1-s2.0-S0944711311000845-main

Page 1

He

EFa

b

c

d

e

f

a

KPBWOBH

I

nlrWtbCmc

CT

0d

Phytomedicine 18 (2011) 976– 984

Contents lists available at ScienceDirect

Phytomedicine

jou rn al hom epage: www.elsev ier .de /phymed

ydroalcoholic extract based-ointment from Punica granatum L. peels withnhanced in vivo healing potential on dermal wounds

.A. Hayounia,e,∗ , K. Miledb, S. Boubakerc, Z. Bellasfarb, M. Abedrabbad, H. Iwaskie, H. Okue, T. Matsuie,

. Limama, M. Hamdi f

Laboratory of Bioactive Compounds at Biotechnology Center, Ecopark of Borj Cedria, BP-901, Hammam Lif 2050, TunisiaExperimental Commodities for Animal Care, Institute of Pasteur, Tunis, TunisiaLaboratory of Anatomo-pathology, Institute of Pasteur, Tunis, TunisiaUR Molecular Physico-chemical, IPEST, La Marsa, TunisiaCenter of Molecular Biosciences, University of the Ryukuyus, Nishihara, Okinawa, JapanMicrobial Ecology and Technology Laboratory, INSAT, Tunis, Tunisia

r t i c l e i n f o

eywords:unica granatum L. peelsiological activitiesound healing

intmentiochemicalistopathological

a b s t r a c t

The present study reports for the first time, the in vivo wound healing potential of Punica granatumL. peels. A 5% (w/w) methanolic extract based-ointment was formulated and evaluated for its woundhealing in guinea pigs. The ointment was applied in vivo on the paravertebral area of twelve excisedwounded models once a day for 10 consecutive days. The ointment significantly enhanced the woundcontraction and the period of epithelialization as assessed by the mechanical (contraction rate, tensilestrength), the biochemical (increasing of collagen, DNA and proteins synthesis) and the histopathologicalcharacteristics. Such investigation was encouraged by the efficiency of the methanolic extract as antimi-crobial and antioxidant. Indeed, the extract showed antioxidant activity as strong as natural and syntheticcompounds (Trolox, BHA, Quercetin). Furthermore, the extract exhibited significant antibacterial and

antifungal activity against almost all tested bacteria: Pseudomonas aeruginosa ATCC 9027, Staphylococ-cus aureus ATCC 25923, Escherichia coli ATCC 25922, Klebsiella pneumoniae, Salmonella anatum, Salmonellatyphimurium, Streptococcus pneumoniae, and fungi Candida albicans, Candida glabrata, Trichopyton rubrumand Aspergillus niger. The formulated ointment might well find use as skin repair agent without hazard tohuman health based on these results and on the fact that it has been well established that the extracts ofpomegranate used in conditions similar to those applied by traditional medicine, showed no toxic effects.

ntroduction

The concept of developing drugs from plants used in indige-ous medical system is much older, while in some cases direct

inks between a local and biomedical use exists, in other cases theelationship is much more complex (Heinrich and Gibbons 2001).

ounds and particularly chronic wounds are major concerns forhe patient and clinician alike, chronic wounds affect a large num-

er of patients and seriously reduce their quality of life. Balick andox (1996) reported that only 1–3% of drugs listed in Western phar-acopoeia are intended for use in the skin and for wounds, by

omparison, at least one third of herbal remedies are for such use.

∗ Corresponding author at: Laboratory of Bioactive Compounds at Biotechnologyenter, Ecopark of Borj-Cedria, BP-901, Hammam-Lif 2050, Tunisia.el.: +216 25600851.

E-mail addresses: [email protected], a [email protected] (E.A. Hayouni).

944-7113/$ – see front matter © 2011 Elsevier GmbH. All rights reserved.oi:10.1016/j.phymed.2011.02.011

© 2011 Elsevier GmbH. All rights reserved.

Wound healing involves a chain of well orchestrated, biochem-ical and cellular events, leading to the growth and regeneration ofwounded tissue. In coetaneous wound healing, the inflammationstage begins immediately after injury, first with vasoconstrictionthat favours homeostasis and releases inflammation mediators. Theproliferative phase is characterized by granulation tissue prolifera-tion formed mainly by fibroblast and the angiogenesis process. Theremodelling stage is characterized by reformulations and improve-ment in the components of the collagen fibers that increases thetensile strength. Although the rate of collagen synthesis slow downafter about three weeks, collagen cross-linking and reorganisationoccur for months after injury in the remodelling phase of repair(Beanes et al. 2003).

