Removal of contaminating viruses from AAV preparations

Abstract

The present invention relates to a separation of viruses of an essentially spherical shape from viruses with a rod-like shape that are comprised in a sample, wherein the sample comprising the viruses is subjected to filtration.

Claims

1. A method for removing baculoviral virions from a sample comprising parvoviral virions and baculoviral virions, which method comprises filtering the sample through a filter that has a pore size of 30-40 nm, thereby removing the baculoviral virions from the sample.

2. The method according to claim 1, wherein the parvoviral virions are adeno-associated virus virions.

3. The method according to claim 1, wherein the baculoviral virions being removed are virions of Autographa californica multicapsid nucleopolyhedrovirus.

4. The method according to claim 1, wherein the filter is a virus filter or an ultrafilter.

5. The method according to claim 1, wherein the filter is a surface and/or a depth filter.

6. The method according to claim 1, wherein the filter comprises a material selected from the group consisting of: regenerated cellulose, polyvinylidene fluoride, polysulfone, polytetrafluoroethylene, polypropylene, modified polyethersulfone, unmodified polyethersulfone, cellulose acetate, cellulose nitrate, and polyamide.

7. The method according to claim 1, wherein the filter has a pore size of 32-38 nm.

8. The method according to claim 6, wherein the regenerated cellulose filter is a cuprammonium-regenerated cellulose filter.

9. The method according to claim 1, wherein the sample is filtered through two or more filters.

10. The method according to claim 1, wherein prior to the filtering step, the sample is prefiltered through a filter with a pore size of 70-200 nm.

11. The method according to claim 1, wherein the pH of the sample is 6 to 10.

12. The method according to claim 1, wherein 1 ml to 200 ml of the sample is filtered per cm.sup.2 of filter surface area.

13. The method according to claim 1, wherein the filtering results in: (a) recovery of at least 85% of the parvoviral virions; and (b) at least a 10.sup.5-fold reduction in baculoviral virion titer.

14. The method according to claim 8 wherein the cuprammonium-regenerated cellulose filter is a hollow fiber filter.

15. The method according to claim 6 wherein the filter is a hollow fiber membrane comprising cuprammonium regenerated cellulose, or a membrane comprising hydrophilic modified polyvinylidenefluoride.

16. The method according to claim 11 wherein the sample pH is 7 to 9.

17. The method according to claim 11 wherein the sample pH is 7.5 to 8.5.

18. The method according to claim 11 wherein the sample pH is about 8.0.

19. The method according to claim 12 wherein 80 to 120 ml of the sample are filtered per cm.sup.2 of the filter surface area.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: Recovery of GC and TP during virus filtration.

(2) The figure show the recovery as % input of genome copies and total particles of the filtrate including correction made for sample volumes, the membrane post use wash and the pool of the aforementioned aliquots.

(3) FIG. 2: Baculovirus reduction during virus filtration

(4) The figure shows the baculovirus titre as measured by TCID.sub.50 of the baculovirus spike material, the virus filter feed (spiked AIEX eluate), nanofiltrate as function of filtration volume (xmL) and the final pool of the virus filter filtrate and virus filter wash. The filtrates and the final pools are the maximum possible titres as the assay returned no well positive for baculovirus.

EXAMPLES

Example 1

(5) Adeno-associated viral vector (AAV) was produced following infection of insect cells with three recombinant baculoviruses as previously described by Urabe et al., 2002 (Hum. Gene Ther 13(16):1935-1943).

(6) Three days after infection the cell culture was detergent lysed and subsequently nuclease treated by the addition of 9 U/mL Benzonase (Merck) and incubation at manufacturer's recommendations.

(7) The crude lysed bulk was subsequently clarified by filtration with a Pall Profile® Star and Pall Supor® filters (Pall Corporation) in series. A viral reduction incubation in the presence of detergent was performed at least 28° C. This material was purified via affinity chromatography using AVB Sepharose HP, GE Healthcare. Briefly, the filtered cell lysate was applied to a 20 cm diameter column (BPG 200/500, GE Healthcare with approximately 6 cm bed height at a linear velocity of 150 cm/hr. The column was washed with phosphate buffered saline until the UV absorbance curve returned to baseline and stabilized. The adsorbed rAAV particles were eluted in acidic media (50 mM sodium citrate adjusted with HCl to pH 3.0) and the column eluate was adjusted immediately with 1/10 volume of 1M Tris-Cl (pH 8.0).

(8) To this neutralized eluate a 10% v/v Baculovirus spike, manufactured as previously described by Urabe et al. (2002; supra) after clarification via centrifugation at 1900 g for 15 minutes and filtration across a 0.2 μm bottle top filter (Corning), was added to provide a mixed population of Baculovirus and AAV for feed to the virus filter.

