The Antioxidant Property of Chitosan Green Tea Polyphenols Complex Induces Transglutaminase Activation in Wound Healing

Chemicals and animals

EGCG, HCl, NaOH, and acetone were purchased from Sigma-Aldrich (Mississauga, Canada). Chitosan was obtained from the Department of Biomedical Engineering, University of Cincinnati (Cincinnati, OH, USA), with an N-deacetylation degree of 90% and an average molecular weight of 150,000. All animal experiments were conducted in accordance with accepted standards of humane animal care. Ten-week-old bred male mice weighing 150–180 g were obtained from the Research Institute of Surgery, China Pharmaceutical University. The mice were housed individually in cages and bred under controlled temperature (24°C±1°C).

Optimization of green tea polyphenols extraction

Since polyphenols are sensitive to environmental factors, it is important to develop an optimal extraction method to shorten the extraction time, reduce organic solvent consumption, and prevent environmental pollution. In general, efficiency of the extraction of polyphenols is influenced by multiple parameters such as temperature, time solvent content, and their effects may be either independent or interactive. ¹⁴ Response surface methodology (RSM) is a collection of mathematical and statistical techniques that are widely applied in the processing industry to determine the effects of several variables for optimizing conditions. ^{15 –17}

In this study, a central composite design was performed with three independent variables on the extraction yield: X ₁, extraction temperature (°C); X ₂, MtOH concentration (%); X ₃, extraction time (h). Generally, crude green tea polyphenols extraction was performed as described by Slinkard and Singleton. ¹⁸ The green tea leaves (10 g) were extracted for polyphenols by MtOH extraction with different concentrations (30%, 40%, 50%, 60% and 70%) for a given time (extraction time ranging from 2 to 6 h), while the temperature of the water bath ranged from 20°C to 80°C. After extraction, the mixture was centrifuged at 1238 g for 15 min, and the supernatant was collected. The residue was extracted twice under the same conditions. Supernatants from extraction were stored in a freezer at −30°C for analysis.

The RSM was applied to the experimental data using a commercial statistical package, Design-Expert version 7.1 (Statease, Inc., Minneapolis, MN, USA). Experiments were randomized in order to minimize the effects of unexplained variability in the observed responses due to extraneous factors. ¹⁹ The range of independent variables and their levels are presented in Table 1. The independent variables and their ranges were chosen based on preliminary experimental results. The adequacy of the model was checked, accounting for R ² and adjusted-R ² . Three-dimensional response surfaces and contour plots for the response are the graphical representations of the regression equations.

CGP complex synthesis

Since the process of transporting bioactive polyphenols through chitosan as a polymer can be consistent with Fick's diffusion law, the mechanism of interaction of polyphenol and chitosan could be ascribed to the development of noncovalent forces in equilibrium with the soluble complex. Since this equilibrium is irreversible under acidic conditions, these soluble complexes may aggregate and precipitate from solution according to Popa et al. ⁷ CGP complex was obtained by interaction between the reactants of polyphenols, and chitosan was prepared according to our previous study. ²⁰ Chitosan was dissolved by stirring in 1 M HCl solution, followed by the neutralization with 1 M NaOH solution, up to pH 6.0. Afterward, the 1% aqueous polyphenols solution was added to chitosan, and the system was stirred continuously at room temperature for 6 h. After the reaction, the precipitate was separated through centrifugation at 2200 g for 10 min. The unreacted polyphenols were removed by repeated washing with distilled water. Finally, the product was washed with acetone and vacuum dried at room temperature. The polyphenol content was expressed as mg gallic acid equivalents (GAE) per 100 g of plant material.

