Biodegradable polyethylene glycol-based poly(β-amino ester) hydrogel for curcumin delivery in wound treatment

Abstract

This report focused on synthesizing a biodegradable material based on poly(β-amino ester) hydrogel with curcumin loaded for application as a wound dressing treatment, as well as examining other physicochemical and physiological properties. The hydrogel was prepared via aza-Michael addition between cystamine and poly(ethylene glycol) diacrylate with and without curcumin loaded during the process to improve bioactivity. Physicochemical analysis showed that the hydrogel had a stable structure and good swelling ability. SEM results indicated that the synthesized gel had little to no pores, suggesting that curcumin is released through the degradation of the hydrogel. Rheological analysis demonstrated that increasing the cross-linking density significantly enhanced the mechanical strength and gelation performance of the hydrogels. All formulations displayed typical elastic gel behavior (G′ > G″), while highly cross-linked hydrogels formed more rigid but less deformable networks with accelerated gelation and improved structural stability. The 0.1% curcumin-loaded poly(β-amino ester) hydrogel exhibited the highest antioxidant activity and was confirmed to be non-hemolytic, have good cytocompatibility and be non-irritating to the mouse skin. In a murine wound model, the hydrogel promoted wound contraction and remarkable restoration of skin architecture such as increased epidermal thickness and hair follicle number, as evidenced by histological analysis. These findings suggest that the curcumin-loaded poly(β-amino ester) hydrogel is a promising drug delivery system, opening new applications in wound dressing for physiological wound healing in the future.

This report focused on synthesizing a biodegradable material based on poly(β-amino ester) hydrogel with curcumin loaded for application as a wound dressing treatment, as well as examining other physicochemical and physiological properties.

Introduction

Hydrogels were among the first biomaterials developed, possessing a porous polymeric network with high water content, tunable mechanical properties, and a molecular diffusion capacity that closely mimic the extracellular matrix, making them particularly attractive for wound dressing and tissue engineering applications.^1–3 Thanks to a three-dimensional network formed by cross-linked polymer chains, hydrogels are capable of absorbing and retaining large amounts of water or biological fluids through hydrophilic groups such as –NH₂, –COOH, and –OH.^4,5 Beyond their swelling capability, hydrogels also possess molecular diffusion capabilities through the polymer network and mechanical strength to withstand compressive or tensile forces.^6,7 Depending on the crosslinking mechanism, hydrogels can be classified as physically or chemically crosslinked systems, which correlates to stimuli responsive designs based on pH, temperature, or redox conditions, enabling controlled drug delivery tailored to pathological microenvironments.^8,9

Among various hydrogel forming polymers, poly(β-amino ester) (PBAE), a polymer synthesized via the Michael addition reaction between an amine and an acrylate, attracted significant interest due to its flexible structure.^10–12 High biocompatibility, biodegradability, and pH sensitivity, making it an ideal material for drug delivery and tissue engineering.^13–15 The pH-sensitive nature of PBAE hydrogels stems from the presence of tertiary amine groups, which induce a transition from a hydrophilic to a hydrophobic state as pH increases from 6.4 to 7.8.^16,17 This physiological transition under mildly acidic conditions can act as an important feature in chronic wounds management, including diabetic ulcers, where local pH is dysregulated.^18–20

Curcumin, a natural polyphenol extracted from turmeric, is a promising agent for wound healing due to its potent anti-inflammatory and antioxidant properties.^21–23 Curcumin inhibits pro-inflammatory cytokines such as tumor necrosis factor-alpha, interleukin-1, and inflammation-related transcription factors such as nuclear factor-kappa B.^24–26 Simultaneously, it acts as an effective antioxidant by neutralizing reactive oxygen species free radicals and activating endogenous antioxidant enzymes, particularly relevant in diabetic wounds, where excessive oxidative stress and prolonged inflammation impair tissue regeneration.²⁷ Regarding mechanisms, curcumin promotes the expression of growth factors like transforming growth factor beta 1 and vascular endothelial growth factor, helping enhance collagen deposition and neovascularization, thereby shortening the inflammatory phase and accelerating recovery speed.^24,27 However, the clinical application of curcumin is limited by its extremely low water solubility, instability, and low bioavailability.^28–30 Recent studies have demonstrated that curcumin-loaded hydrogels can significantly enhance wound healing by reducing inflammation, scavenging reactive oxygen species, and promoting angiogenesis and collagen deposition.^31–33

Despite the extensive development of hydrogel-based drug delivery systems for wound healing, several challenges remain, including limited drug loading of hydrophobic compounds, uncontrolled burst release, and insufficient responsiveness to the complex microenvironment of chronic wounds. In particular, the poor aqueous solubility and instability of curcumin often lead to heterogeneous dispersion and reduced therapeutic efficacy when incorporated into conventional hydrogel matrices.^34–36 Therefore, the design of multifunctional hydrogel systems capable of improving curcumin dispersion while enabling controlled and stimulus-responsive drug release remains highly desirable. In this context, the integration of hydrophilic poly(ethylene glycol) diacrylate (PEGDA) with PBAE networks provides a promising strategy to address these limitations. The hydrophilic PEG chains can create a hydrated microenvironment that facilitates the dispersion of hydrophobic molecules such as curcumin,^37,38 while the PBAE backbone introduces pH-responsive behavior that is relevant to the acidic microenvironment often observed in chronic wounds.^39–41 Furthermore, the incorporation of cystamine as a diamine crosslinker introduces redox-responsive disulfide bonds, which can undergo cleavage in the presence of intracellular reducing agents such as glutathione, thereby enabling controlled degradation of the hydrogel network and enhanced drug release.^42,43 This combination of hydrophilic dispersion, pH responsiveness, and redox-triggered degradation provides a multifunctional platform for improving curcumin delivery in wound healing applications.⁴⁴

Therefore, the aim of this study was to develop a biodegradable PEG-based PBAE hydrogel loaded with curcumin for potential application in acute wound healing. The proposed hydrogel system is designed to enhance curcumin dispersion, provide controlled and stimulus-responsive drug release, and create a moist and protective microenvironment conducive to tissue regeneration.

