Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journal homepage: www.elsevier.com/locate/ijpharm

Construction of a biomimetic chemokine reservoir stimulates rapid in situwound repair and regeneration

Xue-Han Xua, Tie-Jun Yuana, Pei-Wu Yea, Mao-Ze Wanga, Hui-Jian Maa, Zhi-Hong Jiangb,Yong-Pin Zhangc,⁎, Li-Hua Penga,b,⁎

a Institute of Pharmaceutics, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, PR Chinab State Key Laboratory of Quality Research in Chinese Medicine, Macau University of Science and Technology, Macau, PR ChinacGuiyang University of Chinese Medicine, Guiyang, Guizhou, PR China

A R T I C L E I N F O

Keywords:Biomimetic hydrogelChemokineTopical deliveryMoisture microenvironmentWound repair

A B S T R A C T

Skin is the first protection of human body. It is always challenged by a range of external factors, resulting in thewounds of skin. Hydrogel, as a dressing with multiple advantages, causes increasing interests or the applicationsin wound treatment. However, the function and importance of micro-environment of wound region are fre-quently neglected. In this study, we successfully developed a chemokine loaded biomimetic hydrogel as afunctional reservoir to stimulate the rapid in situ recruitment of BMSCs for fast wound repair and regeneration.The biomimetic hydrogel was fabricated by using the Polyvinyl alcohol (PVA) combined with chitosan (CS) asthe hybrid materials. The fabricated hydrogel possesses many features such as the porous structure, highswelling rate and moisture retention property. More importantly, the incorporated chemokine could be releasedwith a sustained manner from the hydrogel and recruited the bone marrow mesenchymal stem cells (BMSCs)significantly both in vitro & in vivo. Moreover, the hydrogel was demonstrated to be highly biocompatible to theskin tissue without any side effect or irritation observed. Topical delivery of chemokine by the biomimetic PVA/CS hybrid material based hydrogel is demonstrated as a promising carrier to accelerate wound repair and re-generation without inducing scar formation and any other negative complications. The PVA/CS/SDF-1 hydrogelwas shown a novel therapeutic system for wound therapy.

1. Introduction

Skin is the largest organ in the human body and serves as a barrierto maintain homeostasis (Hudson, 2006; Madison, 2003; Proksch et al.,2008). It is always challenged by a range of external stress factors, re-sulting in frequent acute wounds (Baum and Arpey, 2005; Singer andClark, 1999). Tough considerable efforts have been focused on thedevelopment of novel therapeutic approaches for wound management,wound therapy remains a clinical challenge.

The healing of skin wound is a complex and multicellular process,requiring the collaborative efforts of several different tissues and celllineages. In this procession, besides the contribution of many cells, forexamples, immune cells, skin fibroblasts, bone marrow mesenchymalstem cells (BMSCs) have been demonstrated to play very important rolein wound healing process (Lu et al., 2018; Qi et al., 2018). After a tissueinjury, BMSCs are mobilized from the bone marrow into the pool ofcirculating cells. These cells then migrate to the injury site, where they

regulate the proliferation and migration of epithelial cells and dermalmesenchymal cells during the early inflammatory phase (Chen et al.,2008). BMSCs also act as the self renewing stem cells that differentiateinto fibroblast-shaped cells and the dermal keratinocytes phenotype;they also increase the number of regenerating appendage-like struc-tures, which serve as the precursors of permanent cutaneous appen-dages (Li et al., 2006). Governed by the Wnt/b-catenin pathway, theaccumulated BMSCs are able to differentiate into fibroblasts (Penget al., 2011) and they are stably present in dermal wounds throughoutthe repair process-especially tissue remodeling process-as well asgreatly contributing to the laying-down of collagen in the ECM(Mauney et al., 2010). BMSCs have long been recognized as a precursorto blood cell lineages (Koerner et al., 2006). However, they have re-cently been identified as giving rise to versatile cell phenotypes, e.g.fibrocytes (Lang et al., 2006) and smooth muscle cells (Jan-Marcuset al., 2010).

Modern knowledge about the way in which signals control wound

https://doi.org/10.1016/j.ijpharm.2019.118648Received 26 April 2019; Received in revised form 7 August 2019; Accepted 25 August 2019

⁎ Corresponding authors at: Guiyang University of Chinese Medicine, Huaxi University Town, Guiyang, Guizhou 550025, PR China (Y.-P. Zhang). Institute ofPharmaceutics, College of Pharmaceutical Sciences, Zhejiang University, 866# Yuhangtang Road, Hangzhou 310058, PR China (L.-H. Peng).

E-mail addresses: gzzhyp@126.com (Y.-P. Zhang), lhpeng@zju.edu.cn (L.-H. Peng).

International Journal of Pharmaceutics 570 (2019) 118648

Available online 26 August 20190378-5173/ © 2019 Elsevier B.V. All rights reserved.

T

cellular and molecular behavior has also promoted the transplantationof stem cells for skin regeneration with less scar formation (Peng et al.,2013a, 2012, 2011; Xie et al., 2016). However, major problems remainunsolved over the past years, such as (i) the lack of long-term in-tegration of the cellular sheets, (ii) the incomplete healing and frequentgeneration of scar tissue, (iii) the inadequate appendage contributionand (iv) overt rejection. Limited clinical successes indicate that theeffective delivery of stem cells is a major challenge which is yet to beovercame (Goldring et al., 2011; Trounson and McDonald, 2015).

In recent years, the topical drug delivery system that controls therelease of chemokine protein by biomaterials at the local sites to guidethe fate of cells to regenerate new tissues in situ has been demonstratedto be capable of avoiding the problem along with cells transplantation,which provides a novel strategy for wound treatment. In the in-siturecruitment, a sufficient concentration of a topically applied chemokineis always needed to be loaded into the vehicle to ensure an adequateconcentration gradient between the formulation and the skin, in orderto attain adequate release of the active protein into the local environ-ment. However, classical patches do not possess the capability to re-lease the entire amount of the drug, and a huge quantity of drugs arewasted once the patch is peeled off from the skin (Nauman et al., 2011).On the other hand, it has been demonstrated that keeping a woundmoist makes wound healing more rapidly with less pain and scarring(Field and Kerstein, 1994; Korting et al., 2011; Svensjö et al., 2000).Wound-healing cells critically require an aqueous environment to live,produce and receive growth factors. If a wound bed dries out, dehy-dration necrosis causes scabs formation. Scabs, which are dead cells, areinefficient in protecting the wound bed from further drying and bac-terial invasion (Junker et al., 2013; Vogt et al., 1995). Characterized bya wound environment, scrabs are toxic to wound-healing cells for thechronic wound (Bucalo et al., 1993; Wysocki et al., 1993). To stimulatethe wound-healing process, moist environment is required to enable thecells to produce the growth factors on which the wound healing de-pends. As a result, for the in-situ strategy, the topical system mustpossess multiple functions in both controlling the release of therapeuticprotein and modulating the microenvironment for tissue regeneration.

Hydrogel, as a novel dressing, is often used in wound repair, of-fering several advantages over the use of preformed dressings. Versatilenatural polymers like gelatin, alginate, synthetic polymers like PVA,PVP and carbomer have been used as dressing material for woundtreatment in the past years (Chen et al., 2017; Dong et al., 2017;Kamoun et al., 2017; Rosa et al., 2017). However, both natural andsynthetic polymers have limitations. Chitosan, also known as deacety-lated chitin, has appropriate viscosity and highly porous structurewhich make it a widely used wound dressing material (Anjum et al.,2016; Lee et al., 2015; Man et al., 2016; Murakami et al., 2010).However, chitosan hydrogel has been identified for its limitations suchas low tensile strength, residues and poor moisture property (Dang andYoksan, 2016; Hoare and Kohane, 2008; Miles et al., 2016). PVA is asynthetic polymer with easy film-forming property, due to its strongwater retention ability, biodegradable, non-carcinogenic, and bio-compatible characters (Jiang et al., 2011). With the combining of PVAand chitosan, the mechanical and physicochemical properties of thehybrid hydrogel might be obtained.

