Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Feb 20.
Published in final edited form as: J Biomed Mater Res A. 2023 May 18;111(10):1577–1587. doi: 10.1002/jbm.a.37556

Dynamically crosslinked thermoresponsive granular hydrogels

Hung-Pang Lee 1, Kathy Xiao Cai 1, Ting-Ching Wang 2, Ryan Davis Jr 1, Kaivalya Deo 1, Kanwar Abhay Singh 1, Tanmay P Lele 1,2,3, Akhilesh K Gaharwar 1,4,5,6
PMCID: PMC12919407  NIHMSID: NIHMS2142604  PMID: 37199446

Abstract

Granular hydrogels are a promising biomaterial for a wide range of biomedical applications, including tissue regeneration, drug/cell delivery, and 3D printing. These granular hydrogels are created by assembling microgels through the jamming process. However, current methods for interconnecting the microgels often limit their use due to the reliance on postprocessing for crosslinking through photoinitiated reactions or enzymatic catalysis. To address this limitation, we incorporated a thiol-functionalized thermo-responsive polymer into oxidized hyaluronic acid microgel assemblies. The rapid exchange rate of thiol-aldehyde dynamic covalent bonds allows the microgel assembly to be shear-thinning and self-healing, with the phase transition behavior of the thermo-responsive polymer serving as secondary crosslinking to stabilize the granular hydrogels network at body temperature. This two-stage crosslinking system provides excellent injectability and shape stability, while maintaining mechanical integrity. In addition, the aldehyde groups of the microgels act as covalent binding sites for sustained drug release. These granular hydrogels can be used as scaffolds for cell delivery and encapsulation, and can be 3D printed without the need for post-printing processing to maintain mechanical stability. Overall, our work introduces thermo-responsive granular hydrogels with promising potential for various biomedical applications.

Keywords: 3D printing, cell delivery, drug delivery, dynamic biomaterials, granular hydrogels, regenerative medicine

1 |. INTRODUCTION

Injectable hydrogels are a promising biomaterial for a wide range of applications, including therapeutics delivery and 3D printing.1–4 However, conventional covalent crosslinking can compromise the important characteristics of injectable hydrogels such as shear-thinning and self-healing, making them mechanically unstable.5,6 To address this limitation, various non-covalent crosslinking systems have been used, including hydrophobic, ionic, and guest–host.7,8 Another promising strategy is the use of dynamic covalent bonds, such as imine, hydrazone, and boronate ester, for designing injectable hydrogels that are shear-thinning and self-healing.9,10 However, many of these hydrogels have rapid degradation and unstable crosslinking, making them unsuitable for long-term load-bearing applications.11,12 Thus, there is an urgent need for injectable hydrogels with post-injection stability for long-term biomedical use.

Granular hydrogels consisting of microgel assemblies have emerged as a popular soft matter for designing injectable biomaterials in recent years. These granular hydrogels offer several advantages over covalently crosslinked bulk hydrogels.13–15 For example, microgels can have the same mechanical properties, such as elastic modulus and stiffness, as bulk hydrogels with the same polymer formulations on a micro-scale. Additionally, granular hydrogels formed through the jamming process of microgel assemblies have good injectability with a shear-thinning effect, while bulk hydrogels typically have poor injectability due to high yield stress.16 When packed and jammed, microgels physically trap each other, providing self-support and structural integrity. During injection under high shear stress, jammed microgels yield to flow when external forces overcome the interparticle friction and physical interaction. Moreover, microgels crosslinked by stable covalent bonds can have a longer degradation period and lower swelling ratio compared to other shear-thinning hydrogels crosslinked by reversible bonds.17 However, a key challenge with microgel assemblies has been interconnecting the microgels to enhance structural integrity and mechanical moduli. Traditional methods of generating secondary crosslinking through photoinitiated reactions or enzymatic catalysis can complicate the injection process and limit the broader use of the microgel assemblies in the body.18–20

One promising approach for designing stable injectable hydrogels is to leverage the sequential change in crosslinking mechanisms from a reversible to a stable network. Injectable hydrogel with secondary autonomous covalent crosslinking by combining noncovalent interaction and covalent crosslinking sequentially are recently developed.11 The reversible host–guest interaction allows the hydrogels to be temporarily injectable, while the thermodynamically stable thiol-ene bonds make the hydrogels mechanically stable as a long-term scaffold. Another approach to achieving post-injection stability in injectable hydrogels is by tuning the reversibility of dynamic crosslinking based on environmental circumstances.21 They used a biocompatible catalyst to accelerate the reversibility of hydrazone bonds, a type of dynamic covalent bond that improves hydrogel injectability for cell delivery.21 As the catalysts diffused from the hydrogel, the hydrogel gained high stability and slowed erosion post-injection. These approaches offer not only readily conformable hydrogels for convenient manipulation during medical procedures but also long-term stability for cell and therapeutic delivery. Overall, leveraging sequential changes in crosslinking mechanisms provides a promising strategy for designing stable injectable hydrogels with a range of biomedical applications.

Here, we present a new approach for programming granular hydrogels with good injectability and a stable crosslinking network by synergizing the advantages of microgels, dynamic covalent bonds, and thermo-responsive cross-linking. First, we selected oxidized hyaluronic acid microgels equipped with abundant aldehyde groups to form hemithioacetal, a dynamic covalent bond generated by thiol-aldehyde addition, with thiolated thermo-responsive polymers. Hemithioacetal, compared to other dynamic covalent bonds, has preferred reaction kinetics and reversible thermodynamics (k1 = 0.27 M−1 s−1, k−1 = 2.1 × 10−3 s−1, and Keq = 1.3 × 102 M−1) to form a shear-thinning and self-healing hydrogel.22 However, the rapidly reversible dissociation of hemithioacetal can lead to fast degradation.23 To address this limitation, we introduced secondary hydrophobic interactions generated by the phase transition of thermo-responsive polymers to stabilize the microgel assemblies. The thermo-strengthening behavior allows the microgel assemblies to be transformed from an injectable to a mechanically stable state at physiological temperature. Overall, this design enables the hydrogel to be developed for use as a therapeutic delivery system and 3D printing ink, with promising potential for a range of biomedical applications.

