Journal of Bioscience and Bioengineering VOL. 110 No. 2, 187 – 193, 2010 www.elsevier.com/locate/jbiosc
Using a fed-batch culture strategy to enhance rAAV production in the baculovirus/insect cell system Yu-Kuo Liu,1,† Ching-Jen Yang,1,2,3,† Chao-Lin Liu,2 Chia-Rui Shen,3 and Lie-Ding Shiau1,⁎ Department of Chemical and Materials Engineering, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333, Taiwan 1 Department of Chemical Engineering and Graduate School of Biochemical Engineering, Mingchi University of Technology, 84 Gungjuan Road., Taishan, Taipei, 243, Taiwan 2 and Department of Medical Biotechnology and Laboratory Science, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333, Taiwan 3 Received 14 July 2009; accepted 6 February 2010 Available online 11 March 2010
Recombinant adeno-associated virus (rAAV) is one of the most promising vectors for human gene therapy. However, the production systems that are currently available have a limited capacity and cannot provide sufficient quantities of rAAV for preclinical or clinical trials. Many novel methods for improving rAAV production have been developed, but few researchers have focused on the culture process. In this study, we use a fed-batch culture system to enhance rAAV yield in the baculovirus/ insect cell system. When the insect cells were co-infected with MOI = 5 of Bac-GFP at a ratio of 1:9:9 (Bac-GFP: Bac-Rep: Bac-VP), the fed-batch culture achieved optimal rAAV yields. In batch culture, the optimal cell density for producing rAAV was found to be 1 × 106 cells/ml, and the highest rAAV yield (1.22 × 108 IVP/ml, 122 IVP/cell) occurred at day 5 post-infection. In the fed-batch culture, rAAV yield reached 2.13 × 108 IVP/ml at day 4 post-infection, and the highest rAAV yield was 2.40 × 108 IVP/ml (240 IVP/cell) at day 5 post-infection. The cost of the batch and fed-batch cultures is similar; however, the rAAV yield was 2.6-fold higher in the fed-batch culture system compared with that in the batch culture system. Therefore, here we demonstrated an economical and efficient strategy for rAAV production. © 2010, The Society for Biotechnology, Japan. All rights reserved. [Key words: Adeno-associated virus; Baculovirus/insect cell system; Fed-batch culture; Batch culture; Large-scale]
Gene transfer has recently emerged as a new and powerful therapeutic strategy. Many reports have compared the advantages and limitations of different gene delivery vectors (1–3). Adenoassociated virus (AAV) is one of the most promising vectors for human gene therapy. The AAV vector, a small, non-enveloped singlestranded (ss) DNA of approximately 4.7 kb, is nonpathogenic to humans and can be used to provide long-term gene expression. Wild-type AAV has two open reading frames (ORFs) with genes encoding for replication (Rep78, Rep68, Rep52, Rep40) and structure (VP1, VP2, VP3). These genes are flanked by inverted terminal repeats (ITRs, 145 bp) located at each end of the genome that are necessary for replication and encapsidation of the viral genome. The production of recombinant adenoassociated virus (rAAV) is commonly achieved by co-transfection with three plasmids containing the therapeutic gene flanked by ITRs, the helper gene and the genes for AAV replication (Rep) and structural (4) proteins in adherent HEK293 cells (5, 6). A serious limitation of current methods is that they do not scale well. Human clinical trials, for example, can require more than 1000–5000 175-cm2 flasks of 293 cells (7). Large-scale production of rAAV is hard to achieve due to low transfection and diffusion efficiencies (6), though the production of
⁎ Corresponding author. Tel.: +886 32118800 ext. 5291; fax: +886 32118668. E-mail address:
[email protected] (L.-D. Shiau). † These authors contributed equally to this work.
rAAV can be performed in suspension systems (8, 9). Thus, large-scale production of rAAV is a major challenge in clinical therapeutic applications. A novel production method has recently been developed for rAAV based on a baculovirus expression vector system (BEVS) in insect cells (7, 10–14). Baculovirus derived from the Autographa californica nuclear polyhedrosis virus (AcNPV) is widely used for scalable protein production in insect cells. The advantage of the baculovirus/ insect system is that it is easy to scale to high cell densities in suspension culture. The system has a strong polyhedrin or p10 promoter that can achieve high expression of foreign proteins while retaining biological activity via post-translation modifications (7, 15). Production of the viral vector in the insect cell system was first used to make rAAV2 composed of three types of baculoviruses for the production of all the structural proteins, replication proteins and the target gene flanked by AAV-ITRs (Bac-RC, Bac-Rep and Bac-GFP), respectively (7). Researchers have attempted to scale up rAAV yield by using different types of bioreactors, including shaking-flask (8, 16), stirredtank (6, 12) or packed-bed reactors (17). However, very little work has focused on the culture process for improving rAAV yield in the baculovirus/insect system. The shaking-flask and stirred-tank are the most commonly used types of bioreactors, but they have the drawback that the large shear stress they impose slows growth in the early growth phase. Alternatively, higher initial cell density could
1389-1723/$ - see front matter © 2010, The Society for Biotechnology, Japan. All rights reserved. doi:10.1016/j.jbiosc.2010.02.004
188
LIU ET AL.
