant virus particles containing intact viral genomes. In some embodiments, the amount of recombinant virus particles having an S value different from that of recombinant virus capsid particles containing intact fully recombinant viral genomes is compared to the total amount of all recombinant virus particles in the preparation by integrating all peaks in the C(S) vs. S plot. In some embodiments, recombinant viral particle variants contain recombinant viral genomes that are smaller (e.g., truncated) or larger than the full-length intact viral genome. Other viral capsidized DNA impurities may also be detected.

[0099] In some embodiments, the present invention provides a method for monitoring the removal of recombinant viral particles containing empty capsids and / or variant genomes during the purification of a preparation of recombinant viral particles (e.g., rAAV, rAd, lentivirus, or rHSV particles), comprising removing a recombinant viral particle sample from the preparation following one or more steps in the purification process, and analyzing the sample for the relative amount of empty capsids using the AUC described herein. A decrease in the relative amount of recombinant viral particles containing empty capsids and / or variant genomes to the complete capsid indicates the removal of empty capsids from the preparation of recombinant viral particles.

[0100] In some embodiments, the present invention provides a method for determining heterogeneity of recombinant virus particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) in a preparation of recombinant virus particles, comprising: a) subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions, wherein the sedimentation of recombinant virus particles is monitored at time intervals (e.g., once or more); and b) plotting the differential sedimentation coefficient distribution value (C(s)) against the sedimentation coefficient (S) in Svedberg units, wherein the presence of a peak in addition to the peak corresponding to a capsid containing an intact viral genome indicates heterogeneity of recombinant virus particles in the preparation. In some embodiments, the species of recombinant virus particles identified by the method of the present invention include, but are not limited to, fully recombinant virus particles containing an intact recombinant viral genome, empty recombinant virus capsid particles, and recombinant virus particles containing a variant recombinant viral genome. In some embodiments, the variant genome is smaller than the intact recombinant viral genome (e.g., a truncated genome). In some embodiments, the variant genome is larger than the intact recombinant viral genome (e.g., aggregates, recombinants, etc.). In some embodiments, the variant genome includes genomes that are smaller or larger than the intact recombinant viral genome.

[0101] In some embodiments, the present invention provides a method for monitoring heterogeneity of recombinant viral particles (e.g., rAAV, rAd, lentivirus, or rHSV particles) during the purification of a preparation of recombinant viral particles, comprising: removing a recombinant viral particle sample from the preparation following one or more steps in the purification process; and determining the relative amount of recombinant viral particles containing intact recombinant viral genomes, empty capsids, and / or variant genomes using the AUC described herein, wherein an increase in the relative amount of recombinant viral particles containing intact viral genomes indicates an increase in heterogeneity of all viral particles in the preparation of recombinant viral particles.

[0102] In the embodiments described above, recombinant virus particles were purified using one or more purification steps. Examples of purification steps include, but are not limited to, equilibrium centrifugation, anion exchange filtration, tangential flow filtration (TFF), apatite chromatography, helper virus thermal inactivation, hydrophobic interaction chromatography, immunoaffinity chromatography, size exclusion chromatography (SEC), nanofiltration, cation exchange chromatography, and anion exchange chromatography.

[0103] In the embodiments described above, the recombinant virus particle comprises a self-complementary AAV (scAAV) genome. In some embodiments, the recombinant AAV genome comprises a first heterogeneous polynucleotide sequence (e.g., a therapeutic transgene coding strand) and a second heterogeneous polynucleotide sequence (e.g., a non-coding or antisense strand of a therapeutic transgene), wherein the first heterogeneous polynucleotide sequence can form intra-chain base pairs with the second polynucleotide sequence along most or all of its length. In some embodiments, the first heterogeneous polynucleotide sequence and the second heterogeneous polynucleotide sequence are linked by a sequence that facilitates intra-chain base pairing; e.g., a hairpin DNA structure. Hairpin structures are well known in the art, for example, in siRNA molecules. In some embodiments, the first heterogeneous polynucleotide sequence and the second heterogeneous polynucleotide sequence are linked by a mutation-introducing ITR. In some embodiments, the scAAV virus particle comprises a monomeric form of the scAAV genome. In some embodiments, the scAAV virus particle comprises a dimeric form of the scAAV genome. In some embodiments, the AUC described herein is used to detect the presence of rAAV particles containing monomeric forms of the scAAV genome. In some embodiments, the AUC described herein is used to detect the presence of rAAV particles containing dimeric forms of the scAAV genome. In some embodiments, the packaging of the scAAV genome into a capsid is monitored by the AUC described herein.

[0104] In the embodiments described above, the rAAV particles are AAV1 capsid, AAV2 capsid, AAV3 capsid, AAV4 capsid, AAV5 capsid, AAV6 capsid (e.g., wild-type AAV6 capsid or variant AAV6 capsid such as ShH10, as described in USPG Pub. 2012 / 0164106), AAV7 capsid, AAV8 capsid, AAVrh8 capsid, AAVrh8R, AAV9 capsid (e.g., wild-type AAV9 capsid or modified AAV9 capsid). As described in Pub.2013 / 0323226, AAV10 capsid, AAVrh10 capsid, AAV11 capsid, AAV12 capsid, tyrosine capsid variants, heparin-binding capsid variants, AAV2R471A capsid, AAVAAV2 / 2-7m8 capsid, AAV DJ capsid (e.g., AAV-DJ / 8 capsid, AAV-DJ / 9 capsid or any other capsid described in USPG Pub.2012 / 0066783), AAV2 N587A capsid, AAV2 E548A capsid, AAV2 N708A capsid, AAV Includes V708K capsid, goat AAV capsid, AAV1 / AAV2 chimeric capsid, bovine AAV capsid, mouse AAV capsid, or AAV capsids described in U.S. Patent No. 8,283,151 or International Publication No. WO2003 / 042397. In the embodiments described, the rAAV particles include at least one AAV1 ITR, AAV2 ITR, AAV3 ITR, AAV4 ITR, AAV5 ITR, AAV6 ITR, AAV7 ITR, AAV8 ITR, AAVrh8 ITR, AAV9 ITR, AAV10 ITR, AAVrh10 ITR, AAV11 ITR, AAV12 ITR, AAV DJ ITR, goat AAV ITR, bovine AAV ITR, or mouse AAV ITR. In some embodiments, the rAAV particles include an ITR derived from one AAV serotype and an AAV capsid derived from another serotype. For example, an rAAV particle may contain a therapeutic transgene having at least one AAV2 ITR capsidized within an AAV9 capsid at its end. Such combinations may be referred to as pseudotyped rAAV particles.

[0105] IV. Virus Particles The methods disclosed herein find use, in particular, in characterizing a target species among various viral particles (e.g., viral particles containing a whole genome compared to viral particles containing a truncated genome and / or viral particles containing DNA impurities).

[0106] In some embodiments, the viral particle is a recombinant AAV particle containing a nucleic acid that includes a transgene having one or two ITRs at its ends. The nucleic acid is capsidized within the AAV particle. The AAV particle also contains a capsid protein. In some embodiments, the nucleic acid includes a control sequence containing transcription start and termination sequences, and a protein-coding sequence(s) of interest (e.g., a therapeutic transgene) operably linked in the transcription direction to the components, thereby forming an expression cassette. The expression cassette is flanked at the 5' and 3' ends by at least one functional AAV ITR sequence. By “functional AAV ITR sequence,” it is meant that the ITR sequence functions for the purpose of rescuing, replicating, and packaging the AAV virion. See Davidson et al., PNAS, 2000, 97(7) pp. 3428-32; Passini et al., J. Virol., 2003, 77(12): pp. 7034-40 and Pechan et al., Gene Ther., 2009, 16: pp. 10-16, which are incorporated herein by reference in their entirety. To carry out some aspects of the present invention, recombinant vectors contain at least the entire sequence of AAV essential for capsidation and physical structure for rAAV infection. The AAV ITR for use in the vectors of the present invention does not need to have a wild-type nucleotide sequence (as described, e.g., Kotin, Hum. Gene Ther., 1994, 5: pp. 793-801), but can be modified by nucleotide insertion, deletion, or substitution, and the AAV ITR can be derived from any of several AAV serotypes. More than 40 serotypes of AAV are now known, and new serotypes and variants of existing serotypes continue to be identified. See Gao et al., PNAS, 2002, 99(18):11854-11866; Gao et al., PNAS, 2003, 100(10):6081-11866 and Bossis et al., J. Virol., 2003, 77(12):6799-110. The use of any AAV serotype is considered to be within the scope of this invention.In some embodiments, the rAAV vector is, without limitation, a vector derived from an AAV serotype, including AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh.10, AAV11, AAV12, tyrosine capsid variants, heparin-binding capsid variants, AAV2R471A capsid, AAVAAV2 / 2-7m8 capsid, AAV DJ capsid, AAV2 N587A capsid, AAV2 E548A capsid, AAV2 N708A capsid, AAV V708K capsid, goat AAV capsid, AAV1 / AAV2 chimeric capsid, bovine AAV capsid, or mouse AAV capsid. In some embodiments, the nucleic acids in the AAV include ITRs such as AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh.10, AAV11, and AAV12. In further embodiments, the AAV particles include capsid proteins such as AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh.10, AAV11, and AAV12. In further embodiments... rAAV particles contain capsid proteins of AAV serotypes derived from Clades AF (Gao et al., J. Virol. 2004, 78(12):6381).

