Multifunctional nanocomposites mediated novel hydrogel for diabetic wound repair

Abstract

The regeneration and repair of diabetic wounds, especially those including bacterial infection, have always been difficult and challenging using current treatment. Herein, an effective strategy is reported for constructing glucose-responsive functional hydrogels using nanocomposites as nodes. In fact, tannic acid (TA)-modified ceria nanocomposites (CNPs) and a zinc metal–organic framework (ZIF-8) were employed as nodes. Subsequent crosslinking with 3-acrylamidophenylboronic acid achieved functional nanocomposite-hydrogels (TA@CN gel, TA@ZMG gel) by radical-mediated polymerization. Compared with a simple physically mixed hydrogel system, the mechanical properties of TA@CN gel and TA@ZMG gel are significantly enhanced due to the intervention of the nanocomposite nodes. In addition, this kind of nanocomposite hydrogel can realize the programmed loading of drugs and release of drugs in response to glucose/PH, to coordinate and promote its application in the regeneration and repair of diabetic wounds and infected diabetic wounds. Specifically, TA@CN gel can remove reactive oxygen species and generate oxygen through its various enzymatic activities. At the same time, it can effectively promote neovascularization, thus promoting the regeneration and repair of diabetic wounds. Furthermore, glucose oxidase-loaded TA@ZMG gel exhibits glucose response and pH-regulating functions, triggering programmed metformin (Met) release by degrading the metal–organic framework (MOF) backbone. It also exhibited additional synergistic effects of antibacterial activity, hair regeneration and systemic blood glucose regulation, which make it suitable for the repair of more complex infected diabetic wounds. Overall, this novel nanocomposite-mediated hydrogel holds great potential as a biomaterial for the healing of chronic diabetic wounds, opening up new avenues for further biomedical applications.

---

Introduction

Diabetes is a chronic disease characterized by persistent hyperglycaemia. Diabetic wounds,^1–3 as one of its main complications, exhibit complex features of delayed healing, which differ from the four processes⁴ of hemostasis, inflammation, proliferation, and re-epithelialization in normal wound healing. Meanwhile, pro-inflammatory cytokines and reactive oxygen species^5,6 are overexpressed in the wound, and the hypoxic microenvironment hinders neovascularization and epithelialization.^7–9 Additionally, the high sugar microenvironment on the wound is more likely to lead to the growth of bacteria^10,11 and thus to the formation of infected diabetic wounds. It further increases the complexity and difficulty of regenerative repair, which greatly reduces the quality of life of patients. Therefore, there is an urgent need to develop novel wound treatment strategies with high efficiency and synergistic effects to overcome the limitations of treating refractory diabetic wounds or infected diabetic wounds.

Compared with the direct use of traditional wound microenvironment regulation (anti-inflammatory and antioxidant), pro-angiogenesis and antibacterial drugs,^12–14 the application of wound repair based on nanomaterials and nano-drug delivery systems^15–17 shows superior therapeutic prospects. This is because the latter may have the following advantages and effects:^18–20 the multienzyme and other multifunctional activities of the nanocomplex itself; the special properties of nanocarriers of good drug loading and release in response to the microenvironment. In particular, nanoceria not only has a variety of enzymatic activities to achieve the clearance of reactive oxygen species,^21–23 but it can also effectively promote angiogenesis,^24–26 which has potential application value in wound regeneration and repair. We also reported earlier^27,28 that a ceria-based nanoenzyme complex can promote the regeneration and repair of diabetic wounds to a certain extent. At the same time, the classic MOF (ZIF-8) has potential application value in more complex diabetic wound regeneration and repair due to its excellent drug loading^29,30 and pH-responsive ability to release drugs. However, the use of a nano-enzyme or nano-drug delivery system in skin wound repair has always faced the problems of effective retention of drugs in the wound and reduction in drug efficacy.^31,32 This is mainly due to the lack of appropriate excipients designed to maintain their effective retention and responsive release^33,34 on the wound surface, so it is difficult to achieve the synergistic effect of multiple drugs and maintain the high efficiency and long-term performance of the drugs. Therefore, there is an urgent need to design and construct a new excipient strategy to effectively integrate nano-enzyme complexes or nano-drug delivery systems in order to produce a better type that could play a synergistic role in refractory diabetic wounds or infected diabetic wounds.

Hydrogels possess excellent hydrophilicity and a three-dimensional (3D) porous structure similar to the extracellular matrix (ECM),^35–37 which can absorb wound exudates, maintain moisture at the wound site, and provide suitable modulus-matched soft tissue.^38–40 Therefore, they are an excellent comprehensive platform for treating diabetic wounds. Among them, nanocomposite-hydrogels^41,42 have received more and more research attention. At present, they are mainly constructed in the following ways:^43–47 (1) the synthesized nanocomposite is formed by physical mixing with a hydrogel; (2) at low temperature, the synthesized nanocomposite is mixed with temperature-sensitive hydrogel, and then heated to form a nanocomposite-hydrogel; (3) a nanocomposite-hydrogel is synthesized by generating reactive nanoparticles in a preformed gel. However, these methods encounter potential drawbacks,^48,49 including: (1) inherent instability of the nanocomposites, leading to particle clustering; (2) relatively weak gelation properties; (3) challenging control over their release from the gel. Therefore, the exploration of a novel and efficient strategy to realize the organic integration of nanocomplexes and hydrogels for responsive and programmed therapy, as well as their comprehensive utilization in regenerative repair for challenging diabetic wounds, is of paramount urgency and significance.

Based on the above considerations, we constructed a novel nanocomposite hydrogel strategy for the treatment of diabetic wounds or infected diabetic wounds (Scheme 1). We selected tannic acid (TA)-modified ceria nanocomposites (CNPs) and a zinc metal–organic frame (ZIF-8) as nodes. Subsequent crosslinking with 3-acrylamidophenylboronic acid yielded glucose-responsive nanocomposite-hydrogels (TA@CN gel, TA@ZMG gel) by radical-mediated polymerization. Compared with a simple physically mixed hydrogel system, TA@CN gel and TA@ZMG gel not only showed better gelation properties, but could also realize the programmed loading and controlled release of drugs. In vitro and in vivo experiments demonstrated that TA@CN gel exhibited abundant enzymatic activity and oxygenation capacity, leading to enhanced collagen deposition and improved angiogenesis, to promote the regeneration and repair of diabetic wounds. More importantly, TA@ZMG gel loaded with metformin and glucose oxidase (GOX) generated glucose/pH-responsive drug release. It also exhibited additional synergistic effects of antibacterial activity, hair regeneration and systemic blood glucose regulation, which make it suitable for the regeneration and repair of more complex infected diabetic wounds. The above results suggest that the functional hydrogel synthesis method designed in this study holds promise as a prospective approach for combating various chronic wound healing processes.

Scheme 1 Synthesis strategy of glucose-responsive intelligent hydrogels based on multifunctional nanocomposite nodes and their mechanism in diabetic and infected diabetic wounds.

Experimental section/methods

Materials

Cerium(III) nitrate hexahydrate (Ce(NO₃)₂·6H₂O), dioctyl sulfosuccinate sodium salt (AOT), polyvinylpyrrolidone (PVP), 2-methylimidazole (HMIM), zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), potassium ferrocyanide (K₄[Fe(CN)₆]·3H₂O) and streptozotocin (STZ) were provided by Sigma-Aldrich (USA). Tannic acid (TA), metformin hydrochloride (Met), 3-acrylamidophenylboronic acid (APBA), acrylamide (AM, 99%), ammonium persulfate (APS, 98%), and N,N,N′,N′-tetramethylethylenediamine (TEMED, 99%) were supplied by Aladdin (Shanghai, China). Glucose oxidase (GOX), probucol (PB) and nitrotetrazolium blue chloride (NBT) were purchased from TCI (Japan). Fetal bovine serum (FBS), penicillin and streptomycin, trypsin and Dulbecco's modified Eagle's medium (DMEM) were purchased from Gibco (USA). A live/dead kit, reactive oxygen detection kit and crystal violet were obtained from Beyotime (China) Millipore deionized water (18.2 MΩ cm) was used in all experiments.

