1. Introduction
Skin wounds represent a global healthcare challenge [
1]. Generally, wound healing is complicated by bacterial infections. In particular, wounds located in mobile body regions (e.g., neck, joints, and extremities) are susceptible to secondary injuries caused by recurrent mechanical stress [
2]. These stressors inevitably prolong healing and exacerbate pain, inflammation, and infections. Therefore, the effective elimination of infections and the promotion of healing remain significant challenges and key research priorities [
3]. Hydrogel patches play an important role in wound management because they maintain a moist wound environment and act as carriers for active substances that support antimicrobial activity and tissue repair [
4], [
5], [
6], [
7]. However, most existing hydrogel patches are designed for treating wounds in static environments and lack mechanical properties compatible with dynamic wounds. Recent advances in mechanically-enhanced hydrogels have primarily focused on passive drug release and antimicrobial activity. Nevertheless, these systems remain limited in addressing complex, dynamic wound conditions [
8]. Consequently, there is a persistent demand for advanced hydrogels that can rapidly adapt to dynamic wounds while enabling precise control of drug release [
9], [
10], [
11], [
12].
In this study, inspired by skin structure and function, a novel mechanically-responsive hydrogel with desired features was developed using liposomes as structural units, as illustrated in
Fig. 1. The skin essentially functions as a natural hydrogel composite in which extracellular matrix (ECM)-connected cells form a network structure that provides mechanical strength [
13], [
14], [
15]. Living cells in this system release bioactive substances that impart hydrogels with biological properties, including antimicrobial activity and immune modulation [
16], [
17], [
18], [
19]. Although some hydrogels, such as polyacrylamide, exhibit favorable mechanical properties, they fail to replicate the complex behavior of skin owing to the insufficient integration of structural and functional attributes [
20]. To accurately mimic skin structure and function, a hydrogel requires three essential components: a matrix, an active carrier, and an interfacial connector between them [
21], [
22], [
23]. Liposomes, which have structures highly similar to that of cell membranes, can respond to physiological stress and modulate active release owing to their softness and deformability [
24], [
25]. Therefore, crosslinking active-carrying liposomes into hydrogel networks represents a promising approach for constructing skin-mimicking hydrogels and may serve as an effective design strategy [
26], [
27], [
28], [
29].
This study presents a mechanically-responsive, skin-mimicking hydrogel prepared using drug-loaded liposomes as structural units (
Fig. 1). Double-bond-modified liposomes were fabricated by inserting distearoylphosphatidylethanolamine polyethylene glycol acrylamide (DSPE-PEG-AM) into the liposomal surface via hydrophobic interactions. Hydrogels featuring a liposome-crosslinked network were formed through radical aqueous polymerization of acrylamide. Integration of polymer chains with deformable liposome crosslinkers significantly enhanced the mechanical properties of the hydrogel compared to those of pure polyacrylamide. In addition, stretching or compressing the hydrogel induced deformation of the embedded liposomes, thereby triggering controlled drug release and imparting the hydrogel with mechanical responsiveness and antimicrobial activity. To evaluate therapeutic potential,
in vivo studies were conducted using both dynamic and static wound models. The hydrogel exhibited antibacterial, anti-inflammatory, and wound-healing activities under both dynamic and static conditions. These findings suggest that the skin-inspired biomimetic paradigm represents a promising strategy for designing hydrogels with tunable structures and functions.
2. Materials and methods
2.1. Materials
DSPE-PEG-AM and SH-PEG1000-rhodamine B (SH-PEG-RhB) were purchased from Shanghai Ponsure Biotech, Inc., China. Lecithin, cholesterol, rifampicin, erythromycin, curcumin, and fluorescein isothiocyanate (FITC) were supplied by Aladdin (China). Acrylamide (99%), N,N′-methylenebisacrylamide (MBAA; 99%), ammonium persulfate (APS; 99.99%), N,N,N′,N′-tetramethylethylenediamine (TEMED; 99%), and 2,2′-azobis(2-methylpropionitrile) (AIBN; 95%) were supplied by Sigma-Aldrich (USA). SYTO 9/propidium iodide (PI) were obtained from Maokangbio (China). The following antibodies were sourced from Servicebio (China): interleukin-6 (IL-6; GB11117), tumor necrosis factor-α (TNF-α; GB115702), and platelet endothelial cell adhesion molecule-1 (CD31; GB15063). Escherichia coli (E. coli; American Type Culture Collection (ATCC) ID. 25922) and Staphylococcus aureus (S. aureus; ATCC ID. 6538) were provided by the Department of Laboratory Medicine in Nanjing Drum Tower Hospital, China.