Results and discussion

Synthesis of hydrogel

Cystamine dihydrochloride was neutralized sodium bicarbonate solution at a molar ratio of 1 : 2. Due to its poor solubility in water, curcumin was initially dissolved in PEGDA 575 to prevent precipitation in the aqueous reaction medium. The resulting curcumin-PEGDA solution was clear and yellow in appearance. Upon mixing with the cystamine-NaHCO₃ solution, the color changed to orange, which may be attributed to the alkaline pH of the reaction environment. The mixture was then incubated at 37 °C for 2 hours under static conditions, leading to the formation of a solid, orange-colored hydrogel, as shown in Fig. 1 and Scheme S1. Hydrogel formulations of varying curcumin concentrations (0.0%, 0.1%, 0.5%, and 1.0%) were prepared and compared, as shown in Table 1 and Fig. S1. For the first step, curcumin was dispersed in PEGDA. The curcumin-free sample (PBAE-CUR-00) appeared a transparent, colorless solution, whereas curcumin-loaded samples exhibited a transparent orange-yellow color, with increasing intensity corresponding to higher curcumin concentrations. Upon mixing with the cystamine – NaHCO₃ solution, curcumin-loaded samples turned dark red-orange, reflecting the pH-dependent structural transition of curcumin. After 2 hours of incubation at 37 °C, PBAE-CUR-00 formed a translucent, homogeneous hydrogel block with gel fraction of 88.7% (Table S1). The PBAE-CUR-01 formulation produced a stable, solid, and orange-yellow hydrogel with homogenous color distribution, indicating effective dispersion of curcumin into the polymer network. In contrast, PBAE-CUR-05 and PBAE-CUR-10 exhibited heterogeneity, with the latter showing intense sedimentation of curcumin particles, likely due to hydrophobicity and low solubility. At higher concentrations, curcumin overwhelmed the dispersion capacity of the hydrogel, thereby self-aggregating, reducing homogeneity.^34–36

Fig. 1. Synthesis process of PBAE hydrogel loaded with curcumin.

Table 1. Components for synthesizing curcumin-loaded PBAE hydrogels.

Gel samples	Components
	Cys (mg)	PEGDA (mg)	NaHCO3 (mg)	Cur (mg)	H2O (µL)
PBAE-CUR-00	58.55	241	43.68	0.00	600
PBAE-CUR-01	58.55	241	43.68	0.96	600
PBAE-CUR-05	58.55	241	43.68	4.86	600
PBAE-CUR-10	58.55	241	43.68	9.91	600

Characterization

The FT-IR spectra of the PBAE hydrogels with different curcumin loadings (PBAE-CUR-00, PBAE-CUR-01, PBAE-CUR-05, and PBAE-CUR-10) are presented in Fig. 2. All samples exhibit a broad absorption band around 3550 cm⁻¹, which is attributed to the stretching vibrations of O–H and N–H groups, indicating the presence of hydroxyl and amine functionalities within the hydrogel network. The bands observed at 2900 and 2866 cm⁻¹ correspond to aliphatic C–H stretching vibrations of the PEG and PBAE backbone. A strong peak at 1731 cm⁻¹ is assigned to the C O stretching vibration of ester groups. Additionally, the characteristic band at 1110 cm⁻¹ corresponds to C–O–C stretching vibrations of ether groups in PEG chains.^45,46

Fig. 2. FT-IR spectra of PBAE hydrogel samples.

The formation of the PBAE hydrogel was further confirmed by ¹³C CP/MAS NMR spectroscopy (Fig. 3). The complete disappearance of the unsaturated C C resonances at 130–134 ppm indicates indicates the complete consumption of the acrylate double bonds. The successful formation of the PBAE network is evidenced by the characteristic ester carbonyl signal at 175.9 ppm (C O), together with the polyethylene glycol backbone resonances at 66.6, 71.7, and 73.1 ppm. Signals at 35.1 and 52.0 ppm are assigned to the methylene carbons of the cystamine backbone. Furthermore, the appearance of new resonances at 38.7 and 56.0 ppm, corresponding to the β-amino ester linkages formed via the aza-Michael addition reaction, provides direct evidence for the successful synthesis of the PBAE hydrogel.⁴⁴

Fig. 3. ¹³C CP/MAS NMR spectrum of PBAE-CUR-00.

The thermal gravimetric analysis (TGA) and thermogravimetric derivative (dTG) results of PBAE hydrogels showed that in the initial stage (below 100 °C), the sample lost about 3% of its original mass mainly due to the evaporation of water and residue solvent in the hydrogel structure (Fig. 4). Most of the mass loss occurred between 220 and 400 °C, with two major decomposition peaks at approximately 350 and 390 °C, accounting for more than 80% of the total mass loss. The first decomposition peak at approximately 350 °C was comparable among the hydrogel samples and may be associated with the thermal degradation of relatively labile linkages, particularly ester and disulfide bonds within the PBAE network. Previous studies have reported the thermal degradation of ester-containing PBAEs and demonstrated the relatively lower thermal stability of polymers containing disulfide linkages.⁴⁷ In contrast, the intensity of the second decomposition peak at approximately 390 °C varied considerably with CUR content. Hydrogels containing higher CUR contents (PBAE-CUR 05) exhibited a more pronounced decomposition peak in this region. This behavior is consistent with the reported thermal decomposition of CUR, which occurs predominantly in the 250–400 °C range,⁴⁸ suggesting that the enhanced peak at approximately 390 °C is likely associated with the thermal decomposition of the incorporated CUR. At higher temperature (400–800 °C), the TGA curve did not show any significant changes which hinting the presence of thermally stable inorganic compounds such as NaCl in the sample. These parts account for 5–20% of the remaining mass. Overall, the TGA results indicate that PBAE-CUR hydrogels remain thermally stable under physiological and processing conditions but undergo substantial degradation at elevated temperatures, consistent with the behavior of biodegradable polymer networks.

Fig. 4. Results of thermogravimetric analysis (TGA) and thermogravimetric derivative (dTG) of the gel samples.

The morphology of PBAE-CUR hydrogel samples was evaluated by scanning electron microscopy as shown in Fig. 5. At a resolution of 50 µm corresponding to magnifications of 5000. The samples were freeze-dried to maintain its structure before analysis. It can be seen that the gel surface appears to be quite dense with a rough, uneven matrix structure without large pores and did not reveal any clear porous structures. Such a compact structure suggests that most of the active compound was predominantly retained inside the polymer network rather than diffusing freely through interconnected pores. The drug release is mainly governed by hydrogel swelling and surface erosion/degradation as previously reported.^49,50 Upon contact with an aqueous medium such as PBS or physiological fluids, the hydrogel absorbs water and begins to swell. The dense network structure also explains why the material tendency to undergo swelling-induced rupture in PBS. A dense network may not be flexible enough to accommodate a large amount of water when swollen which leads to high internal osmotic pressure and subsequent structural rupture. Futher, the EDS mapping results confirmed the elemental composition and the homogeneous distribution of the constituent elements throughout the hydrogel.

Fig. 5. SEM-EDS annalysis of curcumin-loaded PBAE hydrogels.

The strain sweep results for the four hydrogel samples reveal a distinct differentiation in their load-bearing capacity and structural network integrity (Fig. 6). Samples PBAE-CUR-05 and PBAE-CUR-10 continue to demonstrate superior rigidity, with storage moduli (G′) maintained at significantly high levels (ranging from 1000 to nearly 5000 Pa) within the linear viscoelastic region values tens of times higher than those of the PBAE-CUR-00 and PBAE-CUR-01 groups. However, the data also indicates that these highly cross-linked gel systems tend to be more brittle, as evidenced by the abrupt structural collapse (a sharp decline in G′) at a relatively low critical strain threshold (approximately 0.2–0.4%). In contrast, while samples PBAE-CUR-00 and PBAE-CUR-01 exhibit lower mechanical moduli, they maintain structural stability up to higher strain levels (approximately 1%) before yielding. The sudden drop of G′ below the loss modulus (G″) at high strain regions across all samples serves as evidence of the cross-linked network's rupture, causing the material to transition from an elastic gel state to a viscous flow state. Overall, these findings confirm that increasing the concentration or cross-linking from PBAE-CUR-00 to PBAE-CUR-10 significantly enhances gel stiffness but simultaneously reduces the material's toughness and deformability.

