Molecular weight-driven tunable hyaluronic acid-based hydrogels modulate immune polarization in three-dimensional microenvironments

Abstract

Macrophages exhibit phenotypic plasticity that is strongly influenced by their surrounding microenvironment, including extracellular matrix (ECM) components. Hyaluronic acid (HA), a major glycosaminoglycan in ECM, has immunomodulatory effects that are highly dependent on its molecular weight (MW). However, most previous studies have been limited to two-dimensional (2D) culture systems, which were unable to accurately replicate the in vivo environment. In this study, we utilized a three-dimensional (3D) culture system based on HA-based hydrogels to better understand the MW-dependent immunomodulatory effects of HA on macrophages under more physiologically relevant conditions. Three different MWs of HA were chemically modified and crosslinked with PEG-SH₄ to form hydrogels with distinct biophysical properties. Immortalized macrophages were encapsulated within these hydrogels and assessed for the expression of both pro-inflammatory and anti-inflammatory markers. Notably, hydrogels with high-MW HA significantly upregulated the expression of anti-inflammatory markers, indicating that the immunomodulatory effects of HA in 3D culture are affected by its biophysical characteristics. Our findings demonstrate the potential of HA-based hydrogels as customizable ECM-mimetic scaffolds for modulating immune responses in regenerative medicine applications.

Graphical Abstract

1. Introduction

The immune system enables an organism to distinguish between “self” and “non-self” and respond to potential threats such as pathogens.¹ Immune cells play a vital role in orchestrating host defense and maintaining homeostasis.² Upon detecting harmful stimuli, such as viruses or bacteria, immune cells respond by targeting infected tissues through inflammatory responses.³ Among immune cells, macrophages play a key role by detecting pathogens, clearing cellular debris, and coordinating tissue repair.⁴ They also function as antigen-presenting cells to activate adaptive immunity.^{5, 6} Macrophages are exposed to various microenvironments and differentiate into pro-inflammatory or anti-inflammatory phenotypes in response to various signals from the surrounding microenvironment.^{7, 8} These phenotypic changes are primarily regulated by multiple microenvironmental factors, including chemokines, cytokines, and other signaling molecules.^9–11 Pro-inflammatory phenotypic macrophages are involved in pathogen clearance during infection and early inflammatory responses, while anti-inflammatory phenotypic macrophages play a role in promoting inflammation resolution, tissue repair, and angiogenesis.^12–14 The physical properties of the extracellular matrix (ECM) can also influence macrophages as a microenvironment.^{15, 16} The interaction between macrophages and the ECM critically regulates immune responses, cell growth, survival, and differentiation.¹⁷ These interactions are essential for modulating inflammation and promoting tissue remodeling.

Hyaluronic acid (HA) is a major glycosaminoglycan found abundantly in ECM, particularly in inflamed tissues.^18–21 It possesses excellent physicochemical properties, including biocompatibility, low immunogenicity, and tunable mechanical characteristics.^{20, 22, 23} Beyond its structural role, HA also exhibits immunomodulatory effects that are highly dependent on its molecular weight.^{24, 25} Low molecular weight HA (LMW-HA) tends to promote inflammation by activating pro-inflammatory pathways, often through interactions with pattern recognition receptors such as Toll-like receptors (TLR) 2 and 4.^{26, 27} In contrast, high molecular weight HA (HMW-HA) has anti-inflammatory properties, primarily through its interaction with CD44.^{28, 29} CD44 converts ECM-derived signals, including those from HA, into intracellular signals that regulate cell growth, survival, activation, and differentiation.^{30, 31} HMW-HA and CD44 together suppress pro-inflammatory TLR signaling at multiple levels and promote the expression of anti-inflammatory cytokines.²⁸ In immune cells, particularly macrophages, HA-CD44 binding can directly modulate immune function by reducing pro-inflammatory cytokine production and enhancing the expression of anti-inflammatory genes such as Arg1, IL-10, and MRC1.^{32, 33}

Despite extensive research on the immunomodulatory roles of HA based on its molecular weight, most studies have been conducted using two-dimensional (2D) cell culture systems.^{32, 33} While 2D cultures are useful for studying biochemical mechanisms such as cell proliferation, differentiation, and cell–cell interactions, they differ significantly from in vivo environments in both physical and chemical properties.^{34, 35} As a result, cells grown in 2D often behave quite differently from those in living tissues. To overcome these limitations, three-dimensional (3D) cell culture systems using various biocompatible materials have emerged as promising alternatives, offering more physiologically relevant conditions that better simulate the in vivo microenvironment.^{36, 37} These 3D systems provide a more realistic physiological context, allowing for more accurate observation of cellular behavior and complex biological processes.³⁴

Hydrogels have emerged as effective platforms for mimicking the in vivo environment in 3D cell culture systems.^{35, 38, 39} Hydrogels are water-swollen polymer networks and can be made from natural or synthetic materials.⁴⁰ They can offer a tunable microenvironment that supports cell viability, migration, and differentiation. Their ability to replicate the physical and biochemical properties of the ECM makes hydrogels excellent synthetic ECM analogs for 3D cultures.^{41, 42} Moreover, their porous structures enable the encapsulation and controlled release of bioactive molecules, allowing for precise modulation of cellular behavior in 3D environments.^{43, 44} Due to these properties, hydrogel-based materials are considered versatile 3D matrices suitable for applications such as biomedical devices, drug delivery, stem cell engineering, and tissue engineering.⁴⁵

In this study, we investigated the immunomodulatory effects of HA with different molecular weights on phenotypically distinct immortalized macrophages under 3D culture conditions. To this end, three types of HA-based hydrogels were fabricated using low (60 kDa), medium (200 kDa), and high (1,500 kDa) molecular weight HA. We characterized the physical properties of the hydrogels, encapsulated the cells within them, and evaluated macrophage polarization in response to different HA environments.

