3.3.1 Growth profile, sugar consumption, OUR and SOUR
Figure 3a shows the growth profile and sugar consumptions of the transgenic rice cell culture grown in NB+0.5xS medium in the pilot-scale bioreactor using single-stage batch operation with simplified bioreactor processes. The rice cells continued growing until day 8 of cultivation; no lag phase was not observed in this run (data not shown). The initial and final biomass concentrations in the growth phase were 2.05 ± 0.05 and 5.85 ± 0.13 g DW/L, respectively (Table S1). The µmax and τD were 0.16 ± 0.03 day-1 and 4.4 ± 0.9 day, respectively (Table S1), which are comparable to the µmax of 0.20 ± 0.04 day-1 and τD of 3.5 ± 0.8 day obtained in the lab-sale bioreactor run under similar kLa and simplified bioreactor operation (Macharoen et al., 2020) but substantially different from what was obtained in the cyclical semicontinuous operation as described earlier. Tanaka et al., reported that Catharanthus roseus cell suspension cultures in the shake flask with two baffles had lower growth rate and maximum biomass concentration than the shake flask without baffles (Tanaka et al., 1988). Thus, the presence of baffles in the semicontinuous run likely generates different hydrodynamic stress and oxygen transfer capability compared to the batch run, resulting in different µmaxand τD.
Sucrose concentration dramatically declined during the first few days of cultivation and reached undetectable levels by day 5 of cultivation, while glucose concentration became undetectable at day 8 of cultivation, indicating the start of the induction phase, or 0 dpi (Figure 3a). This sugar consumption pattern is also observed in the lab-scale culture using the same bioreactor process and kLa (Macharoen et al., 2020). In other words, maintaining kLa as a scale-up parameter likely provides equivalent growth kinetic parameters, i.e. µmax, τD and rate of sugar consumption between lab-scale and pilot-scale transgenic rice cell suspension cultures.
The OUR pattern in this run (Figure 3b) is similar to cycles 3-4 in the semicontinuous run presented in previous Section 3.1.2 (Figure 1b) in which the maximum OUR of 1.76 mmol O2L-1 h-1 (Table S1) was reached simultaneously with the start of sucrose depletion (2-4 days prior to the induction phase). In addition, the OUR and the metabolic activity (Figure S1b) profiles are alike, indicating that rice cells are likely most metabolically active at the beginning of sucrose depletion. The SOUR reached its maximum at 0.46 mmol O2 g DW-1 h-1 (Table S1), which is equivalent to what obtained in the lab-scale culture under the same conditions (Macharoen et al., 2020), at the early exponential growth phase and then gradually decreased toward the end of growth phase (Figure 3b). A possible explanation for this is that the portion of active cell aggregates to total cell aggregates at the beginning of this run was greater than for cycles 3-4 in which the proportion of accumulated nonviable cells may increase over the long cultivation period. During the induction phase, OUR, SOUR and metabolic activity dramatically decreased as expected. The DO and pH profiles in this run (Figure S2) are also similar to the profiles found in the lab-scale study (Macharoen et al., 2020) suggesting that maintaining kLa in this study is adequate to reproduce growth kinetics (e.g. µmax, τD, OUR and SOUR) and culture conditions (e.g. DO and pH patterns) between a lab-scale (5-L) production to a pilot-scale (40-L) production.