Large-scale PEI-mediated plasmid transfection

Abstract

We have found a way to make possible large-scale plasmid transfection using PEI to produce high titer viral vectors in fixed bed or adherent cell culture bioreactors by using PEI as a transfection agent, while avoiding formation of the PEI-plasmid precipitate which in prior art approaches clogged adherent bioreactor substrates. We have also found a way to improve PEI-based transfection by modifying how pH and CO.sub.2 are managed during transfection.

Claims

1. A method for manufacturing a recombinant lentiviral vector, the method comprising: (a) mixing polyethylene imine (PEI) and plasmid DNA coding for a recombinant lentiviral vector to form a transfection solution, wherein the transfection solution is incubated for longer than 20 minutes and the transfection solution is stirred or mixed throughout the incubation time such that the transfection solution does not form a DNA-PEI complex precipitate; (b) adding cells and the transfection solution to a bioreactor and recirculating the transfection solution until transfection is substantially complete, whereby the plasmid DNA transfects the cells to make producer cells which produce the recombinant lentiviral vector; and then (c) culturing the producer cells in adherent mode in the bioreactor, wherein the bioreactor has a fixed bed volume of at least 5 liters and whereby the producer cells produce the recombinant lentiviral vector; and then (d) harvesting recombinant lentiviral vector.

2. The method of claim 1, wherein the plasmid is present in an amount adequate to produce a PEI:plasmid DNA ratio of about 1:1.5.

3. The method of claim 1, wherein the transfection solution is at least 20 liters in volume.

4. The method of claim 1, wherein the plasmid DNA concentration in the transfection solution is at least 300 nanograms of DNA per cm.sup.2.

5. The method of claim 4, wherein the plasmid DNA concentration in the transfection solution is not more than 400 nanograms of DNA per cm.sup.2.

6. The method of claim 1, further comprising: measuring the formation of DNA-PEI complexes in the transfection solution using light scattering.

7. The method of claim 1, wherein step (b) comprises ceasing addition of CO.sub.2 to the transfection solution, whereby the PEI does not react with added CO.sub.2.

8. The method of claim 1, wherein the bioreactor comprises an automatic pH control mechanism, and wherein the method further comprises allowing the pH of the culture medium to fall naturally during or after transfection, producing an acidic culture medium which prevents PEI-DNA complex precipitate formation.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 is a flow-chart of the process steps for using PEI in a novel fixed-bed bioreactor, the iCELLis® bioreactor.

(2) FIG. 2 is a flow-chart of the process steps for media exchange in a novel fixed-bed bioreactor, the iCELLis® bioreactor.

(3) FIG. 3 shows results from our initial testing of PEI-mediated transfection only by using one plasmid and PEIpro® transfection reagent in flasks by following the manufacturer's instructions, in an iCELLis® bioreactor.

(4) FIG. 4 shows the DNA concentration in our transfection mix (i.e., the plasmid DNA mixed with PEI, before adding that mixture to the host cells).

(5) FIG. 5 shows the effect of prolonged incubation (by mixing) on DNA-plasmid particle size.

DESCRIPTION

(6) The manufacturer of PEIpro® (Polyplus transfection) recommends the use of PEI at 1-6 μl of PEIpro® per μg of DNA for HEK293 cells. For adherent cells, the recommended amount of DNA is 0.1-0.58 μg/cm2, depending of the type of the flask when the total concentration is up to 0.029 μg/μl (Polyplus, PEIpro® in vitro DNA transfection reagent protocol).

(7) First we did “as instructed” by the art in a small scale. A problem came when we tried to scale that up, however, because we realized that the art-recommended approach does not work in a large scale where the working volume is limited. We initially tested PEI mediated transfection only by using one plasmid and PEIpro® transfection reagent in flasks by following the manufacturer's instructions (FIG. 3). We used a DNA amount from 100 to 400 ng/cm.sup.2, evaluated using a DNA:PEI ratio of 1:1. Results showed that 300-400 ng/cm.sup.2 total DNA concentration achieved the highest transfection efficiency (up to 98% positive cells) in our cells. The DNA concentration in our transfection mix (i.e., the plasmid DNA mixed with PEI, before adding that mixture to the host cells) we evaluates was up to 15 μg/ml. Also, we used different PEI ratios, and found that the PEI ratio has an effect on the transfection efficacy (FIG. 4). We found that the best transfection efficacy was achieved by using PEI:DNA ratio of about 1:1.5.