Fig. 6. Frequency sweep rheological analysis of curcumin-loaded PBAE hydrogels.

The frequency sweep results for the hydrogel samples demonstrate a strong correlation with the previously analyzed gelation time data (Fig. 7). Within the investigated frequency range (0.01 to 100 Hz), all samples exhibited the characteristics of a true gel network, with the storage modulus (G′) maintaining a complete dominance over the loss modulus (G″). However, a clear divergence in mechanical strength was observed between the two sample groups: while PBAE-CUR-00 and PBAE-CUR-01 formed soft gel systems with stable G′ values at a lower range (approximately 100–200 Pa), PBAE-CUR-05 and PBAE-CUR-10 displayed significantly higher G′ values reaching nearly 2000 Pa, representing a tenfold increase in stiffness. Notably, in the high-frequency region (above 2 Hz), all four samples particularly the rigid PBAE-CUR-05 and PBAE-CUR-10 gels showed an abrupt decline or significant fluctuations in the moduli. This phenomenon reflects the structural durability threshold, where cross-links begin to rupture or wall slip occurs under the influence of rapid oscillations. Overall, the frequency sweep results confirm that adjusting the composition from PBAE-CUR-00 to PBAE-CUR-10 not only modulates the reaction rate but also allows for flexible control over the mechanical strength of the resulting hydrogels, ranging from soft to highly rigid networks.

Fig. 7. Strain sweep rheological behavior of curcumin-loaded PBAE hydrogels.

Analysis of the rheological profiles for hydrogel samples PBAE-CUR-00 through PBAE-CUR-10 reveals a distinct transition in both gelation kinetics and network strength (Fig. S2). PBAE-CUR-00 exhibited the slowest gelation process, with the crossover point occurring late at 1488.37 s, resulting in a weak gel system with a storage modulus (G′) below 1 Pa. Conversely, samples PBAE-CUR-01 to PBAE-CUR-10 showed signs of pre-gelation during the sample preparation or loading phase, evidenced by elevated and closely aligned G′ and G″ values from the onset of measurement (t = 0). Specifically, PBAE-CUR-01 reached a gel state rapidly within 400–450 s. However, its structure became unstable, showing a decline in mechanical properties after 1400 s. In contrast, PBAE-CUR-05 and PBAE-CUR-10 displayed more complex behavior: following an initial fluctuating phase likely due to the partial disruption of early formed structures under shear forces both samples underwent robust restructuring starting after 800–1000 s. The final results demonstrate that PBAE-CUR-05 and PBAE-CUR-10 formed significantly more durable hydrogel networks, with G′ values reaching approximately 100 Pa. This represents a more than 100-fold increase compared to PBAE-CUR-00 and PBAE-CUR-01, highlighting the highly effective cross-linking capability of these two formulations.

Hydrogel properties

Water content and swelling capacity are critical parameters governing hydrogel performance in biomedical applications, particularly in wound dressing and drug delivery systems. High water content enhances flexibility, facilitates mass transport, and supports volume expansion upon hydration. In the present study, the curcumin-free hydrogel (PBAE-CUR-00) exhibited the highest water content (∼70%) and correspondingly the greatest swelling ratio (Fig. S3 and S4). In contrast, curcumin-loaded formulations (PBAE-CUR-01, PBAE-CUR-05, and PBAE-CUR-10) showed progressively lower water content and a gradual reduction in swelling capacity, with swelling decreasing by approximately 20–40% as curcumin concentration increased. This trend suggests that curcumin incorporation modifies the internal network structure of the hydrogel. The presence of hydrophobic curcumin likely increases effective network compactness through enhanced hydrophobic interactions and possible changes in crosslinking density, thereby reducing intermolecular spacing and limiting water penetration. As a result, both equilibrium water uptake and volumetric expansion are suppressed. The increasing standard deviation observed from PBAE-CUR-00 to PBAE-CUR-10 further indicates greater structural heterogeneity at higher curcumin loadings, consistent with the morphological irregularities observed in SEM analysis. Although PEG-based hydrogels are typically characterized by high water content (>70%) due to strong hydrophilicity, the incorporation of PBAE segments and hydrophobic curcumin modulated this property, reflecting a balance between hydrophilic PEG domains and the more compact crosslinked network. The direct correlation between water content and swelling observed in this study aligns with previous findings by Stocke et al. (2017),⁵¹ who reported that swelling behavior is closely related to the density of intermolecular interactions within the gel; lower bonding density facilitates greater expansion. Overall, increasing curcumin concentration reduced both water retention and swelling capacity, primarily due to its hydrophobic nature and its influence on network compactness. This tunable behavior provides a means to optimize hydrogel mechanical stability and drug release characteristics according to specific wound healing requirements.

It is crucial to investigate the degradation of gels in media such as phosphate buffer solution (PBS) and dithiothreitol (DTT). The degradation behavior directly affects the mechanical properties, especially in drug release rate and biocompatibility of the gel. The curcumin-loaded PBAE hydrogel undergoes hydrolytic degradation in PBS, an environment designed to mimic physiological condition in the body (pH 7.4, 37 °C). Degradation occurs primarily through hydrolysis of ester linkages within the PBAE backbone, enabling gradual breakdown of the network and controlled release of encapsulated curcumin.

In PBS, all formulations exhibited time-dependent degradation accompanied by swelling and progressive softening of the matrix (Fig. S5). For the gel sample without curcumin, PBAE-CUR-00 had a significant degradation after only 8 hours of observation and completely degraded after 24 hours. In contrast, the gels which contain curcumin at concentrations of 0.1%, 0.5% and 1.0% showed better structural retention. During the first 8 hours of soaking at 37 °C, none of these samples showed significant visible changes. PBAE-CUR-01 and PBAE-CUR-05 began to degrade mildly after 8 hours and by 24 hours, the gel mass had disappeared leaving only a small residue as the free curcumin released precipitating out after release. Interestingly, PBAE-CUR-10 had a slight swelling in the first 8 hours but barely degraded even after 24 hours. The structure remained more intact and curcumin precipitation became noticeable only later. Altogether, these observations suggest that increasing the curcumin content can improve the stability of the hydrogel structure, slowing down the hydrolysis of ester bonds in PBS medium. At high concentration (1.0%), curcumin can strongly interact with the polymer network, slowing down the release and increasing precipitation after hydrogel degradation.