2. Materials and Methods

2.1. Materials

HA (MW: 60, 200, and 1,500 kDa) was purchased from Lifecore Biomedical (Chaska, MN). 1-hydroxybenzotriazole hydrate (HOBT) was purchased from Chem-Impex International, Inc. (Wood Dale, IL). 1-Ethyl-3-(3-dimethylaminopropyl) carbodimide (EDC), adipic acid dihydrazide (ADH), triethanolamine (TEA), formamide, and hyaluronidase were acquired from Sigma-Aldrich (St. Louis, MO). Polyethylene glycol tetra-thiol (PEG-SH₄) (MW: 10 kDa) was purchased from Advanced Biochemicals (Lawrenceville, GA). N-acryloxysuccinimide (NAS) was purchased from Thermo Fisher Scientific (Waltham, MA).

For the cell experiment, Dulbecco’s modification of Eagle’s medium (DMEM) was purchased from Corning Inc. (Corning, NY). Fetal bovine serum (FBS), penicillin/streptomycin, and phosphate-buffered saline (PBS) were purchased from Gibco BRL (Grand Island, NY).

2.2. Synthesis of acrylated HA and preparation of hydrogel

Acrylated HA (HA-Ac) was synthesized as previously described.²¹ HA (0.5 mmol) was dissolved in 40 ml of DW, and EDC (0.48 g, 2.5 mmol), HOBT (0.34 g, 2.52 mmol), and ADH (2.2g, 12.6 mmol) were added to the solution. The solution reacted overnight at 37 °C, then dialyzed against 100mM NaCl for 1 day and DW for 1 day, using a dialysis membrane (MWCO 3.5 kD, SpectraPor; Rancho Dominguez, CA). NAS (0.5 g, 3 mmol) was subsequently added to the solution and reacted overnight at 37 °C. Then the solution was dialyzed against 100 mM NaCl for 1 day and DW for 1 day. The product was then lyophilized for 3 days to obtain solid HA-Ac. ¹H NMR 600 MHz spectrometer (Brucker) was used to obtain the degree of acrylation, and it was calculated by comparing peaks from the acryl and methyl groups from the HA residue.

For hydrogel preparation, HA-Ac was dissolved in a 0.3M TEA-buffered solution with pH 8.0. PEG-SH₄ was added as a cross-linker with the same molar ratio of acryl and thiol groups. The solutions were incubated at 37 °C to form gels. The HA-based hydrogels were formed via the Michael-type addition reaction.⁴⁶

2.3. Characterization of hydrogels

Gelation time was evaluated by the inverted tube test. It was performed in a water bath at 37 °C, and the physical state (viscous liquid or gel) of the tested samples was noted by turning a test tube.^{47, 48} The microstructure of the hydrogels was examined using FEI Quanta 250 field-emission scanning electron microscopy (FE-SEM). For SEM imaging, hydrogels were prepared and frozen at −80 °C. Hydrogels were then lyophilized and carefully cut into small pieces. Samples were then imaged using SEM, and pore size (diameter) was manually measured using ImageJ.

In order to measure the swelling properties of the hydrogel, it was incubated in PBS overnight at room temperature. The swelling ratio was measured by comparing the change in the wet weight of the hydrogel before and after incubation. The percentage of water absorbed was calculated by the following eq 1.

$$\begin{array}{c}\mathit{Swelling\; ratio}(\%)=\left\{\right(W_w-W_i)/W_i\}*100\end{array}$$ (1)

Where W_w is the wet weight of the hydrogel, W_i is the initial weight of the hydrogel. Degradation of the hydrogels is monitored by measuring the weight loss of the hydrogel. The pre-swollen hydrogels were added to the PBS and incubated at 37 °C. In addition, to measure the degradation by hyaluronidase, an enzyme that degrades HA in vivo environment,⁴⁹ hyaluronidase was added to PBS to a final concentration of 50 U/ml.

2.4. Measurement of viscoelastic properties of HA-based hydrogels with different molecular weights

To characterize the time-dependent viscoelastic properties of the hydrogels, stress-relaxation tests were conducted using a custom-built, indentation-based mechanical analysis system that has been previously validated with materials of well-defined mechanical properties, including PDMS.^{50, 51} This time-dependent indentation approach enables the extraction of representative viscoelastic parameters, such as elastic modulus, viscosity, and relaxation time, thereby facilitating direct comparisons across experimental groups.^52–55 The system was equipped with a rigid, flat-ended cylindrical indenter to ensure controlled deformation. Each sample was positioned beneath the indenter, which was gradually lowered until contact with the sample surface was detected and an initial force was recorded. The indenter was then slightly retracted and zeroed to establish a consistent baseline prior to testing. Samples were subsequently compressed by 0.4 mm over 30 s, and the resulting force relaxation response was recorded at a sampling rate of 25 Hz using a high-resolution force sensor (PASPORT High Resolution Force Sensor, Pasco Scientific, USA) connected to a Pasco 550 Universal Interface (Pasco Scientific, USA). All measurements were performed in triplicate under ambient laboratory conditions (approximately 20–25 °C) in a dry environment.

After the measuring forces, a theoretical model to convert the measured forces to the elastic moduli was used as following eq 2.

$$\begin{array}{c}F=\frac{2\mathit{ER\delta }}{1-{\nu }^2}\end{array}$$ (2)

where $\text{E}$ is the elastic modulus of the material, δ is the indentation depth, $\text{R}$ is the radius of the force sensor, and $\nu$ is the Poisson ratio of the material ($\nu$). The Poisson ratio was set to be 0.5, assuming that the samples have isotropic and homogeneous material properties. The elastic moduli were converted into the shear moduli using the following eq 3.

$$\begin{array}{c}E=2G\left(1+\nu \right)\end{array}$$ (3)

where G is the shear modulus.

