Electroresponsive hyaluronic acid click hydrogels for combined antibacterial delivery and wound healing applications

Graphical abstract

Highlights

• •Click-crosslinked hyaluronic acid hydrogels were reinforced with functional fibers.
• •Semi-interpenetrating conducting polymers enabled on-demand drug delivery.
• •The composite hydrogel behaves as a robust multifunctional wound dressing.
• •Electrical stimulation enabled sustained chloramphenicol release over eleven days.
• •The bioactive dressings retained cell compatibility and accelerated wound closure.

Abstract

The versatility of electroresponsive hyaluronic acid (HA)-based hydrogels, a promising platforms for wound healing, can be further extended to controlled drug delivery, enabling on-demand release of therapeutic agents. This work reports the development of advanced HA-based electroresponsive hydrogels incorporating electrospun poly(lactic acid) (PLA) fiber mats functionalized with conducting polymers (CPs), such as poly(hydroxymethyl-3,4-ethylenedioxythiophene) and polypyrrole, and loaded with chloramphenicol (CAM). The systems were fabricated through a multi-step approach involving plasma treatment, in situ polymerization of CPs on the fiber mats, and then semi-interpenetration with CP within a click-chemistry crosslinked HA network. The resulting hybrid hydrogels exhibited suitable mechanical properties, swelling behavior, and electrochemical activity, enabling CAM release under both passive and controlled electrically stimulated conditions while maintaining biocompatibility. Electrical stimulation significantly enhanced the initial burst release, while the subsequent release phase showed a sustained profile over an extended period (11 days). The antibacterial activity of the released CAM was confirmed against E. coli and S. aureus using an agar diffusion assay. Although both CPs exhibited comparable CAM release behavior, the polypyrrole-based hydrogel demonstrated superior wound healing performance. Overall, these findings highlight the potential of electroresponsive HA-based composite hydrogels as smart wound dressings capable of controlled and sustained antibiotic delivery.

1. Introduction

The skin, the largest organ of the human body, serves as the primary protective barrier against external factors, including temperature variations, moisture loss, mechanical stress, and pathogenic microorganisms, thereby maintaining homeostasis and preventing disease [1], [2]. A wound is defined as a disruption of the skin’s normal anatomical structure and physiological function caused by factors, such as burns, infections, surgical interventions, or physical or thermal trauma [3]. Wound healing is a complex and hightly regulated process, comprising four overlapped phases (hemostasis, inflammation, proliferation, and remodeling), which are coordinated by dynamic cellular signaling and immune cells, including neutrophils, macrophages, and other leukocytes [4].

Hydrogel-based dressings have emerged as an effective approach for wound management owing to their high water-retention capacity, excellent biocompatibility, and ability to provide controlled topical drug delivery while minimizing patient discomfort during dressing removal [5], [6], [7], [8], [9], [10], [11], [12]. These three-dimensional hydrophilic polymer networks can absorb large amounts of water and biological fluids without dissolving, and can be engineered as smart materials responsive to external stimuli, such as temperature, pH, or ionic strength, enabling on-demand therapeutic release [8], [9], [10], [11], [12]. Such functionality is particularly important because bacterial infection remains one of the main causes of delayed wound healing and chronic wound progression, increasing treatment costs, therapeutic failure, and the risk of severe complications, including sepsis and amputation [13], [14], [15]. Consequently, antibacterial hydrogels incorporating antibiotics, antimicrobial peptides, or metallic nanoparticles have been extensively investigated [16], [17], [18]. However, uncontrolled drug release often limits their long-term efficacy, highlighting the need for stimulus-responsive hydrogels capable of delivering therapeutic agents in a controlled manner [19], [20], [21], [22], [23].

Among these materials, hyaluronic acid (HA)-based hydrogels have attracted considerable attention because of their biodegradability, low immunogenicity, excellent biocompatibility, and intrinsic role in regulating inflammation and cell migration during wound repair [24], [25], [26], [27], [28]. Achievements on electroresponsive HA-based hydrogel were recently reviewed [6]. Among them, two studies deserve special consideration. In 2019, Qu et al. [28] prepared a conducting hydrogel by mixing N-carboxyethyl chitosan and oxidized hyaluronic acid-graft-aniline tetramer, which was loaded with the antibiotic amoxicillin to prevent the infections. More recently, we developed an electroresponsive HA-based hydrogel by semi-interpenetrating a poly(hydroxymethyl-3,4-ethylenedioxythiophene) network [12]. Briefly, HA was functionalized with thiol groups and crosslinked with an alkyne-functionalized three-arm polyethylene glycol through a thiol–yne click reaction to produce the clickHA hydrogel, into which the conducting polymer (CP) was subsequently synthesized in situ by chemical oxidation.

