Injectable thioketal-containing hydrogel dressing accelerates skin wound healing with the incorporation of reactive oxygen species scavenging and growth factor release

Abstract

Wound healing is a complex dynamic process. During the occurrence of skin injury, the excessive reactive oxygen species (ROS) level is associated with sustained inflammatory response, which limits efficient wound repair. Although multifunctional hydrogels are considered ideal wound dressings due to their unique advantages, the development of hydrogel dressings with rapid gelling rates, shape adaptation, and antioxidant function is still a vital challenge. In this work, a ROS-responsive injectable polyethylene glycol hydrogel containing thioketal bonds (PEG-TK hydrogel) was synthesized and utilized to deliver epidermal growth factor (EGF). We adopted bio-orthogonal click chemistry for crosslinking the polymer chains to obtain the EGF@PEG-TK hydrogel with fast gelation time, injectability and shape-adaptability. More interestingly, the thioketal bonds in the PEG-TK hydrogel not only scavenged excessive ROS in the wound sites but also achieved responsive and controlled EGF release to facilitate regeneration. The EGF@PEG-TK hydrogel treatment offered the benefits of protecting cells from oxidative stress, accelerating wound closure, and reducing scar formation in the full-thickness skin defect model. This work provides a promising strategy for developing antioxidant hydrogel dressing for facilitating the repair of wounds.

---

1. Introduction

Skin is the outermost organ of the body, which provides a natural barrier between the body and the external environment, as well as various essential protective functions.^1,2 Once the integrity of the skin is disrupted by acute or chronic injuries, a series of steps for wounded tissue reconstruction and barrier function reestablishment are initiated. The complex skin wound healing process usually involves several overlapping phases, including hemostasis, inflammation, proliferation, and remodeling.³ Reactive oxygen species (ROS) play important roles in the wound healing process. Homeostatic ROS are crucial mediators of intracellular signaling, in particular for wound angiogenesis.⁴ At the early stage of wound healing, thrombogenesis resulting from vascular injury could induce acute inflammation, which attracts neutrophils and macrophages to infiltrate the wound site to release enzymes, pro-inflammatory cytokines, and plenty of ROS. These ROS protect against invading bacteria and pathogens.⁵ When hydrogen peroxide (H₂O₂), one of the representative ROS, performs the bacteriostatic effect on Escherichia coli, its concentration needs to be as high as 500 μM to eradicate the bacteria.⁶ Several studies have revealed that higher H₂O₂ concentration can be found in the early inflammatory phase (2 days after injury) as compared to the later phase (5 days after injury) in the wound site, both of which fall in the range of 100–250 μM.^7,8 Moreover, H₂O₂ at micromolar concentrations also stimulates the proliferation of human fibroblasts and vascular endothelial cells and further facilitates angiogenesis.⁵ However, due to the high reactivity of ROS, the excessive production of ROS at wound sites would destroy cell membranes, protein, and DNA, hinder the occurrence of angiogenesis, as well as delay tissue regeneration. Therefore, introducing antioxidant strategies to develop novel wound dressing with ROS scavenging ability could improve the inflammation microenvironment and accelerate wound healing. The thioketal (TK) bond is a ROS-sensitive covalent bond that can readily react with a few typical ROS, such as the superoxide anion (˙O₂⁻), hydroxyl radical (˙OH), and H₂O₂ at a limiting concentration as low as 100 μM, and then be degraded into acetone and thiols.^9,10 Utilizing the ROS-specific degradation mechanism of TK, TK-containing biomaterials have been employed in constructing responsive drug delivery systems to deliver therapeutic agents to tumors or inflammation sites that are rich in ROS.^9–12 Recently, poly(thioketal) has also been proved to possess good antioxidative properties and can be used to protect cardiomyocytes from cytotoxic ROS.¹³ Moreover, compared to common hydrolytically degradable polymeric implants in tissue engineering, the in vivo degradation rate of TK-containing biomaterials better matches the tissue regeneration rate, which means better healing outcomes.

To accelerate wound closure, minimize scar formation and meet other requirements in the wound healing process, various kinds of wound dressing materials have been developed, including rubbers, membranes, forms, electrospinning fibers and hydrogels.¹⁴ Among them, hydrogels, which possess a three-dimensional porous polymer network, are not only similar to the natural extracellular matrix, but also absorb and retain a large amount of water to maintain the moist wound microenvironment. Moreover, hydrogels can absorb the tissue exudate, allow gas exchange, and relieve the pain of the patient by cooling the wound surface.^15,16 However, in most cases, injury causes irregularly shaped or relatively deep wounds. Considering this dilemma, the injectable hydrogel has attracted much attention due to their capacity for faultlessly filling the wound in three dimensions, avoiding bacterial infection, and non-invasive administration.^17,18 Most injectable hydrogels are formed by the UV-initiated crosslinking of double bonds, the coordination effect between chelates and ions, as well as hydrogen-bond interactions.¹⁷ However, with the introduction of an initiator, UV-irradiation and ions may be harmful to cells.^19–22 Besides, a long gelation time may result in precursor solution diffusing from the target sites, which would compromise the therapeutic effect and induce wound infection.²³ Bio-orthogonal click chemistries are revealed as an alternative approach for fabricating injectable hydrogels with exceptionally fast kinetics, high chemo-selectivity under physiological conditions, and biocompatibility.²⁴ This click chemistry pair, norbornene (Nb)/tetrazine (Tz), possesses aqueous stability under in vivo environment, and the bio-orthogonal reaction between Nb and Tz has been proved to hardly cause apparent cell toxicity.²⁵ The Nb/Tz-based click reaction has already been utilized in cell imaging, drug delivery and release, 3D cell culture and tissue engineering.^26–29 Therefore, a Nb and Tz crosslinked injectable hydrogel with a rapid gelation rate and shape adaptation could be a promising dressing material for skin wound healing.

Wound healing is a complex biological event, which involves many cell types, cellular interactions, the synthesis of numerous factors and various secreted mediators. The growth factors, including epidermal growth factor (EGF), fibroblast growth factor (FGF), transforming growth factor-beta (TGF-β) and platelet-derived growth factor (PDGF) play important roles in wound healing, and the expressions of themselves and their receptors are altered during the healing process.³⁰ Among them, EGF, which combines with the EGF receptor (EGFR), could accelerate the process of re-epithelialization by contributing to keratinocyte proliferation and cell migration.³¹ However, EGF is susceptible to degradation in the proteolytic condition of wound sites and rapidly loses its bioactivity.³² Meanwhile, it is hard to keep EGF in the wound site after direct administration, which will result in decreased therapeutic outcome and increased administration frequency of EGF. Multiple EGF administrations mean aggravation of the patients’ cost, and the diffused EGF may cause side effects in the surrounding healthy tissues. Hydrogels can be employed to deliver therapeutics, antibacterial agents, growth factors, or cells to the wound sites and maintain the activity of payloads.³³ In this study, a ROS-responsive hydrogel wound dressing (EGF@PEG-TK) was developed by integrating TK linkers into the Nb/Tz cross-linked PEG hydrogel and further encapsulating EGF to enhance the wound healing effect (Scheme 1). The prepared PEG-TK hydrogel was injectable and adaptable due to the Nb–Tz biological orthogonal pair. The ROS scavenging experiments verified that the PEG-TK hydrogel had better scavenging capacities against 2,2-diphenyl-1-picrylhydrazyl (DPPH) and ˙OH than the non-responsive PEG hydrogel. When being used as a dressing to treat full-thickness wounds, the EGF@PEG-TK hydrogel improved the injury microenvironment and achieved rapid wound closure via scavenging excessive ROS and releasing EGF synergistically.

Scheme 1 Schematic illustration of the EGF@PEG-TK hydrogel for accelerating wound healing in SD rats.

2. Results and discussion

2.1 Synthesis and characterization of 4-arm-PEG-TK-Nb, 4-arm-PEG-Nb and 4-arm-PEG-Tz

Poly(ethylene glycol) (PEG), which has various molecular masses, architectures, and reactive end groups, is one of the best candidates for bio-orthogonal conjugation in tissue engineering applications.^34,35 To obtain the injectable hydrogel cross-linked by Nb–Tz click chemistry, a series of functionalized PEG macromers, including ROS-responsive 4-arm-PEG-TK-Nb, non-ROS-responsive 4-arm-PEG-Nb, and 4-arm-PEG-Tz, were respectively synthesized.