To determine the viscoelastic properties (i.e., instantaneous shear and elastic moduli, and equivalent viscosity), the estimated shear moduli were fitted to the experimental data using a nonlinear curve fit based on the Prony series, as following eq 4.

$$\begin{array}{c}G\left(t\right)=G_{\mathrm{\infty }}+\sum _{i=1}^N{G}_i{e}^{-\frac{t}{{\tau }_i}}\end{array}$$ (4)

where $G_{\mathrm{\infty }}$ is the long-term equilibrium shear modulus of the material at steady-state conditions, ${\text{G}}_{\text{i}}$ are the shear moduli representing the elastic response at a specific relaxation time, and ${\tau }_{\text{i}}$ are the relaxation times. In this study, we used a two-term Prony series, with N = 2. The three representative shear moduli ($G_{\mathrm{\infty }},G_1$ and $G_2$) and two relaxation times (${\tau }_1$ and ${\tau }_2$) are used to estimate the instantaneous shear modulus, the instantaneous elastic modulus, and the equivalent viscosity. The instantaneous shear and elastic modulus of the sample can be determined by eqs 5 and 6.

$$\begin{array}{c}{G}_{\mathit{instant}}=G_{\mathrm{\infty }}+G_1+G_2\end{array}$$ (5)

where ${\text{G}}_1$ and ${\text{G}}_2$ are the shear moduli corresponding to the relaxation times, ${\tau }_1$ and ${\tau }_2$, respectively.

$$\begin{array}{c}{E}_{\mathit{instant}}=2G_{\mathit{instant}}\left(1+\nu \right)\end{array}$$ (6)

The equivalent viscosity of the sample, ${\mu }_{\text{eq}}$, was calculated by eq 7.

$$\begin{array}{c}{\mu }_{\mathit{eq}}=2\left(1+\nu \right)\left(\frac{{\left(G_1+G_2\right)}^2}{\frac{G_1}{{\tau }_1}+\frac{G_2}{{\tau }_2}}\right)\end{array}$$ (7)

2.5. Cell culture and Treatments

Murine Macrophages (RAW 264.7) were cultured in DMEM with 10% (v/v) FBS and 1% (v/v) antibiotics, containing penicillin and streptomycin in a humidified incubator (5% CO², 37 °C). Macrophages were harvested by scraping and used between passages 7 and 10. Macrophages were plated at a density of 10⁶ cells/ml in a T75 culture flask and allowed to adhere overnight. To make three different states of macrophages, macrophages were cultured for 1 day in media (1) with no activating agent, (2) with 100 ng/ml LPS (Sigma) and 10 ng/ml IFN-γ (BioLegend), or (3) with 20 ng/ml IL-4 (Peprotech). The media was changed to non-activating agent media after 1 day.

2.6. Cytotoxicity and proliferation in vitro

The hydrogels were mixed with cell suspensions (10⁵ cells/construct) and produced a final volume of 10 µl in 96-well plates. The hydrogels with cells were cross-linked for 10 min at 37 °C and were cultured as the above method. The viability of cells in the hydrogels was measured by the LIVE/DEAD™ Viability/Cytotoxicity Kit (Invitrogen, Carlsbad, CA). Several samples of live and dead cells were counted under a Nikon AXR inverted confocal microscope, and only the cell images that focused on the hydrogel were counted. The ratio of live cells in the hydrogel was determined by dividing the number of live cells by the number of total cells. Cell proliferation was assessed using the CCK-8 assay. Fresh medium containing 10% (v/v) of CCK-8 solution was added to each well, and the plate was incubated for an additional 3 h. The absorbance of each well was read on an Epoch plate reader (BioTek) at a wavelength of 450 nm.

2.7. Immunofluorescence staining

For 2D and 3D immunofluorescence staining, samples were washed with PBS and fixed in 4% paraformaldehyde, followed by permeabilization with Triton-X and blocking with 1% bovine serum albumin solution at room temperature. Then they were stained with anti-CD86 antibody and anti-CD206 at 4 °C and subsequently stained with secondary antibodies at room temperature. The detailed information about primary and secondary antibodies was provided in Table S1. Samples were washed with PBS and incubated with Hoechst 33342 to stain nuclei. Samples were imaged with a Nikon AXR inverted confocal microscope and analyzed using ImageJ.

2.8. Quantitative real-time PCR

The hydrogels were mixed with cell suspensions (5 × 10⁵ cells/construct) and produced a final volume of 50 µl in 24-well plates. The hydrogels with cells were cross-linked for 10 min at 37 °C and were cultured as the method above. After culturing for 3 days, the cell-containing hydrogels were homogenized with a tissue homogenizer, and RNA was extracted using the RNeasy Mini Kit (Qiagen, Germany) according to the manufacturer’s instructions. The concentration of RNA was measured using an Epoch plate reader (BioTek), and cDNA was synthesized using the iScript™ cDNA Synthesis kit (Bio-Rad, Hercules, CA) using the 96-well T100 thermal cycler (Bio-Rad, Hercules, CA). Primers were designed for RT-qPCR (Table S2),^56–59 and the RT-qPCR products were assessed using the accumulation level of iQ™ SYBR Green Supermix (Bio-Rad) fluorescence following a manufacturer’s protocol on CFX Connect™ Real-Time PCR Detection System (Bio-Rad). The gene expression level was normalized by the expression of GAPDH, and differences in gene expression were presented as fold ratios from the control group. The relative expression level was calculated as X = 2^–ΔΔCt, where ΔΔC_t = ΔE−ΔC_t and ΔE = C_t,exp – C_t,GAPDH, ΔC_t = C_t,control – C_t,GAPDH.⁶⁰

2.9. Statistical analysis

We performed a one or two-way ANOVA and Tukey’s post hoc test for the multiple comparisons, and GraphPad Prism 9 (La Jolla, CA) was used to perform statistical analyses. Data are shown as mean ± standard error of the mean, and differences of p < 0.05 were considered statistically significant.

