A photothermal-response oxygen release platform based on a hydrogel for accelerating wound healing

During wound healing, oxygen availability and the anti-inflammatory microenvironment play an important role in the formation of new tissue. However, providing continuous and controllable oxygen around the injured tissue while inhibiting inflammation and realizing the synergistic effect of oxygen supply and anti-inflammation is still a major problem affecting the regeneration and repair of wound tissue. Inspired by skin wound pathology and the inflammatory microenvironment, a photothermal response-assisted strategy was developed in this study. We prepared a composite hydrogel system of polydopamine-hyaluronic acid (PDA-HA) hydrogel-loaded calcium peroxide-indocyanine green combined with lauric acid and manganese dioxide (CaO2-ICG@LA@MnO2) nanoparticles that showed excellent photothermal performance under near-infrared (NIR) irradiation and realized the on-off release of oxygen and reactive oxygen species (ROS). Controllable and sustainable oxygen release can promote the regeneration and repair of damaged tissue, and the generated ROS can effectively inhibit the outbreak of inflammation at the initial stage of wound healing. We believe that the system we have developed can be used in a new approach for treating chronic wounds. In this work, we prepared a composite hydrogel system of PDA-HA hydrogel-loaded CaO2-ICG@LA@MnO2 nanoparticles that showed excellent photothermal performance under NIR irradiation and realized the on-off release of oxygen and ROS. Controllable and sustainable oxygen release can promote the regeneration and repair of damaged tissue, and the generated ROS can effectively inhibit the outbreak of inflammation at the initial stage of wound healing. A full-thickness skin defect repair test was carried out in rats with the prepared hydrogel wound dressings and the results showed that hydrogel + NIR group had the most wound healing effective.


Introduction
Wound repair is one of the most complex biological processes in humans 1,2 . Many studies have shown that sufficient oxygen is essential for proper wound healing [3][4][5][6] . However, the environment around wounds is highly hypoxic during wound healing for the following reasons: (I) In addition to the normal consumption of O 2 during aerobic respiration to meet the basic energy requirements, O 2 is also consumed through the generation of new vessels and connective tissue, the production of reactive oxygen species (ROS), and the action of NADPH oxidase in phagocytes 3,5 . (II) Adequate O 2 supply may be prevented due to disruption of the microcirculation and contraction of the vessels in traumatized tissue 5 . (III) Wound dressings may prevent wound tissues from receiving oxygen from the air. Therefore, it is a considerable challenge to design materials that can provide prolonged and sufficient oxygen to hypoxic regions 7,8 .
Nanomaterials and nanotechnologies are expected to provide innovative platforms for tissue repair and regeneration [9][10][11][12] . Among a large number of oxygen-releasing materials, calcium peroxide (CaO 2 ), which decomposes in water to produce O 2 , is preferred due to its high oxygen-generation capacity 13,14 , while manganese dioxide (MnO 2 ) can improve the oxygenation rate as a catalyst. In particular, CaO 2 composite nanomaterials have been widely applied in tissue engineering [15][16][17][18][19] . However, the application of these materials has been greatly limited by the burst release of O 2 , which not only leads to oxidative damage to cells 8 but also significantly reduces the continuous supply of O 2 . Therefore, recent studies have focused on the design of oxygen-releasing materials for controlled O 2 production [16][17][18][19][20] . CaO 2 nanomaterials, because of their small size, can be easily coated, and the coatings can prevent exposure to an aqueous environment by providing a response to stimuli such as temperature, nearinfrared (NIR) irradiation or pH. Thus, controlled release of O 2 can be realized by regulating the coatings. Indocyanine green (ICG), which is responsive to NIR light 21,22 , has been used to coat CaO 2 nanoparticles to realize the controlled release of O 2 . Under NIR irradiation, ICG releases singlet oxygen ( 1 O 2 ), a type of ROS 16,23,24 . 1 O 2 is known to be highly reactive and can effectively damage DNA, proteins, and lipids 25,26 . Therefore, it can also have a positive effect on wound healing in the initial stage of inflammation. On the other hand, as a scaffold material, hydrogel is often used as a wound patch, which can adhere to wounds and release drugs, accelerate wound healing, and provide enough volume to encapsulate these oxygen-releasing materials [27][28][29][30][31] , and the structure is conductive to oxygen transport 32 .
Herein, we designed CaO 2 -ICG@LA@MnO 2 composite nanoparticles with a good NIR response based on well-dispersed CaO 2 nanoparticles. The obtained CaO 2 -ICG@LA@MnO 2 nanoparticles were further encapsulated in a PDA-HA hydrogel. Lauric acid (LA) causes swelling under NIR laser irradiation due to its low melting point (44°C) and the photothermal property of ICG. Then, O 2 is generated as CaO 2 is exposed to an aqueous environment. We achieved the controlled release of O 2 , and then a full-thickness skin defect repair test was carried out in rats with the prepared oxygenreleasing material. The in vivo experiments showed that the strategy effectively promoted wound healing by inducing a synergistic effect of O 2 supply and antiinflammation in the microenvironment through the NIR response (Scheme 1). Scheme 1 Schematic illustration of the preparation and application of the nanocomposite hydrogel. a Synthesis of the PDA-HA hydrogel. b Preparation of CaO 2 nanocomposite particles. c Applications of the nanocomposite hydrogel in wound healing.