Synthesis of TA@NPs

TA@CNPs: ceria nanocomposites were prepared by micro lotion polymerization. Initially, AOT (0.75 g) was dissolved in toluene (50 mL). Afterward, Ce (NO₃)₂·6H₂O (0.01 g) was added to the solution, which was then magnetically stirred at room temperature for 45 minutes. Subsequently, hydrogen peroxide (H₂O₂) (30%, 5 mL) was gradually added dropwise to the mixture while continuing magnetic stirring for 2 h. The transparent solution gradually transformed into a yellowish emulsion. After resting overnight, the solution formed two layers, with the upper layer consisting of ceria nanocomposites dispersed in the AOT phase. In parallel, TA (10 mg) was dissolved in deionized water (3 mL), followed by aspirating 6 mL of the upper layer into the tannic acid solution. Sodium carbonate (120 mg) was added to adjust the pH to approximately 11. The resulting mixture was stirred in an oil bath at 80 °C for 12 h. After standing and layering, the lower aqueous phase was purified through centrifugation at 5000 rpm.

Synthesis of TA@ZIF-8/Met NPs. In a typical experiment, ZIF-8/Met was synthesized using a one-pot encapsulation method. HMIM (180 mg) and Zn(NO₃)₂·6H₂O (20 mg) were separately dispersed in 1 mL of methanol solvent. Then, 3.6 mL of metformin (12.9 mg) and HMIM (180 mg mL⁻¹ in methanol) were mixed and stirred for 5 min. Next, Zn(NO₃)₂·6H₂O (20 mg mL⁻¹ in methanol) was quickly injected into the mixed solution. The mixed system was stirred at room temperature for 48 h. Finally, the product was purified by centrifugation at 7000 rpm for 5 min and washed with methanol 3 times. ZIF-8/Met was collected and dispersed in an aqueous phase. TA (10 mg) was dissolved in deionized water (1 mL). ZIF-8/Met (2 mL) was mixed into the above solution and the pH of the solution was adjusted to 11. The reaction mixture was stirred at room temperature for 2 h and purified by centrifugation at 5000 rpm.

Synthesis of TA@PBNPs. Briefly, K₄[Fe(CN)₆]·3H₂O (55 mg), PVP (1.9 g) was added to HCl solution (0.1 M, 25 mL). The resulting mixture was stirred at 80 °C for 24 h and subsequently purified through centrifugation at 10 000 rpm, followed by three washings with ethanol. Afterwards, TA (10 mg) was dissolved in deionized water (1 mL). PB (2 mL) was mixed into the above solution and the pH of the solution was adjusted to 11. The reaction mixture was stirred at room temperature for 2 h and purified by centrifugation at 10 000 rpm.

Characterization of nanocomposites

Fourier-transform infrared (FTIR) spectroscopy was performed using an FTIR spectrometer (Nicolet iS5, Thermo Fisher Scientific, USA), while ultraviolet-visible (UV-Vis) absorption spectroscopy was performed on a UV-vis spectrophotometer (Ultra-6600A, Rigol, China) to verify the successful chelation of tannic acid and nanocomposites. The average particle sizes of the nanocomposites before and after modification were measured with dynamic light scattering (DLS) (Zetasizer Nano ZS ZEN3600, UK). The microstructures of the CNPs and PB were analyzed using a scanning electron microscope (SEM) (Hitachi SU8020, Japan), and the morphological size of the ZIF-8 was observed with transmission electron microscopy (TEM).

Synthesis of TA@NPs gel

TA@NPs gel was prepared by the following steps. Typically, the above synthesized TA@NP solution was vacuum lyophilized. TA@NPs (1.8 mg) and APBA (22.35 mg) were first separately dissolved in DMSO (500 μL), and then the mixture was vortexed for 30 s and left to stand for 15 min. Second, 100 μL of the above mixture was aspirated, and third, AM solution (870 μL, 7 mol L⁻¹), APS (100 μL, 3.76 mmol L⁻¹) and TEMED (10 μL) were added to the mixture solution to form a hydrogel. The free radical polymerization reaction was carried out at room temperature with a reaction time of 1–2 min.

Characterization of TA@NPs gel

The microstructures of TA@CN gel, TA@ZMG gel and TA@PB gel were analyzed with SEM and TEM. Subsequently, an elemental scanning analysis was performed. A modular compact rheometer (MCR 302, Anton Paar, Austria) was used to evaluate the rheological properties of the hydrogels. The experiment was performed using a time sweep test at 37 °C with a frequency of 10 rads⁻¹ and 1% strain. Strain amplitude sweep tests (γ = 1–1000%) were adopted to detect the critical strain point of TA@NPs gel. Next, the self-healing property of the hydrogel was investigated through 5 cycles of strain amplitude scanning tests. A tensile stress test was carried out first to evaluate the tensile properties of the hydrogel. Then, the adhesion performance of the hydrogel was evaluated using different materials such as tubes, metal and glass slides. In order to study the swelling property of the hydrogel, the hydrogel was immersed in PBS and weighed at 0, 2, 4, 6, 8, 10, 12 h. The swelling ratio was calculated with eqn (1):

Swelling ratio (%) = (W_t − W₀)/W₀ × 100 (1)

The rich enzymatic activity of ceria nanoparticle^50,51 allows the hydrogel to produce scavenging oxidative stress and generate oxygen. The catalase (CAT) enzyme-mimicking activity of TA@CN gel was verified by monitoring the oxygen content generated by using a portable dissolved oxygen tester (DOT Ray Magnetic JPBJ-608, China). Different control groups (H₂O, gel + H₂O, gel + H₂O₂, TA@CN gel + H₂O, TA@CN gel + H₂O₂) were set up and oxygen levels were measured at 0, 10, 20, 30 and 40 min under magnetic stirring. Wherein, H₂O₂ is 10 M, 200 μL. Next, ethylene diamine tetraacetic acid (EDTA, 800 μL, 0.1 mol L⁻¹), NBT (300 μL, 2 mmol L⁻¹), riboflavin (200 μL, 1.2 mmol L⁻¹) and PB (11.4 mL, 10.0 mmol L⁻¹) were mixed to prepare a superoxide dismutase (SOD) enzyme working solution for evaluation of SOD enzymatic activity. PBS, positive control and different concentrations of hydrogel extracts (50 μL) were added to a 96-well plate, and then the working solution (100 μL) was added and protected from light, and the absorbance at 560 nm was measured by a micropore structure analyzer. The inhibition rate of free radicals was calculated using eqn (2):

A = (A_PBS − A₀)/A_PBS × 100 (2)

The inhibition rate of ˙OH radicals by TA@CN gel was determined with H₂O₂ (10 mM) and FeSO₄ (1 mM) solutions. First, gel and TA@CN gel (0.6 g) were weighed separately, and a blank control group and a positive control group were set up. H₂O₂ (500 μL) and FeSO₄ (500 μL) were added to each group, and incubated at 37 °C for 1 h. After cooling down to room temperature, the supernatant of each sample (100 μL) was aspirated and mixed with 3,3′,5,5′-tetramethylbenzidine (TMB) colorimetric solution (1 mM, 100 μL), and the absorbance was measured by the microporous structure analyzer at 650 nm.

Glucose and pH response of TA@ZMG gel

To observe the glucose-responsive degradation behavior of TA@ZMG gel, the hydrogels were immersed in 4 mg mL⁻¹ glucose solution for 2 h. The storage modulus (G′) and loss modulus (G′′) of TA@ZMG gel before and after degradation were measured using a modular compact rheometer with time scan frequency of 10 rad s⁻¹ at 0.5% stress. Similarly, the hydrogels before and after immersion were lyophilized and the microstructures were observed under SEM.

After release from glucose-responsive TA@ZMG gel, GOX undergoes a catalytic oxidation reaction in a high-sugar environment, producing H₂O₂ with antimicrobial capacity, which is detected with a reactive oxygen species detection kit dichlorodihydrofluorescein diacetate (DCFH-DA) (10 μM). Gel and GOX@Gel were each soaked with glucose solution for 12 h, and 500 μL of the leachate was mixed with 10 μL of working solution, and 100 μL of the mixture was added to each well in a 96-well plate for co-incubation for 20 min. The intensity of fluorescence before and after stimulation was detected in real time or time point by time point using a 488 nm excitation wavelength and a 525 nm emission wavelength. The other GOX-catalyzed product, glucuronide, is pH-responsive, so in order to investigate the changes in pH⁵² caused by GOX loaded in TA@ZMG gel, TA@ZMG gel and gel were incubated in glucose solution for 24 h and the pH was measured with a pH meter (Mettler Toledo, USA) at various time points.