Due to the lack of side effects compared to synthetic drugs,approximately 60% of the world’s population relies almost entirelyon plants for medication, and natural products have long beenrecognized as an important source of therapeutically effectivemedicines. Indeed, many plants have been shown to possess ther-

Page 2

omedi

aJ1aeawhhs

pmAtsds2tahcwa2

harubeet

M

C

cW

F(

E.A. Hayouni et al. / Phyt

peutic potential as promoters of wound healing for example,atropha curcas, Aloe barbadensis, Centella asiatica (Villegas et al.997; Shukla et al. 1999). These plants exhibited often antifungal,ntimicrobial, antioxidant, anti-inflammatory activities (Turkoglut al. 2007). North African folklore and tribal medicines employ

number of plants and animal products for treatment of cuts,ounds and burns (Ahmed et al. 1995). Some of these plantsave been screened scientifically for the evaluation of their woundealing activity in different pharmacological models and humanubjects, but the potential of most of the plants remain unexplored.

Punica granatum L. (Punicaceae), commonly calledomegranate, is a large deciduous shrub or small tree usededicinally in Europe, Indo-China, the Philippine Islands, Northfrica, and South Africa. The plant is used in folklore medicine for

he treatment of various diseases, such as ulcer, hepatic damage,nakebite, etc. The rind of the fruit is antihelminthic, useful inysentery and ulcer (Lansky and Newman 2007). The plant alsohows high antioxidant and antiartherogenic activity (Aviram et al.000). Modern uses of pomegranate derived products now includereatment of acquired immune deficiency syndrome (AIDS) inddition to use for cosmetic beautification and enhancementormone replacement therapy, resolution of allergic symptoms,ardiovascular protection, oral hygiene, ophthalmic ointmenteight loss soap and as an adjunct therapy to increase bioavail-

bility of radioactive dyes during diagnostic imaging (Aviram et al.000; Lansky and Newman 2007).

There are only few prospective randomized controlled trials thatave proved the clinical efficacy of the traditional wound healinggents. Accordingly, based on its ethnopharmacological profile andeputed medicinal use in traditional practice, the present study wasndertaken to (i) evaluate, the in vitro, antioxidant and antimicro-ial activities (against various wound pathogens) of the methanolicxtract obtained from the peels of P. granatum fruits (PgME) (ii) tovaluate systematically the possible in vivo wound healing poten-ial of a 5% ointment formulated using this extract.

aterials and methods

hemicals

The anthocyanidins standards were obtained from Apin Chemi-als Ltd. (Abingdon, England). Ellagic acid and gallic acid were from

ako (Japan), all the standards and solvents were HPLC-grade.

ig. 1. Main steps of the experimental protocol: (A) shaving; (B) wound creation under

right) which is shown next and (D) the ointment formulated using PgME.

cine 18 (2011) 976– 984 977

ABTS, �-carotene, were from Sigma–Aldrich (Steinheim, Germany).All other chemicals used in this study were obtained either fromMerck Chemical Company (Darmstadt, Germany) or from FlukaBiochemika (Buchs, Switzerland).

Plant material and preparation of the extract

Pomegranates (Romman as Tunisian vernacular name) werepurchased from a local market in Tunis (October 2008) andidentified at the Aromatic and Medicinal Plants Unit, Biotech-nology Center at Borj-Cedria where a voucher specimen wasdeposited. The peels were dried in shade and than pulverizedby a mechanical grinder. 100 g of the powder were extractedwith aqueous methanol (75%) using a soxhlet extraction appa-ratus, for 8 h. The obtained organic extract was concentrated byrotary evaporation under vacuum at 45 ◦C, using a HACH UV–Visspectrophotometer (Model DR/4000, Colorado, USA). The semi-solid mass (18.7% ± 0.88, methanol free) obtained was used forthe in vitro biological activities and for the formulation of a 5%(w/w) extract-based ointment, using soft white paraffin, as a vehi-cle (Fig. 1D).

HPLC-fingerprint analysis of PgME

Sample preparation. The PgME was prepared to a concentrationof 12 mg/ml, with injection volume of 10 �l. Before injection, PgMEwas centrifuged (4 min at 5000 rpm) and the centrifuged super-natant was filtered through a 0.45 �m PTFE filter (Waters), andsubmitted to HPLC–DAD and HPLC–ESI–MS analysis.

Apparatus. Shimadzu controller (SLC-10A vp) with of a vacuumdegasser (DGU-14A), a binary pumps (LC-10AD vp), an oven (CTO-10AC vp) and a Shimadzu Diode Array Detector (SPD-M10A vp). UVspectra from 200 to 600 nm were recorded for peak characteriza-tion.

Separation. Column: Inertsil C18 ODS-3 column (5 �m,200 mm × 4.6 mm) with a suitable guard column (C18, ODS, 5 �m,4 mm × 3.0 mm). Temperature: 30 ◦C.