In this work, we develop an advanced electroresponsive HA-based system that not only maintains wound-healing capacity of clickHA/PEDOT-MeOH, but also displays improved robustness and electrochemical properties, as well as a prolonged and sustained release of a standard antimicrobial drug. To this end, the correlations among structure, property, and processing of different organic materials, through innovative and proactive design, have been explored by following a multi-step process (Fig. 1). Specifically, plasma-treated electrospun poly(lactic acid) (PLA) fiber mats (Fig. 1a-b) have been coated with a CP to induce electroresponsiveness (Fig. 1c), loaded with the drug (Fig. 1d), and, subsequently, embedded into clickHA hydrogels by in situ formation (Fig. 1e). After this, the fiber-containing clickHA hydrogels have been semi-interpenetrated by a CP network through oxidative chemical polymerization (Fig. 1f). Two different CPs, PEDOT-MeOH and polypyrrole (PPy), were employed to coat the PLA fibers and to semi-interpenetrate the clickHA hydrogel. It should be mentioned that the stability of the selected CPs in a physiological environment far exceeds the frequency of dressing changes, which occurs every 2–7 days depending on the type of wound and the functionality of the dressing. The antimicrobial drug selected for this work is chloramphenicol (CAM), a broad spectrum antibiotic that is highly effective against common wound-pathogenic bacteria, including both Gram-positive and Gram-negative strains.

Fig. 1.

Procedure (multi-step process) used to prepare the advanced electroresponsive HA-based system for controlled CAM release.

2. Methods

Description of the Materials, PLA electrospinning, PLA plasma treatment, characterization methods, biocompatibility and cell viability assays are provided in the Supplementary Information.

2.1. Electroresponsive fibers

Electrospun PLA fibers were subjected to oxygen plasma treatment and, subsequently, coated with PEDOT-MeOH or PPy to obtain PLA/PEDOT-MeOH or PLA/PPy, respectively (Table S1). For coating, plasma-treated PLA fibers were immersed for 24 h in either a 0.1 M hydroxymethyl-3,4-ethylenedioxythiophene (EDOT-MeOH or a 0.05 M pyrrole (Py) solution prepared in 0.2 M HCl. Subsequently, the samples were rinsed with Milli-Q water and transferred into oxidant solutions containing either 0.1 M ammonium persulfate (APS) for EDOT-MeOH or 0.05 M FeCl₃ for Py in 0.2 M HCl for an additional 24 h to promote in situ polymerization. The resulting fiber mats were thoroughly rinsed with Milli-Q water to remove unreacted monomers, initiator, and the excess of acid; and then dried at room temperature. Fiber mats were cut to 1 × 1 cm².

2.2. Loading of CAM

A stock solution of CAM (20 mg/mL) was prepared in ethanol and subsequently diluted to 1 mg/mL using DPBS. An aliquot of 50 µL of this solution was deposited onto 1 cm² of plasma-treated PLA, PLA/PEDOT-MeOH, and PLA/PPy fiber mats. The loading efficiency (LE, in %) was defined as the ratio between the mass of CAM loaded in the fibers (m_f) and the mass of CAM in the loading ethanol solution (m_s):

$$\mathit{LE}(\%)=\frac{m_f}{m_s}\times \text{100}$$ (1)

To quantify drug loading, CAM-loaded fiber samples were incubated in PBS (5 mL) under gentle stirring for 2 weeks to ensure complete drug release. The concentration of CAM in the release medium was determined by UV–Vis spectroscopy using a UV-1900i Plus dual-beam spectrophotometer. The mass of released CAM was calculated from a calibration plot.

2.3. Synthesis of the hydrogel

HA was functionalized with thiol groups (HA-SH) following previously reported procedures [29] A 3-arm alkyne-functionalized polyethylene glycol (PEG) crosslinker was synthesized via esterification of glycerol ethoxylate with propiolic acid, as described elsewhere [12]. Details about the preparation of clickHA hydrogels are provided in the Supplementary Information.

2.4. Semi-interpenetration of fibers-containing clickHA hydrogels

ClickHA hydrogels containing plasma-treated PLA or PLA/PEDOT-MeOH fibers loaded with CAM were placed in a vial filled with 1.5 mL of 0.1 M EDOT-MeOH solution in 0.2 M HCl and allowed to infiltrate for 24 h at room temperature. Subsequently, the solution was then removed and the hydrogels were rinsed with Milli-Q water. Then, 1.5 mL of 0.1 M APS solution in 0.2 M HCl was added to induce chemical oxidation for 24 h at room temperature. Finally, the semi-interpenetrated PEDOT-MeOH hydrogels were removed from the vials and thoroughly washed by immersion in Milli-Q water (three times, each for 20 min) before characterization.

The semi-interpenetration of clickHA hydrogels containing plasma-treated PLA or PLA/PPy fibers loaded with CAM was carried out following the same procedure, using 1.5 mL of a 0.05 M Py solution in 0.2 M HCl for infiltration over 24 h, followed by the addition of 1.5 mL of 0.05 M FeCl₃ solution in 0.2 M HCl for chemical oxidation.

