Method for purifying viruses or virus-like particles using a crosslinked cellulose hydrate membrane

Abstract

The present invention relates to a method for purifying viruses or virus-like particles using a crosslinked cellulose hydrate membrane and to a kit for purifying viruses or virus-like particles and the use thereof.

Claims

1. A method for purifying viruses or virus-like particles from a solution, comprising the following steps: a) providing a mixture M composed of a first solution and a second solution, the first solution containing the viruses or the virus-like particles to be purified and impurities, and the second solution containing a polyalkylene glycol having two or three carbon atoms in the repeating unit, b) loading a crosslinked cellulose hydrate membrane with the mixture M from step a), with the viruses or the virus-like particles attaching to the outer and inner membrane surfaces and being retained, and c) eluting the viruses or virus-like particles retained in the crosslinked cellulose hydrate membrane using a solution which contains no polyalkylene glycol having two or three carbon atoms in the repeating unit or which contains at least 50% less polyalkylene glycol having two or three carbon atoms in the repeating unit than the mixture M from step a), with the result that the viruses or the virus-like particles are obtained in purified form in the eluate.

2. The method as claimed in claim 1, wherein an unpurified, clarified bioprocess solution is used as the first solution.

3. The method as claimed in claim 1, further comprising a step d) which is carried out between step b) and step c), in which step d) the loaded crosslinked cellulose hydrate membrane is washed with a solution which contains as much polyalkylene glycol having two or three carbon atoms in the repeating unit as the mixture M from step a).

4. The method as claimed in claim 1, further comprising a step e), in which step e) the crosslinked cellulose hydrate membrane is sanitized before the loading according to step b) by means of alkaline solution.

5. The method as claimed in claim 1, further comprising a step f), in which step f) the crosslinked cellulose hydrate membrane is equilibrated before the loading according to step b), in which it is flushed with a buffer-containing solution, which contains polyalkylene glycol having two or three carbon atoms in the repeating unit, until a constant conductivity and/or a constant UV signal is measured in the permeate.

6. The method as claimed in claim 1, wherein the first and second solutions, which contain the viruses or virus-like particles to be purified and impurities as well as polyalkylene glycol having two or three carbon atoms in the repeating unit, are mixed with the aid of a mixer.

7. The method as claimed in claim 1, wherein the contact time of the viruses or the virus-like particles with the second solution, which contains polyalkylene glycol having two or three carbon atoms in the repeating unit, is less than one minute before the loading step.

8. The method as claimed in claim 1, wherein the concentration of polyalkylene glycol having two or three carbon atoms in the repeating unit in the mixture M which is contacted with the crosslinked cellulose hydrate membrane is from 1 to 20% by weight, based on the total mass of the mixture M.

9. The method as claimed in claim 1, wherein the polyalkylene glycol having two or three carbon atoms in the repeating unit is selected from the group consisting of polyethylene glycol and polypropylene glycol or mixtures thereof.

10. The method as claimed in claim 1, wherein polyethylene glycol (PEG) having an average molar mass of from 600 to 10 000 daltons is selected.

11. The method as claimed in claim 1, wherein the recovery rate for purified viruses or virus-like particles in the eluate is more than 50%.

12. The method as claimed in claim 1, wherein the concentration of the viruses or the virus-like particles in the eluate is elevated by a factor of from 5 to 10 in comparison with the starting solution.

13. The method as claimed in claim 4, further comprising a step f), in which step f) the crosslinked cellulose hydrate membrane is equilibrated after the sanitization according to step e), in which it is flushed with a buffer-containing solution B, which contains polyalkylene glycol having two or three carbon atoms in the repeating unit, until a constant conductivity and/or a constant UV signal is measured in the permeate.

Description

(1) The present invention will be more particularly elucidated on the basis of the following nonlimiting examples and figures.

(2) FIG. 1 shows, by way of example, the course of the purification of phi X 174 from Exemplary Embodiment 1 comprising the steps of equilibration, loading, washing and elution.

(3) FIG. 2 shows the purification efficiency of the method according to the invention for phi X 174 as an example, measured using size-exclusion chromatography, by means of the chromatogram before and after the purification.