[0107] Various AAV serotypes are used to optimize the transduction of specific target cells or to target specific cell types within specific target tissues (e.g., diseased tissue). rAAV particles may contain viral proteins and viral nucleic acids of the same or mixed serotypes. For example, rAAV particles may contain the AAV9 capsid protein and at least one AAV2 ITR, or the AAV2 capsid protein and at least one AAV9 ITR. In yet another example, rAAV particles may contain capsid proteins derived from both AAV9 and AAV2, and further contain at least one AAV2 ITR. Any combination of AAV serotypes for the production of rAAV particles is provided herein, as each combination is explicitly stated herein.

[0108] In some embodiments, the AAV comprises at least one AAV1 ITR and a capsid protein derived from any of AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV2 ITR and a capsid protein derived from any of AAV1, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV3 ITR and a capsid protein derived from any of AAV1, AAV2, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV4 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV5, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV5 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV6, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV6 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV7, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11, and / or AAV12. In some embodiments, the AAV comprises at least one AAV7 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV8, AAV9, AAVrh.8, AAVrh10, AAV11 and / or AAV12.In some embodiments, the AAV comprises at least one AAV8 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV9, AAVrh.8, AAVrh10, AAV11 and / or AAV12. In some embodiments, the AAV comprises at least one AAV9 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAVrh.8, AAVrh10, AAV11 and / or AAV12. In some embodiments, the AAV comprises at least one AAVrh8 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV8, AAV9, AAVrh10, AAV11 and / or AAV12. In some embodiments, the AAV comprises at least one AAVrh10 ITR and a capsid protein derived from any of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV9, AAV11 and / or AAV12. In some embodiments, the AAV comprises at least one AAV11 ITR and AAV1, AAV2, The AAV comprises a capsid protein derived from any of AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAVrh8, AAV9, AAVrh10 and / or AAV12. In some embodiments, the AAV comprises at least one AAV12 ITR as well as AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAV Contains a capsid protein derived from one of the following: rh8, AAV9, AAVrh10, and / or AAV11.

[0109] Self-complementary AAV virus genome In some embodiments, the present invention provides viral particles containing recombinant self-complementary genomes. Methods of using AAV viral particles having self-complementary genomes and self-complementary AAV genomes are described in U.S. Patents 6,596,535, 7,125,717, 7,765,583, 7,785,888, 7,790,154, 7,846,729, 8,093,054 and 8,361,457, which are incorporated herein by reference in their entirety, and in Wang Z. et al., (2003) Gene Ther 10: pp. 2105-2111. rAAV containing a self-complementary genome rapidly forms double-stranded DNA molecules due to the advantages of its partially complementary sequences (e.g., complementary coding and non-coding strands of the transgene). In some embodiments, the present invention provides an AAV virus particle comprising an AAV genome, the rAAV genome comprising a first heterogeneous polynucleotide sequence (e.g., a therapeutic transgene coding strand) and a second heterogeneous polynucleotide sequence (e.g., a non-coding or antisense strand of a therapeutic transgene), wherein the first heterogeneous polynucleotide sequence can form interstrand base pairs with the second polynucleotide sequence along most or all of its length. In some embodiments, the first heterogeneous polynucleotide sequence and the second heterogeneous polynucleotide sequence are linked by a sequence that facilitates interstrand base pairing; e.g., a hairpin DNA structure. Hairpin structures are well known in the art, for example, in siRNA molecules. In some embodiments, the first heterogeneous polynucleotide sequence and the second heterogeneous polynucleotide sequence are linked by a mutagenesis ITR (e.g., a correct ITR). The mutagenesis ITR includes a deletion of a D region containing a terminal degradation sequence. As a result, when replicating the AAV viral genome, the rep protein does not cleave the viral genome at the mutagenesis ITR so that the recombinant viral genome containing the following: AAV ITR, a first heterologous polynucleotide sequence including a regulatory sequence, a mutagenesis AAV ITR, a second heterologous polynucleotide in the opposite direction to the first heterologous polynucleotide, and a third AAV ITR in the order from 5' to 3' is packaged in the viral capsid.

[0110] In some embodiments, the virus particle is an adenovirus particle. In some embodiments, the adenovirus particle is a recombinant adenovirus particle, for example, a polynucleotide vector containing one or more heterologous sequences (i.e., non-adenovirus-derived nucleic acid sequences) between two ITRs. In some embodiments, the adenovirus particle is missing or contains one or more E1 gene deletion copies that cause the adenovirus to fail to replicate. Adenoviruses contain a linear double-stranded DNA genome within a large (approximately 950 Å), non-enveloped icosahedral capsid. Adenoviruses have a large genome (e.g., instead of the E1 and / or E3 regions) that can incorporate heterologous sequences exceeding 30 kb, making them uniquely suited for use with large heterologous genes. They are also known to infect dividing and non-dividing cells and are not incorporated into the host genome in nature (although hybrid variants may possess this ability). In some embodiments, the adenovirus vector may be a first-generation adenovirus vector with heterologous sequences instead of E1. In some embodiments, the adenovirus vector may be a second-generation adenovirus vector having additional mutations or deletions in E2A, E2B, and / or E4. In some embodiments, the adenovirus vector may be a third-generation vector that has deleted all viral coding genes and retains only the ITR and packaging signals, and requires a helper adenovirus in trans for replication and packaging. These may be gutted adenovirus vectors. Adenovirus particles are being investigated for use as vectors for transient transfection of mammalian cells and as gene therapy vectors. For further information, see, for example, Danthinne, X. and Imperiale, MJ (2000) Gene Ther. 7: pp. 1707-14 and Tatsis, N. and Ertl, HC (2004) Mol. Ther. 10: pp. 616-29.

[0111] In some embodiments, the viral particle is a recombinant adenovirus particle containing nucleic acid with an introduced gene. The use of any adenovirus serotype is considered to be within the scope of the invention. In some embodiments, the recombinant adenovirus vector is a vector derived from adenovirus serotypes, including but not limited to AdHu2, AdHu3, AdHu4, AdHu5, AdHu7, AdHu11, AdHu24, AdHu26, AdHu34, AdHu35, AdHu36, AdHu37, AdHu41, AdHu48, AdHu49, AdHu50, AdC6, AdC7, AdC69, bovine Ad3, canine Ad2, sheep Ad, and porcine Ad3. The adenovirus particle also contains a capsid protein. In some embodiments, the recombinant viral particle contains an adenovirus particle in combination with one or more foreign viral capsid proteins. Such combinations may be referred to as pseudotyped recombinant adenovirus particles. In some embodiments, the exogenous viral capsid protein used in pseudotyped recombinant adenovirus particles is derived from an exogenous virus or another adenovirus serotype. In some embodiments, the exogenous viral capsid protein is derived, non-limitingly, from those containing reovirus type 3. Examples of vector and capsid protein combinations used in pseudotyped adenovirus particles can be found in the following references (Tatsis, N. et al., (2004) Mol. Ther. 10(4): pp. 616-629 and Ahi, Y. et al., (2011) Curr. Gene Ther. 11(4): pp. 307-320). Various adenovirus serotypes are used to optimize the transduction of specific target cells or to target specific cell types within specific target tissues (e.g., diseased tissue). Tissues or cells targeted by specific adenovirus serotypes include, but are not limited to, the lungs (e.g., HuAd3), spleen and liver (e.g., HuAd37), smooth muscle, synovial cells, dendritic cells, cardiovascular cells, tumor cell lines (e.g., HuAd11), and dendritic cells (e.g., HuAd5, HuAd30, or HuAd35 pseudotyped with reovirus type 3).For further information, see Ahi, Y. et al., (2011) Curr. Gene Ther. 11(4): pp. 307-320, Kay, M. et al., (2001) Nat. Med. 7(1): pp. 33-40, and Tatsis, N. et al., (2004) Mol. Ther. 10(4): pp. 616-629.

[0112] In some embodiments, the viral particle is a lentiviral particle. In some embodiments, the lentiviral particle is a recombinant lentiviral particle, for example, a polynucleotide vector containing one or more heterologous sequences (i.e., non-lentiviral nucleic acid sequences) between two LTRs. Lentiviruses are plasmid-sense, ssRNA retroviruses with a genome of approximately 10 kb. Lentiviruses are well known to be incorporated into the genomes of dividing and non-dividing cells. Lentiviral particles are produced, for example, by transfecting a packaging cell line that packages the modified lentiviral genome into lentiviral particles using multiple plasmids (typically the lentiviral genome and genes necessary for replication and / or packaging separated to prevent viral replication). In some embodiments, the lentiviral particle may refer to a first-generation vector lacking the envelope protein. In some embodiments, the lentiviral particle may refer to a second-generation vector lacking all genes except the gag / pol and tat / rev regions. In some embodiments, lentiviral particles may refer to a third-generation vector having a chimeric LTR for transduction that contains only the endogenous rev, gag, and pol genes and does not contain the tat gene (Dull,T See also (1998) J. Virol. 72:8463-71). For further information, see Durand, S. and Cimarelli, A. (2011) Viruses 3:132-59.