Therefore, in this study, chitosan was blended with PVA polymerand fabricated into a hybrid hydrogel with multiple features both toimprove the moisture retention and provide a reservoir to enhance thestem cells recruitment. Physiochemical properties and the protein re-leasing profile from the PVA/CS hydrogel were characterized and op-timized. Upon that, the in vitro & in vivo therapeutic effects of thechemokine loaded PVA/CS biomimetic hydrogel on the wound repairand regeneration were investigated. The schematic design of the PVA/CS hybrid hydrogel loaded with SDF-1 for accelerating the wound re-pair and regeneration was shown in Fig. 1.

2. Materials and methods

2.1. Materials

Chitosan (medium viscosity, 200–400mPa.s, Macklin), Polyvinylalcohol (PVA-124, 54–66mPa.s, Aladdin), 25 wt% glutaraldehyde so-lution, ethanol and glycerol were purchased from Sinopharm ChemicalReagent Co., Ltd. (Shanghai, China). All other chemicals were of re-agent grade and purchased from Sigma unless stated otherwise. SDF-1was purchased from PeproTech (USA); 5-dimethyl-2thiazolyl]-2,5-di-phenyl-2H-tetrazolium bromide (MTT) was purchased from Sigma; 4%buffered paraformaldehyde, Cy3 conjugated goat anti-rabbit IgG H&Land DAPI were purchased from Boster, INC. (Wuhan, China);Dulbecco’s modified Eagle’s medium (DMEM) and fetal bovine serum(FBS) were purchased from Gibco BRL (Gaithersburg, MD, U.S.A.);Purified mouse anti-rat CD31 and CD90 monoclonal antibody waspurchased from Abcam (Cambridge, MA); Masson’s trichrome stainingkit was purchased from Nanjing Keygen, INC. (Nanjing, China);Optimal cutting temperature compound (TISSUE-TEK 4583) was pur-chased from Sakura Finetek USA Inc (USA).

2.2. Preparation and characterization of PVA/CS hydrogel

The PVA/Chitosan (PVA/CS) hydrogel was fabricated by chemicalcross-linking method. Firstly, 1.5 g of the PVA was added to ultrapurewater. After being fully swelled for about 1 h, PVA was put in a waterbath at 85 °C to fully dissolve. Then glacial acetic acid was added topromote the dissolution of chitosan. After that, 0.5 g of chitosan wasmixed with ethanol and added gradually at 65 °C in a water bath todissolve completely. In addition, appropriate amount of glycerin wasalso added and stirred for 5–10mins. Then, 500 μl of glutaraldehyde(2 wt%) solution was added dropwise and the PVC/CS hydrogel wascrosslinked and obtained. The morphology of the PVA/CS hydrogel wasvisualized by using a scanning electron microscope (SEM) (Peng et al.,2015). Briefly, the hydrogel was gently rinsed with H2O and allowed toswell for several hours. Next, the hydrogel was freeze dried and cut toexpose the inner structure, coated with gold for conductance and im-aged with an SEM (SU-70, Hitachi, Japan).

2.3. Rheological investigation

The rheological measurement referred to the method of a previousstudy with some changes (Bhattacharya et al., 2012). Rheologicalmeasurement was performed at 37 °C using a rheometer (MCR 302,Anton Paar, USA). The geometry of choice was a 55mm parallel plate inwhich experiments were conducted at a gap of 1mm between theplates, requiring about 2ml of sample per run. The elastic/storagemodulus (G′) and viscous/loss modulus (G″) of the hydrogels weremeasured as a function of strain and frequency using dynamic oscilla-tory tests. First, a strain sweep test was performed at a constant fre-quency of 1 Hz, in order to determine the linear viscoelastic regime.Next, the viscosity was measured in the shear rate of 0.01–100/s.

2.4. Swelling property

The swelling ratios of the PVA/CS hydrogel were measured grav-imetrically (Shan et al., 2015). The PVA/CS hydrogel were allowed tofully swell in ultrapure water and phosphate-buffered saline (PBS) so-lution respectively. The weight of the hydrating samples was measuredat timed intervals after excess water was removed gently by absorbingthrough filter paper. The experiment was ended when the weight wasno longer increased. The max swelling ratio (SRmax) was calculatedaccording to the following formula:

=−

×M M

MSR 1 0

0100%max (1)

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

2

where M1 is the initial weight of the gel. M0 is the end weight of theswollen gel.

2.5. Moisture retention study

The prepared gel was cut into small pieces and weighed. Next, ac-cording to the maximum swelling condition obtained in the aboveswelling test, different proportions of water were added to obtain thegels with 100%, 80%, 60%, 40%, 20% and 0% of swelling degree, re-spectively. To further study the moisturize retention of hydrogel. 15%,10%, 5% and 1% swelling degree gel were used for the test. Thesesamples were placed in a 37 °C incubator. Hydrogels were weighed andrecorded at intervals until the weight no longer changed. The weightloss rate (R) is defined as follows:

=−

×W Wt

WR 0

0100% (2)

W0 is the weight of the gel in the initial swelling state, and Wt is theweight of the gel at a certain measurement time.

2.6. Moisture retention after topical applied

The skin was split from Sprague-Dawley rats (Shanghai SLACLaboratory Animal Co. Ltd., China.). All animal experimental proce-dures were performed in obedience to guidelines and protocols of theAnimal Experimental Ethics Committee of Zhejiang University. ThePVA/CS hydrogel was prepared by previous method. The PVA and CShydrogel were prepared using the same method described above exceptwithout the CS and PVA, respectively. The prepared gels were cut intosmall pieces. Next, different groups gels were applied to the skin. Thesesamples were placed in room temperature and exposure to room en-vironment. Hydrogels and skin were weighed and recorded at intervals.The weight ratio was calculated as the ratio of the weight of hydrogel ateach time interval to their initial weight, which is 1 g.

2.7. Thermo gravimetric analysis

The PVA/CS hydrogel were prepared using the same method de-scribed above. Thermogravimetric analysis (TGA) was performed undernitrogen atmosphere using a Mettler Toledo model TGA/DSC1 (MettlerToledo, Inc., Columbus, OH), with a heating rate of 10 °C/min.Measurements were analyzed using Mettler ToledoSTARev. 7.01 soft-ware.

2.8. In vitro drug release

The drug release study referred to the method of a previous studywith slightly changed (Chen et al., 2013). Briefly, a dialysis bag (mo-lecular cut-off of 8000–14,000) was prepared according to the protocolprovided by Sigma and then soaked in the release medium overnightprior to the study. Next, SDF-1 was added into the PVA/CS with stirringand dissolved at room temperature with the glutaraldehyde was addeddropwise. After the polymerization, 1 g of the gel was weighed, placedin dialysis bags, which were placed in the centrifugal tubes containing25ml of the release medium. Then, the tubes were placed on a shakingbox at 37 °C and 100 r.p.m. At the fixed timed points, 1 ml of the so-lution was taken and supplemented with the fresh PBS of the samevolume. The protein concentration in the collected sample was assayedby using a SDF-1 Elisa kit (Jiangsu Meimian Industrial Co., Ltd) withthe manufactures’ protocol. The accumulative release ratio was calcu-lated as the ratio of the cumulative mass of SDF-1 released at each timeinterval to their initial input amount in the hydrogel, which is 500 ng inthe gel.

2.9. Preparation of the BMSCs

Bone marrow derived mesenchymal stem cells (BMSCs) were ex-tracted from 3-week-old male Sprague-Dawley rats (Shanghai SLACLaboratory Animal Co. Ltd., China.). All animal experimental proce-dures were performed in obedience to guidelines and protocols of theAnimal Experimental Ethics Committee of Zhejiang University. Theisolation and culture of primary BMSCs were performed according tothe method previously described (Peng et al., 2016). Briefly, the ani-mals were killed and the BMSCs were flushed out of the tibias and fe-murs by using a syringe filled with Dulbecco’s modified Eagle’s mediumsupplemented with 10% fetal bovine serum, L-glutamine, penicillin(50 U/ml), and streptomycin (50 U/ml). After washing, centrifugedcells were resuspended in normal culture medium in a petri dish(Corning, USA) and cultures in a humidified 5% CO2 incubator at 37 °Cfor 72 h. The medium was renewed after every 2 days. When the cellgrowth density reached 80%–90%, the digestion was passaged, and thesecond to fifth generation BMSCs were used for subsequent experi-ments.

