4.2 Virus production in semi-continuous mode
4.2.1 Semi-continuous production of modified vaccinia Ankara virus in AGE1.CR.pIX
44
Table 4. 1. Overview of MVA process parameters, virus titers and productivity obtained in batch, semi-continuous, and continuous experiments.
Expe rime nt a Ce ll Passage Numbe r
Ce ll Conc. at toi [ˣ106 ce lls/mL] i
Me dium Manufacture r
Virus Dilution rate s; F3
b
RT in SVB or VB [h] c
Volume SVB or VB [mL]
Days of Ope ration
[d] d
Maximum Virus Tite r [virions/mL]
Total Numbe r of Virions Produce d
[virions] e
Total Harve st Volume [mL]
Time Yie ld [virions/h]
Space -time Yie ld [virions/(L h)]
BM-A 82 3.2 Biochrom MVA-CR19 B 72 50 8.0 3E+08 2E+10 50 8.2E+07 1.6E+09
BM-B 41 2.5 Merck/Biochrom MVA-CR19 B 72 50 8.0 1E+08 5E+09 50 2.6E+07 5.2E+08
BM-C 41 2.7 Merck/Biochrom MVA-CR19 B 72 50 8.0 3E+07 2E+09 50 8.2E+06 1.6E+08
BM-average f - - Merck/Biochrom MVA-CR19 B 72 50 8.0 1E+08 7E+09 50 3.6E+07 7.3E+08
2 Parallel batchesg - - Merck/Biochrom MVA-CR19 B 72 1290 17.0 1E+08 4E+11 2580 8.9E+08 3.4E+08
2 Parallel batches h - - Merck/Biochrom MVA-CR19 B 72 1290 26.0 1E+08 5E+11 3870 8.7E+08 2.2E+08
SM25-A 50 10.5 Biochrom MVA-CR19 3·D1 = D2 ; F3 = F3 25 65 22.0 2E+09 2E+11 1136 3.7E+08 3.3E+08
SM25-B 90 12.9 Biochrom MVA-CR19 3·D1 = D2 ;F3 = F3 25 65 22.0 2E+09 5E+11 1004 1.0E+09 1.0E+09
SM25-MOCK 82 - Biochrom MOCK 3·D1 = D2 ;F3 = F3 25 65 18.0 MOCK MOCK 726 MOCK MOCK
SM35-A 73 12.1 Biochrom MVA-CR19 2·D1 = D2 ; F3 = F3 35 98 12.0 3E+08 2E+10 816 7.6E+07 9.3E+07
SM35-B 40 7.47 Merck/Biochrom MVA-CR19 2·D1 = D2 ;F3 = F3 35 98 19.0 1E+09 2E+11 1157 4.1E+08 3.6E+08
SM35-C 40 4.42 Merck/Biochrom MVA-CR19 2·D1 = D2 ;F3 = 0 35 65 18.0 3E+05 3E+07 649 7.4E+04 1.1E+05
SM64 73 11.8 Biochrom MVA-CR19 1·D1 = D2 ; F3 = F3 64 198 12.0 6E+08 6E+10 1084 2.0E+08 1.8E+08
SG25 69 5.72 Merck/Biochrom MVA-CR19.GFP 3·D1 = D2 ;F3 = F3 25 62 19.5 1E+08 3E+10 1148 6.0E+07 5.2E+07
SG40 69 6.01 Merck/Biochrom MVA-CR19.GFP 2·D1 = D2 ;F3 = F3 40 120 19.5 6E+09 5E+11 1208 1.0E+09 8.7E+08
T25 50 9.19 PAA MVA-CR19 3 D1 = D2 ; F3 = F3 25 440 21.7 6E+08 6E+11 7100 1.2E+09 1.7E+08
I cell concentration at time of infection.
h two parallel 645 mL batch bioreactors; calculations were carried out assuming 3 cycles (26 d), because it approaches the operational time of the TSB experiment (T25; 3 weeks). The complete time course of such a process is shown in Figure 6.
g two parallel 645 mL batch bioreactors; calculations were carried out assuming 2 batch-cycles, because it approaches the operational time of the SSC cultivations (2 weeks). Note: the TY is valid only for a specific cultivation scale, while the STY is independent of the cultivation scale. The complete time course of such a process is shown in Figure 6.
a T= Two-stage continuous bioreactor; S=semi-continuous small scale cultivation; B=Batch; M=MVA-CR19 strain; G= MVA-CR19.GFP strain; XX = XX hours (25 h,35 h or 64 h) of residence time in the VB or the SVB.
f the average TCID50 titer of batch A, B and C was estimated to be 1x108 virions/mL
d considering a batch with 4 days of cell growth in all processes, 3 days of virus production and 1 day for cleaning and sterilization.
b F3 = D1 (V2 + V1) - F1 with D1 the dilution rate of CB or SCB, V, thevolume of each vessel, and F, the flow rate.
c RT = residence time; VB = Virus Bioreactor; the value shown for batch cultures corresponds to the harvest time (h p.i).
e this value corresponds to the total number of virions produced after adding the virus collected from each harvests. This was calculated by multiplying the TCID50 of each harvest by its volume.
A semi-continuous steady-state production of AGE1.CR.pIX cells was successfully maintained over two weeks of cultivation. Hence, it was important to determine µmax during the batch phase (0.02 h-1), which was slightly higher than in the batch phase of 1 L scale TSB cultivations (0.016 h-1) . Also, the sampling before and after the medium exchanges resulted in reductions of WV that made it difficult reaching a steady state. Accordingly, taking a sample only before the medium exchange was the preferred option in most of the experiments.
