Development of tannin-bridged cerium oxide microcubes-chitosan cryogel as a multifunctional wound dressing.

Efficient resolution of oxidative stress, inflammation, and bacterial infections is crucial for wound healing. To surmount these problems, tannic acid (TA)-bridged CeO2 microcubes and chitosan (CS) (CS-TA@CeO2) cryogel was fabricated through hydrogen bonding interactions as a multifunctional wound dressing. Successful introduction and uniform incorporation TA@CeO2 microtubules enter the CS network. Thus-obtained CS-TA@CeO2 cryogels displayed a suitable porous structure and swelling rate. Cryogels has excellent tissue adhesion, blood cell coagulation and hemostasis, anti-infection, and cell recruitment functions. In addition, the cryogel also showed good antibacterial activity against gram-positive bacteria and gram-negative bacteria. Based on the in vivo study of the multifunctional mixed cryogels, it promotes fibroblasts' adhesion and proliferation and significantly improves cell proliferation and tissue remodelling in wound beds. Furthermore, the chronic wound healing process in infected full-thickness skin defect models showed that cryogels significantly enhanced angiogenesis, collagen deposition and granulation tissue formation by providing a large amount of antioxidant activity. Therefore, this multifunctional mixed cryogels has potential clinical application value.


Background
Skin injury, caused by burns, surgery, trauma or chronic disease (such as diabetes), has become one of the most critical clinical issues [1,2].Generally, the healing of wounded skin is a complex process including four continuous and overlapping phases: hemostasis, in ammation, proliferation and remodeling [3].Although in ammation is an essential wound healing phase, abnormally and persistent in ammation may cause the release of in ammatory cytokines and the elevation of reactive oxygen species (ROS) level, leading to uncontrolled tissue damage and delayed wound healing [4].Meanwhile, it has been documented that excessive ROS production can induce oxidative stress and strong in ammatory response, which makes wounds vulnerable and postpones wound healing process [5].Besides, accumulated excessive ROS in the wound site can severely affect angiogenesis and lead to endothelial dysfunction [6].Thus, the development of multifunctional wound dressings with antioxidant, anti-in ammatory, antibacterial, rapid hemostatic, and angiogenetic properties, which can modulate wound in ammation microenvironment by scavenging excessive ROS and reducing in ammatory factors, is urgently needed [7,8].
Chitosan (CS), a polysaccharide derived from naturally abundant chitin, with favorable biocompatibility, antibacterial activity and effective hemostatic performance, has received intensive attention in biomedical research area, especially as wound dressing material [9][10][11][12].Recently, compositing CS with nanoparticles, such as copper (Cu), zinc oxide (ZnO) and titanium dioxide (TiO 2 ) nanoparticles, has become an effective strategy for fabricating novel CS-based wound dressings with superior wound healing ability [13,14].Ceria (cerium oxide, CeO 2 ) nanoparticles have demonstrated e cient ROS scavenge and cell protecting ability against oxidative stress due to the reversible transition between the reduced (Ce 3+ ) and oxidized (Ce 4+ ) states of cerium ions on the particle surface [15,16].Recently, it has been reported that CeO 2 nanoparticles could accelerate wound healing by promoting cell proliferation and vascularization [17].However, pure CeO 2 nanoparticles have a dose toxicity, that is to say, high concentrations of CeO 2 nanoparticles would bring cytotoxicity [18,19].Besides, the fast release of CeO 2 nanoparticles may cause severe nanotoxicity.Excitingly, immobilizing inorganic particles into organic polymer networks has been proved a possible strategy to address the nanotoxicity issue [20][21][22][23].Tannic acid (TA), a natural polyphenol composing of a central glucose and 10 galloyl ester groups in total, has already been approved by the FDA for applications in food and medicine area [24,25].It is reported that TA can interact with various polymers, such as polyvinyl alcohol (PVA) [26], polypoly(ethylene glycol) (PEG) [27], gelatin [28], and so on, through hydrogen bonding.Thus the surface modi cation of CeO 2 nano-/micro-particles with TA can confer them hydrogen-bonding sites to interact with organic polymers [29,30], which could immobilize CeO 2 nano-/micro-particles in the polymer network to slow down the release of the particles and reduce their toxicity.Moreover, TA can render intrinsic anti-in ammatory and antioxidant properties to the biomaterials [26,27].Besides, the application of antibacterial TA in wound dressing might eliminate antibiotic treatment for wound healing, avoiding severe cytotoxicity and drug resistance [28,29].Moreover, TA can interact with blood-borne proteins, demonstrating excellent hemostatic property [31].Cryogels, prepared via freezing/low temperature cryogelation process, possess highly interconnected and porous structure, thus can absorb the excessive wound exudate and provide a moist environment for wound healing.Moreover, cryogels also can provide an environment similar to the extracellular matrix (ECM) of native tissues, which is conducive to cell migration and oxygen permeabilization to accelerate wound healing [32,33].Herein, TA-bridged CeO 2 microcubes and CS (CS-TA@CeO 2 ) cryogels were fabricated as a multifunctional wound dressing (Scheme 1).The physicochemical composition and structure of the CS-TA@CeO 2 cryogels were characterized, and the biocompatibility, cell proliferation, antibacterial, antioxidant and angiogenic performance, as well as wound closure e cacy of the crogels were investigated both in vitro and in vivo.The obtained multifunctional cryogels exhibited excellent hemostatic, anti-in ammatory, antibacterial, antioxidant and angiogenic properties, as well as good biocompatibility, thus could effectively promote wound healing.

