Efficacy of thermosensitive chitosan/β-glycerophosphate hydrogel loaded with β-cyclodextrin-curcumin for the treatment of cutaneous wound infection in rats.

Wound infection has been a persistent problem that is common and costly. Thermosensitive hydrogel has been demonstrated to be a suitable dressing candidate due to its high moldability, easy administration and ability to maintain a moist topical environment at the wound bed. In the present study, a novel thermosensitive hydrogel was successfully prepared and characterized to have a porous inner structure and a sustained curcumin-releasing profile. The wound healing ability of the hydrogel was investigated in a wound infection model in rats. On analysis, it was observed that the hydrogel complex-dressed wounds exhibited a faster wound closure rate compared with gauze-covered wounds, which was paralleled with improved histological outcomes that were observed. Additionally, the results of in vitro antimicrobial, anti-oxidant, western blot analysis and reverse transcription-quantitative polymerase chain reaction assays indicated that the hydrogel complex had distinct anti-oxidative, antimicrobial and anti-nuclear factor-κB-signaling capacities. These results suggest that this novel hydrogel may be a suitable candidate for facilitating the healing of infected cutaneous wounds in rats.

Introduction

Wound infection has always been a challenging problem, particularly due to its commonality and cost, for both patients and doctors. Bacterial load impairs the topical blood supply, prolongs the inflammation phase and disfigures the appearance upon the eventual healing of the wound (1). Since their discovery, the wide applications of antiseptics and antibiotics have indicated promising effects against bacterial wound infection. However, a large proportion of bacteria have become resistant to antibiotics, and it is considered that antiseptics may have toxic effects and limited antimicrobial efficacy (2). Due to these reasons, novel and effective strategies for the treatment of wound infection are still needed.

Hydrogel has been demonstrated to be an effective wound dressing candidate due to its ability to absorb excessive tissue exudates, maintain a moist topical environment surrounding the wound surface, and allow air to permeate. Moreover, hydrogel can function as an efficient drug delivery system due to its porous structure within humid environments (3,4). In particular, in situ gel-forming hydrogel can be administrated to wounds in a minimally invasive manner by injection, and molds well into irregular wound defects through sol-gel transition. These advantages can result in alleviation of pain, shorter durations of hospital stay and reduced scar tissue (5,6).

Chitosan is well known for its excellent biocompatibility, low toxicity, and low immunostimulatory activity, and thus has long been used to facilitate wound healing (3,7,8). Chitosan alone or in combination with other components exhibits antimicrobial activity (9–11). Former studies have shown that chitosan-β-glycerolphosphate disodium salt (CH/GP) solution can undergo complete thermal-triggered gelation at normal body temperature (3,6). Meanwhile, researchers found have that incorporation of a third component [such as starch (12), collagen (13) or bioactive glass nanoparticles (14)] into the CH/GP system can lead to neotype hydrogels with improved inner structures and specific biological effects (6).

Curcumin is the active component of turmeric, and has long been used in Southeast Asian countries to treat a variety of diseases due to its antioxidant, anti-inflammatory, immunomodulatory and antimicrobial properties (15–19). Notably, curcumin may assist cutaneous wound healing even under complicated conditions, as observed for irradiated wounds (20), burn wounds (17), laser-induced wounds (21), and diabetic wounds (22). Despite these advantages, the low water solubility of curcumin severely limits its bioavailability and clinical efficacy. The curcumin-cyclodextrin complex (CD-Cur) is a modified form of curcumin, in which curcumin is included in the cylinder-like structure of cyclodextrin, that was developed to enhance curcumin bioavailability (23). Indeed, CD-Cur has previously exhibited increased cellular uptake, a longer half-life, and enhanced skin permeability, which thus led to a superior biological effect than curcumin alone (24–26). These results suggest that, with correct application, Cur-CD might be a suitable therapeutic for cutaneous wound healing.

In the present study, CD-Cur was loaded into a CH/GP system to form a novel thermosensitive hydrogel (CH/GP/Cur) for the treatment of a rat wound infection model. CH/GP/Cur was firstly prepared and its drug release rate, degradation rate, and rheology were investigated. Following this, CH/GP/Cur was assayed for its potential wound healing effect and antimicrobial capacity in infected wounds in rats. The anti-oxidative, antimicrobial and anti-NF-κB signaling capacities of CH/GP/Cur were also investigated.

