Crosslinker Architectures Impact Viscoelasticity in Dynamic Covalent Hydrogels

Abstract

Dynamic covalent crosslinked (DCC) hydrogels represent a significant advance in biomaterials for regenerative medicine and mechanobiology. These gels typically offer viscoelasticity and self-healing properties that more closely mimic in vivo tissue mechanics than traditional, predominantly elastic, covalent crosslinked hydrogels. Despite their promise, the effects of varying crosslinker architecture - side chain versus telechelic crosslinks - on the viscoelastic properties of DCC hydrogels have not been thoroughly investigated. This study introduces hydrazone-based alginate hydrogels and examines how side-chain and telechelic crosslinker architectures impact hydrogel viscoelasticity and stiffness. In hydrogels with side-chain crosslinking (SCX), higher polymer concentrations enhance stiffness and decelerates stress relaxation, while an off-stoichiometric hydrazine-to-aldehyde ratio leads to reduced stiffness and shorter relaxation time. In hydrogels with telechelic crosslinking, maximal stiffness and slowest stress relaxation occurs at intermediate crosslinker concentrations for both linear and star crosslinkers, with higher crosslinker valency further increasing stiffness and relaxation time. Our result suggested different ranges of stiffness and stress relaxation are accessible with the different crosslinker architectures, with SCX hydrogels leading to slower stress relaxation relative to the other architectures, and hydrogels with star crosslinking (SX) providing increased stiffness and slower stress relaxation relative to hydrogels with linear crosslinking (LX). The mechanical properties of SX hydrogels are more robust to changes induced by competing chemical reactions compared to LX hydrogels. Our research underscores the pivotal role of crosslinker architecture in defining hydrogel stiffness and viscoelasticity, providing crucial insights for the design of DCC hydrogels with tailored mechanical properties for specific biomedical applications.

Figure 1.

Dynamic covalent crosslinked (DCC) alginate hydrazone hydrogels can be formed with side-chain crosslinking (SCX), linear crosslinking (LX), and star crosslinking (SX) architectures. A) Schematics of SCX or telechelic crosslinking with LX or SX architectures for DCC hydrogels. B) Alginate was functionalized with either hydrazine or aldehyde groups. C) Linear and star PEG-aldehyde crosslinker molecules used for telechelic crosslinking. D) Representative image of two SX hydrogels placed in 6 mm cell culture dish, colored with two food dyes. E) Time evolution of storage and loss modulus of hydrogels with SCX (left), LX (center), and SX (right).

Figure 2.

Rheological characterizations of SCX hydrogels. A-B) Representative of frequency sweep (average of n = 4) and stress relaxation (average of n = 3) tests using formulation of 2 wt% alginate at AG-HYD to AG-ALD ratio of 1:1. C-E) The elastic modulus, loss tangent, and stress relaxation half-times of hydrogels with varying alginate concentration under same AG-HYD to AG-ALD ratio of 1:1. F-H) The elastic modulus, loss tangent, and stress relaxation half-time of hydrogels with varying AG-HYD to AG-ALD ratios, all with the same alginate concentration of 2 wt%. One-way ANOVA with Tukey’s post-hoc test. * P ≤ 0.05, ** P ≤ 0.01, ***P ≤ 0.001; any statistical relationship not indicated means not significant.

Figure 3.

Rheological characterizations of hydrogels with telechelic crosslinking, using crosslinkers with either linear or star architectures. A-B) The elastic modulus and stress relaxation half-time of LX hydrogels and varying concentrations of PEG-dialdehyde crosslinkers of different molecular weight. C-D) The elastic modulus and stress relaxation half-time of SX hydrogels and varying concentrations of 4-arm PEG-dialdehyde crosslinker of different molecular weight. E-F) Elastic modulus and stress relaxation half-time of LX versus SX hydrogels. Multiple unpaired Welch’s t-test with Holm-Šídák multiple comparison. G) DCC alginate hydrazone hydrogels with telechelic crosslinking present correlation of stress relaxation half-time and elastic modulus. Linear regression lines were overlayed, dashed lines for LX hydrogels and solid lines for SX hydrogels. Pearson r coefficient and P value: linear 1 kDa (0.69, *), 2 kDa (0.62, *), and 10 kDa (0.59, *); star 2 kDa (0.92, ****), 10 kDa (0.76, **), and 20 kDa (0.82, **). * P ≤ 0.05, ** P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 4.

DCC hydrogel stiffness and viscoelasticity are crosslinker architecture dependent. A) The elastic modulus, stress relaxation half-time, and loss tangent of hydrazone hydrogels with SCX, LX, and SX collected in this work. B) The stress relaxation half-time and loss tangent of hydrazone hydrogels with SCX, LX, and SX collected in this work. C) Combined data from previous works and this work demonstrate hydrazone hydrogels with SCX, LX, and SX present distinct range of stiffness and stress relaxation half-time with minimal overlaps. SCX (circle), LX (square), and SX (triangle).

Figure 5.

3D cell culture and impact of cell encapsulation on hydrogel mechanical properties. AB) The impact of encapsulating 10 million cells per milliliter in side-chain, linear, and star crosslinking gels on stiffness and stress relaxation half-time. Multiple unpaired Welch’s t-test with Holm-Šídák multiple comparison. C-D) The cross-section area and circularity of MDA-MB-231 cells encapsulated in SCX and SX hydrogels of similar stiffness (17.8 vs 18.7 kPa) but varied stress relaxation time (4,492 vs 1,139 sec). n=148 for SCX and n=62 for SX. Two-tailed unpaired t-test. E) representative images of cell morphology in SCX and SX hydrogels, stained with Calcein-AM dye in green. Scale bar = 100 μm. ns, not significant; * P ≤ 0.05, ****P ≤ 0.0001.