(4) FIG. 3 shows the dynamic binding capacity in the purification of bacteriophage phi X 174 from Exemplary Embodiment 3.

EXAMPLES

Exemplary Embodiment 1: Purification of Bacteriophage phi X 174 from the Clarified Process Solution

(5) The phages to be purified were first generated by infection of a bacteria culture (E. coli C600) in TSB buffer. After the bacteria had been separated off by means of filtration through two filter systems (SartopurePP2 & Sartopore 2, both from Sartorius Stedim Biotech GmbH), there was a phage concentration of 3.810.sup.11 particles/mL. This solution will be referred to below as phage solution.

(6) For the chromatographic purification, use was made of a Hydrosart membrane (Sartorius Stedim Biotech GmbH), i.e., a crosslinked cellulose hydrate membrane, having a pore size of from 3 to 5 m. A filtration unit, Vivapure system (SSB, Gttingen), containing the cellulose hydrate membrane and having a membrane volume (MV) of 0.023 mL was connected to an kta system (kta Prime, GE Healthcare Life Science). Over the experiment, the absolute flow rate was constantly 0.3 mL/min, which corresponds to a relative flow rate of approx. 10 MV/min. The membrane was first equilibrated with a mixture composed of buffer A (50 mM Tris (Merck Millipore), pH 7.0, 150 mM NaCl (Merck Millipore)) and buffer B (50 mM Tris, pH 7.0, 150 mM NaCl, 16% by weight of PEG 6000 (Mw=6000 g/mol, Carl Roth)) in the ratio of 1:4. The final concentration of PEG 6000 in the mixture was 12% by weight.

(7) For the loading, buffer A was replaced with the phage solution. The mixing operation was done in-line in a static mixer of the kta system. After altogether 55 mL of solution (13.7 mL of phage solution and 41.3 mL of buffer B) had been loaded, the membrane was washed. To this end, a mixture composed of buffer A and buffer B in the ratio of 1:4 was used again. The phage particles were eluted from the membrane by using 100% buffer A. The phage recovery was 81.6% (3%).

(8) The virus particles were quantified by means of quantitative real-time PCR on an instrument from Agilent Technologies (Mx3005P qPCR system) using the master mix Brilliant III Ultra-Fast SYBR Green QPCR from Agilent Technologies. The reference used was a plasmid bearing the sequence of the phi X 174 phage (Fermentas GmbH).

(9) The depletion of contaminants with simultaneous enrichment of phage particles was quantified by carrying out size-exclusion chromatography (SEC). To this end, a SEC column (Yarra 3000, Phenomenex) was connected to an HPLC system (UltiMate 3000, Thermo Scientific Dionex) and a 100 l sample was loaded in each case. The absolute flow rate was 1 ml/min and the composition of the running buffer was as follows: 100 mM sodium phosphate and 100 mM sodium sulfate (pH=6.6).

(10) FIG. 1 shows, by way of example, the course of the purification of phi X 174 from Exemplary Embodiment 1 comprising the steps of equilibration, loading, washing and elution.

(11) FIG. 2 shows the purification efficiency of the method for phi X 174 as an example, measured using size-exclusion chromatography, by means of the chromatogram before and after the purification. The impurities (large peak at 12 min) disappear completely and the chromatogram shows only the pure phage peak at approx. 6 min.

Exemplary Embodiment 2: Purification of Bacteriophage phi X 174 from the Clarified Process Solution with Relatively Short Contact Time

(12) The method was carried out in line with Exemplary Embodiment 1 on two further crosslinked cellulose hydrate membrane-containing filtration units which had a higher membrane volume of a) 0.046 mL and b) 0.56 mL compared to Exemplary Embodiment 1. As in Exemplary Embodiment 1, the relative flow rate was adjusted to approx. 10 MV/min in both cases. Owing to the altered membrane volumes compared to Exemplary Embodiment 1, the respective absolute flow rates changed from 0.3 mL/min in Exemplary Embodiment 1 to a) 0.6 mL/min and b) 5.6 mL/min. At the same time, the desired reduction in the contact time from 16.9 min in Exemplary Embodiment 1 to a) 8.3 min and b) 0.93 min appeared with the increase in the absolute flow rate. The experiments were carried out twice in both cases, and the results are depicted in Table 1.