2.5. Release of CAM

Controlled CAM release assays were performed by applying a +0.6 V electrical stimulus for 15 min at predefined times (i.e. 0, 0.5, 1, 2, 3, 4, 6, 24, 48, 72, 96, 120, 144, 168, 192, 216, 240, and 264 h) using an Autolab PGSTAT302N operated with NOVA software. The experiments were carried out in the 3D-printed PLA cell as previously described (Fig. S1). The electrochemical set-up consisted of a Pt counter electrode, an Ag|AgCl reference electrode, and CAM-loaded hydrogels serving as the working electrode. The release medium comprised 5 mL of PBS solution at pH 7.4. Aliquots of the medium, which was fully replaced, were collected at predefined time intervals (i.e. 5, 10, 15 and 30 min, as well as 1, 2, 3, 4, 6, 24, 48, 72, 96, 120, 144, 168, 192, 216, 240, and 264 h) and analyzed by UV–Vis spectroscopy to quantify the released CAM using a calibration plot.

2.6. Bactericidal activity

In order to demonstrate that the bactericidal activity of the drug was not affected by the applied voltage, the agar diffusion test was applied to the CAM released from the conducting hydrogels after 15 min, 6 h, and 72 h, with and without electrical stimulation. Tests were performed against Escherichia coli (E. coli) and S. aureus (S. aureus) using a filter paper disc on tryptic soy agar (TSA), adapted from the EUCAST standard protocol. Briefly, the bacterial inoculum was adjusted to an OD₆₀₀ of 0.1 by diluting the pre-inoculum in PBS. Then, 100 µL of the bacteria suspension were seeded onto TSA plates. Sterile filter paper discs were impregnated with 20 µL of each condition and allowed to air-dry under sterile conditions to prevent liquid spreading in the agar surface. The discs were subsequently placed onto the inoculated TSA plates and gently pressed to ensure full contact with the agar surface. Antibacterial activity was assessed after incubation at 37 °C for 18–24 h by evaluating the inhibition of bacterial growth. The effective CAM concentration in the evaluated samples ranged from 0.023 to 0.059 mg/mL.

2.7. In vitro stimulation for wound healing

The tested hydrogels were placed in well plates and sterilized by UV irradiation for 1 h in a laminar flow hood. Cells were seeded at a density of 7 × 10⁴ cells/mL until confluence was reached. After replace the culture medium by fresh medium, a linear scratch was created in the cell monolayer using a micropipette tip to simulate a wound, immediately prior to electrostimulation. The electrostimulation setup was implemented as previously reported [12]. Transwell inserts were used to allow indirect contact between the hydrogels and the cells through a permeable membrane. Stainless-steel wires, sterilized by UV exposure for 30 min in a laminar flow hood, served as electrodes. Electrical stimulation consisted on the application of a constant voltage of 0.5 V for 15 or 30 min. Intracellular reactive oxygen species (ROS) levels were quantified by measuring the fluorescence (Ex/Em = 488/525 nm) of the oxidized form of 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) using the Invitrogen^TM ROS Fluorometric Assay Kit (Green). A more detailed description of the procedures is provided in the Supplementary Information.

Ethical approval was not required for this study because HFF-1 cells were obtained from a commercial cell repository as an established human cell line. The study did not involve the recruitment of human subjects, the collection of primary human tissue, or access to identifiable donor information.

3. Results and discussion

3.1. Characterization and CAM-loading of PLA/PEDOT-MeOH and PLA/PPy fibers

Fig. S2 shows representative scanning electron microscopy (SEM) micrographs of electrospun PLA fibers, before and after plasma treatment, as well as modified PLA/PEDOT-MeOH and PLA/PPy fibers. Untreated cylindrical PLA microfibers, with an average diameter: 0.78 ± 0.38 μm, exhibit a rough surface texture with small protuberances (Fig. S2a), which has been attributed to the phase separation between the two solvents used during electrospinning. After plasma treatment, PLA fibers display abundant surface pits (Fig. S2b), attributed to oxidation and chain scission of PLA during oxygen plasma exposure. However, the fiber diameter is not significantly affected by the plasma treatment (0.92 ± 0.22 μm). Micrographs of the electroresponsive fibers confirm the successful incorporation of the CPs, which appear as large micrometric agglomerates in PLA/PEDOT-MeOH (Fig. S2c) and as individual nanoparticles in PLA/PPy (Fig. S2d). The average fiber diameters were 0.96 ± 0.26 and 0.98 ± 0.20 μm, respectively, indicating that the polymerization of EDOT-MeOH and Py has no significant effect on fiber thickness. Nevertheless, it is worth noting that the surface features in plasma-treated PLA fibers are barely visible in PLA/PEDOT-MeOH and PLA/PPy samples, which suggests the formation of an ultra-thin CP coating.

