Dual-functional injectable hydrogels as antimicrobial and angiogenic therapeutics for dental pulp regeneration

Abstract

Pulpitis is a painful inflammatory condition of the dental pulp, commonly triggered by bacterial infection, trauma, or repeated dental procedures. Patients often experience heightened sensitivity to temperature changes, spontaneous pain, and, in severe cases, necrosis of the pulp tissue, leading to tooth loss if left untreated. Traditional treatment primarily involves root canal therapy, which removes infected pulp and seals the canal with gutta-percha; however, this approach lacks antimicrobial properties and does not support tissue regeneration. To address these limitations, this study investigates an injectable hydrogel system composed of hyaluronic acid (HA) and carboxymethyl chitosan (CMCs), chemically crosslinked with 1,4-butanediol diglycidyl ether (BDDE). The hydrogel demonstrated excellent injectability, self-healing ability, and mechanical stability. In vitro and in vivo assessments confirmed its biocompatibility, biodegradability, and capability for controlled release of therapeutic agents. Notably, the amoxicillin-loaded hydrogel (HACM/AX) showed significant antibacterial properties; while the erythropoietin-loaded hydrogel (HACM/EPO) significantly enhanced endothelial cell proliferation and migration, facilitating angiogenesis for tissue regeneration. These findings highlight the potential of the HACM hydrogel as a dual-function biomaterial with antimicrobial and pro-angiogenic properties, offering a promising alternative for pulpitis treatment and dental tissue engineering.

---

1. Introduction

Oral diseases represent a significant global health burden, with an estimated 3.5 billion people affected worldwide.¹ These conditions range from common dental caries and periodontal diseases to more severe issues like oral cancer and endodontic diseases such as pulpitis.¹ Pulpitis, which involves inflammation of the dental pulp, is of particular concern due to its painful nature and potential to progress into more severe conditions if left untreated. The dental pulp is a soft tissue comprising nerves, blood vessels, and connective tissues, and its inflammation can be triggered by various factors including bacterial infection (typically following untreated dental caries), trauma, or repeated dental procedures.^2,3 Symptoms of pulpitis range from mild discomfort to severe, sharp pain, often exacerbated by exposure to temperature changes or pressure.⁴ If not addressed in a timely manner, pulpitis can progress from a reversible state, where the pulp can still heal, to irreversible pulpitis, leading to pulp necrosis, abscess formation, and tooth loss.⁵ Therefore, effective intervention is crucial to mitigate the potentially serious consequences of pulpitis.

The treatment of pulpitis depends largely on its classification as reversible or irreversible.⁶ In cases of reversible pulpitis, where the inflammation is mild and the pulp is still capable of healing, treatment typically involves the removal of the irritant, such as decayed tissue, followed by the application of a restorative material to protect the pulp and allow it to recover. However, when pulpitis progresses to its irreversible form, where the damage is too extensive for the pulp to heal, more invasive interventions are necessary.⁷ The standard approach to treating irreversible pulpitis is root canal therapy (RCT). During RCT, the infected or necrotic pulp tissue is removed, the root canal is cleaned and disinfected, and the canal is then filled with a biocompatible material.⁸ Gutta-percha, a plant-derived latex from the Palaquium tree, is commonly used as the filling material in root canal procedures due to its biocompatibility, ease of manipulation, and inert nature. Despite its widespread use, gutta-percha has several drawbacks, including inadequate sealing properties, lack of antimicrobial effects, and its inability to support tissue regeneration.^9,10 These limitations highlight the need for alternative materials that can provide better outcomes, particularly in terms of sealing ability, antibacterial properties, and regenerative potential.

In recent years, hydrogels have emerged as a promising alternative to gutta-percha for use in endodontic treatments such as root canal therapy.¹¹ Hydrogels are three-dimensional, water-rich polymer networks that can be designed to mimic the extracellular matrix (ECM) of natural tissues, making them suitable for a variety of biomedical applications, including tissue engineering and drug delivery.^12,13 When used in pulpitis treatment, hydrogels offer several significant advantages over traditional materials like gutta-percha. For instance, their injectability allows for minimally invasive application, and their capacity to form a gel in situ ensures that they conform perfectly to the shape of the root canal, providing a better seal.^4,14 Moreover, hydrogels can be engineered to possess bioactive properties, promoting tissue regeneration, exhibiting antibacterial activity, and delivering therapeutic agents directly to the site of infection.^15,16 This multifunctional approach is especially beneficial in dental pulp therapy, where the primary goals are not only to eliminate infection but also to regenerate damaged tissues and restore tooth vitality.^17,18 Hydrogels thus offer a more dynamic and versatile alternative to gutta-percha, which acts solely as a passive filler material.

The incorporation of biopolymers into hydrogel formulations further enhances their utility in treating pulpitis. Biopolymers such as hyaluronic acid (HA) and carboxymethyl chitosan (CMCs) are particularly well-suited for use in dental applications due to their biocompatibility, biodegradability, and ability to support tissue regeneration. Hyaluronic acid, a naturally occurring polysaccharide in the human body, plays a key role in wound healing and tissue repair through its ability to promote cell migration and proliferation. In the context of dental pulp therapy, HA can enhance angiogenesis and support the regeneration of damaged pulp tissues.^19,20 Carboxymethyl chitosan, a derivative of chitosan found in the exoskeletons of crustaceans, is known for its antimicrobial properties and its capacity to form biocompatible hydrogels. CMCs can also serve as an effective drug delivery vehicle, enabling the sustained release of antibiotics or other therapeutic agents directly to the site of infection.^21,22 Together, these biopolymers can be used to formulate a hydrogel system that offers superior sealing properties compared to gutta-percha, while simultaneously providing antibacterial protection, supporting tissue regeneration, and delivering drugs. This combination of properties makes hydrogels composed of HA and CMCs a highly attractive option for treating pulpitis and overcoming the limitations of traditional materials.

The use of hydrogels as an alternative filling material offers significant advantages, particularly their capacity to encapsulate and deliver bioactive compounds. These bioactive agents not only mitigate the primary cause of pulpitis, such as bacterial infections, but also promote the healing of damaged tissue and nerves deep within the pulp. Among the most relevant therapeutic compounds for this application are amoxicillin (AX) and erythropoietin (EPO), which exhibit distinct yet complementary properties, making them promising candidates for pulpitis treatment. Amoxicillin, a broad-spectrum β-lactam antibiotic, is widely employed in dental applications due to its high efficacy against prevalent oral pathogens, including Streptococcus and Enterococcus species.^23–25 Since bacterial infection is a leading factor in pulpitis development, AX serves as a crucial antimicrobial agent, effectively eliminating bacterial proliferation and preventing further tissue degradation. Its mechanism of action, which involves disrupting bacterial cell wall synthesis, ensures rapid bacterial clearance, reducing inflammation, and facilitating tissue recovery. Conversely, erythropoietin, traditionally recognised for its role in erythropoiesis, has gained attention in regenerative medicine due to its pleiotropic effects, particularly in tissue repair and angiogenesis.²⁶ EPO enhances cell proliferation, migration, and neovascularisation, all of which are critical for pulp regeneration. Furthermore, it modulates endothelial cell function and stimulates the secretion of pro-survival and anti-inflammatory cytokines, fostering a conducive environment for tissue repair.²⁷ By incorporating both AX and EPO into a controlled-release hydrogel system, a dual-function therapeutic approach could be achieved, simultaneously preventing bacterial infection while promoting pulp tissue regeneration. This synergistic strategy holds significant potential in advancing pulpitis treatment by integrating infection control with regenerative therapy within a single intervention.

In summary, this study hypothesised that an injectable hydrogel could serve as a promising candidate for the treatment of pulpitis and pulp tissue damage. Specifically, a chemically crosslinked hydrogel was developed using HA and CMCs, designed to be injectable and capable of in situ gel formation. This system was intended to prevent bacterial proliferation while simultaneously promoting tissue regeneration through enhanced angiogenesis. The research focused on the synthesis, characterisation, and optimisation of the hydrogel to determine the most suitable formulation for therapeutic application. Mechanical and physical properties were systematically analysed using texture analysis techniques. The injectability of the hydrogel was validated using both in vitro and in vivo models. Furthermore, antimicrobial efficacy was assessed against two common oral bacterial strains using the disk diffusion assay, with hydrogels loaded with AX. Additionally, comprehensive biocompatibility assessments, including cell proliferation and migration assays, were performed on hydrogels loaded with EPO to evaluate their potential to enhance tissue regeneration. Based on these findings, the proposed filling hydrogel system was aimed to have a dual-function injectable biomaterial with both antibacterial properties and tissue-healing capabilities, highlighting its potential for application in pulpitis therapy (Scheme 1).