3. Results and Discussion

3.1. Characterization of HA-based hydrogel by different MWs

We synthesized HA-Ac in a two-step reaction using three different MWs (Fig. 1A). The distinct peaks of the acryl group appear at 5.9 and 6.3 ppm, indicating that the acryl group was successfully conjugated to HA. The substitution degree of the acryl group was calculated to be about 13% through peak integral (Fig. S1). HA-based hydrogels of three different MWs were prepared using PEG-SH₄ as a crosslinker (Fig. 1B). These hydrogels are fabricated via Michael-type addition, a reaction that proceeds rapidly under mild physiological conditions (pH 8 and 37 °C) without harsh external catalysts.^{61, 62} This method allows for precise control over gelation time, network structure, and mechanical properties.^{61, 63, 64} Compared to other hydrogel fabrication methods that utilize UV light or catalysts, Michael-type addition minimizes byproducts and avoids cytotoxicity associated with photoinitiators and UV exposure.^{65, 66} This is particularly crucial for biomedical applications involving cells and sensitive molecules. Furthermore, by conducting the reaction at 37 °C, similar to body temperature, gelation after injection can be achieved under comparable conditions in vivo tests.

Figure 1.

A) Synthetic scheme of conjugating acryl group to the hyaluronic acid, and B) schematic of hydrogel preparation and representative photograph of hydrogel in the gel state. Created in BioRender. Nano, B. (2025) https://BioRender.com/8do5xhv.

Rapid gelation of hydrogels upon injection is known to positively influence scaffold longevity by quickly stabilizing the structure.^{67, 68} However, excessively fast gelation can lead to premature solidification of the hydrogel before it reaches the target site, thereby hindering successful injection.⁶⁹ The ideal gelation time is not universal and depends significantly on the specific clinical scenario and application site. In a surgical environment, a gelation time of approximately 5 to 30 minutes is suggested as optimal.⁷⁰ In bone tissue engineering, hydrogels with a gelation time within 10 minutes are considered effective, balancing injectability with structural stability for bone regeneration.^{69, 71} Similar to bone tissue engineering, hydrogels for nerve regeneration also have gelation times ranging from 5 to 10 minutes, providing an appropriate mechanical microenvironment for nerve tissue repair.^72–74

The gelation time of 2% HA-1500 was approximately 7.5 minutes. At high concentrations, HA-1500 exhibited poor solubility, while at lower concentrations, the gelation time exceeded 10 minutes (Fig. S2). To standardize the gelation kinetics across different hydrogel formulations, the concentrations were adjusted to align with the gelation window of 2% HA-1500 (~ 7–8 minutes). As a result, 2% HA-1500, 4% HA-200, and 8% HA-60 of concentration were selected, as all exhibited desirable gelation times within 10 minutes (Fig. 2A). Our findings demonstrate that gelation time can be manipulated by varying the concentration and molecular weight of HA.

Figure 2.

Characterization of HA-based hydrogels. A) The gelation times of hydrogels were adjusted to be within 10 minutes for the ideal gelation time. B, C) Microstructure of hydrogels via SEM imaging revealed the average pore size of hydrogels. Scale bar = 100 μm. D) The swelling ratio of HA-60 was 311%, HA-200 was 297.8%, and HA-1500 was 188.1% at the peak point (12 hours). E, F) Hydrogel degraded slowly in PBS, and HA-60 and HA-1500 remained more than 80% intact after 7 days. Hydrogel was degraded more quickly than PBS in the solution with added hyaluronidase. A one-way ANOVA with Tukey’s post hoc test for the multiple comparisons. ** p < 0.01, *** p < 0.001, **** p < 0.0001.

SEM imaging of the hydrogels revealed porous hydrogel microstructure (Fig. 2B), and pore size analysis showed that HA-1500 had a significantly larger pore size of 39.69 ± 3.35 μm compared to HA-60 and HA-200 (12.25 ± 0.45 μm and 27.22 ± 3.40 μm, respectively) (Fig. 2C). Although freeze-drying may alter the absolute pore size relative to the hydrated state, all samples were processed under identical freezing and lyophilization conditions. Therefore, SEM analysis provides a reliable qualitative and relative comparison of pore morphology among different hydrogel formulations. Future hydrate-state characterization using a confocal microscope will be needed to validate pore architecture under native conditions.

Pore size is a critical design parameter in porous hydrogels, influencing cell infiltration, proliferation, migration, nutrient diffusion, and vascularization across diverse tissue engineering applications.^{75, 76} Small pore size enhances cell seeding efficiency and surface adhesion, while large pore size facilitates cell migration and neovascularization.^{75, 77} However, excessively small pores can hinder nutrient and oxygen diffusion, leading to poor cell viability and impaired differentiation.⁷⁸ Conversely, large pores may reduce cell attachment due to limited surface area and mechanical strength.^{75, 78} Another important consideration in the design of pore size is the size of cells that must infiltrate the structure. For example, immune cells such as monocytes and neutrophils typically range from 12 to 20 µm in diameter.^{79, 80} In neural tissues, the relevant cell sizes vary more widely, from smaller oligodendrocytes (6–9 µm) to much larger astrocytes, which can reach average diameters of approximately 140 µm.⁸¹

The ideal pore size varies depending on the application, taking into account tissue-specific requirements. For skin regeneration, small (< 2 µm), medium (< 12 µm), and large (> 40 µm) pores support epidermal adhesion, dermal infiltration, and vascularization, respectively.⁷⁵ In bone tissue engineering, small pores (50–100 µm) promote osteoblast adhesion, while larger pores (200–400 µm) support angiogenesis and osteogenesis.^{75, 82, 83} Cardiac and vascular tissues benefit from pores in the 25–60 µm range, balancing cell integration and nutrient transport.^{75, 84, 85} Neural applications, such as spinal cord repair, require pores between 22–200 µm to accommodate axonal infiltration and glial support.⁸¹ Ultimately, pore size should be tailored to the target tissue and therapeutic context to achieve functional tissue regeneration. Our HA-based hydrogel demonstrates broad potential for supporting tissue repair across diverse biological environments.