J. BIOSCI. BIOENG.,
shorten the cultures. However, it is not recommended due to the higher costs. In order to decrease cost as well as to neglect the effect of shear stress, the fed-batch culture appears to be an appropriate way. Also, it shows that the glucose and glutamine can be maintained at low levels in the fed-batch cultures, preventing the accumulation of inhibitory metabolites like lactate and ammonia (18). Higher cell densities can be yielded in a short amount of time, enhancing protein production in the cells (19). Here we demonstrate a more efficient rAAV production strategy using a fed-batch culture for the baculovirus/insect cell system. Using such approach, we are able to start at lower cell densities then achieving high cell densities in a short period of time. After co-infection with the three plasmids of baculoviruses, the fed-batch culture gives the highest rAAV yield. It indicates a rapid and economical way of producing rAAV by the utilization of the fed-batch culture system. MATERIALS AND METHODS Cell culture Spodoptera frugiperda Sf21 cells were grown at 27 °C in spinner flasks containing TNM-FH medium (Sigma, St. Louis, MO, USA) supplemented with 10% FCS. 3T3 cells were maintained in DMEM medium (GIBCO, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum at 37 °C in a 5% CO2 atmosphere. Recombinant baculoviruses Three plasmids, pFBDLSR, pFBDVPm11 and pFBGFPR, were kindly provided by Dr. R.M. Kotin from the National Institutes of Health (Bethesda, MD, USA) and have been described previously (7). The recombinant baculoviruses were generated with the Bac-to-Bac baculovirus expression system (Life Technologies, Carlsbad, CA, USA). The baculoviruses were amplified in Sf21 insect cells grown in suspension in TNM-FH medium. rAAV vector production and purification To produce the rAAV vector, Sf-21 cells were infected with recombinant baculoviruses at a density of 1 × 106 cells/ml. On day 4 post-infection, the cells were harvested and lysed by freeze–thaw, and then heated at 60 °C for 15 min to inactivate the baculovirus for infectious viral particle (IVP) determination. Western blot analysis of viral proteins The primary antibodies used to detect the Cap and Rep proteins were mouse monoclonal antibodies (1:50 dilution; American Research Products, Belmont, MA, USA) against AAV Cap (VP1, VP2 and VP3) and Rep (Rep78 and Rep52) proteins. The blots were detected by using the ECL kit (GE Healthcare, Buckinghamshire, UK). Quantification of baculovirus titer with real-time PCR The amount of baculovirus was quantified with a real-time PCR assay. The viral particles were digested with DNase I at 37 °C for 1 h followed by digestion with proteinase K at 55 °C for 1 h. The viral DNA was purified using a phenol/chloroform/isoamyl alcohol extraction. DNA samples were diluted and quantified by real-time PCR with specific primers (CMV 5′-TATGGAGTTCCGCGTTACATAACTTACGGT-3′ and 5′-GACTAATACGTAGATGTA-CTGCCAAGTAGG-3′; IE15′-CCCGTAACGGACCTCGTACTT-3′ and 5′-TTATCGAGATTTATTTGCATACAA-3′) on an iQ5 PCR system (Bio-Rad, Hercules, CA, USA). Dilution viral DNAof pFBGFPR, pFBDVPm11 and pFBDLSR were used as a genome copy standard. Quantification of infectious viral particle (IVP) titers of rAAV with flow cytometry 3T3 cells were seeded in a 24-well culture plate at 1–5 × 105 cells per well and infected with serial dilutions of the virus in 1 ml culture media. The cells were incubated at 37 °C for 48 hrs, and then harvested and resuspended in 0.5 ml of PBS buffer. Cells that were only infected with the Bac-GFP virus were used as negative controls. For each sample, the percentage of fluorescent cells was determined by flow cytometry using a minimum of 10,000 cells. IVP titer was calculated as follows: Titer (IVP/ml) = % GFP+ cells × dilution factor × number of cells at virus infection Effect of baculovirus ratio and multiplicity of infection (MOI) on rAAV production Sf-21 cells were seeded in 6-well culture plates at 1 × 106 cells per well and infected with different ratios of triple baculovirus (ranging from 1:1:1 to 1:9:9 BacGFP: Bac-Rep: Bac-VP2), with the MOI of Bac-GFP equal to one. Additionally, Sf-21 cells were infected with Bac-GFP at MOIs of 1, 3, 5, 7, 9 and 10 (using a triple baculovirus infection ratio of 1:9:9). At 4 days post-infection, the cells were harvested and lysed by freeze–thaw and the rAAV titer was calculated as described above. rAAV production in spinner flask Sf-21 cells were performed in 500-ml spinner flasks (Bellco, Vineland, NJ, USA) at initial cell density 3 × 105 cells/ml. When the cell density reached 1 × 106 cells/ml, the cells were harvested and re-seeded at initial cell densities ranging from 5 × 105 to 3 × 106 cells/ml at culture volume of 50 ml. The cells were infected with Bac-GFP at MOI = 5 (using a triple baculovirus infection ratio of 1:9:9). At 4 days post-infection, cells were collected and IVP titer was determined. Batch and fed-batch culture The batch and fed-batch experiments were carried out in 125-ml spinner flasks (Bellco). Sf-21 cells were seeded at initial cell 5 density 3 × 10 cells/ml at initial culture volumes of 80 ml in batch culture and 50 ml in fed-batch culture. At fed-batch culture, the 10 ml of fresh media were added on days 1, 2 and 3. Cells were collected and determined daily.