The effects of low-pressure oxygen plasma-treatment on the FTIR and Raman spectra of PLA are consistent with previous reports [30] and are, therefore, not discussed in detail here. When the oxygen pressure is below 0.8 mbar, as in this study, the main effect observed in the FTIR spectrum of PLA is an increase in the transmittance of bands associated with ester group vibrations. Figs. S2e-f compare the FTIR and Raman spectra of plasma-treated PLA, PLA/PEDOT-MeOH, and PLA/PPy. The FTIR spectra of PLA/PEDOT-MeOH and PLA/PPy are dominated by the PLA absorption bands, which are described in the Supplementary Information, with no detectable characteristic signals from PEDOT-MeOH aggregates and PPy nanoparticles. In contrast, the bands of PLA are not present in the Raman spectra of PLA/PEDOT-MeOH and PLA/PPy, which are instead dominated by signals from PEDOT-MeOH and PPy. This behavior is attributed to the well-known resonance Raman effect, whereby the intensity of the CP bands is enhanced when the laser excitation energy matches the electronic transition of the material [31]. Consequently, the broad bands observed in the 1250–1600 cm⁻¹ region for PLA/PEDOT-MeOH and PLA/PPy are assigned to C_α–C_α’ inter-ring stretching, C_β=C_β stretching, and symmetric and asymmetric C_α=C_β stretching modes of the CP.

Fig. S3 shows the differential scanning calorimetry (DSC) thermograms of PLA pellets, electrospun PLA fibers, plasma-treated electrospun PLA fibers, and both PLA/PEDOT-MeOH and PLA/PPy modified fibers. The thermal properties derived from DSC analyses are summarized in Table S2. Neat PLA pellets (control) exhibit T_g and T_m values of 60.8 and 152.6 °C, respectively, which are in good agreement with those provided by the supplier (see Materials subsection). The T_g of electrospun fibers is slightly lower than that of the bulk control, regardless of plasma-treatment and modification with CPs. This behavior is attributed to the high surface area and porosity of the fibers, which increase free volume and enhance chain mobility, as well as to reduced structural constraints in the amorphous phase. In addition, confinement effects at the microscale and processing-induced non-equilibrium chain conformations further promote segmental mobility, leading to a decrease in T_g. A similar trend is observed for T_m, which also shows a slight decrease in all fiber-based samples. The most pronounced changes are found in plasma-treated PLA, likely because of the degradative effect of the plasma. In contrast, the reductions observed in PLA/PEDOT-MeOH and PLA/PPy are more significant than in untreated PLA fibers, which can be attributed to the presence of oligomers that can exert a plasticizing effect.

CAM was loaded into plasma-treated PLA, PLA/PEDOT-MeOH and PLA/PPy fibers using the procedure described in the Methods section, and the CAM loading efficiency (LE, Eq (1) was evaluated by UV–Vis spectroscopy. The characteristic absorption peak of CAM at 278 nm was used to construct the calibration curve (Fig. S4), which was subsequently employed to quantify the LE of the CAM-loaded fibers (Fig. 2a). The LE values obtained for the different fiber systems were comparable (Fig. 2a), ranging from 87.5% ± 7.2% for plasma-treated PLA fibers to 74.7% ± 6.2% for PLA/PPy fibers.

Fig. 2.

(a) LE (Eq. (1)) of CAM in plasma-treated PLA, PLA/PEDOT-MeOH, and PLA/PPy fibers. (b-d) Representative SEM micrographs of (b) clickHA/PLA, (c) clickHA/PLA/PEDOT-MeOH, and (d) clickHA/PLA/PPy. Fibers in the two latter systems are displayed in the insets.

3.2. Characterization of fibers-containing ClickHA hydrogels

CAM-loaded plasma-treated PLA, PLA/PEDOT-MeOH and PLA/PPy fibers were incorporated into clickHA hydrogels following the procedure described in the Methods section. Hereafter, the resulting fibrous scaffolds supported within the hydrogel matrix for CAM release are denoted as clickHA/PLA, clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy, respectively (Table S1).

FTIR and Raman spectra of clickHA, clickHA/PLA, clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy (previously lyophilized) are compared in Fig. S5. In addition to the clickHA hydrogel signals, which are discussed in the Supplementary Information, the FTIR spectrum of clickHA/PLA displays a weak absorption band at 1743 cm⁻¹, which is attributed to the C=O stretching of plasma-treated PLA. Furthermore, the spectra of clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy exhibit shoulders at 1305 and 1460 cm⁻¹, which are assigned to the inter-ring C–C stretching and the intra-ring C=C ring stretching vibrations, respectively. The Raman spectrum of clickHA/PLA is dominated by the hydrogel signals, which are described in the Supplementary Information, and no characteristic PLA bands are detected. As previously observed for PLA/PEDOT-MeOH and PLA/PPy, the Raman spectra of clickHA/PLA/PEDOT-MeOH/and clickHA/PLA/PPy exhibit broad bands in the 1300–1500 cm⁻¹ region, which are attributed to the PEDOT-MeOH and PPy CPs. Nevertheless, the successful incorporation of plasma-treated PLA fibers into the clickHA matrix is confirmed by SEM. Fig. S6 presents a representative SEM micrograph of the pristine clickHA hydrogel. SEM images of uncoated PLA fibers embedded within the clickHA matrix, as well as PLA fibers coated with conducting polymer (CP) agglomerates and nanoparticles, are shown in Fig. 2b-d.