Solvent: A: 0.1% (v/v) aqueous solution of formic acid and B:methanol. Flow rate: 1 ml/min.

Gradient: 0–5 min, 99% B, 5–20 min, linear gradient 1–20% B;20–30 min, linear gradient 20–40% B; 30–35 min, linear gradient40–95% B and maintained constant at 95% B over the next 15 min.

anaesthesia; (C) medication using cetrimide-based cream (left) and our ointment

Page 3

9 omedi

n

H

SiaflritPti

Q

A

a

ˇ

t(

A

M

ilpca

D

tM(ogfittoedtd

i(r(pcfwMiac

78 E.A. Hayouni et al. / Phyt

Detection: 254 nm, 520 nm (anthocyanidins), 366 nm (ellagitan-ins).

PLC–ESI–MS analysis

HPLC coupled to mass spectrometry was performed using aquire 4000 plus ion trap spectrometer fitted with electrosprayonization (Bruker Daltonics, Billerica, MA, USA). The capillary volt-ge was 3500 V. Nitrogen was used as nebulizer gas at 350 ◦C at aow of 8.0 l min−1, nebulizer pressure was 27.5 psi. Spectra wereecorded between m/z 150 and 1400 in negative ion mode. Collisionnduced dissociation (CID) spectra were obtained with a fragmenta-ion amplitude of 1.00 V (MS/MS) using helium as the collision gas.unicalagins A and B were identified by comparison to their respec-ive ions (1083.7 m/z mother ion) and MS/MS (601.0 m/z daughteron) ions as reported (Cedra et al. 2003).

uantification of the antioxidant activity

BTS assayABTS radical scavenging activity of the extract was determined

ccording to (Re et al. 1999) with some modifications.

-Carotene bleaching testThe antioxidant activity of the extract was evaluated according

o a slightly modified version of the �-carotene bleaching methodPratt 1980).

ntimicrobial activity

icrobial strains and culture conditionsThe extract was tested against a panel of 11 wound pathogens

ncluding 7 bacteria: Pseudomonas aeruginosa ATCC 9027, Staphy-ococcus aureus ATCC 25923, Escherichia coli ATCC 25922, Klebsiellaneumoniae, Salmonella anatum, Salmonella typhimurium, Strepto-occus pneumoniae; two yeasts: Candida albicans, Candida glabrata;nd two fungi: Trichopyton rubrum and Aspergillus niger.

etermination of minimum inhibitory concentration (MIC)A broth microdilution susceptibility assay was used for bac-

eria, as recommended by NCCLS, for the determination of theIC (NCCLS 1999). All tests were performed in Nutrient Broth

NB, Oxoid Ltd, Hampshire, UH). Bacterial strains were culturedvernight at 37 ◦C in NB. Bacterial strains were suspended in NB toive a final density of approximately 106 cfu/ml and these were con-rmed by viable counts. Geometric dilutions ranging from 2 mg/mlo 0.062 mg/ml of the methanolic extract from the peels of P. grana-um fruits, were prepared in a 96-well microtitre plate, includingne growth control (NB) and one sterility control (NB + Testedxtract). Plates were incubated under normal atmospheric con-itions at 37 ◦C for 24 h. The bacterial growth was indicated byhe presence of a “pellet” on the well bottom. Amikacin and Clin-amycine were used as positive controls.

A macrodilution broth method was used to determine the min-mal inhibitory concentrations for yeasts and filamentous fungiNCCLS 1997). A serial dilution of the extract was prepared over theange 0.062–2 mg/ml. Recent cultures in Sabouraud Dextrose BrothSDB), in log phase, of each strain were used to prepare the cell sus-ension adjusted to (1–2) × 103 cells/ml for yeasts and (1–2) × 104

ells/ml for fungi. The test tubes were incubated aerobically at 37 ◦Cor 48 h (yeasts) or at 30 ◦C for 3–5 days (dermatophytes) and MICs

ere determined. The fungal growth was indicated by the turbidity.IC was defined as the lowest concentration that reduced growth

n 80%, compared with antifungal-free controls. Herein, Amikacinnd Amphotricin B were used as positive controls and the negativeontrols were used as above.

cine 18 (2011) 976– 984

Animals and experimental protocol

This study was conducted in the Experimental Animal Com-modities at the Institute of Pasteur, Tunis. Male and female guineapigs, weighing 250–300 g, were used. All animals were housed, fedand treated in accordance with the in-house guidelines for animalprotection. Animals were kept for 2 weeks to be acclimatized priorto the investigation. Throughout experimentation period, animalswere given standard pellet diet and water ad libitum.