[0113] In some embodiments, the viral particle is a recombinant lentiviral particle containing nucleic acid including a transgene. The use of any lentiviral vector is considered to be within the scope of the present invention. In some embodiments, the lentiviral vector is derived from lentiviruses, non-limitingly including human immunodeficiency virus-1 (HIV-1), human immunodeficiency virus-2 (HIV-2), simian immunodeficiency virus (SIV), feline immunodeficiency virus (FIV), equine infectious anemia virus (EIAV), bovine immunodeficiency virus (BIV), Jambrana disease virus (JDV), visna virus (VV), and canine arthritis encephalitis virus (CAEV). The lentiviral particle also includes a capsid protein. In some embodiments, the recombinant viral particle comprises a lentiviral vector in combination with one or more exogenous viral capsid proteins. Such combinations may be referred to as pseudotyped recombinant lentiviral particles. In some embodiments, the exogenous viral capsid proteins used in pseudotyped recombinant lentiviral particles are derived from exogenous viruses. In some embodiments, the exogenous viral capsid protein used in pseudotyped recombinant lentiviral particles is vesicular stomatitis virus glycoprotein (VSV-GP). VSV-GP interacts with eccentric cell receptors and provides broad tissue targeting to pseudotyped recombinant lentiviral particles. Additionally, VSV-GP is thought to provide greater stability to pseudotyped recombinant lentiviral particles.In other embodiments, the exogenous viral capsid protein may not be limited to Chandipla virus, rabies virus, Mokola virus, lymphocytic choriomeningitis virus (LCMV), Ross River virus (RRV), Sindbis virus, Semlik Forest virus (SFV), Venezuelan encephalitis virus, Ebola virus Reston, Ebola virus Zaire, Marburg virus, Lassa virus, avian leukemia virus (ALV), Jagzikte sheep retrovirus (JSRV), Molony's mouse leukemia virus (MLV). V) is derived from viruses including gibbon leukemia virus (GALV), feline endogenous retrovirus (RD114), human T-lymphotropic virus 1 (HTLV-1), human foamy virus, Maedivisna virus (MVV), SARS-CoV, Sendai virus, respiratory syncytial virus (RSV), human parainfluenza virus type 3, hepatitis C virus (HCV), influenza virus, chicken plague virus (FPV), or Autographa calihornica multiple nucleus multinucleus virus (AcMNPV). Examples of vector and capsid protein combinations used in pseudotyped lentiviral particles can be found, for example, in Cronin, J. et al., (2005). Curr. Gene Ther. 5(4): pp. 387-398. Various pseudotyped recombinant lentiviral particles are used to optimize the transduction of specific target cells or to target specific cell types within specific target tissues (e.g., diseased tissues).Tissues targeted by specific pseudotyped recombinant lentiviral particles include, but are not limited to, the liver (e.g., pseudotyped with VSV-G, LCMV, RRV, or SeV F proteins), the lungs (e.g., pseudotyped with Ebola, Marburg, SeV F and HN, or JSRV proteins), islet cells (e.g., pseudotyped with LCMV protein), the central nervous system (e.g., pseudotyped with VSV-G, LCMV, rabies, or Mokola proteins), the retina (e.g., pseudotyped with VSV-G or Mokola protein), monocytes or muscle cells (e.g., pseudotyped with Mokola or Ebola proteins), the hematopoietic system (e.g., pseudotyped with RD114 or GALV protein), or cancer cells (e.g., pseudotyped with GALV or LCMV protein). For further information, see Cronin, J. et al., (2005). Curr. Gene Ther. 5(4): pp. 387-398 and Kay, M. et al., (2001) Nat. Med. 7(1): pp. 33-40.

[0114] In some embodiments, the viral particles are herpes simplex virus (HSV) particles. In some embodiments, the HSV particles are rHSV particles, a polynucleotide vector containing one or more heterologous sequences (i.e., not lentiviral nucleic acid sequences) between two TRs. HSV is an enveloped, double-stranded DNA virus with a genome of approximately 152 kb. Advantageously, approximately half of its genes are non-essential and can be deleted to adapt to heterologous sequences. HSV particles infect non-dividing cells. Additionally, they can establish latency in neurons in nature, migrate by retrograde transport, and cross synapses, making them advantageous for neuronal transfection and / or gene therapy techniques involving the nervous system. In some embodiments, HSV particles may be replication-deficient or replication-eligible (e.g., eligible for a single replication cycle through inactivation of one or more late genes). For further description, see Manservigi, R. et al., (2010) Open Virol. J. 4: pp. 123-56.

[0115] In some embodiments, the viral particles are rHSV particles containing nucleic acids including a transgene. The use of any HSV vector is considered to be within the scope of the invention. In some embodiments, the HSV vector is derived from HSV serotypes, including, but not limited to, HSV-1 and HSV-2. The HSV particles also include a capsid protein. In some embodiments, the recombinant viral particles consist of an HSV vector in combination with one or more exogenous viral capsid proteins. Such combinations may be referred to as pseudotyped rHSV particles. In some embodiments, the exogenous viral capsid protein used in pseudotyped rHSV particles is derived from an exogenous virus or another HSV serotype. In some embodiments, the exogenous viral capsid protein used in pseudotyped rHSV particles is vesicular stomatitis virus glycoprotein (VSV-GP). VSV-GP interacts with eccentric cell receptors and provides broad tissue targeting to pseudotyped rHSV particles. Additionally, VSV-GP is thought to provide greater stability to pseudotyped rHSV particles. In other embodiments, the exogenous viral capsid protein may be derived from a different HSV serotype. For example, an HSV-1 vector may contain one or more HSV-2 capsid proteins. Various HSV serotypes are used to optimize the transduction of specific target cells or to target specific cell types within specific target tissues (e.g., diseased tissues). Tissues or cells targeted by specific adenovirus serotypes include, but are not limited to, the central nervous system and neurons (e.g., HSV-1). For further information, see Manservigi, R. et al., (2010) Open Virol J 4: pp. 123-156, Kay, M. et al., (2001) Nat. Med. 7(1): pp. 33-40, and Meignier, B. et al., (1987) J. Infect. Dis. 155(5): pp. 921-930.

[0116] V. Production of viral vectors Numerous methods are well known in the art for the production of rAAV vectors, including transfection, stable cell line production, and infectious hybrid virus production systems including adenovirus AAV hybrids, herpesvirus AAV hybrids (Conway, JE et al., (1997) J. Virology 71(11):8780-8789) and baculovirus AAV hybrids. All rAAV production cultures for the production of rAAV virus particles consist of: 1) suitable host cells, including human cell lines such as HeLa, A549, or 293 cells, or insect cell lines such as SF-9 in the case of baculovirus production systems; 2) suitable helper virus functions provided by wild-type or mutant adenoviruses (such as temperature-sensitive adenoviruses), herpesviruses, baculoviruses, or plasmid constructs that provide helper function; 3) AAV rep and cap genes and gene products; 4) transgenes (such as therapeutic transgenes) having at least one AAV ITR sequence at the end; and 5) suitable cultures that support rAAV production. The process requires soil and culture medium components. In some embodiments, the AAV rep and cap gene products are derived from any AAV serotype. While not essential, the AAV rep gene product is generally of the same serotype as the ITR of the rAAV vector genome, as long as the rep gene product functions to replicate and package the rAAV genome. Suitable media known in the art are used for the production of rAAV vectors. These media include, but are not limited to, modified Eagle medium (MEM), media produced by Hyclone Laboratories and JRH including Dulbecco's modified Eagle medium (DMEM), custom formulations such as those described in U.S. Patent No. 6,566,118, which are incorporated herein by reference in their entirety, in particular with respect to custom media formulations for use in the production of recombinant AAV vectors, and Sf-900 II SFM medium described in U.S. Patent No. 6,723,551. In some embodiments, the AAV helper function is provided by adenovirus or HSV. In some embodiments, the AAV helper function is provided by a baculovirus, and the host cell is an insect cell (e.g., armyworm (Spodoptera frugiperda) (Sf9) cell).

[0117] The preferred rAAV-producing culture medium of the present invention may be supplemented with serum or serum-derived recombinant protein at levels of 0.5% to 20% (v / v or w / v). Alternatively, as is well known in the art, rAAV vectors are produced under serum-free conditions, also known as animal-derived component-free media. Those skilled in the art will recognize that commercial or custom media designed to support rAAV vector production are supplemented with one or more cell culture components well known in the art, including, but not limited to, glucose, vitamins, amino acids, and / or growth factors, to increase the titer of rAAV in the resulting culture.

[0118] In some embodiments, the present invention provides a method for preparing rAAV particles with reduced empty capsids, comprising: a) culturing host cells under conditions suitable for rAAV production, wherein the cells comprise i) a nucleic acid encoding a heterologous transgene having at least one AAV ITR at its end, ii) a nucleic acid comprising AAV rep and cap coding regions, comprising a mutagenerated p5 promoter, wherein expression from the p5 promoter is reduced compared to that of a wild-type p5 promoter, and iii) a nucleic acid encoding AAV helper virus function; b) lysing the host cells to release rAAV particles; c) isolating the rAAV particles produced by the host cells; and d) analyzing the rAAV particles by the analytical ultracentrifugation method described above for the presence of rAAV particles having empty capsids and / or variant genomes. In some embodiments, the p5 promoter of the nucleic acid encoding the AAV rep and cap regions is located at 3' of the rep and / or cap coding regions. In some embodiments, the nucleic acids encoding the AAV rep and cap coding regions are plasmid pHLP, pHLP19, or pHLP09 (see U.S. Patents 5,622,856; 6,001,650; 6,027,931; 6,365,403; 6,376,237; and 7,037,713, the contents of which are incorporated herein in their entirety). In some embodiments, the AAV helper virus function includes adenovirus E1A function, adenovirus E1B function, adenovirus E2A function, adenovirus VA function, and adenovirus E4 or f6 function.

