4.4 Bioreactor process optimization
Producing large numbers of cells efficiently is a key step in the process of creating cultivated meat. Several bioreactor configurations have been suggested [10], but the jury is out as to whether any of them will be practical at the much larger scale and lower cost required to support commercial production. Building out large-scale bioreactor prototypes is an expensive and time-consuming proposition, so making a judgement by empirical means is hampered by cost. Developing a single model specific to one configuration is also likely to be expensive and time-consuming; however, once created, a model can be iteratively refined and tested at a far lower cost than redesigning and testing physical bioreactors. In addition, the methodology used to model any one bioreactor configuration is likely applicable to others with incremental modifications. Indeed, computational modeling approaches have been applied to bioreactor use in cell therapy development [23-28], and these approaches should translate to cultivated meat production.
Here, we briefly detail five bioreactor configurations that have potential utility in cultivated meat - i) suspension growth (w/o microcarriers) in a stirred tank bioreactor, ii) adherent growth (w/ microcarriers) in a stirred tank bioreactor, iii) hollow fiber bioreactors, iv) a continuous flow bioreactor design relying on coated surface and v) a rocking bioreactor design. We also describe instances where computational modeling has been applied to study these reactors. In the future, novel and more holistic modeling of these systems must combine simulation of the cell biology with modeling of physio-mechanical properties (e.g. adherence, elasticity, susceptibility to shear forces) and computational fluid dynamics to get a better understanding of which reactor design is best-suited to achieve the high output requirements of cultivated meat production.