The 125-ml spinner flasks were inoculated 80 × 105 cells in 80 ml media for batch culture and 50 ml media for fed-batch culture. Ten milliliters of fresh media were added into fed-batch culture daily (day 1–day 3). The cells were infected with three baculoviruses (Bac-GFP: Bac-Rep: Bac-VP = 1:9:9) in different types of culture strategies. The details of culture strategies were described in Table 1. In batch 5 and fed-batch 5, the cells were initially infected with three baculoviruses at Bac-GFP MOI = 5 on day 0. In fed-batch 1 2 2 and fed-batch 2 2 1, the cells were initially infected with three baculoviruses at Bac-GFP MOI = 1 and MOI = 2 on day 0, individually. On day 1, fed-batch 1 2 2 and fed-batch 2 2 1 were infected with three baculoviruses at BacGFP MOI = 2; fed-batch 1 2 2 and fed-batch 2 2 1 were infected with three baculoviruses at Bac-GFP MOI = 2 and MOI = 1 on day 2, individually. Cells were collected and IVP titer was determined daily.
RESULTS Production of baculovirus stock The three plasmids provided by Dr. Kotin were used to produce three different types of baculovirus (Bac-GFP, Bac-Rep, and Bac-VP) using the Bac-to-Bac baculovirus expression system. To verify baculovirus expression and genome stability, Sf-21 cells were infected individually with Bac-GFP, Bac-Rep, and Bac-VP in different passages (P1, P2). Bac-GFP expression was examined using fluorescence microscopy. Bac-Rep and Bac-VP were detected by Western blot on day 4 post-infection (Figs. 1A–C). Fig. 1A shows that GFP expression was observed in P1 and P2 of Bac-GFP. Fig. 1B and C shows that the protein expression of Rep78 and Rep52 in Bac-Rep were at the same levels in P1 and P2; similar results were seen in VP1, VP2, and VP3 expression from Bac-VP. These results confirm that cell passages (up to P2) do not decrease the function or expression of the three baculoviruses. rAAV production by baculovirus in Sf-21 cell To substantiate that rAAV was produced in Sf-21 cells with the three baculoviruses, Sf-21 cells were co-infected with Bac-GFP, Bac-Rep and Bac-VP (MOI of 10 for each baculovirus) to produce rAAV. As a negative control, cells were infected with Bac-GFP alone (Fig. 2A). Three days after coinfection, cells were pelleted, lysed and purified with a CsCl density gradient centrifugation. The yield of purified rAAV was detected by real-time PCR (data not shown) and rAAVs were transduced into 3T3 cells. Fig. 2B reveals that many of the 3T3 cells could express GFP 14 hours after transduction with purified rAAV. This result indicates that Sf-21 cells can produce infectious rAAV. Effect of rAAV production at different ratios of triple baculoviruses and MOIs In order to study the effect of the proportion of the three baculoviruses on rAAV production, a range of baculoviruses MOI ratios were tested. Sf-21 cells were coinfected with Bac-GFP at MOI = 1 and equivalent MOIs for Bac-Rep and Bac-VP2 from 1:1 to 9: 9 (Fig. 3A). It was found that the low levels of Bac-GFP, combined with higher level of Bac-Rep and BacVP, produced the highest rAAV yields. When Bac-Rep and Bac-VP levels were decreased, rAAV yield was decreased. The highest rAAV yield was obtained (9.2 × 105 IVP/ml) at for the triple ratio Bac-GFP: Bac-Rep: Bac-VP= 1:9:9. In contrast, the lowest rAAV yield was observed for Bac-GFP: Bac-Rep: Bac-VP= 1: 1: 1 (0.9 × 105 IVP/ml). To determine the optimal baculovirus MOI for rAAV production, Sf-21 cells were co-infected with Bac-GFP, Bac-Rep, and Bac-VP at a ratio of 1:9:9 and were measured with Bac-GFP MOI values of 1, 3, 5, 7, 9, 10 and 100. The rAAV yield was analyzed by flow cytometry on day 4 post-infection (Fig. 3B). An MOI of 1 had the lowest rAAV yield (1.1 × 106 IVP/ml). The rAAV yields increased from MOI 1 to 10, though this effect eventually saturated: the rAAV yield was not significantly different from MOI = 9 to MOI = 100 (data not shown). The most dramatic increase in rAAV yields occurred between MOI = 1 (1.1 × 106 IVP/ml) and 7 (52.8 × 106 IVP/ml). Sf-21 infected with baculovirus at MOI = 9 (81.1 × 106 IVP/ml) had slightly higher rAAV yields than at MOI = 7 (52.8 × 106 IVP/ml) and 5 (36.5 × 106 IVP/ml). Therefore, we chose MOI = 5 for economic considerations.