First, to introduce the ROS-responsive TK linker into the polymer network of the hydrogel, a small molecule containing TK, acetone-[bis-(2-amino-ethyl)-dithioacetal], was synthesized as shown in Fig. S1.† The successful synthesis of corresponding products in the synthetic route was confirmed by ¹H NMR spectra (Fig. S2–S4†). For the preparation of the TK-conjugated 4-arm-PEG, the terminal carboxyl on the 4-arm poly(ethylene glycol) carboxylic acid (4-arm PEG-COOH) was first activated by pentafluorophenol (PFP) to obtain the active 4-arm-PEG-PFP, which exerted higher reactivity. Both ¹H NMR and ¹⁹F NMR spectra of 4-arm-PEG-PFP were recorded to confirm the successful activation (Fig. S5†). Subsequently, TK was conjugated to the 4-arm-PEG through the reaction between the amine group of acetone-[bis-(2-amino-ethyl)-dithioacetal] and the active pentafluorophenyl ester of 4-arm-PEG-PFP. As shown in the ¹H NMR spectrum of 4-arm-PEG-TK in Fig. S6,† the characteristic peaks of methyl (–CH₃) at 1.6 ppm demonstrated that TK was successfully conjugated onto 4-arm-PEG. The grafting ratio of TK to 4-arm-PEG was calculated to be about 100% through the integration area of δ 1.6 ppm (characteristic peak of methyl in acetone-[bis-(2-amino-ethyl)-dithioacetal]) to δ 3.5 ppm (characteristic peak of PEG). To obtain clickable 4-arm-PEG-TK-Nb, 5-norbornene-2-carboxylic acid (Nb–COOH) was firstly activated to Nb-PFP, and then reacted with another unreacted amine group of 4-arm-PEG-TK. In the meantime, the amine-ended 4-arm-PEG-NH₂ was used in place of the amine-ended 4-arm-PEG-TK to react with Nb-PFP to prepare the non-ROS-responsive 4-arm-PEG-Nb as a control. The corresponding ¹H NMR spectra of Nb-PFP, 4-arm-PEG-TK-Nb, and 4-arm-PEG-Nb are respectively shown in Fig. S7,†Fig. 1A (the red line), and Fig. S8,† and the characterized peaks at δ 5.50–6.50 ppm, which belong to the Nb group, could be observed in 4-arm-PEG-TK-Nb and 4-arm-PEG-Nb. Another clickable polymer, 4-arm-PEG-Tz, was successfully synthesized from 4-arm-PEG-COOH and 3-(p-benzylamino)-1,2,4,5 tetrazine by using the benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (PyBOP) coupling reagent. According to the ¹H NMR spectra illustrated in Fig. 1A (the blue one), the characteristic peaks at δ 7.50 and 8.50 ppm demonstrated that the Tz group was successfully grafted onto 4-arm PEG. Based on these results of ¹H NMR, the grafting ratios of Nb to 4-arm PEG or 4-arm-PEG-TK and Tz to 4-arm PEG were about 100%, respectively. Moreover, all these clickable 4-arm-PEG macromers showed excellent water solubility up to 50 wt% and high reactivity of the bio-orthogonal reaction.

Fig. 1 (A) ¹H NMR spectra of 4-arm-PEG-Tz, 4-arm-PEG-TK-Nb, and 4-arm-PEG. (B) Photographs of the gelation and injectable ability of the PEG-TK hydrogel. Representative SEM images of (C) PEG hydrogel, (D) PEG-TK hydrogel, and (E) EGF@PEG-TK hydrogel. Scale bar = 100 μm. (F) Rheological properties of the gelation process of the PEG-TK hydrogel. (G) The G′ and G′′ of the PEG-TK hydrogel (red) and PEG hydrogel (black) with frequencies from 0.1 to 10 Hz. (H) The compression test of the PEG-TK hydrogel (red) and PEG hydrogel (black).

2.2 Fabrication and physiochemical properties of the hydrogel

Non-ROS-responsive PEG hydrogel or ROS-responsive PEG-TK hydrogel made from 4-arm-PEG-Tz and 4-arm-PEG-Nb or 4-arm-PEG-TK-Nb were fabricated both at a precursor concentration of 10 wt%. As shown in Scheme 1, the cross-linking between Nb end groups of 4-arm-PEG-TK-Nb and Tz end groups of 4-arm-PEG-Tz under physiological conditions was conducted by the cycloaddition reaction. The PEG-TK hydrogel was quickly formed in 3 min by mixing precursor solutions at room temperature without external stimulation, which makes this Nb–Tz based gelation system injectable and biocompatible (Fig. 1B). To gain further insight into the gelation process of the hydrogel, the rheological data of the gelation process of 4-arm-PEG-TK-Nb/4-arm-PEG-Tz or 4-arm-PEG-Nb/4-arm-PEG-Tz precursor solutions were monitored by a time sweeping rheological test using a Haake rotational rheometer. During the rheological measurement, after mixing the precursor solutions, the storage modulus (G′) and the loss modulus (G′′) both increased over time, and G′ was lower than G′′. When G′ exceeded G′′, the mixture began to transform solution to solid, and the hydrogel formed a stable inner structure after G′ and G′′ achieved relatively stable values. The time when the value of G′ is equal to the value of G′′ is considered as the gelation time.^36,37 Based on this, the gelation times of the PEG-TK hydrogel at the precursor solution concentration of 5 wt% and 20% were measured to be about 650 s and 30 s, respectively. When the precursor concentration was adjusted to 10 wt%, the gelation times of the PEG-TK hydrogel and PEG hydrogel were 130 s and 155 s (Fig. 1F and Fig. S9†), which are appropriate operation times for the in vivo experiments. Thus, we chose 10 wt% as the precursor concentration of 4-arm-PEG-TK-Nb and 4-arm-PEG-Tz for subsequent experiments.

Next, the porous network of the freeze-dried hydrogels was observed by scanning electron microscopy (SEM). As shown in Fig. 1C–E, the PEG hydrogel, PEG-TK hydrogel, and EGF@PEG-TK hydrogel all showed similar surface and internal microporosities, and the encapsulating of growth factors in the hydrogel did not cause morphology and microstructure changes. All the three hydrogels presented irregular and porous network structures with the pore size ranging from 50 to 200 μm, which contributed to the permeability of gas and nutrients, supplying an appropriate microenvironment for cellular proliferation and vascularization.³⁸ To determine the mechanical strength of PEG-TK and PEG hydrogel, a 1% strain frequency (0.1–10 Hz) sweep test was carried out to monitor the G′ and G′′ after the hydrogels were completely formed. The G′ and G′′ represent the elastic nature and viscous nature of materials, respectively. As shown in Fig. 1G, after the complete gelation of PEG or PEG-TK hydrogels, the G′ values were higher than the G′′ during the whole frequency range from 0.1 Hz to 10 Hz, which indicated that the formed hydrogels exhibited a predominantly elastic behavior.^39,40 Furthermore, the compression test of these two hydrogels was also performed, and the results presented in Fig. 1H show that the maximum compression strength of the PEG-TK hydrogel and PEG hydrogel were about 73.7 kPa and 72.3 kPa, respectively. The rheological and compression tests both demonstrated that the hydrogels have favorable mechanical properties to resist external physical stimulation.⁴¹ Moreover, the adhesive properties of the PEG-TK hydrogel and PEG hydrogel were evaluated by a lap shear test (Fig. S10A†). The results showed that the adhesive strength of PEG-TK and PEG hydrogels could reach about 5 kPa in a porcine skin model (Fig. S10B†), indicating that these injectable hydrogels could remain attached to intact peri-wound skin without falling off and provide an enclosed environment to avoid bacterial infection.³⁷ The tissue adhesion ability of PEG-TK and PEG hydrogel may result from the topological entanglement between PEG chains and tissue.⁴²