(13) A reduction in the contact time from 16.9 min in Exemplary Embodiment 1 to 2a) 8.3 min showed no significant effect on the recovery rate and binding capacity. However, the reduction in the contact time to less than 1 min in 2b) led to a distinct improvement in the recovery rate to more than 97% with, simultaneously, approx. 0% recovery rate in the permeate.

(14) TABLE-US-00001 TABLE 1 Overview of parameters and results in the purification of bacteriophage phi X 174 on crosslinked cellulose hydrate membranes for various membrane volumes (MV) Example 1 Example 2a) Example 2b) MV in mL 0.023 0.046 0.56 Absolute flow rate in mL/min 0.3 0.6 5.6 Contact time in min 16.9 8.4 0.93 Recovery in the eluate in % 81.6 3 79.0 0 97.4 0.7 Recovery in the permeate in 0.7 0.3 1.0 0 0.05 0.01 % Binding capacity (particles/ 1.6 10.sup.14 1.7 10.sup.14 2.5 10.sup.14 mL MV)

Exemplary Embodiment 3: Purification of Bacteriophage phi X 174 from the Clarified Process Solution for Various Residence Times in the Membrane

(15) The method was carried out in line with Exemplary Embodiment 1 on a filtration unit, Vivapure system (SSB, Gttingen), containing the crosslinked cellulose hydrate membrane and having a membrane volume of 0.023 mL, but having an absolute flow rate of 1 ml/min, with the relative flow rate increasing compared to Exemplary Embodiment 1 from 13 MV/min to 43.5 MV/min, and as a result the residence time simultaneously decreasing from 4.6 sec to 1.4 sec. The dynamic binding capacity (DBC 10%), at which a 10% breakthrough of the phage particles, based on the starting solution, could be detected, was calculated. The results are depicted in Table 2 and in FIG. 3. In the case of a residence time of 4.6 sec, the 10% breakthrough of the phage particles, based on the starting solution, was achieved at about 1200 membrane volume (MV), and in the case of a residence time of only 1.4 sec, this was achieved as early as at approx. 200 membrane volume (MV). As a result of an increase in the residence time, it was thus possible to increase, by a factor of 6, the maximum volume of the phage solution that can be purified across the filtration unit before the 10% breakthrough of the phage particles, based on the starting solution.

(16) TABLE-US-00002 TABLE 2 Overview of parameters and results in the purification of bacteriophage phi X 174 on crosslinked cellulose hydrate membranes for various residence times Example 1 Example 3 MV in mL 0.023 0.023 Absolute flow rate in mL/min 0.3 1.0 Relative flow rate in MV/min 13 43.5 Residence time in sec 4.6 1.4 DBC 10% in MV 1200 200

Exemplary Embodiment 4: Purification of Bacteriophage phi X 174 from the Clarified Process Solution for Various Conductivities

(17) The method was carried out in line with Exemplary Embodiment 1 on a filtration unit, Vivapure system (SSB, Gttingen), containing the crosslinked cellulose hydrate membrane and having a membrane volume of 0.023 mL, but having a salt concentration in buffer A and B of 500 mM sodium chloride. The PEG concentration used in the context of the experimental plan was 11% in this series of experiments; therefore, the recovery rates are altogether lower than in the case of 12% PEG. The experiment was carried out twice. There is no significant influence of the salt concentration on the recovery rate, as can be seen on the basis of the results shown in Table 3.

(18) TABLE-US-00003 TABLE 3 Recovery rate in the purification of bacteriophage phi X 174 from the clarified process solution for various conductivities Example 1 Example 4 MV in mL 0.023 0.023 Absolute flow rate in mL/min 0.3 0.3 NaCl concentration in mmol 150 500 Recovery in the eluate in % 64.6 1.0 72.6 2.8

Exemplary Embodiment 5: Purification of Adeno-Associated Virus (AAV2) from Biotechnological Solution

(19) The adeno-associated virus particles (AAV2) were produced using HEK 293 T (ATCC CRL-3216). The cells were transfected using plasmids. The AAV2 stock solution was treated with Benzonase (30 units/ml), clarified by centrifugation (300g, where g=gravity, for 20 min at 4 C.) and filtered across 0.2 m sterile filters (Minisart, Sartorius Stedim Biotech GmbH). This solution is called AAV2 solution.