All hydrogels showed rapid swelling within 1 hour and then slightly increased and tended to reach a peak in the following 12 hours. Swelling behavior also varied with MW, with the HA-60 swelling the most (~311%) and HA-1500 swelling the least (~188%) (Fig. 2D). In previous studies, the swelling ratio decreased as the molecular weight increased due to the increase of entanglement of HA chains.^{86, 87} The low swelling ratio of HA-1500 and the high swelling ratio of HA-60 are consistent with the previous findings. In contrast, HA-200 and HA-60 showed no significant differences in swelling behavior, which could be attributed to the higher MW, but lower concentration of HA-200 compared to HA-60. Hydrogels swell when their polymer network absorbs solvent, causing volume expansion.^{88, 89} This swelling process can generate mechanical pressure in the surrounding environment,⁸⁹ which can be transmitted to the surrounding tissues. In particular, the increased intraspinal pressure due to hydrogel swelling in spinal cord injury can cause injury exacerbation.⁹⁰ Therefore, it is important to precisely control the swelling behavior when designing hydrogels for biomedical applications to ensure the desired mechanical function and minimize adverse effects.⁹¹

Hydrogel degraded slowly in PBS and HA-60 and HA-1500 remained more than 80% intact after 7 days (Fig. 2E). In the solution with added hyaluronidase, hydrogel degraded more quickly than PBS indicating a difference depending on the formulation (Fig. 2F). It degraded faster in the hydrogel with a lower concentration, likely due to the smaller amount of HA maintaining the backbone of the hydrogel, leading to quicker degradation by the enzyme. Our data demonstrates that hyaluronidase can recognize the degradation sites of HA, even though HA is modified by the acryl group and is cross-linked in the gel. This degradability allows them to break down over time in the body, allowing them to be replaced by new host tissue as regeneration progresses.⁹² Non-biodegradable hydrogels can remain in the body permanently, causing foreign body reactions or inhibiting the integration with the tissue.^{93, 94} In addition, biodegradable hydrogels offer significant advantages in controlled drug delivery, allowing for the precise release of medications and bioactive compounds.^{95, 96} Therefore, biodegradable hydrogels offer advantages in the fields of tissue engineering and regenerative medicine.

This data on the properties of hydrogels demonstrates that the gelation time, pore size, swelling behavior, and degradation rate can be controlled by adjusting the MW and concentration of HA. Controlled gelation ensures application-specific adaptability. SEM analysis revealed distinct pore sizes suitable for immune cell infiltration and tissue-specific needs. Swelling and degradation profiles were consistent with physiological relevance, showing stability in PBS and enzymatic sensitivity to hyaluronidase.

3.2. Effects of MWs on the viscoelastic properties of HA-based hydrogels

The viscoelastic properties of the HA-based hydrogel, composed of elastic properties (shear and elastic modulus) and viscous properties (viscosity), were evaluated using a stress-relaxation test, as shown in Fig. 3. Understanding the viscoelastic properties of hydrogels is essential for mimicking the in vivo environment and regulating cellular behavior.^{97, 98} Shear modulus and elastic modulus are representative measures of a material’s elastic properties. The shear modulus reflects a material’s resistance to shear deformation, and the elastic modulus, or Young’s modulus, indicates the stiffness of a material under tensile or compressive stress, describing its resistance to deformation.⁹⁹ Both moduli reflect the material’s elastic properties, which play a critical role in maintaining structural integrity when used as scaffolds.^{100, 101}

Figure 3.

Viscoelastic properties of hydrogels. A, B) Shear and elastic moduli were significantly different with hydrogels (HA-60: 3456 and 10368 Pa, HA-200: 1689 and 5068 Pa, HA-1500: 6237 and 18710 Pa). C) Hydrogels showed similar equivalent viscosities (HA-60: 1572 Pa·s, HA-200: 1862 Pa·s, HA-1500: 1105 Pa·s). A one-way ANOVA with Tukey’s post hoc test for the multiple comparisons. **** p < 0.0001.

The three hydrogels showed significant differences in shear and elastic moduli, while equivalent viscosities were comparable. HA-1500 exhibited a higher shear modulus (6.2 kPa) and elastic modulus (18.7 kPa) compared to HA-200 (1.7 kPa and 5.1 kPa for shear and elastic moduli, respectively) and HA-60 (3.5 kPa and 10.4 kPa, respectively) (Fig. 3A–B). The higher elastic properties of HA-1500 compared to the other formulations are likely attributable to its higher molecular weight, which increases network density and the degree of crosslinking within the hydrogel.^{102, 103} HMW-HA forms longer and more entangled polymer chains with increased crosslinking potential, resulting in enhanced matrix reinforcement and mechanical strength. This is consistent with previous findings showing that the elastic modulus increases with increasing MW in HA-based hydrogels.⁸⁶ In contrast, LMW-HA forms shorter chains with reduced crosslinking capability, yielding softer and more flexible gels. Interestingly, HA-200 exhibited a lower shear and elastic modulus than HA-60 despite its higher molecular weight. This result may be attributed to the lower HA concentration in HA-200 (4%) compared to HA-60 (8%).¹⁰⁴ However, to clarify the relationship between molecular weight and elastic properties, further studies are needed using formulations with equivalent HA concentrations.

Biological tissues display a broad spectrum of mechanical properties,^{105, 106} making it essential to match the mechanical properties of the target tissue to support proper cell behavior and promote successful tissue regeneration.¹⁰⁵ For example, the elastic modulus of soft tissues such as the cortex and spinal cord is 10¹–10⁴ Pa, and that of medium tissues such as cartilage is 10⁵–10⁷ Pa.^{107, 108} The distinct mechanical profiles of these hydrogels suggest their applicability in mimicking various tissue environments.