VOL. 110, 2010
rAAV PRODUCTION IN FED-BATCH CULTURE WITH INSECT CELL
189
TABLE 1. Summary of batch and fed-batch culture using different feeding strategies. Infected cell number × 106
Batch 5 Fed-batch 5 Fed-batch 1 2 2 Fed-batch 2 2 1
80 80 80 80
Day 0
Day 1
Day 2
Day 3
Total volumea
Infected MOIb Bac-GFP
Total volumea
Infected MOIb Bac-GFP
Total volumea
Infected MOIb Bac-GFP
Total volumea
Infected MOIb Bac-GFP
80 50 50 50
5 5 1 2
80 60 60 60
– – 2 2
80 70 70 70
– – 2 1
80 80 80 80
– – – –
a: volume unit is ml b: The MOI ratio of three baculovirus is 1: 9 : 9.
Effect of cell density on rAAV production To optimizing cell density for production of rAAV, Sf-21 cells were cultured in 500-ml spinner flasks at initial cell density 3 × 105 cells/ml. When the cell density reached 1 × 106 cells/ml, the cells were harvested and reseeded at 5 × 105, 1 × 106, 2 × 106 and 3 × 106 cells/ml and supplemented with culture medium in a total volume of 50 ml in 125 ml spinner flasks. Then such cells were infected with three baculoviruses (Bac-GFP MOI = 5, using a triple baculoviruses infection ratio of 1:9:9). On day 4 post-infection, the rAAV yield was obtained by analyzing its GFP expression with flow cytometry (Fig. 4). It was found that the rAAV yield increased following with the increasing cell density from 5 × 105 to 1 × 106 cells/ml. However, if the density was
higher than 1 × 106 cells/ml, the yields declined. Since the cell density of 1 × 106 cells/ml produced the highest rAAV yield (108.9 × 106 IVP/ml), so such density was used for later batch and fed-batch culture experiments. Comparison of batch culture and fed-batch culture In order to obtain high cell densities to improve and optimize rAAV yields, the experiments were run in spinner flasks using two different feeding strategies. Cell growth results for the batch culture and fed-batch culture in 125 ml spinner flasks are shown in Fig. 5. The same initial cell density of 3 × 105 cells/ml was used to seed both batch (80 ml) and fed-batch (50 ml) culture spinner flasks. After receiving fresh medium (arrows) on days 1, 2 and 3 in the fed-batch culture, the total volume of medium was the same as the volume in batch culture; the cell densities (about 1 × 106 cells/ml) were also similar for the two systems. Fed-batch cultures seeded with fewer cells reached cell densities close to those in batch culture by day 3. Additionally, the cell death rate in fed-batch cultures was low: the number of cells was 2fold higher than in the batch culture on day 6. These results suggest that the growth rate of insect cells in the fed-batch culture is faster than in the batch culture and that growth could be achieved over a prolonged period. Results showing cell viability and rAAV yield in batch and fedbatch cultures infected with baculovirus are presented in Fig. 6. The cell viability of batch 5 and fed-batch 5 dropped quickly after infection with baculovirus (Fig. 6A). On day 3 post-infection, the cell viability of batch 5 and fed-batch 5 dropped to 55% and 30%, respectively. Cell viability for both cultures was down to 20% on day 5 post-infection. In batch and fed-batch culture with baculovirus infection, the rAAV yield increased with increasing culture time (Fig. 6B). On day 5 postinfection in batch culture, the best rAAV yield was 1.22 × 108 IVP/ml. This yield could not be increased further by prolonging the culture time. For the fed-batch 5 culture, the rAAV yield was higher than for the batch culture. A yield of 2.13 × 108 IVP/ml for rAAV was attained at day 4 post-infection and increased to 2.40 × 108 IVP/ml at day 5 postinfection. The maximal rAAV yield in fed-batch 5 (213 IVP/cell) was 2.6-fold higher than batch 5 (83 IVP/cell) at day 4 post-infection and 2-fold higher at day 5 post-infection (fed-batch 5: 240 IVP/cell, batch 5: 122 IVP/cell). We also fed baculovirus in the fed-batch culture system (Fig. 7). Fresh media and baculovirus (total Bac-GFP MOI = 5) were fed at the same time. Cell viability in fed-batches 1 2 2 and 2 2 1 increased slightly until day 2 post-infection and then decreased to 60% and 50% at day 3 post-infection (Fig. 7A). Similar results were observed in fedbatches 1 2 2 and 2 2 1. The rAAV yield of fed-batches 1 2 2 (1.30 × 108 IVP/ml) and 2 2 1 (0.97 × 108 IVP/ml) were lower than fed-batch 5 (2.13 × 108 IVP/ml), but higher than batch 5 (0.83 × 108 IVP/ml) at day 4 post-infection (Fig. 7B). On day 5 post-infection, the rAAV yields of fed-batches 1 2 2 and 2 2 1 were 1.46 × 108 IVP/ml and 1.10 × 108 IVP/ml. The fed-batch 5 culture system achieved higher rAAV yields than fed-batches 1 2 2 and 2 2 1.
