SCALABLE LENTIVIRAL VECTOR PRODUCTION SYSTEM COMPATIBLE WITH INDUSTRIAL PHARMACEUTICAL APPLICATIONS

Abstract

The present invention relates to the industrialization of the production of recombinant lentiviral vectors in order to manufacture sufficient materials for therapeutic applications such as gene therapy and/or DNA vaccination, for use in clinical trials and/or commercial use.

Claims

1. A method for the production of a recombinant lentiviral vector, comprising: culturing, in suspension in a serum-free medium, mammalian cells transfected with at least one plasmid adapted for the production of a lentiviral vector, the culture being carried out in a volume of at least 5 L; and harvesting the produced recombinant lentiviral vector from the culture medium.

2. The method according to claim 1, wherein the mammalian cells are HEK293T cells.

3. The method according to claim 1, wherein the harvesting step consists of a single lentivirus harvest.

4. The method according to claim 3, wherein the transfection is a transient transfection and the single harvest is implemented between 48 and 72 hours post-transfection.

5. The method according to claim 1, comprising a transfection step wherein the cells are transfected with a mixture of polyethylenimine (PEI) and plasmids.

6. The method according to claim 5, wherein the PEI is a 20-25 kD linear PEI.

7. The method according to claim 5, wherein transfection is carried out with a total DNA amount of at least 1.5 μg/10.sup.6 cells.

8. The method according to claim 5, wherein the PEI and the plasmids are mixed before transfection according to a N/P ratio of less than 10, wherein N/P refers to the number of nitrogen atoms in the PEI per oligonucleotide phosphate.

9. The method according to claim 8, wherein the N/P ratio is of around 6.

10. The method according to claim 5, wherein the contact time between PEI and the plasmids before addition to the cell culture is between 5 and 30 minutes.

11. The method according to claim 1, wherein sodium butyrate is added to the cell culture 24 hours after transfection of the cells without changing the medium.

12. The method according to claim 11, wherein sodium butyrate is added to the cell culture at a final concentration in the culture of between 2 mM and 12 mM, between 2 mM and 10 mM, or a final concentration of 5 mM.

13. The method according to claim 1, wherein the cells are transfected with four plasmids including a plasmid encoding the envelope proteins (Env plasmid), a plasmid encoding the lentiviral GagPol proteins (Gag-Pol plasmid), a plasmid encoding the lentiviral Rev protein (Rev plasmid) and a plasmid comprising a transgene of interest (TOI) between a lentiviral 3′-LTR and a lentiviral 5′LTR (TOI plasmid).

14. The method according to claim 1, wherein the culture is implemented in a volume of at least 50 L.

15. The method according to claim 1, wherein at least 10.sup.7 infectious genomes/mL are produced.

16. The method according to claim 1, wherein: the cells are 293T cells; transfection of the cells is carried out with a mixture of PEI and the required plasmid(s); sodium butyrate is added 24 hours post-transfection without changing the medium of the culture; and a single harvest of produced lentiviral vectors is carried-out.

17. A cell culture device, wherein said culture device contains a volume of at least 5 L of a serum-free culture medium comprising mammalian cells transfected with at least one plasmid adapted for the production of a lentiviral vector, said cells growing in suspension in said culture device.

18. The cell culture device according to claim 17, wherein the cells are HEK 293T cells.

19. A method for optimizing the production of a lentiviral vector by a mammalian cell grown in suspension in a serum-free medium, transfected with plasmids required for said production, comprising adding sodium butyrate 24 hours post-transfection to a cell culture without changing the medium of the culture.

20. The method according to claim 19, wherein sodium butyrate is added at a final concentration of 5 mM.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0051] FIG. 1 is a graphical representation of the four plasmids used in the study presented in the examples.

[0052] FIG. 2 is a graph representing the test of different transfection conditions in 100 mL spinner flasks with HEK293F cells and measurement of GFP positive cells by flow cytometry.

[0053] FIG. 3 is a graph representing the test of different transfection conditions in 100 mL spinner flasks with HEK293F cells and measurement of the amount of p24 HIV capsid antigen by p24 ELISA testing.

[0054] FIG. 4 is a graph representing the test of different transfection conditions in 100 mL spinner flasks with HEK293T cells and measurement of GFP positive cells by flow cytometry.

[0055] FIG. 5 is a graph representing a test of different transfection conditions in 100 mL spinner flasks with HEK293T cells and measurement of the amount of p24 HIV capsid antigen by p24 ELISA testing.

[0056] FIG. 6 is a graph representing the effect on production yield of two different SFM media for the generation of the transfection complex (F17 MEDIUM and OPTIPROSFM).