The animals were anesthetized with Ketamin chlorhydrate ata dose of 1.5 mg/kg body weight intramuscularly (IM). Then alocal anaesthesia using Lidocaïn hydrate (2%) was applied on theexcision area. This type of anaesthesia prevents any pain and move-ment of the animals at least 2 h after the administration of theanaesthetic solution. The dorsal back skins of the animals wereshaved and ethanol (70%) was used as antiseptic for the shavedregion before creating the wound (Fig. 1A). One excision woundwas inflicted by cutting away a 400 mm2 full-thickness skin in thedorsal inter-scapular. The wound was left undressed to the openenvironment and no local or systemic anti-microbial agents wereused (Fig. 1B). After wound creation, experimental animals wererandomly divided into the following groups (n = 12):

- Group 1: untreated (control) animals- Group 2: wounds treated topically with a simple ointment base

(placebo group)- Group 3: wounds treated topically with reference standard a 2%

(w/w) cetrimide-based cream- Group 4: wounds treated topically with the methanolic extract

based-ointment prepared from P. granatum peels.

The herbal ointment along with standard and vehicle wereapplied (enough to cover the wound) once daily during 10 con-secutive days (Fig. 1C).

Determination of wound contractionFor the determination of the wound contraction, excision

wounds were traced on a transparent paper having a scale, andthe change in wound size at 4 day-intervals was calculated as thepercentage of wound area that healed. The rate of wound contrac-tion was expressed in terms of the percentage of wound area thathad healed: % wound contraction = (healed area/total area) × 100.

Tensile strengthTensile strength of wound represents the effectiveness of wound

healing. Defined as the force required opening the healing skin itis used to measure the completeness of healing. Usually wound-healing agents promote the gaining of tensile strength. For thispurpose the newly repaired tissue including scar was excised tomeasure the tensile strength by mean of a tensiometer. For thequantification one of the edges of the skin was fixed while apply-ing a measurable force to the other one. The load (weight) in gramsrequired to disrupt the wound is determined 20 days after surgery.

Biochemical analyses of the newly formed skinsAnimals were sacrificed on day 8 after wound creation, and the

entire wound on each animal was cut out and stored at −70 ◦Cuntil analyses. Hydroxyproline, the basic constituent of collagen,was taken as a marker of collagen synthesis. Tissues were dried ina hot air oven at 60–70 ◦C to constant weight and were hydrolysedin 6 N HCl at 130 ◦C for 4 h in sealed tubes. The hydrolyzed samples

were adjusted to pH 7.0 and were subjected to chloramines-T oxi-dation for 20 min. The reaction was terminated by addition of 0.4 Mperchloric acid and color was developed with the help of Ehrlichreagent at 60 ◦C (Woessner 1961), and measured at 557 nm using aHACH spectrophotometer.

Page 4

E.A. Hayouni et al. / Phytomedicine 18 (2011) 976– 984 979

ME re

uSTpauBTm

H

a

Fig. 2. HPLC–DAD chromatogram of Pg

For estimations of total protein and DNA content, wet gran-lation tissues were first extracted with TCA by the method ofchneider (1957). Briefly, the tissue was first homogenized in 5%CA and centrifuged. The pellet was washed with 10% TCA, resus-ended in 5% TCA, and kept for 15 min in a water bath maintainedt 90 ◦C. The contents were centrifuged and the supernatant wassed for the determination of DNA content by the method ofurton (1956). The precipitated proteins were suspended in 0.1 Mris–HCl, pH 7.4, and the protein content was estimated by the

ethod of Lowry et al. (1951).

istopathological studiesAnimals were sacrificed on the 4th, 8th, 12th, 16th and 20th days

fter wound creation. Wound tissues were immediately preserved

corded at 254 nm (A) and 366 nm (B).

in 10% buffered formalin and passed through different grades ofalcohol in order to ensure complete dehydration of tissues andthen embedded in paraffin wax. Serial sections of paraffin embed-ded tissues of 6 �m thickness were cut using a microtome andstained with haematoxylin and eosin (H&E). Additional sectionswere stained with Masson Trichrome for the assessment of colla-gen content and maturation within the dermis. Micrographs wereobtained with an Axiophot Zeiss Light Microscope.