2 |. MATERIALS AND METHODS

2.1 |. Synthesis of HA-ALD-MA, PNAM, PNA, and PAM

A total of 400 mg of hyaluronic acid sodium salt (HA, 95%, Thermo Scientific) was dissolved in 100 mL DI water. A total of 270 mg of sodium periodate (NaIO4) was dissolved in 10 mL DI water first and added to the HA slowly. The reaction continued overnight and was terminated by adding 1 mL of ethylene glycol. Then, the solution was dialyzed against DI water for 3 days and lyophilized to obtain HA-ALD. To synthesize methacrylated and aldehyde-modified hyaluronic acid (HA-ALD-MA), 500 mg of HA-ALD was added to 25 mL DI water in an ice bath, and adjusted the pH value to 8.5 using 1 M NaOH. Then, 1.125 mL methacrylate anhydride was added slowly to the solution while controlling the pH value at 8–9 for 4 h. The reaction proceeded at room temperature overnight. Then, the solution was dialyzed against DI water for 3 days and lyophilized to obtain HA-ALD-MA.

Based on our previous paper, poly(N-isopropyl acrylamide-co-acrylamide-co-N, N′-bisacryloylcystamine; PNAC) hydrogels were synthesized first to produce poly(N-isopropyl acrylamide-co-acrylamide-co-2-mercaptoethylacrylamide; PNAM) copolymer later by reductive liquefaction.24 In brief, the monomer solution with the mole ratio of N-isopropyl acrylamide (NIPAM), acrylamide (AM), and N,N′-bisacryloylcystamine (BAC) was designed at 9:1:0.2. For instance, 0.052 g BAC was first dissolved in 70% ethanol and added into the aqueous solution of 1.017 g NIAPM and 0.0071 g AM. Then, 300 μL of N,N, N′,N′-tetramethylethylenediamine (TEMED) was added to the mixture with nitrogen bubbling for 30 min to remove any dissolved oxygen. The polymerization was initiated by adding 600 μL of freshly prepared 10% wt/vol ammonium persulfate and allowed to proceed overnight at room temperature. The resulting hydrogels were placed into 1 L of deionized water to remove unreacted monomers for 2 days. 0.616 g of D,L-dithiothreitol (DTT) was then added to reduce the disulfide bonds of PNAC hydrogels with nitrogen bubbling for 4 h. The resulting solutions were dialyzed against deionized water at pH = 4 to prevent the oxidation of thiol.

Finally, free drying was applied to obtain PNAM copolymers. Poly(N-isopropyl acrylamide-co-acrylamide) (PNA) and poly(acrylamide-co-2-mercaptoethylacrylamide) (PAM) were synthesized by the same method but with varying concentrations of monomers. Ellman’s assay was used to quantify the thiol concentrations of different polymer solutions. In brief, Ellman’s reagent (5,5’-Dithio-bis-2-nitrobenzoic acid, Thermo Scientific) and 0–1.5 mM Cysteine was used to make a standard curve with absorbance at 412 nm. Then, 0.5% of polymers dissolved in DI water were used to react with the Ellman reagent.

2.2 |. Preparation of HA-ALD-MG and dynamic microgel assembly

Aldehyde-modified hyaluronic microgels were fabricated by photocrosslinking and extrusion fragmentation.25 0.5 mL of 4% HA-ALD-MA solution and 0.5% irgacure were added to a 3 mL syringe (BD). The syringe was exposed to UV light for 10 min (20 mW/cm2). The bulk hydrogel formed in the syringe was then extruded by hand through 18G, 21G, 23G, and 27G needles sequentially. Microgels were then washed in pure PBS three times and lyophilized for further uses.

To make aldehyde-modified microgel (HA-ALD-MG) and PNAM microgel assemblies, 10% of HA-ALD-MG and 5% of PNAM were dissolved in PBS separately. Then, the PNAM solution was added to the completely hydrated microgels, and the crosslinking started instantly upon mixing to form microgel assemblies. The final concentration of the microgel assemblies is 5% HA-ALD-MG and 2.5% PNAM.

2.3 |. Hydrogel characterization

All polymers were characterized using 1H-NMR. To measure the sizes of HA-ALD-MG, prior to fabrication, FITC-dextran (2 MDa, 0.1 wt %) was added to microgel precursor solutions. Fluorescence microscopy was used to image the microgels after fabrication, and ImageJ was used to quantify the microgel’s maximum length. For confocal images, PNAM has first conjugated to Alexa Fluor 488 Maleimide by Michael addition. The gelation kinetics, viscoelastic properties, and temperature/time sweeps of the precursor solutions and hydrogel samples were analyzed using an oscillatory stress-controlled rheometer (Discovery HR-2, TA instruments) equipped with 20 mm diameter geometry and a 2 mm gap. To determine the compressive modulus of the hydrogels, cyclic compression testing was performed using ADMET eXpert 7600 system (ADMET, Inc., Norwood, Massachusetts) with an attached load cell of 25 lb. The testing was performed at a strain rate of 1 mm/min on cylindrical crosslinked hydrogel samples ~7.5 mm × 7 mm (diameter thickness) with 30% strain, while compressive modulus was calculated corresponding to 10%–20% strain region of the engineering stress–strain curve.

2.4 |. Drug loading and drug release of DOX

To load DOX into HA-ALD-MG covalently, 100 mg of HA-ALD-MG was dissolved in acetate/acetic acid buffer solution at pH 5.0 in 10 mL, and followed by the addition of 2.5 mg of DOX. The reaction was performed at 45°C for 72 h in darkness. The excess drug was removed by centrifuging twice with 10 mL DI water, which was collected to quantify loading efficacy. DOX concentrations were determined by measuring the absorbance at 490 nm. In vitro drug release assay was performed in PBS at pH 7.4 and 5.0, and PBS contained 10 mM glutathione (GSH).