MOF structure disintegration and drug release

To trace the disintegration of ZIF-8 nanocomposites due to GOX-induced pH reduction, HCl (0.1 M, 20 μL) was added to the ZIF-8@Met solution. The morphology of ZIF-8 before and after the addition of HCl was observed by SEM. The release of metformin loaded in ZIF-8 was evaluated by high performance liquid chromatography (HPLC). TA@ZMG gel was immersed in 4 mg mL⁻¹ glucose solution and 200 μL of the leachate was removed at different time points and supplemented with 200 μL of fresh glucose. The leachate was assayed with methanol and water as mobile phases.

Glucose-responsive degradation of hydrogels

To examine the glucose response properties of hydrogels, first, the hydrogels were immersed in different concentrations of glucose solution (0, 1, 4 and 10 mg mL⁻¹) and the degradation behavior was photographed and recorded after 12 h. Secondly, time-scan tests were performed using a modular compact rheometer at 37 °C with a frequency of 10 rad s⁻¹ and 1% strain.

The time required for complete degradation of the hydrogels was simulated and tested in vitro. TA@CN gel and TA@ZMG gel were each immersed in 4 mg mL⁻¹ glucose solution, and shaken in a constant-temperature shaker at 37 °C for 10 d. The samples were lyophilized after being removed at different time points. The mass residue ratio was calculated from eqn (3):

Mass ratio (%) = W_t/W₀ × 100 (3)

where W₀ is the initial weight of the hydrogel and W_t is the weight of the lyophilized sample measured at different time points.

In vitro cytotoxicity of TA@CN gel and TA@ZMG gel

The cytotoxicity of TA@CN gel and TA@ZMG gel was assessed with the Cell Counting Kit-8 (CCK-8) cell viability kit assay. In short, human umbilical vein endothelial cells (HUVECs) (Sciencell, USA) were seeded in a 96-well plate and cultured for 24 h and then treated with different concentrations of TA@CN gel (6.25, 12.5, 25, 50, 100, 200, 400 mg mL⁻¹) and TA@ZMG gel (equivalent to TA@CN gel) for 8 h. Subsequently, the cells were incubated in CCK-8 (MCE, USA) and fresh medium for 2 h and cell viability was determined using a Varioskan LUX microplate reader (Thermo Fisher Scientific, USA). Live/dead kits were used to assess the viability of HUVECs in different treatment groups (PBS, gel, TA@CN gel, TA@ZMG gel), imaged under a Nikon Eclipse Ti–S inverted fluorescence microscope (Nikon Corporation, Japan).

In vitro cellular evaluation of TA@CN gel

Intracellular CAT enzymatic activity was evaluated using a tris(2,2-bipyridine) ruthenium chloride (RDPP) fluorescent probe. HUVEC cells pre-cultured hypoxically were incubated with RDPP dye (3 μM) for 3–4 h, followed by the addition of 150 μL of different samples (PBS, H₂O₂, gel, TA@CN gel). Except for the PBS group, H₂O₂ (100 mM, 3 μL) was added to each group and incubated for 0.5 h before using a Nikon Eclipse Ti–S inverted fluorescence microscope. In order to study the capacity of TA@CN gel to scavenge intracellular reactive oxygen species (ROS), HUVECs were inoculated in a 6-well plate. Then, after stimulation with H₂O₂ (100 μM), different sample groups (PBS, H₂O₂, gel + H₂O₂, TA@CN gel + H₂O₂) were added and incubated for 12 h. After incubation with DCFH-DA (10 μM) for 30 min, the final fluorescence images were taken using an inverted fluorescence microscope.

HUVEC cells were first cultured in 6-well plates to the appropriate cell density, and then traced along a straight line with a 10 μL pipette. After removing the cell fragments, different treatment groups (PBS, gel, TA@CN gel) were incubated for 24 h. The cell migration was observed at different time points under the microscope, and the migration rate was calculated using ImageJ software to simulate the scratch area before and after cell migration. The mobility was calculated with eqn (4):

MR = (A₀ − A_t)/A₀ × 100 (4)

where A₀ and A_t represent the area at 0 h and t h, respectively.

Transwell (Corning, USA) migration assays were performed to test the migration ability of HUVECs cultured in different leachates (PBS, gel, TA@CN gel). HUVECs were inoculated into 24-well plates, and 5% hydrogel extract was added to the lower chamber. After incubation for 24 h, the upper chamber was taken out and fixed with paraformaldehyde for 30 min, and dyed with 800 μL of crystal violet at room temperature for 20 min. Finally, a Nikon Eclipse Ti–S inverted fluorescence microscope was used to take fluorescence images.

To observe the effect on angiogenesis, the matrix gel was spread in 24-well plates and after 30 min of solidification, pretreated HUVEC cells from different sample groups were inoculated into the well plates. Images of tubule formation at different times were taken using light microscopy, and the number and length of tubules were quantified using ImageJ software.

In vivo healing of full-thickness diabetic wound by TA@CN gel

Male BALB/c mice (20–25 g) were provided by the Laboratory Animal Center of Chongqing Medical University (SYXK2022-0016). All animal experiments were approved by the Animal Ethics Committee of Chongqing Medical University (Chongqing, China). To establish a diabetes model, mice were injected with streptozotocin (STZ) at a dose of 50 mg kg⁻¹ for 5 consecutive days after 2 weeks of continuous feeding on high-fat chow. After monitoring blood glucose levels above 16.8 mmol L⁻¹ for one week, the diabetic mouse model was confirmed to be successful. After intraperitoneal injection of pentobarbital sodium (0.3% per 100 g), a circular wound with a diameter of 0.8 cm was established on the back of the mouse. Subsequently, the mice were divided into three groups (PBS, gel, TA@CN gel) and given different treatments. During the treatment, wound healing photograph of different treatment groups were taken at 0, 3, 5, 7, 10, 14 days and quantified using ImageJ. At the same time, mouse body weight and blood glucose data were collected every 2 days for two weeks.

In vitro antibacterial evaluation of TA@ZMG gel

The antibacterial ability of TA@ZMG gel against Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) was evaluated by measuring absorbance and colony count. The bacteria were first inoculated in Luria–Bertani (LB) medium, and 0.2 g of gel, TA@ZMG gel were each added to 2 mL of bacterial suspension, and incubated at 37 °C for 24 h. Then, the absorbance was measured at 600 nm. The relative viability of the bacteria was calculated using eqn (5):

Relative viability (%) = (A_sample − A_blank)/(A_PBS − A_blank) × 100 (5)

The bacterial suspensions of different samples were diluted separately with LB medium, and 20 μL of dilution was taken, coated on agar plates and incubated at 37 °C for 24 h. After the colonies had grown, the plates of different sample groups were photographed.

TA@ZMG gel in vivo assessment of healing of infected diabetic wounds

Encouraged by the effectiveness of TA@CN gel in diabetic wound healing, we wanted to validate the repair ability of TA@ZMG gel, so we established a diabetic infection model with reference to the above method. Based on the above in vitro antimicrobial assay results, TA@ZMG gel has outstanding antimicrobial potential, so we further extended its application evaluation in infected diabetic wounds. The bacterial infection model was successfully established after adding 50 μL of pre-cultured bacterial solution to diabetic mouse wounds and observing yellow crusting. The mice were divided into three groups and given different treatments (PBS, gel, TA@ZMG gel). After 3 days, wound bacteria from different treatment groups were transferred to LB medium with sterile swabs for plate counts. Wounds were imaged with a digital camera at days 0, 3, 5, 7, 10, and 14 to assess wound healing and, more importantly, to observe the presence of new hair follicles. The wound area was quantified with ImageJ. At the same time, changes in blood glucose and body weight of the mice were monitored every 2 days.

