In vitro and In vivo wound healing studies of

0 downloads 0 Views 2MB Size Report
and the positive fraction containing asiaticoside was further quantified using ...... -regulatory-acceptance/docs-skin-irritation-1/DOC2_OECD-TG-404.pdf. Saad ...
South African Journal of Botany 108 (2017) 163–174

Contents lists available at ScienceDirect

South African Journal of Botany journal homepage: www.elsevier.com/locate/sajb

In vitro and In vivo wound healing studies of methanolic fraction of Centella asiatica extract H.A. Azis a, M. Taher a,⁎, A.S. Ahmed a, W.M.A.W. Sulaiman b, D. Susanti c, S.R. Chowdhury d, Z.A. Zakaria e,⁎ a

Department of Pharmaceutical Technology, Faculty of Pharmacy, International Islamic University Malaysia, Jalan Sultan Ahmad Shah, Bandar Indera Mahkota, 25200, Pahang, Malaysia Department of Basic Medicinal Sciences, Faculty of Pharmacy, International Islamic University Malaysia, Jalan Sultan Ahmad Shah, Bandar Indera Mahkota, 25200, Pahang, Malaysia c Department of Chemistry, Faculty of Science, International Islamic University Malaysia, Jalan Sultan Ahmad Shah, Bandar Indera Mahkota, 25200, Pahang, Malaysia d Tissue Engineering Centre, Universiti Kebangsaan Malaysia Medical Centre, 56000 Kuala Lumpur, Malaysia e Department of Biomedical Sciences, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia b

a r t i c l e

i n f o

Article history: Received 29 March 2016 Received in revised form 5 August 2016 Accepted 13 October 2016 Available online xxxx Edited by V Steenkamp Keywords: Centella asiatica Methanol fraction Asiaticoside Scratch assay Circular excision wound

a b s t r a c t Ethnopharmacological relevance: Asiaticoside is claimed as a bioactive compound capable of wound healing. In order to ensure that the pharmacological activity of the extract is traceable and measurable, the present study attempted to evaluate the bioactivity of rich fractionated extract of asiaticoside. Aim of the study: The current study evaluates the wound healing efficacy via in vitro scratch assay and in vivo circular wound excision model. Materials and methods: The ethanol extract was fractionated into seven fractions via vacuum liquid chromatography. The compound of interest in the fractions was qualitatively identified using thin layer chromatography and the positive fraction containing asiaticoside was further quantified using reverse-phase HPLC. The asiaticoside-rich fraction was subjected to (i) colorimetric MTT (methylthiazoltetrazolium) cytotoxicity assay following incubation with human dermal fibroblast (HDF) and human dermal keratinocyte (HaCaT); (ii) in vitro 12-well plate scratch assay (using HDF and HaCaT cells) and (iii) topically apply (40%, 10% and 2.5%, w/w) on in vivo circular wound excision of rabbits. Data on wound contraction, epithelisation period, hydroxyproline content and histophatological analysis was collected from in vivo study. Results: The results showed that the methanol fraction of the extract contained about 2.4% asiaticoside. Based on the results of colorimetric MTT (methylthiazoltetrazolium) cytotoxicity assay, both HDF and HaCaT showed significant stimulation upon application of the methanolic fraction of extract at concentrations of 100 μg/mL and 0.19 μg/mL. The methanol fraction showed almost no toxicity effect at the concentrations tested since their IC50 could not be determined in concentrations ranging from 100 μg/mL to 0.19 μg/mL. Since all the concentrations tested allowed for more than 90% cell viability, the concentrations chosen for the scratch assay were randomly chosen and designated as highest (100 μg/mL), medium (6 μg/mL) and lowest (0.2 μg/mL) concentrations. In the scratch assay, methanol fraction of extract with concentration of 0.2 μg/mL and 100 μg/mL showed significant effect on HDF and HaCaT compared to the positive control (p b 0.05). In vivo, it was shown that the methanol fraction of the extract induced collagen synthesis. Histopathology data also concluded that dose-dependent effect of the tested extract as a wound healer was present. Conclusions: Taken together, recent findings suggest that methanol fraction of C. asiatica demonstrated remarkable polyvalent activity, and thus has potential as an effective wound healer. In conclusion, the claim of the presence of wound healing properties in C. asiatica had been well supported based on the results obtained in this study. © 2016 SAAB. Published by Elsevier B.V. All rights reserved.

1. Introduction Wounds are an unavoidable and inescapable part of our life. Various types of wounds may require different types of treatments. Fortunately, the human body is equipped with a complex self-healing mechanism. ⁎ Corresponding authors. E-mail addresses: [email protected], [email protected] (M. Taher), [email protected], [email protected] (Z.A. Zakaria).

http://dx.doi.org/10.1016/j.sajb.2016.10.022 0254-6299/© 2016 SAAB. Published by Elsevier B.V. All rights reserved.

Even so, under conditions such as microbial infection, diabetic condition and poor blood circulation, the management of wounds become complicated and sometimes costly (Thakur et al., 2011). In order to overcome these problems, various products have appeared in the market to heal wounds in the shortest time possible and to increase patient compliance by minimizing pain, discomfort and scarring. However, it is important to recognize that wound care should always support the natural healing process to avoid any unwanted complications as well as to ensure a smooth healing process.

164

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

The present study would like to introduce Centella asiatica extract as a traditional herb that has wound healing properties to support the body's natural healing process. C. asiatica (L.) Urban is commonly known as Asiatic pennywort and is known locally as pegaga (Fig. 1). This herb has been consumed as medicine since ancient times especially in the Ayurverdic system of India and in folk medicine in China and Madagascar. Although in Malaysia, C. asiatica is also used by traditional healers in herbal remedies, its popularity is more confined as a vegetable rather than a medicinal plant (Jabatan Perhutanan Semenanjung Malaysia, n.d.). The World Health Organization (WHO) has documented C. asiatica as one of the most important medicinal plants to be conserved and cultivated (Jabatan Perhutanan Semenanjung Malaysia, n.d.) Previous studies on C. asiatica extracts have reported its potential as an antioxidant, antimicrobial agent, agent of collagen synthesis and even as a wound healer (Taemchuay et al., 2009; Hashim et al., 2011; Idrus et al., 2012). Most studies report asiaticoside as the active constituent producing the said effect. Asiaticoside is a triterpenoid compound that is found as a saponin glycoside due to the attachment of sugar molecules to a triterpene unit. The sugar molecules are glucose–glucose– rhamnose (Fig. 2). In various wound healing models, topical application (0.2%–0.4%), injection (1 mg) or ingestion (40 μg/disk) of asiaticoside has been shown to increase hydroxyproline content, improve tensile strength, increase collagen synthesis and remodeling of the collagen matrix, promote epithelialization, stimulate glycosaminoglycan synthesis and elevate antioxidant levels (Shukla et al., 1999a; Shukla et al., 1999b; Somboonwong et al., 2012). This study aimed to validate the pharmacological activity of asiaticoside by using extracts with standardized asiaticoside content. In the present study, the wound healing potential of C. asiatica extract was determined using in vitro scratch assay and in vivo circular wound excision model. The basic wound healing process involves proliferation, migration and functioning of fibroblasts and keratinocytes. Therefore, for the in vitro study, human dermal fibroblast (HDF) and human keratinocyte (HaCaT) cell lines were deemed suitable for use. The in vivo study conducted afterwards confirmed the in vitro results, considering some compounds that show promising activity in vitro may be metabolized into inactive metabolites in vivo (Saad and Said, 2011). However, this does not indicate that the in vivo test is more significant compared to the in vitro test. The latter is more suitable for quick, inexpensive screening tests. While both tests have different roles in research, they actually complement each other (Saad and Said, 2011). According to Fronza et al. (2009), scratch assays used in in vitro studies provide first insights on the positive influence of plant preparations in the formation of new tissue. Although studies on the effectiveness of traditional herbs or plants are quite common, they may be a very helpful step in the commercialization