Also, a drop in cell viability in SVB was observed in all cultivations after some days p.i. This was a consequence of virus replication and, interestingly, was observed from day 10 p.i in SM25-A experiment, and from day 4 p.i. in SM25-B (both experiments were operated at the same RT;
cell viability data of SM25-B is not shown). Most likely, the reason for the delayed viability drop of SM25-A was a dilution of the virus concentration below 1×105 virions/mL after the first harvest.
MVA virus production in semi-continuous mode. SVB was infected with MVA virus after 3-4 days of batch cell growth and the semi-continuous mode with harvesting was started 12 h p.i.
Virus titers showed 8-10 days of a transient phase followed by a stationary phase. TCID50 values between 1·107 and 1×109 virions/mL, were obtained among all SSC experiments.
Virus titers obtained from the SSC showed a transient and a stationary phase. TCID50 values between 1×107 and 1×109 virions/mL were obtained and were in accordance with those of batch cultivations, described in section 4.1.1, and also comparable to those of published data [18] [9].
Also, an important result that was first observed in this small-scale system was that virus titers in the stationary phase oscillated in order of magnitudes not larger than 2 log10 which is clearly less than what was observed previously for IAV [26]. These oscillations were most likely produced by errors in the MVA virus titration assay and by variations in the cell concentration in SVB. This suggested that MVA virus can be produced in continuous mode without interference by defective particles.
Two experiments were operated at 25 h RT in SVB. Virus titers from SM25-A followed a similar pattern to the TSB system from day 4 p.i onwards, reaching similar final titers. The second experiment, SM25-B (white squares, Figure 4.3 D), was more precisely operated and the virus titer dynamics was even closer to those obtained with the TSB system (operated also at 25 h RT in VB, as shown later in section 4.3.1). Therefore, these two experiments suggested that the SSC can serve as a scale down model of the TSB system. Previous works have demonstrated that semi-continuous cultivations can be very reliable to approximate semi-continuous cultures [156]. More recently, mammalian cell kinetics in continuous systems was studied using semi-continuous
47 cultures in shake flasks [157]. Thus, the cell growth and MVA virus titer results showed that these concepts could be also applied for TSB systems using shaker flasks.
Impact of RT and addition of fresh medium in MVA virus titers. Three SSC experiments with 25, 35 and 64 h RT in SVB were carried out as shown in Figure 4.3 D. TCID50 titers obtained with all RT experiments showed a common pattern consisting of an initial transient phase followed by a stationary phase. The MOI of 0.05, used in all experiments, led to initial virus concentrations close to 1.0×105 virions/mL. The fastest increase in virus titers was obtained with 25 h RT in SVB (experiment SM25-B) with approx. 1×108 virions/mL at 48 h p.i. Similar virus titers were obtained with 35 and 64 h RT in SVB but at 144-192 h p.i.(SM35-B and SM64 experiments, respectively).
Figure 4. 3. Semi-continuous propagation of MVA-CR19 virus in a two-stage cultivation system using shaker flasks (SSC). The data of A), B) and C) belong to one representative SSC experiment of a total of nine experiments with variations in RT and medium addition in SVB (SM25-A, Table 4.2). A) Viable AGE1.CR.pIX cells concentration in SCB (circles) and SVB (squares). B) Viability (white) and pH value (grey) of SCB (circles) and SVB (squares). C) Concentration of glucose (white) and lactate (grey) in SCB (circles) and SVB (squares). D) MVA TCID50 titers of the SSC experiments (squares) SM25-A (grey), SM25-B (white), SM35-A (grey with +), SM35-B (black), and SM64 (grey with X). One SSC experiment, SM35-C, was carried out without addition of fresh medium into SVB (white-circles). The dashed line represents the time of infection. The first harvest was carried out 8-12 h p.i..
48 In addition, one experiment was carried out to determine the impact of removing V3 from SVB (SM35-C). As shown in Figure 4.3 D, this experiment resulted in the lowest virus titers among all experiments, with values not exceeding 1×105 virions/mL.
Although 35 and 64 h RT showed comparable titers, 64 h RT would not be a good option for scale-up as long RTs might end up in culture conditions with low levels of glucose and high lactate and ammonia that are clearly not beneficial for both cell and virus propagation.
Interestingly, TCID50 titers at steady state were very similar for different RT experiments and therefore, the TCID50 seems to be independent of the RT.
Finally, one experiment was performed to investigate the impact of removing the addition (V3) of fresh medium on the virus titers (SM35-C, Table 4.1; Figure 4.3 D). The idea of this experiment was to test if the productivity can be increased by keeping virus titers, while decreasing medium consumption. Surprisingly, the lowest TCID50 titers were obtained from this experiment, with values between 1×103 and 1×105 virions/mL. Therefore, addition of fresh medium in the virus bioreactor is an important process variable to keep yields. This is in agreement with previous studies that showed that addition of fresh medium at time of infection is required for optimal virus propagation [158]. Adding fresh medium might play a role on diluting and faster washing-out of signaling molecules and metabolites that are not beneficial for cell growth, metabolism and, therefore, virus replication.
4.2.2 Semi-continuous production of influenza A virus in MDCK.SUS2 cells and in