Results And Discussion
Characterizations of CS-TA@CeO 2 cryogels The chemical structures of CS-TA@CeO 2 cryogels were investigated by FTIR analysis, the results were shown in Figures 1A and 1B.As can be seen, pure TA showed characteristic peaks of C=O (stretching vibration, 1714 cm -1 ) and phenolic hydroxyl groups on the benzene rings (3000-3500 cm -1 ) [34,35].
Compared to pure CeO 2 microcubes, it can be seen that TA@CeO 2 microcubes displayed characteristic peaks of TA at 1714 and 1194 cm -1 , indicating that TA adhesive layer was successfully adhered on the surface of CeO 2 microcubes.As shown in Figure 1B, CS showed characteristic absorption bands at 2878, 1590, 1416 and 1075 cm -1 , which assigned to stretching vibration of C-H, bending of N-H, bending of C-H and stretching of C-O bonds, respectively [36].With the addition of TA@CeO 2 microcubes, the characteristic peak at 3000-3500 cm -1 enhanced, indicating TA@CeO 2 microcubes successfully anchored in CS network.
The X-ray diffractometer (XRD) spectra of CeO 2 , CS and CS-TA@CeO 2 cryogels are presented in Figure 1B.
The thermal properties of CS-TA@CeO 2 cryogels were investigated by thermogravimetry analysis (TGA).
From the TGA curves shown in Figure 1C, two weight loss stages can be seen.The rst weight loss stage was around 100 °C, which was considered to be the release of the trapped water.The second signi cant weight loss stage was from 220 °C to 320 °C, which was attributed to the decomposition of CS polymer chain [37].
Morphology, porosity and swelling ratio evaluation of CeO 2 and CS-TA@CeO 2 cryogels The morphologies of CeO 2 , TA@CeO 2 and CS-TA@CeO 2 cryogels observed by SEM are shown in Figures 2A-J.It can be seen that both CeO 2 and TA@CeO 2 nanoparticles were uniformly dispersed without any aggregation, displaying a cubic structure with a length about 2.5 μm (Figures 2A and 1B).From Figures 2C-F, it can be seen that CS and CS-TA@CeO 2 cryogels displayed interconnected porous structures with a pore diameter of 15-30 μm.After the doping of TA@CeO 2 , the porous structure was kept for CS-TA@CeO 2 cryogels (Figures 2G-J).The quantitative porosity ratios were further determined by an alcohol immersion method.As shown in Figure 2K, the porosity of CS cryogels was 92.32%, while that of CS-TA@CeO 2 -5%, CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels were 86.15%, 84.34% and 80.16%, respectively.The decrease of porosities might be caused by the gravity of the loaded CeO 2 , which suppressed the pore cavities.The pores of CS-TA@CeO 2 cryogel also became more at comparing to that of CS cryogel.The relatively high porosity of CS-TA@CeO 2 cryogels is desirable for cell penetration, as well as hemostasis, oxygen exchange and exudate absorption, bene cial for wound healing [38,39].Furthermore, with the content of TA@CeO 2 increasing, more particles could be observed in the cryogels, indicating the successful incorporation of TA@CeO 2 in CS networks.
Swelling is an essential property of cryogels, enabled the cryogels to absorb excessive exudate of the wound, and provide a moist environment bene cial to wound healing [40][41][42].Also, the porous structure is conducive to cell migration and provides excellent oxygen permeability, which could accelerate wound healing [43,44].Therefore, the swelling ratios of cryogels were evaluated, and the results are shown in Figure 2L.It can be seen that all of the CS and CS-TA@CeO 2 cryogels exhibited swelling ratios all higher than 1400 wt%, which was attributed to the interconnected network and microporous structure of the cryogels.Interestingly, with the addition of TA@CeO 2 , the swelling ratios of the cryogels decreased.The swelling ratio of CS was 1751.64 wt%, and the swelling ratios of CS-TA@CeO 2 with 5, 10, and 15 wt% TA@CeO 2 were 1430.80,1436.02 and 1502.56 wt% respectively, which might be caused by the crosslinking effect raised by the intermolecular hydrogen bonding between CS and TA@CeO 2 [3,40].
Antioxidant activity of CS-TA@CeO 2 cryogels in vitro and in vivo ROS played a critical role in in ammation during wound healing, but overdose ROS would result in oxidative stress, leading to cell and tissue damage [5,45].Herein, the antioxidant activities of pure CS and CS-TA@CeO 2 cryogels were evaluated by testing the scavenging e ciency of DPPH free radicals.As shown in Figures 3A and 3B, CS-TA@CeO 2 cryogels could obviously decrease the absorption peak intensity of DPPH free radicals, and the scavenge percentages were 35.63% (CS-TA@CeO 2 -5%), 59.69% (CS-TA@CeO 2 -10%), and 82.90% (CS-TA@CeO 2 -15%), respectively.However, CS displayed almost no DPPH free radical scavenge ability with only 1.72% radical scavenge.With the increasing of CeO 2 content, the absorption peak intensity of DPPH free radicals decreased signi cantly, indicating that the excellent antioxidant activity of CS-TA@CeO 2 cryogels was endowed by CeO 2 .
Furthermore, the in vivo anti-oxidative activity was also evaluated.ROS is eliminated by the complex antioxidant defense systems in vivo, such as superoxide dismutase (SOD) and catalase (CAT), which are considered as the most important antioxidant enzymes.Thus, increasing the expression levels of SOD and CAT can reduce oxidative stress and promote wound healing [35].As shown in Figures 3C and 3D, for all groups, the SOD and CAT activities increased with healing time until the 14 th day, but decreased at the 21 nd day.Moreover, the groups treated with CS-TA@CeO 2 -15% cryogels showed a remarkable higher SOD and CAT activities comparing to other groups at day 7 and 14.These results further con rmed the antioxidative ability of CS-TA@CeO 2 cryogels by mitigating oxidative stress at the wound sites in vivo.