2.3 ROS-triggered degradation and release behavior of the PEG-TK hydrogel

Natural biomacromolecule-based hydrogels, such as hyaluronic acid, collagen, or gelatin-formed hydrogels, can usually be degraded by the corresponding enzymes, including hyaluronidase and collagenase. ROS are the main byproducts of various biochemical reactions, which can be widely found in living organisms, especially in the injured areas. Taking into account the ROS-responsive property of the TK linker, the ROS-triggered degradation behavior of the PEG-TK hydrogel was studied. The ROS-responsive PEG-TK hydrogel and non-responsive PEG hydrogel were immersed in deionized water that contained different concentrations of H₂O₂ or ˙OH for different periods, and these hydrogels were lyophilized and weighed to calculate the degradation rate. As shown in Fig. 2A, the PEG-TK hydrogel exhibited a concentration- and time-dependent degradation property with the increase in the H₂O₂ concentration and incubation time. The degradation rate of the PEG-TK hydrogel could reach up to 64% and 86% after being incubated with 500 mM H₂O₂ for 7 days and 14 days, respectively. In contrast, no obvious weight loss could be observed in the H₂O₂-treated PEG hydrogel over time due to the lack of ROS-responsive TK groups in the PEG hydrogel (Fig. 2B). Furthermore, CoCl₂ and H₂O₂ were mixed to provide the ˙OH-enriched environment to evaluate the ˙OH-triggered degradation behavior of these two hydrogels. As presented in Fig. 2C and D, both the PEG-TK hydrogel and PEG hydrogel had weight losses under the ˙OH conditions. The ˙OH-induced degradation of the PEG hydrogel could be attributed to the oxidative reaction between the ether bond and ˙OH, which resulted in the breakage of the PEG chain.^43,44 As expected, due to the chain breakage in the TK linker and ether bond, the PEG-TK hydrogel showed a more notable degradation behavior under the same concentration of ˙OH. The degradation rate of the PEG-TK hydrogel reached up to 93% in the solution containing 100 mM H₂O₂ and 2 mM CoCl₂, indicating that the PEG-TK hydrogel was almost all degraded. Fig. 2E shows the macroscopic photographs of the degradation behavior of the PEG-TK hydrogel in the H₂O₂ and ˙OH environments. A commercially mixed food colorant that includes indigo and brilliant blue as ingredients was mixed in the precursor solution and the blue hydrogels were obtained to visualize the degradation process. After immersion in PBS, PBS containing H₂O₂, or PBS containing ˙OH for 7 days, the hydrogels in the ROS-rich environment had swollen to three-fold the initial volume, which also showed a larger volume than the hydrogel in PBS without any ROS. It should be pointed out that the blue colorant would be reduced by H₂O₂ and then lose its blue color, so the hydrogel in the H₂O₂ buffer became colorless with time increased (the third row in Fig. 2E). On day 14, the hydrogels in the ROS-rich environment became liquid and flowed to the bottom of the inverted centrifuge tube due to the cleavage of the TK bonds and degradation of the hydrogel. However, the hydrogel without ROS treatment was still stuck in the tip of the centrifuge tube and just released the colorants into the PBS buffer. These photographs visually illustrate the degradation process of PEG-TK hydrogel in the ROS environment.

Fig. 2 Degradation rates of (A) the PEG-TK hydrogel and (B) PEG hydrogel over time in the solution containing different concentrations of H₂O₂. Degradation rates of (C) the PEG-TK hydrogel and (D) PEG hydrogel over time in the solution containing different quantities of ˙OH generated by mixing H₂O₂ and CoCl₂. (E) Digital photographs of the degradation behavior of the PEG-TK hydrogel under the conditions with ˙OH or H₂O₂ on day 7 and day 14. (F) The release behavior of FITC-BSA from the PEG-TK hydrogel in PBS (black) or PBS containing 10 mM H₂O₂ (red). The P values were calculated by Student's t-test, *p < 0.05, **p < 0.01 and ***p < 0.001.

To demonstrate the potential application of the ROS-responsive PEG-TK hydrogel for the growth factor delivery, fluorescein isothiocyanate-labeled bovine serum albumin (FITC-BSA) was chosen as the model protein to mimic the release behavior of growth factors from the hydrogel. The in vitro release profiles of FITC-BSA from the PEG-TK hydrogel in the PBS buffer containing 10 mM H₂O₂, or not, are shown in Fig. 2F. Under the condition without ROS, the amount of released FITC-BSA reached 8% on day 1. These released proteins were absorbed onto the surface of the hydrogel network, which could be rapidly released into the buffer. With time, the large release of FITC-BSA could not be observed in the PBS group, and the cumulatively released FITC-BSA only reached 25% on day 6. However, in the group that utilized H₂O₂ to mimic the ROS environment, the ROS-triggered hydrogel degradation resulted in an approximately 2-fold greater release of FITC-BSA than the group without ROS. The cumulative release of FITC-BSA reached up to 31% and 58% on day 1 and day 6, respectively. In hydrogel-based drug delivery systems, 100% protein release is hard to achieve due to the limited motion of the entrapped molecules in the highly entangled network.^45,46 These results demonstrated that the PEG-TK hydrogel could be used to encapsulate and deliver proteins, as well as avoid burst release and achieve the controlled slow release of payloads via responding to the ROS changes in the surrounding environment.

2.4 In vitro radical scavenging capability and biocompatibility of PEG-TK hydrogel

To evaluate the antioxidant properties of the PEG-TK hydrogel, DPPH radical and ˙OH scavenging assays were performed. The stable DPPH free radical is the most used reagent to test the ROS scavenging ability of materials; its absorbance at around 517 nm is decreased when DPPH accepts an electronic or hydrogen radical from the antioxidant to form a diamagnetic substance.⁴⁷ In the meantime, the degree of discoloration of the DPPH solution could reflect the scavenging potential of the antioxidant compounds in terms of hydrogen donating ability. First, we evaluated the DPPH scavenging ability of the 4-arm-PEG-TK-Nb macromer over time (Fig. S11†). The absorbance of DPPH at 517 nm was reduced from 0.72 to 0.35 after 24 h treatment with the 4-arm-PEG-TK-Nb macromer. Furthermore, the DPPH scavenging activities of the PEG-TK hydrogel and PEG hydrogel were measured and the calculated scavenging efficiencies are presented in Fig. 3A. Both the PEG-TK hydrogel and PEG hydrogel showed the increased DPPH scavenging effect over time. Interestingly, the PEG hydrogel, which did not contain the ROS-responsive linker in its polymer chain, also showed the ability to consume DPPH. This could be attributed to the imine formed from the click reaction between Nb and Tz, which donated an electron to DPPH to exert the antioxidant effect.^48,49 However, under the synergistic effect of TK bonds and imine groups, PEG-TK hydrogel exhibited a better radical scavenging efficiency, which could reach up to 80% after 24 h treatment. The color change in the DPPH solution with different treatments also verified the better antioxidant potential of the PEG-TK hydrogel (shown in the inset photograph in Fig. 3A). In addition, terephthalic acid (TA), which can be transformed into fluorescent 2-hydroxyterephthalic acid (HTA) by ˙OH, was utilized to evaluate the ˙OH scavenging capability of the PEG-TK hydrogel. As shown in Fig. 3B, a decrease in the fluorescence emission peak at 425 nm could be observed in the presence of the PEG-TK hydrogel for 24 h, which indicated that it can eliminate ˙OH. In comparison, no obvious fluorescence decrease was detected in the presence of the PEG hydrogel. All the above results demonstrated the better removal capacity of PEG-TK hydrogel towards ROS.

Fig. 3 (A) The DPPH radical scavenging capability of the PEG hydrogel and PEG-TK hydrogel over time (inset: photographs of the color change of DPPH mixed with PBS, PEG hydrogel, and PEG-TK hydrogel after 12 h). (B) Clearance of ˙OH by the PEG hydrogel and PEG-TK hydrogel after 24 h treatment. (C) The cell viability of L929 cells that were mixed with PEG hydrogel or PEG-TK hydrogel and further 3D cultured for 1, 3, and 5 days. (D) Representative CLSM images of L929 cells stained by calcein-AM/PI in the L929-PEG hydrogel system and L929-PEG-TK hydrogel system, respectively. Scale bar = 200 μm. (E) Representative CLSM images of ROS formation in the L929-PEG hydrogel system and L929-PEG-TK hydrogel system which exposured to H₂O₂ for 2 days. Scale bar = 200 μm. (F) The corresponding quantitative analysis of DCF signals of L929 cells in (E). The P values were calculated by Student's t-test, *p < 0.05, and **p < 0.01.