Fig. 2. Asiaticoside structure.

of traditional herbs. Accordingly, it is hoped that this study would promote C. asiatica as an agent to overcome poor wound healing among patients. 2. Materials and methods 2.1. Chemicals and reagents All chemicals and solvents used in this study were of reagent or analytical grade. Only selected chemicals were of HPLC grade. Hexane and methanol were purchased from Merck (Germany), ethanol was obtained from R&M Chemicals (UK) while dichloromethane and ethyl acetate were purchased from Fisher Scientific (UK). HPLC grade methanol and acetonitrile were purchased from Fisher Scientific (UK) and Acros Organics (Belgium), respectively. Dulbecco's modified Eagle medium (Sigma, USA), fetal bovine serum (JR Scientific Co., US), penicillin streptomycin (Gibco, USA), 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazoliumbromide (MTT) (Calbiochem, Germany) and plateletderived growth factor-BB (Calbiochem, Germany) were purchased for in vitro study. Solcoseryl jelly 10% (INVIDA, Singapore), aqueous cream (Pharmaniaga, Malaysia), ketamine as HCl (100 mg/mL) and xylazine as HCl (100 mg/mL) (Ilium, Australia), normal saline solution (Opticare, Malaysia), 37% formaldehyde (Merck, Germany), absolute ethyl alcohol (Fisher Scientific, UK), toluene (Fisher Scientific, UK), xylene (Fisher Scientific, UK), Masson's trichrome staining kit (R&M Chemicals, UK), hematoxylin (Harris Formula) and eosin (Surgipath® Leica Microsystem, USA) were purchased for in vivo study. 2.2. Raw material Whole part of C. asiatica (L.) Urban var. Nyonya was collected at Kampung Pandan 2, Kuantan Pahang, Malaysia. The plant was identified by Dr. Shamsul Khamis, Coordinator of Biodiversity Unit, Institute of

Fig. 1. Centella asiatica.

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

Bioscience, University Putra Malaysia, Serdang Selangor, Malaysia. The specimen was deposited at the Herbarium, Kulliyyah of Pharmacy, International Islamic University Malaysia with voucher number PIIUM 0205. 2.3. Collection and extraction of sample Aerial parts of C. asiatica were collected and dried at 35 °C to 40 °C. The dried herbs were ground to obtain a coarse powder (b0.7 mm). The herb was then macerated with 95% ethanol at 1:3 w/v ratios. After three days, the supernatant was collected and the residue was subjected to the same extraction procedure for another two times. The collected supernatants were pooled and the solvent was evaporated using a Rotavapor to obtain the final solvent-free crude extract. The weight of the final crude extract was recorded to determine yield. The extracts were stored at −20 °C until further analysis. Voucher specimen of the collected herbs was also prepared and the species of the herb identified and authenticated. 2.4. Fractionation Vacuum liquid chromatography (VLC) apparatus was prepared by packing a short column with a glass frit (Buchner funnel) using filter papers and 172.2 g dry silica gel 60 (70–230 mesh ASTM; Merck, Germany) in between. The column was developed by rinsing the silica gel with hexane (500 mL). After the silica gel was free from any solvent, the concentrated crude extract in ethanol was applied directly to the top of the developed column. The mobile phase was added portion by portion in sequence and drawn gently under vacuum to collect each portion separately. The solvent system used for VLC was hexane, hexane–dichloromethane (1:1), dichloromethane, dicholoromethane– ethyl acetate (1:1), ethyl acetate, ethyl acetate–methanol (1:1) and methanol. The weight of each fractionated extract was recorded to determine their yield after they had been evaporated using a Rotavapor to remove solvents. 2.5. Thin layer chromatography The fractionated extracts were characterized qualitatively by TLC with silica gel 60F254 aluminum plate (Merck, Germany). The mobile phase used for analysis was methanol:chloroform (2:3). The TLC plate was swabbed with 10% sulphuric acid in ethanol (Somboonwong et al., 2012). The plate was then heated on a hot plate at a constant temperature of 100 °C for 15 min. Spots that appeared were observed and identified. 2.6. HPLC standardization analysis Both methanol fraction of C. asiatica extract and asiaticoside standard was prepared in phosphate buffer saline (0.01 M, pH 7.4) with the concentration of 100 μg/mL. A stock solution (100 μg/mL) of asiaticoside was diluted to obtain solutions of 50, 30, 20, 10 and 5 μg/mL concentrations. All the samples were filtered through a 0.45 μm filter prior to use. HPLC analysis was performed using chromatographic system (model LC-Prominence series, Shimadzu, Japan) consisting of pump model LC-20AD, an injector valve (Rheodyne) and a photodiode array (PDA) detector. Chromatographic separation was performed using a Hypersil GOLD C18 column (5 μm diameter particle size; 250 × 4.6 mm; Thermo Scientific, US) with HPLC grade solvent of water–acetonitrile–methanol in ratio of 50:26:24 as the mobile phase (Sikareepaisan et al., 2011). The injection volume of sample was set for 20 μL at ambient temperature and flow rate was maintained at 1.0 mL/min. The samples were analyzed at 210 nm. The contents of asiaticoside in extract was determined by comparing peak areas of plant samples with those of standards.