Fig. 3a shows the DSC thermograms for clickHA and clickHA/PLA lyophilized hydrogels, while the corresponding thermal parameters are summarized in Table S3. The degradation of HA polysaccharide occurs at around 240 °C [32] and, therefore, the broad endothermic peak found at 162 °C for clickHA has been attributed to the melting of HA and the 3-arm PEG crosslinker. ClickHA/PLA exhibits the T_g and T_m of PLA at 60° and 157 °C, respectively, indicating that the interaction between fiber mats and the hydrogel matrix is weak. In contrast, the DSC thermograms of hydrated clickHA and clickHA/PLA hydrogels (Fig. S7) display two endothermic peaks. The first peak, centered at around 0 °C, corresponds to the melting of ice domains present in the samples. The second one observed at 160 and 164 °C for clickHA and clickHA/PLA, respectively, is attributed to dehydration processes, the melting of HA and 3-arm PEG crosslinker, and in the case of the clickHA/PLA also to the melting of PLA.

Fig. 3.

(a) DSC thermograms of clickHA and clickHA/PLA lyophilized hydrogels (thermograms are offset for clarity). (b-d) Cyclic voltammograms recorded in PBS for: (b) clickHA/PLA, clickHA/PLA/PEDOT-MeOH, and clickHA/PLA/PPy; (c) semi-interpenetrated clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, the corresponding blank (clickHA/PEDOT-MeOH/PLA) and non-semi-interpenetrated clickHA/PLA/PEDOT-MeOH; and (d) semi-interpenetrated clickHA/PPy/PLA/PPy, the corresponding blank (clickHA/PPy/PLA), and non-semi-interpenetrated clickHA/PLA/PPy. (e,f) Representative SEM micrographs of (e) clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and (f) clickHA/PPy/PLA/PPy hydrogels.

Fig. 3b compares the cyclic voltammograms recorded for clickHA/PLA, clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy hydrogels. As can be observed, the incorporation of CP-loaded PLA fibers increases the current density (j) range; however, the enhancement in the electrochemical response of clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy with respect to clickHA/PLA is not significant. This result suggests that an effective percolation network among the CP nanoparticles and agglomerates is not established. Consequently, the electronic contribution to the electrochemical behavior remains limited, and the response is dominated by the ionic contribution associated with the charged carboxylate groups of the clickHA hydrogel. To overcome this limitation, the clickHA matrix was semi-interpenetrated with PEDOT-MeOH and PPy.

3.3. Semi-interpenetrated ClickHA/PLA/CP hydrogels

ClickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy were semi-interpenetrated with PEDOT-MeOH and PPy, respectively, by infiltrating a solution of the corresponding monomer (EDOT-MeOH or Py) into the hydrogel. This step was followed by chemical oxidative polymerization, induced by immersing the infiltrated hydrogel into a solution containing the oxidizing agent (either APS or FeCl₃). It is worth noting that, in this process, the CP nanoparticles and aggregates previously deposited onto PLA fibers were expected to act as polymerization nuclei, thereby facilitating the semi-interpenetration of the hydrogel matrix. To verify this hypothesis, clickHA/PLA semi-interpenetrated hydrogels were also prepared as blanks using the same procedure but in the absence of CP-loaded fibers. Hereafter, the semi-interpenetrated hydrogels prepared using fibers with CP nanoparticles and aggregates are denoted clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy (Table S1), while the corresponding blank hydrogels (prepared with non-modified PLA fibers) are referred to as clickHA/PEDOT-MeOH/PLA and clickHA/PPy/PLA, respectively (Table S1).

Fig. 3c–d compares the cyclic voltammograms of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy with those of the corresponding blanks and non-semi-interpenetrated systems. As observed, the results depend strongly on the type of cp. while semi-interpenetration with PEDOT-MeOH does not lead to significant changes in the electrochemical response, regardless of the presence of PEDOT-MeOH nanoparticles and aggregates on the PLA fibers, a marked improvement is observed for PPY-based systems. This enhancement is particularly pronounced when PPY particles act as polymerization nuclei. in the case, the j variation within scanned potential window was 51% and 26% higher for clickHA/PPy/PLA/PPy than for clickHA/PLA/PPy and clickHA/PPy/PLA, respectively.

Although the FTIR spectra of all CP-containing systems remain dominated by the characteristic bands of clickHA (Fig. S8a), the presence of PEDOT-MeOH and PPy in the semi-interpenetrated clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy, as well as in their corresponding blanks, is clearly confirmed by Raman spectroscopy. As shown in Fig. S8b, the bands in the 1250–1600 cm⁻¹, associated with vibrations of the polyheterocyclic backbone in PEDOT-MeOH and PPy, dominate the spectra of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, clickHA/PEDOT-MeOH/PLA, clickHA/PPy/PLA/PPy and clickHA/PPy/PLA. These results confirm the successful incorporation of PEDOT-MeOH and, more prominently, of PPy via chemical oxidative polymerization into the preformed clickHA/PLA/PEDOT-MeOH, clickHA/PLA/PPy and clickHA/PLA systems, thereby explaining the differences found in their electrochemical behavior. The thermal behavior of clickHA/PLA is not altered by the semi-interpenetrating CPs, as shown in the DSC thermograms (Fig. S9) and the corresponding thermal parameters (Table S4).