Materials and methods

Materials and animals

The chitosan (degree of deacetylation, 85%; average molecular weight, 650 kDa; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), β-cyclodextrin, curcumin (≥95% purity; Sigma-Aldrich; Merck KGaA), and glycerol 2-phosphate disodium salt hydrate (β-GP; Sigma-Aldrich; Merck KGaA) reagents used in this article were all of analytical reagent grade. A total of 36 male Sprague Dawley (SD) rats weighing 200–220 g were purchased from the Laboratory Animal Center of Wuhan University (Wuhan, China). Animals were housed singly in a laminar-flow operating room and were fed with sterile food and water ad libitum. The animal procedures used were in accordance with the protocol approved by the Institutional Animal Care and Treatment Committee of Wuhan University, and the animals were treated humanely throughout the experiments.

Preparation of dialyzed chitosan solution (CH) and β-glycerophosphate solution (GP)

200 mg chitosan was dissolved in 9 ml of aqueous acetic acid (0.1 M), and the insoluble residues were removed by filtration. A dialysis method based on that reported by Kim et al (27) was then used to reduce the acetic acid content. Briefly, the chitosan solution was placed into a 27-mm dialysis bag (MW: 8000; Biosharp, Hefei, China) in distilled water, and the distilled water was changed daily for a week. The solution was then autoclaved, cooled, and stored at 4°C for further use. GP solution was prepared by dissolving 300 mg of GP in 1 ml of distilled water, followed by filtration through 0.22-µm PES syringe filters (Millex™, EMD Millipore, Billerica, MA, USA).

Preparation of β-cyclodextrin-curcumin inclusion complexes (CD-Curs) and determination of curcumin content in CD-Curs

CD-Curs were prepared by mixing β-cyclodextrin (CD) and curcumin (Cur) based on a previously reported method (24). In brief, CD (400 mg) was dissolved in 80 ml deionized water in a 200-ml glass vial with a magnetic bar, and 120 mg curcumin in 5 ml acetone were then added into the CD solution while centrifuged at 400 × g. The solution was stirred for 72 h at 37°C with no cap in order to evaporate the acetone, and then centrifuged at 1,000 × g for 5 min. Subsequently, the supernatant was collected and freeze-dried in a Labconco Freeze Dry System (Labconco, Kansas City, MO, USA) to recover the water soluble CD-Cur.

Cur concentration was measured using a UV-Vis spectrophotometer method (Orion™ AquaMate 8000; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 425 nm. First, a standard curve of curcumin in dimethyl sulfoxide (DMSO; 0–10 µg/ml) was prepared under identical conditions. Subsequently, 1 mg of CD-Cur was dissolved in 5 ml DMSO by gently shaking for 24 h at 37°C away from light. The solution was then centrifuged at 14,000 × g for 3 min, and the resultant curcumin-containing DMSO supernatant was collected to determine the curcumin content in the CD-Curs (24).

Preparation of CH/GP and CD-Cur-loaded CH/GP solution (CH/GP/Cur)

CH/GP solution was prepared by adding 1 ml GP solution dropwise to 9 ml dialyzed chitosan solution under gentle stirring in an ice bath (28). The chitosan and GP concentrations of the CH/GP solution were 2% (w/v) and 3% (w/v), respectively. CH/GP/CD-Cur was prepared via a simple mixing method. First, 200 mg of CD-Cur (containing 30 mg of curcumin) was dissolved into 1 ml deionized water and then added to 9 ml of dialyzed chitosan solution under mechanical stirring. The CH/CD-Cur solutions were then autoclaved at 121°C for 15 min, cooled at room temperature, and stored at 4°C. Finally, CH/GP/Cur was prepared by adding 1 ml GP solution dropwise to CH/CD-Cur solution (10 ml) in an ice bath as above.

