Introduction

The extracellular matrix is more than a passive scaffold for cells; it also provides dynamic biochemical and biophysical signals that regulate cell behavior and fate. Among these signals, ECM mechanics, especially elasticity, have been extensively studied and shown to influence various cellular processes such as cell spreading, migration, proliferation, stemness, differentiation, and cancer cell invasion [1]. Most natural tissues exhibit viscoelasticity or display features of both elastic solids and viscous fluids [2]. For instance, biological tissues such as liver, breast, muscle, skin, and adipose rapidly relax stress upon deformation, while other tissues such as bone, tendon, ligaments, and cartilage have slower relaxation rates [3]. Recently, synthetic ECMs with tunable viscoelastic properties have been developed to show that cells are responsive to differences in viscoelasticity in a variety of ways. In matrices that rapidly relax stress, MSCs undergo more cell spreading, matrix remodeling, and even osteogenesis compared to cells in slower-relaxing matrices [4, 5]. Viscoelasticity has also been shown to influence cancer cell proliferation and migration, intestinal organoid expansion, chondrocyte morphology, and focal adhesion organization of fibroblasts [3]. Mimicking the viscoelasticity of the native ECM is essential to developing accurate cellular microenvironments for in vitro culture models. Various crosslinking techniques have been used to create viscoelastic hydrogels, including ionic bonding, hydrogen bonding, hydrophobic interaction, metal–ligand coordination, supramolecular interaction, and dynamic covalent linkage, providing an expanding chemical toolkit to develop synthetic ECM models [6].

In many biological processes, the mechanical properties of the ECM can change over the course of time. For example, the stiffness of the ECM increases over time in most solid tumors and other fibrotic diseases, due to increased deposition and remodeling of the ECM [7]. In the last decade, hydrogel cell culture platforms have been developed to enable stiffening over time to model these biological processes. An early example utilized methacrylated hyaluronic acid that allowed for temporal matrix stiffening in the presence of cells through a step-wise approach of addition and light-mediated crosslinking [8]. When the hydrogel substrate is stiffened, human MSCs exhibited greater traction and increased area, and showed selective differentiation based on the culture period before or after stiffening [8]. Other light-based approaches have been developed to enable stiffening in PEG, alginate, and GelMA [9, 10]. However, many of these stiffening hydrogels are nearly or purely elastic by the nature of their crosslinking chemistries and do not recapitulate the viscoelastic behavior of native ECM. Here, we present a strategy to achieve temporal stiffening in a 3D hydrogel cell culture platform while maintaining viscoelasticity.

Our approach relies on a system of two orthogonal crosslinking chemistries, one that is reversible and enables viscous flow of the polymer network (ionically crosslinked alginate), and another that is light triggered to stiffen the matrix over time (gelatin methacrylate (GelMA)) (Fig. 1). GelMA is a biomaterial obtained from gelatin, with acrylate groups that form irreversible covalent bonds upon light exposure in the presence of a photoinitiator [11]. GelMA hydrogel mechanical properties can be tuned over the range of stiffness of soft tissues, and it contains cell-binding motifs that enable cell adhesion [11]. Several studies have demonstrated the ability to modulate the stiffness of GelMA hydrogels by varying the crosslinking density, light exposure time, or temperature [12,13,14,15,16]. In our system, GelMA can be stiffened over time by exposure to light, while alginate’s relatively weak ionic crosslinks can unbind and re-bind under load to relax stress. The viscoelastic properties of alginate depend on several factors, including the molecular weight of the polymer, and the type and degree of crosslinking [4]. Higher molecular weight alginate yields slower-relaxing hydrogels with a lower loss modulus, while lower molecular weight alginate yields faster-relaxing hydrogels with a higher loss modulus [4].

Fig. 1
figure 1

Schematic illustrating our approach to achieve a viscoelastic, dynamically stiffening hydrogel platform. 3D GelMA-alginate IPN hydrogels are formed by ionic crosslinking to enable stress relaxation. Stiffness can then be increased by light-mediated crosslinking of GelMA. The degree of stiffening depends on light dosage. Graphics were created with BioRender.com

Full size image

Modeling the viscoelastic properties of the ECM and temporal stiffness changes in the cell microenvironment simultaneously is crucial for developing 3D cell culture systems which can be used for a variety of biomedical applications such as in vitro disease models, drug testing platforms, and tissue engineering. However, it remains a challenge to temporally vary the stiffness of a hydrogel while maintaining its viscoelasticity. Here, we introduce a step-wise dynamic crosslinking approach, ionic then light-mediated covalent crosslinking, to generate viscoelastic hydrogels, which can be stiffened upon light exposure, and investigated the cell viability response to dynamic stiffening.