Scheme 1 Schematic illustration of pulpitis pathogenesis and the proposed therapeutic strategy utilising injectable hydrogels. (A) Pathogenic mechanism underlying pulpitis development. (B) Application of injectable hydrogel with angiogenic and antimicrobial properties for (C) dental pulp regeneration.

2. Materials and methods

2.1. Materials

Hyaluronic acid (HA, 800–1000 kDa) was provided by Aladdin Scientific (CA, United States). Carboxymethyl chitosan (CMCs, degree of carboxymethylation ≥90%, 240 kDa) and 1,4-butanediol diglycidyl ether (BDDE, 60 wt%) were purchased from Shanghai Macklin Biochemical Co., Ltd (Shanghai, China). Sodium hydroxide (96%) was provided by Guangdong Guanghua Sci-Tech Co., Ltd (Guangdong, China). Amoxicillin trihydrate (AX) was obtained from Bio-Pharmachemie (Ho Chi Minh, Vietnam). Erythropoietin (EPO, 2000 IU mL⁻¹) was purchased from Nanogen Pharmaceutical Biotechnology (Ho Chi Minh, Vietnam).

2.2. Characterisation methods

Spectroscopy. Fourier-transform infrared spectroscopy (FTIR, Spectrum Two FTIR, Bruker, United States) was employed to qualitatively assess the chemical structure of the hydrogel. A UV-vis spectrophotometer (CLARIOstar Plus Microplate Reader, BMG Labtech, Germany) was utilised for the quantitative analysis of various substances for different experimental purposes.

Microscopy. The internal pore structure of the hydrogel cross-sections was examined using a scanning electron microscope (SEM, Hitachi S-4800, Hitachi High-Tech, Japan) following freeze-drying with a Martin Christ Alpha 1–2 LD plus system (Martin Christ Gefriertrocknungsanlagen GmbH, Germany). Fluorescence images of stained cells were captured using a fluorescence microscope (Olympus CKX53, Olympus Corporation, Japan). The real-time monitoring and time-lapse imaging of label-free cell growth were recorded with the HoloMonitor (Phase Holographic Imaging, Sweden).

Mechanical properties. The viscosity of the HACM solution was measured using a rotational viscometer (DVE Viscometer, AMETEK Brookfield, United States) with an LV-3C spindle at rotational speeds of 30, 50, 60, and 100 rpm. The lap shear test was performed using a tensile strength tester (RTC TENSILON, A&D Company, Japan) under a 5 N tensile force applied at a speed of 10 mm min⁻¹. The rheological behaviour of the hydrogel was examined with a texture analyser (TA.XTplus, Stable Micro Systems, United Kingdom) under four different modes: frequency sweep, strain sweep, cyclic strain time sweep, and continuous flow.

2.3. Preparation of the HACM hydrogel

HACM was synthesised using HA and CMCs as the primary components, with BDDE serving as the crosslinking agent. Initially, HA and CMCs were dissolved separately in water at specific concentrations, as detailed in Fig. 1. Each solution was stirred at 200 rpm and maintained at 37 °C for 30 minutes to ensure complete dissolution. The two solutions were then combined, and the pH was adjusted to 9.0 using NaOH 1 M before continuing stirring under identical conditions for a further 30 minutes. For drug-loaded samples, AX (5 mg mL⁻¹) and EPO (40 IU mL⁻¹) were incorporated, whereas deionised water was added to the hydrogel formulation without drugs. Subsequently, BDDE was introduced as the curing agent, and the mixture was gently stirred until gel formation occurred and further incubated for 1 hour. Finally, the injectable hydrogel was ready to be used after the hydrogel was washed three times with distilled water to remove any residual BDDE on the surface.

Fig. 1 Characterisation of HACM hydrogels. (A) Photographic images and (B) FTIR spectra of the HACM mixture with and without BDDE crosslinking. Optimisation of gelation time with varied HACM component concentrations: (C) HA, (D) CMCs, and (E) BDDE. (*) The units of HA, CMCs, and BDDE are wt%. (F) Swelling ratio of the HACM samples. (G) SEM images of the HACM hydrogel and (H) pore size distribution calculated from pore dimensions measured using ImageJ software.

2.4. Swelling ratio of the HACM hydrogel

The swelling capacity of the hydrogel was assessed based on its water absorption ability. Initially, HACM hydrogels at varying component concentrations were subjected to freeze-drying, and their mass post-drying was recorded as the dry weight (W_d). These dried hydrogel samples were then immersed in PBS solution at pH 7.4 and maintained at 37 °C. At predetermined time intervals, the samples were retrieved, gently blotted with Kimwipes to remove excess surface water, and subsequently weighed to determine their swollen weight (W_s) at each time point. The swelling ratio of the hydrogel was then calculated using the equation provided below.⁴

2.5. Self-healing and adhesive properties of the HACM hydrogel

The self-healing capability of the HACM hydrogels was investigated by preparing two separate HACM hydrogel samples and staining them differently using food colouring agents (green and yellow), each of which was subsequently cut into two halves. The differently coloured halves were then manually brought into contact and allowed to rest for approximately 5 minutes before being gently pulled apart to evaluate their adhesion. The hydrogels were photographed using a digital camera to record the self-healing properties visually. Furthermore, the adhesive performance of the HACM hydrogels was assessed using various biological tissues, including adipose, muscular, hepatic, and cardiac tissues, which were manually attached onto the surfaces of latex laboratory gloves. Digital images were captured to document the tissue adhesion characteristics. To quantitatively evaluate adhesion strength and load-bearing capacity, a simulated lap shear testing procedure was conducted by joining two pieces of porcine skin, each measuring 5 cm in length, and subsequently applying tensile loads using standardised weights of 5, 20, and 50 g. A similar experimental procedure was also carried out using a dedicated tensile strength tester to corroborate the lap shear results.

2.6. In vitro and in vivo injectability of the HACM hydrogel

The injectability and shear-thinning property of the HACM hydrogel were initially assessed using an in vitro model prior to in vivo experiments. For this purpose, the hydrogel was loaded into a 1 cc syringe fitted with a 26G needle, manually extruded into a silicone mold, and then PBS solution was added to evaluate its shear-thinning properties and structural integrity upon injection. Subsequently, an in vivo evaluation was carried out on female BALB/c mice, aged 8–10 weeks, with body weights ranging from approximately 20 to 25 g. The animal study was approved by the ethics committee of Ton Duc Thang University and was performed in accordance with the guidelines set forth by the National Research Council's Guide for the Care and Use of Laboratory Animals. Specifically, after being anaesthetised using isoflurane, the mice were injected subcutaneously with the HACM hydrogel employing the same syringe and needle configuration as used in the in vitro tests. Following injection, the hydrogels were allowed to stabilise in situ for approximately 30 minutes prior to surgical examination, enabling evaluation of hydrogel localisation. The mice were monitored closely for their general health status throughout the experimental period. Subsequently, animals were euthanised at time points of one and two weeks post-injection. At each endpoint, hydrogel implants were retrieved, and biodegradability was assessed quantitatively by measuring the mass of the recovered samples.

2.7. Controlled drug release of the HACM hydrogel

In the present study, the release characteristics of HACM hydrogels loaded with AX (5 mg mL⁻¹) and EPO (40 IU mL⁻¹) were investigated. Drug-loaded HACM hydrogels with a volume of 1 cm³ were immersed in 8 mL of PBS buffer solution at pH 7.4 and maintained at 37 °C. At 24-hour intervals, 1 mL of the PBS medium was removed and subsequently replenished with an equal volume of fresh PBS. Collected samples were stored frozen prior to further analysis. The concentration of the released AX was determined spectrophotometrically by measuring the absorbance at 272 nm. In contrast, EPO release was quantified using the Bradford method, whereby the protein was stained using Coomassie Brilliant Blue G-250 solution (Sigma-Aldrich, MO, United States), and the absorbance of the resultant solution was measured at 595 nm.