Synthesis of CaO 2 nanoparticles
CaO 2 nanoparticles were synthesized at room temperature according to a previous report 13 . First, a 2 mol L −1 CaCl 2 aqueous solution was prepared as the original solution. Second, 1 mL of CaCl 2 solution was added to 60 mL of methanol, and 1 mL of hydrogen peroxide was mixed into the solution. After a few minutes, 4 mL of ammonia solution was dropped into the mixed solution to initiate the reaction. Finally, the product was separated by centrifugation, washed with ethanol three times at 5000 rpm for 5 min, and dried in a vacuum oven at~60°C for 12 h. Thus, the CaO 2 nanoparticles were successfully fabricated. The samples were measured by X-ray diffraction (XRD) using an X-ray diffractometer (D8 Focus Advance, Bruker) with Cu Kα radiation (λ = 0.15406 nm). Transmission electron microscopy (TEM) and high-resolution TEM (HR-TEM) images of the samples were obtained using a transmission electron microscope (JEM 2100 F, JEOL) at an accelerating voltage of 200 kV equipped with an energy-dispersive X-ray spectrometer (EDX, X-max 80 T, Oxford).

Fabrication of CaO 2 -ICG@LA nanoparticles
First, 25 mg of CaO 2 was mixed into 25 mL of methanol, and the mixed solution was stirred for 20 min after 30 min of ultrasonic treatment. Second, 5 mg of ICG was dispersed in the mixed solution and stirred for 20 min, and then 1 g of LA was added, and the solution was stirred for 6 h. Finally, the product was separated by centrifugation, washed with methanol twice at 10000 rpm for 10 min, and dried at room temperature. Thus, CaO 2 -ICG@LA nanoparticles were successfully fabricated. The morphology of the CaO 2 -ICG@LA nanoparticles was observed with TEM.

Synthesis of CaO 2 -ICG@LA@MnO 2 nanoparticles
MnO 2 nanoparticles (5 mg) were mixed into 20 mL of ethanol with ultrasonic treatment for 60 min, and then 10 mg of CaO 2 -ICG@LA nanoparticles was added. A mixed solution of CaO 2 -ICG@LA@MnO 2 was obtained, and CaO 2 -ICG@LA@MnO 2 nanoparticles were prepared by drying the mixed solution at room temperature. The morphology and element composition of the samples were analyzed by TEM and EDX.

Measurement of O 2 concentration
While vigorously stirring at room temperature, 3 mL of CaO 2 , 3mL of CaO 2 -ICG@LA, or 3 mL of CaO 2 -ICG@LA@MnO 2 methanol solution (0.5 mg mL −1 ) was added to 27 mL of phosphate-buffered saline (PBS, 10 mM, pH 7.4). Then, the O 2 concentration in the solution was recorded in real time with a dissolved oxygen meter (JPSJ-605F, Shanghai Leici Scientific Instrument Co., Ltd.).