2.2. Preparation and characterization of rifampin-loaded liposomes with double-bond modification (RLipo-AM)
Liposomes were synthesized using established protocols [
30]. Dissolve phosphatidylcholine, cholesterol, DSPE-PEG-AM, and rifampin in ethanol at a molar ratio of 10:3:2:5 to form a homogeneous solution. And then evaporate the organic solvent in a rotary evaporator with a 40 °C water bath until a uniform film formed. The lipid film was hydrated in phosphate-buffered saline (PBS) buffer in a water bath and resulting dispersion was processed with probe sonication (60 W, 10 spulse, 5 s interval, repeated for 10 min). The dispersion was then extruded through liposome extruder, with 15 extrusion cycles to control liposome size and improve uniformity. Finally, unencapsulated rifampin was remove by dialysis (molecular weight cutoff 8 000-14 000 Da), yielding RLipo-AM, which were stored at 4 °C. The encapsulated drug was quantified based on its ultraviolet (UV) absorbance at 334 nm following liposome lysis with 1% TX-100.
The resulting RLipo-AM and SH-PEG-RhB were reacted at 37 °C for 1 h in the presence of AIBN. Subsequently, flow cytometry (FCM; BD Accuri™ C6 Plus; BD Biosciences, USA) and laser scanning confocal microscopy (LSCM; Nikon, Japan) analyses were conducted to confirm the success of double-bond modification. The hydrodynamic size as well as zeta potential of RLipo-AM were assessed by dynamic light scattering (DLS). The morphology of RLipo-AM was examined using transmission electron microscopy (TEM).
2.3. Preparation of RLipo-AM Gel
The pre-gel solution was prepared by sequentially adding acrylamide (4 mmol), MBAA (0.1% w/w), APS (0.5% w/w), and TEMED (0.5% w/w) to RLipo-AM dispersion at room temperature (a mass ratio of liposome to MBAA of 35:1). Polymerization of the pre-gel solution was carried out in a sealed glass mold at room temperature for 2 h to obtain the RLipo-AM Gel. A polyacrylamide hydrogel (Blank Gel) without the addition of RLipo-AM was used as a control. For visualization, the hydrogel was fabricated with FITC-loaded liposomes and visualized via three dimensional (3D) LCSM.
2.4. Scanning electron microscopy (SEM) analysis of RLipo-AM Gel
Frozen hydrogels were broken to obtain cross sections and then lyophilized. To verify that the liposomes acted as cross-linking points, stretch the sample to 200% of its initial length prior to freeze-drying. All samples were mounted on aluminum stubs and sputter-coated with an approximately 5 nm layer of gold. SEM imaging was performed using a Hitachi S-3000N microscope (Japan) at an accelerating voltage of 10 kV. Representative micrographs were acquired to evaluate the internal morphology and hierarchical structure of the hydrogels.
2.5. Mechanical characterization
All mechanical experiments were conducted on a universal testing machine (INSTRON, Germany). Tensile fracture tests were conducted using rectangular specimens (10 mm × 5 mm × 1 mm) at a loading rate of 200 mm∙min-1, and cyclic tensile testing was conducted at a loading rate of 5 mm∙min-1 and 60% strain amplitude. Compression tests were conducted using cylindrical specimens (10 mm × 2 mm) at a loading rate of 2 mm∙min-1 until the strain reached 80% strain, while cyclic compression tests were conducted at the same loading rate with a 60% strain amplitude.
2.6. Rheological testing
Disc-shaped hydrogel samples (10 mm in diameter, 1 mm in thickness) were prepared for rheological measurements. All experiments were performed on a rotational rheometer (HAAKE MARS; Thermo Fisher Scientific, USA). Strain sweep tests were conducted at 1 Hz over a strain range of 0.1%-1000%. The linear viscoelastic region (LVE) was defined as the strain range in which the storage modulus (G′) decreased by less than 5% relative to the baseline plateau, with the baseline determined as the average of the three lowest strain points. The loss factor (tanδ) was calculated as the ratio of the loss modulus (G″) to the storage modulus (G′). All reported G′ and tanδ values were obtained within this LVE range.