(8) In our next experiments, the total DNA concentrations per cm.sup.2 were the same, but we used DNA which contained four different plasmids, as is typically used for retroviral vector production. Virus production by producer cells which have been transfected with several plasmids is tricky because the producer cells require a larger volume of plasmid DNA (i.e., several different plasmid constructs) than a typical recombinant protein production where only one plasmid is used to express the one polypeptide of interest. We found that the highest titers were achieved using the best conditions shown in a previous experiments.

(9) The first PEI mediated plasmid transfection in an iCELLis® bioreactor was done by Lennaertz et al. when they produced AAV in a 0.53 m.sup.2 fixed-bed bioreactor. Their results showed that plasmid transfection is feasible in the low bed height laboratory-scale iCELLis® nano bioreactor (Lennaertz et al., 2013).

(10) Our next aim was to test virus production using iCELLis® fixed bed bioreactor with the same conditions than in flasks (manufacturer's instructions) but this time using the largest 4 m.sup.2 iCELLis® Nano bioreactor (fixed-bed comparable for 500 m.sup.2 in iCELLis® 500) (FIG. 1). It was surprisingly observed that actually the recommended transfection conditions are not scalable and applicable for iCELLis® bioreactors, especially in higher bed height (>2 cm) bioreactors due to its limited working volume for high total cell number if the DNA amount would have been kept the same per cell or per cm.sup.2. In other words, if the same plasmid transfection mix would have been used, it would not fit into the bioreactor or would have required full medium exchange during the transfection. The iCELLis® Nano is a small scale equipment where the full medium exchange can be done fast and is not limiting step in a process. In contrast, the iCELLis® 500 at scale, the full medium exchange is not practical process step because it takes time and may be influencing on cell viability due to the fact that during the draining, stirring is closed and the cells on the upper carriers are without the medium. Thus there was a need to decrease the volume in transfection mix which lead to higher DNA (plasmid) concentration in a mix but transfection efficacy was decreased (Run 1, table 1). We were able to avoid formation of aggregation by continuous mixing of the DNA-PEI mix before adding to the cells or with shorter incubation time (table 1). If we would have kept the DNA concentration the same in a iCELLis® nano than in flasks, total of 800 ml transfection mix would have needed which is the maximal working volume in the iCELLis® nano scale. Alternatively, recirculation is required in iCELLis® fixed bed bioreactor when the ≥4 cm bed height is used. In 2 cm fixed bed, lower cell amount do not require that much plasmids (transfection mix) and thus the bioreactor volume is not limiting factor. Thus Lennaertzt et all. did not face any problems in their transfection (Lennaertz et al 2013). We tested the transfection by making plasmid-PEI mix in a large volume (lower concentration), and adding it to the bioreactor but because the bioreactor working volume was exceeded, we used a re-circulation loop (FIG. 2) (run 7, table 1).

(11) Another surprise was seen during the large scale transfection. After adding transfection mix into the bioreactor, everything seem to be normal but when sampling the bioreactor, chemical reaction was observed when normal shape plasmid tube “collapsed” or “melted” because of the medium sample with transfection mix. It was find out that PEI could react with CO.sub.2. Based on all what happened and what was seen, it was concluded that CO.sub.2 flow (pH control) should be shut down to be able to avoid any chemical reaction in a bioreactor. This can be a major safety aspect too.