Fabrication of the hydrogels
DA and HA were used to fabricate the biomimetic PDA-HA hydrogel. Briefly, 0.125 g of DA was dissolved in 2.5 mL of PBS. Minutes later, 125 μL aqueous solution of AP (0.02 g mL −1 ) was added to the solution. Then, 0.2 g HA was added to the mixture with vigorous stirring for~5 min. Finally, the PDA-HA hydrogel was fabricated. Hereby, the nanocomposite (NC) hydrogel was prepared by replacing 2.5 mL of PBS with 2.0 mL of PBS and 0.5 mL of CaO 2 -ICG@LA@MnO 2 methanol solution (1.0 mg mL −1 ). Here, the CaO 2 -ICG@LA@MnO 2 methanol solution was prepared by mixing 20 mg CaO 2 -ICG@LA@MnO 2 and 5 mg ICG into 20 mL of methanol. The microstructures of the obtained DA-HA hydrogel were investigated by scanning electron microscopy (SEM, Quanta 200, FEI). The freeze-dried samples were coated with gold before observation. The accelerating voltage was 25 kV.

Photothermal performance of the hydrogel
To evaluate the NIR photothermal performance of the PDA-HA hydrogel, the as-prepared hydrogel with a mass of 1.0 g was cut into cubes (15 mm × 15 mm × 4 mm) and then exposed to an 808 nm NIR laser (MW-GX-808/ 1~5000 mW, Changchun Laser Optoelectronics Technology Co., Ltd.). The hydrogel temperature was monitored as a function of irradiation time.

Measurement of 1 O 2 generation
SOSG is a highly selective reagent for detecting 1 O 2 . In the presence of 1 O 2 , it emits a green fluorescence similar to fluorescein. To observe this, first, SOSG was diluted to 500 μM using methanol as the solvate, and 45 μL of SOSG solution was added to the NC hydrogel. Then, 1.0 g of NC hydrogel and 2.0 mL of PBS solution were added to a sample cell, and the photoluminescence (PL) intensity of the sample was monitored under NIR (808 nm) laser irradiation. PL spectra were obtained by a PL spectrometer (FLS920P, Edinburgh) using a 450 W Xenon lamp as an excitation source.

Wound healing in a full-thickness skin defect model
A rat full-thickness wound model was used to evaluate the effect of the CaO 2 -ICG@LA@MnO 2 composite hydrogel on wound healing. Rats were randomly divided into 3 groups with 5 rats in each group, which included Group I (control), Group II (NC hydrogel), and Group III (NC hydrogel + NIR). Their backs were shaved, and a rounded full-thickness cutaneous wound was created with a diameter of 8 mm. In Group I, the wounds were treated with 200 μL PBS only. In Groups II and III, the wounds were covered with the NC hydrogel. In Group III, the wounds were irradiated by an NIR laser for 3 min once a day. The recovery of the wounds was evaluated on Days 0, 3, 5, 10, and 14.

Histological and immunofluorescence analysis
The wound tissues were collected on Day 14. The specimens were fixed with 4% paraformaldehyde, embedded in paraffin, and cut into 5 μm thick sections. The histological sections were stained with hematoxylin and eosin (H&E) and Masson trichrome stains after deparaffinization and rehydration. In addition, the main organs (heart, liver, spleen, lung, and kidney) were collected for histological analysis by H&E staining. For immunofluorescence staining, tissue sections were deparaffinized, rehydrated, and subjected to thermally induced antigen retrieval in citrate buffer solution (10 mM, pH 6.0) at 98°C for 10 min. Then, 10% normal goat serum was employed to block nonspecific binding for 70 min. TNF-α and TGF-β primary antibodies and corresponding secondary antibodies were used to evaluate the phenotype of macrophages in wound tissue. The cell nuclei were counterstained with DAPI for 10 min. The H&E-or Masson-stained sections were observed by an optical microscope. Immunofluorescence images were obtained by a fluorescence microscope (Olympus, Japan) and analyzed by ImageJ software.

Live subject statement
Eight-week-old rats were collected from Wenzhou Medical University. Animal experiments were performed according to the protocols approved by the Animal Ethics Committee of Wenzhou Medical University (No. SYXK-2021-0020).