[0119] rAAV-producing cultures are grown under various conditions suitable for the specific host cells used (e.g., over a wide temperature range and over various time periods). As is well known in the art, rAAV-producing cultures include adhesion-dependent cultures, which can be cultured in suitable adhesion-dependent containers such as rotating bottles, hollow fiber filters, microcarriers, and packed bed or fluidized bed bioreactors. rAAV vector-producing cultures can also be grown in disposable systems such as spinner flasks, stirred tank bioreactors, and wave bag systems. This also includes suspension-compatible host cells such as HeLa, 293, and SF-9 cells, which can be cultured using various methods.

[0120] The rAAV vector particles of the present invention can be collected from rAAV-producing cultures by lysing the host cells of the producing culture or by collecting the consumed medium from the producing culture when the cells are cultured under conditions that are well known in the art to result in the release of rAAV particles from intact cells into the culture medium, as further fully described in U.S. Patent No. 6,566,118. Preferred methods for lysing cells are well known in the art and include, for example, multiple freeze / thaw cycles, sonication, micro-solution and treatment with chemicals such as surfactants and / or proteases.

[0121] Numerous methods are well known in the art for the production of adenovirus vector particles. For example, guttid adenovirus vectors, adenovirus vector genomes, and helper adenovirus genomes can be transfected into packaging cell lines (e.g., the 293 cell line). In some embodiments, the helper adenovirus genome may contain a recombination site adjacent to its packaging signal, and both genomes are transfected into packaging cell lines expressing a recombinase (e.g., the Cre / loxP system is used) so that the adenovirus vector of interest is packaged more efficiently than the helper adenovirus (see, for example, Alba, R. et al., (2005) Gene Ther. 12 Suppl 1:S18-27). Adenovirus vectors can be collected and purified using standard methods such as those described herein.

[0122] Numerous methods are well known in the art for the production of lentiviral vector particles. For example, for third-generation lentiviral vectors, a vector containing the target lentiviral genome along with the gag and pol genes is co-transfected with a packaging cell line (e.g., the 293 cell line) along with a vector containing the rev gene. The target lentiviral genome also contains a chimeric LTR that promotes transcription in the absence of Tat (see Dull, T. et al., (1998) J. Virol. 72:8463-71). Lentiviral vectors can be collected and purified using the methods described herein (e.g., Segura MM, et al., (2013) Expert Opin Biol Ther. 13(7):987-1011).

[0123] Numerous methods are well known in the art for the production of HSV particles. HSV vectors can be collected and purified using standard methods, such as those described herein. For example, for a replication-deficient HSV vector, the target HSV genome, which is missing all of the earliest (IE) genes, is transfected into a complementary cell line that provides genes necessary for virus production, such as ICP4, ICP27, and ICP0 (see, e.g., Samanieggo, LA et al., (1998) J. Virol. 72:3307-20). HSV vectors can be collected and purified using the methods described (e.g., Goins, WF et al., (2014) Herpes Simplex Virus Methods). in Molecular Biology 1144:63-79).

[0124] Purification of VI.rAAV vector During collection, the rAAV-producing culture of the present invention may contain one or more of the following: (1) host cell proteins; (2) host cell DNA; (3) plasmid DNA; (4) helper viruses; (5) helper virus proteins; (6) helper virus DNA; and (7) culture medium components, such as serum proteins, amino acids, transferrin, and other low molecular weight proteins. Additionally, the rAAV-producing culture may be selected from the group consisting of AAV1, AAV2, AAV3, AAV4, AAV5, AAV6, AAV7, AAV8, AAVrh8, AAV9, AAV10, AAVrh10, AAV11, or AAV12. The rAAV culture further comprises rAAV particles having an AAV capsid serotype. In some embodiments, the rAAV-producing culture further comprises empty AAV capsids (e.g., rAAV particles containing capsid proteins but not rAAV genomes). In some embodiments, the rAAV-producing culture further comprises rAAV particles containing variant rAAV genomes (e.g., rAAV particles containing rAAV genomes different from intact full-length rAAV genomes). In some embodiments, the rAAV-producing culture further comprises rAAV particles containing truncated rAAV genomes. In some embodiments, the rAAV-producing culture further comprises rAAV particles containing AAV capsidized DNA impurities.

[0125] In some embodiments, the rAAV-producing culture collection is clarified by removing host cell debris. In some embodiments, the producing culture collection is filtered using, for example, a grade DOHC Millipore Millistak+ HC Pod Filter, a grade A1HC Millipore Millistak+ HC Pod Filter, and a 0.2μm Filter Opticap XL1O Millipore Express. The solution is clarified by filtration through a series of depth filters, including an SHC hydrophilic membrane filter. Clarification can also be achieved by various other standard techniques well known in the art, such as centrifugation or filtration through any cellulose acetate filter with a pore size of 0.2 μm or larger, as well as well known in the art.

[0126] In some embodiments, the rAAV-producing culture harvest is further treated with Benzonase® to digest any high molecular weight DNA present in the producing culture. In some embodiments, Benzonase® digestion is carried out under standard conditions well known in the art, including, for example, a final concentration of Benzonase® at 1 to 2.5 units / ml, and a temperature ranging from ambient temperature to 37°C for 30 minutes to several hours.

[0127] rAAV particles are isolated or purified using one or more of the following purification steps: equilibrium centrifugation; flow-through anion exchange filtration; tangential flow filtration (TFF) for concentration of rAAV particles; rAAV capture by apatite chromatography; thermal inactivation of helper viruses; rAAV capture by hydrophobic interaction chromatography; buffer exchange by size exclusion chromatography (SEC); nanofiltration; and rAAV capture by anion exchange chromatography, cation exchange chromatography, or affinity chromatography. These steps are used individually, in various combinations, or in various orders. In some embodiments, the method includes all steps in the order described below. Methods for purifying rAAV particles can be found, for example, in Xiao et al., (1998) Journal of Virology 72:2224-2232, U.S. Patents 6,989,264 and 8,137,948 and WO2010 / 148143. Methods for purifying adenovirus particles can be found, for example, in Bo, H et al., (2014) Eur.J.Pharm.Sci.67C:pp. 119-125. Methods for purifying lentivirus particles can be found, for example, in Segura MM et al., (2013) Expert Opin Biol Ther.13(7):pp. 987-1011. Methods for purifying HSV particles can be found, for example, in Goins, WF et al., (2014) Herpes Simplex Virus Methods in Molecular Biology 1144:pp. 63-79. [Examples]

[0128] The present invention can be understood more fully by reference to the following examples, however they are not intended to limit the scope of the invention. The examples and embodiments described herein are for illustrative purposes only, and various modifications or changes in that regard will be suggested to those skilled in the art and will be understood to be within the spirit and scope of this application and the appended claims. [Examples]

[0129] Characterization of recombinant adeno-associated virus vector preparations by analytical ultracentrifugation. Adeno-associated viruses (AAVs) possess properties that make them attractive as vectors for gene therapy. Wild-type AAV consists of two coding regions (rep and cap) that encode all the structural and regulatory elements necessary for association, replication, and infection. The rep ORF encodes the Rep78 and 68 proteins, which have genome replication function, as well as the Rep52 and 40 proteins, which are involved in single-strand replication and packaging. The cap ORF encodes three structural capsid proteins: VP1, VP2, and VP3. Recombinant AAV vectors were typically produced by triple transfection using a "gutless" vector technique (Xiao, X et al., 1998, J. Virol. 3: pp. 2224-2232). The rep and cap genes are replaced with therapeutic genes, their regulatory elements are sandwiched between 5' and 3' terminal inversion sequences (ITRs), the rep and cap genes are provided trans into separate plasmids, and a third plasmid contributes the required adenovirus helper gene. The viral capsid was assumed to fully assemble, and then the ITR-flanking vector genome was assumed to be inserted into the capsid via a capsid pore (Myers, MW & Carter, BJ, 1980, Virology, 102: pp. 71-82). The resulting population of capsids contained both genome-free capsids (empty capsids) and genome-containing capsids. Additionally, the capsids sometimes contained incomplete portions of the recombinant viral genome. The vector preparations were then purified by affinity chromatography to isolate the capsids from the cell debris and further processed by anion exchange chromatography to concentrate for intact vectors.

[0130] Based on recent approvals for use in gene therapy, adeno-associated virus (AAV) vectors have emerged as a significant class of novel biopharmaceutical drug products. The production of AAV vector products requires analytical methods to monitor product quality in terms of uniformity, purity, and manufacturing integrity; however, to date, no methods supporting AAV vector characterization have been established. To address this need, we investigated the potential use of analytical ultracentrifugation (AUC) as a technique for characterizing the uniformity of AAV vectors.

[0131] method Sample preparation To support accurate AUC assessment, the vector product (AAV2 transgene 2) was highly purified, appropriately buffered, and 5x10⁻¹⁰ 11 The concentration was increased to a level higher than vg / mL. To achieve this, the cell supernatant was purified using AVB affinity chromatography (GE Healthcare) and buffered with PBS, pH 7.2 using a 10K MWCO Slide-a-Lyzer (Thermo Scientific). The product concentration was measured by spectroscopic analysis at 260 nm (OD 260 The optical density was determined by optical density measurement at 260 nm. Sample preparation was performed to obtain reproducible and consistent AUC data, with the target concentration ranging from 0.1 to 1.0 by optical density measurement at 260 nm, using direct dilution in PBS or Amicon. This was either by further enrichment using an Ultra-0.5 / 30K MWCO centrifugal filter device.