(12) We have found several ways to optimize the large scale plasmid transfection to produce high titer viral vectors (or any other biological product) in bioreactors (such as, but not limited to, the iCELLis® fixed bed bioreactor). We have also found a way to improve the safety of the PEI-based production by controlling CO.sub.2 flow during the transfection, and short incubation of the transfection mix (DNA-PEI). When (plasmid) DNA is mixed to the transfection reagent, we have found that there is several factors not previously known to influence influencing on the transfection efficiency do in fact have results-critical effect when transfection is attempted at scale. These factors are:

(13) 1. Plasmid DNA concentration

(14) 2. PEI to plasmid DNA ratio

(15) 3. Incubation time

(16) 4. Mixing during the precipitation

(17) 5. Temperature

(18) 6. Medium

(19) 7. pH

(20) The most optimal conditions may not be practical to perform in large scale to transduce cells in a bioreactor where a high cell density is in relative limited volume. If there is a need to decrease the volume of the transfection mix, increased plasmid DNA concentration may not accomplish optimal DNA-PEI precipitation, and may lead even to DNA aggregation, rendering the DNA in a physical aggregate physically too large to properly transfect a host cell. To prevent aggregation, we surprisingly have found that a shorter transfection mix incubation time before adding the mix to the cells is preferable. This finding was surprising and counter-intuitive because the art teaches that to increase transfection, one should increase the time the plasmid is incubated with the transfection reagent to 20 min.

(21) Continuous Mixing of the Transfection Mix

(22) (DNA-PEI) Prior Addition to the Cells

(23) Similarly, the conventional practice in the art is to combine the plasmid DNA and the transfection reagent, and allow the combination to rest, allowing precipitation, because stirring is thought to interfere with precipitation forming, perhaps by physically moving plasmid DNA away from a transfection reagent. We found that when performed at scale, allowing the mixture to sit tranquil is in fact disadvantageous, and the combination should be stirred or mixed during the incubation time, preferably stirred or mixed for substantially the entire transfection incubation time.

(24) Continuous Mixing and Prolonged Incubation of the

(25) Transfection Mix (DNA-PEI) Prior Addition to the Cells

(26) Similarly, the art teaches precipitation is substantially complete within about 20 minutes, so one should add the mixture to the cells at 20 minutes. Alternatively, we found that when performed at scale, DNA-PEI complex formation depends on the relative concentration of each, and the concentration of both in the medium. We thus surprisingly found that when performed at scale, may continue for longer than twenty minutes, so transfection at scale may require a materially longer incubation than the 20 minute period recommended by the prior art. To avoid the aggregation, it is preferred to stir or mix the transfection mix during the incubation time. It was also observed that prolonging incubation time has an effect on DNA/PEI particle size formation. It was observed that prolonged incubation (by mixing) is increasing the particle size until 35 min, but decreasing the size after that (FIG. 5).

(27) Increasing the transfection volume can be overcome by adding the transfection mix to recirculation loop.

(28) As mentioned above, the art suggests that increased DNA concentration can lead to DNA aggregation, rendering the DNA unavailable for transfection. The art teaches to reduce DNA concentration by perfusion, in effect washing DNA out of the transfection vessel entirely. This works, but it wastes a tremendous amount of plasmid. We surprisingly found a way that one can transfect at scale and overcome the limited volume issue of excessive DNA concentration by re-circulation of the transfection mix and culture medium during the transfection (FIG. 2). This finding was counter-intuitive because the art suggested that re-circulating the culture medium would re-circulate an excess amount of DNA into the transfection vessel. This way the volume of DNA-PEI can be increased, lowering the actual concentration of DNA-PEI during the mixing/incubation time and also after adding the mix to the bioreactor.

(29) We tested the transfection using the 200 ml volume when the full medium enhanced was not required. This way the DNA concentration in a mix increased from 0,015 μg/μl to 0.05 μg/μl. When the PEI was mixed with plasmid, and incubated 15-20 min. at room temperature according to the manufacturer, remarkable visible DNA aggregation was observed. Typically plasmids and PEI should form opal or “cloudy” homogenous precipitation to be able to efficiently transfect the cells. In our case, visual large plasmid aggregation was formed during the incubation. Also transfection efficiency was surprisingly low (40%, measured by sampling the upper carriers from the fixed bed), and productivity decreased. Even though mix should be incubated to allow the DNA and PEI to form a cloudy precipitation, our next experiment was done by limiting the incubation time (<10 min.) when less problematic “too large” aggregation was formed. This improved transfection efficiency. Also we tested the volume increase by doubling the volume of transfection mix into 480 ml when the concentration of the DNA was decreased.

(30) The transfection reagent can be PEIpro® (PolyPlus), jetPEI®, linear PEI or any polyethylene imine derivative. It may also be any other functionally-equivalent transfection reagent.