Results
The preparation of CaO 2 -ICG@LA@MnO 2 nanoparticles was based on the as-synthesized CaO 2 nanomaterials. The CaO 2 nanoparticles were synthesized according to a previous report 13 , as described in the experimental section. A TEM image of the CaO 2 nanoparticles is shown in Fig. 1a. Clearly, the well-dispersed CaO 2 nanoparticles exhibit irregular morphology, and the particle sizes are mainly in the range of 50-150 nm. As shown in Fig. 1b, the typical HR-TEM image of CaO 2 nanoparticles clearly shows the (002) and (110) planes with corresponding lattice spacings of 0.295 and 0.252 nm, respectively. Figure 1d shows the XRD images of the as-prepared CaO 2 nanoparticles. All diffraction peaks of the samples were readily indexed to the standard card of tetragonal CaO 2 (PDF#03-0865). The results indicate that the prepared CaO 2 nanoparticles have a tetragonal structure with cell constants of a = 0.354 nm, b = 0.354 nm and c = 0.591 nm. Subsequently, CaO 2 -ICG@LA nanoparticles were synthesized, and their morphologies were observed by TEM. As shown in Fig.  1c, CaO 2 -ICG@LA nanoparticles have a core-shell structure, in which CaO 2 nanoparticles are wrapped by LA. A certain amount of ICG exists on the surface of CaO 2 and inside of LA. Next, CaO 2 -ICG@LA@MnO 2 nanoparticles were obtained by the simple method described in the experimental section. The α-MnO 2 nanoparticles used were purchased from Jiangsu Xianfeng Nanomaterial Technology Co., Ltd., and their TEM image is shown in Fig. 1e. The MnO 2 nanoparticles have a texture like the fluffy surface of a towel, and larger nanoparticles are~100 nm in size. The HR-TEM image (Fig. 1f) shows that the interplanar distance of the nanoparticles is~0.289 nm, which is consistent with previous reports 33,34 . TEM images of the CaO 2 -ICG@LA@MnO 2 nanoparticles are shown in Fig. 1g. It is evident that there are CaO 2 -ICG@LA and α-MnO 2 structures in this material. The chemical element distributions of the CaO 2 -ICG@LA@MnO 2 nanoparticles were further studied by EDX element mapping analysis techniques, as shown in Fig. S1. Considering that the nanoparticles contain more elements, several typical elements were detected, such as Ca, Mn, O, S and N. As mentioned above, it is very important to build a controllable oxygen release system. As an oxygen release material, CaO 2 has attracted considerable attention due to its high oxygen generation capacity and the biocompatibility of calcium ions. In addition, MnO 2 is used as a catalyst in this reaction. The mechanism of oxygen generation is described by the following chemical equations.
In the aqueous environment, CaO 2 is hydrolyzed to H 2 O 2 and then to O 2 . By limiting the exposure of CaO 2 to water molecules, we can control oxygen generation. Here, due to the hydrophobicity and low melting point of LA, it was chosen to coat the CaO 2 nanoparticles. CaO 2 , CaO 2 -ICG@LA and CaO 2 -ICG@LA@MnO 2 nanoparticles were mixed into PBS solution, and the dissolved oxygen in the solution was measured, as shown in Fig. 1h. Due to the complete exposure to water molecules, CaO 2 nanoparticles generated more oxygen in the first 300 s than the CaO 2 -ICG@LA and CaO 2 -ICG@LA@MnO 2 nanoparticles. LAcoated CaO 2 -ICG@LA nanoparticles generate less oxygen, while CaO 2 -ICG@LA@MnO 2 nanoparticles release more oxygen under the catalysis of MnO 2 . Moreover, after continuous oxygen generation for 30 min, the release of oxygen from CaO 2 -ICG@LA@MnO 2 nanoparticles was further monitored in the range of 25-50°C, as shown in Fig. S2. Below 40°C, the amount of oxygen released increases with temperature, mainly due to dissolved oxygen in PBS. Above 40°C, oxygen production is significantly improved because of the melting of LA.
The physicochemical properties of the synthesized PDA-HA [35][36][37] and NC hydrogels were investigated. The gelation process is shown in Fig. 2a. DA was dissolved in PBS, and then a certain amount of AP aqueous solution was added as an oxidant to induce dopamine polymerization. The mixture turned a clear pale red after stirring for a few minutes. Then, HA was added to the mixture and stirred vigorously for 5 min to prepare the firebrick PDA-HA hydrogel (Fig. 2a). The hydrogel has a double-network structure and overcomes the problem of brittleness. In addition, the short gelation time is beneficial to the application of the PDA-HA hydrogel in wound healing. The PDA-HA hydrogel is flexible enough to construct any shape, as shown in Fig. 2b. It is well known that