Cross-sectional SEM micrographs of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, and clickHA/PPy/PLA/PPy hydrogels are presented in Fig. 3e–f (higher magnification images are displayed in Fig. S10). The most notable differences are observed in the PLA/PEDOT-MeOH and PLA/PPy fibers. In clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, the fibers remain clearly distinguishable, whereas in clickHA/PPy/PLA/PPy they are partially coated by the CP and become more integrated into the hydrogel matrix, which explains the improved of the PPy based systems compared to PEDOT-MeOH ones (Fig. 3c–d).

The distinct behavior of PEDOT-MeOH and PPy within the fiber-reinforced clickHA hydrogel matrix can be attributed to differences in monomer solubility, polymerization kinetics, and interfacial interactions with both the clickHA network and the fibers [33], [34], [35]. Py is significantly more soluble in aqueous acidic media than EDOT-MeOH, enabling faster diffusion throughout the swollen hydrogel and promoting the formation of a continuous PPy network. In contrast, the lower aqueous solubility and slower diffusion of EDOT-MeOH favor polymerization near the hydrogel pore walls and external surfaces, resulting in a less uniform conductive phase with larger polymer domains. Furthermore, the positively charged PPy backbone interacts more strongly with the negatively charged carboxylate groups of clickHA through electrostatic interactions, while hydrogen bonding further enhances polymer integration within the hydrogel network. PEDOT-MeOH, whose positive charge is more delocalized and stabilized by bulkier dopant anions, exhibits weaker interactions with clickHA and therefore a lower degree of interpenetration. Similar differences are expected at the PLA fiber surface, where PPy readily nucleates and forms conformal coatings owing to its rapid oxidative polymerization, whereas PEDOT-MeOH generally displays a lower nucleation density and weaker interfacial adhesion unless the substrate is chemically activated.

The temporal evolution of swelling ratio (SR; Eq S1) is shown in Fig. 4a for all prepared hydrogels. After 9 days, SR values range from 876% ± 5% (clickHA/PPy/PLA/PPy) to 1353% ± 23% (clickHA/PLA/PPy). Notably, the incorporation of the semi-interpenetrating network delays water uptake. While clickHA/PLA, clickHA/PLA/PEDOT-MeOH and clickHA/PLA/PPy swell rapidly, reaching equilibrium within only 8 h, clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, clickHA/PPy/PLA/PPy and their corresponding blanks absorb water only up to only 600–700% over the same period. However, these systems subsequently exhibit a slow but sustained increase in water uptake over the following days. Nevertheless, the swelling capacity of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, clickHA/PPy/PLA/PPy hydrogels remain sufficiently high to maintain a moist environment if used as wound healing patches.

Fig. 4.

(a) Swelling kinetics in PBS (pH 7.4) of the hydrogels prepared in this work. (b) Representative stress–strain curves of clickHA/PLA, clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy.

The mechanical performance of clickHA/PLA (blank), clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels was evaluated by uniaxial compression testing (Fig. 4b). All systems exhibited robust mechanical behavior, characterized by relatively high Young’s modulus and compressive strength (defined as the maximum compressive stress before failure). Specifically, the Young’s modulus of clickHA/PLA, determined to be 24 ± 5 kPa, decreased slightly upon incorporation of PEDOT-MeOH and PPy (17 ± 2 and 21 ± 3 kPa, respectively). This reduction is attributed to interfacial debonding at the fiber surface. In fiber-reinforced hydrogels, the fibers act as microstructural struts that enhance the compressive stiffness by resisting lateral deformation. However, the in situ polymerization of PEDOT-MeOH or PPy modifies the fiber/clickHA interface, reducing the interfacial adhesion between the fibers and the hydrogel matrix. Consequently, the fibers become more prone to interfacial slippage or debonding under compressive loading, resulting in a modest decrease in the overall compressive stiffness of the composite. The compressive strain at break of clickHA/PLA, clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels was 30% ± 4%, 23% ± 6%, and 26 ± 4%, respectively, while the corresponding compressive strength values were 68 ± 18 kPa, 47 ± 17 kPa, and 59 ± 11 kPa, respectively. Overall, the incorporation of PEDOT-MeOH and PPy had only a modest effect on the mechanical properties of clickHA/PLA.

The role of PLA fibers in the mechanical properties of the system is clearly evidenced by comparison with clickHA hydrogels without and with semi-interpenetrating PEDOT-MeOH networks (i.e. clickHA and clickHA/PEDOT-MeOH), which were evaluated using the same procedure. The Young’s modulus increased from 10.8 ± 0.9 kPa for clickHA to 29.7 ± 4.3 kPa for clickHA/PEDOT-MeOH [12], indicating that the incorporation of the PLA fibers enhanced the elastic modulus by approximately 9- and 3-fold, respectively. In addition, the compressive strain at break was 47.5% ± 4.7% for clickHA and 46.4% ± 4.8% for clickHA/PEDOT-MeOH hydrogels, while the corresponding compressive strength values were 17.3 ± 3.9 kPa and 47.7 ± 8.6 kPa, respectively [12].