FIG. 1. Protein expression levels of baculovirus. (A) Observation of GFP (Bac-GFP) expression with fluorescence microscopy. (B, C) Detection of Rep (Bac-Rep) and VP (Bac-VP) protein expression by Western blotting with anti-Rep and anti-VP antibody. All samples were harvested on day 4 post-infection with Sf-21 and presented two passages (P1, P2).
DISCUSSION The baculovirus has been used widely for recombinant protein production, though there is still some discussion on the effects of
190
LIU ET AL.
J. BIOSCI. BIOENG.,
FIG. 2. Confirmation of rAAV production by baculovirus co-infection. 3T3 cells were infected with samples purified by CsCl-banded density gradient centrifugation from Sf-21 cells: (A) infected with Bac-GFP only, (B) co-infected with Bac-GFP, Bac-Rep, and Bac-VP and observed by fluorescence microscopy.
passage. Aucoin et al. (13) demonstrated the effect of passage on protein and rAAV production. Protein production was decreased when the baculovirus was passaged more than two times. Some reports (11, 20) also indicate that unstable passaging could affect protein and rAAV production. In contrast, Hu et al. (17) determined that the virus was stable for up to four passages. To clarify this effect, we maintained virus stocks in early passages and compared protein expression between P1 and P2. There was no difference between P1 and P2 with three baculoviruses (Fig. 1). In order to produce rAAV consistently and avoid the passage effect, early passage from baculovirus stock is recommended. This infection system was based on triple virus co-infection, and we found that the ratio of the three baculoviruses was very important in obtaining optimal results. Meghrous et al. demonstrated that changing the ratios of the three baculovirus MOIs altered the infectious rAAV yields. They also found that the three viruses should be utilized at higher MOIs of 1, respectively (12). Aucoin et al. suggested that a minimum level of BacGFP was needed for high rAAV yields. Moreover, they found that equal amounts of BacRep and BacVP gave the highest rAAV yields. Based on their study, we tested the ratios of three baculoviruses from 1:1:1 to 1:9:9 to try to achieve the highest rAAV yield. The rAAV yields increased with increased ratios of Bac-Rep and Bac-VP to BacGFP. However, the rAAV yields did not rise significantly after the ratio approached 1:8:8 (Fig. 3A). In fact, overexpression of Bac-VP caused production of empty rAAV, which decreased cell viability and interfered with encapsidation of rAAV (13). Furthermore, high levels of Bac-Rep affected capsid protein production. At high levels of Bac-Rep and low levels of BacVP, the ratio of replication to capsid proteins that were produced were not balanced, possibly resulting in degradation of replication
proteins (13). This may explain the low titer of rAAV. Replication proteins are not only necessary for producing rAAV, but are also correlated with increasing and packaging genomes (21). To produce high rAAV titers we decided to use equal ratios of Bac-Rep and BacVP for this system. The optimal MOI of Bac-GFP was an important economic consideration. In this study, we demonstrated that increasing the Bac-GFP MOI increased rAAV yield. The highest rAAV yield was achieved when the MOI of Bac-GFP approached 100. However, the difference between an MOI of 100 and 10 was not significant enough to warrant the added cost (data not shown). We tested the MOI of Bac-GFP ranging from 1 to 10 and determined which MOI gave the highest rAAV yield. MOIs of 9 and 10 had similar yields, and both were higher than MOIs of 7 and 5. The benefits of lower a MOI for scale-up of rAAV production were also considered (22): higher MOI values can decrease production of viral stock and avoid the passage effect. When the MOI was increased from 7 to 10 (Fig. 3B), the rAAV yields were not dramatically increased (about 52.8 × 106–81 × 106 IVP/ml). An MOI of 7 for Bac-GFP-infected insect cells had a lower rAAV yield in fed-batch culture compared to batch culture (data not shown). Although, the rAAV yields were similar in fed-batch cultures with MOI = 5 and batch cultures with MOI = 7, we chose an MOI 5 in the fed-batch culture to reduce capital costs. The effect of cell density was also considered. Some researchers have shown that infecting insect cells at later stages of growth (higher cell density) increases protein yields (23, 24). Meghrous et al. showed that the insect cells infected with an overall MOI of 5 with a ratio of the three baculoviruses equal to 1:1:1, the cell density could be increased up to 7.5 × 106 cells/ml without affecting the cell productivity. They also observed that the cell productivity
VOL. 110, 2010
rAAV PRODUCTION IN FED-BATCH CULTURE WITH INSECT CELL
191
FIG. 5. Sf-21 growth in batch and fed-batch cultures. A cell density of 3 × 105 cells/ml was seeded in the 125 ml spinner flask with 80 ml or 50 ml for batch and fed-batch cultures, respectively. Fresh media (10 ml) was added on days 1, 2, and 3 (arrow) in fed-batch culture. The number of cells was determined by trypan blue exclusion.