Antibacterial activity of CS-TA@CeO 2 cryogels
Bacterial infection has remained a challenging issue, which delays the healing process.Thus, wound dressings with antibacterial ability have a great potential in clinical applications [46][47][48].The antibacterial activity of CS-TA@CeO 2 cryogels was evaluated by measuring the diameter of bacterial inhibition zone rst [49].As shown in Figures 4A and 4B, no inhibition zone was observed for pure CS cryogels (a), while the average diameter of the inhibition zone was around 12.2 mm and 13.1 mm against S. Aureus, and around 11.5 mm and 12.5 mm against E. Coli, for CS-TA@CeO 2 -10% (b) and CS-TA@CeO 2 -15% (c) cryogels respectively.Besides, the continuous inhibitory effect of CS-TA@CeO 2 cryogels on the bacterial growth curve was also investigated and the results are shown in Figure 4C.It can be found that the growth pro les of both bacteria were signi cantly hampered, and the inhibitory effect enhanced with the increase of TA@CeO 2 content.These results con rmed an excellent antibacterial activity of TA and CeO 2 .

In vitro cell viability and proliferation
The biocompatibility and biosecurity are necessary for wound treatment in clinical applications.First, the cytotoxicity and cell proliferation of CeO 2 and TA@CeO 2 particles against broblasts (L929 cells) were evaluated by CCK-8 assay.Although some previous literatures have already proved that CeO 2 can promote cell proliferation and migration [16], the cytotoxicity of CeO 2 is dose-dependent.As shown in Figures 5A, it can be seen that, for CeO 2 treated cells, the cell viabilities re ecting by the optical density (OD) values were similar with that of untreated cells at the CeO 2 concentrations from 0.05 to 0.5 mg/mL.While, 1 mg/mL CeO 2 suspension expressed obvious cytotoxicity (Figure 5A), consistenting with previous report [16].However, for TA@CeO 2 at concentrations as high as 2 mg/mL, still no cytotoxicity against L929 was detected, and the cell viabilities of L929 treated with 2 mg/mL TA@CeO 2 at day 1, 3, and 5 were all signi cantly higher than that of control.These results fully indicated the introduction of TA onto CeO 2 could greatly reduce the cytotoxicity of high-dose CeO 2 , thus improved the superior limit value of CeO 2 when applied in wound healing.Therefore, we choose TA@CeO 2 to be loaded into CS cryogels, and the cell proliferation pro le of which is shown in Figure 5C.It can be seen that from Figure 5C, CS-TA@CeO 2 cryogels signi cantly promoted cell proliferation.In addition, Dead/Live staining was performed and the results are shown in Figure 5D. Figure 5D revealed that a majority of cells for all groups were stained green (live cells) and remained healthy at all time points, and almost no cells were stained red (dead cells).Besides, compared to CS cryogels group, the cell amount signi cantly increased for CS-TA@CeO 2 cryogels.Taken together, these results indicated that the prepared CS-TA@CeO 2 cryogels exhibited excellent biocompatibility, suitable for wound dressing.

Hemostatic characterization in vivo
As the rst step of wound healing process, hemostasis, which can prevent blood loss and avoid aggravation of injury, plays a key role in wound healing [50].In vivo hemostatic capacity of CS and CS-TA@CeO 2 cryogels was evaluated by testing the hemostatic time and the amount of blood loss in a rat tail-amputation model.As shown in Figure 6A, the amount of blood loss for CS-TA@CeO 2 cryogels was signi cantly lower than gauze and control groups (p < 0.01).From Figure 6B, it can be seen that the hemostatic time was reduced from higher than 200 s for the control and gauze groups, to 185.0 ± 4.0, 106.3 ± 4.7, and 83.0 ± 3.6 s for CS, CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels, respectively.In addition, the amount of blood loss was reduced from 0.633 ± 0.012 and 0.483 ± 0.025 g for the control and gauze groups to 0.45 ± 0.022, 0.243 ± 0.012, and 0.127 ± 0.015 g for CS, CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels, respectively (Figure 6C).These results demonstrated that CS-TA@CeO 2 cryogels have favorable hemostatic ability, further facilitating their application as wound dressing.

Wound closure
The wound healing performence of CS-TA@CeO 2 cryogels .From Figure 7A, it can be seen that the wound area of the four groups all gradually decreased over time.The wounds treated with CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels healed signi cantly faster compared to the control and CS groups, especially after day 7.At the 14 th day, the wounds treated with CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels were almost covered with epicortex, while the control and CS groups were not fully epithelialized.The wound healing ratios shown in Figure 8B indicated that the CS-TA@CeO 2 cryogels treated wounds contracted signi cantly faster than the control and CS groups during day 3 to day 21, indicating that the CS-TA@CeO 2 cryogels had a superior wound healing ability than the pure CS cryogels.

