Poster P6 - Glioblastoma Spheroid Growth, Infiltration and Motility in Single and Dual-Stiffness Hydrogels

Introduction: Multicellular spheroids are widely recognized as effective in vitro models for cancer research and are commonly utilized in the development of cancer therapies. Typically, these spheroids are cultivated on non-adhesive surfaces, which do not account for cell-matrix interactions. In our study, we developed a hydrogel-based model for glioblastoma (GBM) spheroids using different hydrogels: polyethylene glycol (PEG), gelatin methacryloyl (GelMA), and methacrylated hyaluronic acid (HAMA). These hydrogels enable manipulation of the spheroid's microenvironment. Our model featured a precise stiffness interface that mimicked the stiffness and biochemical cues of normal or brain tumor tissue on each side. Research has shown that GBM tumors respond to matrix stiffness both in vivo and in vitro, as well as to interfacial mechanical gradients in hydrogels in vitro and topography-related biomechanical cues in vivo. We encapsulated multiforme GBM spheroids within these hydrogels and positioned them at interfaces with varying stiffness and/or biochemical cues to study how these interfaces influence the growth, motility, and therapeutic responses of GBM spheroids.

Materials and

Methods: Spheroids were generated using U87 cells or patient-derived cells, which were cultured in Sylgard microwell molds. The cells were seeded at a density of 1*10^6 cells/mL and centrifuged for 3 minutes at 3000 rpm. After culturing for 24 hours to allow spheroid formation, they were encapsulated in hydrogels. These hydrogels were composed of 4-arm PEG-acrylate (4-arm PEG-Ac), 4-arm PEG-RGDS, PEG-dithiol (PEG-diSH), and an MMP-2 degradable peptide crosslinker (VPMS↑MR). We fabricated hydrogels with varying stiffness: soft hydrogels with a Young’s modulus of 1 kPa and stiff hydrogels of 7 kPa. Additionally, dual hydrogels featured a thin interface with stiffness transitioning between 1 kPa and 7 kPa, where cells were encapsulated in one of the hydrogels and allowed to fall to the interface upon gelation (Figure 1). Hydrogels were also made of GelMA and HAMA. Gelatin and hyaluronic acid were chemically modified in-house to include methacrylate groups, resulting in the formation of GelMA and HAMA enabling their covalent integration into polymer networks via photo-polymerization. The dual GelMA/HAMA hydrogels were formulated with varying ratios/concentrations of the two polymers and spheroid encapsulation in those is ongoing. Cell viability was assessed using live/dead (DiOC/PI) staining. Spheroid size and infiltration were analyzed through microscopic imaging. Hypoxia was measured using Hif-1α staining.

Results, Conclusions, and Discussions: In this study, we developed an in vitro model of GBM using a dual-stiffness hydrogel, wherein spheroids were encapsulated within the upper side of the stiffness interface. To do so, spheroids were added to the upper gel precursor solution and allowed to fall within few microns from the interface during gelation, where distance from the interface was controlled by gelation kinetics (i.e. pH of the precursor solution for the PEG hydrogels) (Figure 2). All spheroids maintained >90% viability for up to 7 days across soft, stiff, and dual hydrogels. GBM spheroids exhibited increased infiltration in soft gels compared to stiff gels, and hypoxia was noted in the cores of spheroids encapsulated within the stiff gels. The reduced invasion in stiff gels may be attributed to higher crosslinking density and a greater number of enzymatically degradable sites, which impede cell migration. In the dual hydrogels, spheroids showed potential for crossing the stiffness boundary in control (stiff-stiff) gels. For dual hydrogels with differing stiffnesses, cells preferentially migrated along and crossed the stiffness interface from soft to stiff. This migration pattern mirrors the in vivo behavior of GBM, where cells tend to migrate towards and favor areas of higher stiffness.
GBM cells secrete higher levels of hyaluronic acid (HA) compared to normal brain cells, indicating that HA is a predominant component of the extracellular matrix (ECM) in glioblastoma tumors. In our current work, we developed a dual-composite hydrogel made of GelMA and HAMA. This approach aims to mimic the tissue composition, stiffness (1-10 kPa), and structure more accurately by varying the ratios of HAMA to GelMA. Our goal is to study cell invasion and migration along and across the GelMA/HAMA interface.
Overall, our results indicate that microenvironment stiffness and stiffness interfaces need to be considered in bioengineered GBM models.

Acknowledgements (Optional):