protein production are dependent on the MOI and medium exchange (25). Therefore, at least three factors including the optimal MOI and cell density as well as fresh medium exchange
FIG. 3. Effect of different ratios of triple baculovirus and MOI on rAAV production. 3T3 cells were infected with samples from Sf-21 cells that had been infected with (A) different ratios of triple baculovirus (Bac-GFP = 1) or (B) different Bac-GFP MOIs (BacGFP: Bac-Rep: Bac-VP = 1: 9: 9) and analyzed by flow cytometry at 48 hours.
could be increased as early as infection at 2.5 × 106 cells/ml in the fresh medium. However, the largest reduction of cell productivity occurred between the infected cells at 1 and 2 × 106 cells/ml without replacing fresh medium, indicating that the medium is essential for the production of rAAV. Such findings are parallel with the previous reports (12, 25), and such may due to the oxygen limitations and nutrients uptake problems at high cell density (25). Also, it has been demonstrated that the effects of initial cell density at infection on
FIG. 4. Effect of cell density at infection on rAAV production. Different cell densities of Sf-21 cells were cultured in spinner flasks with baculovirus (Bac-GFP = 5, Bac-GFP: BacRep: Bac-VP = 1: 9: 9) and the yields of rAAV harvested at 4 days. 3T3 cells were infected with rAAV and analyzed by flow cytometry at 48 hours.
FIG. 6. Batch and fed-batch culture kinetics for the production of rAAV. The time course of cell viability (A) and rAAV yield (B) was measured during the culture. Arrows indicate the time of feeding with 10 ml of fresh medium (Bac-GFP MOI = 5, Bac-GFP: Bac-Rep: Bac-VP = 1: 9: 9).
192
LIU ET AL.
J. BIOSCI. BIOENG., culture required fewer cells in the initial stage and achieved similar cell densities to the batch culture after 4 days (Fig. 5). The fed-batch culture thus reduces the effort associated with preparing large numbers of cells and the culture time required to obtain the infected virus. The viability of insect cells infected with three baculoviruses decreased quickly in the fed-batch culture 3 days after infection. It is possible that the small amount of medium in the fed-batch culture at the initial step caused attachment between the cell and the virus. The baculoviruses attached to insect cells more efficiently and produced higher rAAV particles per cell (batch 122 IVP/cell, fed-batch 240 IVP/ cell) in fed-batch cultures than in batch cultures. In contrast, the cell viability of feeding-baculovirus (fed-batches 1 2 2 and 2 2 1) was slightly increased at 24–48 h post infection and decreased 3 days after infection. These results verify that cell viability is correlated with the number of baculoviruses attached to the insect cell. Furthermore, feeding with fresh medium provides essential nutrients and dilutes inhibitory metabolites for the baculovirus-infected cells so that they can continue producing rAAV particles (25). In this study, we developed a rAAV production process using fedbatch cultures for a baculovirus/insect system. We can cultivate insect cells using fewer cells while quickly achieving high cell densities. After co-infection with three baculoviruses, the fed-batch culture system achieved higher rAAV yields than the batch culture in a shorter time. The batch and fed-batch cultures have very similar costs; however, the rAAV yield was 2.6-fold higher in the fed-batch culture system compared with the batch culture system. We successfully produced high yields of rAAV by using a fed-batch culture system, providing a rapid and economical method for producing rAAV. ACKNOWLEDGMENTS
FIG. 7. Fed-batch culture kinetics for the production of rAAV. The time course of cell viability (A) and rAAV yield (B) was measured during the culture. The arrows indicate the time of feeding with 10 ml of fresh medium and baculovirus (1: Bac-GFP MOI = 1, 2: Bac-GFP MOI = 2, Bac-GFP: Bac-Rep: Bac-VP = 1: 9: 9).