EXAMPLES

(31) Plasmid Transfection

(32) We tested the transfection using the 200 ml volume when the full medium enhanced was not required. This way the DNA concentration in a mix increased from 0,015 μg/μl to 0.05 μg/μl. When the PEI was mixed with plasmid, and incubated 15-20 min. at room temperature according to the manufacturer, remarkable visible DNA aggregation was observed (Table 1). Typically, plasmids and PEI should form opal or “cloudy” homogenous precipitation to be able to efficiently transfect the cells. In our case, visual large plasmid aggregation was formed during the incubation. Also, transfection efficiency was surprisingly low (40%, measured by sampling the upper carriers from the fixed bed), and productivity decreased. Even though mix should be incubated to allow the DNA and PEI to form a cloudy precipitation, our next experiment was done by limiting the incubation time (<10 min.) when less problematic “too large” aggregation was formed. This improved transfection efficiency. Also we tested the volume increase by doubling the volume of transfection mix into 480 ml when the concentration of the DNA was decreased. Best transfection efficacy was achieved when DNA concentration was further increased and DNA-PEI mix as incubated for 7.5 minutes with mixing, before addition to the bioreactor (Table 1).

(33) First we tested PEI mediated transfection only by using one plasmid and PEIpro® transfection reagent in flasks by following the manufacturer's instructions (FIG. 3). We used a DNA amount from 100 to 400 ng/cm.sup.2, evaluated using a DNA:PEI ratio of 1:1. Results showed that 300-400 ng/cm.sup.2 total DNA concentration achieved the highest transfection efficiency (up to 98% positive cells) in our cells. The DNA concentration in our transfection mix (i.e., the plasmid DNA mixed with PEI, before adding that mixture to the host cells) we evaluates was up to 15 μg/ml. Also, we used different PEI ratios, and found that the PEI ratio has an effect on the transfection efficacy (FIG. 4). We found that the best transfection efficacy was achieved by using PEI:DNA ratio of about 1:1.5.

(34) In our next experiments, the total DNA concentrations per cm.sup.2 were the same, but we used DNA which contained four different plasmids, as is typically used for retroviral vector production. Virus production by producer cells which have been transfected with several plasmids is tricky because the producer cells require a larger volume of plasmid DNA (i.e., several different plasmid constructs) than a typical recombinant protein production where only one plasmid is used to express the one polypeptide of interest. We found that the highest titers were achieved using the best conditions shown in a previous experiments (data not shown).

(35) The next aim was to test virus production using iCELLis® fixed bed bioreactor with the same conditions but this time using the largest 4 m.sup.2 iCELLis® Nano bioreactor (fixed-bed comparable for 500 m.sup.2 in iCELLis® 500). It was observed that actually the recommended transfection conditions are not scalable and applicable for iCELLis® bioreactors, especially in a higher bed height (>2 cm) bioreactors due to its limited working volume for high total cell number if the DNA amount would have been kept the same per cell or per cm.sup.2. In other words, if the same plasmid transfection mix would have been used, it would not fit into the bioreactor, or would have required a full medium exchange during the transfection. The iCELLis® Nano is a small-scale equipment where the full medium exchange can be done fast and is not limiting step in a process. In contrast, at the scale of an iCELLis® 500 the full medium exchange is not a practical process step because it takes time and may influence cell viability due to the fact that during the draining, stirring is closed and the cells on the upper carriers are without the medium. Thus, there was a need to decrease the volume in transfection which lead to higher DNA (plasmid) concentration in a mix. If we would have kept the DNA concentration the same, total of 800 ml transfection mix would have needed which is the maximal working volume. Thus, we tested the transfection using the 200 ml volume when the full medium enhanced was not required. This way the DNA concertation in a mix increased from 0.015 μg/μl to 0.05 μg/μl. When the PEI was mixed with plasmid, and incubated 15-20 minutes at room temperature according to the manufacturer's instructions, remarkable visible DNA aggregation was observed. Typically plasmids and PEI should form an opalescent or “cloudy” homogenous precipitation to be able to efficiently transfect the cells. In our case, however, visual large plasmid aggregation was formed during the incubation. Also, transfection efficiency was surprisingly low (40%, measured by sampling the upper carriers from the fixed bed), and productivity decreased. Even though mix should be incubated to allow the DNA and PEI to form a cloudy precipitation, our next experiment was done by limiting the incubation time (<10 min.) when less problematic “too large” aggregation was formed. This improved transfection efficiency. Also we tested the volume increase by doubling the volume of transfection mix into 480 ml when the concentration of the DNA was decreased.