[0132] Acquisition of seismic velocity AUC data Ultracentrifugation velocity analysis (SV-AUC) was performed using a ProteomeLab® XL-I (Beckman Coulter). 400 μL of sample was loaded into the sample sectors of two sector velocity cells, and 400 μL of PBS was loaded into the corresponding reference sector. The sample was placed in a 4-well rotor and allowed to rise to a temperature of 20°C in the instrument. The system was equilibrated, and a high vacuum was maintained for approximately 1 hour. Sedimentation velocity centrifugation was performed at 20,000 rpm, 20°C, and a 0.003 cm radius step setting, without delay or repetition. Absorbance (260 nm) and Rayleigh interference optics were used to simultaneously record radial concentrations as a function of time until the smallest settling component cleared the optical window (1.2 hours). The assay volume was limited to a single sample per run based on absorbance scan recovery times longer than 1 minute, as well as the large size of the AAV and rapid sedimentation.

[0133] AUC Data Analysis The percentage of complete capsids was determined by analyzing approximately 75 scans from each detection method using the SEDFIT (see the worldwide web at NIH / analyticalultracentrifugation.com) continuous size C(S) distribution model. nd Differential regularization was applied to a fitting with a confidence level of F-statistics = 0.68. The following C(S) parameters were kept constant: resolution = 200S, Smin = 1, Smax = 200, and friction ratio = 1.0. RI and TI noise subtraction were applied to shift the meniscus position, allowing the software to select the optimal position. This model fitted the data to the Lamb equation, and the resulting size distribution is expressed in units of fringe or OD. 260 The distribution of sedimentation coefficients resembled a chromatogram, with each peak having an area under it proportional to its concentration. The sedimentation coefficient (Svedberg units) and relative concentration (OD units) were determined for each component in the distribution. Each AUC run was an independent assay, and each analysis was monitored for the following characteristics to ensure the quality of the results: goodness of fit (rmsd), and OD for each peak. 260nm The ratio of interference signals in the fringe (A260 / IF ratio), the consistency of sedimentation coefficients for various aspects between runs, and the overall quality of the scan.

[0134] Absorbance optics (260nm) The extinction coefficient was used to calculate the molar concentration of the intact vector and the actual percentage value of the peak from the absorbance data. Empty capsid (E) 260 / カプシド =3.72e6) and intact vector (E 260 / ベクター The molar extinction coefficients for both (=3.00e7) were calculated based on the published formula (Sommer et al., (2003) Mol Ther., 7: pp. 122-128). Extinction coefficients were available for empty capsids and intact vector peaks. C(S) values ​​were determined using the SEDFIT algorithm described in Schuck (2000) Biophys.J., 78: pp. 1606-19. The molar concentrations of both intact vectors and empty capsids were calculated using Baer's law, and the percentage of complete capsids was calculated from these values. The values ​​are reported as a percentage of complete capsids.

[0135] Generation of AUC standard curve Since it is impossible to experimentally determine the extinction coefficient of fragmented genomes of unknown size and sequence, we established a relationship between the S value and genome size. To achieve this, rAAV vector preparations containing capsidized viral genomes with known nucleotide sizes were analyzed by AUC, and the corresponding S values ​​were determined as described above.

[0136] rAAV production by transient transfection Recombinant AAV vectors were produced by a triple transfection method using a "gutless" vector technique (Xiao, X et al., 1998, J. Virol. 3: pp. 2224-2232). In this method, the rep and cap genes were replaced with therapeutic genes, and their regulatory elements were sandwiched between 5' and 3' terminal inversion sequences (ITRs). The rep and cap genes were provided trans into separate plasmids, and a third plasmid contributed the required adenovirus helper gene. Without being constrained by theory, the viral capsid fully assembled, and then the ITR-flanked vector genome, via the capsid pore... It was assumed that the genome would be inserted into the capsid (Myers & Carter, 1980, Virology, 102: pp. 71-82). The resulting population of capsids contained both genome-free capsids (empty capsids) and genome-containing capsids.

[0137] rAAV production by the production cell platform AAV-producing cell lines are an alternative production platform used to generate clinical-grade rAAV vectors. In this method, HeLa S3 cells adapted for culture in suspension were manipulated to possess a vector sequence and selectable markers, as well as copies incorporating the AAV rep and cap genes necessary for vector replication and packaging (see, e.g., Puro:Thorne et al., (2009) Hum. Gene Ther., 20:707-14). Infection with WT adenovirus, which provides the necessary helper functions for replication, resulted in the production of recombinant AAV vector and adenovirus, which were removed during subsequent purification steps using ion-exchange chromatography.

[0138] Other methods The synthetic transgene was cloned into a plasmid containing an optimal promoter and a bovine growth hormone polyadenylation signal sequence (polyA). The entire transgene expression cassette was then cloned into the previral plasmid vector pAAVDC64 containing an AAV2 terminal inversion sequence. The total size of the resulting AAV genome in each expression plasmid (including regions with ITRs at the ends) was 4–4.6 kb. Recombinant vectors were produced by triple transfection of 293 cells using a helper plasmid expressing the rep2 / cap sequence and adenovirus helper function, the pAd helper (Stratagene, La Jolla, CA USA). The rep / cap helper expressed rep derived from AAV serotype 2, while the cap sequence encoded one of the following sequences: AAV cap 1, 2, 5, 9, or rh8R. The vector was purified by affinity chromatography, and in some cases further purified to remove empty particles (see, for example, Qu et al., (2007) J. Virol. Methods. 140: pp. 183-192).

[0139] result Analytical ultracentrifugation (AUC) using classical boundary sedimentation velocity was used to reveal particle heterogeneity of recombinant adeno-associated virus (rAAV) vector preparations. A mixture containing 20% ​​rAAV2 particles with complete genome and 80% empty capsid was prepared by mixing purified empty capsid with purified genome-containing capsid in a specified ratio. Empty capsid and complete capsid were produced by CsCl2 gradient purification of the empty capsid and complete capsid mixture after triple transfection production. To monitor the migration of rAAV2 particles in response to centrifugal force, the absorbance of this mixture of rAAV2 capsid was scanned at 260 nm along the centrifugation field at specified time intervals. Figure 1A shows a representative scan profile of the AAV2 mixture after centrifugation at 20,000 rpm for 1.2 hours (until the minimum settling species cleared the optical window). The scans represented the acquisition of concentration data as a function of radius r at time t to obtain a series of concentration scans revealing the complete migration pattern of constituent vector particles in the rAAV2 vector preparation. In these sigmoid curves or boundaries, the leading edge of the curve represents species that settle faster (i.e., genome-containing rAAV2 capsids), and the trailing edge represents species that settle slower (i.e., "empty" rAAV2 capsids) (Figure 1A).

[0140] Plotting the differential sedimentation coefficient distribution value, C(S), against the sedimentation coefficient (Svedberg units, S) yielded distinct peaks with unique sedimentation coefficients for both empty capsids and genome-containing capsid species (Figure 1B). The C(S) values ​​were obtained using the SEDFIT algorithm described in Schuck (2000) Biophys.J., 78:1606-19. The determination was made as follows: Extinction coefficients were used according to Table 2 to calculate the molar concentration and percentage values ​​for each capsid species from the absorbance data.

[0141] [Table 3]

[0142] Empty capsid (ε260 / カプシド= 3.72e6) and genome-containing capsids (ε 260 / カプシド= 3e7) Molar extinction coefficients for both were calculated using genome size and a published formula (Sommer et al., 2003). The molar concentrations of both genome-containing and empty capsids were then calculated using Baer's law. The various molar concentrations were used to calculate their relative presence, expressed as a percentage of all capsids (Figure 1). These results demonstrate that AUC can be used to accurately distinguish and quantify empty capsids and genome-containing capsids from heterogeneous vector preparations.

[0143] For the rAAV2 vector preparation shown in Figure 1, capsids containing the complete genome were represented by a peak settling at 94S and accounted for 21% of the vector preparation. Empty capsids settled at an S value of 64S and accounted for 79% of the vector preparation. These settling coefficient values ​​were confirmed by AUC analysis of pure populations of empty (Figure 2A) or genome-containing particles (Figure 2B). The AUC profile of the pure population of rAAV2 empty capsids revealed a single peak with a settling coefficient of 64S, while the AUC profile of the pure population of rAAV2 AUC genome-containing capsids revealed a single peak with a higher settling coefficient of 94S. These results are consistent with values ​​obtained from heterogeneous preparations and further confirm that the AUC method can be used to quantify genome-containing and empty AAV capsids from heterogeneous preparations containing both species.

[0144] The reproducibility of the AUC method was further evaluated by performing five independent AUC runs on the same vector sample (scAAV2 / 9LP2), as shown in Table 3. The sedimentation coefficients for both genome-containing capsids and empty AAV2 capsids were highly reproducible, yielding coefficients with a variation of 0.5–0.6%. Similarly, the relative presence of genome-containing capsids (expressed as a percentage of the total) was determined with a variation of approximately 2% in the coefficient. These results demonstrate that the AUC method for quantifying genome-containing capsids and empty AAV2 capsids is highly reproducible and yields consistent values.

[0145] [Table 4] [Examples]

[0146] Comparison of Interference and Absorbance Detection Methods for AUC We also evaluated Rayleigh interference optics, an alternative optical detection method for AUC. This method measures sample concentration based on the refractive index difference between a reference solution and an AAV-containing sample. Similar to absorbance detection, interference detection can be applied to any rAAV regardless of the genome sequence. Unlike absorbance detection, it requires a extinction coefficient, and interference detection yields an integrated peak that is directly proportional to the concentration.