2.10. Biocompatibility evaluation

In this study, the biocompatibility of the hydrogel with the BMSCs

Fig. 1. Schematic illustration of the preparation of PVA/CS/SDF-1 hydrogel and applied as a multiple functional wound dressing with both controlled releasing effectof recruiting BMSCs and moisture retaining to stimulate a fast wound repair and regeneration.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

3

was investigated by testing the cytotoxicity of the hydrogel liquor toBMSCs. The cytotoxicity test was carried out as follows. (1) The hy-drogel material was sterilized with ultraviolet light for 30min and thencut into sheets with a size of 10mm×10mm×2mm. (2) The hy-drogel was extracted with BMSCs culture medium, and then dilutedwith different weight ratios of 5%, 25%, 50%, 100% for the furtherMTT test. (3) For the MTT (3-[4,5-dimethyl-thiazolyl-2]-2,5-diphenoltetrazolium bromide) assay(Peng et al., 2014), 200 μl liquid/wellmediums mentioned above were added to BMSCs seeded at a density of5× 103 cells/well in 96-well plates, those wells containing cells andnormal culture medium serving as negative control while wells onlycontaining medium serving as Blank well for the purpose of quantifyingthe background OD value. (4) After 48 h of culture, the liquid/culturemedium was replaced with culture medium containing MTT (0.5 mg/ml). (5) After 4 h, the supernatant was aspirated, and 200ml dimethylsulfoxide (DMSO; Sigma, USA) was added to each well. The plate wasmicro-oscillated for 30 s, and then the absorbance at 570 nm wasmeasured on a microplate reader (ELX800, BioTek Instruments Inc.,USA). (6) The viability of the cells cultured in the hydrogel leachingliquor was calculated as the percentage of the negative controls byusing the following equation:

=−

−

×OD s OD b

OD nc OD bViability ( ) ( )

( ) ( )100%

(3)

OD (s) is the OD value of the sample well; OD (b) is the OD value of theblank well; OD (nc) is the OD value of the negative-control well.

2.11. Cells migration assay

To study the capacity of the SDF-1 to induce BMSCs migration andconfirm the adequate concentration of SDF-1, a real-time cell analyzer(RTCA) DP instrument (ACEA Biosciences Inc. USA) was used to eval-uate cells migration (Bird and Kirstein, 2009). Briefly, migration assayswere carried out in a 16-well E-plate (E-Plate 16, ACEA bioscience,USA). Firstly, 165 μl of 25, 100, 250 ng/mL of the SDF-1 containedculture medium with 10% FBS were added in the lower chamber re-spectively, while the lower chamber without SDF-1 served as blankgroup. 35 μl of medium without FBS was added in the upper chamber.After it was balanced in 5% CO2 incubator at 37 °C for 1 h, BMSCs at adensity of 2× 104/ml in 80 μl of medium (0.5% FBS) were placed inthe upper chamber. The upper and lower chambers were assembled andplaced at room temperature for half an hour. Then the migration ofMSCs was recorded for 48 h by observing the change of micro-re-sistance of each well. All data were recorded by the supplied RTCAsoftware.

2.12. Animal study

Twenty-four SD male rats (body weight is 150–200 g) were pur-chased from Shanghai SLAC Laboratory Animal Co. Ltd., China. Allanimals were maintained under constant conditions (25 ± 1 °C), withfree access to standard diet and drinking water. All animal experimentalprocedures were performed in obedience to guidelines and protocols ofthe Animal Experimental Ethics Committee of Zhejiang University. Theanimal study referred to the method of a previous study with slightlychanged (Li et al., 2017a,b). The PVA/CS/SDF-1 hydrogel was pre-pared. Briefly, after preparing a mixed solution of PVA and CS, differentvolume of SDF-1 solution were added in mixed solution and dissolved atroom temperature. The cross-linking agent was added dropwise, stirred,and waited for the hydrogel polymerization. The animals were ran-domly divided into 4 groups: the blank control group, SDF-1 solutiongroup, PVA/CS hydrogel group, PVA/CS/SDF-1 hydrogel group. Sixrats in each group were anesthetized with an intra-peritoneal injectionof 3% phenobarbital sodium solution (0.1 ml/100 g). The hair at theirbacks was removed and their skin was washed with povidone-iodinesolution and wiped with sterile water. A square-shaped full-thickness

wound (1.5×1.5 cm2) was created on the back of each animal. The daywas identified as Day 0. On the next day, PBS, SDF-1 solution, PVA/CShydrogel, and the PVA/CS/SDF-1 hydrogel were topically applied tothe wound site. After the treatments, all groups were dressed withtransparent tegaderm (3M, St Paul, MN). The treatment was adminis-tered every two days within two weeks (day 1–14). And the progressionof wound healing was recorded by taking a picture every 2 days,measuring the wound area. And the weight of the rat was also recorded.The wound healing rate is defined as follows

=−

×WR WR n

WRWound healing rate (0) ( )

(0)100%

(4)

WR(0) is the wound area on day 0; WR(n) is the wound area on day n.

2.13. Histological analysis

The tissues of the wound regions were retrieved 15 days after skinwound treatment. Then tissues including the epidermis, dermis, andhypodermis were fixed in 4% para-formaldehyde for more than 24 h,dehydrated using a series of graded ethanol, embedded in paraffin, andsliced into 8 µm thick sections. Among them, 8 µm of sections werestained with the Masson’s trichrome staining kit according to themanufactures’ protocols. The stained skin sections were then examinedand photographed with Leica image analyzing system (Chen et al.,2012).

2.14. Molecular immunofluorescent staining

The tissues of the wound regions were retrieved on day 7 and 15after the treatment and embedded in an optimal cutting temperaturecompound, followed by freezing and slicing into 10 µm thick sections at−22 °C. To visualize the regenerated micro-vessel and the migration ofBMSCs in vivo, tissue sections were stained with antibodies againstCD31 (Zhang et al., 2016) and CD90 respectively. CD31 and CD90signals were visualized using the Cy3-conjugated secondary antibodiesrespectively. Nucleus were all stained by DAPI. Images of the sampleswere observed under a laser scanning confocal microscopy (IX81-FV1000, Olympus). To quantify the fluorescence intensity, images werephotographed from 6 random areas. All images were post-processedand quantified using Image J software.

2.15. Statistical analysis

Quantitative data were expressed as mean ± standard deviation(SD) values. For comparisons between two groups, means were com-pared using unpaired two-tailed Student’s t-tests. Statistically sig-nificant p values are indicated in figures and/or legends as ∗∗∗,P < 0.005; ∗∗, P < 0.01; ∗, P < 0.05.

3. Results

3.1. Characterization of PVA/CS hydrogel

The PVA/CS hydrogel was prepared by chemical cross-linkingmethod. After stirred homogeneously, CS and PVA were crosslinked bythe glutaraldehyde. We observed the gelation occurred within 10minafter the addition of the crosslinking agent. The morphology andstructure of the PVA/CS hydrogel were studied by SEM (Fig. 2A). Thehydrogel was shown for the highly porous and interconnected networkstructure. The pore size is about 1 μm, which makes it possible to loadthe drug with a high efficiency as well as indicate a certain gas andwater permeability to the gel, which will facilitate the exchange ofcellular nutrients and waste.

The rheology of the prepared hydrogels was shown in Fig. 2B, C.First, a strain sweep test was performed at a constant frequency of 1 Hz,in order to determine the linear viscoelastic region (Fig. 2B). It was

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

4

found that the storage module (G′) was higher than loss modulus (G″),indicated that the hydrogel has a slightly solid-like form. The linearviscoelastic region was under 10% strain, indicated the hydrogel couldmaintain its morphology in a certain degree of mechanical deformation.Furthermore, it was showed a high viscosity in Fig. 2C, indicated thestrong adhesiveness of hydrogel. Thus, it will have good skin followability in vivo and not easily peeled off after repeatedly being swayed.