Antioxidant assay

The total antioxidant activity of CGP complex and other samples (chitosan, EGCG, and polyphenols) were determined according as previously described. ^21,22 The antioxidant ability was calculated as follows: Antioxidant activity (%)=[(Ac−As)/Ac]×100, where Ac is the absorbance of the resulting solution measured at 234 nm and As is the sample solution after reaction with 100%, indicating the strongest antioxidant activity. The reducing power was determined according to the method of Oyaizu (1986), and the absorbance was measured at 700 nm against a blank. ²³ A higher absorbance indicates an increased reducing power. The scavenging ability was determined according to a previously published 1,1-diphenyl-2-picrylhydrazyl (DPPH) method. ²⁴ Chelating ability was performed according to the method of Dinis et al., and the absorbance of the mixture was determined at 562 nm against a blank. ²⁵ Half maximal effective concentration (EC₅₀) value is the effective concentration at which the antioxidant activity was 50% or the absorbance was 0.5, which was obtained from linear regression analysis. Each experiment was repeated thrice (n=3).

Wound-healing assay

Twenty mice were randomly chosen and divided into two groups. Group A received distilled water as control; Group B received CGP complex with Average EC₅₀ value (AE₅₀) (3.50 mg/mL) according to our preliminary research. All animals were anesthetized with 3-cm×3-cm midline areas of their backs delineated and shaved. Two 10-mm-diameter full-thickness wounds were then placed on the dorsum of each mouse in the shaved area. Relative wound surface areas were compared 4 days after injury. For each animal, the wound surface areas were averaged. Excision wounds were formalin fixed and embedded in paraffin. Sections (5–10 μm) were stained with hematoxylin and eosin and photographed under 100× magnification.

In silico analysis

The TGM sequences of the mouse (Mus musculus) were obtained from MGI (www.informatics.jax.org). The gene orthology of annotated mouse nucleotide sequences were retrieved from MGI (Table 2). The distribution of TGMs as positioned in the chromosomes of M. musculus were also obtained from MGI. TGMs were further subjected to bioinformatics servers of FGENESH, ²⁶ and the ORF structures were manually defined. The putative TGM domains were subjected to protein functional analysis using Pfam (http://pfam.sanger.ac.uk). Each of the TGM gene was positioned on mouse chromosomes obtained with MGI genome's database, and the resulting positions were automatically marked on the bar. The identified TGM genes were aligned using ClustalW; then, the phylogenetic tree was generated by bootstrapping (1000 reiterations) with the neighbor-joining method via PHYLIP (version 3.6). Coexpressed gene networks and correlation of expression patterns of TGM genes were retrieved from COXPRESdb (http://coxpresdb.jp). ²⁷ Gene's ontology (GO) released by public database (www.geneontology.org) and Kyoto encyclopedia of genes and genomes (KEGG) of biochemical pathways were applied in this study. Array expression values of annotated TGMs were retrieved from ArrayExpress of the European Bioinformatics Institute (www.ebi.ac.uk/arrayexpress).

Semi-quantitative reverse transcription-PCR

Mouse skin keratinocytes cultures were performed adopted from Dlugosz et al. with some modifications. ²⁸ Generally, the skins of mice were peeled off and floated on the surface of 0.4% trypsin at 4°C for 24 h. Epidermal sections were washed with phosphate-saline buffer (Sigma-Aldrich). The keratinocytes in the suspension were pelleted by centrifugation at 1000 g for 10 min and suspended in the serum-free medium of keratinocytes containing bovine pituitary extract. Total RNA was extracted with the RNeasy Micro kit (Qiagen, Valencia, CA, USA) and treated with RNase-free DNase I (Qiagen) to enable digestion. Reverse transcription (RT) was performed with Script Reverse Transcriptase (Bio-Rad, Hercules, CA, USA) to synthesize first-strand complementary DNA (cDNA). The resulting cDNA was used for PCR with the sequence-specific primer pairs listed in Table 3. The amplification was performed under the following conditions: one cycle at 94°C for 1 min, 35 cycles of 94°C (1 min)–55°C (1 min)–72°C (1 min), and, finally, one cycle at 72°C (10 min). Aliquots of the PCR products were subjected to 0.8% agarose gel electrophoresis and stained with ethidium bromide. The PCR products were visualized under UV light and photographed, which were scanned into a computer. Relative quantitative analysis was performed, and the ratios of target mRNA to housekeeping β-ACTIN mRNA concentrations were calculated. Each experiment was repeated thrice (n=3).