To evaluate redox sensitivity, degradation was further examined in 10 mM dithiothreitol (DTT), a reducing agent capable of cleaving disulfide bonds within the cystamine crosslinks. This condition mimics reductive microenvironments characterized by elevated thiol concentrations, such as inflamed or diabetic tissues, where glutathione (GSH) levels are increased. The cleavage of disulfide bonds disrupts the crosslinked network, significantly accelerating hydrogel degradation. According to Gou et al. (2016), DTT concentrations can be detected as low as 0.07 µM in cells indicating its effective presence in the biological systems. The extracellular DTT concentration is usually maintained at around 1 mmol dm⁻³ in laboratory conditions, making it easy to use in a variety of clinical applications.⁵²

All hydrogels' samples showed degradation over time in 10 mM DTT medium. The degradation rate was quite fast with almost complete degradation occurring within 6 hours for PBAE-CUR-00 (95%). The PBAE-CUR-01, PBAE-CUR-05 and PBAE-CUR-10 samples degraded more slowly than PBAE-CUR-00. PBAE-CUR-00 hydrogel with a less obstructed polymer network structure, disulfide bonds were easily exposed and broken by DTT, leading to faster and more complete degradation (Fig. 8). The presence of curcumin, especially at high concentrations, can create physical interactions such as hydrophobic interactions or hydrogen bonding within the matrix These interactions can increase hydrogel stability making swelling more difficult and reducing the accessibility of DTT to the disulfide bonds. These findings are consistent with previous reports on redox-responsive PBAE systems, confirming the dual degradation mechanism of the hydrogel: hydrolytic degradation under physiological conditions and accelerated cleavage in reductive environments. The combined hydrolytic and redox-responsive degradation profiles demonstrate that the PBAE-CUR hydrogels possess tunable and stimuli sensitive biodegradability. In the context of diabetic wound healing, where elevated oxidative stress and altered redox balance are present, such responsiveness may enable enhanced degradation and drug release specifically at pathological sites. Meanwhile, the improved structural stability observed with moderate curcumin loading (PBAE-CUR-01) suggests an optimal balance between mechanical properties and controlled biodegradation. Overall, the degradation behavior supports the suitability of this system as a biodegradable, stimuli-responsive platform for localized curcumin delivery in wound healing applications.

Fig. 8. In vitro biodegradation of curcumin-loaded PBAE hydrogels in 10 mM DTT medium: (A) image of hydrogel and (B) the rate of hydrogel biodegradation. Data are presented as mean ± SD. *p < 0.05.

The curcumin release profile of the PBAE hydrogel was evaluated using a Franz diffusion cell system with a semipermeable membrane under static conditions. The cumulative release results are presented in Fig. 9. Overall, PBAE-CUR-01 exhibited a rapid release in the early stage (0 to 2 hours). Approximately 20% of the total curcumin was released within the first 30 minutes. This initial burst release can be attributed to curcumin molecules located near or at the hydrogel surface, which are weakly associated with the polymer network and therefore diffuse readily into the receptor medium. Following this early phase, the release rate became more gradual and sustained. At 37 °C, gelation begins to occur, maintaining structural integrity while allowing progressive diffusion of curcumin through the swollen matrix. In the later stage, the system approached equilibrium swelling and partial degradation, the release rate slows down, reaching about 60% after 6 hours. This biphasic pattern suggests a combined mechanism of initial diffusion driven release followed by swelling and degradation-controlled release from the crosslinked PBAE network.

Fig. 9. Release rate of curcumin through the membrane from PBAE-CUR-01 sample.

To further elucidate the release mechanism, the experimental data were analyzed using the Korsmeyer–Peppas kinetic model.^53,54 The fitting results showed an excellent correlation coefficient (R² = 0.9934), indicating that the model accurately describes the release behavior. The release exponent value (n = 0.4797) is close to 0.5, which suggests that the drug release is primarily governed by Fickian diffusion through the hydrated polymer network.⁵⁵ In this mechanism, the release rate is controlled mainly by the concentration gradient and the diffusion of curcumin through the water-filled pores of the hydrogel matrix. The constant k (26.0471) reflects the structural characteristics of the hydrogel, including crosslinking density and network porosity, which influence the overall diffusion rate.⁵⁶ These results demonstrate that the PBAE hydrogel provides a diffusion-controlled drug delivery system capable of sustaining curcumin release, which is advantageous for wound healing applications requiring prolonged therapeutic activity.

Biocompatibility and biological activity

Curcumin-loaed PBAE hydrogel samples were evaluated the antioxidant capacity by using ABTS˙⁺ free radical scavenging method (Fig. 10). The PBAE-CUR-01 sample has better antioxidant activity than the remaining samples despite the low curcumin concentration. In addition, comparing the IC₅₀ results of the samples with free curcumin with IC₅₀ of 4.10 g mL⁻¹, the activity of the PBAE-CUR-01 sample showed superior antioxidant activity with IC₅₀ of 2.38 g mL⁻¹, demonstrating a synergistic effect. Meanwhile, the PBAE-CUR-05 PBAE-CUR-10 sample has less activity than free curcumin. This may be due to the fact that curcumin at high concentrations has poor solubility in polar media, and also partially precipitates and is likely to react with acrylate groups in the gel structure, thereby reducing the actual concentration of curcumin and reducing the antioxidant effect of the gels. Overall, the antioxidant capacity of PBAE hydrogel increased in a synergistic and additive manner when added with curcumin. This demonstrated that the PBAE hydrogel system improved the solubility and dispersion of curcumin more effectively than the use of free curcumin in the ABTS free radical scavenging assay.

Fig. 10. Antioxidant activity of curcumin-loaded PBAE hydrogels: (A) IC₅₀ values based on curcumin concentration and (B) IC₅₀ values based on PBAE hydrogel concentration. Data are presented as mean ± SD. *p < 0.05, **p < 0.01 and ***p < 0.001.

Hemolysis rate is a valid and reliable readout for assessing the hemocompatibility of hydrogels. The mixture of distilled water and red blood cell suspension used as a positive control appeared red, indicative of hemolysis. By contrast, the solutions containing PBAE hydrogel alone (PBAE-CUR-00) at 50, 250 and 500 mg mL⁻¹ were nearly colorless and showed a visible RBC pellet with a slight reddish tinge around it relative to the negative control, suggesting minor hemoglobin release (Fig. 10A). The curcumin-loaded hydrogels showed no significant differences in hemolysis rate compared with PBAE hydrogel (Fig. 10B). Based on the curcumin loading, these formulation concentrations correspond to an estimated curcumin exposure range of approximately 500–5000 µg mL⁻¹. Average hemolysis rates ranged from 1% to nearly 5% across all groups at the tested concentrations, falling below the internationally accepted 5% threshold for good hemocompatibility according to ASTM F756-17 and ISO 10993-4.^57,58 The results indicate that neither the PBAE matrix itself nor curcumin loading up to an estimated 5000 µg mL⁻¹ (or 500 mg mL⁻¹ hydrogel formulation) compromises erythrocyte membrane integrity under these assay conditions (Fig. 11). This concentration range is substantially higher than that used in prior curcumin carrier hemolysis studies such as curcumin-loaded solid lipid nanoparticles up to 200 µg mL⁻¹, curcumin-loaded amphiphilic mixed micelles up to 500 µg mL⁻¹, curcumin-loaded gum arabic aldehyde-gelatin nanogels up to 50 µg mL⁻¹,⁵⁹ and curcumin-loaded polyelectrolyte complex nanoparticles up to 50 µg mL⁻¹.⁶⁰ The concentration range was chosen in this study to match the concentrations applied topically in our in vivo experiments. Moreover, increasing curcumin incorporation into the hydrogel did not affect hemolysis rate, suggesting a dose-independent hemolytic effect. This finding is likely attributable to this specific hydrogel formulation rather than a non-hemolytic property of curcumin itself, as free curcumin or some curcumin carriers may cause dose-dependent changes in the shape of erythrocytes.^61,62