3.2. Swelling and moisture retention studies

After the hydrogel was fully swollen, the SRmax was identified as3073.59 ± 199.84% (Fig. 3A), which means per unit weight of hy-drogel can absorb the liquid about 30 times weight of its own, in-dicating the strong hydrophilicity of the gel. We also examined themaximum swelling rate of the composite hydrogel in PBS. The swellingratio was 106.14 ± 1.82%, suggesting the efficacy of the hydrogel toabsorb the liquid in vivo. According to the maximum swelling ratioobtained by the previous experiment, the hydrogel with 100%, 80%,60%, 40%, 20% and 0% swelling degree gel were examined. It wasdemonstrated that all hydrogels with different swelling degrees reachthe balance state after around 10 h (Fig. 3B, C). There was only slightdifference among the hydrogels with more than 20% swelling degree.However, the equilibrium water loss of these gels was greatly differentfrom the initial hydrogel. It indicated that water loss might be influ-enced by the different swelling rate of the hydrogel. Whereas, there wasan obvious difference between the hydrogel with 1% swelling degreeand 5% degree. The 1% swelling degree hydrogel had about 80% finalwater loss compared to 90% of 5% swelling degree hydrogel. It in-dicated that the lower the hydrogel had swollen, the higher the water itcould contain. Meanwhile, in the wound region, there wasn’t muchfluid the hydrogel can absorb. The PVA/CS hydrogel has a pretty waterretention for the wound treatment in vivo. Overall, this study shows that

the prepared hydrogel have a very strong hydrophilic property andstrong water retention property. It can absorb the exudate in the woundsite and maintain a certain amount of water within a certain period oftime (5–10 h) after topical application, which can keep the moist mi-croenvironment and is advantageous for healing.

3.3. Moisture retention after topical applied and thermogravimetric analysis

The moisture retention was further investigated, hydrogels wereapplied to the skin. As shown in Fig. 3D, the weight ratio of PVA/CShydrogel group were higher than other groups from 0 to 24 h. There aresignificances between the PVA/CS hydrogel group and other groups,with a higher weight ratio, 45.54 ± 3.37%, compare to the PVA andCS hydrogel group with 33.49 ± 4.45% and 34.21 ± 1.57% remainedafter 24 h. Thus, the PVA/CS hydrogel could retain higher moistureafter applied to skin compared to the PVA and CS hydrogel. The me-chanism of PVA/CS hydrogel maintaining moisture was further in-vestigated, as shown in Fig. 3E, that the free water incorporated intohydrogel evaporated fast within 0–100 °C, and the weight was furtherdecreased from 100 to 200 °C, due to the evaporation of the boundwater and other volatile substances, while the remained mass was co-ordinated to the PVA and CS. The weight ratio decreased fast within 0-100 °C, due to the evaporation of free water incorporated into hydrogel,indicated that at normal temperature the moisture retention was owingto the release of free water absorbed in hydrogel when the environmentis dry.

3.4. Release of the SDF-1 from the PVA/CS hydrogel

From the Fig. 4, it was shown that, around 30% of SDF-1 was re-leased in around 2 h, which might be contributed to the prompt diffu-sion of SDF-1 attached on the surface of hydrogel. However, after that,

Fig. 2. (A) SEM image of the morphology and structure of the PVA/CS hydrogel with different magnification. Scale bars are 20 μm (left) or 2 μm (right), respectively.(B) Influence of strain on storage (G′) and loss modulus (G″) of hydrogel. The strain sweep test establishes the range of linear viscoelasticity for the hydrogels. (C) Theshear viscosity of hydrogel as function of shear rate.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

5

60% of SDF-1 was released with a constant manner in the 3rd–48thhour. The accumulative release percentages of the SDF-1 at the 48 hreach to 94.4 ± 3.7%. Because the PVA/CS hydrogel was prepared bya chemical crosslinking based method, the release manner of SDF-1 thatare physically incorporated in the hydrogel was thought to be con-trolled both by the diffusion of biomacromolecules overcoming thestatic effect of polymers and the gradual degradation of the hydrogel.The release of SDF-1 was declining after 48 h, might due to the long-term exposure to outside environment, which influenced the tertiarystructure of proteins, causing undetectable to the Elisa kit.

3.5. The cytotoxicity and recruitment of the PVA/CS hydrogel in BMSCs

From the Fig. 5A, it was demonstrated that, the leaching liquid of allthe tested PVA/CS hydrogel has very low cytotoxicity to the BMSCswithin 48 h, indicating the good biocompatibility of the hydrogel withthe BMSCs. From the Fig. 5B, the cell index increased with the dose ofSDF-1, but declined at the concentration of 250 ng/ml, which might beattributed to the limited interaction of the SDF-1 with the CXCR4 re-ceptors in the BMSCs. Compared with the blank control, the hydrogelwith 100 ng/ml of SDF-1 loaded expressed the strongest increase for themigration of BMSCs (Fig. 5B), which therefore was used in the sub-sequent experiments.

3.6. Wound closure in vivo

The hydrogel was then applied to the wound site in rats and theirinfluence in the wound healing process was investigated. Based on thereleasing profile, the PVA/CS hydrogel was applied to the wound siteevery two days for totally four times. The gloss appearance images ofthe wound sites in the various groups on days 3, 5, 7, 9, 11 and 15 areshown in Fig. 6B. All of the wounds expressed the gradual healingthroughout the experimental period. Wound closure rates determinedby the percentage of wound surface covered by healed skin were shownin Fig. 6C. It was shown that, compared with the blank control, thePVA/CS/SDF-1 hydrogel expressed a trend to accelerate the woundhealing throughout the whole experimental period, with the sig-nificance identified on day 7 and 11, demonstrating the excellenthealing effect of the PVA/CS/SDF-1 in vivo. With the passage of time,the difference in wound healing rate decreased, which might be con-tributed to that the application of PVA/CS hydrogel with SDF-1

Fig. 3. (A) The maximum swelling rate of hydrogel in water and PBS at pH 7.2, respectively. (B) The ratio of lose weight (%) of different groups hydrogel in a largegap of 0%, 20%, 40%, 60%, 80%, 100% swelling rate in water. (C) The ratio of lose weight (%) of different groups hydrogel in a small gap of 0%, 1%, 5%, 10% and15% swelling rate in water. (D) The ratio of weight (%) of different groups hydrogel applied to the skin in room temperature; (E) The thermogravimetric analysis ofthe PVA/CS hydrogel under nitrogen atmosphere with a heating rate of 10 °C/min.

Fig. 4. The accumulated release of SDF-1 from PVA/CS hydrogel within 72 h.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

6

incorporated played an important role in recruiting the BMSCs andpromotes the rapid healing in the early stage of the wound. It is worthmentioned that the wound healing rate of PVA/CS hydrogel treatedgroup was even higher than that of the SDF-1 solution treated group onday 3–11, reminding the advantages of using the PVA/CS as a biomi-metic carrier for the delivery SDF-1 that assist to accelerate the woundhealing by making a moist environment for the optimal wound repair &regeneration. The body weight of all the tested rats was shown inFig. 6D, After the treatment with different samples, the body weights ofthe rats were kept unchanged and increased slowly during the wholeexperimental period, indicating that the rats had good compliance witheach component and no significant systemic effects.

3.7. Histopathological investigation

Fig. 7 shows images of the histological conditions of the skins in thetested groups on day 15 post wounding. The healed skins exhibitedwide areas of a dense dermis and devoid of skin appendages, showingthe characteristics of scars in the blank control. For the SDF-1 solutionand the PVA/CS hydrogel treated groups, some skin appendages couldbe observed. However, the wide area of dense dermis indicating thescar feature was still obvious in these two tested groups. By contrast, inthe PVA/CS-SDF treated group, skins with many skin appendageswithout scar feature were identified. Moreover, the amount of the ap-pendages was significantly higher than that of the SDF-1 solutiontreated group, which suggested the prompt healing with strong re-generation efficiency in this group, which might be partly contributedto the sustained release effect of SDF-1 from the PVA/CS hydrogel thatincreases the stability of the protein and optimized the therapeutic ef-fect of SDF-1.