2.8. Antibacterial properties of the HACM hydrogel

The antimicrobial activity of the HACM hydrogels was evaluated using a disk diffusion assay. Hydrogels loaded with AX (5 mg mL⁻¹) and unloaded hydrogels were compared to assess their antimicrobial effectiveness. Two representative bacterial strains associated with dental infections, namely Gram-positive Bacillus subtilis (B. subtilis, ATCC 6051) and Gram-negative Escherichia coli (E. coli, ATCC 25922), were selected for testing. To carry out the disk diffusion assay, bacterial suspensions were cultivated to a density of 1 × 10⁸ CFU mL⁻¹, after which 1 mL of each suspension was evenly spread onto 9 cm diameter agar plates. Subsequently, the unloaded and AX-loaded HACM hydrogel samples were positioned at the centre of the agar plates and incubated at 37 °C for a duration of 24 hours. Following incubation, digital images of the agar plates were captured, and the diameters of the resulting zones of inhibition were measured using ImageJ software. Amoxicillin (1 mg mL⁻¹) was employed as a positive control, for which filter paper disks with a diameter of 6 mm were soaked in antibiotic solution and placed onto agar plates in the same manner.

2.9. Haemolytic activity of the HACM hydrogel

The haemocompatibility of HACM hydrogels was evaluated using human erythrocytes. Initially, 3.5 mL of citrated human blood was centrifuged at 1500 × g to separate the red blood cells from plasma. After removal of the plasma layer, the erythrocytes were washed several times with PBS buffer, repeating the centrifugation conditions mentioned previously. The resulting erythrocyte suspension was diluted to 4% and employed to assess haemolysis induced by HACM. In the haemolysis test, Triton X-100 solution (0.1%) served as a positive control, PBS acted as the negative control, and HACM samples at varying concentrations (100–1000 μg mL⁻¹) were used as the test samples. Each sample or control was mixed with the erythrocyte suspension at a ratio of 1 : 10, followed by incubation at 37 °C for 24 hours with gentle shaking at 200 rpm. Subsequently, the mixtures were centrifuged, and the absorbance of the supernatants was measured at 545 nm using UV-vis spectroscopy. The percentage haemolysis was calculated according to the equation provided below.²⁸

2.10. Cytocompatibility of the HACM hydrogel

Cell culture. Mouse macrophages (RAW 264.7, ATCC TIB-71, passage 20), mouse aortic smooth muscle cells (MOVAS, ATCC CRL-2797, passages 15–17), and mouse endothelial cells (SVEC4-10, ATCC CRL-2181, passages 10–12) were cultured in low-glucose Dulbecco's modified Eagle's medium (DMEM, Thermo Fisher Scientific, MO, United States) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. All cell types were grown in a humidified incubator at 37 °C under a 5% CO₂ atmosphere.

ROS induction. Reactive oxygen species (ROS) generated by macrophages were evaluated by seeding RAW 264.7 cells at a density of 1 × 10⁴ cells per well. The cells were cultured in DMEM modified with HACM at various concentrations (100–1000 μg mL⁻¹). Cells cultured in standard DMEM without hydrogels served as controls. After incubation for 24 hours under standard conditions, ROS levels produced by macrophages were measured using the fluorescent ROS probe 2′,7′-dichlorofluorescein diacetate (DCFDA, 50 μM, Thermo Fisher Scientific, MO, United States).

Cell viability. Similarly, MOVAS and SVEC4-10 cells were seeded at a density of 1.2 × 10⁴ cells per well and cultured in DMEM for 24 hours to reach approximately 80–90% confluency. Subsequently, the existing medium was replaced with fresh DMEM containing different concentrations of HACM hydrogel (100–1000 μg mL⁻¹) or HACM/EPO (namely (D), EPO concentration of 100 IU mL⁻¹). Untreated DMEM was employed as the control. After a further 24-hour incubation, cells were stained with LIVE/DEAD™ reagents (Thermo Fisher Scientific, MO, United States) containing calcein-AM (green fluorescence, 1 μM) and propidium iodide (red fluorescence, 1 μM), and cell viability was assessed qualitatively by fluorescence microscopy. Additionally, fluorescence intensity (FI) was quantified spectrophotometrically and used to determine cell viability as described by the equation below.

Cell proliferation. Proliferation of endothelial cells was monitored using real-time live-cell imaging with the HoloMonitor system. SVEC4-10 cells were seeded at a density of 8.0 × 10³ cells per well and cultured for 4 hours to facilitate initial attachment. The original medium was then withdrawn and replaced with fresh medium containing HACM (1000 μg mL⁻¹) or HACM/EPO (100 IU mL⁻¹ EPO). Cells were subsequently cultured for 24 hours, and proliferation was assessed by measuring changes in cellular coverage over time. Additionally, morphological features such as convexity, eccentricity, and cell area were analysed for each treatment group.

Cell migration. A scratch assay was utilised to investigate cell migration capabilities relevant to wound healing applications. SVEC4-10 cells were seeded at a density of 2.0 × 10⁴ cells per well and cultured for 4 hours under optimal conditions to establish a monolayer with approximately 95% coverage. A linear scratch was then introduced into the cell monolayer using a P200 pipette tip, and the wells were gently washed with PBS to remove cellular debris. Fresh medium was subsequently added, with untreated samples cultured in standard low-glucose DMEM and treatment groups receiving either HACM or HACM/EPO at 1000 μg mL⁻¹ and an EPO concentration of 100 IU mL⁻¹, respectively. Real-time wound coverage was monitored using the HoloMonitor system, and morphological analyses were subsequently conducted to obtain further insights into cell behaviour.

M2 macrophage polarization. RAW 264.7 M0 macrophages can be polarised into either M1 or M2 phenotypes, depending on the stimulatory cues present in the culture environment.²⁹ In this study, M2 polarisation was assessed through cellular morphology and immunocytochemistry (ICC) staining. Initially, RAW 264.7 cells were cultured in low glucose DMEM supplemented with either HACM hydrogel (1000 μg mL⁻¹) or HACM/EPO (100 IU mL⁻¹ EPO). The control group was maintained in plain culture medium. All samples were incubated for 48 hours, after which the cells were fixed with 4% paraformaldehyde for 30 minutes at room temperature. Subsequently, the cells were blocked and permeabilised using 1% bovine serum albumin and 0.1% Triton X-100. Alexa Fluor® 647-conjugated anti-CD163 antibody (Abcam Limited, Cambridge, United Kingdom), a widely recognised marker of M2 macrophages, was diluted at a ratio of 1 : 200 and applied to the cells overnight at 4 °C. The samples were then rinsed three times with PBS to remove background staining. Nuclei were counterstained using Hoechst dye (1.0 × 10⁻⁵ M, Thermo Fisher Scientific, MO, United States) for 30 minutes at room temperature. Following three additional PBS washes, the samples were visualised using a fluorescence microscope. The FI of the labelled antibodies was also quantified using a microplate reader. Furthermore, DCFDA staining was employed to evaluate intracellular ROS generation in the macrophages.

2.11. Chick embryo chorioallantoic membrane assay

Biocompatibility was evaluated using the chorioallantoic membrane (CAM) assay to investigate the hydrogel's compatibility during in ovo embryonic development. Fertilised eggs were incubated under controlled conditions at 36.5 °C and 65% relative humidity. On day 7 of embryonic development, the eggshell was aseptically opened in a biosafety cabinet previously sterilised by UV exposure. Pre-formed HACM hydrogel (20 mg per egg) was carefully applied to the CAM surface using thumb forceps. In parallel, the HACM/AX/EPO composite (containing 0.25 μg mg⁻¹ AX and 0.25 IU mg⁻¹ EPO) was utilised to assess the synergistic effect of dual-drug incorporation on the hydrogel's biocompatibility. A control group received 20 μL saline per egg. The development of the embryo and the CAM response were documented through photographic imaging at embryonic days 7, 10, and 12. Additionally, the percentage increase in blood vessel density was quantified using ImageJ software to determine the hydrogel's influence on vascularisation.

2.12. Statistical analysis

All experiments were performed with at least three replicates, and results were reported as the mean ± standard deviation. Graphical data presented in this study were plotted using GraphPad Prism software (GraphPad Software Inc., MA, United States). Statistical significance was indicated as follows: ns (not significant) p ≥ 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

3. Results and discussion

3.1. Synthesis and characterisation of the HACM hydrogel

The HACM hydrogel synthesised in this study as an in situ forming scaffold for pulp regeneration comprised two main components, HA and CMCs, and utilised BDDE as the crosslinking agent to stabilise its three-dimensional structure. As shown in Fig. 1A, the HACM solution without BDDE did not undergo gelation, highlighting that HA and CMCs alone lacked an intrinsic gel-forming capability due to the absence of suitable structural modification. Consequently, the addition of BDDE as a crosslinker was essential to stabilise the hydrogel structure.

