Poster S15 - Engineering Hydrogels to Investigate the Role of Hyaluronic Acid on Natural Killer Cell Functions

Introduction: Natural killer (NK) cells hold promise for immunotherapy due to their ability to detect and kill cancer cells without prior antigen priming. However, within solid tumors, NK cells tend to reside in the stroma, with only a small fractions making direct contact with cancer cells. Indeed, NK cell The extracellular matrix (ECM) in solid tumors is known to play a role in NK cell activity; however, it remains unclear if aspects of the ECM acts as a barrier or a substrate. Hyaluronic acid (HA) is a glycosaminoglycan prevalent in lung tumors ECM. During tumor progression, HA gets fragmented from anti-tumorigenic, high molecular weight (HMW) HA into low molecular weight (LMW) HA. LMW HA leads to increased invasiveness of cancer cells, but the effect on NK cell tumor infiltration is unknown. NK cell-cancer cell interactions are typically studied in 2D culture systems which lack the complexity of the 3D tumor microenvironment, leading to poor correlation with in vivo studies. To address this gap, the Sharma Lab developed a 3D poly(ethylene) glycol (PEG)-based hydrogel system that allows the independent manipulation of biochemical and mechanical properties. The goal of this study was to investigate the influence of the molecular weight of HA on NK cell migration and activation in 3D, offering insights crucial for advancing NK cell-based immunotherapies.

Materials and

Methods: Hydrogels were made by photo-polymerizing a mixture of PEG-diacrylate, PEG functionalized with MMP degradable sites, 5 mM of PEG functionalized with integrin binding sites, with or without 500 g/mL of HMW or LMW methacrylated HA (MeHA). To investigate the influence of different molecular weights of HA on NK cell migration, chemoattractant SDF1 was encapsulated in hydrogels containing HMW MeHA, LMW MeHA, and no MeHA. NK-92 cells, a clinically relevant cell line, were cultured with this system and total infiltration into the hydrogel towards the chemokine was quantified. To investigate the influence of HA properties on NK cell effector functions, tumor models were generated by photo-encapsulating A549 metastatic lung cancer cells. NK92 cells were incubated with the tumor models at an effector-to-target ratio of 2:1 for 2 hours, and the media was evaluated for cancer cell death via lactate dehydrogenase (LDH) release.

Results, Conclusions, and Discussions: The addition of HMW and LMW MeHA to the PEG-based hydrogel did not alter the Young’s modulus (Fig. 1A) thus allowing the investigation of the effect of HA properties on NK cell migration, independently. In the absence of cancer cells, there was significantly more NK cell infiltration towards SDF1 in hydrogels containing HMW MeHA (Figure 1B), indicating that HMW HA may act as a substrate in the tumor microenvironment. A549 lung cancer cells remained viable after encapsulation and proliferated over time in culture (Fig. 1C and 1D). Though MeHA did not affect the total number of cells in the hydrogel, it did encourage spheroid formation over time, suggesting an effect of HA on cancer cell migration and cell-cell interactions (Fig. 1D). When the tumor models were incubated with NK cells, an increase in NK cell infiltration in hydrogels containing HMW MeHA is observed (Fig. 1E), but is correlated with a decrease in NK cell effector function (Fig. 1F). This indicates that although the NK cells can infiltrate the tumor models, they may not be able to exert their effector functions on day 7 either due to exhaustion, an immunosuppressive microenvironment, or degradation of HA. Together, these findings suggest that HMW HA provides a substrate to support NK cell infiltration. Future studies will investigate the production of immunosuppressive factors in response to different molecular weights HA, quantify NK cell degranulation in the presence of HA, and the role of CD44, the main receptor that binds HA on NK cell functions. 3D culture systems support studying the mechanistic tri-part interactions between cancer cells, ECM, and NK cells, which is not possible in 2D and difficult in vivo. Importantly, the use of tunable, 3D tumor models may help bridge the gap between 2D and in vivo models by creating a biophysical environment mimetic of some aspects of solid tumors in vivo.

Acknowledgements (Optional):