165

2.7. Cytotoxicity assay test A vial of frozen HDF (human dermal fibroblast) primary cell and HaCaT (human epidermal keratinocyte) cell line, obtained from Cell Tissue Technology Sdn. Bhd. and the American Type Culture Collection (ATCC), respectively, was quickly thawed in a 37 °C water bath. The cells were transferred to a T-flask (25 cm3) containing 9 mL of complete media [Dulbecco's modified Eagle medium (DMEM)], and then incubated at 37 °C in a 5% CO2 humidified incubator. The media was changed one day after plating and every two to three days thereafter. The cells were propagated until the density reached 70%–80% confluence before subculture (Canfield, 2011). To check for cell viability, 100 μL of nearly confluent cells were seeded into a 96-well cell plate and incubated (24 h). At the end of the incubation period, the medium was replaced with 100 μL of DMEM in the presence or absence of different concentrations of methanol fraction and the plate was incubated for another 48 h. Then, 25 μL of 5 mg/mL 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazoliumbromide (MTT) solution was added to each well, then further incubated for another 4 h at 37 °C (Steenkamp et al., 2004). Finally, 100 μL of dimethyl sulfoxide (DMSO) was added to each well and the plate was gently shaken for 10 min to dissolve the purple formazan crystals formed. Cell viability was evaluated using a microplate reader at 570 nm with reference wavelength of 630 nm. The absorbance readings of the test determined the suitable dosage to be used in the scratch assay test. The test was performed in triplicate. 2.8. Scratch assay test The spreading and migration capabilities of HDF and HaCaT cells were assessed using scratch wound assay, which measures the expansion of a cell population on surfaces as described by Fronza et al. (2009). The cells were seeded into 12-well tissue culture dishes in DMEM containing 10% FBS (fetal bovine serum) and 2% penicillin and streptomycin. After the cells had nearly formed a confluent cell monolayer, a linear wound was generated in the monolayer using a sterile 100 μL plastic pipette tip. Any cellular debris was removed by washing the wells with phosphate buffer saline (PBS). The medium (DMEM) used consisted of either platelet-derived growth factor (PGDF; 0.002 μg/mL) (as positive control), asiaticoside (125 μM) or the methanol fraction of C. asiatica extract (0.2 μg/mL, 6 μg/mL and 100 μg/mL). The cells were then incubated for 40 h or 16 h at 37 °C with 5% CO2. The scratched cell layers incubated under the different conditions were then photographed to estimate relative cell migration. The data was analyzed using NIS-Elements Version 4.3 and CapturePro Version 2.5 for Progres® microscope camera from Jenoptik Laser, Optik, Systeme GmbH. The experiment was performed in triplicate. 2.9. Animal care Animal care and handling was carried out as described by Hemmati et al. (2002). Purchased rabbits (New Zealand white albino) of either sex (4 males and 3 females) with weight of 1.8–2.1 kg were used during the study. The animals were supplied by Ladang Ternakan Sharsha, Kuantan, Pahang, Malaysia. The animals were housed individually in aluminum cages throughout the research. They were fed with commercial pellet diet supplemented with fresh vegetables and water ad libitum. All the animals were kept in a holding room illuminated with 12 h light/dark cycles at room temperature of 23 ± 2 °C with relative humidity of 45% to 55%. Study on the rodents were begun after approval from the Institutional Animal Care and Use Committee (IACUC), International Islamic University Malaysia and after an acclimatization period of at least five days.

166

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

2.10. Acute dermal irritation Acute dermal irritation testing of the prepared treatment was carried out to determine the suitable dose to be used for in vivo wound healing activity test. OECD guideline 404 was followed for the study (OECD, 2002). Approximately 24 h before the test, fur was removed by closely clipping the dorsal area of the animal's trunk. Only animals with a healthy intact epidermis by gross observation were selected for the study. The in vivo test was initially performed with the highest concentration of extract (w/w) using one animal. The test substance was moistened with the smallest amount of water. To ensure skin contact, the sufficiently moist test substance was applied with a gauze patch using non-irritating tape. To prevent the animals from interfering with the patches, the trunk of each rabbit was wrapped in an elasticated corset and the animals were returned to their cages for the duration of the exposure period (4 h). Four hours after application, the corset and patches were removed from each animal and any residual test material was removed by gentle swabbing with cotton wool soaked in distilled water. Approximately one hour following the removal of the patches, and 24, 48 and 72 h later, the test sites were examined for evidence of primary irritation and scored according to the scale used in OECD guideline 404. Results were interpreted using calculation of primary irritation index and grading of irritancy potential using the Draize scheme. The scores for erythema and oedema at the 24- and 72-h readings were totaled for the three test rabbits (12 values) and this total was divided by six to give the primary irritation index of the test material. The test material was classified according to the scheme devised by Draize (1959). If irreversible alteration of the dermal tissue is noted in any rabbit, as judged by the study director, which included ulceration and clear necrosis or signs of scar tissue, the test material was classified as corrosive to rabbit skin. Classification according to Draize scheme may, therefore, not be applicable. If a corrosive effect is not observed in the initial test, the irritant or negative response should be confirmed using up to two additional animals, each with one patch, for an exposure period of four hours.

size was observed, measured and photographed using dermal lab. Percentage of wound contraction was calculated as follows: %Wound contraction ¼ ½ðW 0 –W t Þ=W 0   100 where W0 is initial wound size and Wt is specific day wound size. Period of epithelisation was also recorded. Period of epithelisation was recorded as the days needed to reach the end point of complete epithelisation where the scab falls without leaving any raw wound behind (Shenoy et al., 2011; Fikru et al., 2012). 2.13. Hydroxyproline content Hydroxyproline content was determined to quantify collagen synthesis in healed wound tissue. The estimation was done following the methods reported by Shaaban et al. (2014), with some modifications. The tissues (before and after treatment) were excised and dried in a hot air oven at 60 °C–70 °C to constant weight (0.14–1 mg). The excised tissues were then hydrolysed in a sealed glass tube using 6 N HCl (0.5 mL) at 100 °C for 24 h. The hydrolysate was neutralized to pH 7.0 (6 N NaOH, 0.5 mL) and centrifuged for 15 min (40,000 rpm). A total of 40 μL of supernatant was subjected to 25 μL of chloramine-T solution [1 part of 7% chloramine T and 4 parts of acetate buffer (pH 6, 272 g sodium acetate, 50 mL of glacial acetic acid and dissolved in H20 to a final volume of 1 L)] for 20 min. Due to the large number of samples used, 96-well plates were used. Thereafter, 150 μL of Ehrlich's solution was added. The Ehrlich's solution was prepared by dissolving 2 g of 4-dimethylamino-benzaldehyde in 3 mL of 60% perchloric acid and then mixed with 9 mL of isopropanol. The final mixture was incubated at 60 °C for 35 min and then at room temperature for 10 min, and absorbance was measured at 530 nm using a microplate reader. As reported by Wang et al. (2011), hydroxyproline content was determined using a standard curve prepared using various dilutions of 1 mg/mL stock solution of hydroxyproline with final concentrations ranging from 2 to 200 μg/mL (typically 2, 10, 25, 50, 100 and 200 μg/mL). The values of hydroxyproline content were then expressed as μg hydroxyproline/mg skin.