(9) Prior to use, a Planova 35N Filter was prepared according to the manufacture recommendations. Briefly, the storage solution was removed and all air purged by the addition of 40 ml of 60 mM Tris/HCL pH 8.0 and the flushing procedure was performed according to the manufacture recommendations.

(10) The feed flow rate to the virus filter was set at 5 ml/min as recommended by the manufacture (see Table 2) using a peristaltic pump. During virus filtration, the feed pressure and permeate flow were monitored (see Table 2).

(11) TABLE-US-00002 TABLE 2 Virus filtration performance Process Pump Feed volume Permeate time speed pressure processed flow (min) (rpm) (mBar) (mL) (mL/min) 0 3 160 0 5 10 3 160 50 5 20 3 160 100 5 30 3 160 150 5 40 3 160 200 5 50 3 160 250 5 60 3 160 300 5 70 3 160 350 5 80 3 160 400 5

(12) Samples were collected after 50, 250 and 400 ml filtrate were taken for Baculovirus titration via TCID.sub.50 assay for infectious baculovirus using Sf9 cell lines. Results are given in TCID50: 50% tissue culture infective dose, which indicates the amount of virus that is required to produce a cytopathic effect in 50% of inoculated tissue culture sf9 cells.

(13) Samples, circa 1 ml, were taken of the filtrate; these were analyzed on Genome Copies (henceforth referred to as GC) of the gene of interest in AAV, and Total Particle (henceforth referred to as TP) of AAV, to monitor their associated recoveries. GC and TP analysis was performed as described below.

(14) To determine the genome copy concentration of samples, ten-fold serial dilutions of test samples and a qualified working standard are prepared, extraneous DNA is removed by DNase digestion and the encapsidated DNA freed by proteinase K digestion. The released viral DNA is then purified using MagneSil Blue®. Subsequently, the DNA is amplified with quantitative-PCR (Q-PCR) using primers specific for the vector genome sequence. DNA amplification is monitored in real-time by inclusion of the fluorescent DNA binding dye SYBR green. The amount of DNA present in a sample can be calculated by comparing the Ct values found for the test sample with that found for the working standard. A parallel-line-assay design is used to test the serial dilutions of the test sample against the qualified working standard, which provides a ratio. This ratio is transformed to the genome copy concentration (gc/mL) using the working standard genome content.

(15) The total number of AAV particles (both the full particles containing a correct genome (vector particles) as well as empty particles) is determined using gel filtration chromatography. Gel filtration HPLC is performed using a BioBasic SEC-1000 (Thermo Electron Corporation) at a flow rate of 1 mL/min with an aqueous phase of D-PBS at 25° C. Elution is monitored with UV absorbance at 214 nm. The peak areas of the test samples are quantified using a calibration curve of a qualified working standard with a known amount of AAV particles per mL.

(16) The Planova 35N filter was washed using 10 ml of 60 mM Tris/HCl pH 8.0 as per manufactures instructions. The wash was pooled with the filtrate and sampled for baculovirus infectious titre via TCID.sub.50 assay. The wash was sampled for later GC and TP titre determination before being pooled with the filtrate material. The pool of the filtrate and wash were mixed by hand to ensure homogeneity and sampled for later GC and TP titre determination. From the TP and GC titres recovery were calculated and are displayed in FIG. 1

(17) The infectious baculovirus titer of samples is determined using a Tissue Culture Infectious Dose 50% (TCID50) assay. This method is based on the infection of monolayers of SD insect cells with infectious baculovirus in a test sample. A 3-fold serial dilution in culture medium of the test sample is used to infect the cells in octoplicate. The plate is incubated at +28° C. for 7 days. Subsequently, the supernatants are transferred to newly prepared plates and incubated at +28° C. for 7 days. As infection proceeds, the infected cells are not able to remain attached to each other and to the plate surface and form loose cells, i.e., show cytopathogenic effect (CPE), which can be observed microscopically. The titer in log 10 TCID50/mL is calculated using the Spearman-Karber method.

(18) Results

(19) The results for Baculovirus titration are in FIG. 2

(20) The spike solution has an infectious baculovirus titre of 8.1 log.sub.10 TCID.sub.50/mL. The spike solution is added at 10% in the spiked AIEX Eluate, of 7.62 log.sub.10 TCID.sub.50/mL. The TCID.sub.50 value of the Nanofiltrate declines with at least 6 logs, to below 1.511 log.sub.10 TCID.sub.50/mL, the level of quantification of this particular assay. Therefore nanofiltration gives at least a reduction in Baculovirus of 6 logs, therefore Nanofiltration is a feasible unit operation for the discrimination of two virus with comparable sizes in two dimensions.