(622p) Expansion of Human Mesenchymal Stem Cells in Fibrous Bed Bioreactor
Human mesenchymal stem cells (hMSCs) isolated from various adult tissues emerge as primary candidate in cell therapy for a wide range of diseases. The translation of hMSC into clinical application depends on the delivery of therapeutically competent hMSCs at the sufficient quantity. While most cell therapy products to date are based on autologous cells for immunologic compatibility, hMSCs have been used in allogenic transplantation as a potential “off-the-shelf” product. In the production of cells for allogenic transplantation, a scalable production process is crucial for the commercial scale cell production with required quality attributes. The conventional stirred-tank bioreactor provides the scalable possibility, but however has non-homogeneous hydrodynamic environment due to mechanical agitation. As hMSCs are highly sensitive and responsive to the mechanical forces, the potential impact of the biomechanical stimuli on hMSC phenotype and therapeutic potency is yet to be determined. Therefore, the present study investigates the feasibility of hMSC expansion in a 2.5 L CelliGen® 310 Bioreactor packed with Fibra-Cel® disks, providing the scalable potential with low and homogeneous shear stress microenvironment.
Materials and Methods
A 2.5 L CelliGen® 310 bioreactor (Eppendorf) is used in the current study. The 2.5 L glass vessel is jacketed by a water bath layer and capped with a stainless steel head plate. Ports on the head plate are designed for pH and dissolved oxygen probes, seeding and sampling tube, and gas transfer. The vessel was autoclaved with 25.0 grams of Fibra-Cel® disks (Eppendorf) fixed within the packed bed. After sterilization, growth media was delivered into the vessel with pump first, and then hMSCs obtained from plate culture were seeded into the vessel. In the seeding process, 100 rpm agitation speed was used for 1 hour for even-distributed cell attachment. Temperature, pH, and mixture gas (5% CO2 and 95% air) were automatically controlled and monitored by the microprocessor control system. The agitation speed was adjusted to 50 rpm for cell expansion after seeding. 10% of working volume was replaced daily with fresh media and samples were collected before and after media change. The bioreactor operation was repeated twice using identical operation procedure.
Result and conclusion
In our result, hMSCs on the surface of fibers were estimated to be exposed to a much lower shear stress of 0.54 dyne/cm2 in average compared to the conventional stirred-tank bioreactors with suspended microcarriers, based on the dimension and packing density of the fibrous bed and the agitation induced pressure drop through the fibrous bed. After 9 days of expansion, a 9.2-fold increase in cell number with the population doubling (PD) time of 2.8 days (67.2 hours) was achieved. Specific glucose consumption and lactate production are 12.48 pmol/cell/day and 20.95 pmol/cell/day, respectively. Scanning electron microscope (SEM) and fluorescence DAPI staining images of harvested Fibra-Cel® disks show the uniform distribution of hMSCs on the fibers and the potential space for continuous cell growth. hMSCs harvested from the bioreactor maintained their properties based on the analysis of hMSCs surface markers, colony forming unit-fibroblasts (CFU-F) capacity, and multilineage differentiation ability. The results demonstrate the feasibility and the potential of the fibrous bed bioreactor for large scale hMSC expansion.
- T.W. Chiou, S. Murakami, D.I. Wang, W.T. Wu, A fiber‐bed bioreactor for anchorage ‐ dependent animal cell cultures: Part I. Bioreactor design and operations, Biotechnology and bioengineering, 37 (1991) 755-761.
- F. Zhao, T. Ma, Perfusion bioreactor system for human mesenchymal stem cell tissue engineering: dynamic cell seeding and construct development, Biotechnology and bioengineering, 91 (2005) 482-493.
- J. Kim, T. Ma, Perfusion regulation of hMSC microenvironment and osteogenic differentiation in 3D scaffold, Biotechnology and bioengineering, 109 (2012) 252-261.