Methods and materials

Synthesis of alginate, GelMA, and hydrogel preparation

Low-molecular weight alginate was obtained from Dupont Corporation (ProNova UP VLVG; 35 kDa MW) and purified by dialysis against deionized water for 3 days, filtered with activated charcoal, and lyophilized following a previously reported method [4]. Serum and phenol-red free Dulbecco's modified Eagle's medium (DMEM) was used to dissolve alginate at 6 wt%. GelMA was prepared by a common method based on the direct reaction of gelatin with methacrylic anhydride (MA) [17]. The components were mixed to obtain hydrogels with a final concentration of 2% w/v alginate, 0.5% w/v photoinitiator, 5% w/v GelMA, and 25 mM CaSO4. More details are provided in the Supplemental Materials.

Mechanical characterization

In situ rheological experiments were performed with a TA Instruments ARES-G2 stress-controlled rheometer with a UV photocuring fixture (365 nm wavelength, 6 mW cm−2). Solution gelling behavior was monitored using dynamic oscillatory time sweep experiments at 0.5% strain and 1.0 rad s−1 for 2500 s. After the storage modulus reached an equilibrium value, a stress relaxation test was performed for 1000 s at a constant strain of 15%. We measured the elastic modulus after stiffening with a shear rheometer (Anton Paar, MCR 502) using swollen hydrogels. We performed both frequency sweeps and stress relaxation tests. For oscillatory tests, 1% strain was applied while varying frequency from 0.1 to 15 Hz. We measured the shear modulus (G’) and loss modulus (G’’) and converted it to Young's modulus by assuming a Poisson's ratio of 0.5, due to the incompressible nature of water, which constitutes the majority of the gel volume. For stress relaxation tests, 10% strain was applied to the samples, and the resulting stress was measured for 1000 s.

In situ stiffening during cell culture

Mouse bone marrow stromal (D1) cells (passage 6) were cultured within 3D GelMA-alginate hydrogels in 24-well glass-bottomed plates with a cell density of 2 × 106 cells mL−1 with DMEM (4.5 g L−1 glucose) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Hydrogels were stiffened at a user-defined time during the cell culture via radical polymerization by light irradiation (6 mW cm−2, 405 nm). Before stiffening, the media (800 µL per well) was replaced with LAP-containing media (0.5% w/v) and incubated for 30 min at 37 °C, followed by exposure to light. Immediately after exposure, the substrates were washed three times with fresh media.

Statistical analysis

All data obtained from each group were averaged and presented as mean ± standard deviation (SD). Elastic modulus, stress relaxation time, loss factor, and cell viability of each group were compared by one-way analysis of variance (ANOVA) with Tukey post hoc test. Values were considered to be significantly different when the p value was < 0.05.

Results and discussion

We first modified gelatin with methacrylamide groups to form photocrosslinkable hydrogels (GelMA) (Fig. 1). The degree of methacrylation was measured by fluorescamine protein assay, which quantified the number of amino groups before and after the substitution of gelatin [18, 19]. The percent of methacrylation was found to be 98% ± 0.2, and this result is consistent with other studies describing GelMA functionalization [20]. To make the GelMA-alginate IPN hydrogels, first GelMA was dissolved in DMEM containing the photoinitiator lithium phenyl-2,4,6 trimethylbenzoylphosphinate (LAP) and mixed with low molecular weight alginate (35 kDa, Pronova UP VLVG) and CaSO4. Alginate (2% w/v), LAP (0.5% w/v), and CaSO4 (25 mM) concentrations were kept constant for all the experimental groups in this study. After initial ionic gelation of the alginate, the gel could be exposed light for different durations to subsequently increase the stiffness (Fig. 1).