2.7. Tissue adhesion test
The adhesive property of the RLipo-AM Gel was measured using a lap shear test (ASTM F2255) with porcine skin [
31]. Porcine skin was obtained from a local meat market, sourced from commercially slaughtered pigs intended for food use. No animals were sacrificed specifically for this study. All experiments involving porcine skin were conducted following standard biosafety and handling procedures. The skin was carefully cleaned of hair and subcutaneous fat and cut to the required size. A hydrogel sample measuring 10 mm × 10 mm × 1 mm was clamped between two sections of pig skin (30 mm × 10 mm). The ends of pig skin were fixed to the testing machine and stretched at a loading rate of 2 mm∙min
-1 until rupture occurred. Adhesive strength is calculated using the following formula: adhesive strength = maximum adhesive force/contact area.
2.8. Mechanically responsive drug release
Both tensile and compressive cyclic loading tests were conducted to evaluate the drug release of the hydrogel. For tensile cyclic loading, RLipo-AM Gel (60 mm × 25 mm × 2 mm) was secured at both ends using tweezers and immersed in a container with 10 mL of PBS (Fig. S1 in Appendix A). The hydrogel underwent cyclic stretch-release testing at a strain amplitude of 60% and a constant speed of 6 s per cycle. RLipo-AM Gel incubated for the same duration without stretching was served as the control group. After every 50 tensile cycles (5 min for the control group), released medium (1 mL) was sampled and substituted with fresh PBS. Rifampin concentration was determined by absorbance measurement at 334 nm.
For compressive cyclic loading, place RLipo-AM Gel (10 mm × 2 mm) in a 35 mm culture dish containing 2 mL of PBS (Fig. S2 in Appendix A). Cyclic compression was applied to the samples at 60% strain and a speed of 40 mm∙min-1. RLipo-AM Gel incubated for the same duration without compression was served as the control group. After every 50 compression cycles (3 min for the control), release medium (0.5 mL) was sampled. The cumulative drug release was tested using the same method described above.
2.9. Drug penetration assessment
To assess the impact of mechanical stimulation on molecular transport, FITC was employed as a fluorescent model compound in lieu of a therapeutic drug. FITC-loaded hydrogels (10 mm × 10 mm × 2 mm) were placed on fresh, clean porcine skin (20 mm × 20 mm) and subjected to 500 compression cycles at a strain amplitude of 60% and a speed of 20 mm∙min-1. As a control, hydrogels were left stationary on fresh porcine skin at room temperature for the same duration (60 min). Following embedding in optimal cutting temperature compound (OCT), the porcine skin tissues were cryosectioned to yield 10 μm-thick cross-sections. The penetration depth and fluorescence intensity in the porcine skin were observed using a LSCM and analyzed by Image J.
2.10. Cytotoxicity evaluation
RLipo-AM Gel and Blank Gel (100 mg each) were stretched for 250 cycles at a strain of 60% in 2 mL of Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM-F12) complete medium. The solution was sterilized by 0.22-μm membrane filtration. Human umbilical vein endothelial cells (HUVECs) were incubated with the release solution for 24 h. Fresh culture medium was used as control. Cell viability was tested using both Calcein/PI Live/Dead Kit and Cell Counting Kit-8 (CCK-8).
2.11. Hemolysis assays
Fresh whole blood was collected into ethylenediaminetetraacetic acid (EDTA) tubes from healthy rats with ethical approval, and erythrocytes (RBCs) were isolated by centrifugation at 1000g for 5 min. The plasma was discarded and the RBC pellet was washed three times with PBS (pH 7.4) and 2% (v/v) RBC suspension in PBS was prepared. The hydrogel extraction solutions were mixed with an equal volume of 2% RBC suspension (final volume 400 µL per well) in 1.5 mL tube and incubated at 37 °C for 2 h. PBS and 0.1% Triton X-100 served as negative and positive controls, respectively. After incubation, samples were centrifuged at 1000g for 5 min and the supernatant transferred to a 96-well plate. Hemoglobin release was quantified by measuring absorbance (A) at 540 nm using a microplate reader. Percent hemolysis was calculated as: %hemolysis = (Apositive − Anegative)/(Asample − Anegative) × 100.