Histological examination
At pre-set timepoints, the histopathological analysis of the sliced wound tissues treated with CS and CS-TA@CeO 2 cryogels were evaluated by H & E and Masson's trichrome staining, and the results are shown in Figures 7C and 7D.At day 3, massive in ammatory cell in ltration was found in the wound sections for all groups.As shown in Figure 7C, CS-TA@CeO 2 signi cantly enhanced the formation of well-organized granulation tissue at the wound site at day 7.At day 14, in ammatory cells in ltration still could be observed and the wound was not fully reepithelialized in the control group, while more granulation tissues formed and the tissue became denser in CS-TA@CeO 2 groups.The Masson's trichrome staining images re ecting collagen deposition are shown in Figure 7D.Along with wound healing, the wounds underwent a dynamic metabolism of collagen and remodeling.More collagen deposition was found in CS-TA@CeO 2 groups than in other groups at day 7, 14 and 21.As expected, CS-TA@CeO 2 groups also displayed well arranged collagen deposition with dense and aligned bers.In the control and CS groups, the collagen deposition in dermis exhibited in partial incompact and hypogenetic collagen bers.Quantitative analysis (Figure 7E) demonstrated that the collagen densities in CS-TA@CeO 2 -10% (76.67 ± 2.31%) and CS-TA@CeO 2 -15% (76.67 ± 2.31%) groups were much higher than that in control (56.20 ± 2.69%) and CS (71.90± 3.40%) groups.High and aligned collagen deposition is essential for preferable wound healing e cacy.These results indicated that CS-TA@CeO 2 cryogels can promote collagen deposition thus further accelerate wound healing.

Immunohistochemistry staining and real-time quantitative PCR (RT-qPCR) of in ammatory and neovascularization factors
Overdose in ammatory response may lead to further tissue damage and delayed wound healing [41].Therefore, the two typical in ammatory factors, including TNF- (pro-in ammatory) and IL-10 (antiin ammatory), were estimated via immunohistochemistry staining and RT-PCR, to study the effect of CS-TA@CeO 2 cryogels in extenuating the in ammation during wound healing.As shown in Figures 8A and   8B, acute TNF- positive cells or pro-in ammatory cells in ltration, espacially on the wounded tissue sites, was found for the control and CS groups at day 3 and 7. While, the densities of TNF- positive cells signi cantly reduced for CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% groups.This phenomenon was also con rmed by the TNF- expression in mRNA level shown in the Figure 8G.The decay of in ammatory reponse for all groups along with time was also re ected in both the TNF- staining images (Figure 8A), the quantitive analysis results of TNF- positive cells (Figure 8B), as well as the TNF- expression in mRNA level (Figure 8G), but the in ammatory reponse decay rates of CS-TA@CeO 2 groups was much faster than that of the control and CS groups.On the contrary, From Figures 8C and 8D, it can be seen that at day 3 days, there was almost no expression of IL-10 in the control and CS groups, while a small amount of IL-10 expression was found in CS-TA@CeO 2 -10% and CS-TA@CeO 2-15% groups, which was also further con rmed by the RT-PCR results shown in Figure 8H.From day 3 to day 14, the expression of IL-10 gradually increased.Specially, at day 14, a large amount of IL-10 was expressed in CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% groups, while only a little amount of IL-10 expression could be found in the control and CS groups.All the results demonstrating that CS-TA@CeO 2 cryogels can effectively inhibit in ammatory responses and accelerate the wound healing process by promoting anti-in ammatory cytokines expression and inhibiting anti-in ammatory cytokines expression.