Scanning electron microscopy

The morphology of the CH/GP/Cur hydrogel was investigated by scanning electron microscopy [SEM; FEI Quanta 200; Thermo Fisher Scientific (FEI), Eindhoven, Netherlands] at an accelerating voltage of 30 kV. The CH/GP/Cur hydrogel was prepared at 37°C for 3 h, placed into distilled water at 37°C for 24 h, and then lyophilized in the Labconco Freeze Dry System (Labconco) (27). Dry samples were placed on double-sided carbon tape, gold-sputtered and mounted on the SEM stage before scanning.

Assessment of in vitro curcumin release and hydrogel degradation and rheology

The drug release profile of CH/GP/Cur was determined using high-performance liquid chromatography (HPLC). In brief, 4 ml of dialyzed chitosan, 0.5 ml of 30% GP and 133 mg of CD-Cur (containing 20 mg of curcumin) were mixed as above and placed into 6-well plates. The plate was then incubated at 37°C for 10 min to induce sol-gel transition of the CH/GP/Cur hydrogel. Subsequently, 3 ml PBS (0.01 M, pH 7.35) was added into each well and the plates were gently shaken at 37°C. The supernatants were pipetted out at predetermined time points and replaced with the same volume of pre-warmed PBS. After centrifugation at 13,000 × g for 10 min, the supernatant was collected and stored at −20°C until further analysis. Released curcumin was quantified at 425 nm using HPLC (Waters Co., Milford, MA, USA) as described previously (29).

The degradation rate of CH/GP/Cur was determined using an enzymatic degradation method. In brief, CH/GP/Cur was prepared in a 6-well plate, frozen overnight (−80°C) and then lyophilized at −35°C (VirTis AdVantage Wizard 2.0; SP Scientific, Stone Ridge, NY, USA) for three days. The freeze-dried hydrogel was weighed (W0) and then placed in PBS containing 500–1,000 U/ml of lysozyme (kept in a 37°C water bath). At specific time intervals, the hydrogels were carefully retrieved, washed, frozen, lyophilized and weighed (Wt). The degradation rate of CH/GP/Cur was determined based on its reduction in weight, according to the following formula: Weight loss (%)=[(W0-Wt)/W0]×100%. The procedure was conducted in triplicate.

The sol-gel transition characteristics of the CH/GP/Cur were studied by rheometry (AR Rheometer 2000ex; TA Instruments, New Castle, DE, USA) (27). In brief, 1-ml samples of the CH/GP/Cur solution was individually placed into the rheometer, and the rheological properties of the samples were measured at a temperature range of 10–50°C with a constant heating rate of 1°C/min. Changes in the elastic modulus G' (Pa) and the viscous modulus G' (Pa) were recorded as functions of temperature at a fixed frequency of 1 Hz.

Establishment of rat wound infection model and measurement of wound area reduction rate

The rat wound infection model was established based on previously reported methods (2,30). In brief, animals were anesthetized with ketamine hydrochloride (60 mg/kg), the dorsal hair was removed and the exposed skin surface was disinfected with 70% ethanol. Full-thickness round wounds (2 cm in diameter) were then made on the back of the rats to muscle-depth using a scalpel blade, and the wounds were covered with Tegaderm (3M Monrovia, Monrovia, CA, USA). After 24 h, 0.2 ml bacterial suspension (2×109 CFU/ml) of Staphylococcus aureus (ATCC 25923) was injected into the muscle layer (5-mm deep), and another 0.2 ml bacterial suspension was applied topically to the wound. 72 h after the inoculation, photographs of the wounds were taken and tissue samples were collected for bacterial counting.

The 36 rats with infected wounds were randomly divided into 3 groups (12 rats per group) and their wounds were covered with saline-immersed gauze, CH/GP or CH/GP/Cur (containing 20 mg of curcumin), respectively. A piece of Tegaderm (3M Monrovia) was then used to fix the applied dressing. Photographs of the wounds were taken on days 3, 7 and 14. The photographs were taken under identical light conditions with ruler stickers placed at the edges of the wounds as size references. The wound area was measured using PhotoShop CS 6.0 based on the method described by Papazoglous (31). Wound closure rate was calculated as follows: (A0-At)/A0×100%, where A0 was the original wound area and At was the actual area at the different time points.