(36) Improvement for the situations was get when transfection mix was stirred also during the incubation. We concluded that the stirring during the incubation is actually prohibiting the large aggregation when the precipitated molecules are still in reasonable small size and no large aggregation can be formed. The formation of the precipitation was followed by Nanosight™ when the size variation and number of particle can be monitored based on brown movement. Our invention is against the common knowledge that when large transfection is done, the continuous mixing is required or additional the stand-still incubation needs to be shorten than recommended (<20 min).

(37) pH Control

(38) Bioreactors are typically provided with an automatic pH control to maintain the culture medium at a constant pH, automatically adding a basic solution (e.g., a sodium bicarbonate solution) if the culture medium pH falls. We have previously shown (patent number GB 14/17042. 7) that during the transfection if the automatic pH control in the iCELLis® bioreactor is left operational, then the bioreactor will add base solution into the bioreactor, which will cause the formation of a precipitate in the bioreactor. With calcium phosphate transfection, the precipitate, which we believe is a DNA-salt precipitate, is undesirable because it clogs the bioreactor and impedes productivity. We found that by disabling the automatic pH control during (before or just after) the transfection and allowing the pH of the culture medium to fall naturally, the resulting slightly-acidic culture medium prevents precipitate formation and thus increases yield. We here made same observation also with PEI based transfection that there is a need to switch-off the pH control because during the transfection, the system is automatic adding base into the bioreactor and locally this may cause high pH change and lead to either aggregation or detaching of DNA or PEI from the complex.

(39) Re-Circulation Mode

(40) To find the optimal conditions for large-scale transfection, we also tested re-circulation method when the bioreactor was equipped with recirculation instead of perfusion during transfection. With recirculation loop, half of the transfection mixture was added to the bioreactor and the other half to the recirculating medium until the total volume was 1000 ml, and the mixture was added to the bioreactor by recirculating the transfection mixture through the bioreactor. The recirculation loop was replaced with perfusion 24 h post-transfection (“PT”). Critical was to switch of the pH control. Transfection efficacy was comparable, but that might not as practical to perform and requires increased amounts of medium (Table 1, run 7).

(41) Safety Improvement by Switching of CO.sub.2

(42) Another surprise was seen during the large scale transfection. Transfection mix containing DNA, PEI and medium without FBS was done. Base and DO controls as well as perfusion were OFF during transfection, but CO.sub.2 control was ON. Everything seemed to working, and the values on the screen of the iCELLis® were as they were supposed to be. Anything else unusual was not noticed at that point. A 5 ml sample was taken from the bioreactor at 14:00 into a 15 ml Falcon tube for glucose and lactate measurements. Before the sample was taken, the tube was normally shaped. After taking the sample, the operator who took the sample was holding the tube, while emptying the sample bottle back into the bioreactor. After 2-3 min the operator viewed the tube that contained the sample and surprisingly found that the tube had changed its form. It was collapsed/flattened, but no scratches were visible. Later, also a plastic Erlenmeyer flask containing sample from the bioreactor also appeared to be “melted” as if by excess heat. The apparent “melting,” however, was not caused by the heat. It was found that PEI can react with CO.sub.2 causing a chemical reaction. Based on all what happened and what was seen, it was concluded that CO.sub.2 flow should be shut down to be able to avoid any chemical reaction in a bioreactor. This can be a major safety aspect, too.