Favorable cytocompatibility is the prerequisite for the in vivo application of the ROS scavenging hydrogel dressing. First, the proliferation ability of mouse fibroblasts (L929 cells) in 2D culture condition is evaluated through the WST-1 assay and calcein acetoxymethyl ester (calcein-AM)/propidium iodide (PI) staining kit. As for the cell viability and confocal laser scanning microscopy (CLSM) results shown in Fig. S12,† the already cross-linked PEG or PEG-TK hydrogels that were immersed in the cell culture medium did not release toxic substances to compromise the viability of the L929 cells after 5 days of co-incubation. Moreover, PEG-TK and PEG hydrogels as 3D culture scaffolds for L929 cells proliferation were also investigated. Compared to the 2D culture, the 3D cell culture could realistically simulate the microenvironment of biochemical and biophysical properties, as well as provide more accurate results of cytocompatibility.^50,51 The L929 cells in the 1640 medium were mixed with PEG precursor solutions, which were cross-linked to form a cell-hydrogel system via the Nb–Tz bio-orthogonal reaction in a few minutes. According to the results of the WST-1 assay presented in Fig. 3C, the OD values increased steadily from 0.38 and 0.42 on day 1 to 1.29 and 1.30 on day 5 in the L929-PEG hydrogel system and L929-PEG-TK hydrogel system, respectively. The 3D growth of L929 cells in hydrogels was also monitored by CLSM through calcein-AM/PI staining. It has been reported that the cells tended to clump in the microenvironment of the extracellular matrix and form cellular spheres because of a lack of cell-adhesion sites.^52,53 Not surprisingly, some multicellular spheres could be observed on day 5 and scarcely any dead cells were found during the whole co-incubation process (Fig. 3D). Both the cell proliferation and live/dead staining results in 2D and 3D culturing revealed that PEG and PEG-TK hydrogels had good cytocompatibility and would not cause physiological toxicity to wound tissues. The high-water content and porous network structures of PEG-based hydrogels endowed them with the ability to facilitate the permeability of nutrients and gas as well as the transportation of metabolites.⁴⁶ Moreover, the hydrophilic PEG spacer in the scaffold structure could lead to a similar pattern of bioactive compound attached to the surfaces of hydrogel and facilitate cell proliferation by up-regulating the expression of certain proliferation-related genes.^54,55 Taken together, the PEG-TK hydrogel provides an appropriate microenvironment for cell proliferation.

Next, to investigate the intracellular ROS-scavenging ability of PEG-TK, the ROS fluorescence probe, 2′,7′-dichlorofluorescin diacetate (DCFH-DA) was utilized to measure the intracellular ROS level. The cell-hydrogel system was immersed in the oxidative medium, which contained 500 μM H₂O₂ to simulate the pathological oxidative stress microenvironment, and after 2 days, the cell-hydrogel was stained by DCFH-DA. As shown in the CLSM images in Fig. 3E, the intracellular ROS level was largely decreased in the L929-PEG-TK hydrogel system as compared to that in the L929-PEG hydrogel system. The corresponding semi-quantitative analysis of CLSM results showed that the fluorescence intensity of the L929-PEG-TK hydrogel was 46% of that of the L929-PEG hydrogel (Fig. 3F), indicating the efficient antioxidant ability of the PEG-TK hydrogel.

2.5 In vivo foreign body response of the PEG-TK hydrogel

PEG-based materials are considered representative biocompatible and antifouling agents, which have been widely used as implantable biomaterials and medical devices.⁵⁶ However, some studies have reported that PEG-based therapeutics may cause allergic reactions in dermal applications or induce foreign body responses (FBR) after implantation.^56,57 To assess the FBR of PEG and PEG-TK hydrogels, precursor solutions were subcutaneously injected under the back skin of ICR mice and formed PEG or PEG-TK hydrogels. After 14 and 28 days, as shown in Fig. 4A, the residual hydrogels and surrounding skin tissues were excised and collected for subsequent histological analysis. Hematoxylin and eosin (H&E) staining results (Fig. 4B and C) showed a thicker inflammatory capsule surrounding the PEG hydrogels as compared to the PEG-TK hydrogel on both day 14 and day 28, which is consistent with previous reports.⁵⁸ Chronic inflammation always starts 7 days post-implantation and the recruited macrophages would release abundant ROS to attempt to phagocytose the implant, which caused acute immune responses toward the foreign body. The lesser inflammatory response provoked by the PEG-TK hydrogel may be attributed to the ROS scavenging ability of TK bonds to alleviate the FBR. In addition, there were more disintegrated holes in the residual PEG or PEG-TK hydrogels as time increased. However, the volume of the residual PEG-TK hydrogel was smaller than that of the residual PEG hydrogel group, indicating a faster degradation rate of the PEG-TK hydrogel. This phenomenon is consistent with the in vitro degradation process.

Fig. 4 (A) The gross view of the residual hydrogel and subcutaneous tissue that was resected for FBR analysis. (B) Histology (H&E staining) of FBR to PEG and PEG-TK hydrogels on days 14 and 28 post-injection in ICR mice. The red arrowhead indicates the hole formed during the hydrogel degradation process. Scale bar = 100 μm. (C) The quantification of inflammation thickness surrounding PEG or PEG-TK hydrogels. The P values were calculated using Student's t-test, **p < 0.01.

2.6 In vivo ROS-scavenging ability of the PEG-TK hydrogel

A high level of ROS in wounded skin will induce oxidative stress and lead to further tissue injury. In the early inflammatory stage, the concentration of H₂O₂ is far greater than that in normal tissue.⁴ In addition to H₂O₂, the bursting production of superoxide radical (˙O₂⁻) would increase to the peak after being injured for 2 days.^4,59 Encouraged by the efficient ROS-scavenging ability of the PEG-TK hydrogel in cell experiments, we then investigated whether it could decrease ˙O₂⁻ levels in a skin wound healing model. Two days after the wound defect modeling and hydrogel treatment, the ROS levels in the injury tissue sites were stained by dihydroethidium (DHE), whose fluorescence changed from blue to red after entering the cell membrane to react with ˙O₂⁻. From the CLSM images of DHE stained sections (Fig. 5A) and the corresponding quantitative statistics results (Fig. 5B), it could be observed that the levels of ROS were significantly higher in the blank group and PEG hydrogel-treated group when compared to the PEG-TK hydrogel-treated groups. The fluorescence signals in the PEG-TK hydrogel group and growth factor-loaded EGF@PEG-TK hydrogel group diminished, which decreased to 70% and 47%, respectively. These results confirmed the protective role of the PEG-TK hydrogel when used as an antioxidant wound dressing from the oxidative stress generated at injury sites. Moreover, the released EGF from the EGF@PEG-TK hydrogel could reduce the intracellular ROS levels of fibroblasts through up-regulating the expression of antioxidant enzymes, including superoxide dismutase-1 (SOD-1) and catalase (CAT).⁶⁰ Therefore, a better synergistic effect on alleviating oxidative stress could be obtained by combining the ROS-responsive TK linker and EGF in one EGF@PEG-TK hydrogel wound dressing.

Fig. 5 (A) Representative CLSM images of DHE stained wound sections of SD rats with different treatments. Scale bar = 200 μm. (B) Quantification analysis of the ROS scavenging ability of different hydrogels by measuring the mean grey values of DHE signals in (A). The P values were calculated using Student's t-test, **p < 0.01.

2.7 Enhanced regeneration of PEG-TK hydrogels in a full-thickness skin defect model

To further evaluate the wound healing efficacy of the EGF@PEG-TK hydrogel, the in vivo experiments were performed in the full-skin defect model on SD rats. After the surgery of wound defect modeling, various precursor solutions were directly injected into the wound site, which diffused and filled the wound area and then gelled to cover the wound area completely. The changes in wound sites along the time of transplantation were recorded in photographs, and the side-by-side comparisons of the wounds treated with different hydrogels on days 0, 5, 10, 15, and 20 are shown in Fig. 6A. It was found that the wound areas that were treated with the PEG-TK hydrogel or EGF@PEG-TK hydrogel were smaller than the blank control and PEG hydrogel-treated group. From the wound closure curves (Fig. 6B) obtained from photo analysis software, wounds treated with PEG-TK and EGF@PEG-TK hydrogels exhibited that the closure rates of wound healing reached up to 88% and 90%, respectively, both of which were higher than the rates of the control group (79%) and PEG hydrogel-covered group (81%). These results indicated that the introduction of TK linker in the hydrogel dressing has a beneficial effect on wound healing. Interestingly, at the early wound regeneration stage, the EGF@PEG-TK hydrogel treatment led to an obvious shrinkage and a higher slope of closure curve, which demonstrated that the TK linker and EGF synergistically accelerated the wound healing phase transition from inflammation to proliferation.