[0147] Pure populations of empty capsids and genome-containing AAV2 capsids were mixed in a 1:1 ratio and analyzed by AUC using both interference (Figure 3A) and absorbance detection (Figure 3B). Interference detection revealed two populations of AAV capsids in approximately the predicted ratio of 43% empty and 57% genome-containing (Figure 3A). Both detection methods yielded similar abundance ratios. However, a comparison of the peak sizes generated by both methods illustrated a discrepancy between peak height and concentration in absorbance detection (compare the sizes of the “empty capsid” peaks in Figures 3A and 3B). The data generated by both methods are compared in Table 4. Ratio of interference signal to absorbance signal (A 260nmThe / IF function can be used in a format similar to the 260 / 280 ratio of absorbance data, which helped identify peaks in the C(S) distribution.

[0148] [Table 5]

[0149] While interference optics offers precision and resolution, it requires high-concentration samples. Furthermore, interference optics can be affected by mismatches between the reference and AAV sample buffers. However, AAV samples typically contain low protein concentrations, and a perfect match between the AAV sample and the reference buffer may be necessary. [Examples]

[0150] Influence of production method on AAV vector heterogeneity Previous examples have demonstrated that the AUC method is a highly accurate and reproducible method for degrading and quantifying empty capsids and genome-containing AAV capsids from heterogeneous mixtures. This capability may be advantageous for various applications in assessing the quality of AAV vector preparations. For example, a major problem in producing pure AAV vector preparations is the presence of capsids containing partial or fragmented genomes. To illustrate the usefulness of the AUC method for degrading these species, AAV vectors produced by two different methods, named the “triple transfection” and “producing cell line” methods, were analyzed by AUC.

[0151] AAV2 vectors containing transgene 2 were produced using either the triple transfection method (Figure 4) or the production cell line method (Figure 5). See Example 1 for descriptions of these methods. Following chromatographic purification, both vector preparations were analyzed by AUC. Figure 6A shows a schematic diagram of this AAV2 vector genome.

[0152] The AUC profiles of the vector preparations produced by these methods were distinctly different. Using the cell line production method, 74% of the capsids contained a complete genome and were represented by the 92S species (Figure 6B). 19% were empty capsids, and the remainder contained fragmented genomes (75S species, 7%). In contrast, 82% of the capsids produced by the triple transfection method were empty, 64S species (Figure 6C), 11% contained fragmented genomes (76S and 84S species), and only 8% had a complete genome (94S).

[0153] These results demonstrate that vector preparations produced using the production cell line technology are of high quality and primarily contain capsids with complete genomes. The majority of capsids produced by the triple transfection method were empty, with a greater proportion containing fragmented genomes. These results also highlight the ability of AUC to degrade capsids containing fragmented genomes, in addition to complete genome-containing and empty capsids. Furthermore, they exemplify the ability of AUC to assess the quality and uniformity of vector preparations. [Examples]

[0154] Use of AUC to evaluate the removal of empty capsids from vector preparations The AUC method was evaluated as a tool for monitoring the removal of empty capsids using chromatography (see Qu et al., (2007) J. Virol. Methods, 140: pp. 183-92 for details on the method). Separation of empty capsids from genome-containing capsids was performed using anion exchange chromatography (Figure 7A). AUC was performed on the degradation peak to demonstrate that genome-containing rAAV2 particles were concentrated in the fraction after elution from the resin ("Complete Genome Capsid" in Figure 7A).

[0155] As shown in Figure 7B, AUC analysis revealed that genome-containing capsids accounted for 94% of the vector preparation upon elution from the column. This subsequent fraction yielded a single peak with a sedimentation coefficient of 92S. In contrast, the rAAV2 vector preparation before the chromatography step (Figure 7C) had a considerable level of empty capsids. AUC analysis revealed two peaks with S values ​​of 63 and 93, with a 63S peak (empty capsids) corresponding to 52% of the total capsid population. These results demonstrate that chromatography is highly effective in removing empty capsids from AAV vector preparations. Importantly, they suggest that applying the AUC method to assess vector quality in purification is beneficial. The usefulness of this method has been demonstrated. The AUC method is a useful tool for evaluating various vector purification protocols or techniques. [Examples]

[0156] Evaluation of viral genome integrity using AUC method As illustrated in Example 3, AAV vector preparations may contain a capsid packaged with a fragmented genome in addition to the complete genome and empty seed. The main problem in producing homogeneous AAV preparations for therapeutic and research applications is the presence of a capsid containing a fragmented genome, which can result in abnormal expression or absence of expression of the target transgene. In fact, heterogeneity in AAV vector preparations has been reported to arise from the packaging of fragmented genome or AAV capsidized DNA impurities (Kapranov et al., (2012) Hum. Gene Ther., 23: pp. 46-55). Therefore, AUC was evaluated as a tool to quantify abnormal packaging of fragmented genome in rAAV vector preparations.

[0157] Since it is impossible to experimentally determine the extinction coefficient of fragmented genomes of unknown size and sequence, we established a relationship between the S value and genome size. To achieve this, rAAV vector preparations containing capsidized viral genomes of known sizes were analyzed by AUC, and their corresponding S values ​​were determined as shown in Table 5. A standard curve was then created to correlate genome size and S value (Figure 8). This shows a very linear correlation (R) between the sedimentation coefficient and genome size. 2 This demonstrated that (=0.9978).

[0158] [Table 6]

[0159] To demonstrate the usefulness of AUC for detecting genome fragments, a self-complementary vector containing AAV2 ITRS, a minimal CBA promoter, and an EGFP transgene were packaged in an AAV9 capsid (AAV2 / 9minCBAEGFP; see schematic diagram in Figure 9A). Vector particles were purified to eliminate empty capsids and analyzed by AUC. A standard curve was then used to assign genome size to each degraded genome containing a capsid. Approximately 25% of the vector preparations settled as 101S species, corresponding to a capsidized genome of approximately 4.3 kb (Figure 9A). This 101S peak corresponds to a double-stranded dimeric vector genome with a predicted size of approximately 4.3 kb. However, the majority (75%) of the vector preparations settled with an S value of 82, corresponding to a vector genome size of approximately 2 kb (Figure 9A), and consistent with single-stranded monomer packaging. Packaging of monomer genomes containing self-complementary vectors is well documented and is often the result of accidental terminal degradation at pseudo-trs-like sequences, despite the presence of ITRs with mutant D sequences (McCarty et al., (2001) Gene Ther., 8:1248-54).

[0160] Figure 9B shows an alkaline Southern blot of the same vector, scAAV9 EGFP, revealing two vector populations containing genomes of approximately 4.3 kb and 2 kb in size, demonstrating the AUC data in Figure 9A. The Southern blot also confirmed that monomeric viral genomes were preferentially packaged over dimeric genomes. Interestingly, AUC analysis of the single-stranded AAV9 EGFP vector (approximately 4 kb) had an S measurement of 99 S, revealing a single major peak corresponding to approximately 4.1 kb and 84% capsid presence by the standard curve (Figure 9C). These results suggest that single-stranded AAV vectors are packaged in a more uniform manner than double-stranded vectors. Again, consistent with the AUC method, Southern blot analysis of this vector preparation revealed uniform capsidation of the viral genome of the predicted size of approximately 4 kb (lane 2, Figure 9B). These results demonstrate that the AUC method can be used to measure the size of AAV vector genomes, yielding genome size data consistent with standard Southern blotting techniques. Using the AUC method, we found that single-stranded AAV vectors produce more uniform vector preparations than double-stranded ones. These results indicate that the AUC method is a powerful tool for identifying and quantifying capsid species containing incomplete genomes derived from vector preparations. [Examples]

[0161] Using AUC to evaluate factors that influence vector genome packaging Next, we used the AUC method as a tool to identify factors that influence the packaging of intact AAV vector genomes.

[0162] As discussed in Example 3, transient transfection-based production of rAAV vectors requires the use of three plasmids, including a rep / cap helper, an ITR vector plasmid, and a pAd helper (see Figure 4). AUC was used to evaluate the effect of the rep / cap helper on vector genome packaging for both single-stranded and self-complementary AAV vectors. First, a self-complementary AAV vector with an EGFP transgene (Figure 10A) was produced using one of two methods. In the first method (Figure 10B), a helper plasmid was used in which rep78 / 68 expression was driven by an endogenous p5 promoter ("WT Rep" construct). In the second method (Figure 10C), the helper was modified to reduce 78 / 68 expression by moving the p5 promoter downstream of the cap2 sequence and introducing mutations into the TATA box ("pHLP Rep" construct). The complete scAAV2 EFGP capsid was predicted to have a sedimentation coefficient of 100S in the dimeric genome morphology and 80S in the monomeric genome morphology (Figure 10A).

[0163] AUC analysis of these scAAV2EGFP vector preparations revealed significant differences in vector genome packaging. In the presence of reduced rep78 / 68 (pHLP), more than half (55%) of the vector preparations contained a dimeric genome, represented by 100S (Figure 10C). This was the predicted sedimentation coefficient for capsids containing a 4.4kb dimeric genome. In contrast, scAAV2EGFP preparations produced with complete complementation of rep78 / 68 had significantly less dimeric genome packaging (26%), with the majority of the capsid containing monomeric genome and sedimenting at 80S (Figure 10B). These results suggest shifting the P5 promoter of the helper plasmid and rep78 We revealed significant differences in genome packaging induced by a reduction in / 68 protein levels.