[0057] FIG. 7 is a graph showing the transfection at the optimal molar ratio (1:1:2:1) of plasmids on the production of two different products (different TOI) having different sizes. The assay was performed in spinner flasks at 100 mL under optimal transfection conditions.

[0058] FIG. 8 is a graph showing the impact of sodium butyrate addition strategy on productivity, measurement of the p24 concentration in supernatant.

[0059] FIG. 9 is a graph showing the impact of sodium butyrate addition strategy on productivity, measurement of the infectious genome (IG) concentration in supernatant.

[0060] FIG. 10 is a graph showing the impact of sodium butyrate addition strategy on productivity, measurement of the ratio physical particles/infectious particles (PP/IP) in supernatant.

[0061] FIG. 11 is a graph representing the average of 6 productions of HIV-VSVG-WASp in spinner flasks at 100 mL with addition of sodium butyrate 24 hpt at a 5 mM final concentration in the culture.

[0062] FIG. 12 is a graph showing the comparison between suspension protocol at 100 mL with HEK293T cells grown in suspension in a serum-free medium and the standard in 10-stack Cell Factories for production of HIV-VSVG-WASP lentiviral vector, IG results and PP/IP ratio in supernatant at 48 hpt.

[0063] FIG. 13 is a graph representing the evaluation of the suspension process of the invention at different scales (100 mL to 50 L) in different cell culture devices (spinner, wave and stirrer tank) and comparison with conventional adherent cells process using serum.

EXAMPLES

[0064] The aim of this study was to produce a lentiviral vector at a scale compatible with industrial applications, in a bioreactor in suspension in a serum-free media. Advantageously, the process has been developed up to 50 L and the production is readily adaptable to at least 100 L, 200 L bioreactor scale, or even at least 1000 L.

[0065] For recombinant lentivirus production we used 4 plasmids (see strategy in FIG. 1).

Materials and Methods

Cell Culture:

[0066] All vector production and cell culture were done with an anchorage dependent HEK293T working cell bank (WCB), initially growing in the presence of fetal bovine serum which was adapted for suspension growth in serum free media and a new working cell bank was established. Cells were cultured in modified F17 MEDIUM supplemented with PLURONIC F68 (Invitrogen), GIBCO Anti-Clumping Agent (Invitrogen) and 4 mM GLUTAMAX (Invitrogen).

[0067] For the process development described, different culture containers were used under controlled conditions:

[0068] for the 100 mL scale: spinner flask (Techne, UK) under controlled conditions (37° C., 120 rpm); and

[0069] for larger scale: glass bioreactor (B-DCU 2 L-10 L, Sartorius), wave bioreactor (CULTIBAG RM 10 L-25 L, Sartorius), single use stirrer tank bioreactor (CULTIBAG STR 50 L, Sartorius) under controlled conditions (pH=7.2, pO2=50%, 37° C. and a specific agitation according to the system).

Vectors and Plasmids:

[0070] The W1.6-huWASP-WPRE vector described in Zanta-Boussif et al., 2009, Gene Ther., 16(5):605-19, was produced by transient transfection of 293T cells using 4 plasmids consisting of pCCL W1.6-huWASP-WPREmut6-K transfer plasmids combined with the GagPol, VSV-G, and Rev plasmids respectively coding for HIV-1 gag-pol and rev genes and for the unrelated vesicular stomatitis virus G glycoprotein. All plasmids contain the kanamycin resistance gene. Further details are provided in Merten et al., cited supra.

[0071] The HIV-VSVG-GFP vector was produced using the same reagent except for the transgene plasmid which is pRRLSINcPPT-PGK-eGFP-WPRE.

Sodium Butyrate:

[0072] Sodium butyrate is commercially available (sodium butyrate ≥98.5% (GC)|Sigma-Aldrich). A stock solution is prepared at 500 mM in customized F17 MEDIUM and 0.22 filtered.

Medium:

[0073] F17 MEDIUM (Invitrogen) is customized by supplementation with PLURONIC F68 at 0.08%, GIBCO Anti-Clumping Agent at 0.01%, and GLUTAMAX at 4 mM final concentration.

Lentiviral Vector Production:

[0074] Suspension culture vessels or bags were seeded at 0.2×10.sup.6 cells/mL. Transfection was performed 72 h after seeding. Cell density was between 0.8 and 1.3×10.sup.6 cells/mL at the time of transfection.