Fig. 6 (A) Representative photographs of the wound sites during 20 days with the treatments of PBS, PEG hydrogel, PEG-TK hydrogel, and EGF@PEG-TK hydrogel, respectively. Scale bar = 0.5 cm. (B) Quantification of the wound closure process for different groups. (C) Representative H&E staining images of wound sections on day 20 post-surgery. Scale bar = 500 μm. The dashed black lines outline the advanced edges of wounds, and the black arrows mark the space of granulation tissue. The white arrow and blue arrow in the magnified green box mark the interior of hair follicles and external root sheath of hair follicles, respectively. Scale bar in the green box = 100 μm. (D) Quantitative results of skin thickness on day 20 in various treatment groups. (E) Representative images of H&E (enlarged images of the center position of regeneration tissue) and CD31 staining tissues from the wound area after the surgery for 20 days. Scale bar = 100 μm. The red arrows mark the blood vessels. The P values were calculated using Student's t-test, *p < 0.05, and **p < 0.01.

Histological examinations were performed to evaluate the healing process. In the first week, the H&E staining examination of wound tissues in the control group and PEG hydrogel-treated group demonstrated some neutrophils, which were more in number than those in the PEG-TK hydrogel group and EGF@PEG-TK hydrogel group (Fig. S13†). Neutrophils play dual roles in wound healing. Although the recruited neutrophils could kill invading microorganisms and stimulate other immune cells to eliminate infections, the released toxic antimicrobial substances (including ROS) and proteases from abundant neutrophils also cause damage to the host tissue and delay the healing process.⁶¹ The PEG-TK and EGF@PEG-TK hydrogels, which exhibited ROS eliminating ability, decreased the FBR and inflammation reaction in the wound section, further resulting in less recruitment and activation of neutrophils. With time, on day 20, the neutrophils were found only sparsely throughout the wound in all groups. The fibroblasts and strand-like collagen bundles were visible in the groups treated with PEG-TK and EGF@PEG-TK hydrogels. Next, the granulation tissue was evaluated via the degree to which it filled the wound site, including the granulation tissue spacing and regenerated skin thickness. As shown in Fig. 6C, the contraction gap of the EGF@PEG-TK hydrogel group (1932.2 ± 101.3 μm) is much smaller than that in the PEG-TK hydrogel group (2380.0 ± 42.4 μm), the PEG hydrogel group (3156.3 ± 71.0 μm) and the blank control group (3367 ± 73 μm). In addition to the spacing of granulation tissue, their thickness, including epidermis and dermis, also reflects the degree of wound healing.⁶² The results of quantitative analysis on skin thickness (Fig. 6D) showed that the skin thickness of the PEG-TK hydrogel-treated group and EGF@PEG-TK hydrogel-treated group were up to ∼913 μm and ∼1078 μm, respectively, which were higher than that of the blank group (∼583 μm) and PEG hydrogel-treated group (∼736 μm). Moreover, new hair follicles (the magnified image in the green box of Fig. 6C) could be found in the boundary of regeneration granulation tissue in the EGF@PEG-TK group, which suggested the critical influence of the EGF@PEG-TK hydrogel on the proliferation and differentiation of epithelial cells. Collagen is the largest class of fibrous extracellular matrix (ECM), which plays an important role in wound healing.⁶³ After 20 days of treatment, the regularly arranged collagen could also be found in the Masson's trichrome staining images of PEG-TK and EGF@PEG-TK hydrogel-treated groups (Fig. S14†), which further demonstrated the maturity of the regenerated tissue. Angiogenesis is a critical event in the wound-healing process. The delivery of nutrition and oxygen from the newly formed blood vessels is essential for granulation tissue formation and wound repair.⁶⁴ In this study, the immunohistochemical staining of platelet endothelial cell adhesion molecule-1 (CD31) was utilized to evaluate angiogenesis in all groups. As exhibited in Fig. 6E, the strong positive staining of CD31, which indicated numerous blood vessels, was presented in the wound section of the EGF@PEG-TK hydrogel-treated group. This result indicated that the released EGF from the EGF@PEG-TK hydrogel promoted angiogenesis for the wounds.

The above evidence suggested that the EGF@PEG-TK hydrogel promoted granulation tissue formation, collagen deposition, epithelialization, and vascularization, further enhancing the healing efficacy. It can be speculated that the mechanism of enhanced healing efficacy was related to the injectable and adhesion properties, the ROS-scavenging ability of the PEG-TK hydrogel, as well as the loaded EGF, which could stimulate the keratinocyte proliferation, fibroblast function, angiogenesis, and collagenase activity.⁶⁵ First, the injectable and adhesion properties caused the PEG-TK hydrogel to adapt to the inhomogeneous wound site and join the wound tissue tightly.⁴² Moreover, the high-water content, porous structure, and excellent mechanical properties of the hydrogel dressing could keep the wound moist, benefit gas exchange, and resist external physical stimulation.^15,66 Second, the ROS-scavenging ability of the PEG-TK hydrogel, which is attributed to the introduction of the TK linker and the imine groups formed from the click bio-orthogonal reaction, could eliminate the excessive ROS during tissue regeneration, provide a suitable microenvironment to prevent cell aging, decrease the inflammatory response, as well as synergistically promote the wound healing with the released functional molecules. Third, the encapsulation of EGF in the hydrogel network and ROS-responsive EGF release in situ prolonged the bioavailability of EGF in the wound healing process. The ROS-responsive TK linker not only consumed the excessive ROS in the wound site but also achieved the controlled release of EGF. The ROS-responsive chemical groups and released EGF synergistically alleviated oxidative stress, protected the cells and tissue from further damage in the early stage of wound healing, promoted the formation of granulation tissue, and finally accelerated wound recovery.

3. Conclusions

We prepared an injectable hydrogel composed of a 4-arm-PEG network with ROS-cleavable thioketal linkers as a multifunctional wound dressing for EGF controlled release. The bio-orthogonal pair between Nb and Tz endowed the hydrogel with a fast gelation time, injectability and shape-adaptability. The ROS scavenging capacity, ROS-triggered degradation, and EGF release behavior, as well as biocompatibility of the hydrogel, were systematically verified in vitro and in vivo. Finally, it was found that the EGF@PEG-TK hydrogel could promote skin wound healing via synergistically scavenging ROS and releasing EGF in the full-thickness skin defect model. Given the excellent properties of the PEG-TK hydrogel dressing, including injectability, good biocompatibility, strong antioxidant ability, and ROS-responsive degradability, it is a promising candidate for biomedical applications.

4. Experimental section

The materials used in this study and the detailed synthetic procedures for acetone-[bis-(2-amino-ethyl)-dithioacetal] and Nb-PFP are described in the ESI.†

4.1 Synthesis of 4-arm-PEG-PFP

4-Arm-PEG-COOH (1 g, 0.05 mmol), PFP (184 mg, 1 mmol) and N,N′-dicyclohexylcarbodiimide (DCC) (211 mg, 1 mmol) were dissolved in dry 1,4-dioxane. After stirring overnight under a nitrogen atmosphere at room temperature, the mixture was concentrated and precipitated in cool diethyl ether. The final product was achieved through centrifugation and vacuum drying overnight. The ¹H NMR and ¹⁹F NMR spectra of 4-arm-PEG-PFP are shown in Fig. S5.†

4.2 Synthesis of 4-arm-PEG-TK

4-Arm-PEG-PFP (1 g, 0.05 mmol) and N,N-diisopropylethylamine (DIPEA) (104 μL, 0.6 mmol) were dissolved in 50 mL of DCM and added dropwise into 200 mL of the acetone-[bis-(2-amino-ethyl)-dithioacetal] (777.44 mg, 4 mmol)/DCM solution under the ice-bath. Afterwards, the mixture was allowed to return to room temperature and stirred overnight. The 4-arm-PEG-TK was obtained by precipitation in cool diethyl ether and drying in a vacuum overnight. The ¹H NMR spectrum of 4-arm-PEG-TK is shown in Fig. S6.†

4.3 Synthesis of 4-arm-PEG-TK-Nb

4-Arm-PEG-TK (1 g, 0.05 mmol), Nb-PFP (180 mg, 0.6 mmol) and DIPEA (104 μL, 0.6 mmol) were dissolved in 20 mL of DCM and stirred overnight at room temperature. Afterward, the mixture was concentrated and precipitated in cool diethyl ether. After being isolated via centrifugation, the obtained macromer was transferred to the dialysis tube (MWCO = 3500 Da) and dialyzed against deionized water for 3 days. Finally, the solution was lyophilized to obtain 4-arm-PEG-TK-Nb. The ¹H NMR spectrum of 4-arm-PEG-TK-Nb is shown in Fig. 1A.