3.8. CD31 immunofluorescent identification

CD31 immunohistochemical staining is widely used for observingangiogenesis at the molecular level (Peng et al., 2013b). Through themolecular immunohistological staining for endothelial cellular proteinCD31, angiogenesis in the wounded skin was identified. It was de-monstrated that the PVA/CS/SDF-1 hydrogel can obviously enhancethe formation of the microvessels in contrast to the other tested groups.From the Fig. 8A, no newly formed blood vessels were detected in theblank control and PVA/CS hydrogel treated groups. However, micro-vessels with obvious vascular structure can be observed in the PVA/CS/SDF hydrogel and the SDF-1 treated group. Compared with the blankcontrol, both the microvessels density of the SDF-1 and PVA/CS/SDF

hydrogel treated groups were significant (Fig. 8B). Moreover, PVA/CS/SDF-1 expressed the trend to further enhance the formation of themicrovessels than that of free SDF-1 solution, which give the furtherevidence for the enhanced therapeutic effect of SDF-1 by using thePVA/CS to incorporate and control the release.

3.9. CD90 immunofluorescent analysis

To investigate the migration of BMSCs, CD90 immunofluorescentanalysis was used to identify the migration of BMSCs in vivo. From theFig. 9, it was found that the strong fluorescence indicating the largeamount of BMSCs have been observed in the healed skins of the PVA/CS/SDF-1 hydrogel treated group. By contrast, invisible fluorescenceindicating the few BMSC have been migrated to the wound site. Asshown in Fig. 9B, the PVA/CS/SDF-1 hydrogel group has the highestmean fluorescence intensity compared to the other tested groups. Theseresults give the further evidence for that the accelerated wound repairand regeneration in the PVA/CS/SDF-1 treated group might be con-tributed to the function of BMSCs and the micro-environment main-tained.

4. Discussion

Wound repair has conventionally relied on the transplantation ofstem/skin cells. However, cell transplantation has been hampered bycrucial issues like low cells survival rate, incomplete differentiation,limited outgrowth (Von Bahr et al., 2012). In addition, most of thetechniques above are time-consuming, invasive, frequently induce theinflammation and immune resistance. Developing a therapeutic systemthat can stimulate the fast-wound closure with the regeneration of skinappendages remains an ultimate however elusive goal for wound re-storation. In recent years, increasing studies demonstrated the bidir-ectional interaction between the stem cells and their microenviron-ment, which brought into sharp relief in the context of endogenousstem cell based therapeutic strategies (Abrajano et al., 2010; Miller,2006). However, all these in situ therapeutic systems share the chal-lenges to recruit the stem cells efficiently and the construction of anadvantageous microenvironment for the wound healing.

Some bio-scaffolds that mimic the natural microenvironment havebeen reported (Faraj et al., 2007; Lai et al., 2014; Yesmin et al., 2017).But significant obstacles including burst release of drugs, insufficientcellular adhesion support, and slow scaffold degradation rate remain tobe overcame before the full potential of bio-scaffold based stem celltherapies being realized (Kan and Lu, 2010; Li and Dai, 2018). In

Fig. 5. (A) The cell viability of BMSCs cultured in the leaching liquors of the PVA/CS hydrogel, diluted with different weight ratios of 5%, 20%, 50%, 100%,respectively for 48 h. (B) The cell index curves reflected the migration of cells induced by different concentration of SDF-1 solution, measured by a real-time cellanalyzer (RTCA) DP instrument. The concentrations of the leaching liquid were 0 ng/ml (blank control), 25 ng/mL, 100 ng/mL, 250 ng/mL, respectively.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

7

addition, it has been demonstrated that wound-healing cells criticallyrequire an aqueous environment to live as well as produce and receivegrowth factors which is often neglected by the researchers (Huang andLiu, 2012; Tang et al., 2017). To address these issues, we developed aPVA-chitosan hybrid biodegradable hydrogel that combines the ad-vantages both of the natural and synthetic biomaterials with capacity to

control the delivery of chemokine with a sustained manner and keepthe surrounding microenvironment in moisture. Furthermore, it showna degrade speed matching to the healing process. The constructed PVA/CS-SDF-1 hydrogel has demonstrated a highly porosity and inter-connected networks, which makes it an efficient carrier to load andcontrol the release of the biomacromolecules. The porous structure also

Fig. 6. (A) Graphical process of the experiment. (B) Representative skin wound photographs of different groups from Day 1 to Day 15, the interval is 2 days; (C) Thequantitative evaluation and statistical analysis of wound healing percentage from Day 3 to Day 13, the interval is 2 days. The PVA/CS/SDF-1 hydrogel significantlycontributed to the wound healing compared with the Blank group. (D) The body weight of different groups from Day 0 to Day 15.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

8

ensures a certain water permeability to the gel, and facilitates the ex-change of cellular nutrients and waste. As a result, PVA/CS hydrogelshowed the excellent swelling ratio with the SRmax was3073.59 ± 199.84% in water and 106.14 ± 1.82% in PBS, respec-tively. 2 interfaces were created after gel applying, the bottom oneinterfaces the gel and wound site while the upper one interfaces the geland the air. It is sufficient for the PVA/CS hydrogel to absorb the fluidin wound site where there are only few fluids with SRmax106.14 ± 1.82%. And the lower swelling rate of hydrogel was alsodemonstrated to have a higher moisture retention, as shown in Fig. 3C,the 1% swelling degree hydrogel has about 80% final water loss com-pared to 90% of 5% swelling degree hydrogel. Meanwhile, the PVA/CShydrogel not only absorb the fluid in wound site, but also absorb themoisture from the air with higher SRmax in DI water in the upper in-terface.

Based on this, we further evaluated the moisture retention property

of the tested hydrogel composed by the single or blended materials afterapplied to skins. As shown in Fig. 3D, PVA/CS shows the highest waterretention capacity with 45.54 ± 3.37%, which is higher than those ofPVA or CS hydrogel with 1.36 and 1.33 folds, respectively, with sig-nificances indicating the PVA/CS hydrogel could retain higher moistureafter applied to skin. The mechanism of PVA/CS hydrogel maintainingmoisture was further investigated. As shown in Fig. 3E, the free waterincorporated into hydrogel evaporated fast within 0–100 °C, and theweight was further decreased from 100 to 200 °C, due to the evapora-tion of the bound water and other volatile substances, while the re-mained mass was coordinated to the PVA and CS. The featured propertyof PVA/CS makes it highly efficient in maintain the local environmentin moisture to promote the healing with less scar formation and pain(Field and Kerstein, 1994; Li et al., 2017a,b; Lu et al., 2017)

These results demonstrated the features of the PVA/CS in not onlydelivering the SDF-1 with a controlled manner for the enhanced BMSCs

Fig. 7. Masson’s trichrome stained sections of the healed skins of day 15 after treatment. All images were taken under the same magnification. The scale bar is200 μm.

Fig. 8. Immunofluorescent staining for CD31-posi-tive microvessels at the wound healing region 15dpost-wounding. Red fluorescence indicates CD31positive expression, and blue fluorescence indicatesthe DAPI stained in the cells nucleus. Scale bars is100 µm. Micro vessels wers indicated by the whitearrow. (B) Histogram showing the quantification ofCD31-positive microvessels density. All images weretaken under the same magnification, n= 5.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

9

recruitment, but also is capable in retaining the water and keeping thewound microenvironment in moisture that prevents the scar formationand improve the healing quality, as well as relieving the pain inducedby wound dryness. The SDF-1 loaded PVA/CS hydrogel provide a newoption for the wound treatment with the potential to stimulate a rapidwound repair and regeneration.