Histological analysis

Wound tissues at different times (3, 7, 14 days) were subjected to H&E, Masson's three-color staining and picric red staining to assess wound healing. First, skin samples were fixed with 4% paraformaldehyde, dehydrated and embedded in paraffin to form 5 μm thick sections. For immunohistochemical analysis, after sealing with 5% goat serum for 30 min at room temperature, the de-paraffinized and re-hydrated treated wound tissues were prepared with anti-CD31 binding antibody (ab1990121:2000), anti-VEGF antibody (ab1316, Abcam, 1:200), anti-α smooth muscle actin antibody (ab7817, 1 : 200), anti-IL-6 antibody (ab214419, Abcam, 1 : 200) and anti-IL-1 antibody (ab1316, Abcam, 1 : 200) and incubated overnight at 4 °C.

In vivo safety assessment

To verify the biocompatibility of TA@CN gel and TA@ZMG gel, treated mice were anesthetized and blood was collected. Blood samples were used for routine blood testing and isolated serum samples were used for biochemical analysis.

Statistical analysis

Each experiment was repeated at least three times. The experimental data were statistically analyzed, and the results were expressed as mean ± standard error (SD). One-way analysis of variance (ANOVA) was used for analysis of the significant difference of the experimental data. The data are indicated with *p < 0.05, **p < 0.01, and ***p < 0.001.

Results and discussion

Preparation and characterization of TA@CN gel

Nanocomposites are a beacon of hope for biological applications due to their exceptional properties. In view of this, the development of multifaceted platforms capable of seamlessly integrating nanoparticle enzymes is of great significance and great potential. Based on the above premise, we selected ceria nanocomposites (CNPs) and a zinc metal–organic framework (ZIF-8) as nano nodes. In Fig. 1A, we present the synthesis strategy employed to fabricate CNPs and their hydrogel complexes. As shown in Fig. 1B, the CNPs observed under TEM are very homogeneous spherical nanoparticles. To facilitate subsequent bond formation within the hydrogels we employed TA, which possesses a phenolic hydroxyl structure, for surface modification of the nanocomposites. The successful modification was confirmed by characteristic absorption peaks at 1265 nm in the FTIR analysis presented in Fig. S1A (ESI†). DLS analysis further demonstrated increased hydrated particle sizes in the modified nanocomposites (Fig. S2A, ESI†), confirming the successful surface coating. Subsequently, the hydrogel system was constructed by a free radical polymerization reaction. Following TA modification on the surface, the abundant phenolic hydroxyl groups were crosslinked with APBA through borate ester bonds, ultimately mediating the formation of glucose-responsive hydrogels. By carefully comparing the different compositions and ratios of the resulting hydrogels, we confirmed the optimal synthetic formulation as AM (7 M) combined with APS and TEMED at a ratio of 10 : 1, as specified in Table S1 (ESI†). Ensuring an appropriate pore size and porosity is crucial to providing an ideal microenvironment for cell migration and proliferation without compromising the scaffold's mechanical properties. SEM image characterization of the hydrogel in Fig. 1C shows that the pore size of the TA@CN gel is about 10 μm. Elemental analysis of the hydrogel confirms the successful loading of Ce (Fig. 1D). This finding was further reinforced by the elemental localization scan of the hydrogel depicted in Fig. S3A (ESI†), clearly illustrating the uniform dispersion of elements within the hydrogel structure. Lastly, TA@CN gel was subjected to strain amplitude scanning tests using rheometry at a constant frequency of 10 rad s⁻¹ (Fig. 1E). The point of intersection between decreasing G′ and increasing G′′ at a specific stress value indicates the critical transition between solid and liquid states. Based on these results, we successfully constructed a hydrogel system with rheological properties using CNPs as nodes.

Fig. 1 (A) Schematic of cerium oxide nanoparticle-mediated hydrogel synthesis. (B) TEM image of CNPs. (C) SEM image of TA@CN gel. (D) Elemental scan analysis of TA@CN gel. (E) Strain sweep measurement of storage modulus (G′) and loss modulus (G′′) of TA@CN gel. (F) Schematic of ZIF-8 nanoparticle-mediated hydrogel synthesis. (G) TEM image of ZIF-8. (H) SEM image of TA@ZMG gel. (I) Elemental scan analysis of TA@ZMG gel. (J) Strain sweep measurement of storage modulus (G′) and loss modulus (G′′) of TA@ZMG gel.

Preparation and characterization of TA@ZMG gel

Encouraged by the successful construction of the hydrogel system with CNPs as nodes, and recognizing the remarkable drug-loading capabilities of ZIF-8, we astutely employed a similar approach to construct TA@ZMG gel, with ZIF-8 serving as the pivotal node (Fig. 1F). The SEM image of Fig. 1G shows the rhombic dodecahedral structure of ZIF-8. Successful modification with tannic acid was also confirmed by FITR (Fig. S1B, ESI†) and DLS analysis (Fig. S2B, ESI†). Subsequently, a network-like hydrogel system with uniform pore size was constructed by a free radical polymerization reaction. Interestingly, the SEM images featured in Fig. 1H provide unequivocal proof of the presence of ZIF-8 nanoparticles, further substantiating the efficacious construction of the nanogel system. Elemental analysis of Fig. 1I detected the presence of elemental Zn, while the elemental distribution scan performed on the hydrogel illustrates the homogeneous distribution of Zn (Fig. S3B, ESI†). Finally, in order to verify that TA@ZMG gel has the rheological properties of a hydrogel, a strain amplitude scanning test was carried out using a rheometer at a constant frequency of 10 rad s⁻¹. The convincing results shown in Fig. 1J demonstrate the ability of the hydrogel to maintain a stable morphology over a defined strain range.

Drawing upon the synthesis and characterization of TA@CN gel and TA@ZMG gel as detailed previously, it is justifiable to surmise that the synthesis of glucose-responsive hydrogels utilizing nanoparticles as fundamental components can serve as a versatile and overarching approach in fabricating glucose-responsive hydrogels for the purpose of managing diabetic wounds. Consequently, we deliberately opted for Prussian blue nanoparticles (PBNPs), a material characterized by its distinct anisotropic properties, and ingeniously devised a Prussian blue hydrogel complex (TA@PB gel) employing the aforementioned comprehensive synthesis strategy (Fig. S4A, ESI†). The success of tannic acid modification was first confirmed by FTIR analysis and DLS analysis (Fig. S4B and C, ESI†). Subsequent SEM images in Fig. S4D and E (ESI†) showed the cubic structure of PBNPs and the hydrogel network structure of TA@PB gel. Elemental analysis and strain amplitude testing of the hydrogels were also performed and are shown in Fig. S4F and G (ESI†). The generalizability of this method of nanogel composite synthesis was further demonstrated.

Mechanical properties and enzymatic activity testing of TA@CN gel

Ceria nanocomposites have garnered considerable attention as classical nanocomposites due to their remarkable enzymatic activity. Consequently, we expected TA@CN gel to retain enzymatic functionality, effectively scavenging ROS and converting hydrogen peroxide into oxygen, as depicted in Fig. 2A.

Fig. 2 (A) Mechanism of action of TA@CN gel hydrogels. (B) Dynamic step-strain under repeated deformation with 1% strain for 20 s and 1000% strain for 20 s. (C) Time scan measurement of G′ and G′′. (D) Comparison photographs of TA@CN gel tensile test. (E) Examination of wearability, healing, and adhesion of TA@CN gel. (F) Water absorption rate of hydrogel during 12 h of dissolution. (G) Oxygen production of gel, TA@CN gel at different time points under incubation with H₂O₂. (H) Free radical inhibition rates of different concentrations of TA@CN gel leachate. (I) Ability of gel and TA@CN gel to scavenge hydroxyl radicals. (J) Schematic of TA@ZMG gel programmed drug delivery. (K) Time scan measurement of G′ and G′′ before degradation. (L) Time scan after TA@ZMG gel degradation. (M) SEM image of TA@ZMG gel before degradation. (N) SEM image of hydrogel after degradation. (O) The amount of H₂O₂ produced by gel and TA@ZMG gel. (P) pH changes in different control groups over 24 h. (Q) and (R) SEM images of ZIF-8 nanocomposites before and after disintegration. Data are shown as mean ± SD (n ≥ 3), *p < 0.05, **P < 0.01, ***P < 0.001.