Tissue bacteria counting

Tissue bacterial load was investigated using an agar plate bacteria counting method. In brief, on days 0, 3 and 7 after establishment of the wound infection models, the granulation tissue was harvested, washed with normal saline and weighed. Tissue homogenate was then prepared in 1 ml of PBS and centrifuged (800 × g for 3 min at 4°C), and the resultant supernatant was diluted stepwise. Subsequently, 0.1 ml of each diluted supernatant was plated in duplicate on LB agar plates and cultured at 37°C, and after 24 h, the number of bacterial colonies on each plate was counted. Tissue bacterial loads were then calculated and expressed as l g CFU/g tissue. All the procedures were conducted using aseptic technique.

Histopathological examination

At 14 days after the different dressings were applied to the infected wounds, the granulation tissues were removed from the rats in each group. The tissues were fixed in 10% buffered formalin for 24 h at 4°C, embedded in paraffin, and then sectioned perpendicular to the wound surface into 4-µm sections. Hematoxylin and eosin (H&E) and Masson's trichrome staining were then performed according to standard procedures. Histological evaluation was performed by two observers blinded to the prior treatments using light microscopy.

A Sircol™ collagen assay (Biocolor, County Antrim, UK) was also conducted to quantify the soluble collagen content at 14 days post-wounding as a supplementary method to the Masson's trichrome staining. 1×1 cm tissue strips were harvested from each wound, and the samples were prepared and assayed following the manufacturer' protocol.

In vitro assay of CH/GP/Cur hydrogel antimicrobial activity

A disc diffusion method was used to evaluate the antimicrobial activity of the CH/GP/Cur hydrogel. The CH/GP/Cur hydrogel sheets were made into 15-mm-diameter discs (casted and shaped in 24-well plates), autoclaved, and placed in the center of agar culture plates (S. aureus; ATCC 25923). The plates were then incubated for 1 day at 37°C. A CH/GP hydrogel sheet and normal saline-immersed gauze were set as controls. A vernier caliper was used to measure the inhibition zone.

Analysis of CH/GP/Cur hydrogel anti-oxidant capacity

At 7 days post-wounding, animals were euthanized and granulation tissue samples were collected. 10% (w/v) wound tissue homogenate was then prepared in neutral phosphate buffer (50 mM) and centrifuged (1,000 × g for 10 min at 4°C). The resultant supernatant was then subjected to superoxide dismutase (SOD) (32), catalase (CAT) (33), and lipid peroxidation (LPx) (34) assays following the previously described protocols.

Western blot analysis of granulation tissue protein expression profile

At 7 days post-wounding, tissue homogenates of the isolated granulation tissue samples were prepared in RIPA buffer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) with a homogenizer. The tissue homogenates were then centrifuged at 1,000 × g for 10 min. The above steps were performed on ice to avoid protein denaturation and degradation. Following centrifugation, the resultant supernatants were subjected to western blot analysis following a standardized protocol (35). The primary antibodies used in this study were anti-IκBα (Santa Cruz Biotechnology, Inc., Dallas, TX, USA), anti-TGFβ1 (SSanta Cruz Biotechnology, Inc.), anti-pAKT (Cell Signaling, USA) and anti-PI3K (Cell Signaling Technology, Inc., Danvers, MA, USA).

Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis of granulation tissue gene expression profile

After washing with PBS containing DEPC, total RNA was immediately extracted from the isolated granulation tissues on day 7 using a commercialized kit (RNeasy Plus kit; Qiagen GmbH, Hilden, Germany) following the manufacturer's protocol. cDNA was generated using a reverse transcription system (ImProm-II; Promega Corporation, Madison, WI, USA), and qPCR was performed using a Real-Time PCR system (ABI7900; Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The relative amount of mRNA in each sample was normalized to the transcript levels. The sequences of the PCR primers used are listed in Table I.

Table I.

Primer sequences for reverse transcription-quantitative polymerase chain reaction.

Genes	Forward primers	Reverse primers
NF-κB	5′-GGCAGCACTCCTTATCAACC-3′	5′-GAGGTGTCGTCCCATCGTAG-3′
β-actin	5′-CACGATGGAGGGGCCGGACTCATC-3′	5′-TAAAGACCTCTATGCCAACACAGT-3′

NF-κB, nuclear factor-κB.