Statistical analysis

Statistical analysis was performed using the general linear model of SAS 9.0 (SAS Institute, Inc., Cary, NC, USA, 2009). P<.05 was considered statistically significant.

Modeling fitting and extraction procedure optimization

The experimental data for yield of green tea polyphenols (mg GAE/100 g) under different treatment conditions and polynomial response surface model fitted to the response variable (Y) are presented in Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/jmf). After the regression analysis, the effect of treatment variables as the linear, quadratic, and interaction terms were tested for adequacy and fitness by analysis of variance (Table 4). The lack of fit illustrated in Table 4 showed an insignificant P value for selected variables, meaning that the model was sufficiently accurate for predicting the relevant response. It was reported that when R ² (coefficients of determination) approaches the unity, the better empirical model fits the actual data. ¹⁶ The statistical data also demonstrated that the proposed regression model for polyphenols yield was adequate, possessing no significant lack of fit and with satisfactory values of the R ² (Table 4). Our results showed that the R ² value for the response variable is higher than 0.95, indicating that the regression model was suitable to explain the behavior. In addition, adj-R ² was calculated and an adj-R ² (0.91) of over 90% always supports the model adequacy. Moreover, the model of polyphenols yield was significant (P<.0001) to the yield. From the analysis just described, it was found that the extraction temperature and MtOH content were the most important factors (P<.0001).

The results of the polyphenol yield affected by extraction temperature, time, and MtOH content are shown in Figure 1a–c, respectively. The maximum predicted value was shown by the surface confined in the smallest ellipse in the contour diagram. Figure 1a showed the interaction between extraction temperature and MtOH content at constant extraction time (4 h). It could be found that the yield of polyphenols increased with increasing MtOH concentration and temperature. Generally, higher temperatures and MtOH concentration result in a hydrolysis of a plant cell wall that favors polyphenols release and, thus, increases the polarity of solvent. ¹⁹ In addition, the polyphenols yield increased in a parabolic manner with a constant MtOH concentration (50%) as well as an increase in extraction temperature and treatment time (Fig. 1b). However, the yield of polyphenols was not pronounced with increases in time and MtOH concentration (Fig. 1c), probably due to the degradation of polyphenols. The suitability of the model for predicting optimum response values is presented in the following equation: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}Y = & 73.58 + 14.28{X}_1 + 11.89{X}_2 + 5.54{X}_3 \\ & + 1.40{X_1}{X_2} - 2.92{X}_1{X}_3 + 0.41{X}_2{X}_3 \\ & + 1.35{X}_1^{2} + 1.23{X}_2^{2} + 1.38{X}_3^{2} \end{align*} \end{document}

OPEN IN VIEWER

FIG. 1.

Response surface plots showing the effect of (a) extraction time and percent of MtOH, (b) extraction time and percent of MtOH, and (c) extraction time and temperature on the yield of green tea polyphenols. Color images available online at www.liebertpub.com/jmf

These results demonstrated that a higher yield of polyphenols could be obtained under the optimal extraction conditions: at temperature 77°C for 6 h, and MtOH concentration at 71%. This set of conditions was determined to be feasible by the RSM approach and was also used to validate the experiments and to predict the values of the responses using the model equation.