For mechanical characterization of IPN hydrogels, in situ rheology was performed between an 8-mm-diameter parallel plate tool with a stress-controlled rheometer at 37 °C. Initial gelation resulted from calcium crosslinking the alginate network and entrapping GelMA. The hydrogels were then exposed to light for 300 s (6 mW cm−2 at 365 nm, Omnicure 2000, Lumen Dynamics) to initiate radical crosslinking and stiffening of GelMA. As shown in Fig. 2A, the values of the storage modulus (G′) at 1 rad s−1 increased from ~ 1.6 kPa in the gels made of GelMA-alginate at a low GelMA concentration (2.5% w/v) to ~ 5 kPa after photocrosslinking. When the GelMA concentration was increased to 5% w/v, the storage modulus was increased from ~ 2.0 to ~ 21.0 kPa after light exposure (Fig. 2B). Thus, the range of tunable elastic moduli for the GelMA-alginate IPN hydrogels vary from ~ 3 to ~ 60 kPa, depending on the concentration of GelMA.

Fig. 2
figure 2

GelMA-alginate hydrogels can be stiffened by light while maintaining stress relaxation capacity. A Time sweep rheology of ionically crosslinked GelMA-alginate hydrogels prior to light exposure, and stiffened hydrogels as a result of photocrosslinking (365 nm light, 5 min exposure, 6 mW cm−2). B The magnitude of stiffness change can be enhanced by incorporating a high concentration of GelMA (5%). C Stress relaxation tests at 15% strain show substantial relaxation for stiffened GelMA-alginate gels (2.5% GelMA), similar to alginate-only controls. GelMA-only controls behave nearly elastically. D Stress relaxation in high concentration GelMA-alginate stiffened gels is maintained. For all groups, one representative experiment of at least three replicates are shown

Full size image

These results are consistent with previous research that reported that the elastic modulus of GelMA hydrogels increased proportionally with polymer concentration or light exposure [21]. We then characterized the stress relaxation profile of the dual crosslinked IPN hydrogels. Purely covalently crosslinked GelMA resulted in a predominantly elastic material, as expected since covalent crosslinks are permanent and lead to elastic gels (Fig. 2C, D) [22]. However, incorporation of alginate within the hydrogels generates varying degrees of stress relaxation independent of the stiffness of the composite (Fig. 2C, D). The stress relaxation rate of GelMA-alginate IPNs was nearly as fast as control hydrogels made from only low molecular weight (LMW) alginate, even after stiffening the gel by crosslinking the GelMA network (Fig. 2C, D). We found that increasing the GelMA concentration in the gel increases the stiffness without significantly altering the stress relaxation profile.

In order to maximize the cytocompatibility of this system, we subsequently used a visible-light source (405 nm). To simulate a potential experimental timeline in a 3D cell culture experiment, we initially (Day 1) formed a GelMA-alginate hydrogel through Ca2+ crosslinking and by briefly exposing it to light for 5 s for photocrosslinking (30 mJ cm−2) (Fig. 3A). On the second day, the hydrogels were subjected to longer periods of light exposure to increase their stiffness (Fig. 3A). On day 2, a sharp increase in storage modulus was observed when the initially soft hydrogel (0.7 kPa ± 0.24) was exposed to 405 nm light for either 15 s (90 mJ cm−2) or 60 s (360 mJ cm−2), resulting in elastic moduli of 6.4 kPa ± 0.48 and 17.9 kPa ± 0.04, respectively (Fig. 3B). The viscoelastic characteristics of the hydrogels were assessed by conducting loss factor measurements and stress relaxation tests. The results indicated that the pure GelMA hydrogel exhibited a substantially lower tan delta value (0.04 ± 0.01) in comparison to the GelMA-alginate IPN hydrogels, which is expected since GelMA hydrogels are usually found to be predominantly elastic (Fig. 3C) [23]. In addition, no significant difference was observed between the tan delta values of GelMA-alginate IPN hydrogels, (Fig. 3C). Stress relaxation measurements revealed that after stiffening, the GelMA-alginate IPN hydrogels still maintained their viscoelasticity, with relaxation half-times of 779 s, 562 s, and 523 s for soft, soft-medium, and soft-stiff groups, respectively (Fig. 3D, E). We did not observe any significant changes between the stress relaxation times of gels exposed to different light intensities (Fig. 3E). Importantly, the stress relaxation rates are independent of hydrogel stiffness. While the addition and covalent crosslinking of the GelMA network resulted in slower relaxation rates than LMW pure alginate (~ 100 s), these gels still exhibited faster stress relaxation than previously reported gels formed from higher molecular weight alginates and are within the range at which cells are responsive to stress relaxation [4].