2.11. Preparation of topical formulation

2.14. Histopathological studies

The methanol fraction of C. asiatica extract (MF) was formulated in aqueous cream at 40%, 10% and 2.5% concentration (w/w). The formulation with a concentration of 2.5% (w/w) was chosen since it was the lowest concentration where the pH found to be acidic as same as its original extract. The other concentrations were increments of that concentration (four times).

The regenerated tissues were observed qualitatively under light microscope for keratinisation, epithelisation, inflammation, fibroblastic proliferation and neovascularisation. The excised wound tissues were immediately fixed in 10% neutral buffered formalin (24 h) and dehydrated after dissection. Then, they were embedded in paraffin wax to ensure complete dehydration. Serial sections of paraffinembedded tissues of 6 μm thickness were cut using microtome and stained with hematoxylin and eosin staining and Masson's trichrome staining.

2.12. In vivo wound healing activity (excision wound) Each rabbit received all the treatments and were their own control. Treatment 1: Control (no treatment) Treatment 2: Treated with MF40% Treatment 3: Treated with MF10% Treatment 4: Treated with MF2.5% Treatment 5: Treated with solcoseryl jelly 10% Treatment 6: Control (blank aqueous cream) The hair on the neck region was shaved 24 h before the test. The rabbits were generally anesthetised with subcutaneous injection of ketamine:xylazine (0.4 mL:0.1 mL, 40 mg/kg:10 mg/kg). Ethanol (70%) was used as an antiseptic for the shaved region before introducing the wound. A circular excision wound was made using a biopsy punch (4 mm in diameter) as mentioned by Shailajan et al. (2011). The wounding day was counted as day zero. The wounds were treated topically twice daily (20 mg) until the wounds completely healed. Wound

2.15. Statistical analysis Statistical evaluation was carried out with IBM SPSS statistics version 20 and with Microsoft Office Excel 2007. Significant differences between the treated groups and the control were determined by one-way ANOVA using Kruskal–Wallis test, at a significance level of p b 0.05. 3. Results 3.1. Physical properties and yield of C. asiatica extracts The ethanolic extract of C. asiatica is sticky and dark in color. The yield of crude extract was 4.63%. Among all the seven fractions eluted, only the MF contained asiaticoside, based on thin layer chromatography (TLC) tests. The MF was a dark yellow powder. Asiaticoside

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

167

Fig. 3. Cell viability of HDF.* The mean difference is significant compared to controls (P b 0.05).

quantification in the MF was estimated from the standard calibration curve, which was constructed from the chromatograms of different concentrations of asiaticoside standards using HPLC. The percentage of asiaticoside in MF was quantified to be around 2.4%. 3.2. Cytotoxicity assay test Based on the cytotoxicity assay test using different concentrations of MF, the IC50was not found within the range of 0.19–100 μg/mL. All the treatments showed cell viability percentage above 90%. The result of cell viability on HDF and HaCaT is shown in Figs. 3 and 4, respectively. However, based on the cell viability results, three concentrations (0.2, 6 and 100 μg/mL) were chosen for the scratch assay. The reason for choosing 100 μg/mL of MF as the highest concentration for the scratch assay was because, at this concentration, the percentage of cell viability was highest and the mean differences showed significant effects compared to the control (p b 0.05). The lowest concentration of 0.19 μg/mL also resulted in significant mean differences compared to the control. However, instead of using 0.19 μg/mL, a concentration of 0.2 μg/mL was selected after rounding that value. The concentration of 6 μg/mL was chosen due to the abnormal trends observed between concentrations of 0.19 and 100 μg/mL, where there was less decrement in cell viability. It was hypothesized that at this concentration, varying results would be observed in the scratch assay test.

3.3.1. Scratch assay test of human dermal fibroblast As mentioned before, fibroblasts play a big role in wound repair activity. Hence, the result for human dermal fibroblast (HDF) scratch assay was critically analyzed. The migration rate of cells tested for 40 h are shown in Fig. 5. All the treatments used resulted in higher cell migration rates compared to untreated cells (control) or PDGF-treated cells (positive control) except for MF at 6 μg/mL concentration. Compared to positive controls, MF at concentrations of 100 and 0.2 μg/mL showed significant migration rates (p b 0.05). Fig. 6 shows the healing percentage of the scratched cells after 40 h. The trend of the graph was the same as previous tests of cell viability, the MF at concentrations of 100 and 0.2 μg/mL exerted significant healing percentages compared to positive control (p b 0.05). Fig. 7 shows the migration rate of the HDF cells with each treatment against time. Treatment from untreated control, positive control, MF6 and MF0.2 showed the same pattern while the other two, which were MF100 and asiaticoside, showed different trends of migration rate even to each other. 3.3.2. Scratch assay test of human epidermal keratinocyte Fig. 8 illustrates the relative cell migration of human epidermal keratinocyte (HaCaT) from the scratch assay test done. The results are almost the same from the previous scratch assay carried out on HDF. MF100 and MF0.2 managed to significantly affect cell migration (p b 0.05) compared to PDGF-treated cells (positive control). However, MF6 still resulted in higher relative cell migration compared to the positive control.

3.3. Scratch assay test

3.4. Acute dermal irritation test

Using the scratch assay, cell migration in response to an artificial injury on HDF and HaCaT cells was observed under phase-contrast microscope and bright-field microscope, respectively.

3.4.1. Preparation of sample The MF (paste form) was applied to the test site. The absorption of the test material was not determined. The pH of the test material was

Fig. 4. Cell viability of HaCaT.* The mean difference is significant compared to controls (P b 0.05).

168

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

Fig. 5. Migration rate of HDF for forty hours.* The mean difference is significant compared to positive controls (P b 0.05).

determined prior to commencement of the study and was found to be acidic (pH 4.73). 3.4.2. Changes in rabbit body weight Changes in body weight of the tested rabbits were determined during the test. Any sudden changes indicated signs of stress or other abnormalities. From the data, there were no abrupt body weight changes in all tested rabbits. 3.4.3. Individual skin reactions Indication of acute dermal irritation on the tested treatment was decided based on the calculation of primary irritation index and grading of irritancy potential (Table 1). Based on the calculations made, the MF extract did not cause irritation to rabbit skin, even at concentrated doses. Since the dosage used in further experiments were more dilute than the tested doses, the extracts were deemed suitable and safe to use.

3.5.2. Percentage of wound contraction All the treatments used significantly (p b 0.05) increased the percentage of wound contraction compared to untreated wounds. Majority of the wounds fully contracted on the ninth day (Table 2). Table 3 shows the mean wound contraction values for the treatments used.

3.5.3. Period of epithelisation Untreated wounds represent normal healing. In the present study, normal healing completed wound contraction at 10.75 ± 0.50 days while wounds treated with vehicle control achieved complete wound contraction at 10 ± 0.82 days. Wounds treated with MF extracts showed comparable results to the positive control and significantly (p b 0.05) better results compared to the untreated control and vehicle control groups (Table 4).