(43) TABLE-US-00001 TABLE I Transfection optimization and the influence in the transfection efficacy and the total yield Run Transfection number Transfection mix efficiency Yield 1 PEI and DNA 300 ng/cm.sup.2 in 120 ml IMDM without serum 35-40% 4.05E+09 (+glut + P/S) -> separate mixes (a′120 ml). Visual aggregation seen. Mixes in 50 ml Falcons, vortexing pairwise and pouring into glass beaker -> 15 min incubation RT without mixing PEI 2 and DNA 300 ng/cm.sup.2 in 240 ml medium without serum 43% 2.83E+09 (+glut + P/S) -> separate mixes (a′240 ml) Mixes in 500 ml erlenmayer -> PEI mix was poured into DNA mix -> Mixing with magnetic stirring -> 10 min incubation RT without mixing 3 PEI and DNA 300 ng/cm.sup.2 in 240 ml mediumwithout serum 75% 2.09E+09 (+glut + P/S) -> separate mixes (a′240 ml) Mixes in 500 ml erlenmayer -> PEI mix was poured into DNA mix -> Mixing strongly with magnetic stirring -> 10 min incubation RT without mixing 4 PEI and DNA 300 ng/cm.sup.2 in 240 mlmedium without serum 65-70% 3.17E+09 (+glut + P/S) -> separate mixes (a′240 ml) Mixes in 500 ml erlenmayer -> PEI mix was poured into DNA mix -> mixing with magnetic stirring -> 5 min incubation RT mixing gently 5 PEI and DNA 400 ng/cm.sup.2 in 320 ml medium without serum 60-80% 5.20E+09 (+glut + P/S) -> separate mixes (a'320 ml) Mixes in 500 ml and 1 liter erlenmayer -> PEI mix was poured into DNA mix -> mixing with magnetic stirring -> 7.5 min incubation RT mixing gently 6 PEI and DNA 300 ng/cm.sup.2 in 240 ml medium without serum 50-60% 9.06E+09 (+glut + P/S) -> separate mixes (a′240 ml) Mixes in 500 ml erlenmayer -> PEI mix was poured into DNA mix -> mixing with magnetic stirring -> 5 min incubation RT mixing gently 7 DNA 336 ng/cm.sup.2 in 240 ml medium without serum (+glut + P/S) NA -> separate mixes (a'240 ml) Mixes in 500 ml erlenmayer -> PEI mix was poured into DNA mix -> mixing with magnetic stirring -> 5 min incubation RT mixing gently