Example for WASp Production (Best Mode):

[0075] The four plasmids used in this study are represented in FIG. 1. Different amounts of total DNA have been tested. Different concentrations have been tested but in the most optimal conditions total DNA was used at an amount of around 2 μg/10.sup.6 cells. The transfection reagent used was JETPEI (Polyplus product) with an N/P ratio of about 6. DNA and JETPEI were respectively diluted in culture media before being gently mixed for approximately 10 min. This mixing led to the formation of a transfection complex, which was directly added to the cell culture. Twenty-four hours after transfection, sodium butyrate was added for a final concentration of approximately 5 mM. Conditioned media containing the lentiviral vector particles were harvested 72 h after transfection for analytical purposes.

Titration:

[0076] Physical particles produced were quantified by measuring the amount of p24 (HIV capsid protein) using a specific ELISA kit. Infectious particles were titrated after infection of a cell line susceptible to lentiviral vector infection using qPCR (TaqMan) as previously described in Merten et al. (supra).

Results

[0077] A—Description of Adaptation of HEK293T Cells to Suspension Culture in Chemically Defined Media in the Absence of Serum

[0078] Source of the HEK293T Cell Line:

[0079] Cells came from a vial of an adherent, GMP master (working) 293T cell bank cultured in DMEM at 10% FBS.

[0080] Adaptation to Suspension in the Serum-Free Media:

[0081] Cells were thawed in a T75 tissue culture flask (DMEM+10% FBS). After 2 passages and amplification in a T175 tissue culture flask, we performed a complete media change on adherent cells, replacing DMEM with modified F17 MEDIUM (serum-free media). 48 h after media change, all cells were detached from the support and viability was still around 90%. Cells were continuously cultivated in F17 in T175 tissue culture flask. After 3 passages in F17 and amplification in T225 tissue culture flask, cells were seeded in spinner flask at 50 ml in suspension condition (using single-use spinner flask, Corning). A cell bank of 54 vials (50×10.sup.6 cells/vial) of 293T cells in suspension was generated at passage 8 (P8).

[0082] Generation of Cell Bank:

[0083] The formulation for cryoconservation is: 80% F17, 10% DMSO and 10% methylcellulose 1%.

[0084] B—Production of Lentiviral Vector Expressing the Green Fluorescent Protein in Classical HEK293F Vs. HEK293T

[0085] One of the aims of our work was to establish a process for manufacturing lentiviral vectors in suspension culture in the absence of serum for industrial applications.

[0086] Initially, experiments were performed with HEK293F cell line, a commercially available cell line adapted for suspension culture in the absence of serum.

[0087] To generate the HIV-GFP lentiviral vectors by the 4-plasmid transfection system described in Zanta-Boussif et al., 2009, Gene Ther., 16(5):605-19, HEK293F cells were seeded in 100 mL spinner flasks at 1E+06 cells/mL. Different transfection conditions were tested at a scale of 100 mL. Variable parameters were the amount of total DNA used per 1×10.sup.6 cells, as well as the molar ratio between the 4 plasmids/1E+06 cells. Although variations are possible in these parameters the contact time for the complex formation (10 min) with transfection reagent (JETPEI) and the ratio DNA/PEI (N/P=6) were fixed as the optimal conditions for lentivirus production.

[0088] The DNA/PEI complex was generated in 10 mL of OPTIPROSFM (Invitrogen), a chemically defined medium. After 10 minutes of contact, the DNA/PEI complex mix was added directly into the culture.

[0089] To assess efficiency of transfection, cell cultures were analyzed by flow cytometry to measure green fluorescent protein expression.

[0090] In addition cell culture supernatants were subjected to p24 ELISA testing to measure the concentration of the HIV p24 capsid antigen which is indicative of the presence of lentiviral particles.

[0091] The results are presented in FIGS. 2 and 3.

Transfection Efficiency 48 h Post Transfection

[0092] Results show that even if HEK293F can be efficiently transfected in certain conditions of DNA concentration and ratio (2.5 μg, 1:1:1:1 and 1:1:1:2, respectively), very little amounts (<50 ng/mL) of p24 antigen can be detected. An amount of p24 above 150 ng/mL is indicative of an efficient lentiviral production whereas a lower value is essentially due to free p24. We can correlate the amount of p24 with the amount of physical particles using an ELISA kit (Alliance HIV-1 P24 ANTIGEN ELISA Kit (480 Test), PerkinElmer) which gives this information: 1 ng p24=1.2×10.sup.7 PP.

[0093] C—Production of HIV-GFP from HEK293T

[0094] The production of the lentiviral vector HIV-GFP was performed in similar conditions using HEK293T cells. The efficiency of transfection and the concentration of p24 antigen in the cell culture supernatants were determined at 48 h post-transfection.