4.4 Synthesis of 4-arm-PEG-Nb

To synthesize the non-ROS-responsive clickable 4-arm-PEG-Nb as a control macromer, a similar procedure to 4-arm-PEG-TK-Nb was performed by using 4-arm-PEG-NH₂ instead of 4-arm-PEG-TK. The molar ratio of 4-arm-PEG-NH₂, Nb-PFP, and DIPEA was 0.5 : 6 : 6, and the post-processing was the same as 4-arm-PEG-TK-Nb. The ¹H NMR spectrum of 4-arm-PEG-Nb is shown in Fig. S8.†

4.5 Synthesis of 4-arm-PEG-Tz

4-Arm-PEG-COOH (1 g, 0.05 mmol), DIPEA (126 μL, 0.76 mmol), PyBOP (260 mg, 0.5 mmol) and 3-(p-benzylamino)-1,2,4,5 tetrazine (44.9 mg, 1.2 mmol) were dissolved in 20 mL of DCM and stirred overnight under a nitrogen atmosphere at room temperature. Afterward, the mixture was concentrated and precipitated in cool diethyl ether. The final product was achieved through centrifugation and the precipitate was dispersed in deionized water and dialyzed (MWCO = 3500 Da) against water for 3 days. Finally, the solution was lyophilized to obtain 4-arm-PEG-Tz. The ¹H NMR spectrum of 4-arm-PEG-Tz is shown in Fig. 1A.

4.6 Preparation of hydrogel scaffolds

The hydrogel was formed through the click reaction between Nb and Tz. Briefly, 10% of precursor solutions of 4-arm-PEG-TK-Nb, 4-arm-PEG-Nb, and 4-arm-PEG-Tz were prepared. To fabricate the PEG-TK hydrogel, equal volumes of the precursor solutions of 4-arm-PEG-TK-Nb and 4-arm-PEG-Tz were mixed uniformly. The PEG hydrogel was fabricated through a similar method to 4-arm-PEG-Nb and 4-arm-PEG-Tz.

4.7 Characterization of hydrogels

First, the internal microscopic morphologies of the hydrogels were observed via SEM (FEI Quanta 250 FEG). Briefly, hydrogels were frozen in liquid nitrogen, followed by freeze-drying. The lyophilized hydrogels were sputter-coated with gold for imaging. Rheological tests of the hydrogels were investigated on a Haake rotational rheometer (Malvern Kinexus pro⁺). G′ and G′′ of the hydrogels were measured using a Haake rotational rheometer with a 1% strain frequency (0.1–10 Hz) sweep. In addition, the gelation times of the hydrogels were measured using the Haake rotational rheometer with 1% strain and 1 Hz frequency. To measure the compressive properties of the scaffolds (diameter 7 mm, height 9 mm) in hydrated conditions, the hydrogels were analyzed using a universal material testing machine with a 10 N loading unit at a cross-head speed of 1 mm min⁻¹ at room temperature. The compressive modulus was calculated as the slope of the linear elastic region of the stress–strain curve.

4.8 ROS-triggered degradation in vitro

The degradation tests of hydrogels under excessive H₂O₂ or ˙OH environments were performed by measuring the mass loss of the hydrogels. First, 50 μL of hydrogels were immersed in deionized water to achieve the swelling equilibrium. After that, the swelled hydrogels were immersed in the water containing H₂O₂ or ˙OH (˙OH generated from the reaction of H₂O₂ with CoCl₂), and the ROS-abundant aqueous solution was changed once a day. The hydrogels were then lyophilized and weighed on day 0 (m_e), day 7 (m_w) and day 14 (m_w). The degradation behavior was estimated according to (m_e − m_w)/m_e × 100%. Moreover, in order to observe the ROS-triggered degradation process, a food colorant named tiffany blue, which was composed of indigo and brilliant blue, was used to stain the PEG-TK hydrogel. Then, 20 μL of PEG-TK hydrogels stained by the blue colorant were immersed in 60 μL of PBS, PBS containing H₂O₂, or PBS containing ˙OH, respectively. The centrifuge tubes were inverted and photographed on day 0, day 7, and day 14 to record the real situation.

4.9 DPPH radical scavenging ability

The DPPH radical scavenging capabilities of the 4-arm-PEG-TK-Nb macromer and the hydrogels were determined by a UV-Vis spectrophotometer. Firstly, 100 μL of 4-arm-PEG-TK-Nb aqueous solution (500 mg mL⁻¹) was mixed quickly with 500 μL of 100 μM DPPH ethanol solution in the dark. The UV-vis spectra of the mixture (DPPH/4-arm-PEG-TK-Nb) around 517 nm were recorded at different time points during a 24 h period. After that, 200 μL crosslinked PEG hydrogel or PEG-TK hydrogel were soaked in 100 μM DPPH ethanol solution in the dark. The absorbance of the soaking solution (A_sample) or original DPPH ethanol solution (A_control) at 517 nm was measured at different time points. The ability to clear DPPH radicals was calculated according to (A_control − A_sample)/A_control × 100%.

4.10 ˙OH radical scavenging ability

TA, which could capture ˙OH to produce the fluorescent HTA compound, was utilized as the probe for detecting the amount of ˙OH radical. The ˙OH scavenging ability of hydrogels was evaluated by measuring the fluorescence emission intensity of the generated HTA (λ_ex = 320 nm, λ_em = 425 nm). Here, 50 μL of PBS, the crosslinked PEG hydrogel or PEG-TK hydrogel were immersed in 500 μL of solution that contained TA (0.05 mM) and H₂O₂ (100 mM) in the dark, respectively. After 24 h, the fluorescence spectrum of each mixture was recorded by a fluorescent spectrometer.

4.11 ROS-triggered protein release from the PEG-TK hydrogel

Fluorescent BSA-FITC was used as a model protein to mimic the release behavior of EGF from the PEG-TK hydrogel. First, 0.5 mg mL⁻¹ of BSA-FITC was mixed with the precursor solutions (25 μL of 4-arm-PEG-TK-Nb and 25 μL of 4-arm-PEG-TK-Tz), and these precursor solutions were mixed thoroughly to form the BSA-loaded PEG-TK hydrogel. These BSA-loaded PEG-TK hydrogels were then immersed in 1 mL of releasing buffer of PBS or PBS containing 10 mM H₂O₂ in the dark, respectively, and the release experiments were performed at 37 °C in the dark. At specific time points, 200 μL of buffer was taken out and measured via the fluorescence spectrophotometer (λ_ex = 488 nm, λ_em = 520 nm) to calculate the released content of FITC-BSA. After that, the buffer was added back to the releasing medium, and shaking was continued in the dark.