hydrogels with self-healing properties can greatly improve the lifespan of dressings, and the self-healing capability of the PDA-HA hydrogel was further evaluated. In Fig. 2c, a simple self-healing process of the hydrogel is shown. We cut two squares at the same position and patched two pieces of different hydrogels into a new square, one of which was stained with ICG. Self-healing was achieved 5 min after physical contact. Simultaneously, the hydrogel exhibited excellent bioadhesion performance on the skin, as shown in Fig. 2d. In addition, the microstructure of the hydrogel was observed by SEM. It can be seen from Fig. 2e that the hydrogel is porous with an average pore size of~42 μm.
Hydrogels are a fascinating platform for use as threedimensional (3D) scaffolds for tissue repair and delivery of therapeutic molecules and cells 38 . The physicochemical and biological properties of hydrogels can be enhanced by combining them with bioactive and functional nanomaterials. NC hydrogels with ICG photothermal properties were obtained by introducing ICG into PDA-HA hydrogels. It is well known that ICG is a photothermal agent that absorbs NIR light (~800 nm). In addition, it is an NIR clinical imaging agent that has been approved by the U.S. Food and Drug Administration 21,22 . Photothermal therapy is based on the photothermal effect and has been applied to clinical treatment [39][40][41] . Therefore, the photothermal properties of the NC hydrogel were investigated. Under irradiation with an 808 nm NIR laser with different power densities (0.3, 0.5 and 1.0 Wcm −2 ), it was found that the temperature of the hydrogel increases with increasing laser irradiation time and power density, as shown in Fig. 2f. The photothermal stability of the NC hydrogel under repeated NIR laser irradiation was further investigated, and the heating (laser on) and cooling (laser off) processes for 5 cycles are shown in Fig. 2g. The heating profiles are similar for all cycles, with no significant change during the cycles. ICG generates 1 O 2 under 808 nm laser irradiation 16,24 . With the in-depth study of 1 O 2 , an increasing number of important applications have been found, such as photodynamic therapy, sterilization, and blood pressure regulation 22,25,42 . Wound healing involves four overlapping stages, namely, hemostasis, inflammation, proliferation and remodeling, in which inflammation occurs throughout the whole process of wound healing. ICG can effectively inhibit the outbreak of inflammation at the initial stage of wound healing. SOSG is currently the most widely used 1 O 2 probe and is detected by monitoring its PL at 528 nm 43,44 . In general, with proper control, the intensity of the green fluorescence signal can be correlated with the 1 O 2 concentration. As shown in Fig. 2h, the increase in PL intensity confirms that the 1 O 2 concentration increases with time.
To investigate the promoting effect of the hydrogel on wound healing, a full-thickness skin defect repair experiment was conducted in rats. The rats were randomly divided into 3 groups with 5 rats in each group, which included Group I (control), Group II (NC hydrogel), and Group III (NC hydrogel + NIR). A rounded full-thickness cutaneous wound was created on their backs and monitored over time. The rats in in Group III had their wounds treated with NC hydrogel and were irradiated by an 808 nm laser (1.0 W cm −2 ), and photothermal images were recorded by an infrared thermal imaging camera from 0 to 3 min (Fig. 3a). On Day 10, the rats in Group III exhibited the best wound healing, in which the wounds were completely closed. Wounds were not fully closed in Groups I and II (Fig. 3b). Moreover, the wound healing rate curve also demonstrated that the wound healing of rats in Group III was faster than that in the other groups (Fig. 3c). After 3 d of treatment, the wound area of Group III (NC hydrogel + NIR) was reduced to nearly 50% of its original size, while 89% remained for Group I (control group) and 65% for Group II (NC hydrogel). As shown in Fig. 3b, Group I required 7.1 d for half of the wound to be recovered, while Group II required 5.2 d. In addition, considerable redness and swelling were observed in Group II after 3 d of treatment. Inflammation is a basic pathological process dominated by the defense response when biological tissues are exposed to certain stimuli, including trauma, infection and other damaging factors. Local manifestations of inflammation include redness, swelling, fever, pain, dysfunction, and