[0095] The results are presented in FIGS. 4 and 5.

[0096] Results show that at a similar efficiency of transfection (˜90% at 2 and 2.5 μg DNA, ratio 1:1:2:1), HEK293T cells are more efficient than HEK293F at generating p24 antigen and therefore HIV lentiviral vector particles (198 ng/mL and 314 ng/mL).

[0097] In conclusion, these experiments allowed us to determine efficient conditions to generate lentiviral vectors in HEK293T cells at a small scale. Those conditions were further investigated to evaluate the feasibility to manufacture lentiviral vectors in suspension in the absence of serum at a scale allowing industrial applications.

[0098] D—Optimization of the Lentiviral Vector Production Process in HEK293T Cells in Suspension in the Absence of Fetal Bovine Serum

[0099] D1—Elimination of the OPTIPROSFM Medium for PEI/DNA Complex Generation

[0100] To simplify the process, we investigated the possibility to generate the PEI/DNA complex directly in the F17 MEDIUM so as to avoid the use of a different media (OPTIPROSFM).

[0101] Lentiviral vector production was performed at a 100 mL scale in a spinner flask using the transfection conditions determined in previous experiments, i.e., use of HEK293T cells, plasmid molar ratio of 1:1:2:1 and 2.5 μg total DNA/1×10.sup.6 cells. Cells and supernatants were harvested for testing and results are presented in FIG. 6.

[0102] Results show that there is no major difference in p24 concentration if generated from PEI/DNA complexed in the OptiPro media vs. F17. Using F17 media only throughout the process, rather than using two different types of media, constitutes a major improvement towards adapting the process to industrial scale.

[0103] D2—Importance of Using Plasmid Molar Ratio Instead of Plasmid (DNA) Mass Ratio in the Production System—Flexibility of the Process Thanks to the Use of a Molar Ratio

[0104] The lentiviral vector system of production used in the present experiments involves 4 plasmids. Three of those (Gag-Pol plasmid, VSV-G plasmid and Rev plasmid) are common to all lentiviral vectors as they encode trans-acting functions necessary for the formation of the lentiviral particles themselves, i.e., the structural elements (vector capsid, VSV-G envelope), enzymatic proteins (reverse transcriptase, integrase), and regulatory factor of expression (Rev protein). The only factor that varies is the plasmid encoding the vector genome. Because the transgene expression cassette encoded by the vector genome plasmid can come in different sizes (different promoters, cDNAs), the final amount of plasmid necessary for the generation of functional particles can vary from vector to vector, and with different expression cassettes.

[0105] Therefore, given that the molar ratio 1:1:2:1 (Env plasmid:Gag-Pol plasmid:Rev plasmid:TOI plasmid) gave the best results, we wanted to verify that by keeping the molar ratio intact, we could reproducibly maintain lentiviral production yields even if the size of the plasmid varied. Thanks to this molar ratio, which keeps the number of each plasmid molecule intact independently of their size in base pairs (and therefore their weight), we can guarantee the optimal transfection conditions regardless of the product.

[0106] FIG. 7 shows an example where we compared a lentiviral vector encoding the GFP protein cDNA (total size of the plasmid=7388 bp) with a lentiviral vector encoding the Wiskott-Aldrich protein (WASp) cDNA (total size of the plasmid=9780 bp).

[0107] The results show that by keeping the ratio of plasmid equal in terms of number of molecules (which conversely affects the amount of individual plasmid used), the yields of lentiviral vectors remain intact. These results suggest that the method of production of lentiviral vectors in HEK293T cells in suspension in the absence of serum with a total of 2.5 μg/1E+06 cells at a molar ratio of 1:1:2:1 can be used for different vectors independently of size. Of course, although optimal, these conditions may be varied starting from these values so that a person skilled in the art can adapt the parameters to the particular cells and plasmids used for production of the desired lentiviral vector.

[0108] D3—Improvement of Productivity: Use of Sodium Butyrate

[0109] Sodium butyrate has been reported to enhance the production of lentiviral vectors in an adherent cell system. (Gasmi et al Mar. 1999). We wanted to determine whether sodium butyrate would be useful in such suspension cultures. However, if that was the case, the use of sodium butyrate should not make the process more cumbersome, the priority being to keep the process applicable to industrial applications. Therefore we decided to test whether the addition of sodium butyrate post-transfection without media change could positively impact lentiviral vector yields in the conditions previously established (HEK293T, suspension culture, absence of serum, 2.5 μg/mL plasmid at a ratio of 1:1:2:1).