2.12. Antibacterial ability
A bacterial suspension (500 μL) of E. coli or S. aureus (1 × 106 CFU∙mL-1) was mixed with an equal volume of Luria-Bertani (LB) medium and the extracts of Blank Gel or RLipo-AM Gel, respectively. Hydrogel extracts were prepared by stretching 100 mg of hydrogels for 250 times at 60% strain in 2 mL of LB medium, followed by sterilization through a 0.22-μm sterile filter. After 12-, 48- and 72-h incubation, bacterial growth was assessed by optical density at 600 nm (OD600) measurement, and inhibition rates were calculated. Additionally, diluted bacterial suspensions (100 μL) was smeared onto LB agar plates and incubated inversely for 24 h at 37 °C. Following photography, colony counts were obtained and used to calculate the inhibition rate.
2.13. Biofilm formation experiment
Bacterial suspension (500 μL, 1 × 108 CFU∙mL-1) was inoculated into a 24-well plate with a coverslip and an equal volume of hydrogel extract was then added, and biofilm formation was induced by incubation for 24 h. After staining with SYTO 9/PI, the biofilms were visualized using LSCM imaging.
2.14. Inhibition zone assay
To perform the assay, bacterial suspensions (100 μL, 1 × 108 CFU∙mL-1) are spread onto LB agar plates to form a bacterial lawn. Filter paper disks (approximately 10 mm in diameter) were sterilized by UV irradiation, and then soaked in the hydrogel extract. To evaluate antibacterial efficacy, the soaked filter paper disks were placed on LB agar plates, followed by 24-h incubation and measurement of the inhibition zone diameters.
2.15. In vivo wound infection and healing experiment
The animal experiments were approved by the Animal Investigation Ethics Committee of Nanjing Drum Tower Hospital (approval No. 2024AE02013) and conducted in accordance with ethical guidelines. To evaluate in vivo therapeutic efficacy, a full-thickness excision wound was established in Sprague-Dawley rats (n = 6). A 10-mm circular incision was made along the midline of the skin on the neck and back of anesthetized rats. Subsequently, S. aureus suspension (20 μL, 1 × 108 CFU∙mL-1) was applied to wound site. At 24-h post-infection (designated Day 0), the wound beds were treated with either Blank Gel or RLipo-AM Gel, whereas the control wounds were treated by applying 3 M Tegaderm film (USA). The hydrogels were replaced every two days, with a total of two applications. On Day 2, residual bacteria were collected from the wound sample with a sterile cotton swab, followed by immersed in LB medium (1 mL) and incubated for 1 h. Spread bacterial suspension (100 μL) onto LB agar plates. Photographs of wound were taken on Days 0, 2, 4, 6, 8, and 11. Ex vivo wound samples were fixed for histological analysis and inflammatory factor evaluation.
2.16. In vivo skin safety evaluation
Long-term
in vivo toxicity of the hydrogels was evaluated following previously reported procedures [
32]. Disc-shaped samples (10 mm in diameter) of RLipo-AM Gel were implanted at subcutaneous sites (10 mm incisions) in rats. The negative control was incised only. The incisions were closed with sutures and disinfected with povidone-iodine. Animals were maintained under standard conditions for 21 d, after which they were euthanized and the implantation sites were collected for histological analysis (hematoxylin and eosin (H&E) staining).
2.17. Statistical analysis
Experimental results are expressed as the mean ± standard deviation (SD). Statistical comparisons were made by Student’s t-test and one-way analysis of variance (ANOVA). The following symbols denoted statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001; p > 0.05 was considered not significant (ns).