Meanwhile, the rheological measurements showed that the linearviscoelastic region was under 10% strain, indicated the hydrogel couldmaintain its morphology in a certain degree of mechanical deformation.Furthermore, it was showed a high viscosity in Fig. 2C, indicated thestrong adhesiveness of hydrogel, which can help to maintain its mor-phology in a certain degree of strain and adhesion to skin. All thesefeatures of ensures the good elasticity and plasticity of the PVA/CShydrogel for the application to the skin wound treatment. Moreover,the further loading of the SDF-1 in the PVA/CS hydrogel contributesimportantly to the wound repair and regeneration. The incorporatedchemokine released with a controlled manner within the 48 h avoidsthe long exposure of SDF-1 to the outside environment, reducing thecleavage and optimized its function. Mechanistically, besides thebinding of SDF-1 to the CXCR4 receptor in the BMSCs that recruits theBMSCs vigorously, SDF-1 can also binds to the immune cells, such asneutrophils and macrophages deriving from mobilized monocytes(Bollag and Hill, 2013). Although the amount of immune cells are notso much, however, they can remain for several weeks at the wound site(Larouche et al., 2018) and play the roles of removing the cellulardebris, damaged matrix, microbes and secreting additional chemokines,growth factors, and matrix digesting enzymes, stimulating the macro-phages shift phenotype from a pro-inflammatory into an anti-in-flammatory one and subsequently promote tissue regeneration(Kunkelet al., 2017; Mahdavian et al., 2011; Sa et al., 2016). From the in vivoinvestigation, it was shown that, compared with the other testedgroups, in which, almost no skin appendage can be observed and thecollagens were deposited with random and loose manner, the healedskin in the PVA/CS/SDF-1 treated group were regenerated promptlyalong with many skin appendages regenerated. These in vitro & in vivoresults together demonstrated the excellent efficacy of the PVA/CS as abiomimetic carrier with multiple functions for delivering the ther-apeutic protein and regulating the wound microenvironment.

5. Conclusion

PVA/CS was demonstrated as a novel biomimetic carrier with the

multiple functions to deliver the therapeutic protein with controlledmanner and keep the wound site as a moist microenvironment for fastwound repair along with the prompt regeneration of skin appendages.PVA/CS/SDF-1 was shown promising therapeutics with strong efficacyto stimulate the wound healing with skin appendages regeneration.

Declaration of Competing Interest

The authors declare that they have no known competing financialinterests or personal relationships that could have appeared to influ-ence the work reported in this paper.

Acknowledgments

The study was supported by National key research and developmentprogram of China (2018YFC1105404), National Natural ScienceFoundation of China (81473145), Open project of State Key Laboratoryof Quality Research in Chinese Medicine (Macau University of Scienceand Technology, MUST-SKL-2016-11) funded by the Macao Science andTechnology Development Fund, Macau Special Administrative Region,China.

References

Abrajano, J.J., Qureshi, I.A., Gokhan, S., Molero, A.E., Zheng, D., Bergman, A., Mehler,M.F., 2010. Corepressor for element-1–silencing transcription factor preferentiallymediates gene networks underlying neural stem cell fate decisions. Proc. Natl. Acad.Sci. 107https://doi.org/10.1073/pnas.0906917107. 16685 LP – 16690.

Anjum, S., Arora, A., Alam, M.S., Gupta, B., 2016. Development of antimicrobial and scarpreventive chitosan hydrogel wound dressings. Int. J. Pharm. 508, 92–101. https://doi.org/10.1016/j.ijpharm.2016.05.013.

Baum, C.L., Arpey, C.J., 2005. Normal cutaneous wound healing: clinical correlation withcellular and molecular events. Dermatologic Surg. 31, 674–686. https://doi.org/10.1097/00042728-200506000-00011.

Bhattacharya, M., Malinen, M.M., Lauren, P., Lou, Y.R., Kuisma, S.W., Kanninen, L., Lille,M., Corlu, A., Guguen-Guillouzo, C., Ikkala, O., Laukkanen, A., Urtti, A., Yliperttula,M., 2012. Nanofibrillar cellulose hydrogel promotes three-dimensional liver cellculture. J. Control. Release 164, 291–298. https://doi.org/10.1016/j.jconrel.2012.06.039.

Bird, C., Kirstein, S., 2009. Real-time, label-free monitoring of cellular invasion and mi-gration with the xCELLigence system. Nat. Methods 6, v–vi. https://doi.org/10.1038/nmeth.f.263.

Bollag, W.B., Hill, W.D., 2013. CXCR4 in epidermal keratinocytes: crosstalk within theskin. J. Invest. Dermatol. 133, 2505–2508. https://doi.org/10.1038/jid.2013.271.

Bucalo, B., Eaglstein, W.H., Falanga, V., 1993. Inhibition of cell proliferation by chronicwound fluid. Wound Repair Regen. 1, 181–186. https://doi.org/10.1046/j.1524-475X.1993.10308.x.

Chen, X., Peng, L.H., Li, N., Li, Q.M., Li, P., Fung, K.P., Leung, P.C., Gao, J.Q., 2012. The

Fig. 9. (A) Immunofluorescent staining for theCD90-positive BMSCs at the wound healing region7d after treatment. Red fluorescence indicates CD90expression (indicated by the white arrows), and blueindicates DAPI stained cells nucleus. Scale bars is100 µm. (B) Histogram showing the mean fluores-cence intensity of CD90. All images were takenunder the same magnification. N=5.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

10

healing and anti-scar effects of astragaloside IV on the wound repair in vitro and invivo. J. Ethnopharmacol. 139, 721–727. https://doi.org/10.1016/j.jep.2011.11.035.

Chen, X., Peng, L.H., Shan, Y.H., Li, N., Wei, W., Yu, L., Li, Q.M., Liang, W.Q., Gao, J.Q.,2013. Astragaloside IV-loaded nanoparticle-enriched hydrogel induces woundhealing and anti-scar activity through topical delivery. Int. J. Pharm. 447, 171–181.https://doi.org/10.1016/j.ijpharm.2013.02.054.

Chen, L., Tredget, E.E., Wu, P.Y.G., Wu, Y., 2008. Paracrine factors of mesenchymal stemcells recruit macrophages and endothelial lineage cells and enhance wound healing.PLoS ONE 3, e1886.

Chen, H., Xing, X., Tan, H., Jia, Y., Zhou, T., Chen, Y., Ling, Z., Hu, X., 2017. Covalentlyantibacterial alginate-chitosan hydrogel dressing integrated gelatin microspherescontaining tetracycline hydrochloride for wound healing. Mater. Sci. Eng., C 70,287–295. https://doi.org/10.1016/j.msec.2016.08.086.

Dang, K.M., Yoksan, R., 2016. Morphological characteristics and barrier properties ofthermoplastic starch/chitosan blown film. Carbohydr. Polym. 150, 40–47. https://doi.org/10.1016/j.carbpol.2016.04.113.

Dong, Y., Sigen, A., Rodrigues, M., Li, X., Kwon, S.H., Kosaric, N., Khong, S., Gao, Y.,Wang, W., Gurtner, G.C., 2017. Injectable and tunable gelatin hydrogels enhancestem cell retention and improve cutaneous wound healing. Adv. Funct. Mater. 27,1–12. https://doi.org/10.1002/adfm.201606619.

Faraj, K.A., van Kuppevelt, T.H., Daamen, W.F., 2007. Construction of collagen scaffoldsthat mimic the three-dimensional architecture of specific tissues. Tissue Eng. 13,2387–2394. https://doi.org/10.1089/ten.2006.0320.

Field, C.K., Kerstein, M.D., 1994. Overview of wound healing in a moist environment.Am. J. Surg. 167, 2–6. https://doi.org/10.1016/0002-9610(94)90002-7.

Goldring, C.E.P., Duffy, P.A., Benvenisty, N., Andrews, P.W., Ben-David, U., Eakins, R.,French, N., Hanley, N.A., Kelly, L., Kitteringham, N.R., Kurth, J., Ladenheim, D.,Laverty, H., McBlane, J., Narayanan, G., Patel, S., Reinhardt, J., Rossi, A., Sharpe, M.,Park, B.K., 2011. Assessing the safety of stem cell therapeutics. Cell Stem Cell 8,618–628. https://doi.org/10.1016/j.stem.2011.05.012.

Hoare, T.R., Kohane, D.S., 2008. Hydrogels in drug delivery: progress and challenges.Polymer (Guildf) 49, 1993–2007. https://doi.org/10.1016/j.polymer.2008.01.027.