Antioxidant properties

All samples showed consistent dose-dependent antioxidant activity and exhibited moderate to high antioxidant activities of 64%±2.2%–87%±0.8% at 10 mg/L (Fig. 2a). However, at 1 mg/mL, the antioxidant activities were 27%±1.2%, 35%±0.8%, 40%±1.7%, and 45%±2.5% for chitosan, EGCG, polyphenols, and CGP treatment, respectively. Among those, the CGP obtained from the interaction of chitosan and polyphenol was capable of increased inhibition of oxidation, Similarly, all samples showed consistent reducing power with an increased concentration. At 10 mg/mL, the reducing ability of CGP was 0.87±0.02; whereas others were in the range of 0.62±0.04–0.68±0.01 (Fig. 2b). However, the scavenging capability of polyphenols in DPPH radicals showed a slight decrease from 4–10 mg/mL, which may be due to the degradation of polyphenols. The scavenging ability of CGP was 75%±1.5% at 10 mg/mL (Fig. 2c), while chitosan, EGCG, and polyphenols showed scavenging abilities of 64%±1.3%, 68%±2.4%, and 70%±1.0% at 10 mg/mL, respectively. From 8 to 10 mg/mL, polyphenols and chitosan showed a plateau of scavenging abilities at 60%±1.4%–68%±2.5%, indicating that the chitosan and polyphenols were not as effective for DPPH radicals scavenging as CGP. At 10 mg/mL, the chelating ability of all samples on ferrous ions was in the range of 78%±1.1%–84%±1.0% of chitosan, EGCG, and polyphenols, respectively (Fig. 2d). It seemed as though all ferrous ions of CGP (92%±1.4%) were chelated at 10 mg/mL, but the effect of ion-chelating ability of CGP was complex. These results could be attributed to the CGP complex, enabling the binding ability with the phenol hydroxyl groups and amino groups of the chitosan molecules.

OPEN IN VIEWER

FIG. 2.

Antioxidant assay of CGP. (a) Antioxidant property of CGP; (b) reducing power of CGP; (c) scavenging property of CGP; and (d) chelating property of CGP. Each value is expressed as mean±SD (n=3). CGP, chitosan green tea polyphenols.

The antioxidant properties are presented in Table 5 and summarized as EC₅₀ values for each assay. EC₅₀ values were in the range of 3.10±0.24–5.30±0.16 mg/mL for polyphenols, EGCG, and chitosan, respectively, in which CGP (2.32±0.34 mg/mL) exhibited the greatest efficacy. Both EC₅₀ values of reducing powers and scavenging abilities in DPPH radicals were in the range of 3.05±0.25–6.25±0.24 mg/mL and 5.19±0.37–6.48±0.45 mg/mL, indicating that CGP was most potent for these two antioxidant attributes. With regard to chelating abilities in ferrous ions, the EC₅₀ values were in the range of 3.51±0.18–4.70±0.32 mg/mL. Overall, CGP was the most powerful in antioxidant activity, scavenging ability, and chelating ability in ferrous ions.

Wound-healing assay

The mice were administered 1.0 AE₅₀/kg body weight according to our preliminary test. The wound healing of control and CGP treatment were recorded as the days from injury to complete epithelization (Fig. 3). By day 4, wounds treated with CGP showed more epithelial tissue recovery than the control (Fig. 3a, c). On day 10, a notable wound closure occurred, because the epithelization of CGP treatment was complete (Fig. 3b). In contrast, wounds of controls healed more slowly at day 10 (Fig. 3d). Furthermore, the healing tissues of the wounds were evaluated by examining the histopathology of the CGP treatment and control samples at 4, 8, and 10 days after wounding (Fig. 4). At day 4, the epidermis was fully interrupted; the wound gap was filled with necrotic materials and inflammatory exudates; and no granulation tissue was observed at the time (Fig. 4a). CGP treatment demonstrated that wound gaps were filled with necrotic debris, and proliferation of blood vessels was formed with earlier granulation tissues (Fig. 4d). On the 8th day, the necrotic and inflammatory materials were filled by the proliferation of fibroblasts, which started filling up the gap (Fig. 4b). However, a large amount of necrotic debris was still observed. CGP treatment resulted in less acute inflammatory material, and the epidermis was smaller, which was filled by forming granulation tissues. The intact epidermal tissue was under growing and only a minimal gap was observed at the epithelial level (Fig. 4e), On the 10th day, the control group was still filling the gap and was superficially surrounded by fibrosis containing residual inflammatory debris (Fig. 4c). The CGP treatment of wounds improved significantly, the regrowth of the epidermis was nearly complete, and the wound scar was clearly formed (Fig. 4f). Therefore, the histological observations indicated that CGP was found to be effective in wound healing due to an acceleration of the reformation of granulation tissues in the wound-healing process.