3. Results and discussion
3.1. Design and preparation of liposome crosslinked hydrogels
As a proof of concept, liposomes encapsulating rifampicin were prepared to serve as drug carriers within the hydrogels. The liposome loading efficiency was approximately 43.7%. To enable crosslinking, the liposomes were surface-modified with olefin groups by incorporating DSPE-PEG-AM into the lipid bilayer via hydrophobic anchoring. Acrylamide-functionalized liposomes (Lipo-AM) were validated using a thiol-Michael addition reaction with rhodamine B-containing sulfhydryl groups (SH-PEG-RhB). Flow cytometry analysis revealed an approximately 2.5-fold increase in the mean fluorescence intensity (MFI) of Lipo-AM compared to that of liposomes without double bonds (Lipo) (
Figs. 2(a) and
(b)). LSCM revealed a distinct red fluorescence emission from Lipo-AM (Fig. S3 in Appendix A). DLS showed that the average diameter of the liposomes was approximately 150 nm, with a slight reduction in size following DSPE-PEG-AM modification, likely due to steric hindrance from the highly hydrated poly(ethylene glycol) (PEG) shell (
Fig. 2(c)). All liposomes exhibited negative zeta potentials, with a modest reduction observed for RLipo-AM, likely due to the electronegative hydroxyl groups of rifampicin (
Fig. 2(d)). The liposome particle size determined by TEM was in general agreement with the DLS results (
Fig. 2(e)).
Lipo-AM Gels were fabricated via free radical polymerization in aqueous solution, utilizing acrylamide as the primary monomer, with Lipo-AM and MBAA as crosslinker, APS as the initiator, and TEMED as the catalyst. Fourier-transform infrared (FTIR) analysis was conducted to confirm the covalent integration of the liposomes into the hydrogel network. The vinyl C-H signals of Lipo-AM at ∼800-900 and ∼910-990 cm
−1 were observable in the Blank Gel and Lipo-AM physical blend sample, however, these signals disappeared in the Lipo-AM Gel (Fig. S4 in Appendix A), indicating the successful occurrence of the Michael-type addition reaction. A polyacrylamide hydrogel without Lipo-AM was used as the control (Blank Gel). This hydrogel can serve as a versatile platform for encapsulating various bioactive molecules, such as rifampicin, erythromycin, curcumin, and the fluorescent probe FITC (Fig. S5 in Appendix A). LSCM demonstrated 3D fluorescence images of the FITC-encapsulated liposome-crosslinked hydrogel (
Fig. 2(f)). SEM analysis demonstrated that both the lyophilized RLipo-AM Gel and control polyacrylamide hydrogels exhibited porous 3D structure at low magnification (
Fig. 2(g), Fig. S6 in Appendix A). At a high magnification (6000×), the RLipo-AM Gel displayed distinctly dispersed nanospheres throughout the cross-section, a feature not observed in the Blank Gel (
Fig. 2(h)).
3.2. Mechanical characterization of hydrogels
Hydrogels with varying MBAA equivalents were prepared, and a 35:1 (RLipo-AM:MBAA) mass ratio was found to provide an optimal balance between the mechanical toughness and network stability (Fig. S7 in Appendix A). The mechanical properties of the RLipo-AM Gel were systematically evaluated and compared with those of the Blank Gel. Visually, the RLipo-AM Gel could be stretched over ten times its original length, even when twisted or knotted. Moreover, the hydrogel was sufficiently stiff to withstand substantial compressive loads without being damaged (
Fig. 3(a)). The RLipo-AM Gel exhibited a significantly higher maximum tensile strain ((1206.39 ± 129.17)%) than that of the Blank Gel ((690.42 ± 166.89)%), indicating an approximately 1.75-fold improvement in flexibility (
Figs. 3(b) and
(c)). Rheological strain sweep analysis was conducted to assess the mechanical reinforcement and energy dissipation capability of the hydrogel. The RLipo-AM Gel exhibited a wider LVE (0.1%-54.6%) and a higher loss factor (tan
δ, 0.10 ± 0.001) compared with those of the Blank Gel (0.1%-29.4%; 0.045 ± 0.004), indicating enhanced network stability and energy dissipation (Fig. S8 in Appendix A). These improvements were attributed to the presence of deformable liposomes as crosslinking points [
33]. Unlike traditional rigid small-molecule crosslinkers, which often undergo sudden fracture under applied stress, deformable liposomes can rearrange and redistribute applied mechanical loads, allowing energy absorption and dissipation over a wider strain range [
34], [
35], [
36]. Similar to the ECM-cell network in skin, membrane fluidity enables cells to undergo deformation in response to stretching forces.