In contrast to the trend observed in elastic properties, viscosity measurements showed an opposite pattern among the formulations (Fig. 3C). HA-200 exhibited the highest viscosity (1862 Pa·s), followed by HA-60 (1572 Pa·s) and HA-1500 (1105 Pa·s). However, the differences were much smaller and not statistically significant. This may be because HMW-HA can induce stronger elastic properties within the network due to its longer polymer chains, while its lower HA concentration reduces the overall gel network density, potentially leading to minimal differences in viscosity. Conversely, LMW-HA has shorter chains and thus lower elasticity, but its higher HA concentration can help maintain viscosity at a level comparable to other formulations. Nevertheless, further studies are needed to clarify the relationship between HA molecular weight and viscosity, as it remains poorly understood.

3.3. Cell viability and proliferation

Cell viability and proliferation were confirmed using Live/Dead staining and CCK assay using the resting state of macrophages (Fig. 4A-B). We selected HA-60 as the internal control, given that HA inherently has immunomodulatory effects. In this study, we aim to investigate how different MWs of HA-based hydrogels affect immune responses under 3D culture conditions. All hydrogels exhibited excellent cytocompatibility, with more than 80% of cells remaining viable on each gel after 24 hours (Fig. 4C-D). However, cell proliferation after 3 days was highest in HA-1500 (Fig. 4E). This is probably due to the interaction between CD44 and HA. CD44 plays a key role in regulating cell adhesion and migration.¹⁰⁹ HA-CD44 interaction can promote macrophage proliferation by inducing the expression of growth factors such as insulin-like growth factor-1 (IGF-1),¹¹⁰ and the p38 mitogen-activated protein kinase (MAPK) pathway is known to be involved in regulating macrophage proliferation.^{111, 112} CD44 requires at least three HA repeat units for binding, with improved affinity beyond that.^{113, 114} When the degree of modification is the same, lower MW HA binds less efficiently to CD44 than higher MW HA because it has fewer binding sites on CD44. Therefore, these different numbers of binding sites would affect the different proliferation of macrophages within the hydrogel. These results indicate that higher MW HA better supports macrophage proliferation, which could be advantageous for cell-hydrogel interactions.

Figure 4.

Viability and proliferation of macrophages cultured with hydrogels. A) Schematic illustration of the experimental plan to confirm viability and proliferation. Created in BioRender. Nano, B. (2025) https://BioRender.com/u2sfgkh. B) Representative photograph of hydrogel with 3D culture. C) Immunofluorescent images of live (green) and dead (red) cells in hydrogels after 1 day in culture. Scale bar = 100 μm. High-magnification images of the dashed boxes. Scale bar = 20 μm. D) Quantitative analysis of live cells. All formulations showed high cell viability of more than 80%. E) The cell proliferation rate was significantly higher in HA-1500 after 3 days. A one or two-way ANOVA with Tukey’s post hoc test for the multiple comparisons. **** p<0.0001.

3.4. The polarization of macrophages in 2D culture

To evaluate the functional phenotype of macrophages within hydrogels, we use macrophages in three different states: Resting, pro-inflammatory (stimulated with LPS/IFN-γ), and anti-inflammatory (stimulated with IL-4) states. Prior to hydrogel encapsulation, phenotypic switching of macrophages was confirmed in 2D culture. Pro- and anti-inflammatory marker (CD86 and CD206) expressions are determined after 1 day of treatment in 2D culture. CD86 and CD206 were highly regulated in LPS/IFN-γ and IL-4 stimulated cells, respectively (Fig. S3). When stimulated with LPS/IFN-γ, the area of RAW 264.7 cells increased, which is consistent with previous findings.¹¹⁵ It ensured that the differentiation into each state occurred properly after 1 day of stimulation.

After differentiation into each state, the expression of inflammation-related genes was compared in 2D culture after 3 days (Fig. S4). All pro-inflammatory markers, CCL2, TNF-α, iNOS, and CD86, increased in the LPS/IFN-γ treatment group. CCL2 showed the highest relative expression. There was no significant difference between the IL-4 group and the resting group. In the case of anti-inflammatory markers, only CD206 significantly increased in the IL-4 group compared to other groups, and its expression in the LPS/IFN-γ group decreased compared to the resting group. However, other anti-inflammatory markers were expressed more in the LPS/IFN-γ group, which is consistent with other studies showing increased anti-inflammatory markers after LPS treatment.^116–119 When comparing the resting group and the IL-4 group, anti-inflammatory genes showed an increasing trend, and CD206 and TGF-β1 showed significant differences. According to this gene expression data, the difference in the expression of CCL2 and CD206 genes on day 3 can be considered an accurate marker for confirming RAW 264.7 cell differentiation.

3.5. Immunomodulation of the resting state of macrophages cultured in hydrogels

Initial evaluation of macrophage activation by HA-hydrogels with different MWs was performed utilizing the resting state of macrophages (Fig. 5A). In the resting state, phenotypical change of macrophages varied depending on the MW of HA. The proportion of CD86-positive cells did not significantly differ among the HA molecular weights. On the other hand, the proportion of CD206-positive cells was significantly higher in HA-1500 than in HA-60 and HA-200 (Fig. S5 and 5B-C). Gene expression analysis mirrored this trend, with pro-inflammatory markers not showing significant differences according to MW of HA (Fig. 5D). On the other hand, among anti-inflammatory markers, CD206 gene expression increased significantly at HA-1500 compared to HA-60 and HA-200. TGF-β1 (p = 0.1763 vs HA-60, p = 0.7821 vs HA- 200) and IL-10 (p = 0.0696 vs HA-60, p = 0.0906 vs HA-200) showed an increasing trend at HA-1500, but there was no statistical significance (Fig. 5E).

Figure 5.