In addition, the neovascularization of the wounds was evaluated by immunostaining staining of CD31 (an endothelial cell marker) 3, 7, 14 and 21 days after treatment.As shown in Figures 8E and 8F, CS and CS-TA@CeO 2 cryogels treated groups displayed signi cantly improved CD31 expression, compared to the control group, indicating the formation of more capillaries with mature structures in CS and CS-TA@CeO 2 groups.This was further con rmed in mRNA level as shown in the RT-PCR results of CD31 (Figure 8I).A signi cant increase of CD31 expression was observed in CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% groups, while the expression of CD31 was little in other groups.These results suggested that TA@CeO 2 loading can improve angiogenesis and promote wound healing.At day 3, 7 and 14, the expression of CD31 in mRNA level increased much faster for CS-TA@CeO 2 -15% group than other groups (Figure 8I).Interestingly, as shown in Figure 8I, at day 21, the expression of CD31 decreased, this might impyed a return of CD31 expression to normal levelsand the wound healing process almost completed.
Synthesis of CeO 2 and TA@CeO 2 microcubes CeO 2 microcubes was synthesized via a facile hydrothermal method.Brie y, 2.17 g Ce(NO 3 ) 3 •6H 2 O was dissolved in a mixture solution (~100 mL) containing 10 g CH 3 COONa, 10 mL CH 3 COOH and 60 mL water.After being dissolved completely, the solution was then transferred into a 100 mL Te on-lined stainless steel autoclave and heated to 150 °C for 24 hours.After cooling to room temperature, the product was collected, centrifuged and washed with deionized water and ethanol for several times, and freeze-dried for further characterizations.
For the synthesis of TA coated CeO 2 (TA@CeO 2 ) microcubes, 1 g TA power was dissolved in 10 mL deionized water with vigorous stirring.Then, 1 g CeO 2 microcubes was dispersed in TA solution and stirred at room temperature for 12 hours.Subsequently, the crude product was centrifuged and washed with deionized water and ethanol for several times.After being freeze-dried, puri ed TA@CeO 2 microcubes was obtained.
Synthesis of CS-TA@CeO 2 cryogels Chitosan (CS) solution was prepared by dissolving CS powder in a 2% (v/v) acetic acid aqueous solution.A certain amount of TA@CeO 2 (5, 10, and 15 wt% to CS) was dispersed in 1 mL deionized water and the mixture was sonicated for 20 minutes, until a homogenous suspension was obtained.After that, TA@CeO 2 solutions were gradually added to CS solution, and stirred magnetically for 2 hours at room temperature.Then, the CS or CS-TA@CeO 2 solution was added into the wells of 24-well plates with 1 mL/well, the plates were then frozed at -20 °C.After freeze-drying, the cryogels were further washed with 2 wt% NaOH solution and distilled water to remove excessive acid, and freeze-dried again.The CS, CS-TA@CeO 2 -5%, CS-TA@CeO 2 -10% and CS-TA@CeO 2 -15% cryogels were obtained.