To verify this hypothesis, RLipo-AM Gel strips were lyophilized under 200% tensile strain. SEM analysis revealed that liposomes were deformed into flattened morphologies aligned with the gel stretching direction (Fig. S9 in Appendix A). The RLipo-AM Gel and Blank Gel exhibited similar maximum tensile stresses (
Fig. 3(d)). A reduction in Young’s modulus was observed for the RLipo-AM Gel ((32.97 ± 1.40) kPa) compared to that of the Blank Gel ((53.27 ± 3.40) kPa), with both values being close to Young’s modulus of human skin (Fig. S10(a) in Appendix A). The compressive properties of the hydrogels displayed similar trends (
Figs. 3(e) and
(f), Fig. S10(b) in Appendix A). At 80% compressive strain, the compressive stress and modulus of the RLipo-AM Gel were (538.17 ± 62.71) kPa and (136.00 ± 28.92) kPa, respectively, which were slightly lower than those of the Blank Gel. In addition to its excellent tensile properties, the RLipo-AM Gel exhibited high elasticity and rapid recovery. Cyclic stretching studies were conducted to assess energy dissipation of the Rlipo-AM Gel. The hydrogel exhibited minimal curve deviations and negligible hysteresis during cyclic loading (
Fig. 3(g)). The observed hysteresis loops were likely due to the deformation and recovery of the liposome crosslinking sites, which consume a portion of the energy [
37]. Overlapping hysteresis loops during compression loading/unloading indicated that the hydrogel recovered its original shape after undergoing a certain degree of deformation (
Fig. 3(h)).
3.3. Adhesion performances of hydrogels
Compared with conventional dressings, hydrogels with appropriate tissue adhesion offer clear advantages for wound treatment [
38], [
39]. The adhesion of the hydrogel to finger joints at various angles and to porcine skin under twisting and bending are illustrated in
Fig. 3(i). Shear adhesion of the hydrogel to the porcine skin was evaluated using a tensile testing machine, as schematically illustrated in
Fig. 3(j). The RLipo-AM Gel exhibited enhanced adhesion strength compared to that of the Blank Gel (
Figs. 3(k) and
(l)), likely resulting from liposome-induced increases in surface energy and enhanced intermolecular interactions (e.g., van der Waals forces and hydrogen bonding) at the hydrogel-skin interface. The hydrogel exhibited an adhesion strength of approximately 0.43 N (per 1 cm
2 contact area), which exceeds its self-weight by over 400-fold, enabling robust adhesion even on complex curved surfaces and during dynamic motion.
3.4. Mechanoresponsive drug release and cytotoxicity evaluation of hydrogels
Mechanical stimulation enables the immediate release of drugs in response to external forces (
Fig. 4(a)). This approach eliminates the risk of chemical toxicity and side effects and offers improved biosafety compared to passive drug delivery or traditional chemical and biological stimulation methods [
40], [
41]. In this study, the drug-release behavior of the hydrogels was evaluated under various mechanical conditions (Figs. S1 and S2). When the hydrogel was subjected to 250 tensile cycles at 60% strain (equivalent to the strain experienced during human finger joint flexion), approximately 39.6% of the drug was released. This value was significantly higher than the static release rate of 17.6% (
Fig. 4(b)). The sustained burst-free drug release profile indicated that the liposomes remained stable within the hydrogel network, which is consistent with previous reports on PEGylated liposomes integrated into hydrogel networks [
42], [
43].
To further clarify the release mechanism, the cumulative release profiles were fitted using zero-order, Higuchi, and Korsmeyer-Peppas models (Fig. S11 in Appendix A). As summarized in Table S1 in Appendix A, drug release in the absence of mechanical stimulation (no mech) mainly followed a Fickian diffusion process (n = 0.423), indicating diffusion-controlled release. Under mechanical stimulation (mech), the release rate constant (k) increased markedly. The higher Korsmeyer-Peppas exponent (n = 0.572) reflected a transition to anomalous transport involving both diffusion and polymer relaxation. These results demonstrate that mechanical stimulation accelerates drug diffusion and enables tunable mechanoresponsive release behavior.
Similarly, compressive deformation facilitates drug release. Under 60% compressive strain, corresponding to a compressive stress equivalent to a plantar pressure of approximately 120 kPa, the hydrogel released approximately 8.4% of the drug over 250 cycles. This value was notably higher than the passive release level of 3% observed in the absence of compression during the same 15-min period (
Fig. 4(c)). To better simulate the dynamic mechanical environment, cyclic “stimulus-rest” experiments were performed. The RLipo-AM Gels were subjected to repeated stretching cycles, followed by rest periods, and the drug release was monitored over multiple cycles (Fig. S12 in Appendix A). Each tensile cycle triggered an accelerated release of the encapsulated drug, while periods of rest resulted in slower release. These results demonstrate the reversible and controllable mechanoresponsive behavior of the hydrogel. In summary, mechanically-responsive hydrogels can regulate drug release in dynamic wound environments, thereby providing an advanced and promising strategy for dynamic wound treatment.