3.5. In vivo wound healing activity (excision wound)

3.5.4. Hydroxyproline content There were no significant differences in hydroxyproline content between all the treatments. However, all of them showed increments of more than 50% compared to old tissue excised (Table 5).

3.5.1. Changes in rabbit body weight Changes in body weight of rabbits for the in vivo test were determined. There were no sudden changes in rabbit weight that would indicate signs of stress or other abnormalities. From the data collected, there were no abrupt body weight changes in all tested rabbits.

3.5.5. Histopathological observation Fig. 9 shows the gross observation of one of the wounds made and also the stained normal tissue excised to produce that wound. It was observed that the normal tissue excised included both the epidermis

Fig. 6. Healing percentage of HDF.* The mean difference is significant compared to positive controls (P b 0.05).

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

169

Fig. 7. Migration rate of HDF.

and dermis layer. Hence, the wounds made were categorized as severe injuries. Fig. 10 shows the observation of wound area and tissues excised from the wound area on day 10 after the treatment. All the tissues lost their dermal appendages. There was a marked formation of keratin and epithelial layer in all the tissues. The granulation tissues also contained blood vessels (reddish), fibroblastic cells (bluish violet in H&E stain, pinkish in trichrome stain), inflammatory cells (purplish red in H&E stain, blackish red in trichrome stain) and collagen (pinkish pale in H&E stain, bluish in trichrome stain). Minimal mature collagen was observed in tissues that were untreated and the vehicle control group. Fibrinoid necrosis was absent in all of the treated tissues. The summary of the tissue observation is tabulated in Table 6. 4. Discussion Crude extracts contain all of the plant constituents in various amounts. The efficacy of crude extracts may be due to the synergistic action of its combined phytoconstituents. However, since each batch of extraction may result in different amounts of phytoconstituents, a product with consistent chemistry quality is not reproducible. On the other hand, standardized extracts contain one or more components present in specific and guaranteed amounts. With that, the pharmacological activity of the extract is traceable and measurable. Therefore,

the present study focused on evaluating the bioactivity of rich fractionated extracts of asiaticoside compound. TLC tests and HPLC standardization analyses indicated that the sample most suitable for use for further experiments was the methanol fraction (MF) containing standardized asiaticoside concentration of 2.4%. Even the results of the investigations could not conclude asiaticoside as the primary phytochemical responsible for that particular event, the results would surely contribute new information to this field of study. The basic wound healing process involves proliferation, migration and functioning of fibroblasts and keratinocytes. Therefore, for the in vitro study, human dermal fibroblast (HDF) and human keratinocyte (HaCaT) cell lines were chosen as the most suitable cell lines to be used. Platelet-derived growth factor (PDGF) was used as a positive control in the scratch assay test. PDGF can increase the formation of granulation tissue that is important in improving the rate of healing. Based on research conducted by Fronza et al. (2009), low concentrations of 0.5, 1 and 2 ng/mL of PDGF resulted in good correlation between dose and increase in fibroblast number compared to the control. In contrast, higher concentrations of 4 and 15 ng/mL of PDGF resulted in lesser migration. Therefore, 2 ng/mL of PDGF was chosen as the concentration for positive controls. In the same assay, asiaticoside was used at a concentration of 125 μM, which is equal to 119.89 μg/mL. This concentration was chosen because previous research conducted by Lee et al. (2012) has shown that this is the minimum concentration of asiaticoside where

Fig. 8. Relative cell migration of HaCaT.* The mean different is significant compared to positive controls (P b 0.05).

170

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

Table 1 Table of acute dermal irritation calculation. Skin reaction

Observation time (h)

Erythema/eschar formation

1 24 48 72 Oedema formation 1 24 48 72 Sum of 24 and 72 h reading: 0.33 Primary irritation index (s/6): 0.33/6 Classification: non-irritant

Table 3 Percentage of wound contraction. Individual scores (rabbit no. and sex)

Total

Female 1

Female 3

0 0 0 1 0 0 0 0

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

0 0 0 0.33 0 0 0 0

the number of treated fibroblast cells started increasing steadily. From the same paper, Lee et al. (2012) found that asiaticoside did not influence the growth rate of keratinocytes. Hence, the current study chose to use asiaticoside in the HDF scratch assay study only. In the cell viability tests conducted, the results of the treatments used on both HDF and HaCaT cells showed almost the same trend. A slight reduction in cell viability between concentrations of 0.19 and 100 μg/mL was observed. Additionally, significant changes in cell viability percentage at the highest concentration of 100 μg/mL were observed. The test was then repeated to a concentration of 1000 μg/mL. The IC50 of the results could not be determined in those ranges (data not shown). Considering that 1000 μg/mL concentration is quite high for cells in a 96-well plate, it is concluded that the toxicity effect of MF is quite low. The rate of migration refers to speed of cell migration within certain time duration. Normally, the cells migrate towards empty spaces. However, the rate of migration is affected by the environment (Liang et al., 2007). The results of the migration test indicate that the MF increased the rate of cell migration compared to untreated control. The other treatments also sped up the cell migration rate better than PDGF (positive control), except for MF6. Since asiaticoside did speed up the migration rate better than PDGF, it could be assumed that another

Table 2 Percentage of wound contraction calculation. Treatment

Male 1

Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream) Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream) Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream) Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream)

Male 2

Male 3

Male 4

Wound size (mm) Day 0

Day 9

5 6 6 6 5 5 5 4 4 5 4 5 5 4 4 4 4 4 5 5 5 4.5 4 5

1 0.5 0.5 0.5 1 0.5 0.5 0 0 0 0 0 1 0 0 0 0 0.5 1 0.5 0 0 0 0.5

% Wound contraction Male 1 Male 2 Male 3 Male 4

Male 3

Rabbit no. and sex

Treatment

Wound contraction (%)

80 91.7 91.7 91.7 80 90 90 100 100 100 100 100 80 100 100 100 100 87.5 80 90 100 100 100 90

Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream)

80 91.7 91.7 91.7 80 90

90 100 100 100 100 100

80 100 100 100 100 87.5

80 90 100 100 100 90

Mean of % wound contraction (SD) 82.5 95.4 97.9 97.9 95.0 91.9

± ± ± ± ± ±

0.05 0.05* 0.04* 0.04* 0.10* 0.06*

*p b 0.05 when compared to untreated control.