[0164] Single-stranded AAV5 factor IX vector, AAV5FIX (Figures 11A-B), and single-stranded AAV5hSMN vector (Figures 11C-D) were generated using a rep / cap helper in which the AAV2 cap sequence was replaced by the AAV5 cap sequence, although they differed in rep expression as described above. Based on the nucleotide size (4.3kb) of the FIX expression cassette, the predicted sedimentation coefficient for the AAV5 FIX vector capsid was approximately 10¹S. AUC analysis of AAV2 / 5FIX in the presence of reduced rep78 / 68 ("pHLP19 Rep") revealed a uniform profile in which most vectors (90%) settled at the predicted size, with an S value of approximately 10¹S (Figure 11A). In contrast, AAV5 FIX vectors generated using a rep / cap5 helper expressing wild-type levels of rep78 / 68 protein ("WT Rep") yielded a significantly different AUC profile (Figure 11B). Instead of a major peak at 101S, this profile revealed further capsid heterogeneity, with the majority of AAV5 FIX (80%) settling at a low S value of 86S, which is thought to represent the packaging of the fragmented genome. Furthermore, in this vector sample, only 15% of the AAV5 FIX vector capsids settled at the correct S value of approximately 104S (Figure 11B).

[0165] AAV5 SMN vectors constructed using these same wild-type and mutant p5rep / cap helpers also exhibited remarkably different AUC profiles. As seen in the single-stranded AAV5FIX vector, AAV5 SMN vectors generated in the presence of reduced rep78 / 68 showed less heterogeneity by AUC analysis, exhibiting a single capsid species precipitation with an S value of 101S, consistent with packaging a genome of approximately 4.4kb (Figure 11C). In contrast, AUC profiles for the same vector genome packaged with “wild-type” levels of rep78 / 68 protein revealed three distinct AAV vector species with precipitation coefficients of 100S (predicted S value for a complete vector genome of 4400nt), 92S (corresponding to a fragmented genome of approximately 3300nt), and 80S (corresponding to a fragmented genome of 2000 nucleotides) (Figure 11D). These results confirmed significant differences in genome packaging induced by a helper plasmid P5 promoter shift using two additional AAV vectors.

[0166] Further analysis of AAV5SMN and AAV5FIX vector preparations was performed by Southern blotting of the vector DNA. Consistent with AUC, Southern blotting of AAV5SMN produced at wild-type rep78 / 68 protein levels revealed packaging of full-length (4.4kb) and fragmented (<4.4kb) SMN genomes (Figure 12A, lane 1). In contrast, AAV5SMN vectors produced in the presence of reduced rep78 / 68 protein contained many capsids including the full-length SMN genome (Figure 12A, lane 2). Interestingly, comparison of the two AAV5FIX vector preparations by Southern blotting revealed the presence of the full-length FIX genome even when the vector was produced in the presence of wild-type rep78 / 68 (Figure 12B, lane 2). However, AUC analysis of this AAV5FIX vector (Figure 11B) showed that 80% of the capsids contained fragmented genomes (approximately 3000 nucleotides) that were not detected by the FIX probe.

[0167] Further analysis of the FIX vector preparations generated under two experimental conditions was performed by generating probes to separate regions of the vector plasmid, including the skeletal region. A map of this vector is provided in Figure 13. Figure 14 shows Southern blot analysis using these probes to compare these FIX vector preparations generated under different conditions. This is shown. As shown in Figure 14A, both vector preparations (pHLPrep, lane 1; WTrep, lane 2) contained the hFIX transgene. However, Figure 14B lane 2 confirms that the vector genome species (Figure 11B) that settled at 86S in the WT Rep preparation was approximately a 3kb fragment. Furthermore, this species is Amp R It reacted with a specific probe (Figure 14B, lane 2), suggesting that packaging upstream of the 5'ITR occurred in a rep-dependent manner. In contrast, Amp was present in rAAV5 FIX vector preparations generated in the presence of reduced levels of rep68 / 78. R No evidence of contained fragments was found (Figure 14B, lane 1).

[0168] Amp R The results were also evaluated by Q-PCR using primers and probes specific to the target. Q-PCR showed approximately 35% amplification. R Titer was detected in AAV5FIX vector preparations generated in the presence of "wt" rep, in contrast to less than 1% when the same vector plasmid was used to generate the AAV5FIX vector in the presence of reduced rep68 / 78 (data not described). These results further highlight the usefulness of AUC analysis in revealing the presence of packaged genomes that are not detected by gene-specific Southern blot analysis.

[0169] The packaging capabilities of AAV vectors have been extensively studied, and numerous reports demonstrate good transduction with vectors packaging large AAV genomes, although these have later been shown to be fragmented into subgenome-length DNA. To further explore the applicability of the AUC method, we evaluated the heterogeneity of AAV vectors produced using large genomes. Expression cassettes containing a full-length CBA promoter driving the expression of a β-phosphate diesterase transgene were packaged as a 5.4kb large genome (Figure 15A) or as a 4.6kb wild-type-sized genome (Figure 15B). To generate the 4.6kb genome, the CBA promoter was cleaved by reducing the size of introns as previously reported (Gray, SJ et al., (2011) Hum. Gene Ther. 22(9):1143-1153).

[0170] As shown in Figure 15A, the AUC profiles of AAV vector preparations generated using large vector genomes demonstrated that approximately half of the preparations settled as a 93S species, consistent with packaging a fragmented vector genome of approximately 3.5kb. Thirty percent of the preparations corresponded to another subgenome vector species of approximately 4.9kb, settling at 105S. There was no evidence of packaging for the full-length 5.4kb genome, which was predicted to settle at 108–109S. In contrast, AUC analysis revealed that under the control of a shortened CBA promoter, the same transgene primarily settled as a 102S vector species, consistent with packaging for the predicted full-length vector genome of 4.6kb (Figure 15B). These results demonstrate the usefulness of AUC analysis in profiling AAV vectors with large genomes, and that this profiling is essential given the observed event of genome fragmentation.

[0171] This embodiment demonstrates the highly effective AUC method for analyzing the genomic size of AAV vector capsids in heterogeneous preparations. By degrading the genomic-containing capsid by size (e.g., dimeric and monomeric genomes, or partial fragments thereof), the AUC method becomes a powerful tool for assaying the quality of AAV vector preparations produced under various conditions. Furthermore, results from three different vector systems demonstrate that the AUC method is broadly useful for quality control and condition optimization to obtain improved AAV vector preparations. Importantly, the AUC method can detect fragmented genomes that cannot be detected by Southern blotting. While Southern blotting relies on the presence of DNA probe sequences for detection, the AUC method is sequence-independent. It was demonstrated that AUC is an effective tool for analyzing AAV genomes of a specific type. Overall, these results demonstrate a highly advantageous and effective implementation of the AUC method for analyzing multiple types of AAV vector preparations that were found to exhibit dramatically variable effects on genome packaging. [Examples]

[0172] Characterization of recombinant adenovirus vector preparations by analytical ultracentrifugation. Adenovirus (Ad) vectors possess properties that make them attractive as vectors for gene therapy. The production of Ad vector products requires analytical methods to monitor product quality in terms of uniformity, purity, and integrity of the manufacturing process. To address this requirement, we investigated the potential use of analytical ultracentrifugation (AUC) as a technique for characterizing the uniformity of Ad vectors.

[0173] method Sample preparation To support accurate AUC assessment, recombinant adenovirus serotype 2 vector (Ad2) was prepared and highly purified by CsCl gradient ultrafiltration to concentrate genome-containing particles. Product concentrations were measured by spectroscopic analysis at 260 nm (OD). 260The optical density was determined by optical density measurement at 260 nm. Sample preparation was performed to a target concentration of 0.1 to 1.0 by optical density measurement at 260 nm, either by direct dilution in PBS or by further concentration using an Amicon Ultra-0.5 / 30K MWCO centrifugal filter device, in order to generate reproducible and consistent AUC data.

[0174] Acquisition of seismic velocity AUC data Ultracentrifugation velocity (SV-AUC) analysis was performed using a ProteomeLab® XL-I (Beckman Coulter). 400 μL of sample was loaded into the sample sector of two sector velocity cells, and 400 μL of PBS was loaded into the corresponding reference sector. The sample was placed in a 4-well rotor and equilibrated in the instrument to a temperature of 20°C, maintaining high vacuum for 1 hour. Sedimentation velocity centrifugation was performed at 6,000 rpm, 20°C, and a 0.003 cm radius step setting, without delay or repetition. Rayleigh interference optics was used to simultaneously record radial concentrations as a function of time until the smallest settling component cleared the optical window (1.2 hours). Assay volume was limited to a single sample per run based on absorbance scan recovery times longer than 1 minute, as well as the large size and rapid sedimentation of Ad2.

[0175] AUC Data Analysis The percentage of complete capsid was determined by analyzing approximately 75 scans using interferometric detection with a SEDFIT (see the worldwide web at NIH / analyticalultracentrifugation.com) continuous-size C(S) distribution model. nd Differential regularization was applied to the fitting at a confidence level of F-statistics / ratio = 0.68. The following C(S) parameters were kept constant: resolution = 250S, Smin = 10, Smax = 1500, and friction ratio = 1.86935. RI and TI noise subtraction were applied, and the meniscus position was shifted, allowing the software to select the optimal position. This model fits the data to the Lamb equation, and the resulting size distribution is in units of fringe or OD.260 The distribution of the settling coefficient was like a chromatogram, with each peak having an area under it proportional to the concentration in units. The settling coefficient (in Svedberg units) and relative concentration (in OD units) were determined for each component in the distribution. Each AUC run was an independent assay, and each analysis was monitored for subsequent characteristics to ensure the quality of the results: goodness of fit (rmsd), OD for each peak. 260nm / Ratio of interference signals in the fringe (A260 / IF ratio), agreement of various sedimentation coefficients between runs The overall quality of the scan. For this representative example, the rmsd score was 0.006584.