OPEN IN VIEWER

FIG. 3.

Observation of excision wound healing: (a) Control (4th day), (b) CGP treatment (4th day), (c) Control (10th day), and (d) CGP treatment (10th day). Color images available online at www.liebertpub.com/jmf

OPEN IN VIEWER

FIG. 4.

H&E stained section of the skin tissue granulation of (a–c) control group on 4th, 8th, and 10th day of wound healing, (d–f). CGP treatment on 4th, 8th, and 10th day of wound healing (100×magnifications). Color images available online at www.liebertpub.com/jmf

In silico identification

TGMs have been proposed to participate in the process of wound healing, and they are activated in all layers of the skin, corresponding to the epidermis, the dermis, and the panniculus carnosus; particularly, plasma TGMs are activated by thrombin and contribute to the formation of a fibrin clot at the site of injury. ^29,30 So, a comprehensive in silico investigation of the TGM gene family of M. musculus from the recently available online database, its chromosomal location, domain structure, phylogenetic tree construction, coexpression gene network generation, and array expression value collection have been attempted.

TGM sequences of M. musculus obtained from MGI were performed using various bioinformatics tools. The presence of the conserved TGM domain is a typical feature for consideration as a member of the TGM family. The domains of TGM proteins revealed highly conserved sequences in all the seven TGM proteins. The phylogenetic tree with bootstrapping values revealed the existence of two major groups. Group A was further divided into two subgroups: A1 and A2 (Fig. 5a). The TGM gene family with high bootstrapping values suggested common origins of TGM genes for each subgroup. The putative gene structures of the predicted TGM genes in terms of ORF distribution patterns are shown in Figure 5b, and the TGM7 seemed to contain a long ORF structure. However, the putative domains of TGM proteins with 109–117 amino acids were 100% conserved in six TGM proteins, while only TGM7 showed a variation and less amino acid residues (67 aa) (Fig. 5c). All the TGM genes were distributed on three chromosomes as provided in the chromosome map of MGI (Supplementary Fig. S1). Chromosome 2 maintained a maximum of five TGM genes, whereas the other two were located on chromosome number 9 and 14. The investigations suggested that TGM1, TGM2, and TGM3 were essential to early wound healing of the injured skin. The minor expression of TGM1 in simple epithelia, including endothelial cells, might play a role in wound healing. ^31,32 TGM2 might be induced in response to acute-phase injury cytokines. ^33,34 TGM3 could catalyze the substrates to form a complex that was related to the differentiation of the stratified squamous epithelia. ^35,36 Thus, these TGMs may be involved in inflammatory and endothelial cell's construction by stabilizing the extracellular matrix, and the end product of the process is the replacement of the fibrin matrix with granulation tissue. ^35,37 As the reason of that fact, coexpression data of TGM1, TGM2, and TGM3 in mouse were obtained from COXPRESdb. On that basis, a gene network of coexpressed genes for TGM1, TGM2, and TGM3 was constructed to focus on their functional relationships based on the correlation ranks. The density of genes in correlation spaces and the genes encoding the proteins in each organelle were identified by the KEGG annotations (Fig. 6). Genes that were strongly coexpressed were automatically picked, added to the network, and represented as small gray-shaded nodes. Generally, three coexpressed genes were directly associated with TGM1 as the following annotations: Hmox with the annotation of response to oxidative stress (GO: 0006979) and oxidation reduction (GO: 0055114), Ms4a6d with the annotation of membrane formation (GO: 0016020), and Crym with the annotation of cytoplasm formation (GO: 0005737) (Fig. 6a). Coexpressed genes directly connected to TGM2 were identified by the following GO annotations: Ccrl2 with the annotation of membrane formation (GO: 0016021) and plasma membrane formation (GO: 0005886), Tmem176p with the annotation of integral to membrane formation (GO: 0016021), Tinagl1 with the annotation of extracellular region construction (GO: 0005576), and cytoplasm formation (GO: 0005737) (Fig. 6b). Sprrs involved in the wound-healing process with the annotation of cytoplasm keratinization (GO: 0031424) was identified as the TGM3 coexpressed gene (Fig. 6c). Furthermore, the correlation of expression patterns among TGM1, TGM2, and TGM3 was found and shown in Supplementary Figure S2. The correlation value between TGM1 and TGM2 (0.285) was higher and similar to TGM2 and TGM3 (0.234) (Supplementary Fig. S2a, b). However, the correlation value between TGM1 and TGM3 was lower (0.208) (Supplementary Fig. S2c). To evaluate the expression patterns of TGM1–3 against their human (Homo sapiens) counterparts in the wound-healing process, the array express values of TGM genes in epidermis were collected from ArrayExpress (Supplementary Table S2). Accordingly, the expressions of TGM1–3 genes were highly involved in the up processes of building mouse skin epidermis, while the human TGM1–3 genes were involved in both up and down processes of constructing fibroblast and skin epidermis.