We gratefully acknowledge Dr. R. M. Kotin for providing the baculovirus constructs used in this study. This project is supported by grants from National Science Council (098-2811-B-182-023 to CJY and CRS, 98-2320-B-182-014 to CRS) and Chang Gung University and Chang Gung Memorial Hospital (EMRPD170241 and CMRPD150381 to CRS). References
are crucial. In our data (Fig. 4), the cells that produced the largest rAAV yield for Bac-GFP at MOI 5 were at a density of 1 × 106 cells/ ml. Although, higher cell densities produced higher rAAV yields with other MOI, the process took longer and used more reagents. The benefit of decreasing the number of cells while using a suitable MOI was an important consideration in this research. Based on our studies, we demonstrated that the optimal cell density for economic efficiency is 1 × 106 cells/ml. In general, the typical cell culture media have excess nutrients of requirements by cells, such as glucose and glutamine. The excess nutrients induced an unnecessarily high uptake and produced waste of metabolites, like as lactate and ammonia, to inhibit cell growth. The waste of metabolites limited the cell growth approach high density and product yields. Fed-batch culture appears to be the appropriate strategy to achieve high cell density and enhanced protein yields. This strategy is able to maintain glucose and glutamine at low levels and shift cell metabolism to an efficient state with reduced waste metabolites production (18, 19). Also, fedbatch cultures have been shown able to avoid inappropriate osmolality problems and to reduce the medium costs (23, 26, 27). Moreover, there was no extra requirement for other devices and preventing the inconvenient handling of large volumes. It has been proposed that using fed-batch cultures could improve cell density and protein yields over batch culture methods in many production systems (19, 23, 28, 29). However, few researchers have focused on the baculovirus/insect system (23, 27). In this work, the fed-batch
1. Kay, M. A., Glorioso, J. C., and Naldini, L.: Viral vectors for gene therapy: the art of turning infectious agents into vehicles of therapeutics, Nat. Med., 7, 33–40 (2001). 2. Mohr, L. and Geissler, M.: Gene therapy: new developments, Praxis, 91, 2227–2235 (2002) (Bern 1994). 3. Mancheno-Corvo, P. and Martin-Duque, P.: Viral gene therapy, Clin. Transl. Oncol., 8, 858–867 (2006). 4. Caplen, N. J.: Gene therapy progress and prospects. Downregulating gene expression: the impact of RNA interference, Gene. Ther., 11, 1241–1248 (2004). 5. Xiao, X., Li, J., and Samulski, R. J.: Production of high-titer recombinant adenoassociated virus vectors in the absence of helper adenovirus, J. Virol., 72, 2224–2232 (1998). 6. Durocher, Y., Pham, P. L., St-Laurent, G., Jacob, D., Cass, B., Chahal, P., Lau, C. J., Nalbantoglu, J., and Kamen, A.: Scalable serum-free production of recombinant adeno-associated virus type 2 by transfection of 293 suspension cells, J. Virol. Methods, 144, 32–40 (2007). 7. Urabe, M., Ding, C., and Kotin, R. M.: Insect cells as a factory to produce adenoassociated virus type 2 vectors, Hum. Gene. Ther., 13, 1935–1943 (2002). 8. Hildinger, M., Baldi, L., Stettler, M., and Wurm, F. M.: High-titer, serum-free production of adeno-associated virus vectors by polyethyleneimine-mediated plasmid transfection in mammalian suspension cells, Biotechnol. Lett., 29, 1713–1721 (2007). 9. Park, J. Y., Lim, B. P., Lee, K., Kim, Y. G., and Jo, E. C.: Scalable production of adenoassociated virus type 2 vectors via suspension transfection, Biotechnol. Bioeng., 94, 416–430 (2006). 10. Sollerbrant, K., Elmen, J., Wahlestedt, C., Acker, J., Leblois-Prehaud, H., LattaMahieu, M., Yeh, P., and Perricaudet, M.: A novel method using baculovirusmediated gene transfer for production of recombinant adeno-associated virus vectors, J. Gen. Virol., 82, 2051–2060 (2001). 11. Kohlbrenner, E., Aslanidi, G., Nash, K., Shklyaev, S., Campbell-Thompson, M., Byrne, B. J., Snyder, R. O., Muzyczka, N., Warrington Jr., K. H., Jr., and Zolotukhin,
VOL. 110, 2010
12.
13.
14.
15. 16.
17.
18.
19.
20.