Huang, Y.C., Liu, T.J., 2012. Mobilization of mesenchymal stem cells by stromal cell-derived factor-1 released from chitosan/tripolyphosphate/fucoidan nanoparticles.Acta Biomater. 8, 1048–1056. https://doi.org/10.1016/j.actbio.2011.12.009.

Hudson, T.J., 2006. Skin barrier function and allergic risk. Nat. Genet. 38, 399–400.https://doi.org/10.1038/ng0406-399.

Jan-Marcus, D., Wiebke, B., Philipp, S., Soenke, W., Harald, T., Sedding, D.G., 2010.Time-course analysis on the differentiation of bone marrow-derived progenitor cellsinto smooth muscle cells during Neointima formation. Arterioscler. Thromb. Vasc.Biol. 30, 1890–1896. https://doi.org/10.1161/ATVBAHA.110.209692.

Jiang, S., Liu, S., Feng, W., 2011. PVA hydrogel properties for biomedical application. J.Mech. Behav. Biomed. Mater. 4, 1228–1233. https://doi.org/10.1016/j.jmbbm.2011.04.005.

Junker, J.P.E., Kamel, R.A., Caterson, E.J., Eriksson, E., 2013. Clinical impact uponwound healing and inflammation in moist, wet, and dry environments. Adv. WoundCare 2, 348–356. https://doi.org/10.1089/wound.2012.0412.

Kamoun, E.A., Kenawy, E.R.S., Chen, X., 2017. A review on polymeric hydrogel mem-branes for wound dressing applications: PVA-based hydrogel dressings. J. Adv. Res. 8,217–233. https://doi.org/10.1016/j.jare.2017.01.005.

Kan, E.M., Lu, E.A.L., 2010. Stem Cell Therapy for Spinal Cord Injury. Curr. Med. Chem.17 (36), 4492–4510. https://doi.org/10.2174/092986710794182971.

Koerner, J., Nesic, D., Romero, J.D., Brehm, W., Mainil-Varlet, P., Grogan, S.P., 2006.Equine peripheral blood-derived progenitors in comparison to bone marrow-derivedmesenchymal stem cells. Stem Cells 24, 1613–1619. https://doi.org/10.1634/stemcells.2005-0264.

Korting, H.C., Schöllmann, C., White, R.J., 2011. Management of minor acute cutaneouswounds: importance of wound healing in a moist environment. J. Eur. Acad.Dermatology Venereol. 25, 130–137. https://doi.org/10.1111/j.1468-3083.2010.03775.x.

Kunkel, S.L., Kimball, A.S., Boniakowski, A.E., Jacobs, B.N., Gallagher, K.A., 2017.Macrophage-mediated inflammation in normal and diabetic wound healing. J.Immunol. 199, 17–24. https://doi.org/10.4049/jimmunol.1700223.

Lai, H.-J., Kuan, C.-H., Wu, H.-C., Tsai, J.-C., Chen, T.-M., Hsieh, D.-J., Wang, T.-W., 2014.Tailored design of electrospun composite nanofibers with staged release of multipleangiogenic growth factors for chronic wound healing. Acta Biomater. 10, 4156–4166.https://doi.org/10.1016/j.actbio.2014.05.001.

Lang, H., Ebihara, Y., Schmiedt, R.A., Minamiguchi, H., Zhou, D., Smythe, N., Liu, L.,Ogawa, M., Schulte, B.A., 2006. Contribution of bone marrow hematopoietic stemcells to adult mouse inner ear: mesenchymal cells and fibrocytes. J. Comp. Neurol.496, 187–201. https://doi.org/10.1002/cne.20929.

Larouche, J., Sheoran, S., Maruyama, K., Martino, M.M., 2018. Immune regulation of skinwound healing: mechanisms and novel therapeutic targets. Adv. Wound Care 7,209–231. https://doi.org/10.1089/wound.2017.0761.

Lee, S., Valmikinathan, C.M., Byun, J., Kim, S., Lee, G., Mokarram, N., Pai, S.B., Um, E.,Bellamkonda, R.V., Yoon, Y.S., 2015. Enhanced therapeutic neovascularization byCD31-expressing cells and embryonic stem cell-derived endothelial cells engineeredwith chitosan hydrogel containing VEGF-releasing microtubes. Biomaterials 63,158–167. https://doi.org/10.1016/j.biomaterials.2015.06.009.

Li, X., Dai, J., 2018. Bridging the gap with functional collagen scaffolds: tuning en-dogenous neural stem cells for severe spinal cord injury repair. Biomater. Sci. 6,265–271. https://doi.org/10.1039/c7bm00974g.

Li, H., Fu, X., Ouyang, Y., Cai, C., Wang, J., Sun, T., 2006. Adult bone-marrow-derivedmesenchymal stem cells contribute to wound healing of skin appendages. Cell TissueRes. 326, 725–736. https://doi.org/10.1007/s00441-006-0270-9.

Li, L.J., Gao, S.Q., Peng, L.H., Wang, X.R., Zhang, Y., Hu, Z.J., Gao, J.Q., 2017a.

Evaluation of efficacy of aloin in treating acute trauma in vitro and in vivo. Biomed.Pharmacother. 88, 1211–1219. https://doi.org/10.1016/j.biopha.2017.01.174.

Li, Y., Han, Y., Wang, X., Peng, J., Xu, Y., Chang, J., 2017b. Multifunctional hydrogelsprepared by dual ion cross-linking for chronic wound healing. ACS Appl. Mater.Interfaces 9, 16054–16062. https://doi.org/10.1021/acsami.7b04801.

Lu, G., Ding, Z., Wei, Y., Lu, X., Lu, Q., Kaplan, D.L., 2018. Anisotropic biomimetic silkscaffolds for improved cell migration and healing of skin wounds. ACS Appl. Mater.Interfaces 10, 44314–44323. https://doi.org/10.1021/acsami.8b18626.

Lu, Z., Gao, J., He, Q., Wu, J., Liang, D., Yang, H., Chen, R., 2017. Enhanced antibacterialand wound healing activities of microporous chitosan-Ag/ZnO composite dressing.Carbohydr. Polym. 156, 460–469. https://doi.org/10.1016/j.carbpol.2016.09.051.

Madison, K.C., 2003. Barrier function of the skin: “La Raison d’Être” of the epidermis. J,Invest. Dermatol. 121, 231–241. https://doi.org/10.1046/j.1523-1747.2003.12359.x.

Mahdavian, B., Veer, W.M Van Deer, Egmond, M Van, Niessen, F.B., Beelen, R.H.J., 2011.Immunobiology Macrophages in skin injury and repair. Immunobiology 216,753–762. https://doi.org/10.1016/j.imbio.2011.01.001.

Man, Z., Hu, X., Liu, Z., Huang, H., Meng, Q., Zhang, X., Dai, L., Zhang, J., Fu, X., Duan,X., Zhou, C., Ao, Y., 2016. Transplantation of allogenic chondrocytes with chitosanhydrogel-demineralized bone matrix hybrid scaffold to repair rabbit cartilage injury.Biomaterials 108, 157–167. https://doi.org/10.1016/j.biomaterials.2016.09.002.

Mauney, J., Olsen, B.R., Volloch, V., 2010. Matrix remodeling as stem cell recruitmentevent: a novel in vitro model for homing of human bone marrow stromal cells to thesite of injury shows crucial role of extracellular collagen matrix. Matrix Biol. 29,657–663. https://doi.org/10.1016/j.matbio.2010.08.008.

Miles, K.B., Ball, R.L., Matthew, H.W.T., 2016. Chitosan films with improved tensilestrength and toughness from N-acetyl-cysteine mediated disulfide bonds. Carbohydr.Polym. 139, 1–9. https://doi.org/10.1016/j.carbpol.2015.11.052.

Miller, R.H., 2006. The promise of stem cells for neural repair. Brain Res. 1091, 258–264.https://doi.org/10.1016/j.brainres.2006.01.073.