In vitro response of the resting state of macrophages cultured with hydrogels A) Schematic illustration of the experimental plan for assessing immunomodulation effects on the resting state of macrophages. Created in BioRender. Nano, B. (2025) https://BioRender.com/pbaiihk. B) Quantitative analysis of images for pro-inflammatory and anti-inflammatory markers. The proportion of CD206-positive cells was significantly higher in HA-1500 than in HA-60 and HA-200. C) Images of cells in hydrogel stained for pro-inflammatory marker CD86 (green) and anti-inflammatory marker CD206 (red). Scale bar = 50 µm. High-magnification images of the dashed boxes. Scale bar = 20 µm. Relative gene expression of D) pro-inflammatory genes and E) anti-inflammatory genes. CD206 gene expression increased significantly at HA-1500 compared to HA-60 and HA-200. A one-way ANOVA with Tukey’s post hoc test for the multiple comparisons. ** p < 0.01, *** p < 0.001, n = 3–4 per group.

To summarize, in resting macrophages, anti-inflammatory markers such as CD206 increased in the HMW-HA hydrogel, while there was no change in pro-inflammatory markers. These findings suggest that the HMW-HA hydrogel itself biases macrophages toward an anti-inflammatory phenotype. Similar phenotypical shifts in immune cell behavior have been observed in another study, with HMW-HA-rich matrices spontaneously inducing anti-inflammatory polarization ¹²⁰.

3.6. Immunomodulation of pro-inflammatory macrophages cultured in hydrogels

We investigated phenotypic changes in macrophages pre-activated with LPS and IFN-γ to determine the effect of phenotypic reprogramming from pro-inflammatory to anti-inflammatory (Fig. 6A). As a result of immunofluorescence staining of pro-inflammatory macrophages, the proportion of CD86-positive cells was maintained high at approximately 55–60% in the HA-60 and HA-200 but decreased sharply in the HA-1500 to less than 5%. On the other hand, the proportion of CD206-positive cells increased in HA-1500, which was significantly higher than that in HA-60, but there was no significant difference compared to that in HA-200 (Fig. S6 and 6B-C). Gene expression analysis confirmed that pro-inflammatory markers were suppressed by the HMW-HA hydrogel, but iNOS and CD86 showed no difference by MW of HA (Fig. 6D). However, CCL2 significantly decreased in HA-200 and HA-1500 compared to HA-60. TNF-α in HA-1500 also significantly decreased compared to HA-60. Meanwhile, anti-inflammatory genes were upregulated. CD206 showed a significantly higher level at HA-1500 compared to HA-200, and IL-10 also significantly increased at both HA-200 and HA-1500 compared to HA-60. Arg1 and TGF-β1 genes slightly increased and decreased at HA-1500, respectively, but there was no statistical difference (Fig. 6E).

Figure 6.

In vitro response of LPS/IFN-γ activated macrophages cultured with hydrogels. A) Schematic illustration of the experimental plan for assessing immunomodulation effects on pro-inflammatory macrophages. Created in BioRender. Nano, B. (2025) https://BioRender.com/ndkijtt. B) Quantitative analysis of images. The proportion of CD86-positive cells decreased sharply in the HA-1500 to almost less than 5% and the proportion of CD206-positive cells was significantly higher than in HA-60. C) Images of cells in hydrogel stained for pro-inflammatory marker CD86 (green) and anti-inflammatory marker CD206 (red). Scale bar = 50 µm. High-magnification images of the dashed boxes. Scale bar = 20 µm. Relative gene expression of D) pro-inflammatory genes and E) anti-inflammatory genes. A one-way ANOVA with Tukey’s post hoc test for the multiple comparisons. * p < 0.05, ** p < 0.01, **** p < 0.0001, n = 3–4 per group.

In summary, HA-1500 partially reprogrammed pro-inflammatory macrophages toward an anti-inflammatory profile. These results align closely with the previous finding that HMW-HA decreases the expression of pro-inflammatory genes and increases the expression of anti-inflammatory genes in LPS-stimulated macrophages.^{32, 33} This suggests that the HMW-HA hydrogel may help modulate acute inflammation by shifting macrophages actively away from a pro-inflammatory state.

3.7. Immunomodulation of anti-inflammatory macrophages cultured in hydrogels

We also investigated phenotypic changes in macrophages preactivated with IL-4 (Fig. 7A). In IL-4 pre-activated macrophages, HMW-HA hydrogel also modulated cellular phenotypes. Immunofluorescence data showed higher CD206 positivity on HA-1500 versus HA-60 and HA-200 (Fig. S7 and 7B-C). In gene expression data, pro-inflammatory genes (iNOS and CCL2) were reduced in HA-1500 (Fig. 7D), and CD206 and TGF-β1 transcripts were upregulated relative to HA-60 (Fig. 7E). Similar to the LPS/IFN-γ stimulated case, HA-1500 suppressed pro-inflammatory gene expression and enhanced anti-inflammatory gene expression. Importantly, CD86 remained low on HA-1500 in anti-inflammatory conditions, indicating that HMW-HA hydrogel does not induce conflicting signals in anti-inflammatory cells.

Figure 7.

In vitro response of IL-4 activated macrophages cultured with hydrogels. A) Schematic illustration of the experimental plan for assessing immunomodulation effects on anti-inflammatory macrophages. Created in BioRender. Nano, B. (2025) https://BioRender.com/7mjsucf. B) Quantitative analysis of images. The proportion of CD206-positive cells was significantly higher at HA-1500 than at HA-60 and HA-200. C) Images of cells in hydrogel stained for pro-inflammatory marker CD86 (green) and anti-inflammatory marker CD206 (red). Scale bar = 50 µm. High-magnification images of the dashed boxes. Scale bar = 20 μm. Relative gene expression of D) pro-inflammatory genes and E) anti-inflammatory genes. A one-way ANOVA with Tukey’s post hoc test for the multiple comparisons. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, n = 3–4 per group.

Previous studies have shown that LMW-HA fragments stimulate TLR4-mediated NF-κB signaling and drive macrophages toward a pro-inflammatory phenotype, whereas HMW-HA tends to induce immunosuppression ²⁷. Importantly, those findings were obtained in 2D culture. To better mimic an in vivo environment, we evaluated HA-based hydrogels with different MW in a 3D culture system. Our data indicated that macrophages in all three activation states exhibited a shift toward a more anti-inflammatory profile in the HMW-HA hydrogel. However, this result alone cannot fully attribute the effect to the MW of HA, because physical properties of the biomaterial such as stiffness and pore structure are also known to regulate macrophage polarization.