In vitro drug permeability of the hydrogel through porcine skin was further evaluated by encapsulating FITC as a fluorescent probe in the liposomal units. Confocal fluorescence imaging of porcine tissue sections revealed a significant increase in the average penetration depth to 617.5 μm after 500 cycles of 60% compressive strain, compared to 254.8 μm for the sample without compression (Figs. 4(d)-(f)). These findings demonstrated that mechanical stress on the hydrogel promoted drug release and facilitated deeper tissue penetration. The cell viability test demonstrated that neither the Blank Gel nor the RLipo-AM Gel induced cytotoxicity in HUVEC, revealing their superior cell compatibility (
Figs. 4(g) and
(h)). To further assess biocompatibility, hemolysis assays were performed. All hydrogel samples exhibited hemolysis rates below 1%, indicating excellent hemocompatibility (Fig. S13 in Appendix A). Collectively, these results demonstrate favorable biocompatibility of the hydrogels.
3.5. Antibacterial ability of hydrogels
Bacterial infections, especially associated with biofilms, represent the most serious challenges in wound treatment [
44]. Therefore,
in vitro antimicrobial activity of the RLipo-AM Gel was further evaluated at 12, 48, and 72 h. As anticipated, compared to the control group, the RLipo-AM Gel-treated group exhibited strong antibacterial efficacy, effectively inhibiting the growth of both
E. coli and
S. aureus (
Figs. 5(a)-(c), Figs. S14 and S15 in Appendix A). In contrast, the Blank Gel and the liposome-only hydrogel group (Lipo-AM Gel) showed negligible antibacterial activity at 48 and 72 h (Fig. S15 in Appendix A), indicating that rifampin is the primary source of antibacterial activity.
The strong antimicrobial activity of RLipo-AM Gel prompted further evaluation of its efficacy against biofilm formation. Live/dead staining demonstrated that the drug solution released from the stretched RLipo-AM Gel effectively inhibited biofilm formation, whereas treatment with Blank Gel failed to completely prevent biofilm formation (
Fig. 5(d)). In contrast, dense, mature biofilms were observed in the control group. Biofilm inhibition at 24, 48, and 72 h was further confirmed by crystal violet staining. The results demonstrated that the drug solution released from the stretched RLipo-AM Gel eradicated biofilms by killing more than 95% of bacteria (
Figs. 5(e) and
(f), Fig. S16 in Appendix A). Bacterial growth inhibition was assessed using the paper-diffusion method. The drug released from the stretched RLipo-AM Gel exhibited the strongest antibacterial activity, as evidenced by the largest inhibition zones (10.1 mm for
E. coli; 29.5 mm for
S. aureus) compared with those of PBS and Blank Gel treatments, highlighting its outstanding antimicrobial efficacy (
Fig. 5(g), Fig. S17 in Appendix A).
3.6. In vivo anti-infection and wound treatment
Previous studies on wound dressings have largely overlooked the necessity for their application in body regions that undergo stretching. The dorsal region of rats exhibits minimal movement and was therefore used as a static wound model [
45]. In contrast, the neck region frequently experiences compression, stretching, and twisting motions, making it a suitable model for representing wounds in stretchable areas [
45], [
46], [
47]. Accordingly,
S. aureus-infected full-thickness skin wound models were developed on the neck and dorsa of rats to represent dynamic and static wounds, respectively, and were used to evaluate
in vivo antibacterial and wound-healing effects of the RLipo-AM Gel (
Fig. 6(a)). Compared to the control and Blank Gel, treatment with the RLipo-AM Gel resulted in a significant reduction in
S. aureus colonies that were recovered from infected tissues on Day 2 (
Fig. 6(b), Fig. S18 in Appendix A). Concurrently, a marked reduction in the infiltration of inflammatory factors (including IL-6 and TNF-α) was observed at the wound site in the RLipo-AM Gel group (
Fig. 6(c)).