Fig. 11. Hemolysis evaluation of the PBAE-CUR hydrogels. (A) Photographs of the solutions containing RBC suspension mixed with distilled water (positive control), PBS pH 7.4 (negative control), PBAE-CUR-00, PBAE-CUR-01, PBAE-CUR-05 and PBAE-CUR-10 at different concentrations (50, 250 and 500 mg mL⁻¹). (B) Quantification of hemolysis rates (%) of all PABE-CUR hydrogels at all tested concentrations. Data are presented as mean ± SD.

The cytocompatibility of the hydrogel formulations on L929 mouse fibroblast cells was evaluated using a cell viability assay. As shown in Fig. 12, cell viability exhibited a clear concentration-dependent decrease for all tested formulations. At the highest concentration (100 mg mL⁻¹), cell viability was markedly reduced to below 10%, indicating substantial cytotoxicity. As the concentration decreased to 50 mg mL⁻¹, cell viability increased, although considerable differences were observed among the formulations. A further decrease in concentration resulted in a pronounced improvement in cell viability. At 25 mg mL⁻¹, the PBAE-CUR-00 and PBAE-CUR-01 formulations exhibited high cell viability (>90%), whereas the PBAE-CUR-05 and PBAE-CUR-10 formulations showed substantially lower viability. At 12.5 and 6.25 mg mL⁻¹, the cell viability of PBAE-CUR-05 and PBAE-CUR-10 progressively increased, while PBAE-CUR-00 and PBAE-CUR-01 maintained high cell viability (>90%). At the lowest concentration tested (3.125 mg mL⁻¹), all formulations exhibited relatively high cell viability (approximately 89–96%), approaching that of the control group. These results demonstrated that formulations with higher CUR contents (PBAE-CUR-05 and PBAE-CUR-10) exhibited greater cytotoxicity than the formulation with a lower CUR content (PBAE-CUR-01). This observation is consistent with previous reports of the concentration-dependent cytotoxicity of CUR.⁴⁴

Fig. 12. Cytotoxicity properties of PBAE-CUR hydroges on L929 mouse fibroblast cells at different concentrations (n = 3). Data are presented as mean ± SD.

Given the superior antioxidant activity, non-hemolytic profile, and cytocompatibility of PBAE-CUR-01, this formulation was chosen for further safety testing. Topical treatment with PBAE-CUR-01, applied once to the dorsal back skin of shaved mice, did not cause irritation, erythema, eschar formation or edema over 5 days (Fig. S6). This resulted in a zero-irritation score, indicating good topical tolerability. This finding is consistent with non-irritating profile reported for other curcumin-loaded hydrogels.^63,64 However, given that the skin irritation test was conducted only with PBAE-CUR-01 rather than with all formulations, this result suggests that only this specific concentration is well tolerated on intact skin, while PBAE formulations with higher curcumin loading remain to be tested.

To investigate the wound-healing effect of PBAE-CUR-01, the same amount was applied to the excisional wound in mice, after which they were monitored closely over 21 days. The gross appearance of wounds and percentage of wound contraction across four treatment groups (saline, curiosin gel, PBAE-CUR-00 and PBAE-CUR-01) are shown in Fig. 13 and Table S2. On the day of wound induction, exposed underlying muscles were uniformly observed across all groups, confirming consistent full-thickness wound excision. By day 4, all wounds had developed a dry eschar covering the wound bed with well-defined edges. Signs of purulent discharge were seen around the wound periphery in PBAE-CUR-00 group, while no visible pus was evident in the saline, curiosin gel and PBAE-CUR-01 groups. Despite progressive wound contraction, the eschar persisted in all groups by day 7. The curiosin gel group appeared to contract the most at this time point but it did not significantly differ from the other groups. By day 14, the extent of wound closure was substantial in all groups, with the most pronounced contraction observed in the PBAE-CUR-01 group relative to day 7; however, no significant differences were detected. Eschar coverage was reduced in almost all groups at this time point, though there was still minimal pus discharge over the wound surface in the PBAE-CUR-00 group. At day 21, all groups showed nearly complete macroscopic wound closure without necrosis. Notably, although PBAE-CUR-01 did not accelerate wound healing at the early stages, it significantly increased the percentage of wound contract at day 21 after wound induction compared to saline (p = 0.023) and PBAE-CUR-00 (p = 0.045), as shown by repeated measures two-way ANOVA with Geisser-Greenhouse correction (treatment: F(3,10) = 0.9105, p = 0.4702; time: F(1.61,16.10) = 139.5, p < 0.0001; interaction: F(9,30) = 1.33, p = 0.263). Regarding wound contraction, it can be concluded that wounds treated with PBAE-CUR-01 showed promoted wound healing during the tissue remodeling phase (the later stage), but had no effect on wound contraction during the early stages (inflammatory and granulation tissue proliferation phases). However, the evaluation of wound contraction represents only a surface-level assessment and does not clearly reflect the degree of underlying skin tissue regeneration.

Fig. 13. Time course of wound contraction. (A) Representative images of wound closure up to 21 days post-wounding in the saline, curiosin gel, PBAE-CUR-00 and PBAE-CUR-01 groups. (B) Quantification of the percentage of wound closure over 21 days in the saline (n = 3), curiosin gel (n = 3), PBAE-CUR-00 (n = 4) and PBAE-CUR-01 (n = 4) groups. Data are presented as mean ± SD. *p < 0.05. Scale bar = 5 mm.

Histopathological examination revealed distinct skin architecture and tissue regeneration patterns across four treatment groups (Fig. 14A). The saline group exhibited poor-quality tissue regeneration, characterized by incomplete re-epithelialization and disorganized collagen patterns. In contrast, skin tissues from the curiosin gel group were dominated by dense collagen bundles which may be associated with fibrosis and scar formation. Moreover, the epidermis appeared irregular and the dermis was composed of elongated clear space without any adipocytes, suggestive of poor deep tissue regeneration. A similar pattern was observed in PBAE-CUR-00 although this group showed a relatively more uniform epidermal layer. The PBAE-CUR-01 group showed the most complete and organized epidermal structure among the groups, with a well-stratified multi-layered architecture. Epidermal thickness was significantly greater than that of other three groups (p = 0.0004 vs. saline, p = 0.025 vs. curiosin gel, p = 0.0089 vs. PBAE-CUR-00), with an average of approximately 48.1 µm (Fig. 14B). The epidermis functions to protect the skin from environmental factors, prevent water loss, and guard against infection. Increased epidermal thickness is generally considered indicative of epithelial cell proliferation and migration to cover the wound area, representing a crucial stage in the wound healing process, helping to restore the skin's protective barrier. These findings suggest that PBAE-CUR-01 gel may be associated with both external wound closure and signs of epidermal recovery.