2.5 |. In vitro assays

Breast cancer cells (MCF-7, ATCC, USA) and fibroblast (NIH/3T3, ATCC, USA) were cultured in normal growth media (DMEN, Gibco, USA), supplemented with 10% FBS (Life Technologies, USA) and 1% penicillin/streptomycin (100 U/100 μg/mL; Life Technologies, USA) at 37°C with 5% CO2. The metabolic activity of breast cancer cells was measured by Alamar blue using the indirect method. Briefly, 100 μL of DOX-loaded microgel assemblies was immersed in 1 mL of cell culture media to extract the released DOX. The released solutions were then transferred to MCF-7 cells with normal media as control. The cell viability tests of MCF-7 were done by live/dead staining and flow cytometry with the microgel assemblies placed in inserted wells. Cells were co-cultured with the hydrogels for 48 h and stained with calcein AM and ethidium homo-dimer. These cells were analyzed using BD accuri C6, where cell populations were initially gated based on forward scatter/side scatter profiles, with subsequent gating placed on fluorescent signal plots to quantify cell viability. To test the cell viability of the microgel assemblies with and without modification with cell adhesion peptides, Maleimide-RGD was used to conjugate to PNAM. In brief, 1 mM maleimide-RGD (Cyclo[Arg-Gly-Asp-D-Phe-Lys(Mal)], Vivitide) was dissolved in PBS at pH 8, which was further used to dissolve PNAM. 2.5% PNAM-RGD was then mixed with 5% HA-ALD-MG to obtain cell-adhesive microgel assemblies. 3T3 cells were seeded on the microgel assemblies, which metabolic activities and cell morphology were studied by alarmar blue, cytoskeletal, and nuclear staining, respectively. To show the cell viability of encapsulated cells after injection, 3T3 cells were encapsulated in the RGD-functionalized microgel assemblies and assessed by Live/Dead assays after injection. For 3D cell encapsulation, 1% gelatin methacryloyl (GelMA) and 3.5% HA-ALD-MA were used as precursor solutions for fabricating cell-adhesive microgels. Immediately before encapsulation, cells were trypsinized for 5 min, then 10 mL of media was added, and the resulting cell-media suspension was centrifuged for 5 min. The super-natant was removed, and the cell pellet was resuspended in a small amount of media, which was gently mixed into the PNAM solution and transferred to the HA-ALD-MGs slowly by pipetting. After 10 min, to ensure the formation of crosslinking, the samples were transferred to an incubator for secondary crosslinking, cell culture, and further analysis.

2.6 |. 3D printing

The 3D printing was done on a custom extrusion printer26 with a screw extruder-based printing head. The designs were generated in Solidworks 2019 and converted to STL files. The STL files were sliced and converted into G-code using Slic3r software. Repetier Host v.2.1.1 was used as a user interface for controlling the 3D printer. Layer height was kept at 200 μm, layer width was set at 600 μm, and printing speed was 10 mm/s. 3D printing was performed with a 410 μm or 1 mm tapered extruder tapered tip attached to the extruder. The filament uniformity extruded filaments were quantified using ImageJ. To determine filament uniformity, in each image of the 2D snaking lines, 15 thickness measurements were taken along each of the four long lines and compared to the specified thickness in the. stl files.

2.7 |. Statistical analysis

All results are expressed as the mean ± deviation (n = 3–100). Statistical analysis of all quantitative data was performed via one-way or two-way ANOVA with post hoc Tukey tests or unpaired t-test using GraphPad Prism (v 6.01).

3 |. RESULTS AND DISCUSSION

3.1 |. Preparation of thiolate- and thermo-responsive polymers and aldehyde-modified hyaluronic acid microgels

Amine, hydrazide, and hydroxylamine groups are commonly used nucleophiles that can form dynamic covalent bonds with aldehyde groups, resulting in self-healing, injectable, and moldable hydrogels for various biomedical applications.27,28 Imine, hydrazone, and oxime have different equilibrium constants, association rates, and dissociation rate constants, which affect the gelation kinetics, stiffness, and stress relaxation rates of the hydrogels. More recently, thiol groups have been used to react with aldehyde groups to form hemithioacetal-crosslinked hydrogels. Due to the rapid exchange rate of thiol and hemithioacetal, the hydrogel can be readily adaptable for convenient manipulation in examples of injectable hydrogels and 3D printing.22,23,29 However, the high reversibility of hemithioacetal also causes the relatively fast degradation of the hydrogel, limiting its biomedical applications for long-term use. To equip the hydrogel with injectability, self-healing ability, and mechanical stability together, we have designed a thermo-responsive microgel assembly using hydrophobic interaction and thiol-aldehyde addition (Figure 1).

FIGURE 1.

FIGURE 1

Open in a new tab

Injectable and thermo-responsive granular hydrogels consisting of microgel assemblies formulated by synthetic polymers (PNAM) and hyaluronic acid-based microgels. (A) Thermo-responsive and thiolated polymers (PNAM) were blended with oxidized microgels to form shear-thinning and self-healing microgel assemblies by dynamic covalent crosslinking. Due to the hydrophobic interaction of PNAM at 37°C, the microgel assemblies became mechanically stable. (B) Methacylated and aldehyde modified hyaluronic acid (HA-ALD-MA) was used to fabricate oxidized HA microgels (HA-ALD-MG) by UV crosslinking and fragmentation. The thiol groups of PNAM can react with the aldehyde groups of HA-ALD-MGs to form hemithiolacetal. After heating, due to hydrophobic interaction of PNAM, the polymer aggregation facilitated the formation of disulfide bonds, which contributed to the structural mechanical stability of microgel assemblies.

First, we have synthesized a thermo-responsive and thiol-functionalized polymer, poly(N-isopropyl acrylamide-co-acrylamide-co-2-mercaptoethylacrylamide) (PNAM) to conjugate with aldehyde-modified microgels, based on our previous study. PNAM was prepared by a two-step process (Figure S1). First, PNAC hydrogels containing disulfide bonds were synthesized by free-radical polymerization of N-isopropyl acrylamide (NIPAM), acrylamide (AM), N, N′-bisacryloylcystamine (BAC) with a monomer ratio of 9:1:0.2, respectively. Then, the hydrogels were liquified by D,L-dithiothreitol (DTT), followed by dialysis and lyophilization to obtain PNAM, the structure of which was confirmed by 1H-NMR spectrum in deuterium oxide (D2O) (Figure S2). The other two polymers, PNA and PAM, were also synthesized through the same method with different monomer formulations as control groups.