The chemical modifications resulting from the BDDE-mediated crosslinking reaction were further confirmed through changes observed in FTIR spectral bands (Fig. 1B). Specifically, the uncrosslinked HACM mixture exhibited characteristic stretching vibrations corresponding to overlapped N–H and O–H, C–H, and C=O bonds around 3290, 2920, and 1660 cm⁻¹, respectively. Additionally, the presence of bending vibrations of N–H and O–H bonds at approximately 1580 and 1410 cm⁻¹ further indicated the incorporation of CMCs in the HACM composite. Given that polysaccharide-based materials typically contain abundant hydroxyl groups, a strong absorption band due to C–O stretching at around 1030 cm⁻¹ was also observed. These vibrations generally appeared stronger near 1000 cm⁻¹ due to the contribution of primary hydroxyl groups. However, upon crosslinking with BDDE, a noticeable shift towards higher wavenumbers (around 1130 cm⁻¹) occurred, which was attributed to the formation of secondary hydroxyl groups following BDDE-induced crosslink formation. These spectral shifts validated the successful establishment of crosslinking networks between BDDE and the hydroxyl groups of both HA and CMCs.

Following the fundamental characterisation of gel formation, optimisation of gelation time was critical for preparing the HACM hydrogel. Initially, the influence of HA concentration on the gelation time of HACM was investigated (Fig. 1C). Results revealed an inverse relationship between HA concentration and gelation time, with increased HA concentrations resulting in shorter gelation times. Specifically, at the highest HA concentration tested, HACM gelled within 5.3 ± 0.5 minutes. Notably, even in the absence of HA, a hydrogel structure formed from CMCs alone, albeit at a much longer duration of 68.7 ± 3.1 minutes, demonstrating the capability of BDDE to interact effectively with hydroxyl and amino groups present on CMCs. To explore this further, the gelation time was also investigated as a function of CMCs concentration, as shown in Fig. 1D. At 0 wt% CMCs, HA alone (3.2 wt%) still formed a hydrogel with a gelation time of 28.7 ± 2.1 minutes. However, the addition of 0.9 wt% CMCs markedly shortened the gelation to 6.2 ± 0.6 minutes. Increasing the CMCs concentration further to 2.3 wt% caused a modest reduction, reaching the shortest gelation time of 2.8 ± 0.2 minutes. These findings indicated that the gelation time depended significantly on both CMCs and HA concentration, despite BDDE traditionally being recognised as a curing agent for HA-based cosmetic fillers. This phenomenon can be explained theoretically by the diverse reactivity of BDDE epoxy groups with different nucleophilic functionalities (such as hydroxyl, amino, and carboxymethyl groups) found on glucosamine units of CMCs. While reactions between BDDE and non-ionisable hydroxyl groups occurred predictably, thus yielding a linear decrease in gelation time with increased hydroxyl density, the presence of ionisable amino and carboxyl groups introduced complexity. The reactivity of these groups strongly depended on the environmental pH. As the HACM system was adjusted to pH 9.0, the amino groups were predominantly deprotonated, and carboxyl groups were ionised, enhancing the reactivity of the amine groups with BDDE and consequently amplifying the effect of CMCs concentration on the gelation time. Finally, the concentration-dependent effect of the BDDE crosslinker itself was evaluated, and as anticipated, the gelation time decreased proportionally with increasing BDDE concentration (Fig. 1E).

Subsequently, the water absorption capability of HACM hydrogels was evaluated to optimise swelling characteristics suitable for root canal filler applications. The swelling properties of different HACM formulations, with the concentrations of individual components clearly indicated in the inset of Fig. 1F, showed that decreasing concentration of HA resulted in a corresponding reduction in hydrogel swelling, attributable to the hydrophilic nature of HA, known for its excellent water-retention capacity.³⁰ Additionally, as both HA and CMCs actively participated in crosslinking reactions with BDDE, higher polymer concentrations consequently led to increased crosslinking density and, thus, diminished swelling. Conversely, at excessively high polymer concentrations, as observed in the HACM1 sample, the swelling was initially moderate within the first 15 minutes but increased markedly thereafter, ultimately reaching the highest value among the tested samples. This phenomenon likely arose from uneven crosslinking, causing structural heterogeneity and variable water uptake. Furthermore, in HA-rich hydrogels, the polymer network could become overly dense due to strong water affinity, slowing water penetration into the hydrogel core. Interestingly, the HACM2 sample exhibited notably lower swelling compared with HACM3 and HACM4, despite its relatively high polymer content, indicating the system reached a concentration plateau. This finding implied that HACM2 possessed a higher crosslinking density, subsequently confirmed by calculations presented in the ESI,† Table S1. Indeed, crosslinking densities calculated from swelling ratios indicated HACM2 had the highest density of 175.0 ± 14.2 nmol mL⁻¹, whereas HACM1 had the lowest at 23.4 ± 5.9 nmol mL⁻¹. The greater crosslinking density suggested improved mechanical properties, essential for soft and sensitive tissues such as dental pulp. Consequently, the hydrogel composed of 3.2 wt% HA, 2.3 wt% CMCs, and 4 wt% BDDE was considered optimal and designated as HACM in subsequent experiments. This optimised HACM formulation was then freeze-dried and its morphology characterised by SEM, as depicted in Fig. 1G. SEM imaging of the hydrogel cross-section revealed a homogeneous porous structure with predominantly circular and oval-shaped pores. Quantitative analysis using ImageJ software determined the pore sizes to range around 71.6 ± 7.8 μm, facilitating efficient cell migration and adhesion, thereby enhancing the suitability of HACM as a bioscaffold for dental pulp regeneration.

3.2. Self-healing and adhesive properties of the HACM hydrogel

To evaluate the self-healing capability of HACM hydrogels, a straightforward experiment was conducted by pairing two halves of separately prepared hydrogels stained with green and yellow food dyes, as illustrated in Fig. 2A. Following a period of physical contact, the two halves formed robust connections and maintained adhesion even under manual tension. These observations suggested strong interfacial adhesion, likely facilitated by abundant dipolar groups capable of interacting via hydrogen bonding and van der Waals forces, enabling the two hydrogel sections to effectively heal and integrate into a cohesive structure.

Fig. 2 Self-healing and tissue adhesion properties of the HACM hydrogel. (A) Self-healing capability of HACM. (B) Adhesion ability on different tissue types. (C) Mechanical evaluation of the HACM hydrogel using a basic shear lap test with weights ranging from 5–50 g.

Since tissue adhesion is a critical property that enables a root canal filler to remain in place during application, it is essential to assess its adhesion ability on various tissue types. In this study, adhesion tests were conducted using porcine adipose and muscle tissues as well as chicken liver and heart tissues. As shown in Fig. 2B, HACM demonstrated strong adhesion properties, enabling different tissue samples to firmly attach to the surface of a rubber glove.

The mechanism of adhesion can be attributed to similar interactions observed in the self-healing process, where functional groups on the hydrogel surface form bonds with the corresponding groups present on biological tissues. These interactions primarily involve hydrogen bonding and van der Waals forces. Additionally, covalent bonding may occur between unreacted epoxy groups of BDDE crosslinkers with functional groups present on tissue surfaces, further enhancing adhesion strength.

Similarly, the adhesion between two skin samples was evaluated. To preliminarily assess the mechanical durability of the HACM adhesive, standard weights of varying masses were applied to induce tensile stress on the hydrogel. Results indicated that HACM was able to withstand a tensile load of 50 g without showing signs of detachment or deformation. The strong adhesion observed is a crucial property for pulpitis therapy, as an effective root canal filler must remain securely in place to protect the damaged pulp tissue and maintain structural integrity.

3.3. Mechanical properties of the HACM hydrogel: shear thinning, lap shear, and rheology

To further evaluate the tensile strength of the HACM hydrogel system, a lap shear test was conducted using a tensile strength tester. This experiment was performed on a larger scale, where two porcine skin samples were cut into rectangular strips measuring 12 cm in length and 3 cm in width, with their thickness recorded at approximately 6 mm using a calliper. A volume of 200 μL of HACM hydrogel was applied onto the surface of one skin sample and evenly spread over an area of 2 cm × 3 cm before being pressed against the second skin sample. To enhance adhesion, a 50 g weight was placed on the overlapped region, ensuring firm contact. The sample was then clamped at both ends and subjected to tensile force in opposite directions, as illustrated in the inset of Fig. 3C. The tensile test results, averaged over three replicates, demonstrated that the maximum displacement before detachment was 17.8 ± 2.4 mm, with the ultimate tensile force recorded at 0.9 ± 0.2 N (Fig. 3C). Based on these findings, the shear strength of the HACM hydrogel was calculated to be 1575.1 ± 264.2 N m⁻², while the toughness of the material was determined to be 1.6 ± 0.3 kJ m⁻³.