Fig. 3
figure 3

3D viscoelastic GelMA-alginate hydrogels can be dynamically stiffened. A Schematic diagram showing the stiffening strategy after 24 h in culture medium. B The elastic modulus can be increased as a function of light exposure (n ≥ 3, P < 0.05). C Loss tangent, a measure of viscoelasticity (ratio of loss modulus to storage modulus), is significantly higher for GelMA-alginate gels compared to GelMA controls. (n ≥ 3, P < 0.05) D Stress relaxation tests for GelMA-alginate hydrogels and E quantification of relaxation half-time from stress relaxation tests in D. Neither the loss tangent nor the relaxation half-time was significantly different with respect to light exposure (n = 3, P < 0.05). Results are reported as mean ± standard deviation

Full size image

To test the biocompatibility of the system, mouse bone marrow stromal (D1) cells were cultured within 3D GelMA-alginate hydrogel in 24-well glass-bottomed plates with a cell density of 2 × 106 cells mL−1 and hydrogels were stiffened on days one and two of the cell culture via radical polymerization using light irradiation (405 nm) (Fig. 4A). Before stiffening, the media was replaced with LAP-containing media and incubated for 30 min at 37 °C, followed by exposure to light for 5 s (30 mJ cm−2), 15 s (90 mJ cm−2), and 60 s (360 mJ cm−2) (Fig. 4A). Immediately after exposure, the substrates were washed with fresh media, and after 24 h, live cells were stained with calcein AM, while dead cells were stained with ethidium homodimer. Cells cultured on tissue culture plastic (TCP) were used as a control group for the light exposure and gel. Results from the live-dead assay showed that up to 85% of cells were viable after 360 mJ cm−2 of light irradiation, with no significant differences with respect to duration of light exposure (Fig. 4B, C). Thus, the stiffening approach we have shown here is amenable to maintaining viable cells in 3D culture.

Fig. 4
figure 4

Dynamic stiffening in 3D GelMA-alginate matrices is cytocompatible. A Experimental timeline to evaluate viability of mouse bone marrow stromal (D1) cells exposed to 405 nm light for 5 s (30 mJ cm−2), 15 s (90 mJ cm−2), and 60 s (360 mJ cm.−2) in 3D GelMA-alginate hydrogels. Images were taken on days 1 and 2 after dynamic stiffening. Viable cells are labeled with calcein AM (green), and dead cells are labeled with ethidium homodimer (red). B Quantification of cell viability. Results are reported as mean ± standard deviation (n = 3) (P < 0.05)

Full size image

This system demonstrates an advance in mimicking dynamic microenvironmental conditions of the native ECM in many tissues. For example, this GelMA-alginate IPN could be applied to investigations of stem cell lineage commitment and maintenance in 3D hydrogels, which has been reported within the range of our system, where adipogenesis is favored in hydrogels with an elastic modulus of 2.5–5 kPa, while osteogenesis is favored in those with a modulus of 11–30 kPa [15]. This system can also be used in promoting the maturation of human stem cell-derived cardiomyocytes (hSC-CMs). In fact, researchers have developed techniques to differentiate human stem cells into cardiomyocytes, but these cells are often immature for practical applications such as drug testing and cardiac disease modeling [24]. To address this problem, hSC-CMs can be matured in a microenvironment that mimics the stiffening of heart tissue during development [25]. The GelMA-alginate IPN hydrogel we developed in this study can be used for this purpose with its stiffness within the range of embryonic myocardium (~ 6 kPa) and healthy murine myocardium (10–40 kPa). In addition, dynamically stiffened matrices provide a useful platform for investigating how mechanical changes in the microenvironment of cardiomyocytes can activate the regulation of their function [26]. Notably, our system's stiffness range also falls within the pathophysiological range of healthy tissue and malignant tumors such as breast, liver, pancreas, and lung cancers [27,28,29,30].

In conclusion, we have developed a biocompatible 3D hydrogel platform in which stiffness can be dynamically tuned independent of its viscoelastic features, such as stress relaxation time and loss tangent. The incorporation of alginate in the IPN hydrogel provides viscoelasticity, while radical polymerization between chains of GelMA enabled a temporal modulation of stiffness from ~ 1 to ~ 18 kPa. Importantly, the changes in stiffness did not significantly alter the stress relaxation rate of the hydrogels. Over 85% of the cells remained viable after stiffness tuning demonstrating good cytocompatibility. We believe that this new IPN hydrogel system represents a significant advancement in the development of 3D cell culture systems and will enable researchers to better investigate how cells interact with their dynamic microenvironment.