To produce HA-ALD-MG, we first synthesized a dual modification of methacrylated and oxidized hyaluronic acid. High-molecular weight hyaluronic acid was oxidized using sodium periodate to generate aldehyde groups and reduce the viscosity of the polymer solution. Methacrylate anhydride was then introduced to the oxidized hyaluronic acid to create HA-ALD-MA, resulting in the addition of alkenes to the polymer. It is worth noting that the methacrylation process required a nonviscous solution to allow for proper dispersion of the methacrylate anhydride in the aqueous phase. The oxidation and methacrylate degrees of HA-ALD-MA were determined by the 1H-NMR spectrum in deuterium oxide (D2O), which were approximately 8.6% and 33%, respectively (Figure S3).

3.2 |. Assembling microgels by thermo-responsive polymers

HA-ALD-MGs were prepared by the fragmentation of photo-crosslinked bulk hydrogels. 4% HA-ALD-MA solution mixed with photoinitiators was transferred to 3 mL syringe and exposed to UV light (20 mW/cm2) for 10 min to ensure the completion of crosslinking. The gelation occurred in less than 2 min for HA-ALD-MA, reaching plateau storage moduli (G′) of 3000 Pa for HA-ALD-MA (Figure 2A). Then, HA-ALD-MGs were fabricated with the extrusion fragmentation of the HA-ALD-MA hydrogel through a gradient size of needles from 18G to 27G. The microgels’ maximum length was determined by measuring the longest length of the fluorescently labeled particles (Figure 2B). HA-ALD-MGs with a maximum length of around 50 28μm were then collected, washed, and lyophilized for long-term storage. The sizes of the microgels produced by extrusion fragmentation are less uniform than by microfluidic devices and batch emulsion.25 However, compared to the other two methods, extrusion fragmentation could produce more microgels at a time and have similar rheological properties, injectability, and porosity. Prior to assembling the microgels with PNAM, 10% HA-ALD-MGs were incubated in PBS at 37°C for 30 min to ensure full hydration. Five percent of the PNAM solution was then mixed with the hydrated HA-ALD-MGs, and the gelation occurred in a few minutes with increased viscosity. It was noted that if the hydration was not completed, the gelation would take a longer time to reach a plateau. The porosity of the microgel assemblies was determined by the area of the fluorescently labeled PNAM with Rhodamine (Figure 2C and Figure S4). The void space of the microgel assemblies was around 12.17% ± 4.21%, which allowed the PNAM and HA-ALD-MG closely interact with each other and accelerate the thiol-aldehyde addition.

FIGURE 2.

FIGURE 2

Open in a new tab

The fabrication of microgel and granular hydrogels consisting of microgel assemblies. (A) The gelation kinetic of methacrylated and aldehyde-modified hyaluronic acid (HA-ALD-MA) by UV photocrosslinking. 4% of HA-ALD-MA reached a plateau of storage modulus at 3000 Pa within 2 min of 20 mW/cm2 UV light. (B) aldehyde-modified microgels (HA-ALD-MGs) were produced by the fragmentation of methacrylated and aldehyde-modified hyaluronic acid bulk hydrogel with gradient sizes of needles from 18G to 27G. (C) The confocal image of the microgel assembly formed by PNAM (Red) and HA-ALD-MGs (Green). The PNAM phase occupied approximately 12% of the total volume, and the microgel sizes were around 50 μm, measured by ImageJ and fluorescent images. (D) The shear rate sweep analysis of the microgel assembly indicated its good injectability. (E) 1% and 500% of strain were applied to the microgel assembly to test its self-healing ability. Two pieces of the microgel assemblies can bind together in a few minutes.

The microgel assemblies demonstrated shear-thinning and self-healing properties due to the high reversibility of thiol-aldehyde addition. These properties are essential for the development of injectable hydrogels and hydrogel inks. The shear-thinning behavior was measured for the microgel assemblies using steady shear flow sweeps from low to high shear rates relevant to extrusion and injection processes.30 The microgel assemblies had approximately a four-order magnitude decrease in viscosity with increasing shear rates across the range tested (10−1–103 S−1; Figure 2D). When the hydrogel assemblies were injected and sheared through a needle, the hemithiolacetal between the HA-ALD-MGs and PNAM polymers dissociated. Because of the fast dissociation rate of hemithiolacetal, the microgel assembly can be deformed easily, which gives it a shear-thinning characteristics. The microgel assembly can be easily injected through a 26G needle and recover its structural integrity in a few seconds. The good injectability might also be ascribed to the reduced number of crosslinking required for forming a stable microgel assembly compared to conventionally bulk hydrogels.31 The self-healing behavior of the microgel assembly was assessed by applying alternative 1% and 500% strains (Figure 2E). The storage modulus quickly dropped once the high strain was applied and rapidly recovered once low shear rates were applied. The microgel assembly was observed to efficiently self-heal within seconds after being cut into two pieces and reattached. This remarkable self-healing behavior can help maintain the mechanical and structural stabilities of the hydrogels, which is crucial for various applications, such as 3D printing and therapeutic delivery.

3.3 |. Thermo-responsive stabilization of the HA-ALD-MG and PNAM microgel assembly

While the rapid reversibility of hemithioacetal enabled the microgel assembly to be injectable and self-healing, it also led to quick disintegration due to thiol and aldehyde dissociation. To address this limitation, we incorporated thermo-responsiveness into the microgel assembly to stabilize it after injection by introducing additional hydrophobic interaction at body temperature. The lower critical solution temperature of PNAM was adjusted to 36:5°C by varying the ratio of NIPAM and AM, which was also measured by the temperature sweep of rheological tests (Figure 3A). Two other polymers, PNA and PAM, were also synthesized as controls to investigate the effects of thiol groups and phase transition of the polymers assembling the HA-ALD-MGs. PNAM showed typical thermo-gelling behavior of phase transition polymers by increasing storage modulus at a 2-order of magnitude above its LCST. PNA had a slightly higher storage modulus at 37°C and lower LCST than PNAM. However, PAM had no phase transition behavior and remained flowable across all temperatures tested. For thiol content, PAM and PNAM had minimally different concentrations of thiol groups at 0.5 wt % of the polymers, but PNA had no thiol groups, assessed by Ellman’s assay (Figure 3B).