3.4. Release of bioactive CAM

The release of CAM from clickHA/PLA (control), clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels was investigated under both passive conditions and in the presence of electrical stimulation (+0.6 V for 15 min at predefined times). The antimicrobial drug was initially loaded into PLA, PLA/PEDOT-MeOH and PLA/PPy fiber meshes (Fig. 2) prior to their incorporation into the clickHA hydrogel and the subsequent formation of a semi-interpenetrating CP network. The passive and electrostimulated kinetics release profiles were monitored over a period of 11 days and are presented in Fig. 5a. Quantification was performed by UV–Vis using the calibration plot displayed in Fig. S4.

Fig. 5.

(a) Release of CAM from clickHA/PLA (control), clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy without and with electrical stimulation (ST). (c) Cumulative CAM release after 11 days without (empty bars) and with (filled bars) electrical stimulation.

For all systems, including the CP-free control, the initial burst release was significantly higher under electrical stimulation than under passive conditions, differences reaching 8% for both clickHA/PLA and clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, and 21% for clickHA/PPy/PLA/PPy. Following this initial stage, the release rate ranged between 0.08 and 0.11% per hour for clickHA/PLA, clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy, regardless of whether electrical stimulation was applied. Due to this similarity in release rates during this phase, the differences observed in the cumulative CAM release after 11 days are primarily attributed to the effect of electrical stimulation on the initial burst release (Fig. 5b). This prolonged (11 days) sustained release was not achieved in previous CAM-loaded polyester fibers systems [36] and conducting hydrogels [19], where, following the initial burst release, the drug release reached a plateau after only a few minutes and a few hours, respectively.

The release mechanism of CAM is governed by changes in its interactions with the fibers. Upon loading, the antibiotic is retained primarily through interactions between its polar functional groups and the doped CPs, or with the surface functional groups generated by plasma treatment in PLA. However, the latter interactions are relatively weak compared with the electrostatic interactions established between the positively charged CP backbone and the compensating dopant anions.

Under passive conditions, CAM is released as a result of the dynamic equilibrium between CAM···CP and CAM···water competing interactions, leading to diffusion of the antibiotic into the surrounding medium. Upon applying an electrical potential of +0.6 V, the CP chains undergo further oxidation, as is observed by the potential peak at around +0.5 V in the corresponding cyclic voltammograms (Fig. 3b-d), increasing the density of positive charges along the polymer backbone. This enhanced oxidation strengthens the electrostatic association between the CP chains and their counter-anions, thereby modifying the balance of intermolecular interactions at the polymer interface. As a consequence, CAM···CP interactions become less favorable, while CAM···water interactions are promoted, facilitating the diffusion of CAM from the polymer coating into the surrounding medium. Therefore, electrical stimulation accelerates drug release by shifting the interaction equilibrium toward desorption and diffusion of the antibiotic. In the case of PLA alone, the effect of the stimulation is much smaller because PLA only undergoes unspecific oxidation (usually associated to chain scissions) when an electric voltage is applied.

The bactericidal activity against E. coli and S. aureus of CAM released from clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels under passive and electrostimulated conditions was evaluated after 15 min, 6 h, and 72 h by measuring the inhibition zones formed against E. coli and S. aureus on TSA plates. As controls, freshly prepared (<24 h) CAM solutions (20 mg in 100 µL of PBS) and solutions prepared one week in advance were also evaluated. The agar diffusion test showed similar trends for E. coli and S. aureus (Fig. S11). The inhibition zones were more pronounced for E. coli, with diameters ranging from 3.3 to 3.8 cm (Fig. 6a), than for S. aureus, which showed diameters between 2.5 and 2.8 cm (Fig. 6b). These results indicate that, although CAM is a broad-spectrum antibiotic, it is more effective against the Gram-negative bacterium E. coli than against the Gram-positive bacterium (S. aureus). Furthermore, neither electrical stimulation nor release time significantly affected the bactericidal activity of the released CAM. Overall, these results demonstrate that the released CAM is bioactive and exerts a clear inhibitory effect on bacterial growth.

Fig. 6.

Diameter of the inhibition zone in the agar diffusion tests for (a) E. coli and (b) S. aureus exposed to prepared CAM solutions (control) and CAM released from clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy without and with electrostimulation (ST). (c) Cell viability values for the HFF-1 cell line after exposure for 24 h, 48 h and 7 days to clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels. (d) Wound size indicate the scratch gap for the two studied hydrogels without and with electrostimulation (ST; 15 min). The control corresponds to assays without hydrogel. Error bars indicate the standard deviation (n = 3).