Examination of rheological performance using a rheometer confirmed the resilience of the TA@CN gel network after five cycles of testing at 1000% strain (Fig. 2B). Notably, the storage and loss moduli were successfully restored upon reducing the strain to 1%, indicating excellent self-healing properties. Furthermore, an oscillation time scan traced the kinetic process of hydrogel viscoelasticity over time, revealing favorable rheological properties (Fig. 2C). Investigation of the hydrogel's mechanical properties revealed a remarkable six-fold increase in tensile capacity for TA@CN gel, as shown in Fig. 2D. Subsequent experiments evaluated wearability, healing, and adhesion capabilities, while the hydrogel weight ratios at different time points (Fig. 2F) demonstrated the excellent swelling equilibrium properties of TA@CN gel. As shown in Fig. S5 (ESI†), tensile tests were performed to evaluate the effects of nanoparticle and borate bonds on the hydrogel properties. The hydrogels with the addition of borate bonds could stretch up to ten times their own length, and their tensile properties were significantly better than those of the other groups. Subsequent modulus test results showed that their energy storage modulus was also significantly higher than that of the other groups (Fig. S6, ESI†). This indicates that the introduction of borate ester bonds has positive significance for the improvement of the rheological properties of the hydrogel. Importantly, the inclusion of dynamic borate ester bonds conferred glucose-responsive properties upon the hydrogel. To validate the glucose dependence of degradation, the hydrogel morphology was observed in Fig. S7 (ESI†) after immersion with different concentrations of glucose solution. This was confirmed by further rheological analysis (Fig. S8, ESI†). As the concentration of glucose solution increased, the modulus of the hydrogel gradually decreased, suggesting an increase in degradation. Next, we verified the enzymatic activity of TA@CN gel. CAT acts as an enzymatic scavenger, catalyzing the decomposition of H₂O₂ into molecular oxygen and water. By effectively removing hydrogen peroxide from the body, CAT plays a crucial role in preserving cellular integrity and is a key component of the biological defense system. Measurement of oxygen production using a dissolved oxygen meter confirmed the remarkable CAT activity of TA@CN gel, as depicted in Fig. 2G. Additionally, the produced oxygen is clearly observable in Fig. S9 (ESI†). SOD-like activity was further assessed by co-incubating different concentrations of TA@CN gel leachate with a SOD working solution, revealing excellent concentration-dependent SOD activity (Fig. 2H). The ability of TA@CN gel to scavenge ˙OH radicals was subsequently evaluated using a chromogenic solution of TMB, demonstrating exceptional free radical scavenging ability (Fig. 2I). Collectively, these results underscore the remarkable enzymatic activity exhibited by TA@CN gel.

Programmed drug delivery by TA@ZMG gel

After evaluating the simple nanohydrogels above, we proceeded to explore drug-encapsulated nanohydrogels. Rhombic dodecahedral ZIF-8 and Met were combined to construct TA@ZMG gel using the generic synthesis method described earlier. Fig. 2J illustrates the programmed drug release process of TA@ZMG gel. Our hypothesis is that after TA@ZMG gel responds to glucose in the wound, the borate ester bond breaks, releasing GOX and decomposing it into glucuronic acid. This process triggers a pH-drop-mediated disintegration of the MOF structure, leading to the subsequent release of metformin. The successful loading of metformin in TA@ZMG gel was confirmed by the characteristic absorption peak of the guanidine group at 1168 nm in the IR results (Fig. S10, ESI†), and the UV-visible spectral results supported these findings (Fig. S11, ESI†). The HPLC standard curve in Fig. S12 (ESI†) further detected the encapsulation rate of Met, which was determined to be 70%.

Initially, we assessed the self-healing properties of TA@ZMG gel using the same method, and the results presented in Fig. S13 (ESI†) were similar to those of TA@CN gel, indicating that both gels possess good self-healing properties. The glucose-responsive degradation process of the hydrogels was captured through an oscillation time scan, and the changes in modulus of TA@ZMG gel before and after immersion in 4 mg mL⁻¹ glucose are compared in Fig. 2K and L. It is evident that the modulus of TA@ZMG gel reversed after 2 h of immersion in the glucose solution, indicating a phase change in the hydrogel. To further confirm this change, the three-dimensional (3D) network structure of the hydrogel before and after immersion was observed through SEM imaging. As shown in Fig. 2M, the hydrogel exhibited an ordered network structure with clearly visible uniform pore sizes. However, after soaking in the glucose solution for 2 h, the morphology of the hydrogel changed and showed a degradation trend (Fig. 2N). After the hydrogel degraded, the loaded GOX was programmed to be released, and its catalytic product H₂O₂ could act as an antimicrobial agent. Hence, we measured the absorbance of gel and TA@ZMG gel to evaluate the production of H₂O₂. Fig. 2O demonstrates that the group of materials containing GOX produced more H₂O₂, indicating the catalytic reaction between GOX and glucose. Glucuronic acid, another catalytic product, was able to mediate a decrease in pH. This was confirmed by incubating gel and TA@ZMG gel with 4 mg mL⁻¹ glucose solution for 24 h, and measuring the change in pH at different time points. As shown in Fig. 2P, the pH of TA@ZMG gel solution significantly decreased, providing evidence that the addition of GOX in TA@ZMG gel caused the decrease in solution pH. With the decrease in solution pH, the structure of ZIF-8 imidazole disintegrated under acidic conditions. The SEM images in Fig. 2Q and R illustrate the morphology of ZIF-8 before and after disintegration. Initially, it had a complete rhombic dodecahedral structure, which became fragmented after 2 h of reaction with the GOX solution. These results demonstrate the dual response drug release process of TA@ZMG gel.

The degradation rate of hydrogels determines the rate of drug release, which is important for subsequent in vitro and in vivo applications, so we investigated the degradation of TA@CN gel and TA@ZMG gel based on glucose response in vitro. As shown in Fig. S14 (ESI†), after 10 days of degradation, the residual mass ratios of TA@CN gel and TA@ZMG gel were only 14.70% and 11.98%. The slightly faster degradation of the latter than the former may be due to the loading of GOX.

In vitro cellular evaluation of TA@CN gel

Moving on to the in vitro cellular evaluation of TA@CN gel, we validated its enzymatic activity at the chemical level and further explored its bioactive effects at the cellular level, as depicted in Fig. 3A. CNPs were released from the hydrogel to act on cells, with the expectation of exerting enzymatic activity and promoting cell migration and angiogenesis. To ensure the compatibility of the gel with cells, we first assessed its cytotoxicity by incubating HUVECs with different concentrations of gel extract for 24 h. The samples at various concentrations did not exhibit any cytotoxicity compared to the PBS group (Fig. S15, ESI†). Additionally, there was no significant difference in cell viability between the three sample groups, as observed through fluorescence images obtained from calcein-AM/PI cell staining (Fig. S16, ESI†). These results indicate that TA@CN gel is biocompatible with cells. Furthermore, we evaluated its enzymatic activity in the following experiments. We observed oxygen production using an RDPP fluorescent probe in pre-hypoxic cultured HUVEC cells. As oxygen quenches the fluorescence of the RDPP probe, only the TA@CN gel group in Fig. 3B did not exhibit significant green fluorescence, indicating that the hydrogel produced oxygen. This analysis suggests the presence of CAT-like activity in the loaded CNPs, which convert H₂O₂ to O₂. The intracellular ROS levels were further measured using a DCFH-DA assay. Both the positive control and gel groups exhibited intense fluorescence, while the material group showed weaker fluorescence, attributed to the antioxidant effect of CNPs (Fig. 3C).

Fig. 3 (A) Schematic representation of TA@CN gel promoting cell migration and angiogenesis. (B) RDPP probe to detect intracellular oxygen levels in HUVECs after different treatments. (C) DCFH-DA assay detecting intracellular ROS level of HUVECs after different treatments. (D) Digital images of HUVECs after treatment with PBS, gel, and TA@CN gel for 0, 12, and 24 h. (E) Images showing HUVEC migration after different treatments in a transwell migration assay. (F) Images of tube formation after different treatments. (G) Schematic diagram of the diabetic mouse trauma model established and treatment given. (H) Photographs of wounds in different sample groups after 0, 3, 5, 7, 10 and 14 days of treatment. (I) H&E and Masson's trichrome staining of different treatment groups at 3, 7 and 14 days. (J) Immunohistochemical images of VEGF and platelet endothelial cell adhesion molecule-1 (CD31) after day 3, day 7 and day 14 of different treatments. (K) Quantification of new collagen in wound tissue on day 7 and day 14 of different treatments. (L) Quantification of VEGF after 14 days of different treatments. (M) and CD31. Data are shown as mean ± SD (n ≥ 3), * p < 0.05, **P < 0.01, ***P < 0.001.