S.: Successful production of pseudotyped rAAV vectors using a modified baculovirus expression system, Mol. Ther., 12, 1217–1225 (2005). Meghrous, J., Aucoin, M. G., Jacob, D., Chahal, P. S., Arcand, N., and Kamen, A. A.: Production of recombinant adeno-associated viral vectors using a baculovirus/ insect cell suspension culture system: from shake flasks to a 20-L bioreactor, Biotechnol. Prog., 21, 154–160 (2005). Aucoin, M. G., Perrier, M., and Kamen, A. A.: Production of adeno-associated viral vectors in insect cells using triple infection: optimization of baculovirus concentration ratios, Biotechnol. Bioeng., 95, 1081–1092 (2006). Negrete, A. and Kotin, R. M.: Strategies for manufacturing recombinant adenoassociated virus vectors for gene therapy applications exploiting baculovirus technology, Brief. Funct. Genomic. Proteomic., 7, 303–311 (2008). Luckow, V. A.: Baculovirus systems for the expression of human gene products, Curr. Opin. Biotechnol., 4, 564–572 (1993). Farson, D., Harding, T. C., Tao, L., Liu, J., Powell, S., Vimal, V., Yendluri, S., Koprivnikar, K., Ho, K., Twitty, C., Husak, P., Lin, A., and Snyder, R. O., et al.: Development and characterization of a cell line for large-scale, serum-free production of recombinant adeno-associated viral vectors, J. Gene Med., 6, 1369–1381 (2004). Huang, K. S., Lo, W. H., Chung, Y. C., Lai, Y. K., Chen, C. Y., Chou, S. T., and Hu, Y. C.: Combination of baculovirus-mediated gene delivery and packed-bed reactor for scalable production of adeno-associated virus, Hum. Gene. Ther., 18, 1161–1170 (2007). Lee, Y. Y., Wong, K. T., Nissom, P. M., Wong, D. C., and Yap, M. G.: Transcriptional profiling of batch and fed-batch protein-free 293-HEK cultures, Metab. Eng., 9, 52–67 (2007). Kuwae, S., Ohda, T., Tamashima, H., Miki, H., and Kobayashi, K.: Development of a fed-batch culture process for enhanced production of recombinant human antithrombin by Chinese hamster ovary cells, J. Biosci. Bioeng., 100, 502–510 (2005). Negrete, A., Yang, L. C., Mendez, A. F., Levy, J. R., and Kotin, R. M.: Economized large-scale production of high yield of rAAV for gene therapy applications exploiting baculovirus expression system, J. Gene Med., 9, 938–948 (2007).
rAAV PRODUCTION IN FED-BATCH CULTURE WITH INSECT CELL
193
21. Bleker, S., Pawlita, M., and Kleinschmidt, J. A.: Impact of capsid conformation and Rep-capsid interactions on adeno-associated virus type 2 genome packaging, J. Virol., 80, 810–820 (2006). 22. Maranga, L., Brazao, T. F., and Carrondo, M. J.: Virus-like particle production at low multiplicities of infection with the baculovirus insect cell system, Biotechnol. Bioeng., 84, 245–253 (2003). 23. Elias, C. B., Zeiser, A., Bedard, C., and Kamen, A. A.: Enhanced growth of Sf-9 cells to a maximum density of 5.2 × 10(7) cells per mL and production of betagalactosidase at high cell density by fed batch culture, Biotechnol. Bioeng., 68, 381–388 (2000). 24. Chico, E. and Jager, V.: Perfusion culture of baculovirus-infected BTI-Tn-5B1-4 insect cells: a method to restore cell-specific beta-trace glycoprotein productivity at high cell density, Biotechnol. Bioeng., 70, 574–586 (2000). 25. Carinhas, N., Bernal, V., Yokomizo, A. Y., Carrondo, M. J., Oliveira, R., and Alves, P. M.: Baculovirus production for gene therapy: the role of cell density, multiplicity of infection and medium exchange, Appl. Microbiol. Biotechnol., 81, 1041–1049 (2009). 26. Marteijn, R. C., Jurrius, O., Dhont, J., de Gooijer, C. D., Tramper, J., and Martens, D. E.: Optimization of a feed medium for fed-batch culture of insect cells using a genetic algorithm, Biotechnol. Bioeng., 81, 269–278 (2003). 27. Jardin, B. A., Montes, J., Lanthier, S., Tran, R., and Elias, C.: High cell density fed batch and perfusion processes for stable non-viral expression of secreted alkaline phosphatase (SEAP) using insect cells: comparison to a batch Sf-9-BEV system, Biotechnol. Bioeng., 97, 332–345 (2007). 28. Gurramkonda, C., Adnan, A., Gabel, T., Lunsdorf, H., Ross, A., Nemani, S. K., Swaminathan, S., Khanna, N., and Rinas, U.: Simple high-cell density fed-batch technique for high-level recombinant protein production with Pichia pastoris: application to intracellular production of Hepatitis B surface antigen, Microb. Cell Fact, 8, 13 (2009). 29. Faulkner, E., Barrett, M., Okor, S., Kieran, P., Casey, E., Paradisi, F., Engel, P., and Glennon, B.: Use of fed-batch cultivation for achieving high cell densities for the pilot-scale production of a recombinant protein (phenylalanine dehydrogenase) in Escherichia coli, Biotechnol. Prog., 22, 889–897 (2006).