compound was present in the MF that may have also contributed to the increased rate in cell migration. Percentage of healing was calculated by measuring the area covered in HDF in forty hours. The results supported previous results on migration rate. The results indicate that untreated cells only healed the wounds to about 50% while other treatments healed the wounds better. Increase in healing by more than half in wounds treated using the extracts showed promising results for use as a wound healer. The measurement readings were influenced by the initial gap size of the cells. The narrower the gap of the cell, the better the possibility for 100% healing. Repeated tests were carried out using the same treatments (data not shown). As a consequence, the result of percentage of healing above was plotted only from the same batch to reduce standard deviation values. Although the values do not accurately describe the effect, the same trend present in replicates of the test provides an insight into the effect displayed. Calculating the migration rate for the whole study and for certain period intervals helps define cell migration activity at different points. Evaluating the migration rate at certain period intervals helped provide a clearer picture on how treatment affected cell migration. Migration rate in untreated cells was high for the first three periods and decreased in the later hours. This is normal because the fibroblast cells at the confluent state compete between themselves to individually migrate to the empty space and in due time, the cells are released from the crowded space and are able to move individually (Liang et al., 2007). Reduced migration rates during the later hours is common either due to a decrease in the supplement or because the wound gap narrows, causing some cells to sense a lack of empty space, which leads to less migration. Cells treated with PDGF, MF6 and MF0.2 showed almost the same pattern of migration rate. The only difference was the rate of migration at certain time intervals that resulted in different rates of cell migration at the end of the time period (40 h). The data used to calculate the overall rate of migration (Fig. 5) was based on the area at the end of the total duration, set at forty hours. An increase in the rate of migration with increasing time may conclude that the treatment contains a component that indirectly influences cell migration. A higher rate of cell migration during the first hours of treatment that decreases with time may be due to the presence of a compound that directly influences the rate of cell migration. The

Table 4 Period of epithelisation. Treatment

Period of epithelisation (day) Male 1 Male 2 Male 3 Male 4

Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream)

11 9 9 9 10 11

10 8 8 8 9 9

11 8 9 8 7 10

11 9 9 9 8 10

Mean of epithelisation period (SD) 10.75 ± 0.50 8.50 ± 0.58** 8.75 ± 0.50** 8.50 ± 0.58** 8.50 ± 1.29** 10.00 ± 0.82

⁎p b 0.05 when compared to untreated control. ⁎⁎p b 0.05 when compared to both untreated control and vehicle control.

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174 Table 5 Hydroxyproline content of treated animal tissue. Treatment

Total (μg hydroxyproline/mg skin)

Control (no treatment) MF40% MF10% MF2.5% Solcoseryl jelly 10% Control (blank aqueous cream)

Day 0

Day 10

8.49 6.74 6.09 6.25 6.79 8.94

24.86 20.77 24.10 17.43 21.05 25.05

± 1.8618 ± 0.6462 ± 1.2303 ± 0.5635 ± 1.9106 ± 0.8342

± 7.5665 ± 3.5168 ± 3.4773 ± 2.7617 ± 6.4142 ± 4.4454

decreasing effect afterwards may be due to a decrease in levels of that compound (Wang et al., 2011). Hence, based on the observed trends, asiaticoside may directly induce cell migration. MF100 also showed higher migration rate during the first hours, but migration rate still increased afterwards. MF100 might contain components that directly and indirectly induce cell migration. Comparing the results in Figs. 5 and 7, it is deduced that although there are other components that might indirectly affect the cell migration rate of MF100, asiaticoside remains the direct influencer of cell migration, giving high rate of cell migration, as supported by data from MF6 and MF0.2. This test was conducted using a live imaging confocal microscope. Mathematically, we could not measure the cell proliferation in this study without using any markers. However, from observations of the recorded video, the cells not only migrate but they also actively proliferated to cover up the empty space. Hence, migration rates in this study were not only due to cell movement but also due to cell proliferation. Relative cell migration and healing were both measured in percentage values. However, both of them refer to different parameters, as healing percentage defines the whole area of the gap while relative cell migration only measures the gap distance on the one particular side of the cell gap. Fibroblasts migrate individually and may disperse in different directions. Upon contact, keratinocytes migrate individually and also as a sheet, making their movement more structured and arranged (Clark and Henson, 1988). Hence, relative cell migration of keratinocytes involving measurement of gap distance is considered reliable data. Relative cell migration of HaCaT treated with MF100, MF6 and MF0.2 showed almost the same trend as the cell viability tests. MF100 and MF0.2 resulted in higher relative cell migration compared to PDGF, which is the positive control used in this study. Hence, the overall

171

results of the in vitro study support the use of our extracts as a promising wound healer. Before conducting detailed excision wound test on the animals, the sample prepared needed to be analyzed in order to collect detailed information on the sample and determine the suitable dosage to be used for the experiment. For the present study, pH of the sample was measured. It has been proven that the role of wound dressing pH is important during the healing process, especially in chronic wounds. Prolonged chemical acidification of the wound dressing could increase tissue oxygen availability through oxygen dissociation and reduce the histotoxicity of the bacterial end product, thus stimulating the healing process. In addition, chronic wound fluids contain elevated protease levels that could hinder the healing process (Houghton et al., 2005). Protease activity is extremely pH-sensitive. It is highest between pH 7 and pH 8, and decreases rapidly in the presence of acidity (Al-waili et al., 2011). The pH of wounds is usually neutral to alkaline whereas the pH of normal skin is acidic (pH 5.5). Hence, tuning protease activity may accelerate wound healing. The pH of the MF extract used was around 4.7 to 4.8, which is acidic. This pH may potentially keep the wound in acidic conditions, which induces fibroblast proliferation in order to stimulate the healing process. The MF extract was determined as safe for use, as concentrated extracts did not cause any dermal irritation. In vivo studies are quite challenging considering they involve the maintenance of living creatures. Any stress or disease to the living creature would affect the results of the experiment. Besides keeping the animal houses clean and comfortable and feeding the animals correctly, observing their behavior and physical abnormality is also important. Since this present study did not intervene with the rabbits' normal diet, only their body weights were recorded as an indication of their clinical signs. Throughout the present study, no sudden changes to any of the rabbits used were observed. Their behavior remained normal except for female rabbits in which they were excluded from the in vivo wound healing study since the stress observed among them. Wound contraction was recorded every three days and stopped once epithelisation was achieved. There was no significant contraction from day one until day six. Wound contraction increased rapidly on the following days. Wound contraction is defined as the reduction of the unhealed area. The more the wound reduction, the better the efficacy of the medication used. The results obtained showed comparable efficacy between the commercial product (solcoseryl jelly 10%) and the MF

Fig. 9. Skin and tissue observation at day 0. F = fibroblast; C = collagen; and HF = hair follicle; magnification: 100×.

172

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

Fig. 10. Skin and tissue observation at day 10. Magnification: 100×.

extract. The reduction of wound treated with MF extract was significantly higher compared to the untreated wound. There was no dosedependent effect at different concentrations of MF extract used.