The literature has reported^24–26 that CNPs can induce angiogenesis through the upregulation of vascular endothelial growth factor (VEGF) and expression of tumor necrosis factor-α (TNF-α). To determine whether TA@CN gel promotes vascular endothelial cell migration, we conducted a scratch test. After incubating HUVEC cells under various conditions for 24 h, the cells treated with TA@CN gel exhibited more rapid migration then the control group (Fig. 3D), indicating that its ability to promote migration is based on its loading of CNPs. The quantification of the migration area and migration ratio of different groups can be seen in Fig. S17 (ESI†). The results of the transwell migration assay were consistent with the scratch assay, as a large number of cells in the TA@CN gel treated group passed through the membrane (Fig. 3E). Furthermore, we investigated the effect of ceria nanoenzymes on angiogenesis. HUVECs pretreated under different conditions were transferred to the substrate and incubated for 12 h. The cell tube length in Fig. 3F was significantly increased compared to the control group, which aligns with the quantitative results of vessel length obtained from ImageJ (Fig. S18, ESI†). The enhanced angiogenesis after CNP treatment confirms the apparent ability of ceria dioxide nanozymes to promote angiogenesis. Thus, these results collectively support the notion that TA@CN gel may facilitate diabetic wound healing.

In vivo evaluation and histochemical analysis of TA@CN gel

Encouraged by the positive impact of cerium nanozymes on cell migration and the promotion of tube formation in vitro, we conducted experiments to test the effectiveness of TA@CN gel in promoting wound healing in vivo. The diabetic mouse model was established using the procedure of Fig. 3G, in which mice were first injected with STZ after high-fat chow feeding, and then the success of the modeling was confirmed by monitoring the blood glucose level of the mice, which remained above 16.7 mmol L⁻¹ for a week. Subsequently, we established a diabetic mouse back wound model and randomly divided the mice into three groups (PBS, gel, TA@CN gel) for treatment. After a 14 day treatment period, the progress of wound healing at different time points was assessed (Fig. 3H), and it was observed that the TA@CN gel group showed significant healing after 7 days. Quantification of the wound area (Fig. S20, ESI†) revealed that the wounds treated with TA@CN gel were almost completely healed after 14 days, with a healing rate of approximately 90%. These findings suggest that the TA@CN gel group effectively promoted wound healing. Throughout the treatment period, the body weight of the mice in each group was monitored, and no significant changes were observed in any of the treated groups (Fig. S21, ESI†).

To further examine the changes in epithelialization and collagen during the healing process, H&E staining and Masson's trichrome staining were performed on the skin sections. As shown in Fig. 3I, after 14 days of treatment, thicker crusts were still present in both the PBS and gel control groups, indicating a delayed healing process in chronic wounds. In comparison, the TA@CN gel group exhibited minimal scab formation, and even the growth of new hair follicles, sebaceous glands, sweat ducts, and squamous epithelium was observed after 14 days of treatment. Additionally, the TA@CN gel group showed significantly higher collagen deposition compared to the control group after 14 days. Importantly, the TA@CN gel group exhibited the densest collagen deposition, indicating that the functional hydrogel accelerated wound recovery by promoting collagen deposition. To further assess collagen deposition in the wound skin, mature type I collagen was selectively stained red or orange using picrosirius staining. In Fig. S22 (ESI†), a large amount of mature type I collagen was observed in the TA@CN gel after 14 days of treatment, while only a small amount was present in the PBS and gel groups. Quantitative analysis of collagen (Fig. 3K) provided further evidence that TA@CN gel not only promoted the growth of total collagen but also facilitated collagen maturation.

Since local neovascularization plays a crucial role in accelerating the healing of diabetic wounds, immunohistochemical staining was performed to assess the expression of VEGF and cell adhesion molecule-1 (CD31). VEGF is a highly specific pro-vascular endothelial cell growth factor that promotes increased vascular permeability, extracellular matrix degeneration, vascular endothelial cell migration, proliferation, and angiogenesis by upregulating its expression level. Fig. 3J shows that VEGF expression was significantly higher in the TA@CN gel group than in the other two control groups after treatment, and the quantification of VEGF expression (Fig. 3L) further demonstrated the upregulation of VEGF by TA@CN gel. CD31, a marker of neovascularization, is involved in leukocyte migration, angiogenesis, and integrin activation. The staining results of CD31 in Fig. 3J indicate clear formation of neovascularization in the TA@CN gel group. Quantification of neovascularization (Fig. 3M) showed significant differences between the TA@CN gel group and the other two groups. Additionally, α-smooth muscle actin (α-SMA) staining was performed to verify the effect of TA@CN gel on angiogenesis. As shown in Fig. S23 (ESI†), the TA@CN gel group exhibited robust neovascularization during the treatment period, while only minimal neovascularization was observed in the other two groups.

Finally, immunohistochemical staining was conducted to detect the expression levels of inflammatory cytokines, including interleukin-6 (IL-6), interleukin-1 (IL-1), and TNF-α. The staining results (Fig. S24, ESI†) showed that the PBS and gel groups still exhibited more severe inflammation compared to the TA@CN gel group, which showed a decreasing trend in the expression of inflammatory factors. This indicates that TA@CN gel effectively reduced the expression of inflammatory factors in the wound and promoted wound healing. The results of immunohistochemical staining collectively demonstrate that TA@CN gel effectively promotes diabetic wound healing and exerts anti-inflammatory and pro-angiogenic effects.

Evaluation of antimicrobial activity in vitro

Classical ZIF-8 nanocomposites demonstrate remarkable antibacterial efficacy. Building upon our previous investigation on diabetic wounds, our aim was to further explore the application of drug-loaded ZIF-8 hydrogel in the treatment of infected diabetic wounds. Consequently, we employed the synthesized TA@ZMG gel in infected diabetic wounds through programmable drug delivery (Fig. 4A). Initially, the degradation of ZMG occurs in response to elevated glucose concentration in diabetic wounds, leading to the release of GOX and ZIF-8. The catalytic product of the former induces a decrease in pH, resulting in the disintegration of ZIF-8 and subsequent release of metformin. This process proves beneficial for reducing wound blood glucose levels and promoting the growth of hair follicles and sweat glands. Notably, the disintegration of ZIF-8 nanocomposites inhibits bacterial growth. The antibacterial properties of TA@ZMG gel were evaluated in vitro through the measurement of bacterial suspension absorbance at 600 nm and a plate counting method. As illustrated in Fig. 4B and C, TA@ZMG gel exhibited some degree of antibacterial activity against both S. aureus and E. coli. The bacterial survival rate of the samples was significantly lower than in the PBS and gel groups, and statistical analysis also showed differences between the groups. This conclusion is further supported by the results of plate counts, which demonstrated a substantial reduction in the number of bacteria in the sample group (Fig. 4D). It is evident that this difference can be attributed the presence of ZIF-8 loaded in TA@ZMG gel.

Fig. 4 (A) Schematic diagram of ZMG programmed release to inhibit bacteria and promote hair follicle growth in infected diabetic wounds. (B) and (C) Bacterial survival rates of S. aureus and E. coli after different treatments (PBS, gel, TA@ZMG gel). (D) Photographs of S. aureus and E. coli agar plates after various treatments (PBS, gel, TA@ZMG gel). (E) Establishment of bacterial infection models, and plate counts after different treatments. (F) Schematic diagram of the diabetic mouse trauma model established and treatment given. (G) Photographs of wounds in different sample groups after 0, 3, 5, 7, 10 and 14 days of treatment. (H) Wound size in different treatment groups. (I) Changes in whole-body blood glucose in mice after 1, 3, 5 and 7 days of different treatments. (J) H&E staining and Masson's trichrome staining of different treatment groups at day 3, 7 and 14. (K) Quantitative analysis of the number of new hair follicles. Data are shown as mean ± SD (n ≥ 3), *p < 0.05, **P < 0.01, ***P < 0.001.