FIGURE 3.

FIGURE 3

Open in a new tab

Postinjection stability of the microgel assemblies was attributed to the phase transition ability and thiol groups of PNAM. (A) The temperature sweep analysis of PNAM, PNA, and PAM showed the phase transition behavior of PNAM and PNA by their enhancing storage modulus along the increment of temperature from 25 to 40°C. (B) The thiol concentrations of 0.5% PNAM, PNA, and PAM polymers were measured by Ellman’s assay (n = 3, mean ± SD). (C) The temperature sweep analysis of the PNAM, PNA, and PAM microgel assemblies. (D) Change in the transparency of the PNAM microgel assemblies from 25 to 37°C. (E) The strain sweeps of the PNAM, PNA, and PAM microgel assemblies at 25 and 37°C. (F) The yield stress of the PNAM, PNA, and PAM microgel assemblies at 37°C were obtained from the strain sweeps. (n = 4, mean ± SD). (G) The maximum compressive stress and compressive modulus of the PNAM, PNA, and PAM microgel assemblies before and after being incubated in PBS at 37°C for 48 h (n = 3, mean ± SD). Statistics analysis was performed using one-way or two-way ANOVA with *p < .05, **p < .01, ***p < .001, and ****p < .0001.

Then, 5% of PNAM, PNA, and PNAM were blended with 10% HA-ALD-MGs to form microgel assembles. The temperature sweeps of three microgel assemblies showed that the thiol content and phase transition behavior affected their storage modulus at room and body temperature differently (Figure 3C). First, the microgel assemblies of PNAM and PAM had higher storage modulus than that of PNA due to the covalent crosslinking of thiol-aldehyde addition at 25°C. Moreover, the microgel assemblies of PAM had the highest storage modulus at 25°C, probably due to the extra hydrogen bonds between acrylamide and hyaluronic acid. When the temperature increased to 37°C, the modulus of the microgel assemblies of PNAM and PAN gradually enhanced due to the phase transition behavior of NIPAM. The hydrophobic force of the polymers formed secondary physical crosslinking, which further strengthened and stabilized the microgels. Interestingly, for the microgel assemblies of PAM, its modulus slightly decreased at 37°C. The decreasing transparency of the PNAM microgel assembly was also observed when immersed in warm water, indicating the phase transition of PNAM (Figure 3D).

Next, other important rheological properties of injectable hydrogels are the yield stress and strain, which indicate the stress and strain required to trigger flow in a hydrogel.32,33 Low yield stress and strain can enhance the printability and injectability of hydrogels carrying therapeutic drugs and cells. On the other hand, higher yield stress and strain mean greater resilience and persistence of hydrogels in the body and greater retention of therapeutics. Therefore, we programmed the microgel assembly to be injectable in a syringe at 25°C and mechanically stable in the body at 37°C. The yield strain and yield stress of the microgel assemblies were assessed by strain sweep from 0.1% to 500% at a constant angular frequency (ω = 10 rad/s; Figure 3E). The microgel assemblies of microgel alone (HA-ALD-MG) and PAS showed no significant increment of yield strain from 25 to 37°C, which remained at 20% and 150%, respectively. Also, PAS showed a significantly higher yield strain than HA-ALD-MG due to the formation of hemithiolacetal crosslinking. However, for the microgel assemblies of PNA and PNAM, the yield strain increased from 5% to 10% and 140% to 225%, respectively. Then, the yield stress of microgel assemblies was acquired from the yield point where G′ and G″ intersected each other (Figure 3F). The yield stress of the PAM microgel assembly remained at approximately 2000 Pa, but that of the PNAM microgel assembly increased from 1500 to 3000 Pa after heating. Moreover, the microgel assemblies of PNAM had the highest yield stress among all groups at 37°C due to the dual crosslinking of thiol-aldehyde addition and hydrophobic interaction.

To study the mechanical stability of the microgel assemblies after injection into the body, we measured the max stress and elastic modulus of the microgel assemblies before and after 48 h of incubation in PBS at 37°C (Figure 3G). First, due to the weak association between the microgels and PNA, the microgel assembly of PNA had the lowest max stress and modulus at 0 h, was not stable in PBS, and disintegrated after incubation. Also, since all the measurements were done at 25°C, the max stress of PNAM was only slightly higher than that of PAM but significantly higher than that of PNA at 0 h. However, at 48 h, the max stress and elastic modulus of PNAM were both significantly higher than those of PAM, indicating that the introduction of secondary hydrophobic interaction could restrict the disintegration of microgel assemblies. No significant loss in the max stress and modulus of PNAM at 0 and 48 h was found, showing that the aggregation of the PNAM polymers at 37°C might reserve the hemithioacetal crosslinking and form new disulfide bonds by self-oxidation. Moreover, the weight loss and swelling ratio of microgel assemblies were restricted because of intraparticle covalent crosslinking and interparticle hydrophobic interaction (Figure S5). Therefore, the thiol-aldehyde addition and thermo-responsibility of the PNAM microgel assemblies were successfully programmed to fabricate a mechanically stable hydrogel after injection.