Statistical analysis

Statistical analysis was conducted using SPSS 13.0 software (SPSS, Inc., Chicago, IL, USA). Data comparisons were performed using one-way ANOVA followed by Student's-Newman-Keuls or Independent Samples t-tests. A two-tailed value of P<0.05 was considered to indicate a statistically significant value.

Results

Physicochemical characterization of CH/GP/Cur hydrogel

Micro and macro appearance of CH/GP/Cur hydrogel

The inner and surface morphology of the Ch/GP/Cur hydrogel was studied using SEM. Ch/GP/Cur exhibited an inter-connected, thick fibrillar inner structure (Fig. 1A) with irregular pore sizes on the surface (Fig. 1B). The macro appearance and the sol-gel transition process are depicted in Fig. 1C; the vial on the left shows Ch/GP/Cur solution under room temperature while the vial on the right shows the gel phase of Ch/GP/Cur when placed at 37°C for 10 min.

Figure 1.

Characterization of CH/GP/Cur hydrogel. Scanning electron micrographs of (A) the inner structure and (B) surface morphology of the lyophilized CH/GP/Cur hydrogel (magnification, ×200); (C) A photo showing the temperature-triggered sol (left, room temperature) to gel (right, 37°C for 10 min) transition; (D) in vitro curcumin release behavior of CH/GP/Cur; (E) in vitro degradation profile of CH/GP/Cur; (F) rheological analysis of the CH/GP/Cur solution; the arrow head indicates the transition point where G' (elastic modulus) outweighs G' (viscous modulus). CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin.

In vitro curcumin release behavior of CH/GP/Cur

A curcumin release profile of the CH/GP/Cur hydrogel was generated using a HPLC method, as shown in Fig. 1D. It was observed that curcumin was released slowly in a sustained manner rather than in a rapid-burst release; 39.95±6.90% of the curcumin was released within 2 weeks. This curcumin release profile indicated the potential of the CH/GP/Cur composite as an in situ gel system with sustained drug release capacity.

In vitro degradation behavior of CH/GP/Cur

To investigate the stability of CH/GP/Cur when applied on topical wounds, the degradation profile of CH/GP/Cur was determined by an enzymatic degradation method. As shown in Fig. 1E. 53.26±5.41% of the CH/GP/Cur hydrogel was degraded within 14 days.

Rheological properties of CH/GP/Cur

Fig. 1F shows the changes in the elastic modulus (G') and viscous modulus (G') of the CH/GP/Cur hydrogel as functions of temperature. The sol-gel transition took place at around 29°C, which was characterized by a marked increase in G'. Following this, G' showed a stable increase as a function of temperature while G' showed little increase. The higher G' value compared with G' indicated successful sol-gel transition.

Determination of curcumin content in the CD-Cur inclusion complex

A total of 400 mg of CD and 120 mg of Cur yielded 384.73 mg of CD-Cur. Based on the absorption value at 425 nm of standard samples of curcumin in DMSO, a standard curve was generated. CD-Cur was then dissolved in DMSO and analyzed by UV-Vis spectrophotometry. The results of this showed that 1 mg of CD-Cur contained 148.09±5.01 µg of curcumin.

Effect of gauze, CH/GP or CH/GP/Cur application on infected wound healing

Wound appearance

Images of wounds were taken on days 0, 3, 7 and 14 after CH/GP/Cur, CH/GP or gauze were applied on the infected wounds. As shown in Fig. 2A, on day 0, the wounds in all three groups exhibited milky-white purulent exudate, and the conditions of the wounds were comparable among the three groups. Meanwhile, on days 3 and 7, the saline gauze-treated wounds showed varying degrees of necrosis, pathological fluid oozing, redness, and swelling around the wound edge, all of which were evident to a more severe extent in the gauze-treated wounds when compared with those treated with CH/GP. In comparison, the CH/GP/Cur-treated wounds appeared relatively dry, exhibited little slough and no erythematous rash around the wounds, and healthy granulation tissue that was light pink and granular in appearance had started to fill the wounds.

Figure 2.

CH/GP/Cur accelerated infected wound healing and reduced topical bacterial load. Appearance of (A) infected wounds, (B) wound closure rate and (C) tissue bacterial load following treatment with gauze, CH/GP or CH/GP/Cur for 0, 3, 7 and 14 days post-wounding. Data are shown as the mean ± standard deviation; **P<0.01 and *P<0.05, one-way ANOVA. CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin.