In vivo evaluation and histochemical analysis of TA@ZMG gel

Further evaluation was conducted in vivo to assess the efficacy of TA@ZMG gel in combating infection and promoting wound healing in diabetic mice. Prior to this, the biocompatibility of TA@ZMG gel was evaluated through CCK-8 and calcein-AM/PI cell staining using HUVECs. Fig. S25 and S26 (ESI†) demonstrate the safe applicability of TA@ZMG gel in vivo. As depicted in Fig. 4E, after establishing the diabetes model, bacterial fluid of S. aureus was introduced into the wound to induce bacterial infection. Different treatments (PBS, gel, TA@ZMG gel) were administered to the infected diabetic wounds. The results of wound bacterial plate counts (Fig. 4F) revealed limited anti-infective effects in the PBS and gel groups, whereas the TA@ZMG gel treatment group witnessed a significant reduction in the number of bacteria. This finding confirms that TA@ZMG gel acts as a protective barrier against bacterial infection in wounds. The impacts of different treatment groups (PBS, gel, TA@ZMG gel) on wound healing after 14 days of treatment are presented in Fig. 4G and H. The wound area was simulated and analyzed using ImageJ (Fig. S27, ESI†). Remarkably, TA@ZMG gel exhibited excellent wound healing and significantly improved infection compared to other controls, owing to the anti-infective properties of the nanogel. Additionally, we monitored changes in body weight and systemic glucose levels of the mice during the treatment period and observed no significant alterations in the body weight of each group (Fig. S28, ESI†). Surprisingly, continuous decreases in blood glucose values were observed in the TA@ZMG gel treated group while no significant changes occurred in the PBS and gel groups. Blood glucose values in the TA@ZMG gel group steadily decreased on days 1, 3, 5, and 7, eventually reaching the normal blood glucose range (Fig. 4I). This outcome suggests that the Met loaded in TA@ZMG gel plays a role in regulating systemic blood glucose levels through programmed drug release.

Based on the presence of metformin in TA@ZMG gel, its potential to promote hair follicle and sweat gland growth in diabetic wounds was anticipated. Consequently, we performed H&E and Masson's trichrome staining on skin sections. As demonstrated in Fig. 4J, TA@ZMG gel exhibited partial epithelialization and follicular growth after 7 days, while dense follicular and granulation tissue growth was observed after 14 days of treatment. Quantification of the number of hair follicles revealed a significant increase in the TA@ZMG gel group compared to the PBS and gel groups (Fig. 4K). This distinction may be attributed to the positive effect on the growth of hair follicles and sweat glands exerted by Met in the hydrogel. Furthermore, inflammatory infiltration was assessed through H&E staining results. The PBS and gel groups displayed more pronounced inflammatory infiltration, whereas the TA@ZMG gel group exhibited minimal inflammation. This observation further substantiates the potential antibacterial effect of TA@ZMG gel. Masson's trichrome staining results (Fig. 4J) indicated that the TA@ZMG gel group exhibited more pronounced collagen deposits than the other two groups. Picrosirius red staining results (Fig. S29, ESI†) revealed the presence of orange type I collagen in TA@ZMG gel after 7 days, and a substantial deposition of mature type I collagen after 14 days. Similarly, immunohistochemical staining for VEGF and CD31 demonstrated the ability of TA@ZMG gel to assist in neovascularization (Fig. S30, ESI†). Finally, we focused on monitoring the expression of inflammatory factors following treatment for infected diabetic wounds. Pro-inflammatory cytokines, IL-6, IL-1, and TNF-α, contribute to the release of inflammatory mediators involved in the inflammatory response triggered by infection. Inflammatory factors tend to be highly expressed in infected wounds. Immunostaining results for inflammatory factors (Fig. S31, ESI†) exhibited a significant decrease in inflammatory expression in the TA@ZMG gel group after 14 days of treatment, while substantial inflammation persisted in the other groups. This reduction in inflammatory factors can be attributed to the down-regulation exerted by TA@ZMG gel, in addition to its anti-infective effects. Histological and immunohistochemical analyses confirm the remarkable effectiveness of TA@ZMG gel in combating infection, promoting wound healing, and its potential to stimulate hair follicle regeneration.

In vivo safety assessment

The safety of TA@CN gel and TA@ZMG gel should not be disregarded; hence, potential adverse effects post-treatment were investigated. As depicted in Fig. S32 (ESI†), the hematological and biochemical indices of all groups of mice remained within normal ranges, implying favorable biocompatibility of TA@CN gel and TA@ZMG gel.

Conclusions

In conclusion, we utilized nanocomposite nodes to fabricate multifunctional hydrogel complexes (TA@CN gel and TA@ZMG gel) with glucose-responsive properties aimed at promoting healing of diabetic wounds and infected diabetic wounds. TA@CN gel and TA@ZMG gel exhibit commendable rheological and mechanical properties, rendering them effective for diabetic wound applications. Notably, TA@CN gel demonstrates significant enzymatic activity in scavenging ROS and stimulating wound vascular growth. In addition, TA@ZMG gel achieves antimicrobial activity, follicle regeneration,^53–55 and even regulation of systemic blood glucose levels in infected wounds through glucose/pH-responsive^56–58 drug release. We firmly believe that this multifunctional hydrogel, with its straightforward synthesis process and robust drug exchange capabilities, holds great promise as a combination therapy for the effective regeneration of wounds under various complex conditions, including infected diabetic wounds.

Author contributions

All authors participated in drafting the article and critically revised important content, and endorsed the final version. Yingjuan Zhou: design of experiments, materials preparation and evaluation, writing – original draft, writing – review & editing; Jiaxin Yang: assistance of writing and reviewing of the paper, investigation, methodology; Yan Li: methodology, formal analysis; Xin Shu and Yucen Cai: methodology, data curation; Ping Xu and Wenyan Huang: formal analysis, investigation; Zhangyou Yang: supervision, project administration and funding support; Rong Li: supervision, writing-reviewing and funding support.

Conflicts of 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.

Acknowledgements

Yingjuan Zhou, Jiaxin Yang and Yan Li contributed equally to this work. This work was supported by the National Natural Science Foundation of China (No. 82173456), Chongqing Municipal Foundation (CSTC2021jcyj-msxm3803), and the Natural Science Foundation of Chongqing (Grant No. CSTB2022NSCQ-MSX1084).