[0110] Experiments were performed at the 100 mL scale, in spinner flasks. Sodium butyrate was prepared in customized F17 MEDIUM and added post transfection at a final concentration of 5 mM. The effect on lentiviral vector production was assessed by measurement of p24 antigen by ELISA and by measuring the concentration of infectious particles using qPCR (TaqMan). Measuring both parameters allows for the calculation of the PP/IP ratio (total number of particles to infectious particles) which is an indicator of the quality of a lentiviral vector preparation. The quality of the production is considered acceptable when the PP/IP ratio is between 100-250 (results commonly observed for GMP production at (Genéthon).

[0111] Initially we tested different strategies for sodium butyrate added in a 100 mL spinner flask. One strategy was to add it 6 h post transfection directly in the culture. Another strategy was to perform a complete media change 24 h post transfection and add the sodium butyrate in the fresh media used to resuspend cells. Finally, the strategy giving the best results was to add the sodium butyrate directly in the culture 24 h post transfection without media change.

[0112] Note: during the media change cells were centrifuged 5 min at 500 g before resuspending them in fresh F17.

[0113] We performed an experiment in parallel with three spinners to confirm previous experiments, results are in FIGS. 8, 9 and 10.

[0114] Results show that adding sodium butyrate at a final concentration of 5 mM, 24 h post-transfection increases vector productivity between 3-4 fold concerning infectious particles and that there is also an increase in the amount of p24 produced.

[0115] FIG. 10 presents the ratio PP/IP that we had for this experiment.

[0116] This graph shows that sodium butyrate allows not only an increase of productivity but also keeps an acceptable quality of the production by giving a PP/IP ratio in the acceptable range (100-250).

[0117] The robustness of this strategy was assessed by doing 6 spinners with the same protocol; results are represented in FIG. 11.

[0118] These experiments confirm that the better harvest time is 48 hpt according to the IG concentration and the quality of the production regarding the PP/IP ratio at 48 hpt.

E-Scale Up.

[0119] E1—Demonstration that the Lentiviral Vector Production Method in Suspension-Grown Cells in the Absence of Serum Gives Similar Results to the Conventional Lentiviral Vector Production System in Adherent Cells in the Presence of Serum

[0120] Commonly, large-scale productions of lentiviral vectors for research or clinical purposes are performed using transfection of HEK293 adherent cells in the presence of serum. For reason of vector titers, HEK293T are the most commonly used cells. The production protocols are essentially based on the use of 2-stack or 10-stack cell factories or equivalent multitray systems. See Schweizer and Merten, 2010 Current Gene Therapy 10(6):474-486, most particularly part 2.3 (“Large Scale Process, Including Transient Transfection”).

[0121] This adherent-cell based protocol has been compared to the optimal method defined above in which HEK293T cells were grown in suspension in the absence of serum, and transfected with 2.5 μg DNA/1×10.sup.6 cells with a plasmid molar ratio of 1:1:2:1 with sodium butyrate added at 5 mM 24 hpt without media change.

[0122] FIG. 12 shows a comparison between suspension protocol at 100 mL with HEK293T and the standard in 10-stack cell factories for production of HIV-VSVG-WAS lentiviral vector.

[0123] Results demonstrate that the suspension system generates lentiviral vectors in similar yields and quality to the adherent cell system.

[0124] E2—Demonstration that the Lentiviral Vector Production System in Suspension in the Absence of Serum According to the Invention can be Scaled Up and Applied to Industrial Applications

[0125] The scalability of the optimal process of lentiviral production described above (HEK293T cells in suspension in the absence of serum with 2.5 μg/mL DNA/1e6 cells at a plasmid ratio of 1:1:2:1 with sodium butyrate) was evaluated over various volumes of culture in terms of yields of particles (p24 ELISA) and infectious particles (qPCR, TaqMan) in the case of a lentiviral vector HIV-VSVG-WASp. Ratios of PP/IP as described before were calculated for each scale and plotted in the histogram presented in FIG. 13 in comparison with results obtained with the conventional production method in adherent cells in the presence of serum.

[0126] Results show that vector productivity (number of infectious genomes, IG) and quality (PP/IP) of the novel system of lentiviral vector production is maintained over a wide range of culture volumes and that they favorably compare with those obtained with the conventional method of production implementing adherent cells grown in a serum-containing medium (same quality and productivity for all scales and competitive with the Cell Factories process).

[0127] These results show that the novel process of lentiviral vector production in suspension combines efficiency with practicality and can therefore be used in industrial-scale applications of lentiviral vectors.