(C-114) Control of jammed microgel properties via electrostatic interactions for cancer invasion studies

Pancreatic ductal adenocarcinoma (PDAC) has been, and continues to be, one of the deadliest cancers. Patients are typically diagnosed after the cancer has metastasized, leading to poor prognosis, with a typical 3% five-year survival rate (1). The extracellular matrix (ECM) of PDAC plays an important role in PDAC progression, where an increase in collagen creates a stiffer environment, and the upregulation in hyaluronic acid produces increased intratumoral pressure (2,3).  This is a complex environment, with both healthy and sick tissue, flowing through dynamic stages. To create an in vitro model that properly models these dynamics, their mechanical interactions, and their effects on PDAC invasion, a tunable ECM mimic must be designed.

Jammed microgels have emerged as an important class of materials for fine control of hydrogels for in vitro models. Due to their shape, jammed systems have an inherent porosity (4, 5) and the microgels can move past one another via an applied force, giving the material dynamic properties (6,7). To enhance and modify this dynamic behavior, electrostatic interactions can be introduced. By using an anionic biopolymer, hyaluronic acid, to formulate the microparticles, the system mechanics can be tuned by altering interparticle interactions through inclusion of gelatin with varying charges. This can influence how cells move within the particle-based hydrogel, offering a means to study the influence of mechanical properties on migratory cells and related processes, like cancer metastasis. Work here focuses on the design of the microgel system and interactions with PDAC spheroids.

Gelatin (300 bloom, type A) was modified to be cationic via EDC-coupling of ethylene-diamine (8), or anionic through reaction with succinic anhydride (9). Gelatins type A and B were also used as received to examine intermediate charges with the microgels. Norbornene-modified hyaluronic acid (NorHA) was prepared as previously described via an HA-TBA intermediate that was reacted anhydrously with 5-norbornene-2-carboxylic acid (7, 10). All materials were purified via dialysis or ultrafiltration then lyophilized. Zeta potentials of all materials were measured using a zetasizer.

To prepare particles, NorHA was dissolved with DTT and a photoinitiator then homogenized in oil with surfactant. UV light was used to crosslink the particles. The microgels were washed with isopropanol then rehydrated in 70% ethanol and then PBS. Jamming of particles was achieved via centrifugation at high speeds. Gelatin solutions were added based on weight of the jammed NorHA particles.

Rheological characterizations of NorHA particles with different gelatin materials were conducted using a parallel plate system. Frequency and strain sweeps were conducted on freshly-prepared samples. For the cationic gelatin samples, strain sweeps were also conducted at varying amounts of compression to examine poroelastic effects.

PDAC cells modified with GFP-luciferase were cultured and formed into spheroids. The spheroids were harvested on day 3 of culture. The spheroids were placed into the support materials and monitored overnight using an on-stage microscope set to 37°C and with medical gas flow. Brightfield and fluorescent images were taken every 15 minutes for 16 hours.

When combined with cationic gelatin, jammed NorHA microgels produce a gel-like material that can hold together microgels at lower packing densities (Fig. A). The charge on the gelatin can be easily modified, with a zeta potential range of -17.5mV to +13.4mV, and a charge of -6.6mV on the NorHA microgels (Fig. B). Under confinement, the charge on the gelatin affects interparticle interactions, with increasing strain-softening behavior as the added gelatin charge becomes increasingly negative. Additionally, yielding occurs at lower strains as the gelatin becomes more negatively-charged (Fig. C). For microgels with cationic gelatin, strain sweeps were conducted at different compressions of 0, 1000, or 2000 um. As compression increases in a sample with 3% cationic gelatin and with medium-packing density, the strain-stiffening behavior becomes more apparent, and with a higher slope prior to yielding (Fig. D). This data may suggest poroelastic behavior as the material is compressed.

PDAC spheroids showed more sprouting from the spheroid in microgels with cationic gelatin compared to those with GelA or GelB (Fig. E and F). Packing density also played a role, where the higher packing density appears to show higher invasion compared to the lower packing density samples (Fig. E and F). The addition of CCL-2, a chemokine, also appears to increase PDAC invasion in the medium-packed microgels (Fig. E and F). This data suggests that interparticle interactions may affect PDAC migration through a granular medium where the gel-like jammed microgels provide a stiff environment for cells to move through. Chemokines can also be included to elucidate effects on migration. Work is ongoing in quantifying changes in morphology and migration.

The work shown here illustrates two points. One, interparticle interactions can be modified through electrostatic interactions. The addition of cationic gelatin formed a gel-like material with NorHA microgels, whereas negatively-charged gelatin increased electrostatic repulsions, causing more liquid-like behavior. Second, PDAC migration appears to have been modified based on these interparticle interactions, with more sprouting in samples with cationic gelatin. The tunable system here shows promise in further work in other cancer cell types, as well as in modeling and studying behaviors in non-cancerous cells.

We would like to acknowledge the Bauer group at the University of Virginia for kindly providing us with the PDACs

References (Optional): : (1) Siegel, R. L., CA Cancer J Clin. 2018; 68:7-30
(2) Drifka, C.R., Oncotarget. 2016; 7:76197-76213
(3) Kim, P.K., eLife. 2021; 10:e62645
(4) Griffin, D.R., Nat Mater. 2015; 14(7): 737-744
(5) Seymour, A.J., Adv Healthcare Mater. 2021; 10: 100644
(6) Bhattacharjee, T. Sci Adv. 2015; 1:e1500655
(7) Highley, C.B. Adv Sci. 2019; 6: 1801076
(8) K, J. RSC Adv. 2015; 5:89521
(9) Madl, A.C., ACS Macro Lett. 2020; 9:619-626
(10) Gramlich, W.M. Biomaterials. 2013; 34: 9803-9811