Tissues samples were evaluated for the following histological

criteria: the extent of re-epithelization, the maturation and orga-nization of the epidermal squamous cells, the thickness of thegranular cell layer, the degree of granulation tissue formation, andfor collagenisation and scar formation in the dermis. For the epi-dermal changes, the adjacent non-traumatized skin served as the

Page 5

9 omedi

ipcwbaf

S

mad

Fb

80 E.A. Hayouni et al. / Phyt

nternal control for the assessment of epidermal collagen fibers,rominent vascularity and inflammation considered as an indi-ation for immature granulation tissue. Contrary, the presence ofell formed and horizontally-oriented collagen fibers, as evident

y Masson Trichrome stain, along with scarce inflammation cellsnd inconspicuous blood vessels were considered as an indicationor the maturity of the scar.

tatistical analysisFor the in vitro tests, all data were expressed as

eans ± standard errors of triplicate measurements. One-waynalysis of variance (ANOVA) was carried out to identify theifferences between treated groups and controls. The statistical

ig. 3. The antioxidant activities of the methanolic extract from P. granatum peels as deleaching test. Results are mean ± S.E of three parallel measurements. In (B) error bars ar

cine 18 (2011) 976– 984

comparisons between variables were performed with Student’st-test. Significances were calculated by comparing rates of healingin group treated either with PgME ointment or with cetrimidecream versus controls and in group treated with PgME ointmentversus cetrimide cream group. p-Values less than 0.05 wereconsidered to be significant.

Results and discussion

Fingerprint analysis of PgME

Fig. 2 shows the chemical pattern of PgME with ellagi-tanins (punicalagin A, punicalagin B, gallic acid and ellagic

termined by: (A) the ABTS Free radical scavenging activity and (B) the �-carotenee too small to be seen.

Page 6

omedicine 18 (2011) 976– 984 981

aa3gab1dgpeha

A

dAheBirah2cauoat

c7suipgg

otaepfim

bafspf(oEcasto

Table 1Minimum inhibitory concentrations of PgME on wound pathogens using micro- andmacrodilution methods.

MIC

P. granatum extract Antibiotic

Staphylococcus aureus ATCC 25923 >2 2a

Streptococcus pneumoniae 1 0.125b

Escherichia coli ATCC 25922 0.5 2a

Klebsiella pneumoniae >2 1b

Pseudomonas aeruginosa ATCC 9027 0.5 1a

Salmonella typhimurium 0.25 1a

Salmonella anatum 0.25 2a

Candida albicans 0.5 0.5a

Candida glabrata 2 0.125a

Trichopyton rubrum 0.125 15.62c

Aspergillus niger >2 15.62c

MIC, minimum inhibitory concentration. Values are given as mg/ml for the methano-

E.A. Hayouni et al. / Phyt

cid) being the major compounds. Some anthocyanidins werelso detected namely delphinidin-3,5-diglucoside; quercitin-,4′-O-diglucoside; pelargonidin-3-glucoside and quercitin-4′-O-lucoside (Spiraeoside). It has been reported that pomegranate is

rich source of anthocyanidins with delphinidin-3,5-diglucosideeing a major anthocyanin in pomegranate juice (Harborne967; Du et al. 1975). The seed coat of this fruit containselphinidin-3-glucoside, delphinidin-3,5-diglucoside, cyanidin-3-lucoside, cyanidin-3,5-diglucoside, pelargonidin-3-glucoside, andelargonidin-3,5-diglucoside (Du et al. 1975). In addition, it is wellstablished that pomegranate peel contains substantial amounts ofydrolyzable tannins punicalagins, punicalin, ellagic acid and galliccid (Nasr et al. 1996; Tanaka et al. 1986).

ntioxidant activity of PgME

Fig. 3A illustrates the decrease in absorbance of ABTS•+ radicalue the scavenging ability of the PgME. At used concentrations, theBTS•+ radical scavenging activity of PgME was found significantlyigher than quercitin (p ≤ 0.05, at least). Furthermore, the extractxhibited a scavenging potency, as strong as the positive controlHA. It is well established that extracts obtained using high polar-

ty solvents (such as methanol) were considerably more effectiveadical scavengers than those using less polarity solvents (Zhound Yu 2004). The peel of this fruit had recently been reported toave higher antioxidant activity than its pulp and seed (Li et al.006). Furthermore, Ricci et al. (2006) reported the antioxidantapacities of some extracts from P. granatum arils, juice and rindnd correlated them to their polyphenols contents. In another workndertaken by Okonogi et al. (2007) it was reported that the extractf pomegranate peel showed the highest antioxidant activity, withn IC50 of 0.003 mg/ml in a DPPH assay. This extract exhibited alsohe highest TEAC value of 4.07 mM/mg.

In addition, according to the results presented in Fig. 3B, wean notice that our extract, used at 500 ppm, showed more than5% bleaching of the �-carotene solution, which is as strong as theynthetic antioxidants Trolox and BHA. García-Alonso et al. (2004)sing the TBARS method found markedly similar results when

nvestigating the antioxidant capacities of aqueous extracts fromomegranate peels. Indeed, they pointed out that the fruits withreater antioxidant activity are all rich in anthocynanidins, sug-esting that these pigments could be contributing to this activity.