4.12 Cytocompatibility of hydrogel

L929 mouse fibroblasts were purchased from the National Collection of Authenticated Cell Cultures (CAS typical Culture Collection Committee cell library). L929 cells were cultured in 1640 medium with 10% FBS, 100 IU ml L⁻¹ penicillin, and 100 μg mL⁻¹ streptomycin at 37 °C with a 5% CO₂ atmosphere. First, for the 2D cytocompatibility evaluation, the PEG or PEG-TK hydrogels were first crosslinked in the molding under a sterile super clean bench and transferred into the 24-well plate. And then, the medium containing L929 cells was added to each well for co-incubating at an initial cell density of 1.0 × 10⁴ cells per well. The L929 cells directly seeded in the blank wells without hydrogel acted as the control. After co-incubating for the different periods, the cell viabilities were quantified using the WST-1 cell viability assay on day 1, day 3, and day 5. Specifically, the culture medium was pipetted and transferred to a 96-well plate and the OD value of each well was read at 450 nm via a microplate reader. For the visualization of cell proliferation in vitro, the co-incubated L929 cells were stained with the live/dead (calcein-AM/PI) kit and further observed by CLSM. Subsequently, to evaluate the proliferation of L929 cells encapsulated in the 3D PEG or PEG-TK hydrogel scaffold, L929 (10⁷ cells per mL) were separately mixed into the hydrogel precursors (each volume was 25 μL) and formed PEG or PEG-TK hydrogel scaffold loaded with cells. And then, these hydrogels loaded with L929 were transferred into a 24-well plate for continuing incubation and monitored by the similar WST-1 assay and live/dead assay.

4.13 Intracellular ROS scavenging ability

The DCFH-DA assay was performed to evaluate the protective function of PEG-TK hydrogels for L929 cells against ROS. L929 cells were encapsulated in PEG or PEG-TK hydrogel and cultured for 24 h, and then the medium of each sample was replaced by the fresh medium containing 500 μM H₂O₂. After treatment with H₂O₂ for another 2 days, the cells were stained with DCFH-DA and Hoechst33342 to assess the intracellular ROS level. The fluorescence intensity of the generated DCF in each sample was observed by CLSM and further quantified by Image J software.

4.14 Foreign body response evaluation

All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Suzhou Institute of Nano-Tech and Nano-Bionics, and approved by the Animal Experimental Ethics Committee of Suzhou Institute of Nano-Tech and Nano-Bionics. The ICR mice (female, 4-week-old) were purchased from Shanghai SLAC Laboratory Animal Co. Ltd. Firstly, 20 μL of mixed sterile precursor solution (10 μL 4-arm-PEG-Tz solution and 10 μL 4-arm-PEG-TK-Nb solution or 10 μL 4-arm-PEG-Tz solution and 10 μL 4-arm-PEG-Nb solution) was injected subcutaneously into ICR mice (n = 3). The hydrogel depot formed within a few minutes. After 14 days and 28 days treatment, these mice were euthanized, and the hydrogels were retrieved along with the surrounding tissues for H&E staining.

4.15 In vivo antioxidant activity

Sprague Dawley (SD) rats (female, weight ≈ 250 g) were purchased from Shanghai SLAC Laboratory Animal Co. ltd. First, the full-thickness round skin defects (1.0 cm in diameter) were prepared on the dorsal sides of SD rats. The hydrogel precursors were injected into the wound sites and then crosslinked to form various hydrogel dressings (PEG, PEG-TK, or EGF@PEG-TK hydrogel) to cover the wound sites. After 2 days of hydrogel dressing treatment, the rats were sacrificed and the wound specimens including full-thickness skin layers were retrieved and processed via cryo-sectioning. The ˙O₂⁻ fluorescent probe, DHE, was utilized to stain the sections. The fluorescence intensity of DHE observed by CLSM could reflect the ROS level in the wound sites.

4.16 In vivo wound healing

The established full-thickness skin wounds were covered with either the PEG hydrogel, PEG-TK hydrogel, or EGF@PEG-TK hydrogel. All these hydrogel dressings were changed every two days. The changes in each wound site were recorded by photographing on days 0, 5, 10, 15, and 20, and the wound areas were measured by Image J software to calculate the wound closure rate. The percent wound closure was calculated as follows:

Percent wound closure (%) = (A_t/A₀) × 100%

where A_t is the wound area after the treatment time interval “t”, and A₀ is the initial wound area on day 0. After 20 days of hydrogel dressing treatment, the entire wound area and adjacent normal skin were harvested and fixed in 4% paraformaldehyde for 1 day, and then the tissue was embedded in paraffin and cross-sectioned to 3–5 μm thick slices. The tissue sections were stained with H&E or Masson's trichrome staining kits for histological analysis.

Author contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Strategic Priority Research Program of Chinese Academy of Science (XDA16040700), the National Natural Science Foundation of China (No. 31971326 and 81801837), the Science Foundation of Jiangsu Province (No BK20180258) and the China Postdoctoral Science Foundation (No. 2019T120475).

