(C-107) Hydrazone and Strain-Promoted Azide-Alkyne Cycloaddition Cross-Linked Hyaluronic Acid and Poly(ethylene glycol) Injectable Hydrogen for Cell Therapies

Injectable hydrogels are hydrophilic polymer networks typically cross-linked through nonpermanent interactions, such as covalent adaptable bonds.  Covalent adaptable networks (CANs) enable a hydrogel to yield upon an applied force, extrude through a needle via bulk flow, and self-heal.  These properties render CANs useful to fulfill roles in non-invasive procedures to implant materials, as they possess the ability to encapsulate and protect cells, and mold to cavity-specific shapes.  However, the adaptable cross-links lead to increased erosion and a lack of long-term stability.  As a result, researchers often rely on the introduction of a small amount of irreversible cross-links (such as covalent bonds), in combination with the adaptable cross-links, to stabilize the hydrogel post-injection [1]. Because covalent bonds decrease the injectability of the formulation, typically the reaction is polymerized after injection through external stimuli [2]. However, in spontaneous reactions, controlling the gelation kinetics is important for the successful translation of the material.  In this work, we introduce two design elements in a hyaluronic acid-poly(ethylene glycol) (HA-PEG) injectable formulation: 1. A slow forming strain-promoted azide-alkyne cycloaddition (SPAAC) cross-link to stabilize and slow down gelation kinetics and 2. the addition of a small concentration of a monofunctional methyl-poly(ethylene glycol)₄-hydrazide (m-PEG₄-Hyd) to further control gelation kinetics.  Together, these approaches allow tuning of the gelation kinetics, network stability, and final material properties.

The hydrazone and SPAAC cross-linked hydrogels were formulated with three macromers: HA functionalized with hydrazides (HA-Hyd), HA functionalized with aldehydes (HA-Ald), and PEG functionalized with Bicyclo[6.1.0]nonyne (PEG-BCN) all synthesized following established protocols [3]. HA-Hyd was additionally functionalized with a small number of azides by premixing it with benzaldehyde-PEG₃-azide, allowing for both hydrazone and SPAAC cross-links to be formed upon mixing with HA-Ald and PEG-BCN. Hydrogels were formulated with an increasing percent of SPAAC bonds ([BCN]/([Hyd] + [azides])) with a ~3 w/v% final polymer content, a ratio of Hyd:Ald of 1:0.35, and a BCN:azide ratio of 1:1. Degradation and network stability of 0-12% SPAAC hydrogels were quantified over time using a uronic acid assay and ImageJ. A shear rheometer was used to monitor network formation for the 0-12% SPAAC hydrogels (1 Hz, 0.5% strain) and determine the final shear storage (G’) modulus. To estimate the speed of network formation, the time to reach 90% of G’_max was compared between formulations.  The injectability of the 12% SPAAC-containing hydrogel was characterized by shear rheology by applying a strain cycle (500% strain). The 12% SPAAC formulation was modified by addition of m-PEG₄-Hyd, (0, 0.24, or 0.48 mM) to slow down the speed of network formation, as measured by shear rheology.  Finally mesenchymal stromal cells (MSCs) were encapsulated in 0% SPAAC, 12% SPAAC, and 12% SPAAC + 0.48 mM m-PEG₄-Hyd hydrogels and MSC viability was assessed post-injection as a function of the applied force required for injection.

With an increase in the percent of SPAAC bonds present in the hydrogel formulation (0-12%), the degradation profiles varied over time. By day 14, the normalized HA erosion was observed to range from ~90% to ~60% HA content in the 0% to 12%, respectively.  Further, the network stability was visually apparent, as degradation and erosion also led to changes in the normalized surface area (Fig 1A). Additionally, rheological measurements of the shear storage modulus showed no significant differences between the 0% and 12% SPAAC hydrogels (1,250 ± 50 Pa vs 1,300 ± 100 Pa).  Interestingly, the rheological measurements of the shear storage modulus showed changes in the hydrogel gelation kinetics, observed by differences in the time to reach 90% of G’_max, ranging from ~150 seconds to ~4000 seconds, with increasing SPAAC content (Fig 1B). These results demonstrated the ability to tailor the degradation, network stability, and gelation kinetics by small changes in SPAAC content without altering the initial hydrogel mechanical properties. To investigate post-injection self-healing, shear rheology was used to introduce a strain cycle ~120 second after initial mixing of the macromers used for the 12% SPAAC formulation. Notably, no significant differences in G’ were measured with or without the strain cycle (Fig 1C). Finally, the effect of slowing down hydrogel formation on the force to inject encapsulated MSCs and their viability after injection was assessed. The force of injection was found to decrease with increasing SPAAC content ( >30 N to ~7 N) and further increase the viability (~45% to ~70%) (Fig 1D). Together, these results demonstrate two design elements that together allow for control over the stability and the time for network formation to complete to improve an injectable hydrogel for cell therapies.

The authors thank the United States National Institutes of Health for funding this work (Grant: R01 DE016523) and the Department of Education Graduate Assistance in Areas of National Need fellowship for supporting M. W. J.

1. Chem. Rev. 2023, 123, 2, 834-873

2. J Biomed Mater Res A. 2018, 106, 4, 865-875

3. Adv Healthc Mater. 2022, 11, 14, e2200393