Materials and Methods

Cells, medium and shake-flask cultures

Three clones derived from CHO Cell Line B (Wild-type Clone 47, BCAT1 KO Clone 83, and BCAT1 KO Clone 90) utilizing a glutamine synthetase expression system and expressing a recombinant antibody were used in the present study (Mulukutla et al., 2019). Three types of proprietary medium were used: “Medium A” is the production medium for traditional and HiPDOG fed-batch production bioreactors. “Medium B” is an enriched nutrient solution used as feed medium for both modes of production bioreactor. Medium C is the passaging medium used for cultivation and experimentation in shake-flask batch cultures. Medium A and B are identical in composition to those used previously (Mulukutla et al., 2019).
Medium A is a chemically defined, protein-free, amino acid fortified version of DMEM:F12 medium with adjusted levels of vitamins, trace elements, sodium bicarbonate and potassium chloride, also containing polyvinyl alcohol. Medium A was modified with additional sodium bicarbonate and glucose for non-HiPDOG fed-batch production bioreactors. Medium B is a chemically defined, protein-free concentrated feed composed of amino acids, vitamins, and trace elements (a subset of those in Medium A). Medium B was supplemented with additional alanine, cysteine, tyrosine, and zinc to prevent depletion of these compounds in high-density cell culture processes.
The specific impact of lactate on cell growth of BCAT1 KO Clone 83 and WT Clone 47 was assessed using shake-flask cultures. Medium C, a chemically defined and protein-free lean passaging medium was modified with 0 to 10 g/L lactate via sodium lactate (Sigma-Aldrich®, St. Louis, MO) at concentrations similar to those achieved in fed-batch and HiPDOG bioreactors. An osmolality control matching the 8 g/L lactate condition was also made using Medium C modified with sodium chloride. WT Clone 47 sodium lactate concentrations used were a subset of those used for Clone 83, included for comparability. Cells were inoculated at 0.1 × 106 cells/mL in triplicate 125mL shake-flasks for each condition with a working volume of 30 mL, and cultivated for 6 days on a shaking platform (Orbital Shaker, Bellco Glass, Inc.) in a humidified incubator (ThermoFisher Scientific) maintained at 36.5 °C and 5% carbon dioxide. Viable cell density, glucose, lactate, and ammonia concentrations in the cell culture medium were measured on days 0, 3, and 6 using a Nova Bioprofile FLEX Analyzer (Nova Biomedical, Waltham, USA).

Bioreactor setup for fed-batch and HiPDOG cultures

Cell Line B clones were cultivated in pairs of bioreactors where the first of each pair employed a typical titrant-based lower pH dead-band and the second, the HiPDOG control strategy (Gagnon et al., 2011). For HiPDOG bioreactors, pH dead-bands during and after HiPDOG were 7.125 +/- 0.025 and 7.10 +/- 0.20 respectively. Each fed-batch bioreactor had its pH tightly controlled to match its HiPDOG partner during the HiPDOG operational phase, while utilizing the same pH dead-band thereafter. Medium A was used as the production medium, while Medium B was used as feed medium for all vessels. Inoculation cell densities were 1.5 × 106 cells/mL in an initial working volume of 1L, with temperature and agitation set to 36.5 oC and 259 rpm respectively. In HiPDOG reactors, feed medium was added based on upper pH dead-band during HiPDOG control, operational from day 2 to day 6. During this period each fed-batch reactor had constant semi-continuous feed rates set daily to match the cell specific feed rate of its partner HiPDOG reactor. Fed-batch reactors were additionally provided glucose to target 2.5 g/L as-needed. Following the end of HiPDOG control, feed rates for all vessels were set daily to target a final cumulative cell specific feed rate of 1.75 pL/cell/day, resulting in a total feed medium addition of 20 to 60% of the culture starting volume. Post-HiPDOG glucose levels were targeted at 2.5 g/L by feeding additional glucose as necessary for all vessels. Viable cell density, glucose, lactate, and ammonia concentrations in the cell culture medium were measured daily using a Nova Bioprofile FLEX Analyzer (Nova Biomedical, Waltham, USA). Titer analysis was performed by Protein A HPLC (model 1100 HPLC, Agilent Technologies, Inc., Santa Clara, CA, protein A column model 2-1001-00, Applied Biosystems, Foster City, CA). WT clone 47 conditions were run in duplicate, however data beyond day 11 for the duplicate HiPDOG condition are unavailable due to a controlling probe failure.
Additional methods pertaining to amino acid concentration analysis via UPLC and spent medium metabolite analysis via NMR are identical to those utilized in the prior study (Mulukutla et al., 2019).