It is consented that reactive oxygen species (ROS) are deleteri-us to wound healing process due to the harmful effects on cells andissues (Aliyeva et al. 2004). The wound site is rich in both oxygennd nitrogen reactive species along with their derivatives. The pres-nce of these radicals will result in oxidative stress leading to lipideroxidation, DNA breakage, and enzyme inactivation, includingree-radical scavenger enzymes. Evidence for the role of oxidantsn the pathogenesis of many diseases suggests that antioxidants

ay be of therapeutic use in these conditions.In this context, topical applications of methanolic extract-

ased ointment, with free-radical-scavenging properties, could beble to significantly improve wound healing and protect tissuesrom oxidative damage. It has been suggested that free radicalcavenging and antioxidant activities of extracts from variousarts of P. granatum play an important role in prevention ofree radical-related diseases, including aging, wounds and ulcersHarmam 2001). In addition, anthocyanidins prevented lipid per-xidation of cell or liposome membranes (Tsuda et al. 1996).llagitanins, namely punicalagin, had shown remarkable pharma-

ological activities including anti-inflammatory hepatoprotectivend antigenotoxic activities. One of the important factors respon-ible for all of the above activities of punicalagin could be ascribedo its antioxidant activity which has been attributed to the presencef 16 dissociable –OH groups (Lin et al. 2001).

lic extract and as �g/ml for antibiotics: amikacin, clindamycine, amphotericin B.a Amikacin.b Clindamycine.c Amphotericin B.

Antimicrobial activity of the methanolic extract

The inhibitory effects of the crude-methanolic extract fromthe peels of P. granatum against 11 pathogens are given inTable 1. Among the test microorganisms, Trichopyton rubrumwas found to be the most sensitive with MIC = 0.125 mg/ml.Salmonella typhimurium and Salmonella anatum came next withMIC = 0.25 mg/ml, closely followed by Escherchia coli, Pseudomonasaeruginosa, Candida albicans (MIC = 0.5 mg/ml). Streptococcus pneu-moniae was the less sensitive strain (MIC = 1 mg/ml). Similar resultswere reported in many other studies, where various extractsobtained from different parts of P. grantum, were tested againstlarge panels of microorganisms. For example, Navarro et al. (1996),recorded MICs values of 0.62 mg/ml (against Staphylococcus aureusATCC 6538) and 10 mg/ml (against Escherchia coli ATCC 8937, Pseu-domonas aeruginosa ATCC 10231 and Candida albicans ATCC 10231)when they tested a methanolic extract from the fruit shell. In pre-vious research, ellagitanins inhibited the growth of the pathogenicGram-negative bacteria E. coli (IC50 = 9.2 �M) and P. aeruginosa(IC50 = 3.2 �M). It is also likely that ellagitanins act synergisti-cally when combined with anthocynanidins and other phenols toenhance their action against bacteria (Reddy et al. 2007). Since MICvalues of P. granatum extract were substantially low namely againstTrichopyton rubrum (125 �g/ml), and taking into accounts the highantioxidant activity of the extract, it was proposed to undertake itsin vivo wound healing capacity.

Mechanical survey of the wound healing process

The restoration of a functional barrier during the healing pro-cess is dependent on the successful regeneration of new skin witharchitecture that closely resembles the injured tissues. Wound con-traction in different groups is shown in Fig. 4. It clearly appears thatwound contraction started immediately from day 4, in all groups.However, on later days, the rate is much faster in the positive con-trol group and the P. grantaum extract-based ointment group thanin the control groups (untreated and placebo groups). On day 8,animals of both treated groups exhibited significant increase inthe percentage of wound contraction as compared to untreatedand placebo groups. On day 16 post-surgery, It may be seen that

the positive control group showed 85% healing, whereas grouptreated with P. granatum ointment showed 83.5% healing, whencompared to the controls (43% and 54% for untreated and placebogroup, respectively). On day 20, contrary to the control groups, noscars were observed on animals treated with certimide cream or

Page 7

982 E.A. Hayouni et al. / Phytomedicine 18 (2011) 976– 984

F and ta d signa . One-s ound

Pcs

mwtsagpTi(iisC1

B

s

TE

Rt

ig. 4. Percentage reduction of wound size in control groups (untreated and placebo)t 4-day intervals. Results are presented as means ± S.E (n = 4–6). Results were founnd also in cetrimide group versus the same control groups (*p < 0.05 or **p < 0.01)imilar potency and wound closure. The upper serial photos show the kinetics of w

. granatum ointment. This may be considered as an indication foromplete healing and hence on this day measurements of tensiletrength were carried out.