Wound closure rate

In addition to superficial wound appearance, the closure rate of the infected wounds was determined in each of the three groups as the percentage reduction in the wound area. As shown in Fig. 2B, the CH/GP/Cur group exhibited a faster healing rate (63.89±4.62% on day 7; 94.67±2.29% on day 14) compared with the CH/GP (39.67±10.26% on day 7, P<0.01; 85.84±8.83% on day 14, P<0.05) and gauze (18.17±9.83% on day 07, P<0.01; 71.31±8.24% on day 14, P<0.05) groups. No statistically significant differences were detected between the groups on day 3.

Tissue bacteria count on wound surface

Tissue bacterial load was investigated using an agar plate bacteria counting method on days 0, 3 and 7 after the wound infection models were established. As shown in Fig. 2C, tissue bacterial loads (expressed as lg CFU/g tissue) were significantly reduced in the CH/GP/Cur group on both days 3 and 7 when compared with the gauze and CH/GP groups; on day 3, tissue bacterial load was 4.46±1.51 in the CH/GP/Cur group, 8.29±1.80 (P<0.05) in the CH/GP group, and 10.70±2.27 (P<0.01) in the gauze group; on day 7, bacterial load was 2.51±0.61 in the CH/GP/Cur group, 5.48±1.85 (P<0.05) in the CH/GP group and 8.01±1.06 (P<0.01) in the gauze group.

Histopathological analysis results

As indicated in Fig. 3, the histopathologies of the infected wounds treated with gauze, CH/GP or CH/GP/Cur were investigated by H&E and Masson staining. As depicted in Fig. 3Cc, wounds treated with CH/GP/Cur exhibited advanced re-epithelialization characterized by 4–5-layer structured epidermis, which was similar to epithelium under normal physiological conditions. In addition, extensive extracellular matrix deposition was observed for the CH/GP/Cur-treated wounds, and the cells and collagen fibers appeared compact and well-aligned. In comparison, re-epithelialization in the CH/GP-treated wounds was apparent to a lesser extent, though migration of peripheral epithelial cells underneath the scab tissue was still observed (Fig. 3Bb). In the gauze-treated group, any re-epithelialization was too minor to be seen, and marked levels of inflammatory cells were present in the granulation tissue (Fig. 3Aa) instead of epithelial and extracellular matrix components.

Figure 3.

CH/GP/Cur facilitated epithelialization of infected wounds. H&E stained sections of the granulation tissue in the gauze (A, ×100 magnification; a, ×400 magnification), CH/GP (B, ×100 magnification; b, ×400 magnification), and CH/GP/Cur (C, ×100 magnification; c, ×400 magnification) groups on day 14 after application of the treatments. CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin.

In order to clarify the extent of extracellular matrix deposition, we performed Masson's trichrome staining to visualize extracellular matrix collagen bundles. Additionally, a Sircol collagen assay was also used to quantify the collagen contents (Fig. 4). As depicted in Fig. 4Cc, CH/GP/Cur-treated wounds exhibited more densely packed collagen bundles (stained in blue) with marked parallel arrangement when compared with the other groups (Fig. 4Aa and Bb). Meanwhile, results of the Sircol collagen assay (Fig. 4D) indicated that CH/GP/Cur-treated wounds contained the highest level of collagen (42.17±5.87 µg/mg) compared with those treated with CH/GP (29.87±4.39 µg/mg, P<0.05) and gauze (14.16±4.29 µg/mg, P<0.01) on day 14.

Figure 4.