Characterizations
The morphology of the relevant samples was observed by scanning electron microscopy (SEM, XL30 FESEM FEG, PHILIPS) at an acceleration voltage of 20 kV.
The crystalline phases of the relevant samples were examined by X-ray diffractometer (XRD, MiniFlex-600, Rigaku Corporation, Japan) using Cu-Kα radiation at 40 kV voltage and 40 mA current.And the spectra were recorded at a 2θ from 10° to 90° with a scan rate of 10 °/min.Thermogravimetry analysis (TGA) was carried out with a thermogravimetric analyzer (209F3-ASC, Germany, Netzsch).The relevant samples were heated from room temperature to 800 °C at a heating rate of 10 °C/min under N 2 atmosphere.

Porosity and swelling ratio tests
The porosities of CS and CS-TA@CeO 2 cryogels were determined using an alcohol immersion method described in previous literature [51].Typically, CS and CS-TA@CeO 2 cryogels were weighted (W 1 ) and immersed in ethanol until saturated (the total weight was recorded as W 2 ).The porosity was calculated using the following equation: where V is the volume of the sample and ρ is the density of ethanol.Three parallel experiments were carried out for each specimen and the results were averaged.
The swelling performance of CS and CS-TA@CeO 2 cryogels was evaluated by gravimetric method [11].Brie y, the samples were weighted (W 1 ) and placed in deionized water at 37 °C for 2 hours.After that, the samples were blotted with a lter paper to absorb surface excess deionized water and weighted (W 2 ).
Three parallel experiments were carried out and the swelling ratio was calculated using the following equation:

Antioxidant activity assay
The antioxidant e ciency of CS and CS-TA@CeO 2 cryogels was evaluated by the method of scavenging the stable 1, 1-diphenyl-2-picrylhydrazyl (DPPH) free radical. [1]DPPH was dispersed in methanol with a concentration of 0.1 mM.Then, 5 mg CS-TA@CeO 2 cryogels was added into 3 mL DPPH solution and incubated in a dark place for 30 minutes.After that, wavelength scanning was performed using UV-vis spectrophotometer (SHIMADZU UV-2550) at pre-set time-points.The absorbance at 516 nm was measured and used to calculate the DPPH degradation using the following equation [52]: where A B , A S were the absorption of the blank (DPPH + methanol) and the absorption of the tested specimen (DPPH + methanol + specimen), respectively.