Fig. 3 Qualitative evaluation of the mechanical properties of the HACM hydrogel. (A) Experimental setup for measuring the viscosity of the HACM mixture. (B) Relationship between viscosity and shear stress measured at different shear rates. (C) Illustration of the lap shear test setup using two porcine skin samples bonded by HACM hydrogel and its corresponding tensile strength. (D) Texture analyser setup and the effect of applied force on the three-dimensional structure of the hydrogel. Rheological characterisation of the HACM hydrogel under (E) frequency sweep, (F) strain sweep, (G) continuous flow, and (H) cyclic strain time sweep conditions. For the cyclic strain sweep, the shaded region indicates the application of high strain (500%), while the unshaded region represents the application of low strain (0.2%) in the rheometer setup.

Additionally, the rheological behaviour of the HACM hydrogel was examined to determine whether it exhibited Newtonian (constant viscosity) or non-Newtonian (shear-dependent viscosity) flow characteristics. The experimental setup is depicted in Fig. 3A, while the results are presented in Fig. 3B. The data revealed that an increase in shear rate was directly proportional to shear stress. However, according to Newton's law, the dynamic viscosity of the hydrogel decreased with increasing shear rate, indicating shear-thinning behaviour, a characteristic typical of injectable hydrogels designed for biomedical applications.

To gain deeper insights into the mechanical performance of HACM hydrogels, their rheological properties were evaluated under different conditions using a texture analyser, as illustrated in Fig. 3D. Initially, oscillatory frequency sweeps were conducted to determine the viscoelastic characteristics of the hydrogel. The results presented in Fig. 3E demonstrated that G′ (storage modulus) remained consistently higher than G′′ (loss modulus) across the frequency range, indicating a predominantly elastic, solid-like behaviour. However, within the frequency range of 10⁰ to 10¹, a crossover region was observed, suggesting a transition between solid- and liquid-like behaviour. For strain sweep analysis (Fig. 3F), a sharp decrease in G′ was evident at 10^0.5, indicating that the hydrogel network began to yield under increasing strain, resulting in a progressive loss of mechanical integrity. The observed G′ and G′′ crossover suggested that HACM hydrogels were not excessively rigid, making them suitable for injection-based applications. One of the most crucial characteristics of injectable hydrogels is their shear-thinning behaviour after crosslinking. A shear-thinning hydrogel exhibits a decrease in viscosity under applied shear stress, enabling it to flow smoothly through a syringe in a liquid-like state while rapidly recovering its solid-like structure once shear stress is removed. This behaviour was confirmed by the continuous flow experiment shown in Fig. 3G, which demonstrated that HACM retained its shear-thinning properties even in the hydrogel state. Similarly, the cyclic strain sweep experiment (Fig. 3H) further validated this property, showing a significant reduction in storage modulus during high-strain application, followed by its recovery under low-strain conditions. This rapid shift in flow state reinforced the self-healing capability and injectability of HACM hydrogels, confirming their potential for in situ gel formation and biomedical applications.

3.4. In vitro and in vivo injectability of HACM hydrogel

Following the preliminary assessment of the mechanical properties of the hydrogel, its injectability was investigated using both in vitro and in vivo models, as shown in Fig. 4A and B, respectively. In the in vitro experiment, the HACM hydrogel was extruded through a 26G needle into a silicone mold to evaluate its ability to pass through a syringe smoothly. Then, PBS solution was added to demonstrate the high stability of the shaped hydrogel upon injection. As expected, after extrusion, the hydrogel was capable of maintaining its shape, as evidenced by its successful moulding into a specific structure representing the acronym “KHUD” (an abbreviation for the Faculty of Applied Sciences, where the study was conducted). Notably, the hydrogel could be lifted out of the solution while preserving its gel-like consistency, further confirming its stability and shape retention following injection.

Fig. 4 Injectability and biodegradation of the HACM hydrogel in in vitro and in vivo models. Evaluation of injectability in the (A) in vitro model using PBS and (B) in vivo model in BALB/c mice. (C) Mass and morphology of hydrogel implants collected at 30 minutes, 1 week, and 2 weeks post-subcutaneous injection.

For the in vivo injectability assessment, a subcutaneous injection model was employed using BALB/c mice to evaluate the local gel-forming ability of HACM hydrogel. Following injection, a small subcutaneous bulge was observed on the dorsal region of the mice, indicating successful hydrogel deposition. After implantation, the animals exhibited no complications and maintained normal behavior, including regular food and water intake, with no signs of distress or discomfort associated with the implant. To assess the gel formation post-injection, mice were sacrificed after 30 minutes, and the implanted hydrogel was retrieved and examined. The results confirmed that HACM successfully formed a stable gel in situ without visible deformation. Further evaluations at 1 week and 2 weeks post-injection were conducted to investigate biodegradation, with the collected hydrogels analysed for mass loss. As illustrated in Fig. 4C, the hydrogel mass significantly decreased over time, with less than 50% of the initial weight remaining after 2 weeks, demonstrating the biodegradable nature of the HACM hydrogel, which is attributed to its biopolymer composition.

3.5. Release profiles of drug-loaded HACM hydrogels

For a root canal filler based on hydrogel materials, it is essential to exhibit controlled drug release to prolong therapeutic efficacy in pulpitis treatment. This function enables the sustained delivery of both antimicrobial agents and wound-healing compounds, ensuring effective infection control and tissue regeneration. The three-dimensional polymeric network, hydrophilic nature, and abundant surface functional groups of the hydrogel facilitate efficient drug loading and controlled release. As anticipated, the results in Fig. 5A and B confirmed the controlled release behaviour of the drug-loaded HACM hydrogels. In Fig. 5A, the release profile of AX-loaded HACM hydrogel exhibited a linear pattern over time, indicating stable drug release without an initial burst effect, achieving 86.8 ± 9.2% cumulative release after 8 days. Similarly, as shown in Fig. 5B, the EPO-loaded HACM hydrogel displayed a comparable release profile, with a cumulative release of 89.7 ± 8.1% at 8 days. Overall, both release profiles exhibited the characteristic controlled-release behaviour of polymeric hydrogels. To further investigate the release kinetics, established mathematical models were applied, with regression results summarised in Table S2 (ESI†). The best-fitting models were found to be zero-order kinetics and the Korsmeyer–Peppas model. The zero-order model describes a constant drug release rate over time, independent of the remaining drug concentration in the hydrogel, which aligns with the sustained release observed for both drug-loaded HACM hydrogels. The Korsmeyer–Peppas model, on the other hand, provided deeper insights into the release mechanism. The release constant (n) values were calculated as 1.0079 for HACM/AX and 0.9841 for HACM/EPO, both of which are relatively high. These values suggest that the primary drug release mechanism follows Super Case II transport.³¹ This release process is governed by polymer swelling and relaxation dynamics, rather than concentration gradient-driven diffusion. The observed behaviour can be attributed to the high water-retention capacity and strong hydrophilicity of hyaluronic acid-based hydrogels.¹⁹ Overall, the controlled release properties of HACM hydrogels offer significant advantages for pulpitis therapy, as they can enhance antimicrobial efficacy and promote wound healing by facilitating vascularisation and tissue regeneration.

Fig. 5 Release profiles of HACM loaded with (A) AX and (B) EPO. (C) Antimicrobial activity of HACM and HACM/AX evaluated using the disk diffusion method. (D) Measurement of the inhibition zone diameter for different hydrogel formulations.

3.6. Antibacterial properties of the HACM hydrogel

As previously discussed, pulpitis is primarily caused by bacterial invasion within the root canal, making the antimicrobial properties of the hydrogel a critical aspect of its therapeutic potential. In this study, the disk diffusion assay was employed to evaluate the antibacterial activity of HACM hydrogels. The bacterial strains B. subtilis and E. coli were selected as representative Gram-positive and Gram-negative species, respectively, commonly associated with oral infections.³² The results presented in Fig. 5C demonstrated that both HACM and HACM/AX hydrogels exhibited antibacterial effects, as indicated by the formation of inhibition zones surrounding the samples. The inhibition zone diameters, measured and summarised in Fig. 5D, revealed that HACM/AX exhibited superior antibacterial activity compared to the control, against both Gram-positive and Gram-negative bacteria. This highlights the advantage of the sustained release of antibiotics from the HACM hydrogel. This enhanced antibacterial efficacy was attributed to AX, a broad-spectrum β-lactam antibiotic widely used for various bacterial infections. Notably, the HACM hydrogel itself also displayed a mild antibacterial effect, likely due to the presence of CMCs. The positively charged amino groups of CMCs have been reported to disrupt bacterial membranes, leading to bacterial cell lysis and inhibition of growth.²⁶ Although HA does not exhibit direct antibacterial properties, its high water-retention capacity may contribute to bacterial repulsion, thereby limiting microbial adhesion and colonisation.³³ These findings suggest that HACM/AX is a promising alternative to conventional antibacterial materials, offering an effective means of simplifying root canal disinfection while simultaneously promoting tissue regeneration.