OPEN IN VIEWER

FIG. 5.

(a) The phylogenetic tree of Mus musculus TGM genes constructed from a complete alignment of seven TGM gene sequences by neighbor-joining method (gap complete deletion) with 1000 bootstrapping analysis. The groups of homologous genes identified are indicated. The scale bar corresponds to 0.1 estimated nucleotide substitutions per site. (b) OFR structure of TGM genes, black bars represent the ORF; while the line represents the CDS. The size of exons and introns can be estimated using the horizontal scale. (c) The TGM domain for each TGM protein is marked with a black box. The length of the TGM proteins can be estimated using the horizontal scale. aa indicate the amino acid. TGM, transglutaminase.

OPEN IN VIEWER

FIG. 6.

Co-expressed gene network of (a) TGM1, (b) TGM2, and (c) TGM3 gene networks in COXPRESdb are drawn based on the rank of correlation using the correlation rank, which partially normalizes the density of genes in correlation space. Yellow represents target gene, light gray represents additional genes, and orange represents commonly appearing genes. Color images available online at www.liebertpub.com/jmf

Temporal expression of TGM

An in silico analysis indicated that TGMs, particularly TGM1–3, play important roles in the wound-healing process. Activation of TGMs by CGP was confirmed by expression profiling of each TGM mRNA in skin specimens by RT-PCR through 10 days after excision (Fig. 7) and 2 days posthealing. After excision, there was an increase in each TGM mRNA level. On comparing the mRNA expression levels against β-actin in the control group, it was shown that the peak level for TGM1 mRNA occurred at 24 h, while peak levels of TGM2 and TGM3 were at 72 and 168 h, respectively (Fig. 7a). All the mRNA levels decreased after the peak time and were normalized in 10 days in the control group. However, after CGP treatment, increased expression levels of TGM1 and TGM2 remained constant for 10 days, while the expression profile of TGM3 was similar to the control treatment (mouse received distilled water) (Fig 7b). Moreover, the ratios of the dramatic increase in expression levels of TGM1–3 were observed at 24 h after the wound healing (Fig. 8a). For control treatment, the expression values of CGP treatment appeared to be larger (Fig. 8b) at 24 h, after which the expression levels of all TGM1–3 were closely normalized for both the control and the CGP treatment group from 24to 48 h. According to the results, the significant increase in expression levels of TGM1–3 under CGP treatment after the excision wound model in the present study was correlated with the temporal expression levels of TGM genes. This result could be attributed to the activation of TGMs expression by CGP treatment, and the functionality of CGP treatment may be related to its antioxidant properties of skin function homeostasis.

OPEN IN VIEWER

FIG. 7.

Reverse transcription-PCR analysis of mRNA expression of epidermis fibroblast TGM1–3 during skin wound healing in mice of (a) Control and (b) CGP treatment groups.

OPEN IN VIEWER

FIG. 8.

Temporal expression of TGM1–3 for epidermal fibroblasts after excision. The data of 0 h after excision represent the results obtained from normal mice: (a) Control, (b) CGP treatment. Values with the same letter in the same column do not differ significantly at P<.05; each value is expressed as mean±standard deviation (n=3).