Fig. 14. Histopathological examination of skin tissue. (A) Representative images of skin tissues from saline, curiosin gel, PBAE-CUR-00 and PBAE-CUR-01 at the wound site stained with H&E at 21 days post-wounding. Quantification of (B) epidermal thickness, (C) dermal thickness and (D) number of hair follicles in the saline (n = 3), curiosin gel (n = 3), PBAE-CUR-00 (n = 4) and PBAE-CUR-01 (n = 4) groups. Data are presented as mean ± SD. *p < 0.05, **p < 0.01 and ***p < 0.001. Scale bar represents 100 µm.

In the PBAE-CUR-01 group, the dermal layer displayed a balanced and organized collagen network with the presence of adipocytes although dermal thickness did not significantly differ in all groups (Fig. 14C). The dermis plays a crucial role in supporting skin structure, providing nutrition to the epidermis, contributing to immunity and wound repair. When the dermis is damaged due to trauma, aging, or disease, the skin loses elasticity, become dry, wrinkled, and more susceptible to damage. All treatment groups showed a consistent regenerative effect on the dermis, contributing to the skin's foundational structure. In the hypodermis, the number of follicles in mice treated with PBAE-CUR-01 was significantly higher than in the saline (p = 0.0009) and PBAE-CUR-00 (p = 0.0032) groups (Fig. 14D). This finding may reflect either greater preservation of existing follicles during wound healing or contribution to new follicle formation. Distinguishing between these two possibilities would require additional evidence such as time-course follicle counts or staining for a specific marker for follicle neogenesis. Altogether, the results suggest that PBAE-CUR-01 may support the regeneration of accessory skin structures, especially within the hypodermis, which could contribute to a more comprehensive regeneration of skin tissue structure; however, whether this particular hydrogel can confer long-term improvement in skin function requires further assessment.

Curcumin possesses potent antioxidant and anti-inflammatory properties. However, its clinical application is often limited by poor solubility and rapid degradation upon direct topical application.⁶⁵ In this study, we developed the PBAE hydrogel that encapsulates curcumin within the hydrogel matrix, thereby improving its dispersibility. The release study showed approximately 60% cumulative release of curcumin within the 6 hours for PBAE-CUR-01. In the context of wound healing, this initial burst release may instead be advantageous as it delivers a high load of curcumin to the wound site during the early inflammatory phase. Curcumin has been extensively reported to suppress inflammation mediated by multiple signaling pathways including nuclear factor-κB (NF-κB), mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) and Janus kinase/signal transducer and activator of transcription (JAK/STAT).⁶⁶ Whether PBAE-CUR-01 exerts its anti-inflammatory effects through these pathways remains to be determined. Wound closure did not differ among groups from day 1 to day 14. This is not unexpected as a single dose of treatment may have been too transient to produce a measurable difference in wound closure at earlier time points. However, PBAE-CUR-01-treated wounds showed significantly greater contraction at the day 21 than saline- and PBAE-CUR-00-treated groups. Accumulating evidence suggested that prolonged oxidative stress and inflammation can delay wound closure.⁶⁷ Given that PBAE-CUR-01 showed the highest antioxidant activity among the tested formulations, it can be reasoned that PBAE-CUR-01 confers initial combined anti-inflammatory and antioxidant effects that prime wound microenvironment to be more conducive to remodeling at later stages.

In rodents, wound closure is driven predominantly by contraction,⁶⁸ hence, macroscopic contraction alone may not be sensitive enough to reflect differences in tissue regeneration among treatment groups. Histological analysis was performed to examine changes at the tissue level. Both epidermal thickness and hair follicle number increased following PBAE-CUR-01 treatment by day 21 compared with saline and PBAE-CUR-00 groups, suggesting improvements in structural tissue regeneration accompanying wound contraction. It is plausible that the observed effects on wound healing may be, at least in part, attributable to curcumin released as the hydrogel degrades over time. Although curcumin has been shown to promote fibroblast proliferation, angiogenesis, and collagen synthesis,⁶⁹ whether curcumin-loaded PBAE hydrogel produces comparable effects remains to be confirmed. Such parameters have been widely used to characterize wound healing quality in other hydrogel-based wound dressing studies.^70,71 In this study, histological analysis was limited to H&E staining, which does not allow quantitative assessment of collagen density and organization. A more substantiated analysis in future studies using Masson's trichrome staining would provide more definitive evidence on the quality of tissue remodeling. Altogether, our results suggest that PBAE-CUR-01 hydrogel is associated with improved wound contraction and histological signs of tissue regeneration at the later stages of wound healing.

This study has some limitation. The small sample size in this study may have resulted in insufficient power to detect a real but modest difference between day 1 to day 14, therefore, further confirmation is needed to distinguish this from a true absence of early effect. Additionally, the improvement observed in acute wound healing following a single application may plausibly reflect a contribution from the antioxidant activity of the hydrogel and the burst release of curcumin which may exert anti-inflammatory effects. Nevertheless, chronic wounds such as diabetic wounds, which are characterized by persistent inflammation and oxidative stress, may benefit from sustained curcumin release to maintain therapeutic levels over time rather than a single burst release. Further optimization of the hydrogel, along with repeated dosing, may be necessary for applications to chronic wounds.

Experimental

Chemicals

Curcumin 97% (Sigma, USA), cystamine dihydrochloride salt 98% (Shanghai Yuanye Bio-Technology, Taiwan), polyethylene glycol diacrylate (PEGDA) 99% (M_n = 575 g mol⁻¹, Macklin, China), 2,2,2-tribromoethanol (Avertin) 97% (Sigma, USA), acetic acid glacial 99.5% (Xilong, China), alloxan 98% (Sigma, USA), ammonium acetate 98% (Meck, Germany), absolute ethanol 99.7% (Xilong, China), potassium hydroxide 85% (Xilong, China), sodium hydrogen carbonate 99.5% (Xilong, China), sodium hydroxide 99% (Merch, Germany). Other chemical agents were analytical-grade purity and used as received without further purification.

Hydrogel preparation

Dissolved 43.68 mg of NaHCO₃ in an Eppendorf tube with 600 µL of distilled water. Adding a 58.55 mg of cystamine dihydrochloride and vortexed for 1 minute until the cystamine was completely dissolved in the NaHCO₃ solution. Curcumin was first dissolved with PEGDA to avoid incomplete dissolution. A suitable amount of curcumin was weighed and dissolved in 241 µL of PEGDA, vortexed thoroughly for 5–10 minutes to ensure the curcumin was well-dispersed in the mixture. The synthesis ratios were based on Tabel 1. The cystamine mixture was then mixed with the curcumin-PEGDA mixture, and vortexed thoroughly for 5 minutes until the mixture was evenly dispersed. After mixing, the mixture was incubated statically at 37 °C for 2 hours. The formation of the curcumin hydrogel in the Eppendorf tube was observed after 2 hours.