[0176] result Analytical ultracentrifugation (AUC) using classical boundary sedimentation velocity was used to reveal particle heterogeneity in recombinant adenovirus serotype 2 vector (rAd2) vector preparations. To monitor the migration of rAd2 particles in response to centrifugal force, this mixture of rAd2 capsids was scanned along the centrifugation field at specified time intervals using interferometric optics. The scans represented the acquisition of concentration data as a function of radius r at time t to obtain a series of concentration scans revealing the complete migration pattern of constituent vector particles in the rAd2 vector preparation. Plotting the differential sedimentation coefficient distribution value, C(S), against the sedimentation coefficient (Svedberg units, S) yielded different peaks containing rAd2 species with intrinsic sedimentation coefficients (Figure 16). C(S) values ​​were determined using the SEDFIT algorithm described in Schuck (2000) Biophys.J., 78:1606-19.

[0177] Regarding the rAd2 vector preparations shown in Figure 16, 87.8% of the rAd2 vector preparations that precipitated with an S value of 731 were consistent with vector preparations mainly composed of genomes containing capsids. This data confirms that adenovirus particles can be degraded by AUC.

Claims

1. A method for determining the molar concentrations of various individual virus particles in a heterogeneous mixture of virus particles containing a recombinant adeno-associated virus (rAAV) vector capsidized within a viral capsid, a) A step of subjecting a heterogeneous mixture of virus particles to analytical ultracentrifugation under boundary sedimentation velocity conditions to generate a sedimentation boundary, wherein the analytical ultracentrifugation is performed at approximately 3,000 rpm to approximately 20,000 rpm. b) A step of measuring the rate of movement or migration of a sedimentation boundary, wherein the movement or migration of various individual virus particles in a heterogeneous mixture of virus particles produces different sedimentation boundaries, each different sedimentation boundary corresponds to a species of degradable virus particle, and the heterogeneous mixture of virus particles includes a complete genome, a fragmented genome, and an empty capsid that does not contain a genome. c) A step of determining the sedimentation coefficient of individual virus particles in a heterogeneous mixture of virus particles, d) A step of quantifying the molar concentration of various individual virus particles in a heterogeneous mixture of virus particles, The method, including the method described above.

2. The method according to claim 1, further comprising the step of determining the size of the various individual virus particles by comparing the sedimentation coefficient of the various individual virus particles with a standard curve created from the sedimentation coefficients of virus particles containing recombinant AAV genomes of known nucleotide sizes, after step (c) and before step (d).

3. The method according to claim 1 or 2, wherein the analytical ultracentrifugation method is performed at approximately 10,000 rpm to approximately 20,000 rpm or approximately 15,000 rpm to approximately 20,000 rpm.

4. The method according to any one of claims 1 to 3, wherein boundary sedimentation is performed at a temperature of approximately 4°C.

5. The method according to any one of claims 1 to 4, wherein centrifugal separation is performed for approximately 0.5 hours to approximately 2 hours.

6. The total concentration of virus particles in the heterogeneous mixture before step (a) is 5 x 10 11 The method according to any one of claims 1 to 5, wherein the concentration is higher than vg / mL.

7. The total concentration of virus particles in the heterogeneous mixture before step (a) is approximately 1 x 10 11 vg / mL to approximately 1 x 10 13 The method according to any one of claims 1 to 5, wherein the concentration is vg / mL.

8. The method according to any one of claims 1 to 7, wherein the virus particle contains the AAV9 capsid protein.

9. The method according to claim 8, wherein the virus particle comprises at least one AAV2 ITR.

10. The method according to any one of claims 1 to 9, further comprising the step of determining the relative ratio of various individual virus particles in the heterogeneous mixture of virus particles.

11. A method for determining the size of one or more fragmented genomes in a preparation of viral particles containing a recombinant adeno-associated virus (rAAV) vector capsidized within a viral capsid, a) A step of subjecting the preparation to analytical ultracentrifugation under boundary sedimentation velocity conditions to generate one or more sedimentation boundaries, wherein the analytical ultracentrifugation is performed at approximately 3,000 rpm to approximately 20,000 rpm. b) A step of measuring the rate of movement or migration of one or more sedimentation boundaries, wherein the movement or migration of the virus particles gives rise to different sedimentation boundaries, each of which corresponds to a degradable virus particle, and one or more virus particles contain a fragmented genome; and a step of determining the sedimentation coefficient of the virus particles containing one or more fragmented genomes in the preparation; c) A step of determining the size of one or more fragmented genomes as a function of the sedimentation coefficient of the viral particle containing the one or more fragmented genomes, The method, including the method described above.

12. The method according to claim 11, wherein step (c) comprises determining the size of one or more fragmented genomes by comparing the sedimentation coefficient of one or more viral particles containing the fragmented genome with a standard curve created from the sedimentation coefficients of viral particles containing recombinant AAV genomes of known nucleotide sizes.

13. The method according to claim 11 or 12, wherein the fragmented genome is a subgenome DNA molecule.

14. The method according to any one of claims 11 to 13, comprising the step of determining the size of all fragmented genomes in the preparation.

15. The method according to claim 14, further comprising the step of quantifying all of the aforementioned fragmented genome.

16. The method according to any one of claims 11 to 15, wherein the virus particle contains the AAV9 capsid protein.

17. The method according to claim 16, wherein the virus particle comprises at least one AAV2 ITR.

18. A method for quantifying individual variant virus particles of one or more species containing fragmented recombinant adeno-associated virus (rAAV) genomes in a heterogeneous mixture of virus particles, a) A step of subjecting a heterogeneous mixture of virus particles to analytical ultracentrifugation under boundary sedimentation velocity conditions to generate a sedimentation boundary, wherein the analytical ultracentrifugation is performed at approximately 3,000 rpm to approximately 20,000 rpm. b) A step of measuring the rate of movement or migration of a sedimentation boundary, wherein the movement or migration of various individual virus particles in a heterogeneous mixture of virus particles produces different sedimentation boundaries, each different sedimentation boundary corresponds to a species of degradable virus particle, and the individual virus particles include empty particles that do not contain a genome, particles that contain a complete genome, and particles that contain a fragmented genome. c) A step of determining the genome size of one or more individual variant virus particles of a species in a heterogeneous mixture of virus particles, and d) A step of determining the amount of one or more species of variant virus particles in the heterogeneous mixture of virus particles, The method, including the method described above.

19. The method according to claim 18, wherein various individual variant virus particles in the heterogeneous mixture of virus particles are completely decomposed.

20. The method according to claim 18 or 19, wherein the step of determining the genome size of individual variant virus particles of one or more species in the heterogeneous mixture of virus particles includes comparing the sedimentation coefficient of the variant virus particle with a standard curve created from the sedimentation coefficients of virus particles containing recombinant AAV genomes of known nucleotide sizes.

21. A method for quantifying rAAV particles having a complete viral genome in a heterogeneous mixture of recombinant adeno-associated virus (rAAV) particles, wherein the heterogeneous mixture includes multiple types of rAAV virus particles, including rAAV virus particles having a complete viral genome, rAAV virus particles having a fragmented genome, and rAAV virus particles without a genome, and the method is (i) A step of generating a sedimentation boundary by subjecting the heterogeneous mixture of rAAV virus particles to analytical ultracentrifugation under boundary sedimentation velocity conditions. (ii) A step of measuring the rate of movement or migration of the sedimentation boundary, wherein the movement or migration of each rAAV virus particle species in the heterogeneous mixture of rAAV virus particles produces different sedimentation boundaries, and each different sedimentation boundary corresponds to a degradable rAAV virus particle species. (iii) A step of determining the genome size of the degradable rAAV virus particle species in a heterogeneous mixture of rAAV virus particles, and (iv) A step of determining the amount of rAAV virus particles having a complete viral genome in a heterogeneous mixture of rAAV virus particles, The method, including the method described above.

22. The method according to claim 21, further comprising step (iii) comparing the sedimentation coefficient of the rAAV virus particles with a standard curve prepared from the sedimentation coefficients of rAAV virus particles having an rAAV genome of known nucleotide size.

23. The method according to claim 21 or 22, wherein the analytical ultracentrifugation method is performed at approximately 3,000 rpm to approximately 20,000 rpm.

24. A method for determining the percentage of rAAV virus particles having a complete viral genome in a heterogeneous mixture of recombinant adeno-associated virus (rAAV) particles, wherein the heterogeneous mixture is The method includes multiple types of rAAV virus particles, including rAAV virus particles having a complete viral genome, rAAV virus particles having a fragmented genome, and rAAV virus particles without a genome, and the method is (i) A step of generating a sedimentation boundary by subjecting the heterogeneous mixture of rAAV virus particles to analytical ultracentrifugation under boundary sedimentation velocity conditions. (ii) A step of measuring the rate of movement or migration of the sedimentation boundary, wherein the movement or migration of each rAAV virus particle species in the heterogeneous mixture of rAAV virus particles produces different sedimentation boundaries, and each different sedimentation boundary corresponds to a degradable rAAV virus particle species. (iii) A step of determining the molar concentration of the degradable species in the heterogeneous mixture of rAAV virus particles, and (iv) A step of determining the percentage of rAAV virus particles having a complete viral genome in a heterogeneous mixture of rAAV virus particles, The method, including the method described above.

25. The method according to claim 24, further comprising the step of determining the genome size of each of the degradable rAAV virus particle species after step (ii) and before step (iii).

26. The method according to claim 24 or 25, wherein the analytical ultracentrifugation method is performed at approximately 3,000 rpm to approximately 20,000 rpm.

Images