3.4 |. Granular hydrogels consisting of microgel assembly for drug delivery

This injectable and thermo-responsive microgel assembly can find good use in controlled drug delivery. Doxorubicin (DOX), an anticancer drug, was used as a model drug to be loaded with the microgel assembly. The amine group of DOX can be utilized to form Schiff base linkages with the aldehyde groups of HA-ALD-MGs, which can facilitate the sustainability of DOX release.34 HA-ALD-MGs were incubated with DOX at 45°C for 72 h to generate the conjugation of imine bonds. The loading efficiency of DOX onto HA-ALD-MGs was 52:79%±4:85%, which is close to the reported literature.35 To fabricate the DOX-loaded granular hydrogels, 2.5% of the DOX-loaded microgels, 2.5% of the unloaded microgels, and 2.5% of PNAM were blended (Figure 4A). Since DOX has occupied the aldehyde groups of HA-ALD-MGs, unreacted HA-ALD-MGs were incorporated to crosslink with PNAM. Despite reduced hemithioacetal bonds, the combination of dynamic crosslinking and hydrophobic interaction was still able to stabilize the DOX-loaded granular hydrogels. The DOX-loaded granular hydrogels were shear-shinning and had higher storage modulus and yield strain than HA-ALD-MGs alone, indicating the successful formation of thiol-aldehyde addition (Figure S6). Due to the reversibility of Schiff base linkages, environmental pH value, amino acid, or thiol ligand could promote the disassociation of DOX and HA-ALD-MGs. Therefore, the release profiles indicated that the DOX-loaded microgel assembly with PNAM was pH- and redox-responsive. The acidic environment and elevated levels of GSH in tumors can be used as stimuli to precisely control drug release from injected hydrogels. The restricted DOX release in the physiological environment can mitigate the damage to healthy tissue surrounding the tumor. The accumulative released DOX at pH 5.0 was slightly higher than pH 7.0, and both reached plateaus around 10 days. Moreover, the accumulative released DOX of the microgel assemblies in 10 mM GSH approached almost 100% after 20 days. The results indicated that free thiol groups presented in the environment have better effects on degrading the microgel assemblies and triggering drug release because of exchanging of Schiff base linkages, thiol-aldehyde addition, and disulfide bonds.

FIGURE 4.

FIGURE 4

Open in a new tab

Injectable granular hydrogels consisting of microgel assembly for controlled drug release. (A) DOX was used as a model drug and covalently loaded to aldehyde-modified microgels (HA-ALD-MGs) by Schiff base reaction. 2.5% DOX-loaded HA-ALD-MGs, 2.5% HA-ALD-MGs, and 2.5% PNAM were mixed to form the DOX-loaded microgel assembly. (B) The release profile demonstrated that the DOX-loaded microgel assemblies are pH and redox responsive because of the formation of imine and disulfide bonds. (C) Live/dead assays of MCF-7 cells co-cultured with the microgel assembly (−DOX), DOX-loaded microgel assembly (+DOX; Control: TCPS, Calcein: live cells, EthD-1: dead cells). (D) Alarma blue assays of the MCF-7 cells cultured with the extracted solutions of −DOX and +DOX. All viability values were normalized to the control (Control: TCPS; n = 3, mean ± SD). (E) Live/dead assays were quantified by flow cytometry (n = 3, mean ± SD). Statistics analysis was performed using two-way ANOVA with *p < .05, **p < .01, ***p < .001, and ****p < .0001.

Breast cancer cells (MCF-7) co-cultured with the DOX-loaded microgel assemblies with (+DOX) and without (−DOX) DOX displayed that the released DOX had a good anticancer effect (Figure 4C). The −DOX group had cell viability as good as the negative control, but the +DOX group found began to cause cell death on day 3 and had less than 30% metabolic activity compared to the control group (Figure 4D). In live/dead assay and flow cytometry test, the same results were obtained that the negative control and –DOX groups had no adverse effect on the cell viability, but more than half of MCF-7 cells began to die after 48 h of incubation with the DOX-loaded microgel assemblies (Figure 4E). These results demonstrated the use of the microgel assembly for stimuli-responsive therapeutic delivery. Moreover, it also demonstrated that individual microgels could act in different roles in a complex microgel assembly.

3.5 |. Granular hydrogels consisting of microgel assemblies for cell encapsulation and 3D printing

The hydrogel inks for 3D extrusion-based printing should equip good injectability, shear-thinning, and self-healing ability. Earlier we designed a range of hydrogel inks with shear-thinning and self-healing characteristics.36–39 Based on similar approaches, we evaluated the use of granular hydrogels containing microgel assemblies for potential 3D printing. The microgel assemblies were able to form hydrogel filaments with 0.4 and 1 mm extrusion tips (Figure 5A). The uniformity, assessed by the size ratios of the printed line and 3D model, are 0.95 ± 0.13 and 2.19 ± 0.45 with 1 and 0.4 mm tips, respectively. Due to the clogging with the 0.4 mm tips, the microgel assemblies had better printability with 1 mm tips, demonstrated by the printed grid and snake line structures. Also, the PNAM microgel assembly was able to be printed as a cylinder with 50 layers and 1 cm height (Figure 5B). Moreover, the printed scaffold needed no UV light or introduction of crosslinkers for postinjection stability. After printing, the cylinder was then incubated at 37°C to make the phase transition of PNAM to reinforce the microgel assembly because of the generation of hydrophobic aggregation and disulfide bonds. This secondary crosslinking process is convenient and can be homogenously applied to the whole scaffold. The reinforced scaffold was able to hold liquid without leaking. The scaffold was also compressive and elastic, which could recover its shape after being pressed. To apply the printed microgel assemblies to tissue engineering, adhesive peptides (RGD) were conjugated to PNAM by Michael addition. The fibroblast seeded on the surface of microgel assembles with or without RGD showed that adhesive peptides could increase cell metabolic activity and promote cell adhesion (Figure 5C). Moreover, the microgel assembly with RGD conjugation could also be used for cell delivery, in which the cell viability of encapsulated 3T3 cells remained above 95% 30 min after injection (Figure S7). Next, 3D cell encapsulation of fibroblast in the PNAM microgel assemblies was performed. 1% GelMA was incorporated into 3.5% HA-ALD-MG to enhance cell adhesive ability, which showed no significant change in rheological properties (Figure S8). Live/dead assays on both day 1 and day 7 showed good cell viability, indicating the microgel assemblies can be used for cell encapsulation and maintain stability without introducing additional initiators or crosslinkers (Figure 5D).

FIGURE 5.