Characterization

FT-IR spectra were recorded using a Nicolet 6700, Thermo Scientific (USA) with a resolution of 4 cm⁻¹ in the wavenumber range of 4000–400 cm⁻¹. Data was recorded in transmittance mode. Samples were pressed into a pellet with KBr before analysis. Solid-state ¹³C CP/MAS NMR spectrum was recorded on a JNM-ECA-400II spectrometer (JEOL, Japan). The thermal properties of PBAE hydrogels will be analyzed using thermogravimetric analysis (STA PT 1600, Linseis GmbH). The sample was heated under N₂ gas from 0 °C to 800 °C at a heating rate of 10 °C min⁻¹. Scanning electronic microscope (SEM) – the S-4800 FESEM Hitachi (Japan) was used to analyze the hydrogel surface morphological. The PBAE hydrogel was freeze-dried at −40 °C for 8 hours and sputter-coated with gold before analyzed, and imaged to reveal surface roughness. The rheological properties of the prepared hydrogels were evaluated using a HAAKE MARS 60 rheometer (Thermo Scientific™). To determine the storage (G′) and loss (G″) moduli, frequency sweep tests were performed from 0.1 to 100 rad s⁻¹ at a constant strain of 1%. Additionally, the mechanical stability under shear stress was investigated via strain sweep measurements (1–100%) at a fixed frequency of 0.1 rad s⁻¹. All experimental procedures were executed at 37 °C and 95% relative humidity.

Hydrogel properties

Water content

A sufficient amount of hydrogel was weighed and recorded the mass of the wet gel (W_w). The gel was dried at 60 °C in the static incubate until reached a constant mass. The dry mass after drying (W_d) was recorded. The experiment was repeated 3 times to obtain the average result. The water content was determined using eqn (1).

1

where W_w is the wet gel weight and W_d is the gel weight after drying.

Swelling ratio

The swelling behavior of the hydrogels was evaluated in phosphate-buffered saline (PBS, pH 7.4) at 37 °C. Prior to the experiment, hydrogel samples were collected, gently blotted with filter paper to remove surface moisture, and dried at 60 °C until a constant weight was obtained (W_d). The dried samples were then immersed in PBS and incubated at 37 °C. At predetermined time intervals, the samples were removed, lightly blotted to eliminate excess surface solution, and weighed immediately (W_s). The swelling ratio (%) was calculated according to eqn (2).

2

where W_s is the weight of the swollen gel, and W_d is the weight of the dry gel.

Biodegradation test

The hydrogel samples were weighed (W_i) and immersed in degradation solutions (4 mL of pH 7.4 phosphate buffer and 10 mM DTT solution) at 37 °C. Hydrogel samples were removed from the solution, dried, and weighed (W_f) at specified time intervals. The experiment was repeated 3 times to obtain the average result. The stability of the hydrogel was determined using eqn (3).

3

where W_i is initial gel weight, and W_f is the weight of immersed gel after drying.

In vitro curcumin released

The PBAE hydrogel released ability was defined using static diffusion cells based on the Franz method.22 PBAE-CUR-01 hydrogel samples were placed in the donor chamber. The recipient solution was PBS buffer (pH = 7.4) with 1% Tween 80 surfactant. A cellulose membrane with a skin-like structure was used to separate the two chambers, ensuring no air bubbles were trapped in the recipient chamber, and was stabilized for 30 minutes. The receptor solution was stirred with a magnetic stirring bar to maintain a constant compound concentration and temperature by circulating water at 32 °C (simulating skin temperature).²³ A cellulose acetate membrane with a pore size of 0.22 µm was fixed between the donor and receptor chambers. The donor chamber was loaded with approximately 1 g of PBAE-CUR-01 (containing 0.1% curcumin) hydrogel. Samples for permeation detection were taken hourly for 6 hours, and immediately after taking a sample, 1 mL of buffer solution was returned to the receptor chamber to ensure the volume remained unchanged. The samples were then measured using a UV-Vis spectrophotometer at the known curcumin wavelength of 426 nm. The cumulative percentage of released curcumin (%) was calculated using eqn (4).

4

where, R_t is the percentage of curcumin released at time t, C_t is the measured curcumin concentration at time t (mg mL⁻¹), V is the total volume of release medium (mL), and m is the initial amount of curcumin loaded into the hydrogel (mg). The Korsmeyer–Peppas model was selected to analyze the drug release kinetics (eqn (5)).

5

where, k is the kinetic constant specific to the hydrogel system, n is the diffusional exponent, M_t (mg) is the cumulative amount of curcumin released into the medium at time t (min), and M_∞ (mg) is the amount released at equilibrium.

Biocompatibility and bioactivity evaluation

Antioxidant activity

The antioxidant activity of PBAE hydrogels was assessed using an optimized ABTS^·+ radical scavenging assay conducted in a pH 4.5 acetic acid buffer under dark conditions. A 120 mL buffer solution was prepared by dissolving 9.252 g of ammonium acetate in distilled water, adding 8.4 mL of glacial acetic acid, and adjusting the volume to achieve a pH of 4.5. The ABTS^·+ radical solution was generated by mixing equal volumes of 7 mM ABTS and 2.45 mM potassium persulfate (K₂S₂O₈) stock solutions and incubating the mixture in the dark for 12–16 hours. The resulting ABTS^·+ solution, exhibiting green fluorescence, was diluted with the buffer to achieve an absorbance of 0.74 ± 0.03 at 734 nm, measured using a UV-Vis spectrophotometer (Thermo UV-Vis EVOLUTION 220, USA). Test samples of curcumin and PBAE hydrogel samples, with concentrations ranging from 3.39 to 54.29 µM (equivalent to 400–6666 µg mL⁻¹ gel mass), were incubated with the ABTS^·+ solution in the dark for 6 minutes, and their absorbance was measured at 734 nm. Ascorbic acid, used as the reference standard, was tested at concentrations of 3.25 to 50 µM. The antioxidant capacity was quantified by calculating the percentage inhibition using eqn (6). The IC₅₀, representing the concentration required to inhibit 50% of the ABTS^·+ radical, was determined for each sample to compare their antioxidant efficacy.

6

where A_c is the absorbance of the ABTS^·+ control and A_s is the absorbance of the test sample.

Hemolysis test

Mouse whole blood was collected into a heparinized tube and centrifuged (1000×g for 10 min) to collect erythrocytes (RBCs). The RBC pellet was washed three times with PBS (pH 7.4) and then diluted to prepare a 2% (v/v) RBC suspension. The hydrogel samples (PBAE-CUR 00, 01, 05, 10) at three concentrations of 50, 250 and 500 mg mL⁻¹ were incubated with an equal volume of RBC suspension at 37 °C for 1 hour, and then centrifuged at 1100×g for 8 min at room temperature. The absorbance of the supernatant was measured at 540 nm using a SpectraMax Mini AFL microplate reader (Molecular Devices, USA), with PBS and distilled water serving as the negative and positive controls, respectively. The hemolysis ratio (%) was calculated as follows (eqn (7)):

7

where A_s, A_n, and A_p represent the absorbance values of the sample, negative control and positive control. Each experiment was performed in triplicate.