3.7. Biocompatibility of the HACM hydrogel

Biocompatibility assessments are among the most critical criteria for determining whether a biomaterial is suitable for clinical application. One key aspect of this evaluation is cell compatibility, which includes haemocompatibility testing. In this study, haemolysis assays were conducted to examine the interaction between HACM hydrogels and erythrocytes. As shown in Fig. 6A, the haemolysis rate remained below 2.5% across all tested concentrations, with no significant difference compared to the negative control (PBS). This observation was further supported by qualitative analysis presented in the inset of Fig. 6A. In samples containing HACM hydrogels, erythrocytes retained their cellular integrity and sedimented after centrifugation, indicating minimal interaction with the hydrogel. In contrast, the positive control (Triton X-100) caused complete erythrocyte lysis, resulting in the release of haemoglobin, which dispersed evenly throughout the solution, giving it a distinct red colour. These findings indicate that HACM hydrogel is haemocompatible, posing no significant risk to red blood cells. This haemocompatibility may be attributed to the highly anionic nature of HA, which likely counterbalances the positively charged amino groups of CMCs, thereby stabilising erythrocytes and preventing undesired membrane interactions.

Fig. 6 Biocompatibility evaluation of the HACM hydrogel. (A) Haemolysis activity assessment of HACM. (B) ROS induction levels in RAW 264.7 macrophages (statistical analysis: comparison between the control and other samples). (C) Fluorescence imaging and cell viability results from the LIVE/DEAD assay. Fluorescence intensity-based cell viability measurements of (D) MOVAS and (E) SVEC4-10 cells.

Next, the ROS levels were assessed in RAW 264.7 macrophages. ROS is a widely used biomarker for evaluating the immune response of cells to foreign materials.⁴ When compared to the control group, which consisted of cells maintained in an optimal culture medium, no significant differences were observed across HACM hydrogel samples at varying concentrations. In fact, most hydrogel-treated samples exhibited a slight reduction in ROS induction levels, which can be attributed to the high biocompatibility of polysaccharides. Additionally, hyaluronic acid HA, a primary component of the extracellular matrix, plays a key role in cell membrane structure.³⁴ The absence of macrophage stimulation further suggests that HACM hydrogels may reduce inflammation associated with foreign body responses when applied in vivo, particularly in root canal filler applications.³⁵

Subsequently, biocompatibility was evaluated using MOVAS and SVEC4-10 cells, which represent mouse smooth muscle cells and endothelial cells, respectively, as illustrated in Fig. 6C–E. Both cell lines exhibited healthy proliferation when cultured with HACM hydrogel at varying concentrations, as demonstrated by calcein-AM staining (green fluorescence) and cell morphology analysis. Notably, from a concentration of 250 μg mL⁻¹, both MOVAS and SVEC4-10 cells exhibited a marked increase in proliferation, which continued up to 1000 μg mL⁻¹, confirming the biocompatibility and proliferative potential of HACM hydrogel. Slightly better results were observed for HACM/EPO, as expected, given that EPO is a well-established cell proliferation stimulant.²⁶ In the assessed cell models, EPO promotes proliferation by binding to the EPO receptor, which subsequently activates the JAK2/STAT5, MAPK/ERK, or PI3K/Akt signalling pathways.^28,36 Concurrently, HA has been reported to engage with CD44 receptors, with minor involvement from the receptor for HA-mediated motility.³⁷ These interactions trigger downstream signalling cascades, ultimately enhancing cell survival and proliferation.³⁸ The observed proliferative effects of HACM hydrogels, both in the presence and absence of proliferative agents, underscore their potential applications in wound healing, angiogenesis, and soft tissue regeneration. These properties highlight the therapeutic suitability of HACM hydrogels for pulpitis treatment, where enhanced cell proliferation and tissue repair are essential for functional recovery.

3.8. M2 macrophage polarisation

M2 polarisation represents a crucial process through which macrophages adapt their functional roles in response to specific signals and factors derived from the microenvironment, transitioning from a pro-inflammatory to an anti-inflammatory phenotype.²⁹ This shift supports tissue repair and promotes immune homeostasis. The induction of the M2 phenotype also indirectly reflects the biocompatibility of biomaterials, implying that the material does not elicit detrimental immune responses, thereby facilitating its biomedical utility without immune interference. In this assessment, DCFDA was once again employed to evaluate the ROS production of RAW 264.7 macrophages incubated with HACM and HACM loaded with EPO. As illustrated in Fig. 7A, green fluorescence intensity was markedly higher in the control and HACM groups, whereas the EPO-loaded sample demonstrated reduced fluorescence, indicating lower ROS induction. Quantitative analysis using a microplate reader (Fig. 7C) confirmed this trend, with HACM/EPO reducing ROS levels to 60.0 ± 0.7 RFU × 10³, compared to 73.3 ± 4.9 RFU × 10³ in the control. This result aligned with prior expectations, given the well-established anti-inflammatory properties of EPO,³⁹ and suggested potential M2 polarisation associated with the material.

Fig. 7 ROS level and M2 macrophage polarisation. (A) and (C) Qualitative and quantitative results of ROS induction measured using DCFDA. (B) Fluorescence images obtained by staining with conjugated anti-CD163 antibodies and counterstaining with Hoechst. (D) Fluorescence intensity of Alexa Fluor 647-labelled antibodies. (E) Representative bright-field images illustrating the morphology of M0 and M2 macrophages.

To validate this hypothesis, CD163, a well-established M2 macrophage marker, was utilised via a conjugated primary antibody for ICC staining. As shown in Fig. 7B, anti-CD163 fluorescence intensity varied considerably across the groups after 48 hours of incubation. Notably, the cell morphology differed significantly: cells treated with plain media or HACM hydrogel primarily displayed the M0 phenotype with rapid proliferation. In contrast, the HACM/EPO-treated group exhibited lower cell density but a predominance of large spindeloid macrophages with elongated cell bodies and apical cytoplasmic extensions (Fig. 7E). Some also presented as multinucleated giant cells with extensive cytoplasmic projections, further reinforcing the evidence for M2 polarisation.⁴⁰ Measurement of FI at an excitation wavelength of 650 nm and an emission wavelength of 665 nm revealed that CD163 staining in the HACM/EPO group reached 19.2 ± 5.1 RFU × 10³, substantially higher than the control (4.6 ± 1.9 RFU × 10³), with the HACM-only sample showing no statistically significant difference. These findings collectively suggested that the material, when supplemented with EPO, substantially enhanced biological activity, implying its efficacy in promoting tissue regeneration and wound healing without triggering macrophage-mediated inflammation.

3.9. Proliferation of the HACM hydrogel

To further evaluate the proliferative potential of the HACM hydrogel, a proliferation assay was conducted by monitoring endothelial cell growth over time. The cell density progression after 24 hours, starting from an initially uniform seeding density, is illustrated in Fig. 8A, while the corresponding cell confluency measurements are presented in Fig. 8B. Compared to the control group, the HACM hydrogel at 1000 μg mL⁻¹ did not exhibit a statistically significant increase in cell proliferation. However, the overall cell coverage remained consistently higher over time. With an initial cell density of approximately 8.0 × 10³ cells per well, the HACM hydrogel-treated group reached a confluency of 68.8 ± 8.2%, whereas the control group only achieved 41.1 ± 9.3% confluency after 24 hours. Moreover, in the presence of HACM/EPO, which contains EPO with pro-angiogenic properties, both qualitative imaging observations and cell density measurements indicated a higher proliferative potential. Notably, cells cultured with HACM/EPO exhibited a significant increase in confluency from 12 hours onwards, with a marked rise before the end of the monitoring period, ultimately reaching 82.8 ± 3.7% confluency.

Fig. 8 Proliferative capacity of SVEC4-10 endothelial cells in response to HACM and HACM/EPO treatment. (A) Real-time imaging captured over 24 hours with three-dimensional morphology reconstruction. (B) Endothelial cell coverage under different treatments over time. (C) Hull convexity analysis of endothelial cells. (D) Distribution of eccentricity and cell area measurements.