REFERENCES

(44) Chahal P S, Schulze E, Tran R, Montes J, Kamen A A. “(2014) Production of adeno-associated virus (AAV) serotypes by transient transfection of HEK293 cell suspension cultures for gene delivery. Chiorini J A, Kim F, Yang L, Kotin R M. (1999) Cloning and characterization of adeno-associated virus type 5. Coleman, J. E., Huentelman, M. J., Kasparov, S., Metcalfe, B. L., Paton, J. F., Katovich, M. J., Semple-Rowland, S. L., and Raizada, M. K. (2003). Efficient large-scale production and concentration of HIV-1-based lentiviral vectors for use in vivo. Physiol Genomics 12, 221-228. Cortin V, Thibault J, Jacob D, Gamier (2004) A. High-titer adenovirus vector production in 293S cell perfusion culture. Biotechnol Prog. 20(3):858-63. Dormond E, Perrier M, Kamen A. (2009) From the first to the third generation adenoviral vector: what parameters are governing the production yield? Biotechnol Adv. 27(2):133-44. Ferreira T B, Ferreira A L, Carrondo M J, Alves P M. (2005) Effect of re-feed strategies and non-ammoniagenic medium on adenovirus production at high cell densities. J Biotechnol. 119(3):272-80. Follenzi, A. and Naldini, L. (2002). Generation of HIV-1 derived lentiviral vectors. Methods Enzymol. 346, 454-465. Geraerts, M., Michiels, M., Baekelandt, V., Debyser, Z., and Gijsbers, R. (2005). Upscaling of lentiviral vector production by tangential flow filtration. J. Gene Med. 7, 1299-1310. Iyer P, Ostrove J M, Vacante D. (1999) Comparison of manufacturing techniques for adenovirus production. Cytotechnology. 30(1-3):169-72. Karolewski, B. A., Watson, D. J., Parente, M. K., and Wolfe, J. H. (2003). Comparison of transfection conditions for a lentivirus vector produced in large volumes. Hum. Gene Ther. 14, 1287-1296. Kamen A, Henry O. (2004) Development and optimization of an adenovirus production process. J Gene Med. 6 Suppl 1:S184-S192. Koldej, R., Cmielewski, P., Stocker, A., Parsons, D. W., and Anson, D. S. (2005). Optimisation of a multipartite human immunodeficiency virus based vector system; control of virus infectivity and large-scale production. J. Gene Med. 7, 1390-1399. Kuroda, H., Kutner, R. H., Bazan, N. G., and Reiser, J. (2008). Simplified lentivirus vector production in protein-free media using polyethylenimine-mediated transfection. J. Virol. Methods. Lennaertz, A., Knowles, S., Drugmand, J C., and Castillo, J. (2013) Viral vector production in the Integrity® iCELLis® single-use fixed-bed bioreactor, from bench-scale to industrial scale. 7 (Suppl 6), 59-60. 1-1-2013. BMC Proceedings. Lesch H P, Heikkila K M, Lipponen E M, Valonen P, Müller A, Räsänen E, Tuunanen T, Hassinen M M, Parker N, Karhinen M, Shaw R, Ylä-Herttuala S. Process Development of Adenoviral Vector Production in Fixed Bed Bioreactor: From Bench to Commercial Scale.Hum Gene Ther. 2015 August; 26(8):560-71. doi: 10.1089/hum.2015.081. Liu H, Liu X M, Li S C et al. (2009) A high-yield and scaleable adenovirus vector production process based on high density perfusion culture of HEK 293 cells as suspended aggregates. J Biosci Bioeng. 107(5):524-9. Meuwly F, Ruffieux P A, Kadouri A, von SU. (2007) Packed-bed bioreactors for mammalian cell culture: bioprocess and biomedical applications. Biotechnol Adv. 25(1):45-56. Naldini, L., Blomer, U., Gage, F. H., Trono, D., and Verma, I. M. (1996a). Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc. Natl. Acad. Sci. U.S.A 93, 11382-11388. Petiot E, Cuperlovic-Culf M, Shen C F, Kamen A. (2015) Influence of HEK293 metabolism on the production of viral vectors and vaccine. Vaccine. November 4; 33(44):5974-81. Rajendran R, Lingala R, Vuppu S K et al. (2014) Assessment of packed bed bioreactor systems in the production of viral vaccines. AMB Express. 4:25. Reiser, J. (2000). Production and concentration of pseudotyped HIV-1-based gene transfer vectors. Gene Ther. 7, 910-913. Rodrigues A F, Carmo M, Alves P M, Coroadinha A S. (2009) Retroviral vector production under serum deprivation: The role of lipids. Biotechnol Bioeng. 104(6):1171-81. Segura, M. M., Garnier, A., Durocher, Y., Coelho, H., and Kamen, A. (2007). Production of lentiviral vectors by large-scale transient transfection of suspension cultures and affinity chromatography purification. Biotechnol. Bioeng. 98, 789-799. Sena-Esteves, M., Tebbets, J. C., Steffens, S., Crombleholme, T., and Flake, A. W. (2004). Optimized large-scale production of high titer lentivirus vector pseudotypes. J. Virol. Methods 122, 131-139. Slepushkin, V., Chang, N., Cohen, R., Gan, Y., Jiang, B., Deausen, D., Berlinger, D., Binder, G., Andre, K., Humeau, L., and Dropulic, B. (2003). Large-scale Purification of a Lentiviral Vector by Size Exclusion Chromatography or Mustang Q Ion Exchange Capsule. BioProcessing Journal 2, 89-95. Tiscornia, G., Singer, O., and Verma, I. M. (2006). Production and purification of lentiviral vectors. Nat. Protoc. 1, 241-245. Wu S C, Huang G Y, Liu J H. (2002) Production of retrovirus and adenovirus vectors for gene therapy: a comparative study using microcarrier and stationary cell culture. Biotechnol Prog. 18(3):617-22 Wang, X., Olszewska, M., Qu, J., Wasielewska, T., Bartido, S., Hermetet, G., Sadelain, M., and Riviere, I (2015) Large-scale Clinical-grade Retroviral Vector Production in a Fixed-Bed Bioreactor. 38[3], 127-135. 3-4-2015. J Immunother