FIGURE 5

Open in a new tab

The applications of microgel assemblies in 3D printing and cell encapsulation. (A) 3D printing of the microgel assemblies extruded by a tapered extruder with 0.41 and 1 mm diameter. The snake line structures were used for assessing uniformity, and the printed grids showed that the microgel assemblies have better printability with a 1 mm tip (scale bar: 1 cm). (B) The printed microgel assembly was further reinforced by incubation at 37°C overnight. The printed cylinder was compressive and elastic. (C) The modification of cell adhesive peptide (RGD) to PNAM improved cell adhesion and metabolic activities of 3T3 fibroblasts seeded on the microgel assemblies (n = 3, mean ± SD). Cell morphology of the 3T3 fibroblasts seeded on the RGD-modified microgel assemblies (blue: nucleus, purple: actin). (D) Live/dead assays of 3T3 fibroblasts encapsulated in the microgel assemblies on days 1 and 7 (n = 4, mean ± SD). The results showed good cell viability on both days. (Calcein: live cells, EthD-1: dead cells) Statistics analysis was performed using one-way or two-way ANOVA with *p < .05, **p < .01, ***p < .001, and ****p < .0001.

4 |. CONCLUSION

The granular hydrogels, composed of the microgel assembly, were successfully programmed for postinjection stability using thiol-aldehyde addition and the phase-transition behavior of the thermo-responsive polymers. These granular hydrogels exhibited good shear-thinning and self-healing ability due to the dynamic covalent bonds of hemithiolactal at room temperature. Additionally, the granular hydrogels demonstrated improved mechanical stability of yield stress and modulus due to the formation of hydrophobic interaction and disulfide bonds at body temperature. This makes them suitable for a variety of biomedical applications, including therapeutics delivery, cell delivery, cell encapsulation, and 3D printing. Covalent loading of DOX to the aldehyde-modified microgels was achieved via Schiff base reaction, allowing the DOX-microgel assembly to exhibit pH and redox-responsive release profiles. The 3D printed granular hydrogels showed good fidelity and could be stabilized with simple incubation at 37°C. Overall, this study presented the first injectable and thermo-responsive granular hydrogels with sequentially programmed properties, which have the potential for a range of biomedical applications.

Supplementary Material

Supplementary Information

Additional supporting information can be found online in the Supporting Information section at the end of this article.

ACKNOWLEDGMENTS

Akhilesh K Gaharwar acknowledges financial support from the National Institute of Biomedical Imaging and Bioengineering (NIBIB) of NIH, the Director’s New Innovator award (DP2 EB026265), National Institute of Neurological Disorders and Stroke (R21 NS121945), Peer Reviewed Medical Research Program (PRMRP) of Department of Defense (DOD) (W81XWH2210932), President’s Excellence Fund (X-Grants) from Texas A&M University. Tanmay P. Lele acknowledges financial support from the CPRIT-established investigator award grant # RR200043. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies. Some of the schematics in the article were created using Biorender.

Funding information

Cancer Prevention and Research Institute of Texas, Grant/Award Number: RR200043; National Institute of Biomedical Imaging and Bioengineering, Grant/Award Number: DP2EB026265; National Institute of Neurological Disorders and Stroke, Grant/Award Number: R21NS121945; U.S. Department of Defense, Grant/Award Number: W81XWH2210932

DATA AVAILABILITY STATEMENT

All data is freely available over public repository.