Murakami, K., Aoki, H., Nakamura, Shingo, Nakamura, Shinichiro, Takikawa, M.,Hanzawa, M., Kishimoto, S., Hattori, H., Tanaka, Y., Kiyosawa, T., Sato, Y., Ishihara,M., 2010. Hydrogel blends of chitin/chitosan, fucoidan and alginate as healing-im-paired wound dressings. Biomaterials 31, 83–90. https://doi.org/10.1016/j.biomaterials.2009.09.031.

Nauman, E.B., Patel, K., Karande, P., 2011. Design of optimized diffusion-controlledtransdermal drug delivery systems. Drug Dev. Ind. Pharm. 37, 93–102. https://doi.org/10.3109/03639045.2010.495751.

Peng, L.H., Fung, K.P., Leung, P.C., Gao, J.Q., 2011. Genetically manipulated adult stemcells for wound healing. Drug Discov. Today 16, 957–966. https://doi.org/10.1016/j.drudis.2011.07.009.

Peng, L.-H., Niu, J., Zhang, C.-Z., Yu, W., Wu, J.-H., Shan, Y.-H., Wang, X.-R., Shen, Y.-Q.,Mao, Z.-W., Liang, W.-Q., Gao, J.-Q., 2014. TAT conjugated cationic noble metalnanoparticles for gene delivery to epidermal stem cells. Biomaterials 35, 5605–5618.https://doi.org/10.1016/j.biomaterials.2014.03.062.

Peng, L.-H., Huang, Y.-F., Zhang, C.-Z., Niu, J., Chen, Y., Chu, Y., Jiang, Z.-H., Gao, J.-Q.,Mao, Z.-W., 2016. Integration of antimicrobial peptides with gold nanoparticles asunique non-viral vectors for gene delivery to mesenchymal stem cells with anti-bacterial activity. Biomaterials 103, 137–149. https://doi.org/10.1016/j.biomaterials.2016.06.057.

Peng, L.H., Tsang, S.Y., Tabata, Y., Gao, J.Q., 2012. Genetically-manipulated adult stemcells as therapeutic agents and gene delivery vehicle for wound repair and re-generation. J. Control. Release 157, 321–330. https://doi.org/10.1016/j.jconrel.2011.08.027.

Peng, L.H., Wei, W., Qi, X.T., Shan, Y.H., Zhang, F.J., Chen, X., Zhu, Q.Y., Yu, L., Liang,W.Q., Gao, J.Q., 2013b. Epidermal stem cells manipulated by pDNA-VEGF165/CYD-PEI nanoparticles loaded gelatin/β-TCP matrix as a therapeutic agent and gene de-livery vehicle for wound healing. Mol. Pharm. 10, 3090–3102. https://doi.org/10.1021/mp400162k.

Peng, L.H., Mao, Z.Y., Qi, X.T., Chen, X., Li, N., Tabata, Y., Gao, J.Q., 2013a.Transplantation of bone-marrow-derived mesenchymal and epidermal stem cellscontribute to wound healing with different regenerative features. Cell Tissue Res.352, 573–583. https://doi.org/10.1007/s00441-013-1609-7.

Peng, L.H., Wei, W., Shan, Y.H., Zhang, T.Y., Zhang, C.Z., Wu, J.H., Yu, L., Lin, J., Liang,W.Q., Khang, G., Gao, J.Q., 2015. Β-cyclodextrin-linked polyethylenimine nano-particles facilitate gene transfer and enhance the angiogenic capacity of mesench-ymal stem cells for wound repair and regeneration. J. Biomed. Nanotechnol. 11,680–690. https://doi.org/10.1166/jbn.2015.1970.

Proksch, E., Brandner, J.M., Jensen, J.M., 2008. The skin: an indispensable barrier. Exp.Dermatol. 17, 1063–1072. https://doi.org/10.1111/j.1600-0625.2008.00786.x.

Qi, C., Xu, L., Deng, Y., Wang, G., Wang, Z., Wang, L., 2018. Sericin hydrogels promoteskin wound healing with effective regeneration of hair follicles and sebaceous glandsafter complete loss of epidermis and dermis. Biomater. Sci. 6, 2859–2870. https://doi.org/10.1039/C8BM00934A.

Rosa, R.M., Silva, J.C., Sanches, I.S., Henriques, C., 2017. Simultaneous photo-inducedcross-linking and silver nanoparticle formation in a PVP electrospun wound dressing.Mater. Lett. 207, 145–148. https://doi.org/10.1016/j.matlet.2017.07.046.

Sa, L., Estecha, A., Samaniego, R., Sa, S., Sa, P., 2016. The chemokine CXCL12 regulatesmono-mkp differentiation and RUNX3 expression. Blood 117, 88–98. https://doi.org/10.1182/blood-2009-12-258186.The.

Shan, Y.H., Peng, L.H., Liu, X., Chen, X., Xiong, J., Gao, J.Q., 2015. Silk fibroin/gelatinelectrospun nanofibrous dressing functionalized with astragaloside IV induceshealing and anti-scar effects on burn wound. Int. J. Pharm. 479, 291–301. https://doi.org/10.1016/j.ijpharm.2014.12.067.

Singer, A.J., Clark, R.A.F., 1999. Cutaneous wound healing. N. Engl. J. Med. 738–746.Svensjö, T., Pomahac, B., Yao, F., Slama, J., Eriksson, E., 2000. Accelerated healing of

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

11

full-thickness skin wounds in a wet environment. Plast. Reconstr. Surg. 106.Tang, Q., Luo, C., Lu, B., Fu, Q., Yin, H., Qin, Z., Lyu, D., Zhang, L., Fang, Z., Zhu, Y., Yao,

K., 2017. Thermosensitive chitosan-based hydrogels releasing stromal cell derivedfactor-1 alpha recruit MSC for corneal epithelium regeneration. Acta Biomater. 61,101–113. https://doi.org/10.1016/j.actbio.2017.08.001.

Trounson, A., McDonald, C., 2015. Stem cell therapies in clinical trials: progress andchallenges. Cell Stem Cell 17, 11–22. https://doi.org/10.1016/j.stem.2015.06.007.

Vogt, P.M., Andree, C., Breuing, K., Liu, P.Y., Slama, J., Helo, G., Eriksson, E., 1995. Dry,moist, and wet skin wound repair. Ann. Plast. Surg. 34, 493–499 discussion 499–500.

Von Bahr, L., Batsis, I., Moll, G., Hägg, M., Szakos, A., Sundberg, B., Uzunel, M., Ringden,O., Le Blanc, K., 2012. Analysis of tissues following mesenchymal stromal cell therapyin humans indicates limited long-term engraftment and no ectopic tissue formation.Stem Cells 30, 1575–1578.

Wysocki, A.B., Staiano-Coico, L., Grinnell, F., 1993. Wound fluid from chronic leg ulcers

contains elevated levels of metalloproteinases MMP-2 and MMP-9. J, Invest.Dermatol. 101, 64–68. https://doi.org/10.1111/1523-1747.ep12359590.

Xie, S.-Y., Peng, L.-H., Shan, Y.-H., Niu, J., Xiong, J., Gao, J.-Q., 2016. Adult stem cellsseeded on electrospinning silk fibroin nanofiberous scaffold enhance wound repairand regeneration. J. Nanosci. Nanotechnol. 16, 5498–5505. https://doi.org/10.1166/jnn.2016.11730.

Yesmin, S., Paget, M.B., Murray, H.E., Downing, R., 2017. Bio-scaffolds in organ-re-generation: clinical potential and current challenges. Curr. Res. Transl. Med. 65,103–113. https://doi.org/10.1016/j.retram.2017.08.002.

Zhang, C.Z., Niu, J., Chong, Y.S., Huang, Y.F., Chu, Y., Xie, S.Y., Jiang, Z.H., Peng, L.H.,2016. Porous microspheres as promising vehicles for the topical delivery of poorlysoluble asiaticoside accelerate wound healing and inhibit scar formation in vitro & invivo. Eur. J. Pharm. Biopharm. 109, 1–13. https://doi.org/10.1016/j.ejpb.2016.09.005.

X.-H. Xu, et al. International Journal of Pharmaceutics 570 (2019) 118648

12