CH/GP/Cur facilitated collagen deposition and remodeling in infected wounds. Masson's trichrome staining of granulation tissue sections from the gauze (A, ×100 magnification; a, ×400 magnification), CH/GP (B, ×100 magnification; b, ×400 magnification), and CH/GP/Cur (C, ×100 magnification; c, ×400 magnification) groups on day 14 after application of the treatments. (D) A Sircol collagen assay indicated that CH/GP/Cur-treated wounds contained the highest level of collagen. Data are shown as the mean ± standard deviation; **P<0.01 and *P<0.05, one-way ANOVA. CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin

Anti-oxidant, antimicrobial and anti-NF-κB signaling capacities of CH/GP/Cur

In vitro antimicrobial activity of CH/GP/Cur hydrogel

The in vitro antimicrobial activity of the CH/GP/Cur hydrogel was investigated by a disc diffusion method (Fig. 5). A distinct bacterial inhibition zone (0.48±0.12 cm) was measured around the CH/GP/Cur hydrogel, as shown in Fig. 5C. By contrast, the inhibition zone was significantly smaller around the CH/GP hydrogel (0.16±0.08, P<0.01; Fig. 5B) and no inhibition zone was observed around the normal saline-immersed gauze (Fig. 5A).

Figure 5.

Possible mechanisms of action involved in the wound healing process. Antimicrobial activity of (A) normal saline-immersed gauze, (B) CH/GP and (C) CH/GP/Cur. Red arrows indicate the bacterial inhibition zone. (D and E) Selected protein and gene expression profiles of the NF-κB pathway in the different groups. (F-H) The effect of the different dressings on (F) superoxide dismutase, (G) catalase and (H) lipid peroxidation levels in the wounded tissue on day 7 post-wounding. Data are shown as the mean ± standard deviation; **P<0.01 and *P<0.05, one-way ANOVA. CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin.

Anti-oxidant capacity analysis

The local levels of SOD, CAT and malondialdehyde (MDA, a product of LPx) were assayed to evaluate the extent of oxidative stress at the wound sites. As shown in Fig. 5F and G, SOD and LPx levels were the lowest in the CH/GP/Cur-treated wounds when compared with the CH/GP- and gauze-treated wounds. By contrast, CAT levels were higher in the CH/GP/Cur-treated group than in the other groups. These results indicated the effective anti-oxidant capacity of CH/GP/Cur.

Effect of CH/GP/Cur on NF-κB signal pathway

A selection of proteins were chosen to evaluate the effect of the different dressings on the NF-κB signal pathway. As shown in Fig. 5D, on day 7 post-wounding, downregulation of P13K and p-AKT could be seen in the western blot bands of the CH/GP/Cur group, while IκBα (inhibitory protein of the P13K/AKT/NF-κB signaling pathway) was expressed to a greater extent than in the other groups. Meanwhile, the results of RT-qPCR (Fig. 5E) indicted a significant downregulation of NF-κB mRNA.

Discussion

The present work aimed to develop a novel CD-Cur-loaded CH/GP hydrogel with the capacity to facilitate cutaneous infected wound healing. As shown by the results, the thermo-sensitive hydrogel was successfully prepared and the drug release rate, degradation rate and rheological properties were determined. Additionally, the in vivo animal experiments showed that CH/GP/Cur-treated wounds had a faster wound closure rate and improved histological outcomes compared with those in the other groups on establishment of the full-thickness excision wound infection rat model. Further results indicated that possible mechanisms underlying the wound-healing capacity of CH/GP/Cur may involve anti-oxidative, antimicrobial and anti-NF-κB signaling effects.

Thermosensitive gel-forming systems have gained increasing attention as wound dressings due to their advantages. In the present study, a CD-Cur-loaded CH/GP hydrogel was developed for cutaneous infected wound healing. As shown in Fig. 1F, at 29°C, the modulus of the CH/GP/Cur solution changed dramatically, and was characterized by G' (elastic modulus) outweighing G' (viscous modulus). This result indicated that CH/GP/Cur was a kind of thermo-sensitive hydrogel, and that the sol-gel transition could be triggered below or at body temperature. Furthermore, according to Fig. 1D, curcumin could be released from CH/GP/Cur in a sustained manner, potentially due to the porous inner structure (Fig. 1A) and moist topical environment.

Chitosan itself can accelerate wound healing, inhibit bacterial infection, modulate collagen deposition and promote smoother scarring (3,9). Former studies also showed that curcumin was able to facilitate wound healing (5,17,20). In the present study, incorporation of CD-Cur into a CH/GP hydrogel resulted in anti-infection and pro-healing effects, and the micro-environment underneath CH/GP/Cur became more hospitable for epithelial migration, revascularization, and granulation tissue formation; however, the underlying mechanisms are likely complex. Thus, the present study mainly focused on the anti-oxidant, antimicrobial and anti-NF-κB signaling capacities of CH/GP/Cur.