References

1. X. Zhao, H. Wu, B. L. Guo, R. N. Dong, Y. S. Qiu and P. X. Ma, Biomaterials, 2017, 122, 34–47.
2. R. Xu, G. X. Luo, H. S. Xia, W. F. He, J. Zhao, B. Liu, J. L. Tan, J. Y. Zhou, D. S. Liu, Y. Z. Wang, Z. H. Yao, R. X. Zhan, S. S. Yang and J. Wu, Biomaterials, 2015, 40, 1–11.
3. G. P. Chen, Y. R. Yu, X. W. Wu, G. F. Wang, J. N. Ren and Y. J. Zhao, Adv. Funct. Mater., 2018, 28, 1801386.
4. S. Roy, S. Khanna, K. Nallu, T. K. Hunt and C. K. Sen, Mol. Ther., 2006, 13, 211–220.
5. C. Dunnill, T. Patton, J. Brennan, J. Barrett, M. Dryden, J. Cooke, D. Leaper and N. T. Georgopoulos, Int. Wound J., 2017, 14, 89–96.
6. P. A. Hyslop, D. B. Hinshaw, I. U. Scraufstatter, C. G. Cochrane, S. Kunz and K. Vosbeck, Free Radicals Biol. Med, 1995, 19, 31–37.
7. V. B. Mehta and G. E. Besner, Growth Factors, 2007, 25, 253–263.
8. V. B. Mehta, Y. Zhou, A. Radulescu and G. E. Besner, Growth Factors, 2008, 26, 301–315.
9. J. Ding, Y. Yao, J. Li, Y. Duan, J. R. Nakkala, X. Feng, W. Cao, Y. Wang, L. Hong, L. Shen, Z. Mao, Y. Zhu and C. Gao, Small, 2020, 16, e2005038.
10. M. S. Shim and Y. N. Xia, Angew. Chem., Int. Ed., 2013, 52, 6926–6929.
11. J. R. Martin, M. K. Gupta, J. M. Page, F. Yu, J. M. Davidson, S. A. Guelcher and C. L. Duvall, Biomaterials, 2014, 35, 3766–3776.
12. J. S. Kim, S. D. Jo, G. L. Seah, I. Kim and Y. S. Nam, J. Ind. Eng. Chem., 2015, 21, 1137–1142.
13. Y. J. Yao, J. Ding, Z. Y. Wang, H. L. Zhang, J. Q. Xie, Y. C. Wang, L. J. Hong, Z. W. Mao, J. Q. Gao and C. Y. Gao, Biomaterials, 2020, 232, 119726.
14. D. Chouhan, N. Dey, N. Bhardwaj and B. B. Mandal, Biomaterials, 2019, 216, 119267.
15. X. Y. Ding, G. Li, P. Zhang, E. Jin, C. S. Xiao and X. S. Chen, Adv. Funct. Mater., 2021, 31, 2011230.
16. X. Qiu, J. M. Zhang, L. L. Cao, Q. Jiao, J. H. Zhou, L. J. Yang, H. Zhang and Y. P. Wei, ACS Appl. Mater. Interfaces, 2021, 13, 7060–7069.
17. R. Dimatteo, N. J. Darling and T. Segura, Adv. Drug Delivery Rev., 2018, 127, 167–184.
18. Y. Gao, Z. Li, J. Huang, M. Zhao and J. Wu, J. Mater. Chem. B, 2020, 8, 8768–8780.
19. H. Q. Xu, J. Casillas, S. Krishnamoorthy and C. X. Xu, Biomed. Mater., 2020, 15, 055021.
20. E. R. Ruskowitz and C. A. DeForest, ACS Biomater. Sci. Eng., 2019, 5, 2111–2116.
21. J. Xiao, S. Chen, J. Yi, H. Zhang and G. A. Ameer, Adv. Funct. Mater., 2017, 27, 1604872.
22. N. S. Fagali, C. A. Grillo, S. Puntarulo and M. A. Fernández Lorenzo de Mele, Colloids Surf., B, 2015, 128, 480–488.
23. A. G. Gristina, Science, 1987, 237, 1588–1595.
24. J. Apinun, S. Honsawk, S. Kuptniratsaikul, J. Jamkratoke and S. Kanokpanont, Asian Biomed., 2018, 12, 273–279.
25. V. Delplace, P. E. B. Nickerson, A. Ortin-Martinez, A. E. G. Baker, V. A. Wallace and M. S. Shoichet, Adv. Funct. Mater., 2020, 30, 1903978.
26. B. L. Oliveira, Z. Guo and G. J. L. Bernardes, Chem. Soc. Rev., 2017, 46, 4895–4950.
27. S. T. Koshy, R. M. Desai, P. Joly, J. Y. Li, R. K. Bagrodia, S. A. Lewin, N. S. Joshi and D. J. Mooney, Adv. Healthcare Mater., 2016, 5, 541–547.
28. H. S. Han, N. K. Devaraj, J. Lee, S. A. Hilderbrand, R. Weissleder and M. G. Bawendi, J. Am. Chem. Soc., 2010, 132, 7838–7839.
29. S. T. Koshy, D. K. Y. Zhang, J. M. Grolman, A. G. Stafford and D. J. Mooney, Acta Biomater., 2018, 65, 36–43.
30. N. Wathoni, T. Rusdiana, A. N. Hasanah, A. R. Pratama, M. Okajima, T. Kaneko, A. F. A. Mohammed, B. W. Putera and H. Arima, J. Adv. Pharm. Technol. Res., 2020, 11, 74–80.
31. S. Barrientos, O. Stojadinovic, M. S. Golinko, H. Brem and M. Tomic-Canic, Wound Repair Regen., 2008, 16, 585–601.
32. H. Kim, W. H. Kong, K. Y. Seong, D. K. Sung, H. Jeong, J. K. Kim, S. Y. Yang and S. K. Hahn, Biomacromolecules, 2016, 17, 3694–3705.
33. L. M. C. Aguilar, S. M. Silva and S. E. Moulton, J. Controlled Release, 2019, 306, 40–58.
34. D. L. Alge, M. A. Azagarsamy, D. F. Donohue and K. S. Anseth, Biomacromolecules, 2013, 14, 949–953.
35. A. Isaac, F. Jivan, S. J. Xin, J. Hardin, X. H. Luan, M. Pandya, T. G. H. Diekwisch and D. L. Alge, ACS Biomater. Sci. Eng., 2019, 5, 6395–6404.
36. X. Y. Liu, M. M. Hao, Z. J. Chen, T. Zhang, J. Huang, J. W. Dai and Z. J. Zhang, Biomaterials, 2021, 272, 120771.
37. C. Ghobril and M. W. Grinstaff, Chem. Soc. Rev., 2015, 44, 1820–1835.
38. Q. L. Loh and C. Choong, Tissue Eng., Part B, 2013, 19, 485–502.
39. L. Han, M. H. Wang, P. F. Li, D. L. Gan, L. W. Yan, J. L. Xu, K. F. Wang, L. M. Fang, C. W. Chan, H. P. Zhang, H. P. Yuan and X. Lu, ACS Appl. Mater. Interfaces, 2018, 10, 28015–28026.
40. R. S. H. Chang, J. C. W. Lee, S. Pedron, B. A. C. Harley and S. A. Rogers, Biomacromolecules, 2019, 20, 2198–2206.
41. L. Yan, K. Han, B. Pang, H. Jin, X. X. Zhao, X. G. Xu, C. M. Jiang, N. Cui, T. L. Lu and J. L. Shi, Chem. Eng. J., 2021, 414, 128836.
42. Y. Gao, J. J. Chen, X. Y. Han, Y. D. Pan, P. Y. Wang, T. J. Wang and T. Q. Lu, Adv. Funct. Mater., 2020, 30, 2003207.
43. M. A. Schubert, M. J. Wiggins, J. M. Anderson and A. Hiltner, J. Biomed. Mater. Res., 1997, 34, 519–530.
44. J. R. Martin, P. Patil, F. Yu, M. K. Gupta and C. L. Duvall, Biomaterials, 2020, 263, 120377.
45. S. Koutsopoulos, L. D. Unsworth, Y. Nagai and S. G. Zhang, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 4623–4628.
46. S. Dutta, P. Samanta and D. Dhara, Int. J. Biol. Macromol., 2016, 87, 92–100.
47. S. B. Nimse and D. Pal, RSC Adv., 2015, 5, 27986–28006.
48. A. C. Knall and C. Slugovc, Chem. Soc. Rev., 2013, 42, 5131–5142.
49. X. F. Bao, J. H. Zhao, J. Sun, M. Hu and X. R. Yang, ACS Nano, 2018, 12, 8882–8892.
50. K. Duval, H. Grover, L. H. Han, Y. Mou, A. F. Pegoraro, J. Fredberg and Z. Chen, Physiology, 2017, 32, 266–277.
51. M. W. Tibbitt and K. S. Anseth, Biotechnol. Bioeng., 2009, 103, 655–663.
52. K. W. Chen, Y. Y. F. Fen, Y. Zhang, L. Yu, X. X. Hao, F. Shao, Z. Z. Dou, C. F. An, Z. M. Zhuang, Y. H. Luo, Y. Wang, J. R. Wu, P. Ji, T. Chen and H. N. Wang, ACS Appl. Mater. Interfaces, 2019, 11, 36458–36468.
53. D. N. Kim, J. Park and W. G. Koh, J. Ind. Eng. Chem., 2009, 15, 124–128.
54. J. M. Goddard and J. H. Hotchkiss, Prog. Polym. Sci., 2007, 32, 698–725.
55. S. J. Liu, T. Jiang, R. Q. Guo, C. Li, C. F. Lu, G. C. Yang, J. Q. Nie, F. Y. Wang, X. F. Yang and Z. B. Chen, ACS Appl. Bio Mater., 2021, 4, 2769–2780.
56. D. H. Zhang, Q. Chen, C. Shi, M. Z. Chen, K. Q. Ma, J. L. Wan and R. H. Liu, Adv. Funct. Mater., 2021, 31, 2007226.
57. K. Knop, R. Hoogenboom, D. Fischer and U. S. Schubert, Angew. Chem., Int. Ed., 2010, 49, 6288–6308.
58. L. Zhang, Z. Q. Cao, T. Bai, L. Carr, J. R. Ella-Menye, C. Irvin, B. D. Ratner and S. Y. Jiang, Nat. Biotechnol., 2013, 31, 553–556.
59. N. Ojha, S. Roy, G. He, S. Biswas, M. Velayutham, S. Khanna, P. Kuppusamy, J. L. Zweier and C. K. Sen, Free Radicals Biol. Med., 2008, 44, 682–691.
60. J. M. Kim, J. E. Choo, K. N. Kim and Y. S. Kim, Clin., Cosmet. Invest. Dermatol., 2018, 11, 505–513.
61. T. A. Wilgus, S. Roy and J. C. McDaniel, Adv. Wound Care, 2013, 2, 379–388.
62. L. Mao, S. M. Hu, Y. H. Gao, L. Wang, W. W. Zhao, L. N. Fu, H. Y. Cheng, L. Xia, S. X. Xie, W. L. Ye, Z. J. Shi and G. Yang, Adv. Healthcare Mater., 2020, 9, 2000872.
63. G. S. Schultz, G. Ladwig and A. Wysocki, World Wide Wounds, 2005, 2005, 1–18.
64. J. Li, Y. P. Zhang and R. S. Kirsner, Microsc. Res. Tech., 2003, 60, 107–114.
65. J. Hardwicke, D. Schmaljohann, D. Boyce and D. Thomas, Surgeon, 2008, 6, 172–177.
66. M. Schafer and S. Werner, Pharmacol. Res., 2008, 58, 165–171.

---

Footnote

† Electronic supplementary information (ESI) available: The materials used in this study, the synthesis procedures of acetone-[bis-(2-amino-ethyl)-dithioacetal] and Nb-PFP, ¹H NMR spectra, rheological properties of PEG hydrogel, adhesive strength, DPPH radical scavenging ability of 4-arm-PEG-TK-Nb macromer, the cell viability of 2D cultured L929 cells in the hydrogels, H&E and Masson staining images. See DOI: 10.1039/d1bm01179k

---