To gain deeper insights into cell morphology, a detailed analysis was conducted to evaluate the effects of proliferation dynamics on cellular structure. As shown in the last column of Fig. 8A, three-dimensional reconstructions revealed that cells cultured under standard conditions tended to be rounder and more spindle-shaped, whereas those exposed to stimulator-containing environments (HACM and HACM/EPO) exhibited more elongated shapes with extended filopodia. Hull convexity measurements, which quantify the extent to which a cell's shape deviates from a perfect convex boundary, are presented in Fig. 8C.^41,42 A lower convexity value indicates greater morphological irregularities, which often arise from cellular adaptations to external stimuli, including changes in surface curvature.⁴³ Similarly, cell eccentricity and area, two key morphometric parameters, were assessed to further evaluate morphological changes, as illustrated in Fig. 8D. The flow cytometry plots of these parameters demonstrated that cells cultured in HACM/EPO exhibited a reduced cell area, with the majority concentrated below 500 μm², and an eccentricity distribution shifting towards 0.8. In contrast, control samples displayed a broader distribution of cell area, averaging approximately 750 μm². These findings reinforce previous observations regarding the morphological alterations induced by growth-promoting stimuli in the microenvironment. The results further indicate that HACM/EPO influences cell proliferation and structural adaptation, promoting SVEC4-10 cell growth and vascularisation, which are essential for effective wound healing.

3.10. Wound healing promoted by the HACM hydrogel

For dental pulp tissue, which is densely vascularised with small blood vessels, promoting angiogenesis is essential for nutrient supply and soft tissue regeneration. As previously discussed, the HACM hydrogel, with its three-dimensional network, serves as a biological scaffold that facilitates cell adhesion and tissue reconstruction. However, effective tissue regeneration requires not only cell proliferation but also cell migration, enabling the repopulation of damaged areas. To assess cell migration, an in vitro scratch assay was performed. The initial wound gap, manually introduced, is shown at 0 hours in Fig. 9A, with the scratch area horizontally aligned in the image. After 6 hours, the control group, cultured in standard medium, exhibited randomly dispersed cells with limited directional migration. In contrast, HACM-treated cells showed a more concentrated distribution along the scratch margins. Notably, cells treated with HACM/EPO demonstrated a uniform, well-organised cell layer, migrating evenly from both edges of the scratch. After 12 hours, differences in wound coverage became increasingly apparent. The HACM/EPO-treated group exhibited the highest closure rate, covering 95.7 ± 2.1% of the scratch area. This was followed by the HACM-treated group, which achieved 73.7 ± 4.1% coverage, while the control group exhibited the lowest wound closure, with only 58.7 ± 12% coverage.

Fig. 9 Migratory capacity of endothelial cells cultured in HACM and HACM/EPO-enriched environments. (A) Real-time wound healing images captured over 12 hours, along with three-dimensional morphology of representative cells under different culture conditions. (B) Quantification of cell coverage measured from the scratch assay. (C) Convexity analysis of SVEC4-10 cells following culture in activation-inducing environments. (D) Distribution of eccentricity and cell area in angiogenic cells.

The real-time quantification of cell coverage area was measured hourly, as shown in Fig. 9B and the videos attached in the ESI,† to determine the time required to achieve 50% wound closure. The control group reached this threshold at 11.95 hours, while the HACM-treated group exhibited a slightly reduced closure time of 10.19 hours. In contrast, the HACM/EPO-treated group demonstrated a substantially faster migration rate, achieving 50% coverage within 6.76 hours. In line with the proliferation assay, additional morphometric parameters were assessed once cells reached full confluency (100% coverage). As illustrated in Fig. 9C, cell convexity was significantly reduced following HACM hydrogel treatment, with the effect being more pronounced in the presence of EPO. These findings were further corroborated by eccentricity and cell area measurements, presented in Fig. 9D. Specifically, eccentricity values exhibited a notable shift towards 1, while the cell area was markedly reduced, displaying distinct separation between treated and control groups. These morphological alterations indicate a cellular response to stimulatory factors in the culture environment, leading to the formation of irregular, branched, or protrusive shapes, as observed in the three-dimensional reconstructions in Fig. 9A.⁴² Such morphological adaptations are typically associated with enhanced migratory activity, cellular spreading, and angiogenic behaviour, suggesting that stimulator-induced migration follows a distinct pathway compared to spontaneous cell proliferation. Thus, in addition to promoting endothelial cell proliferation, HACM/EPO exhibits strong pro-angiogenic properties, facilitating the migration and interconnection of vascular cells. This capability is expected to support neovascularisation, thereby enhancing wound healing efficiency.

3.11. Chick embryo chorioallantoic membrane assay

The results of the CAM assay provided insight not only into the biosafety of the test materials but also into their capacity to promote neovascularisation in the presence of bioactive stimuli.⁴⁴ In this study, both parameters were evaluated concurrently using HACM hydrogel co-loaded with AX and EPO, aiming to deliver a comprehensive assessment of the dual-drug system. As illustrated in Fig. S1 (ESI†), fertilised eggs exposed to HACM maintained normal embryonic development, with the hydrogel remaining stable atop the chorioallantoic membrane. By day 12, the embryos demonstrated healthy development, including embryonic repositioning, indicating no adverse effects attributable to HACM exposure. Similarly, embryos in the HACM/AX/EPO group also exhibited normal morphology and growth throughout the observation period, suggesting that the co-administration of AX and EPO had no detrimental impact on embryogenesis.

In addition, a notable increase in vascular density was observed, particularly in the HACM/AX/EPO group. Quantitative analysis of fold change in vascular area (Fig. S2, ESI†) revealed minimal change across all groups by day 10. However, by day 12, the HACM/AX/EPO group demonstrated a marked increase in vascularisation, reaching 171.5 ± 71.6%, while the HACM-only group exhibited a moderate enhancement at 81.2 ± 30.7%. In contrast, the control group treated with saline showed no abnormalities but only a modest increase in vascular area (12.2 ± 12.0%). These results confirmed that the angiogenic potential of EPO remained effective when co-delivered with AX and further supported the overall biocompatibility of the hydrogel formulation in the in ovo model.

4. Conclusions

In this study, we first aimed to synthesise and optimise an injectable hydrogel as a potential replacement for conventional materials used in pulpitis therapy, facilitating angiogenesis and promoting the repair of damaged soft tissue. The developed hydrogel was derived from two naturally occurring polysaccharides, hyaluronic acid and carboxymethyl chitosan, and was chemically crosslinked using 1,4-butanediol diglycidyl ether. To summarise, the HACM hydrogel system was optimised based on gelation time, shear-thinning characteristics, and swelling ratio to ensure in situ gel formation while preventing excessive swelling. The optimised formulation, comprising 3.2 wt% HA, 2.3 wt% CMCs, and 4 wt% BDDE, was selected for subsequent experiments. The hydrogel demonstrated excellent adhesion to various tissue types and exhibited self-healing properties. Further mechanical strength assessments confirmed its shear-thinning behaviour, a key characteristic for injectability, and lap shear testing revealed a fracture strength of approximately 0.9 N. The hydrogel's injectability was successfully demonstrated in both in vitro and in vivo models, where it also exhibited favourable biodegradation characteristics. The biocompatibility of HACM was validated through tests on erythrocytes, macrophages, muscle cells, and endothelial cells. Additionally, the hydrogel's ability to encapsulate and sustain the release of bioactive compounds (AX) conferred broad-spectrum antibacterial activity against B. subtilis and E. coli. Notably, the HACM hydrogel loaded with EPO showed significant pro-angiogenic effects, enhancing endothelial cell proliferation and cell migration, indicating its potential for tissue repair via rapid vascularisation. Moreover, the macrophage polarisation towards the M2 phenotype, alongside findings from the CAM assay, further substantiated the material's capacity to promote tissue regeneration and stimulate neovascularisation. In future studies, the further optimisation of hydrogel properties may be achieved by modifying or incorporating alternative material components. This approach aims to tailor the mechanical characteristics not only to suit soft tissues such as dental pulp, but also to broaden the applicability of the hydrogel for the regeneration of more structurally demanding hard tissues. Overall, these findings underscore the therapeutic potential and suitability of the HACM hydrogel for treating nerve and soft tissue damage associated with bacterial-induced pulpitis, reinforcing its applicability in dental pulp regeneration.

Author contributions

Nhu-Quynh Thi Nguyen: writing – review & editing, methodology, investigation, formal analysis, data curation, conceptualization. Cuong Hung Luu: writing – original draft, methodology, investigation, formal analysis, data curation, conceptualization. Ngoc-Dan Ho Nguyen: investigation, formal analysis, data curation. Duy Quoc Vo: investigation, formal analysis, data curation. Ngoc Hong T. Luu: investigation, formal analysis. Nhu-Y Ngoc Ha: investigation, formal analysis. Thanh-Truc Ngoc Vo: investigation. Que-Phuong Le Huynh: data curation. Thavasyappan Thambi: writing – review & editing. Chau T. T. Ngo: supervision, methodology, conceptualization, administration. Giang V. H. Phan: supervision, methodology, conceptualization, administration.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

This work is supported by Ton Duc Thang University. It is also supported by the Queensland node of the Australian National Fabrication Facility, Griffith University (ANFF).

