Multicellular spheroids are well-accepted in vitro cancer models as they possess most of the solid tumor hallmarks. They are typically grown in tissue culture polystyrene (TCP) dishes and very rarely in biomaterial scaffolds. However, it has been shown that spheroids grown in a biomaterial behave differently than spheroids grown in TCP, yet the nature of these differences has not been characterized. This study is aimed at developing a novel method for the production of uniform hydrogel-based glioblastoma spheroid model and elucidating the differences between a traditional and a hydrogel-based spheroid in terms of extracellular matrix (ECM) production as well as drug responsiveness. Specifically, we use polyethylene glycol (PEG) hydrogel because of its biocompatibility, ability to independently tune mechanical and biochemical properties and stiffness and viscoelastic properties similar to native neural tissues.
The PEG hydrogel used in this study was based on the Michael-type addition of 4-arm PEG-Acrylate and PEG-diester-dithiol (diSH) crosslinker and was degradable due to the nature of the in-house synthesized crosslinker. The resultant hydrogel was nanoporous with a mesh size of ~15 nm. Cysteine terminated RGD ligand was added covalently at a stoichiometric deficit to PEG acrylate groups when needed to elicit cell attachment. The cells were encapsulated prior to gelation and allowed to grow for up to two weeks. Here, we will specifically discuss three methods for PEG droplet formation and subsequent microsphere production: manual pipetting at an oil-air interface, automated pipetting in an oil bath through the use of a programmable syringe pump, and electrospraying. Our results indicated that the manual pipetting method allowed us to produce uniform microspheres in the range of 500-1200 μm but it was very time consuming and subject to experimental error. The automated method alleviated the issue allowing for the quick and easy production of uniform microspheres albeit in a larger size (>700 μm). The electrospraying method allowed us to produce microspheres in the size range of 70-700 μm, but some loss in uniformity was experienced. Higher than 90% cell viability and high cell loading (U-87 and U251 glioblastoma cell lines; ~60x106 cells/ml) could be achieved with either method. However, for the electrospraying method we evaluated the parameters which were most likely to contribute to lower cell viability, such as needle gauge, flow rate, and voltage and determined the range of parameters that can give us small (~100-200 μm) microspheres and high cell viability. For example, we determined that a voltage of 5-10 kV would be good to accomplish the above mentioned goals, while 15 kV resulted in cell viability of less than 30%.
Upon further microsphere characterizations, we also observed that while microsphere swelling and degradation were independent of microsphere size, they were dependent on the presence of cells: cell-containing microspheres swelled less and degraded slower (~3-fold differences). These results suggested to us that the cells were either exerting a force on the hydrogel matrix, secreting their own extracellular matrix (ECM) or both. Thus, our current work is focused on characterizing the ECM proteins secreted by the cells and comparing those to ECM proteins secreted by spheroids grown on a non-cell-adhesive TCP. Concurrently, we will also be testing for differences in drug responsiveness. Presently, we have confirmed that cells grown as a spheroid are much more resistant to drugs: when treated with paclitaxel, cells grown on TCP as a monolayer showed an IC50 of 10 nM, while cells grown as spheroids retained ~90% viability even at 1000 nM of the drug.
In conclusion, we have developed several methods for the production of hydrogel-based multicellular spheroids. Cells were more resistant to chemotherapeutics when grown in spheroids as opposed to a monolayer. Ongoing work is aimed at comparing spheroids grown on TCP and in PEG hydrogels to determine whether the addition of hydrogels improves the predictive capacity of the platform. Future studies will also focus on determining mechanisms of drug resistance in both systems with focus on protein expression, including ECM secretion by cells.
See more of this Group/Topical: Materials Engineering and Sciences Division