Finally, Fig. 6c-d shows that clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels maintain the biocompatibility and wound healing properties previously observed for clickHA/PEDOT-MeOH. Biocompatibility was assessed using commercial HFF-1 cells (ATCC-SCRC-1041.1) by measuring cell viability after 24 h, 48 h and 7 days. In general, biocompatible hydrogels maintain high cell viability (> 80–90%) within the first 24–48 h while, over longer periods (e.g. 7-day), viability may vary depending on factors such as the hydrogel degradation, cell proliferation, and the accumulation of acidic byproducts. As shown in Fig. 6c, both materials exhibited high cell viability (>120%, >115% and >145% at 24 h, 48 h and 7 days, respectively). These results indicate that biocompatibility in preserved after incorporating PLA/CP fibers, forming a semi-interpenetrating CP-network within the hydrogel matrix, and replacing PEDOT-MeOH by PPy.

The performance of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels as electroresponsive wound dressings was investigated using fibroblast cells. Prior to the assay, a scratch was made using a pipette tip to generate an artificial wound in the cell monolayer. The experiment was conducted without electrostimulation at all (control systems that mimics a non-stimulated wound closure process) and with an applied electrical stimulus applied of 0.5 V for 15 min. In addition to electroresponsive wound dressings, assays were conducted without considering any dressing material (control). Following electrostimulation, cells were further incubated for 1 h at 37 °C and 5% CO₂ before fixation and staining. The wound gap was then visualized by microscopy and quantified. The results are shown in Fig. 6d as the relative wound size with respect to t = 0 min, while representative micrographs are shown in Fig. S12.

In absence of electrical stimulation, after 1 h 15 min, the wound gap in the presence of clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH hydrogel was approximately 7% smaller than that of the control (i.e. without hydrogel), whereas for clickHA/PPy/PLA/PPy it was about 9% larger. Upon applying a +0.5 V stimulus for 15 min, the wound gap decreased in all cases after 1 h, with reductions of approximately 15% for the control and 26% for the hydrogel-containing samples, confirming the beneficial effect of electrostimulation on wound healing. Notably, under electrostimulated conditions, the wound gap was about 18% and 6% smaller for clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH and clickHA/PPy/PLA/PPy hydrogels, respectively, compared to the control. Accordingly, the fibroblast migration under electrostimulation in the control confirms that the wound closure is not only driven by electrically induced directional cell migration (galvanotaxis) but also by the intrinsic properties of the conductive hydrogels. These results indicate that the presence of conducting hydrogels, particularly clickHA/PEDOT-MeOH/PLA/PEDOT-MeOH, further enhances wound closure. Importantly, this demonstrates that the developed systems for the controlled delivery of broad-spectrum bacteriostatic agents preserve both the biocompatibility and wound healing capability of clickHA/PEDOT-MeOH. On the other hand, the intracellular reactive oxygen species (ROS) levels (Fig. S13) indicate that, as expected, cellular oxidative stress increased over time regardless of the application of electrical stimulation. These results suggest that electrical stimulation did not constitute an additional source of cellular stress under the experimental conditions investigated.

4. Conclusions

Electroresponsive hydrogel platforms based on click-crosslinked HA and electroactive PLA fiber mats have been successfully developed for controlled drug delivery. The incorporation of PEDOT-MeOH and PPy on PLA fibers and semi-interpenetrating the hydrogel matrix enabled the fabrication of electrically responsive systems with enhanced functionality and tunable drug release behavior. The developed hydrogels demonstrated effective CAM loading and controlled release under electrostimulated conditions, while preserving biocompatibility and wound healing capacity. Electrical stimulation primarily influenced the initial burst release, while long-term release profiles remained sustained and comparable across systems. This dual behavior provides opportunities for both immediate and prolonged therapeutic action. Importantly, the released CAM retained its antibacterial activity against both tested gram-negative and gram-positive bacteria, with a more pronounced effect observed for E. coli. Moreover, the bactericidal performance was not significantly affected by stimulation conditions or release time, which confirms the stability and bioactivity of the drug throughout the process. Although the PPy-containing hydrogel exhibited superior wound healing performance, no significant differences were observed between the PEDOT-MeOH- and PPy-based semi-interpenetrating hydrogels with respect to chloramphenicol (CAM) release. Overall, the combination of biocompatible clickHA-based hydrogels, electroactive fiber reinforcement, and stimulus-responsive drug delivery offers a versatile platform for advanced wound healing applications. These systems represent a promising strategy for achieving controlled, on-demand drug release while maintaining sustained antimicrobial efficacy.

CRediT authorship contribution statement

Pilar A. Haro-Gutiérrez: Writing – review & editing, Validation, Investigation, Formal analysis, Data curation. Júlia Sanz-Farnós: Writing – review & editing, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Juan Carlos Ahumada: Writing – review & editing, Methodology, Investigation, Formal analysis. Maria-Pau Ginebra: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization. Maria M. Pérez-Madrigal: Writing – review & editing, Resources, Investigation, Funding acquisition, Formal analysis. Carlos Alemán: Visualization, Supervision, Resources, Methodology, Investigation, Funding acquisition, Formal analysis, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Data 1

Methods and complementary experimental data.

Data availability

All data will be made available on request to the authors.