However, this data may have some limitations since the measurement of wound size was recorded by day and not by viewing the actual reduction event. Thus, although wound contraction was recorded as zero on

Table 6 Summary of tissue observation at day 10. Treatment

Keratinisation Epithelisation Inflammatory cell infiltration Fibroblastic proliferation Neovascularisation/angiogenesis Fibrinoid necrosis

Control (no treatment) 40% MF 10% MF 2.5% MF Solcoseryl jelly 10% Control (blank aqueous cream)

+++ +++ +++ +++ +++ +++

+++ +++ +++ +++ +++ +++

+++ ++ ++ +++ + +++

+, slight; ++, moderate; +++, marked; ++++, extensive; −, absent.

++ +++ +++ +++ ++++ ++

+++ ++ +++ +++ + +++

− − − − − −

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

day nine, this does not indicate that wound size was zero on that particular day. The wound size may have been zero on previous days. Data for period of epithelisation may explain the wound condition better. Nevertheless, it was observed that during the healing period, no indication of infection to the wounds which could elongate the time of wound contraction was present. Epithelisation period is the period where there is complete formation of epithelium over a denuded surface. If epithelisation period is slow, a scar will form over many weeks, even months. In contrast, faster epithelisation periods help the regeneration process of healing. Compared to the untreated and vehicle control groups, the treatment group showed a significantly faster period of epithelisation. Among the treatments, only 10% MF extract showed slightly slower periods of epithelisation. Otherwise, the other two concentrations of MF extract showed comparable results to the positive control. Although this data explains the wound condition better compared to the previous data of wound contraction, the data still has certain limitations. The limitation concerns the method for measurement of duration for epithelisation. In the study, the period of epithelisation was recorded just by observing the end size of the wound regardless of the initial size of the wound at day zero. The smaller the size of the initial wound, the higher the possibility for the epithelium to form completely. Hence, histopathology results of the tissue would better explain the wound condition. Concentration of hyroxyproline is a measure of concentration of collagen. Collagen is a constituent of growing cells. Higher concentrations of hydroxyproline indicates faster rate of wound healing (Agarwal et al., 2009). Although both untreated tissue and vehicle-treated tissue showed higher hydroxyproline content, the increment was only two times higher compared to the initial tissue. In contrast, the other treatments increased hydroxyproline content up to three times their initial values except with treatment using 2.5% of MF extract, which showed similar results to the control groups. Estimation of hyroxyproline content is a complicated and lengthy work. During the investigation process, each step was ensured to be handled carefully to get accurate results. Histopathology facilitates the study of the tissue under microscopic examination. Normal tissue stained with H&E and trichrome indicates that normal tissue has a thin epidermis layer and dermis layer rich in appendages such as hair follicles and sebaceous glands. Since the normal tissue excised contained both the epidermis and dermis, the wound was categorized as severe injury (Walter and Israel, 1996). The tissue undergoes a repairing process consisting of connective tissue deposition, which commonly occurs when the supporting structure of the tissue is severely damaged. With this information, it was predicted that tissue healing should occur. The new regenerated tissue would appear without any skin appendages. At the beginning of the healing process, the tissue is expected to show a blood vessel appearance due to either neovascularisation or angiogenesis and also infiltration of the inflammatory cells in the area (Deonarine et al., 2007). The later stages of healing should show elevated presence of fibroblasts and collagen, as well as keratinisation and epithelisation. Fibrinoid necrosis is a form of tissue death that occurs due to stress, such as excessive inflammation and so on. None of the treatment regimens induced necrosis on the wound areas. It was concluded that none of the treatments would cause tissue death. From the tissue observation in Table 6, positive control (Solcoseryl jelly) showed best results among all the treatments as only slight appearance of blood vessel and infiltration of inflammatory cell was observed. Treatment using 40% MF extract was second best, followed by 10% MF extract and 2.5% MF extract. By observing the histopathology data, the MF extract showed a dose-dependent effect as a wound healer. During the final phase of the wound healing process, which is remodeling or maturation phase, type III collagen is gradually degraded by the stronger type I collagen (Hashim et al., 2011). Thus, before the end of the healing process, instead of thinning of the epidermis size due to differentiation, collagen type I will be present. The easiest way to spot type I collagen is by looking at the collagen structure. Type I collagen is

173

organized horizontally (Fitzpatrick and Morelli, 2011). This structure can be spotted mostly in tissue treated with positive control and the least on tissues of untreated and vehicle control group. The present in vivo study shows reproducible wound healing potential in the methanol fraction of C. asiatica extract. 5. Conclusions In conclusion, recent findings suggest that methanol fraction of C. asiatica extract (MF) possess wound healing properties. The standardized 2.4% asiaticoside extract (MF) showed cell migration ability in cells with fibroblast and keratinocyte morphologies and enhanced the proliferation and remodeling process of healing. With the said results, the ethnopharmacological approach in selecting this plant for study is useful for current knowledge contribution and future research needs. Declaration of conflict of interest The authors declare no conflict of interest. Acknowledgements The authors are thankful to the Ministry of Higher Education, Malaysia and International Islamic University Malaysia, Malaysia for providing financial support and the facility to carry out this research. Financial support was provided under MyRA Incentive Research Grant Scheme (MIRGS) with project number MIRGS13-01-001-0006. Highest gratitude is also expressed to the Tissue Engineering Centre, UKM Medical Centre, Malaysia for their permission to use their facilities and also their assistance in performing the research work. References Agarwal, P.K., Singh, a., Gaurav, K., Goel, S., Khanna, H.D., Goel, R.K., 2009. Evaluation of wound healing activity of extracts of plantain banana (Musa sapientum var. Paradisiaca) in rats. Indian Journal of Experimental Biology 47, 32–40. Al-waili, N.S., Salom, K., Al-ghamdi, A.A., 2011. Honey for wound healing, ulcers, and burns; data supporting its use in clinical practice. Scientific World Journal 766–787. http://dx.doi.org/10.1100/tsw.2011.78. Canfield, T., 2011. SOP: Propagation of mouse NIH-3T3 embryonic fibroblast Retrieved May 12, 2013. http://encodeproject.org/ENCODE/protocols/cell/mouse/NIH-3T3_ Stam_protocol.pdf. Clark, R.A.F., Henson, P.M., 1988. Overview and General Considerations of Wound Repair. The Molecular and Cellular Biology of Wound Repair. Plenum Press, New York, pp. 3–22. Deonarine, K., Panelli, M.C., Stashower, M.E., Jin, P., Smith, K., Slade, H.B., Norwood, C., Wang, E., Marincola, F.M., Stroncek, D.F., 2007. Gene expression profiling of cutaneous wound healing. Journal of Translational Medicine. http://dx.doi.org/10.1186/14795876-5-11. Draize, J.H., 1959. Appraisal of the Safety of Chemicals in Foods, Drugs and Cosmetics. Dermal Toxicity. Association of Food and Drug Officials of the United States, Austin, Texas, pp. 46–59. Fikru, A., Makonnen, E., Eguale, T., Debella, A., Mekonnen, G.A., 2012. Evaluation of in vivo wound healing activity of methanol extract of Achyranthes aspera L. Journal of Ethnopharmacology 143, 469–474. Fitzpatrick, J.E., Morelli, J.G., 2011. Structure and Function of the Skin. Dermatology Secrets Plus, 5th ed. Mosby Inc., Missouri, p. 14. Fronza, M., Heinzmann, B., Hamburger, M., Laufer, S., Merfort, I., 2009. Determination of the wound healing effect of Calendula extracts using the scratch assay with 3T3 fibroblasts. Journal of Ethnopharmacology 126, 463–467. http://dx.doi.org/10.1016/j.jep. 2009.09.014. Hashim, P., Sidek, H., Helan, M.H.M., Aidawati, S., Palanisamy, U.D., Ilham, M., 2011. Triterpene composition and bioactivities of Centella asiatica. Molecules 16, 1310–1322. http://dx.doi.org/10.3390/molecules16021310. Hemmati, A.A., Arzi, A., Amin, M., 2002. Effect of Achillea millefolium extract in wound healing of rabbit. Journal of Natural Remedies 2, 164–167. Houghton, P.J., Hylands, P.J., Mensah, A.Y., Hensel, A., Deters, A.M., 2005. In vitro tests and ethnopharmacological investigations: wound healing as an example. Journal of Ethnopharmacology 100, 100–107. http://dx.doi.org/10.1016/j.jep.2005.07.001. Idrus, R.B.H., Chowdhury, S.R., Manan, N.A.B.A., Fong, O.S., Adenan, M.I., Saim, A.B., 2012. Aqueous extract of Centella asiatica promotes corneal epithelium wound healing in vitro. Journal of Ethnopharmacology 140, 333–338. http://dx.doi.org/10.1016/j. jep.2012.01.023. Jabatan Perhutanan Semenanjung Malaysia, n.d. Pegaga [WWW Document]. URL http://www.forestry.gov.my/herba/pegaga_en.pdf