Indeed, the results showed that the tensile strength of ani-als treated with cetrimide cream or the PgME-based ointmentas significantly greater than those of the untreated group (con-

rol) and the placebo group. We recorded 506 g and 494 g (notatistically significant difference between treated groups: p > 0.05)s breaking forces in certimide-cream group and PgME ointmentroup, respectively. Meanwhile, in the untreated animals and thelacebo group, only 389 g and 412 g were recorded, respectively.he increase in tensile strength of treated wounds may be due toncrease in collagen concentration and stabilization of the fibersUdupa et al. 1995). The results obtained in this study were sim-lar to those given by of Aloe vera on collagen characteristicsn healing dermal wounds in rats (Chithra et al. 1998). Also, aimilar effect has been observed with the ethanolic extract ofentella asiatica on the rat dermal wound healing (Suguna et al.996).

iochemical survey of the wound healing process

The collagen, DNA and total protein contents of granulation tis-ues on day 8 are given in Table 2. Wounds treated with P. granatum

able 2ffects of topical treatment for 8 days on selected biochemical markers of wound healing

Untreated Placebo

Hydroxyproline 62.6 ± 1.22 73.2 ± 1.1

DNA 2.56 ± 0.03 2.98 ± 0.09

Total proteins 52.3 ± 0.51 65.5 ± 0.33

esults are given in mg/g wet weight tissue. Values are mean ± S.D (n = 6 animals). There ao ointment group.

* As compared to untreated or placebo group: p < 0.05.** As compared to untreated or placebo group: p < 0.01.

reated groups (with topical application of cetrimide cream or P. granatum ointment),ificant in ointment group versus untreated or placebo group (*p < 0.05 or **p < 0.01)way ANOVA also supported t-test analysis and showed that both treatments haveclosure in P. granatum ointment group.

extract-based ointment as well as the cetrimide cream exhibited avery significant increase in hydroxyproline levels as compared tothe untreated group (63% and 61.9% increase, respectively) and tothe vehicle group (74.2% folds and 72.5% rise, respectively). Col-lagen is the predominant extracellular protein in the granulationtissue of a healing wound and there is a rapid increase in the synthe-sis of this protein in the wound area soon after an injury. In addition,to provide strength and integrity to a tissue matrix collagen playsan important role in haemostasis and subsequent epithelializationrequires also collagen.

Furthermore, the DNA levels were also increased by 77.5% in thewound receiving topical applications of P. granatum extract-basedointment as compared to the untreated animals. Similar resultswere recorded in the certimide-treated group which showed a sig-nificant DNA content increase (70.1% rise when compared to theuntreated animals). In parallel, similar behaviour was noticed, con-cerning the rates of the total proteins. Indeed, it was observed that P.granatum and certimide groups had 74.9% and 73.6%, higher proteincontents, when compared to the vehicle, respectively. Such obser-

vations are corroborated by the collagen/DNA ratio in granulationtissue. This parameter may be taken to be an index of the synthesisof collagen per cell in the wound area. The order of the recordedratios was as follows: P. granatum ointment (29.9) > cetrimidecream (27.7) > placebo (24.56) > untreated (24.45).

in guinea pig excision models.

Cetrimide cream P. granatum ointment

101.1 ± 0.87** 98.7 ± 1.01**

3.65 ± 0.05** 3.3 ± 0.08*

89 ± 0.53** 87.4 ± 0.41**

re no statistically significant difference (p > 0.05) when comparing cetrimide group

Page 8

E.A. Hayouni et al. / Phytomedicine 18 (2011) 976– 984 983

F 0×) s( de crea

cHtppabtseommcriapt

H

tio

Fpsgo

ig. 5. Histological evaluation of wound repair by Haematoxylin and Eosin (H&E 20B) control and (C) P. granatum treated wounds; on day 12: (D) vehicle, (E) cetrimind (H) P. granatum treatment.

The protein and DNA content of granulation tissues indi-ate the levels of protein synthesis and cellular proliferation.igher protein and DNA contents (compared to the untreated con-

rols) of the treated wounds suggest that P. granatum ointment,robably through an unknown mechanism, stimulates cellularroliferation. Even though, it was recently established that anqueous extract of pomegranate peel had a potent dermal effecty stimulating dermal fibroblast proliferation and collagen syn-hesis while inhibiting the major collagen-degrading enzyme inkin (matrix metalloproteinase-1), but had no growth-supportingffect on keratinocytes (Aslam et al. 2006). The collagen/DNA ratiof the granulation tissues also suggests that P. granatum oint-ent may increase the synthesis of collagen per cell. The collagenolecules synthesized are laid down at the wound site and become

rosslinked to form fibers. Wound strength is acquired from both,emodelling of collagen, and the formation of stable intra- andnter-molecular crosslinks. P. granatum ointment-treated woundslso showed an increased rate of wound contraction, leading to arompt healing as confirmed by decreased period of epithelializa-ion when compared to untreated control wounds.