Firstly, we addressed the antimicrobial capacity of the CH/GP/Cur hydrogel. Previous results have indicated that chitosan functions as an antimicrobial agent mainly through an interaction between its positively charged NH3+ group and the negatively charged microbial cell membrane. This interaction may subsequently change the permeability of the microbial cell membrane, weaken the function of peptidoglycans and ultimately lead to electrolyte imbalance (9,10). With the incorporation of CD-Cur, this effect may be enhanced due to the antimicrobial effect of curcumin, which has been reported for various bacteria, including H. pylori, P. aeruginosa, K. pneumoniae and V. vulnificus (15). While the underlying mechanisms remain largely remain unknown, one hypothesis is that curcumin functions by interfering with bacterial growth and metabolism (15). Thus, CH/GP/Cur may reduce bacterial load through either synergistic or additive effects of chitosan and curcumin. Further study is still needed to elucidate the precise mechanism.

Secondly, due to the incorporation of CD-Cur, we addressed the anti-oxidant capacity of the CH/GP/Cur hydrogel. During the inflammatory phase of the wound healing process, various reactive oxygen species (ROS) are generated (16,36,37). In the case of infected wounds, ROS are even more abundant due to high bacterial load. Excessive levels of ROS are often associated with oxidative stress and impaired wound healing (16,36). SOD and CAT are enzymes that aid to scavenge excessive ROS, and hence the levels of SOD and CAT are considered to reflect the level of oxidative stress. In the present study (Fig. 5F-H), the levels of SOD and LPx were dramatically lower in the CH/GP/Cur-treated group than in the CH/GP and gauze groups. By contrast, CAT levels were decreased in the CH/GP/Cur-treated group, likely due to the conversion of scavenged ROS into hydrogen peroxide, which then stimulates the expression of CAT. These results were analogous with those of previous research (5,37), and indicated that the incorporation of CD-Cur into CH/GP may enhance ROS scavenging, as previously reported both in vitro and in vivo (16,36,38).

CH/GP/Cur significantly reduced bacterial load and oxidative stress in the wound site, and this process has been illustrated in Fig. 6. This resulted in an improved micro-environment for the cells present in the wound. NF-κB is a transcription factor that helps to govern the expression of genes involved in immune and inflammatory responses. NF-κB can be activated by various stimuli, such as wound infection and elevated oxidative stress (39). In brief, ROS initially phosphorylate PI3K/AKT, which leads to phosphorylation and activation of NF-κB, and subsequent degradation of the inhibitory protein IκB, which normally retains NF-κB in the cytoplasm. Consequently, NF-κB enters the nucleus and promotes the expression of target genes (40,41). Consistent with former studies on curcumin and the NF-κB signaling pathway (16,35), lower expression levels of pI3K and p-AKT were observed in the CH/GP/Cur-treated groups in our western blot analysis, while IκB was upregulated. This result was further confirmed by RT-qPCR, which identified a downregulation of NF-κB. The above results indicate that CH/GP/Cur may suppress the NF-κB signaling pathway and thus the inflammatory response, possibly by reducing the levels of external stimuli for the receptor of the NF-κB pathway. Meanwhile, we concluded that oxidative stress was alleviated due to the ROS scavenging capacity of curcumin, and bacterial load was lowered due to the antimicrobial capacity of CH/GP/Cur. Indeed, a former study showed that a combination of an anti-microbial reagent (such as chitosan) with a compound with immunomodulatory and anti-inflammatory activities (such as curcumin) can collectively aid to reduce the inflammatory responses to infection (42).

Figure 6.

Illustration showing the suppressive effects of CH/GP/Cur on bacterial load and oxidative stress in the infected wounds, which consequently created a relatively healthy micro-environment. CH/GP/Cur, chitosan/β-glycerophosphate-β-cyclodextrin-curcumin.

In the present study, CH/GP/Cur hydrogel was prepared and characterized, and was demonstrated to facilitate the healing of infected cutaneous wounds in rats. The underlying mechanisms of this wound healing process may be attributed to the anti-oxidative, antimicrobial, and anti-NF-κB signaling capacities of CH/GP/Cur.