174

H.A. Azis et al. / South African Journal of Botany 108 (2017) 163–174

Lee, J.H., Kim, H.L., Lee, M.H., You, K.E., Kwon, B.J., Seo, H.J., Park, J.C., 2012. Asiaticoside enhances normal human skin cell migration, attachment and growth in vitro wound healing model. Phytomedicine 19, 1223–1227. http://dx.doi.org/10.1016/j.phymed. 2012.08.002. Liang, C.-C., Park, A.Y., Guan, J.-L., 2007. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nature Protocols 2, 329–333. http://dx.doi.org/10.1038/nprot.2007.30. OECD, 2002. OECD Guideline for Testing of Chemicals: Acute Dermal Irritation/Corrosion 404 [WWW document] URL http://ihcp.jrc.ec.europa.eu/our_labs/eurl-ecvam/ validation-regulatory-acceptance/docs-skin-irritation-1/DOC2_OECD-TG-404.pdf. Saad, B., Said, O., 2011. Safety Assessment, in: Greco–Arab and Islamic Herbal Medicine: Traditional System, Ethics, Safety, Efficacy, and Regulatory Issues. John Wiley & Sons, Inc, Hoboken, New Jersey, p. 286. Shaaban, A.A., Shaker, M.E., Zalata, K.R., El-Kashef, H.A., Ibrahim, T.M., 2014. Modulation of carbon tetrachloride-induced hepatic oxidative stress, injury and fibrosis by olmesartan and omega-3. Chemico-Biological Interactions 207, 81–91. http://dx.doi. org/10.1016/j.cbi.2013.10.008. Shailajan, S., Menon, S., Pednekar, S., Singh, A., 2011. Wound healing efficacy of Jatyadi Taila: in vivo evaluation in rat using excision wound model. Journal of Ethnopharmacology 138, 99–104. http://dx.doi.org/10.1016/j.jep.2011.08.050. Shenoy, R.R., Sudheendra, A.T., Nayak, P.G., Paul, P., Kutty, N.G., Rao, C.M., 2011. Normal and delayed wound healing is improved by sesamol, an active constituent of Sesamum indicum (L.) in albino rats. Journal of Ethnopharmacology 133, 608–612. http://dx.doi.org/10.1016/j.jep.2010.10.045. Shukla, A., Rasik, A.M., Dhawan, B.N., 1999a. Asiaticoside-induced elevation of antioxidant levels in healing wounds. Phytotherapy Research 13, 50–54.

Shukla, A., Rasik, A.M., Jain, G.K., Shankar, R., Kulshrestha, D.K., 1999b. In vitro and in vivo wound healing activity of asiaticoside isolated from Centella asiatica. Journal of Ethnopharmacology 65, 1–11. Sikareepaisan, P., Ruktanonchai, U., Supaphol, P., 2011. Preparation and characterization of asiaticoside-loaded alginate films and their potential for use as effectual wound dressings. Carbohydrate Polymers 83, 1457–1469. http://dx.doi.org/10.1016/j.carbpol. 2010.09.048. Somboonwong, J., Kankaisre, M., Tantisira, B., Tantisira, M.H., 2012. Wound healing activities of different extracts of Centella asiatica in incision and burn wound models: an experimental animal study. BMC Complementary and Alternative Medicine 12, 103. http://dx.doi.org/10.1186/1472-6882-12-103. Steenkamp, V., Mathivha, E., Gouws, M.C., van Rensburg, C.E.J., 2004. Studies on antibacterial, antioxidant and fibroblast growth stimulation of wound healing remedies from South Africa. Journal of Ethnopharmacology 95, 353–357. http://dx.doi.org/10. 1016/j.jep.2004.08.020. Taemchuay, D., Rukkwamsuk, T., Sakpuaram, T., Ruangwises, N., 2009. Antibacterial activity of crude extracts of Centella asiatica against Staphylococcus aureus in bovine mastitis. Kasetsart Veterinarians 19, 119–128. Thakur, R., Jain, N., Pathak, R., Sandhu, S.S., 2011. Practices in wound healing studies of plants. Evidence-Based Complementary and Alternative Medicine 1–17. http://dx. doi.org/10.1155/2011/438056. Walter, J.B., Israel, M.S., 1996. Wound Healing. General Pathology. W.B. Saunders Company, New York, pp. 165–180. Wang, J.-P., Ruan, J.-L., Cai, Y.-L., Luo, Q., Xu, H.-X., Wu, Y.-X., 2011. In vitro and in vivo evaluation of the wound healing properties of Siegesbeckia pubescens. Journal of Ethnopharmacology 134, 1033–1038. http://dx.doi.org/10.1016/j.jep.2011.02.010.