Wound size was monitored by photographic imaging on Days 0, 2, 4, 6, 8, and 11 post-treatment at both the nape and dorsal sites in the rats (
Figs. 6(d) and
(e)). Quantitative assessment of the wound area revealed significant differences between the RLipo-AM Gel group and the other groups (
p < 0.05, particularly on Days 2-6), indicating accelerated wound healing in the RLipo-AM Gel group at identical time points (
Figs. 6(f) and
(g)). This effect was more pronounced during the early stages of wound healing and may be attributed to the ability of RLipo-AM Gel to effectively eliminate bacterial infections. Moreover, the RLipo-AM Gel promoted wound healing in both static and dynamic wound models. These results validate the superior antibacterial properties and wound healing ability of the RLipo-AM Gel.
To further evaluate the microscopic aspects of wound healing, histological analysis of tissues on Day 11 post-treatment was conducted. H&E staining showed that the RLipo-AM Gel group promoted wound closure, as evidenced by smaller epithelial gaps and a higher degree of epithelialization compared to the other groups (
Figs. 7(a) and
(b)). Masson’s trichrome staining revealed a significant increase in the newly formed collagen deposition in the RLipo-AM Gel-treated group (
Figs. 7(c) and
(d)). Angiogenesis, a key indicator of tissue regeneration, was evaluated using CD31 immunofluorescence staining to assess neovascularization across different groups. As shown in
Figs. 7(e) and
(f), the RLipo-AM Gel treatment group exhibited increased neovascular density. Notably, these indicators displayed consistent trends in both static and dynamic wound models. Long-term cutaneous toxicity evaluation demonstrated no apparent tissue abnormalities or significant inflammatory responses compared to the control group (Fig. S19 in Appendix A). Additionally, the safety of the hydrogel was confirmed by H&E staining of the major organs, which revealed no evidence of systemic toxicity (Fig. S20 in Appendix A). Collectively, these findings confirm that the RLipo-AM Gel enhances tissue regeneration and remodeling in both infected static and dynamic wounds.
4. Conclusion
A novel anti-infective mechanoresponsive hydrogel inspired by skin structure and function was developed for loading antimicrobial agents with mechanically-triggered release. Divinyl-functionalized liposomes were employed as crosslinkers to construct a hydrogel network through free-radical polymerization of acrylamide. Unlike conventional mechanically-responsive hydrogels or systems based on physically entrapped liposomes [
48], [
49], [
50], [
51], [
52], the liposome-crosslinked network covalently integrates DSPE-PEG-AM-modified liposomes as nano-crosslinkers. This design establishes a hierarchically organized soft-hard coupled architecture that enhances energy dissipation and enables translation of external mechanical stimuli into controllable drug release. This hierarchically organized soft-hard structure improves mechanical adaptability and supports controllable drug release upon mechanical deformation, representing a significant step toward mechanically robust and functionally responsive hydrogels.
The results demonstrate that the RLipo-AM Gel exhibits excellent mechanical performance, including high tensile strength, elasticity, rapid recovery, and a Young’s modulus comparable to that of skin. The RLipo-AM Gel also showed appropriate tissue adhesion and superior biocompatibility. Additionally, the RLipo-AM Gel enabled modulation of drug release in response to mechanical stimuli, including stretching and compression, demonstrating strong in vitro antimicrobial activity and anti-biofilm formation capabilities. When treating infected full-thickness wounds at both nape and dorsal sites in rats, RLipo-AM Gel effectively eradicated the infection in the early stages and subsequently promoted epithelialization, collagen deposition, and angiogenesis, thereby accelerating healing of dynamic neck wounds.
This design concept offers considerable potential for clinical applications. Chronic wounds located in mechanically dynamic regions, such as the elbow, knee, and other joint areas, are particularly challenging to manage [
53], [
54]. This is because conventional dressings often cannot accommodate frequent deformations and frequently fail to retain therapeutic agents [
55], [
56]. The mechanoresponsive characteristics of this hydrogel provide a distinct advantage in such settings, whereby physiological movements can actively facilitate controlled drug release while maintaining conformal adhesion and structural stability. In summary, this study presents a versatile mechanoactive hydrogel platform with strong potential for treating various infection-prone wounds that are subjected to mechanical stress.
PDF
(6318KB)