Antibacterial studies
The antibacterial performance of the CS and CS-TA@CeO 2 cryogels was evaluated using S. aureus (ATCC 25923) and E. coli (ATCC 25922) as Gram-positive and Gram-negative bacteria, respectively.Firstly, the bacteria suspensions (10 6 CFU mL -1 ) were spread on the solidi ed agar gel in Petri dish for inoculation.Then, the CS, CS-TA@CeO 2 -5% and CS-TA@CeO 2 -10% cryogels were placed on top of agar plates, respectively.After being cultured for 16 hours at 37 °C, the inhibition zone diameters were measured to assess the inhibition effect of the samples against the tested bacterium.
To further investigate the antibacterial ability of the CS-TA@CeO 2 cryogels, 50 mg sterilized samples were put into 10 mL of sterilized PBS (pH 7.4) solution containing 200 μL bacterial suspension (10 6 CFU mL -1 ), and the mixture was incubated at 37 °C under stirring at 150 rpm.At pre-set time points, the suspension was collected and diluted, and the optical density (OD) was recorded by a micro-reader at 570 nm.
The proliferation of L929 cells was evaluated using CCK-8.Brie y, CS and CS-TA@CeO 2 cryogels were sterilized by soaking in 75% ethanol followed by UV-irradiation for another 4 hours.Then, the samples were washed with sterile PBS (pH 7.4) and cell culture medium, and put into 24-well tissue culture polystyrene plates (TCPs, Costar, USA).After that, L929 cells were seeded in 24-well plates with a cell density of 1×10 4 cells/well, and incubated for 1, 3 and 5 days.At pre-determined time-points, the absorbance at 450 nm was recorded using a plate-reader after the cells were incubated with CCK-8 for 1 hour.Five parallel wells were set for each sample and the results were averaged.
The cell morphology and activity were also evaluated by Live/Dead assay according to the manufacturer's instruction.Typically, the cells were cultured for pre-set periods, and incubated with Live/Dead staining media for 10 minutes.Then, the stained cells were observed under an inversed uorescent microscope (Olympus CKX41, Tokyo, Japan) after washing with sterile PBS (pH 7.4).

Hemostatic evaluation experiments
The in vivo hemostatic ability of CS-TA@CeO 2 cryogels were evaluated using the rat tail amputation model [3,24].Typically, fteen male Sprague-Dawley (SD) rats with a weight of 230-250 g were randomly divided into ve groups equally, which were treated with gauze, CS, CS-TA@CeO 2 -5%, CS-TA@CeO 2 -10% respectively.And a negative control without any treatment was also set.Before the experiment, SD rats were anesthetized with sodium pentobarbital and the tails were cut at the 30% length of tail to the end, followed by exposing the tail in air for 15 s.After that, gauze, different treatments were applied to evaluate the hemostatic performance by recording the weight of blood loss and the hemostatic time.

Animal model
To evaluate the wound healing performance of CS-TA@CeO 2 cryogels, twelve male Sprague-Dawley (SD) rats (240-280 g) were used for the full-thickness skin wound healing experiments.All animal experiments were conducted in compliance with the Animal Experimental Committee of Southern Medical University (Approval No. SYXK2016-0167).After anesthetization, four wounds with a diameter of 1.2 cm were created on the back of each rat.After that, the wounds were covered with CS, CS-TA@CeO 2 -5% and CS-TA@CeO 2 -10% cryogels respectively, and PBS treatment group was also set as control.After pre-set time points, the macroscopic photographs of the treated wounds were recorded.The wound sizes were measured by an Image J software, and the wound closure ratios were also calculated using the following equation [53].
where A 0 and A t are the initial wound area and wound area at indicated time of "t", respectively.