systemic reactions such as fever and changes in peripheral blood leukocyte counts. The redness and swelling in Group II were attributed to the stimulation of the hydrogel coating, while those symptoms were not observed in Group III due to the presence of 1 O 2 and photothermal effects. On Day 5, the redness and swelling of Group II gradually reduced. In addition, the wound healing rates of Group II and Group III became slow on the surface, which could be related to the internal healing of the wounds. On Day 10, based on external observations, the wounds of Group III were completely closed, with small scars surrounded by normal fur tissue. The wounds of Group II were small with white patches and no surrounding hair, which was attributed to hydrogel shedding. Notably, after 10 d, more than 99% of the total area healed in Group III, and~13% and 5% of wound tissue remained in Group I and Group II, respectively.
Moreover, H&E staining and Masson staining were performed to further analyze the wound repair process from a histological perspective. As shown in Fig. 4, fully closed wounds were formed in all 3 groups, but wound closure in Groups II and III was better than that in Group I. Masson staining mainly reflects the generation of collagen fibers, and a larger blue area represents more collagen fibers, indicating that the effect of repair is better. It was clear that better tissue structures were formed in Group III than in the other groups. All H&E staining (Fig. 4a) and Masson staining (Fig. 4b) showed that the treatment given to Group III (NC hydrogel + NIR irradiation) leads to better wound closure and has a better therapeutic effect on full-thickness skin incisions.
Immunofluorescence analysis was conducted according to the changes in proinflammatory and anti-inflammatory Fig. 5 Immunofluorescence analysis of wound healing process. Representative images of (a) M1 macrophage (TNF-α) and (b) M2 macrophage (TGF-β) immunofluorescence staining on Day 14 and statistical data of the relative intensity of (c) TNF-α and (d) TGF-β (n = 4) (**P < 0.01, ****P < 0.0001, ns: no significant difference). macrophages during wound recovery. As mentioned above, inflammation is present throughout the wound healing process. Therefore, changes in wound inflammation in rats were observed by tissue immunofluorescence staining. In the process of wound healing, there are two types of macrophages, proinflammatory (M1) and antiinflammatory (M2) macrophages 45,46 , whose markers change to a certain extent. Wound inflammation can be assessed indirectly by investigating corresponding markers in tissues. Hence, immunofluorescence staining of macrophages was performed on wound sections during the proliferation phase (Day 14) to clarify the role of the NC hydrogel in the conversion of M1 macrophages into the M2 phenotype. As shown in Fig. 5a, b, NC hydrogel + NIR treatment reduced the proportion of TNF-α M1-labeled positive cells, and the infiltration and distribution of TGF-β M2-labeled positive cells were enhanced compared with other groups. These results can be further observed in Fig. 5c, d, which show statistical data of the relative intensity of TNF-α (c) and TGF-β (d), respectively. M2 macrophages secrete anti-inflammatory mediators during wound healing, and our results further confirm that NC hydrogel + NIR can increase the presence of the macrophage phenotype relevant to tissue healing and regeneration.
To determine the biosafety and side effects of the NC hydrogel, H&E-stained slices of major organs such as the heart, liver, spleen, lung, and kidney were evaluated after 14 days of treatment (Fig. 6). The results showed no apparent histological abnormalities or lesions, confirming the biosafety of the NC hydrogel in living organisms 17,47,48 .

Conclusions
Well-dispersed CaO 2 nanoparticles were synthesized, and their morphology and structure were characterized by TEM and XRD. CaO 2 -ICG@LA@MnO 2 composite nanoparticles were constructed by using CaO 2 nanoparticles, and their oxygen release could be efficiently regulated by LA. Furthermore, NC hydrogels with photothermal properties were designed to achieve continuous release of oxygen upon 808 nm laser irradiation. The control group, hydrogel group, and hydrogel (laser irradiation) group were subjected to a full-thickness skin defect repair test in rats, and the hydrogel (laser irradiation) group had the most effective wound healing. This was further confirmed by H&E staining, Masson staining, and immunofluorescence analysis.