References

1. B. A. Dye, J. Dent. Res., 2017, 96, 361–363.
2. L. F. Donaldson, J. Physiol., 2006, 573, 2–3.
3. C. Zhan, M. Huang, X. Yang and J. Hou, Int. Endod. J., 2021, 54, 85–99.
4. C. M. Phan, C. H. Luu, M. Murugesan, T.-N.-Q. Nguyen, N.-Y. N. Ha, H. L. Ngo, N.-D. H. Nguyen, Z. Pan, V. H. G. Phan, Y. Li and T. Thambi, Int. J. Biol. Macromol., 2025, 284, 137939.
5. J. Baginska and W. Stokowska, Med. Princ. Pract., 2013, 22, 555–560.
6. B. F. Aliev and L. A. Gardashova, Procedia Comput. Sci., 2017, 120, 539–546.
7. R. Bhat, S. Shetty, P. Rai, B. K. Kumar and P. Shetty, J. Oral Biosci., 2024, 66, 272–280.
8. D. R. Nixdorf, E. J. Moana-Filho, A. S. Law, L. A. McGuire, J. S. Hodges and M. T. John, J. Endod., 2010, 36, 224–230.
9. R. Prakash, V. Gopikrishna and D. Kandaswamy, Endodontology, 2005, 17, 32–36.
10. V. Vishwanath and H. M. Rao, J. Conservative Dent. Endod., 2019, 22, 216–222.
11. H. Li, D. Zhang, P. Bao, Y. Li, C. Liu, T. Meng, C. Wang, H. Wu and K. Pan, ACS Nano, 2024, 18, 16395–16412.
12. P. L. Thi, T. Y. N. Tran, H. C. Luu, D. L. Tran, T. T. H. Thi and D. H. Nguyen, J. Bioact. Compat. Polym., 2022, 37, 252–266.
13. V. H. G. Phan, M. Murugesan, P. Manivasagan, T. L. Nguyen, T.-H. Phan, C. H. Luu, D.-K. Ho, Y. Li, J. Kim, D. S. Lee and T. Thambi, Pharmaceutics, 2022, 14, 709.
14. V. H. G. Phan, T. Thambi, B. S. Kim, D. P. Huynh and D. S. Lee, Biomater. Sci., 2017, 5, 2285–2294.
15. C. H. Luu, G. Nguyen, T.-T. Le, T.-M. N. Nguyen, V. H. Giang Phan, M. Murugesan, R. Mathiyalagan, L. Jing, G. Janarthanan, D. C. Yang, Y. Li and T. Thambi, Gels, 2022, 8, 246.
16. Z. Xie, W. Jiang, H. Liu, L. Chen, C. Xuan, Z. Wang, X. Shi, Z. Lin and X. Gao, Adv. Healthcare Mater., 2024, 13, 2303709.
17. V. H. G. Phan, T. M. D. Le, G. Janarthanan, P.-K. T. Ngo, D. S. Lee and T. Thambi, J. Ind. Eng. Chem., 2021, 96, 345–355.
18. S. Song, J. Zhang, Y. Fang, W. Li, H. Zeng, Z. Fang, T. Wang, Y. Xie, C. Liu and J. Wang, OR, 2025, 1.
19. V. H. G. Phan, M. Murugesan, P. P. T. Nguyen, C. H. Luu, N.-H. H. Le, H. T. Nguyen, P. Manivasagan, E.-S. Jang, Y. Li and T. Thambi, Colloids Surf., B, 2022, 219, 112859.
20. C. B. Highley, G. D. Prestwich and J. A. Burdick, Curr. Opin. Biotechnol, 2016, 40, 35–40.
21. J. Cao, P. Wu, Q. Cheng, C. He, Y. Chen and J. Zhou, ACS Appl. Mater. Interfaces, 2021, 13, 24095–24105.
22. Z. Shariatinia, Int. J. Biol. Macromol., 2018, 120, 1406–1419.
23. G.-M. Lemnaru, R. D. Truşcă, C.-I. Ilie, R. E. Ţiplea, D. Ficai, O. Oprea, A. Stoica-Guzun, A. Ficai and L.-M. Diţu, Molecules, 2020, 25, 4069.
24. A. Huttner, J. Bielicki, M. N. Clements, N. Frimodt-Møller, A. E. Muller, J. P. Paccaud and J. W. Mouton, Clin. Microbiol. Infect., 2020, 26, 871–879.
25. L. Bai, R. L. Reis, X. Chen, J. Su and C. Liu, OR, 2025, 1.
26. P.-K. T. Ngo, D. N. Nguyen, H.-P. Nguyen, T.-H. H. Tran, Q.-N. D. Nguyen, C. H. Luu, T.-H. Phan, P. K. Le, V. H. G. Phan, H. T. Ta and T. Thambi, Int. J. Biol. Macromol., 2024, 279, 135329.
27. Y. Li, H. Peng, W. Tang, D. Gu, S. Ren, Y. Yu, J. Yang and L. Miao, Mater. Des., 2023, 225, 111540.
28. M. Murugesan, R. Mathiyalagan, Z. M. Ramadhania, J. Nahar, C. H. Luu, V. H. G. Phan, D. C. Yang, Q. Zhou, S. Chan Kang and T. Thambi, Bioact. Mater., 2025, 49, 154–171.
29. S. Chen, A. F. U. H. Saeed, Q. Liu, Q. Jiang, H. Xu, G. G. Xiao, L. Rao and Y. Duo, Signal Transduction Targeted Ther., 2023, 8, 207.
30. Y. Li, V. H. Giang Phan, Z. Pan, X. Xuan, H. Y. Yang, C. H. Luu, T.-H. Phan, T. M. D. Le and T. Thambi, Eur. Polym. J., 2024, 220, 113426.
31. S. Dash, P. N. Murthy, L. Nath and P. Chowdhury, Acta Pol. Pharm., 2010, 67, 217–223.
32. S. Alghamdi, Saudi J. Biol. Sci., 2022, 29, 318–323.
33. C. H. Luu, N.-T. Nguyen and H. T. Ta, Adv. Healthcare Mater., 2024, 13, 2301039.
34. M. Dovedytis, Z. J. Liu and S. Bartlett, Eng. Regener., 2020, 1, 102–113.
35. T.-H. H. Tran, C. H. Luu, K.-T. T. Nguyen, M.-A. L. Hoang, Q.-K. Pham, C. M. Phan, N.-K.-L. Thai, H. T. Nguyen, T. Thambi and V. H. G. Phan, Macromol. Biosci., 2025, 25, 2400351.
36. D. Kuhrt and D. M. Wojchowski, Blood, 2015, 125, 3536–3541.
37. C. X. Yang, Y. W. Liu, Y. Q. He and F. Gao, Fenzi Xibao Shengwu Xuebao, 2006, 39, 495–501.
38. K. C. M. L. Elvitigala, W. Mubarok and S. Sakai, Biomolecules, 2024, 14, 604.
39. I. Silva, C. Alípio, R. Pinto and V. Mateus, Biomed. Pharmacother., 2021, 139, 111558.
40. F. Heinrich, A. Lehmbecker, B. B. Raddatz, K. Kegler, A. Tipold, V. M. Stein, A. Kalkuhl, U. Deschl, W. Baumgärtner, R. Ulrich and I. Spitzbarth, PLoS One, 2017, 12, e0183572.
41. A. J. Mclaughlin, A. J. Kaniski, D. I. Matti and B. Xhabija, Anticancer Res., 2023, 43, 3891–3896.
42. J. Lobo, E. Y.-S. See, M. Biggs and A. Pandit, J. Tissue Eng. Regener. Med., 2016, 10, 539–553.
43. S. Yadav, R. Joshi and A. Majumder, Langmuir, 2024, 40, 17590–17600.
44. V. H. G. Phan, B.-P. T. Nguyen, N. Y. Nguyen, C.-N. D. Tran, Q.-N. D. Nguyen, C. H. Luu, P. Manivasagan, E.-S. Jang, D. C. Yang, D. U. Yang, Y. Li, J. Conde and T. Thambi, Int. J. Biol. Macromol., 2025, 308, 142254.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tb00559k

‡ These authors contributed equally.

---