(20) The purification was carried out on a filtration unit, Vivapure (SSB, Gttingen), containing the crosslinked cellulose hydrate membrane and having a pore size of from 3 to 5 m and a membrane volume of 0.046 mL, connected to an kta system (kta Prime, GE Healthcare Life Science). Over the experiment, the absolute flow rate was constantly 0.6 mL/min. The membrane was first equilibrated with a mixture composed of buffer A (0.5 M NaCl, 0.05 M Tris, pH 7.0, conductivity 53.7 mS/cm) and buffer B (16% by weight of PEG 6000, 0.5 M NaCl, 0.05 M Tris, pH=7.0, conductivity 24.4 mS/cm) in the ratio of 1:4. The final concentration of PEG 6000 in the mixture was 12% by weight.

(21) For the binding of the AAV2 particles, buffer A was replaced with the AAV2 solution. The mixing operation was done in-line in a static mixer of the kta system. After altogether 55 mL of solution (13.7 mL of AAV2 solution and 41.3 mL of buffer B) had been loaded, the membrane was washed. To this end, a mixture composed of buffer A and buffer B in the ratio of 1:4 was used again. The AAV2 particles were eluted from the membrane by using 100% buffer A.

(22) The AAV2 particles were quantified by means of quantitative real-time PCR on an instrument from LightCycler 480 (Roche Life Science) using the master mix LightCycler Faststart Master SYBR Green. The reference used was an internally produced plasmid bearing the sequence of the AAV2 particle.

(23) The recovery rate of the purified AAV2 particle was 81.8%, with simultaneous breakthrough of 0.18% in the permeate and 0.15% in the wash solution. The binding capacity was 2.5210.sup.13 particles per mL of membrane volume.

(24) For comparison, the purification of the AAV2 particles from the same biotechnological solution was carried out using the method of affinity chromatography. To this end, the virus solution was guided across a chromatography column, packed with AVB Sepharose affinity resin, having a column volume of 0.2 ml at a flow rate of 1.2 ml/min.

(25) The recovery rate of the affinity chromatography-purified AAV2 particles was 30.0%. The binding capacity was 1.8210.sup.12 particles per mL of membrane volume.

(26) The results in Table 4 show that the purification method described according to the invention achieves a binding capacity higher by a factor of 10 compared to conventional affinity chromatography. Similarly, a higher recovery rate (recovery) is achieved.

(27) TABLE-US-00004 TABLE 4 Overview of parameters and results in the purification of AAV2 on crosslinked cellulose hydrate membranes on Hydrosart compared to the conventional affinity-chromatography method Purification as per method according Affinity to the invention chromatography MV in mL 0.046 0.2 Particle concentration in the 1.42 10.sup.12 1.42 10.sup.12 starting solution in particles/mL Particle concentration in the 2.62 10.sup.9 2.08 10.sup.11 breakthrough in particles/mL Particle concentration in the eluate 1.16 10.sup.12 3.64 10.sup.11 in particles/mL Binding capacity in particles/mL 2.52 10.sup.13 1.82 10.sup.12 MV Recovery rate in the eluate in % 81.8 30.0

Exemplary Embodiment 6: Purification of Adenovirus Type 5 from Biotechnological Solution

(28) The adenovirus type 5 particles (AD5) were produced using HEK 293 (ATCC CRL-1573). The AD5 stock solution was treated with Benzonase (50 units/ml), clarified by centrifugation (300 g for 20 min at 4 C.) and filtered across 0.45 m sterile filters (Minisart, Sartorius Stedim Biotech GmbH), resulting in a concentration of 4.510.sup.9 particles/mL being obtained. This solution is called AD5 solution.

