During process development (PD), single-use (SU) bioreactors can streamline cell-culture workflows and reduce risks for contamination and operator errors. But transferring cultures from laboratory- to benchtop- and pilot-scale vessels requires consideration of fundamental bioreactor principles. In October 2021, Ann D’Ambruoso (manager of product applications and marketing) and Cristina Bernal Martinez (applications support engineer, both at Getinge) reviewed factors for scale-up success and presented a case study involving transfer of cell cultures to Applikon AppliFlex SU stirred-tank (ST) reactors.
The Presentations
Scale-Up Principles: D’Ambruoso explained that scale-up involves establishing similar geometric and physical variables across culture volumes and reactor sizes. Selecting a robust vessel design can facilitate that work. For instance, AppliFlex ST bioreactors leverage hemispherical bottoms to ensure effective mixing of culture fluids, and three-dimensional printing enables precise production of the vessels’ SU stirring assemblies and headplate components. Then, PD scientists must determine which factor to hold constant across reactors — and that depends on culture morphology and key process parameters.
The volumetric oxygen transfer coefficient (kLa) is a helpful scaling factor if O2 concentration is a critical parameter. The value represents the influences of physical factors (e.g., stirring speed, gassing rate, working volume, and even media composition) upon oxygen transfer in a vessel. Although kLa is usually determined by a static N2 gassing-out method, it can be modeled mathematically. Creating a model for several reactor conditions at a scale of interest can help PD teams to understand the effects of stirring and gassing on oxygen transfer.
Mixing power:volume ratio (P/V) might be a suitable scaling factor if oxygen transfer is not a critical culture parameter, D’Ambruoso continued. P/V reflects culture homogenization and gas–liquid transfer. It also can indicate a culture’s potential for shear stress.
Shear-sensitive cultures can be scaled by impeller-blade tip speed. Generally, large vessels must be set to slower stirring speeds to hold tip speed constant across reactors.
Case Study: Martinez described scale-up strategies that her company considered when collaborating with blood-services provider Sanquin and Delft University of Technology, both based in the Netherlands. She explained that production of transfusion-ready erythrocytes from precursors could help to meet high global demand for blood. Each transfusion unit requires 2 × 1012 red blood cells (RBCs) — a quantity that cannot be achieved in static bioreactor conditions. ST reactors could produce RBCs in sufficient quantities, but such reactors are prone to turbulence, and erythroblasts are shear sensitive. Thus, the team sought to identify scale-up conditions that minimized shear stress.
Studies began with a 14-day culture process. Hematopoietic stem cells (HSCs) were grown in Petri dishes for eight days to encourage differentiation into proerythroblasts. Those cells were transferred to 0.5-L AppliFlex or glass bioreactors and cultured in repeated-batch mode for six days. In all reactors, cell concentrations increased from
106 cells/mL to 1.5–2.0 × 106 cells/mL daily, and all cultures yielded comparable total numbers of cells. Subsequent analysis showed that AppliFlex SU and glass vessels yielded lower numbers spontaneously differentiated cells (an undesired phenotype) and higher terminal viability than did cultures from static reactors.
Considering erythroblasts’ shear sensitivity, the team chose tip speed as the primary scale-up parameter. HSCs from three donors were grown in dishes for eight days, transferred to 0.5-L AppliFlex or glass vessels for three days, and finally cultivated in 3-L AppliFlex and glass reactors for six days (working volume = 2.5 L). Cultures in the static and AppliFlex reactors grew at similar rates and yielded comparable quantities of cells. Thus, Martinez explained, SU ST vessels can be used for erythroblast expansion, and such a process can be scaled according to tip speed without diminishing culture performance or inducing unwanted differentiation.
Questions and Answers
Can a process be scaled across vessels with different impeller types? It is advisable to choose vessels with the same impellers. For instance, tip speed can be adjusted across vessels, but flat-blade and Russian impellers create more dynamic mixing environments than marine and pitch-blade impellers do, raising distinct shear concerns.
Which is more important to monitor during scale-up, shear stress from gas sparging or that from tip speed? That depends on the size and quantity of bubbles generated by a sparging strategy. Such factors can inhibit culture growth. But stirring is a known source of shear, making it a helpful factor to begin scale-up consideration.
The recorded webcast is available to view: Watch Now.