Notes and references

1. A. S. MacLeod, Cell Host Microbe, 2019, 25, 638–640.
2. J. B. Cole and J. C. Florez, Nat. Rev. Nephrol., 2020, 16, 377–390.
3. S. Patel, S. Srivastava, M. R. Singh and D. Singh, Biomed. Pharmacother., 2019, 112, 108615.
4. B. K. Sun, Z. Siprashvili and P. A. Khavari, Science, 2014, 346, 941–945.
5. M.-Q. Tong, C.-T. Lu, L.-T. Huang, J.-J. Yang, S.-T. Yang, H.-B. Chen, P.-P. Xue, L.-Z. Luo, Q. Yao, H.-L. Xu and Y.-Z. Zhao, Acta Biomater., 2023, 157, 467–486.
6. R. Shi, H. Li, X. Jin, X. Huang, Z. Ou, X. Zhang, G. Luo and J. Deng, Acta Biomater., 2022, 152, 425–439.
7. Z. Gao, Q. Wang, Q. Yao and P. Zhang, Pharmaceutics, 2021, 14, 6.
8. T. Khaliq, M. Sohail, M. U. Minhas, S. Ahmed Shah, N. Jabeen, S. Khan, Z. Hussain, A. Mahmood, M. Kousar and H. Rashid, Int. J. Biol. Macromol., 2022, 197, 157–168.
9. Y. Liu, C. Li, Z. Feng, B. Han, D.-G. Yu and K. Wang, Biomolecules, 2022, 12, 1727.
10. K. Y. Choi, Md. T. Sultan, O. Ajiteru, H. Hong, Y. J. Lee, J. S. Lee, H. Lee, O. J. Lee, S. H. Kim, J. S. Lee, S.-J. Park, J. G. Eden and C. H. Park, Biomedicines, 2021, 10, 27.
11. L. A. Lavery, E. C. Ryan, J. Ahn, P. A. Crisologo, O. K. Oz, J. La Fontaine and D. K. Wukich, Clin. Infect. Dis., 2020, 70, 1573–1579.
12. I. D. Teixeira, E. Carvalho and E. C. Leal, Antibiotics, 2023, 12, 467.
13. W. Hu, H. Yu, X. Zhou, M. Li, L. Xiao, Q. Ruan, X. Huang, L. Li, W. Xie, X. Guo and P. Yao, Burns, 2022, 48, 1452–1461.
14. R. Weinberg Sibony, O. Segev, S. Dor and I. Raz, IJMS, 2023, 24, 17147.
15. W. Qin, Y. Wu, J. Liu, X. Yuan and J. Gao, IJN, 2022, 17, 6007–6029.
16. M. Roberts, Y. Mohammed, M. Pastore, S. Namjoshi, S. Yousef, A. Alinaghi, I. Haridass, E. Abd, V. Leite-Silva, H. Benson and J. Grice, J. Controlled Release, 2017, 247, 86–105.
17. K. Korschelt, M. N. Tahir and W. Tremel, Chem. – Eur. J., 2018, 24, 9703–9713.
18. J. Wu, X. Wang, Q. Wang, Z. Lou, S. Li, Y. Zhu, L. Qin and H. Wei, Chem. Soc. Rev., 2019, 48, 1004–1076.
19. K. McNamara and S. A. M. Tofail, Adv. Phys.: X, 2017, 2, 54–88.
20. Z. Wang, R. Zhang, X. Yan and K. Fan, Mater. Today, 2020, 41, 81–119.
21. V. Baldim, N. Yadav, N. Bia, A. Graillot, C. Loubat, S. Singh, A. S. Karakoti and J.-F. Berret, ACS Appl. Mater. Interfaces, 2020, 12, 42056–42066.
22. V. Patel, M. Singh, E. L. H. Mayes, A. Martinez, V. Shutthanandan, V. Bansal, S. Singh and A. S. Karakoti, Chem. Commun., 2018, 54, 13973–13976.
23. Y. Wu and H. T. Ta, J. Mater. Chem. B, 2021, 9, 7291–7301.
24. J. Xiang, J. Li, J. He, X. Tang, C. Dou, Z. Cao, B. Yu, C. Zhao, F. Kang, L. Yang, S. Dong and X. Yang, ACS Appl. Mater. Interfaces, 2016, 8, 4489–4499.
25. S. K. Nethi, H. S. Nanda, T. W. J. Steele and C. R. Patra, J. Mater. Chem. B, 2017, 5, 9371–9383.
26. I.-S. Park, C. Mahapatra, J. S. Park, K. Dashnyam, J.-W. Kim, J. C. Ahn, P.-S. Chung, D. S. Yoon, N. Mandakhbayar, R. K. Singh, J.-H. Lee, K. W. Leong and H.-W. Kim, Biomaterials, 2020, 242, 119919.
27. X. Yu, X. Fu, J. Yang, L. Chen, F. Leng, Z. Yang and C. Yu, Mater. Today Bio, 2022, 15, 100308.
28. X. Fu, X. Yu, J. Jiang, J. Yang, L. Chen, Z. Yang and C. Yu, Nat. Commun., 2022, 13, 6528.
29. H. Zhang, Q. Li, R. Liu, X. Zhang, Z. Li and Y. Luan, Adv. Funct. Mater., 2018, 28, 1802830.
30. X. Zhao, L. Chang, Y. Hu, S. Xu, Z. Liang, X. Ren, X. Mei and Z. Chen, ACS Appl. Mater. Interfaces, 2022, 14, 18194–18208.
31. Q. Yang, C. Mo, P. Cui, S. Zhou, L. Qiu, P. Jiang, Y. Xuan, H. Huang, C. Wang and J. Wang, Life Sci., 2023, 329, 121930.
32. M. Liu, X. Wei, Z. Zheng, Y. Li, M. Li, J. Lin and L. Yang, IJN, 2023, 18, 1537–1560.
33. Z. Han, M. Yuan, L. Liu, K. Zhang, B. Zhao, B. He, Y. Liang and F. Li, Nanoscale Horiz., 2023, 8, 422–440.
34. Q. Zeng, X. Qi, G. Shi, M. Zhang and H. Haick, ACS Nano, 2022, 16, 1708–1733.
35. S. Yao, Y. Zhao, Y. Xu, B. Jin, M. Wang, C. Yu, Z. Guo, S. Jiang, R. Tang, X. Fang and S. Fan, Adv. Healthcare Mater., 2022, 11, 2200516.
36. J. R. Bardill, M. R. Laughter, M. Stager, K. W. Liechty, M. D. Krebs and C. Zgheib, Acta Biomater., 2022, 138, 73–91.
37. Z. Xu, S. Han, Z. Gu and J. Wu, Adv. Healthcare Mater., 2020, 9, 1901502.
38. S. Correa, A. K. Grosskopf, H. Lopez Hernandez, D. Chan, A. C. Yu, L. M. Stapleton and E. A. Appel, Chem. Rev., 2021, 121, 11385–11457.
39. C. Liu, Q. Zhang, S. Zhu, H. Liu and J. Chen, RSC Adv., 2019, 9, 28299–28311.
40. X. Xu, V. V. Jerca and R. Hoogenboom, Mater. Horiz., 2021, 8, 1173–1188.
41. W. Zhang, Z. Yang, M. Zhang, J. He, S. Li, X. Sun and P. Ni, J. Mater. Chem. B, 2023, 11, 4922–4933.
42. L. Yang, F. Liang, X. Zhang, Y. Jiang, F. Duan, L. Li and F. Ren, Chem. Eng. J., 2022, 427, 131506.
43. L. Russo, A. Gloria, T. Russo, U. D’Amora, F. Taraballi, R. De Santis, L. Ambrosio, F. Nicotra and L. Cipolla, RSC Adv., 2013, 3, 6286.
44. A. Ronca, L. Ambrosio and D. W. Grijpma, J. Appl. Biomater. Biomech., 2012, 10, 249–258.
45. L. Sheeney-Haj-Ichia, G. Sharabi and I. Willner, Adv. Funct. Mater., 2002, 12, 27.
46. V. Pardo-Yissar, R. Gabai, A. N. Shipway, T. Bourenko and I. Willner, Adv. Mater., 2001, 13, 1320.
47. S. R. Sershen, S. L. Westcott, N. J. Halas and J. L. West, Appl. Phys. Lett., 2002, 80, 4609–4611.
48. P. Thoniyot, M. J. Tan, A. A. Karim, D. J. Young and X. J. Loh, Adv. Sci., 2015, 2, 1400010.
49. P. Lavrador, M. R. Esteves, V. M. Gaspar and J. F. Mano, Adv. Funct. Mater., 2021, 31, 2005941.
50. S. Dong, Y. Dong, B. Liu, J. Liu, S. Liu, Z. Zhao, W. Li, B. Tian, R. Zhao, F. He, S. Gai, Y. Xie, P. Yang and Y. Zhao, Adv. Mater., 2022, 34, 2107054.
51. T. Naganuma, Nano Res., 2017, 10, 199–217.
52. P. Yang, Z. Zhu, T. Zhang, W. Zhang, W. Chen, Y. Cao, M. Chen and X. Zhou, Small, 2019, 15, 1902823.
53. P. Zhao, B. Sui, N. Liu, Y. Lv, C. Zheng, Y. Lu, W. Huang, C. Zhou, J. Chen, D. Pang, D. Fei, K. Xuan, C. Hu and Y. Jin, Aging Cell, 2017, 16, 1083–1093.
54. P. Kokhabi, S. Dadkhahfar and R. M. Robati, Med. Hypotheses, 2023, 179, 111155.
55. M. Chai, M. Jiang, L. Vergnes, X. Fu, S. C. De Barros, N. B. Doan, W. Huang, J. Chu, J. Jiao, H. Herschman, G. M. Crooks, K. Reue and J. Huang, Cell Rep., 2019, 27, 3413–3421.e3.
56. T. Liang, L. Zou, X. Guo, X. Ma, C. Zhang, Z. Zou, Y. Zhang, F. Hu, Z. Lu, K. Tang and C. M. Li, Adv. Funct. Mater., 2019, 29, 1903026.
57. L. Fu, C. Qi, Y. Hu, J. Lin and P. Huang, Adv. Mater., 2019, 31, 1808325.
58. C. Hu, L. Long, J. Cao, S. Zhang and Y. Wang, Chem. Eng. J., 2021, 411, 128564.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3tb02283h

‡ These authors contributed equally to this work.

---