REFERENCES

  • 1.Daly AC, Riley L, Segura T, Burdick JA. Hydrogel microparticles for biomedical applications. Nat Rev Mater. 2020;5(1):20–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Li J, Mooney DJ. Designing hydrogels for controlled drug delivery. Nat Rev Mater. 2016;1(12):1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Zhang YS, Khademhosseini A. Advances in engineering hydrogels. Science. 2017;356(6337):eaaf3627. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Gaharwar A, Singh I, Khademhosseini A. Engineered biomaterials for In situ tissue regeneration. Nat Rev Mater. 2020;5:686–705. [Google Scholar]
  • 5.Gaharwar AK, Avery RK, Assmann A, et al. Shear-thinning nanocomposite hydrogels for the treatment of hemorrhage. ACS Nano. 2014; 8(10):9833–9842. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Thakur A, Jaiswal MK, Peak CW, et al. Injectable shear-thinning nanoengineered hydrogels for stem cell delivery. Nanoscale. 2016; 8(24):12362–12372. [DOI] [PubMed] [Google Scholar]
  • 7.Meng L, Shao C, Cui C, Xu F, Lei J, Yang J. Autonomous self-healing silk fibroin injectable hydrogels formed via surfactant-free hydrophobic association. ACS Appl Mater Interfaces. 2019;12(1):1628–1639. [DOI] [PubMed] [Google Scholar]
  • 8.Shuai F, Zhang Y, Yin Y, Zhao H, Han X. Fabrication of an injectable iron (III) crosslinked alginate-hyaluronic acid hydrogel with shear-thinning and antimicrobial activities. Carbohydr Polym. 2021;260: 117777. [DOI] [PubMed] [Google Scholar]
  • 9.Rizwan M, Baker AE, Shoichet MS. Designing hydrogels for 3D cell culture using dynamic covalent crosslinking. Adv Healthc Mater. 2021; 10(12):2100234. [DOI] [PubMed] [Google Scholar]
  • 10.Chakma P, Konkolewicz D. Dynamic covalent bonds in polymeric materials. Angew Chem. 2019;131(29):9784–9797. [DOI] [PubMed] [Google Scholar]
  • 11.Rodell CB, MacArthur JW Jr, Dorsey SM, et al. Shear-thinning supramolecular hydrogels with secondary autonomous covalent crosslinking to modulate viscoelastic properties in vivo. Adv Funct Mater. 2015;25(4):636–644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Wang H, Zhu D, Paul A, et al. Covalently adaptable elastin-like protein–hyaluronic acid (ELP–HA) hybrid hydrogels with secondary thermoresponsive crosslinking for injectable stem cell delivery. Adv Funct Mater. 2017;27(28):1605609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Feng Q, Li D, Li Q, Cao X, Dong H. Microgel assembly: fabrication, characteristics and application in tissue engineering and regenerative medicine. Bioact Mater. 2022;9:105–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Feng Q, Li D, Li Q, et al. Dynamic nanocomposite microgel assembly with microporosity, injectability, tissue-adhesion, and sustained drug release promotes articular cartilage repair and regeneration. Adv Healthc Mater. 2022;11(8):2102395. [DOI] [PubMed] [Google Scholar]
  • 15.Muir VG, Weintraub S, Dhand AP, Fallahi H, Han L, Burdick JA. Influence of microgel and interstitial matrix compositions on granular hydrogel composite properties. Adv Science. 2023;10:2206117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Highley CB, Song KH, Daly AC, Burdick JA. Jammed microgel inks for 3D printing applications. Adv Sci. 2019;6(1):1801076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Chen MH, Chung JJ, Mealy JE, et al. Injectable supramolecular hydrogel/microgel composites for therapeutic delivery. Macromol Biosci. 2019;19(1):1800248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Edwards SD, Hou S, Brown JM, et al. Fast-curing injectable microporous hydrogel for In situ cell encapsulation. ACS Appl Bio Mater. 2022; 5:2786–2794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Griffin DR, Weaver WM, Scumpia PO, Di Carlo D, Segura T. Accelerated wound healing by injectable microporous gel scaffolds assembled from annealed building blocks. Nat Mater. 2015;14(7):737–744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hou S, Lake R, Park S, Edwards S, Jones C, Jeong KJ. Injectable macroporous hydrogel formed by enzymatic cross-linking of gelatin microgels. ACS Appl Bio Mater. 2018;1(5):1430–1439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lou J, Liu F, Lindsay CD, Chaudhuri O, Heilshorn SC, Xia Y. Dynamic hyaluronan hydrogels with temporally modulated high injectability and stability using a biocompatible catalyst. Adv Mater. 2018;30(22): 1705215. [DOI] [PubMed] [Google Scholar]
  • 22.Hua Y, Gan Y, Zhang Y, et al. Adaptable to mechanically stable hydrogels based on the dynamic covalent cross-linking of thiol-aldehyde addition. ACS Macro Lett. 2019;8(3):310–314. [DOI] [PubMed] [Google Scholar]
  • 23.Hua Y, Gan Y, Li P, et al. Moldable and removable wound dressing based on dynamic covalent cross-linking of thiol-aldehyde addition. ACS Biomater Sci Eng. 2019;5(8):4048–4053. [DOI] [PubMed] [Google Scholar]
  • 24.Lee HP, Lokhande G, Singh KA, Jaiswal MK, Rajput S, Gaharwar AK. Light-triggered in situ gelation of hydrogels using 2D molybdenum disulfide (MoS2) Nanoassemblies as crosslink epicenter. Adv Mater. 2021;33(23):2101238. [DOI] [PubMed] [Google Scholar]
  • 25.Muir VG, Qazi TH, Shan J, Groll Jr, Burdick JA. Influence of microgel fabrication technique on granular hydrogel properties. ACS Biomater Sci Eng. 2021;7(9):4269–4281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Chimene D, Deo KA, Thomas J, Dahle L, Mandrona C, Gaharwar AK. Designing cost-effective open-source multihead 3D bioprinters. GEN Biotechnol. 2022;1(4):386–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Li S, Pei M, Wan T, et al. Self-healing hyaluronic acid hydrogels based on dynamic Schiff base linkages as biomaterials. Carbohydr Polym. 2020;250:116922. [DOI] [PubMed] [Google Scholar]
  • 28.Lou J, Mooney DJ. Chemical strategies to engineer hydrogels for cell culture. Nat Rev Chem. 2022;6:726–744. [DOI] [PubMed] [Google Scholar]
  • 29.Zhang Y, Li X, Bai G, Wei W, Liu X. Hyperbranched polymer with dynamic thiol–aldehyde crosslinking and its application as a self-healable bioadhesive. J Mater Chem B. 2021;9(29):5818–5828. [DOI] [PubMed] [Google Scholar]
  • 30.Carnicer V, Alcázar C, Orts M, Sánchez E, Moreno R. Microfluidic rheology: a new approach to measure viscosity of ceramic suspensions at extremely high shear rates. Open Ceram. 2021;5:100052. [Google Scholar]
  • 31.Feng Q, Li D, Li Q, et al. Assembling microgels via dynamic crosslinking reaction improves printability, microporosity, tissue-adhesion, and self-healing of microgel bioink for extrusion bioprinting. ACS Appl Mater Interfaces. 2022;14(13):15653–15666. [DOI] [PubMed] [Google Scholar]
  • 32.Chen MH, Wang LL, Chung JJ, Kim Y-H, Atluri P, Burdick JA. Methods to assess shear-thinning hydrogels for application as injectable biomaterials. ACS Biomater Sci Eng. 2017;3(12):3146–3160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Grosskopf AK, Roth GA, Smith AA, Gale EC, Hernandez HL, Appel EA. Injectable supramolecular polymer–nanoparticle hydrogels enhance human mesenchymal stem cell delivery. Bioeng Transl Med. 2020;5(1):e10147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Feng X, Li D, Han J, Zhuang X, Ding J. Schiff base bond-linked polysaccharide–doxorubicin conjugate for upregulated cancer therapy. Mater Sci Eng C. 2017;76:1121–1128. [DOI] [PubMed] [Google Scholar]
  • 35.Song J, Xu B, Yao H, et al. Schiff-linked PEGylated doxorubicin pro-drug forming pH-responsive nanoparticles with high drug loading and effective anticancer therapy. Front Oncol. 2021;11:656717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Deo KA, Jaiswal MK, Abasi S, et al. Nanoengineered ink for designing 3D printable flexible bioelectronics. ACS Nano. 2022;16(6):8798–8811. [DOI] [PubMed] [Google Scholar]
  • 37.Gold KA, Saha B, Rajeeva Pandian NK, et al. 3D bioprinted multicellular vascular models. Adv Healthc Mater. 2021;10(21):2101141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nava-Medina IB, Gold KA, Cooper SM, et al. Self-oscillating 3D printed hydrogel shapes. Adv Mater Technol. 2021;6(12):2100418. [Google Scholar]
  • 39.Peak CW, Singh KA, Adlouni M, Chen J, Gaharwar AK. Printing therapeutic proteins in 3D using nanoengineered bioink to control and direct cell migration. Adv Healthc Mater. 2019;8(11):1801553. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
NIHMS2142604-supplement-Supplementary_Information.docx (5.1MB, docx)

Data Availability Statement

All data is freely available over public repository.

RESOURCES