Cytotoxicity test

In vitro cytotoxicity of the various hydrogel formulations was evaluated using a Resazurin assay in accordance with ISO 10993-5:2009 guidelines.⁷² Extraction fluids were prepared by immersing hydrogel samples in complete culture medium at a weight-to-volume ratio of 0.1 g mL⁻¹ for 24 h at 37 °C under humidified 5% CO₂ conditions. Concurrently, L929 mouse fibroblast cells were seeded into 96-well plates at a density of 10⁴ cells per well in 100 µL of medium and incubated overnight to facilitate cell attachment. The culture medium was then replaced with 100 µL of the prepared hydrogel extract, followed by an additional 24 h incubation period. Subsequently, 100 µL of resazurin solution was introduced to each well, and the microplate was incubated for 4 h under standard culture conditions. The absorbance was recorded at 590 nm using a microplate reader (USA). All assays were executed in triplicate, and data are expressed as mean ± standard deviation.

Animals

All animal experiments were approved by the Ton Duc Thang University Animal Ethics Committee (Number 02/HDTVDD) and conducted according to the good practice guidelines for the care and use of animals for scientific purposes. Adult male Swiss Albino mice were purchased from the Pasteur Institute, Ho Chi Minh City, and were acclimated to the laboratory environment for at least 2 weeks before the experiments. P80 mice weighed 32 ± 5 g at the start of the experiment. Animals were housed in laboratory polycarbonate cages on a 12 h light/dark cycle and maintained at a controlled temperature of 23 ± 2 °C, with free access to standard rodent chow and water.

Skin irritation test

The skin irritation test was performed in mice using a modified Draize method.⁷³ The dorsal skin surface was shaved (1 cm × 1 cm) and mice were left untreated for 24 hours to observe for any abnormal skin responses. Mice were then topically applied once with 100 µL (100 mg) of PBAE-CUR 01 hydrogel containing 0.1% w/w curcumin on the shaved skin. Signs of local skin reactions including erythema, eschar formation and edema were monitored and visually scored daily up to 5 days.

Excision wound model

Experimental mice were anesthetized with a single injection of avertin (250 mg kg⁻¹, i.p.) and the back hair shaved and disinfected prior to wound induction. A 1-cm diameter circular full-thickness excision wound was created on the dorsal skin of each mouse. Mice with skin wounds were randomly assigned to four groups: (1) control group with wound treated with saline, (2) curiosin group, in which the wound were treated with a commercialized curiosin gel intended for promoting wound healing, (3) PBAE-CUR-00 group, where PBAE hydrogel without curcumin was applied to the wound, (4) PABE-CUR-01 group, in which the wound was covered by 0.1% curcumin-containing PBAE hydrogel. Treatment was initiated immediately following the wound induction, and the wound was protected with a 3 M Tegaderm film. Daily monitoring of wound and animal behaviour was provided throughout the experiment.

Wound contraction

Wound appearance was photographed daily for macroscopic evaluation including assessment of appearance (eschar or necrosis, slough, and granulation tissue), depth (exposed muscle, subcutaneous tissue or skin surface), and purulent discharge (Table S1). Wound areas were quantified from images obtained on the day of wound induction and on days 4-, 7-, 14- and 21-days post-wounding by tracing the wound boundaries using ImageJ software. The percentage of wound contraction was calculated as follows (eqn (8)):

8

where A₀ represents the initial wound area on day 0, and A_n represents the wound area measured on the following days (n = 4, 7, 14, or 21).

Histopathological analysis

On day 21 post-treatment, all mice were euthanized by cervical dislocation, followed by excision of the wound. Tissues were fixed in 10% neutral buffered formalin for 48 hours at 22 °C before being embedded in paraffin for sectioning. Tissue sections of 7-µm thickness were cut using a rotary microtome (Slee medical, Germany) and subsequently deparaffinized and stained with hematoxylin and eosin (H&E) to assess wound architecture, i.e., epidermal and dermal thickness, and the number of hair follicles.

Statistical analysis

All physiochemical properties and in vitro experiments were performed in triplicate to ensure reproducibility. The number of animals used for the in vivo experiments is indicated in the corresponding figure legends. The results are expressed as the mean ± standard deviation (SD). All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software). Changes in wound contraction over time were analysed using repeated measures two-way ANOVAs, followed by Tukey's post-hoc test. Histopathological parameters were analysed using a one-way ANOVA with Tukey's post-hoc analysis. Statistical significance was defined as p < 0.05.

Conclusions

This study successfully synthesized curcumin-loaded poly(β-amino ester) hydrogels via the Michael addition reaction between cystamine dihydrochloride and poly(ethylene glycol) diacrylate. The formulation containing 0.1% curcumin yielded a homogeneous and stable hydrogel. Structural analysis using FT-IR and ¹³C CP/MAS NMR confirmed the formation of the PBAE hydrogel and its structural correspondence to the initial precursors following synthesis. Rheological characterization confirmed that increasing the cross-linking density effectively accelerated gelation and enhanced the mechanical strength and structural stability of the hydrogels, although at the expense of reduced deformability. Water content and swelling ratios tended to decrease with increased curcumin loading, suggesting that curcumin enhances cross-linking density and reduces water retention capacity. Degradation of hydrogels in both PBS (pH 7.4) and DTT (10 mM) showed that the degradation rates increased over time, while higher curcumin concentrations slowed down the degradation process. As the result of in vitro delivery, the PBAE hydrogel sample exhibited an initial burst release followed by a more controlled and sustained release, reaching approximately 60–80% cumulative release after 6 hours. In vivo studies successfully established a murine wound model. Gross observations indicated that all treatment groups facilitated wound improvement. Notably, the PBAE-CUR-01 promoted superior wound healing at the later stage, as reflected by a higher percentage of wound contraction at day 21 and more comprehensive microscopic regeneration of skin architecture including greater epidermal thickness and increased density of hair follicles. Collectively, these findings suggest that PBAE-CUR-01 holds potential as an effective therapeutic strategy for regenerative wound management.

Author contributions

Phuong Tram Nguyen: investigation, writing – review & editing. Ngoc Anh Tam Tran: investigation, writing – original draft. Hoang Hai Le: investigation. Que Lam Nguyen: investigation. Hoang Lich Phan: writing – review & editing. Hieu Minh Doan: investigation. Hoan Ngoc Doan: investigation. Hai Huu Nguyen: investigation. Phu Phong Vo: investigation. Khoi Minh Le: investigation. Tongsai Jamnongkan: investigation, writing – review & editing. Ut Dong Thach: writing – review & editing, supervision, methodology, funding acquisition, conceptualization.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: the synthesized PBAE hydrogel photos, gelation time results, water content, swelling ratio, degradation of PBAE hydrogel in PBS (Fig. S1 to S5). Strauss scoring scale and results of macroscopic assessment of wound characteristics (Tables S1 and S2). See DOI: https://doi.org/10.1039/d6ra04388g.