(52f) Influence of Microgel Fabrication Technique on Granular Hydrogel Properties

Introduction: Bulk hydrogels traditionally used for tissue engineering and drug delivery have numerous limitations, such as restricted injectability and a nanoscale porosity that reduces cell invasion and mass transport. An evolving approach to address these limitations is the fabrication of microgels that can be assembled into multifunctional granular hydrogels.¹ Granular hydrogels can be injectable, exhibit tunable porosity through the extent of microgel packing, and can be post-crosslinked into stable structures.² These properties have progressed granular hydrogels as injectable therapeutics and for 3D printing applications.^2,3 There are numerous techniques that are used to fabricate microgels,⁴ but no studies have yet investigated how the technique affects the resulting granular hydrogel properties. To address this need, we investigated the influence of three microgel fabrication techniques (e.g., microfluidic devices [MD], batch emulsions [BE], and extrusion fragmentation [EF]) on granular hydrogel properties (Fig 1a).

Methods: Microgels were fabricated from 3wt.% norbornene-modified hyaluronic acid (NorHA, ~15% of repeat units modified with norbornene, synthesized as described previously⁵), 0.05wt% i2959 (initiator), and dithiothreitol (DTT, crosslinker) across all methods. MD microgels were generated using a single-channel droplet generator of precursor solution with light mineral oil containing 2wt.% Span80; BE microgels were generated through the dropwise addition of 1mL of precursor solution to 30mL of light mineral oil containing 2wt.% Span80 stirring at 350rpm; EF microgels were generated through the gelation of precursor solution in a syringe and serial extrusion through 18G, 25G, and 30G needles. All microgels were photocured (UV light, 20mW/cm², 10min), washed in PBS, and jammed by vacuum. Microgel size was determined by adding 0.05wt.% 2MDa FITC-dextran to the precursor solution and quantifying fluorescent microscopy images using ImageJ. Degree of consumption (DoC%) of norbornene groups in microgels was determined by incubating microgels in hyaluronidase for 72h, freezing at -20°C for 24h, lyophilizing for 24h, suspending the dry sample in D₂O, and quantifying norbornene peak integration using ¹H-NMR spectra. Mechanical properties of granular microgels were characterized using a rheometer with parallel plate geometry (1000μm gap size, 1Hz) or under compression with a dynamic mechanical analyzer (0.05Nmin^-1) after post-crosslinking in the presence of excess crosslinker and UV light. Data is presented as mean ± standard deviation, with statistical analysis via ANOVA and a Tukey’s post hoc comparison (*p<0.05, **p<0.01).

Results and Discussion: Across all methods, DTT concentration in the precursor solutions was tuned such that ~50% of norbornene groups were consumed during microgel fabrication (Fig 1b). EF microgels required 3mM DTT, whereas MD and BE microgels required 5mM DTT. We hypothesize that the increased amount of DTT required in MD and BE fabrication methods may be due to interactions between hydrophobic norbornene groups and oil that interfere with microgel crosslinking, which is the focus of ongoing investigations.

Average microgel diameter across fabrication methods was ~100μm (Fig 1c, scale bars = 200μm). MD microgels were relatively homogeneous in diameter (Fig 1cii), whereas BE and EF microgels were heterogenous (Fig 1ciii,iv). EF microgels were non-uniform in shape, whereas MD and BE microgels were spherical. The generation of 1mL of granular hydrogel required ~1, 2, and 6hrs of total fabrication time (microgel synthesis to jamming) for the EF, BE, and MD methods, respectively. Additionally, MD microgels required ~2 days of microfluidic device fabrication before particle generation.

EF microgels exhibited a higher storage modulus (G’) (~2kPa) compared to MD or BE microgels (~0.15kPa) (Fig 1d). All granular hydrogels were shear-thinning and exhibited self-healing properties (G’ recovery of >90% after 2min of 500% strain, data not shown). Post-crosslinking was possible with MD and BE granular hydrogels, leading to compressive moduli of ~3kPa and ~2kPa, defined as slope between 5%-10% strain. Granular hydrogels from EF microgels did not post-crosslink. We hypothesize that norbornene-oil interactions in MD and BE methods may alter the number of free norbornene groups present at the microgel surface, thus allowing successful post-crosslinking of microgels, which is the focus of ongoing investigations.

To probe the effects of microgel degree of crosslinking on resultant mechanical properties of granular hydrogels, EF microgels were fabricated with varying DTT concentrations (1, 3, 5 mM), resulting in ~20, 50, and 90% consumption of norbornene groups, respectively. Across DTT concentrations, all EF microgels had an average diameter of ~100μm (Fig 1eii). Storage modulus (G’) varied significantly, from ~0.1kPa at 1mM DTT to ~4kPa at 5mM DTT (Fig 1eiii). Interestingly, EF microgels fabricated with 1mM DTT were able to post-crosslink into stable structures with a compressive modulus of ~2.5kPa. Ongoing investigations include determining the effects of the norbornene/oil interface in MD and BE fabrication methods as well as quantifying the porosity of granular hydrogel formulations.

Conclusions: Granular hydrogels are useful in many biomedical applications. Here, we show that the fabrication technique not only influences microgel uniformity and shape, but also granular hydrogel mechanical properties. These results, along with consideration of microgel fabrication rates, will inform multifunctional granular hydrogel design and utility in numerous applications.

¹Riley, et al. Curr. Opin. Biotechnol., 2019. 60, 1-8.

²Mealy, et al. Adv. Mater., 2018, 30, 1705912.

³Xin, et al. Biomater. Sci., 2019, 7, 1179.

⁴Daly, et al. Nat. Rev. Mat., 2020. 5, 20–43.

⁵Gramlich, et al. Biomaterials, 2013. 34, 9803-9811