We therefore considered how the physical properties of hydrogel might influence our findings. Indeed, stiffness is a well-established cue for macrophage phenotype. Increasing the compressive modulus of the hydrogel has been shown to promote a shift toward an anti-inflammatory profile.^{121, 122} The substrate with 0.1–4 kPa stiffness promotes pro-inflammatory polarization, and the substrate with 25–40 kPa stiffness contributes to anti-inflammatory polarization.¹²² Similarly, pore size profoundly affects macrophage behavior, with larger pores promoting pro- to anti-inflammatory polarization and the secretion of anti-inflammatory cytokines.^123–125 For instance, in a study utilizing 3D-printed scaffolds, those with larger pores demonstrated significantly increased anti-inflammatory macrophage infiltration, enhanced vascularization, and improved bone regeneration compared to smaller pores.¹²⁶ Similarly, implants with medium to large pore sizes were associated with a significantly higher number of CD163⁺ pro-regenerative macrophages than those with smaller pores.¹²⁷ These studies consistently show that larger pores and higher stiffness promote anti-inflammatory polarization.

Notably, our HA-1500, which had substantially larger pores and higher stiffness than the lower-MW hydrogels, elicited an anti-inflammatory macrophage phenotype. This suggests that immunomodulation in the 3D culture system may be driven not only by HA MW, but also by MW-associated differences in the physical properties of the hydrogel. As a result, the observed immunomodulatory effects appear to reflect the combined influence of MW-associated biochemical and physical properties. Future studies using mechanics-matched hydrogel comparisons across distinct MWs will be needed to more definitively isolate the MW–dependent biochemical effects of HA on macrophage polarization.

Another limitation of this study is the focus on short-term culture time points for assessing macrophage viability and inflammatory markers in HA hydrogels. While these early time points capture acute cellular responses following encapsulation, they do not account for the effects of hydrogel evolution and degradation on macrophage behavior beyond the initial phase. As the HA matrix gradually breaks down, the resultant changes in mechanical properties and the release of lower MW HA fragments could modulate macrophage polarization and function. Increasing pore size due to degradation can facilitate cell spreading and infiltration, which has been associated with a pro-regenerative phenotype.^123–125 On the other hand, a softer matrix and the release of lower MW HA fragments over time might encourage pro-inflammatory macrophages.^{121, 122, 128} Therefore, future work should incorporate extended culture durations (e.g., >14 days) to define long-term macrophage dynamics in these HA hydrogel systems. In addition, RAW 264.7 cells were used in this study as an initial in vitro screening model to investigate macrophage responses to HA-based hydrogels. While this cell line is a well-established screening model and widely employed in biomaterials research, it does not fully recapitulate the biological complexity of primary macrophages.¹²⁹ Therefore, the MW-associated responses observed here should be followed up with studies using primary macrophages (e.g., murine bone-marrow-derived macrophages or human monocyte-derived macrophages) and ultimately validated through in vivo animal models. Future research will extend these findings using primary macrophages and animal models to evaluate how different MWs of HA-based hydrogels modulate immune responses in physiologically relevant settings.

Collectively, our results demonstrate that tuning HA concentration and MW provides a versatile platform to control both the biophysical properties of hydrogel and immune cell responses. By adjusting MW and crosslinking density, hydrogels can be engineered with specific stiffness and porosity to drive immune cell activation toward either pro-inflammatory or anti-inflammatory states. This tunability enables the design of HA-based hydrogels optimized for specific tissues. For instance, applications for soft tissues or skin may require very low stiffness or fine pores, whereas cartilage demands higher stiffness. Overall, our design framework thus highlights the broad potential of HA-based hydrogels as adaptable platforms for reprogramming of immune responses and tissue-specific regenerative applications.

Conclusion

In this study, we demonstrated that HA-based hydrogels with tunable MWs provide a powerful platform for modulating immune responses and tailoring physical properties for specific tissue applications. By adjusting the MW and concentration of HA, we controlled critical hydrogel parameters including gelation kinetics, pore size, stiffness, swelling, and degradation rate. These properties are known to affect immune cell phenotype and tissue compatibility. Notably, HMW-HA hydrogels promoted an anti-inflammatory phenotype across all macrophage states in 3D culture, indicating their potential to mitigate excessive inflammation.

Beyond immunomodulatory capacity, the mechanical tunability of HA-based hydrogels enables their adaptation to a wide range of regenerative medicine applications. For instance, stiffer and more porous HA-1500 is well-suited for cartilage or spinal cord repair, where higher mechanical strength or cell infiltration is required. In contrast, softer hydrogels with smaller pores may better support skin regeneration, where more delicate matrix environments are favorable. Collectively, these findings demonstrate the potential of HA-based hydrogels as customizable ECM analogs for 3D cell culture, organoid models, and tissue-specific regenerative therapies.

Supplementary Material

Supporting Information Available

The following files are available free of charge.

¹H NMR spectra of HA and HA-Ac (Figure S1); Gelation time of hydrogels by the inverted tube test (Figure S2); Immunofluorescence image and analysis in 2D culture at 1 day after treatment (Figure S3); Gene expression of pro- and anti-inflammatory genes in 2D culture (Figure S4); Overview images of resting macrophages cultured with hydrogels (Figure S5); Overview images of LPS/IFN-γ activated macrophages cultured with hydrogels (Figure S6); Overview images of IL-4 activated macrophages cultured with hydrogels (Figure S7); Primary and secondary antibodies for immunofluorescence staining (Table S1); Primer sequence for RT-qPCR (Table S2)

Data availability

The datasets generated during this study are publicly available in the Dryad Digital Repository at DOI: 10.5061/dryad.280gb5n4g).