(29) The purification was carried out on a filtration unit, Vivapure (SSB, Gttingen), containing the crosslinked cellulose hydrate membrane and having a pore size of from 3 to 5 m and a membrane volume of 0.023 mL, connected to an kta system. Over the experiment, the absolute flow rate was constantly 0.3 mL/min. The membrane was first equilibrated with a mixture composed of buffer A (0.5 M NaCl, 0.05 M Tris, pH=7.0, conductivity 53.2 mS/cm) and buffer B (13.3% by weight of PEG 6000, 0.5 M NaCl, 0.05 M Tris, pH=7.0, conductivity 28.4 mS/cm) in the ratio of 1:4. The final concentration of PEG 6000 in the mixture was 10% by weight. For the binding of the AD5 particles, buffer A was replaced with the AD5 solution. The mixing operation was done in-line in a static mixer of the kta system. After altogether 55 mL of solution (13.7 mL of AD5 solution and 41.3 mL of buffer B) had been loaded, the membrane was washed. To this end, a mixture composed of buffer A and buffer B in the ratio of 1:4 was used again. The AD5 particles were eluted from the membrane by using 100% buffer A.

(30) The AD5 particles were quantified by means of quantitative real-time PCR on an instrument from Roche Life Science (LC480) using the master mix Brilliant III Ultra-Fast SYBR Green QPCR. The reference used was an AD5 internal standard of viral DNA.

(31) The recovery rate of the purified AD5 virus from the elution was 82.3% with a binding capacity of 1.410.sup.12 particles per ml of membrane volume.

Exemplary Embodiment 7: Influence of the Pore Size of a Crosslinked Cellulose Hydrate Membrane on the Performance Parameter Permeability

(32) Crosslinked cellulose hydrate membranes of the Hydrosart membrane type (Sartorius Stedim Biotech GmbH) having different pore sizes were compared with respect to their permeabilities in the application of PEG-containing solutions, as exist in the method according to the invention. To this end, membranes having two different pore sizes were subjected to flow-through by the PEG-containing solution under application of 1 bar of pressure; the flow-through was collected and determined volumetrically. The solution used corresponded to the mixture M composed of buffer A and B according to Exemplary Embodiment 1 and containing 12% PEG.

(33) For comparison, the permeability of the membrane having the pore size of from 3 to 5 m was set at 100%. In the case of the membrane having a pore size of 1 m, the permeability was only 30.8% compared to the permeability of the membrane having a pore size of from 3 to 5 m.

(34) TABLE-US-00005 TABLE 5 Overview of parameters and measurement values in the determination of the influence of pore size on the performance parameter permeability for crosslinked cellulose hydrate membranes in the case of flow-through with the PEG-containing mixture M from Exemplary Embodiment 1 Membrane 1 Membrane 2 Pore size in m 3 to 5 1 Mixture M containing 12% PEG 86.6 35.7 Permeability (ml/min*bar*cm.sup.2) Comparison % 100 30.8

Exemplary Embodiment 8: Influence of Sanitization on the Performance Parameter Permeability in the Comparison of Crosslinked and Uncrosslinked Cellulose Hydrate Membrane

(35) A crosslinked cellulose hydrate membrane of the Hydrosart membrane type (Sartorius Stedim Biotech GmbH) of 1 m pore size was compared with respect to its permeability in the application of PEG-containing solutions after sanitization, with an uncrosslinked cellulose hydrate membrane of 1 m pore size.

(36) To this end, the membranes were sanitized by application of 1 N sodium hydroxide solution for 30 min and equilibrated with buffer A (approx. 150 MV). Thereafter, permeabilities of the PEG-containing mixture M were determined, as described in Exemplary Embodiment 7.

(37) TABLE-US-00006 TABLE 6 Overview of parameters and measurement values in the determination of the performance parameter permeability for crosslinked and uncrosslinked cellulose hydrate membranes after sanitization with sodium hydroxide solution Membrane 2 Membrane 3 Crosslinked Uncrosslinked Mixture M containing 12% PEG 35.5 5.56 Permeability (ml/min*bar*cm.sup.2) Comparison % 100 14.1

(38) After sanitization, the permeability of the uncrosslinked cellulose hydrate membrane for the PEG-containing mixture M according to Exemplary Embodiment 1 is only 14% in comparison with the crosslinked cellulose hydrate membrane.