Histology and immunohistochemistry evaluation
For histological examination, the harvested wound tissues at pre-determined time-points were xed with 4% paraformaldehyde and embedded in para n, and sliced into 4 μm to prepare the pathological slides.
After that, hematoxylin and eosin (H & E) and Masson's trichrome staining of the sliced tissue sections were conducted to analyze the microscopic wound regeneration in different phases.
Immunohistochemical staining of TNF- and IL-10 were performed to evaluate the in ammatory reaction in granulation tissues.Similarly, immunohistochemical staining of CD31 was also conducted followed a standard protocol to assess the angiogenesis.

Real-time quantitative PCR (RT-qPCR) analysis
The expression of in ammatory and angiogenic related genes (tumor necrosis factor- (TNF-α), interleukin-10 (IL-10) and platelet endothelial cell adhesion molecule-1 (CD31)) by wounded tissue after being treated with different samples was analyzed by real-time quantitative PCR (RT-qPCR) assay.The total RNA was isolated from granulated tissue for complementary DNA (cDNA) synthesis.Finally, RT-qPCR (iQSYBR Green Supermix Detection System, Bio-Rad) analysis was performed with a house keeping gene GAPDH on Light Cycler 480 SYBR green (Roche) and analyzed using Q-PCR instrument (Bio-Rad, USA).

Determination of antioxidant enzyme activities
The oxidative stress markers, including activities of catalase (CAT) and superoxide dismutase (SOD) in rats wound tissue homogenate were determined using the assay kits (SOD CAT, CUSABIO Co, Wuhan, China), according to the manufacturer's instruction.

Statistical analysis
Statistical analysis was performed using one-tailed Student's t-test by Statistical software (SPSS).The experimental results were expressed as mean ± standard deviation (SD).The differences were considered statistically signi cant when the p-values were < 0.05.* and ** represent p < 0.05 and p < 0.01, respectively.

Conclusions
Excessive reactive oxygen species (ROS), strong in ammatory response, and bacterial infections would make the wound sites vulnerable and greatly slow down wound healing process.To address these problems, a highly effective and well-known rare earth-based ROS scavenger, ceria (cerium oxide, CeO 2 ), was rstly modi ed with tannic acid (TA) to give TA coated CeO 2 (TA@CeO 2 ) microcubes, which was further introduced into chitosan (CS) to fabricate TA-bridged CeO 2 microcubes and CS cryogels (CS-TA@CeO 2 ) through hydrogen bonding interactions between TA and CS.Bene t from the highly interconnected and porous structure (porosity > 80%), the CS-TA@CeO 2 cryogels could absorb the excessive wound exudate and provide a moist environment for wound healing.The CS-TA@CeO 2 cryogels exhibited excellent antioxidant ability evidenced by scavenging more than 82.9% ROS in vitro and signi cantly increasing the antioxidant enzyme levels in vivo.The anti-in ammatory ability of the cryogels was also con rmed by the downregulated expression of the in ammatory cytokine, TNF-α and the upregulated expression of the anti-in ammatory cytokine, IL-10.The CS-TA@CeO 2 cryogels can promote the adhesion and proliferation of mouse broblasts (L929) cells.In addition, the CS-TA@CeO 2 cryogels showed excellent antibacterial, hemostatic and angiogenic activities.The in vivo full-thickness skin wound healing experiments on Sprague-Dawley (SD) rats demonstrated that the multifunctional CS-TA@CeO 2 cryogels can signi cantly accelerate wound healing through providing considerable antioxidant activity, promoting angiogenesis, and increasing collagen deposition.The developed multifunctional CS-TA@CeO 2 cryogels hold a great potential in clinical applications as wound dressing to promote wound healing, espacially when excessive ROS, in ammatory response, or bacterial infection was involved.The development strategy of CS-TA@CeO 2 cryogels could also be expanded to other organic/inorganic composite material designs for more broad biomedical applications such as wound healing and bone regeneration. Characterizations of TA, CeO2, TA@CeO2, CS and CS-TA@CeO2: (A, B) FTIR spectra, (C) XRD patterns, (D) thermogravimetry analysis (TGA) curves.