isthopathalogical survey of the wound healing process

Haematoxylin and eosin (H&E) stained sections of granula-ions tissue collected on various days were examined for cellularnfiltration, neo-vascularisation, epithelial regeneration and matrixrganization. Day 4 sections showed increased cellular infiltration

ig. 6. Histological analysis of wound-edge tissue obtained on day 20 of the experimenrocedure (400×) which results in blue-black nuclei, blue collagen/cytoplasm, and keratinpongiosis of the epidermis with persistence of inflammatory cells in the dermis. (B) Cetrranatum treatment showing thin well-formed epidermis with hair follicle formation in thf the references to color in this figure legend, the reader is referred to the web version o

taining of granulation tissue on different day of healing. (A) Normal skin; on day 4:am treatment, and (F) P. granatum treatment; on day 20: (G) cetrimide treatment

in animals topically treated with the P. granatum ointment (Fig. 5C)than control (Fig. 5B) and, no epidermal regeneration was observed.On day 12, a well-advanced organization of granulation tissue andon-going epithelialization was observed in treated groups, eitherwith certimide cream or P. granatum ointment (Fig. 5E and F,respectively) than the vehicle (Fig. 5D) and the control. The his-tological examination showed that the original tissue regenerationwas much faster in the skin wound treated with extract ointment orcetrimide cream without any oedema, congestion or inflammatorydisorders (results not shown).

On day 20, animals treated with the extract-based ointment(Fig. 5H) and the certimide-based cream (Fig. 5G) showed very closeprofiles when compared to one another and when both groups werecompared to the control and the vehicle. There was full thicknessepidermal regeneration which covered completely the wound area.The epidermis was thick and disorganized, especially when com-pared with the adjacent normal skin. The keratin layer was thickwith focal parakeratosis. In both groups the granular layer was wellformed but 2–3 cells in thickness. The basal layer was well formedin both groups as it was in the control and the vehicle groups. Inaddition, in the dermis, maturation and organization of the col-lagen fibers as judged by the Masson Trichrome stain was similar

(Fig. 6B and C). The dermis was cellular with proliferation of fibrob-lasts, laying down disorganized and poorly oriented collagen fibers.Prominent capillary-sized blood vessels were seen. Scattered col-lections of inflammatory cells, consisting mainly of macrophagesand few neutrophils were also present throughout the whole thick-

t. Neutral buffered formalin-fixed sections were stained with Masson Trichrome/muscle fibers/intracellular fibers all stained red. (A) Vehicle treated group showingimide cream treatment showing maturation of the epidermis and the dermis. (C) P.e dermis and no inflammatory cells in a well organized dermis. (For interpretation

f the article.)

Page 9

9 omedi

nfshnt

C

tyfan

ipiprbr

ttlpmPHiawinn

A

ptt

R

A

A

A

A

B

B

84 E.A. Hayouni et al. / Phyt

ess of the dermis. Whereas, in vehicle group (Fig. 6A) there was aailure of re-epithelization of the wound. Indeed, there was per-istence, within the dermis, of immature granulation with fewaphazardly-oriented collagen fibers. Inflammatory cells, predomi-antly neutrophils, were still detected within the granulation tissueogether with blood vessels which were prominent and dilated.

onclusion

In addition to its value as a table fruit, pomegranate prepara-ions have been used for ages in various folk medicines. In recentears there is important focus on the bioactive moieties of its dif-erent parts which showed a large panel of activities: antioxidant,nti-inflammatory, angiogenic, anti-cancer, skin repair. . . these areamely attributed to the polyphenolic and lipophilic fractions.

The different phases of the wound healing process overlap anddeally a plant-based remedy should affect at least two differentrocesses before it can be said to have some scientific support for

ts traditional use. Since there is a definite role of free radicals in theathogenesis of wound, the antioxidant activity was studied. Theesults indicate that the PgME possesses potent antioxidant activityy inhibiting lipid peroxidation and increasing the potency of freeadicals scavenging.

In the field of wound healing, there are several unknowns,his includes the wound itself. Mechanical and biochemical surveyogether with the histological results showed that the methano-ic extract-based ointment from Tunisian pomegranate exhibitsotent healing properties on excision wounds, in guinea pigsodel. This again validates the potent wound healing activity of

. granatum extracts as claimed by the ethnopharmacological data.owever, the exact mechanism of the healing process of wound

s not clearly understood. Indeed, one has to remember that therere a number of parameters which are involved in the healing ofound including epithelization, antioxidant defense and biochem-

cal changes (hydroxyproline). Therefore, further approaches areeeded to clearly elucidate the full mechanism of action of suchatural preparations.

cknowledgments

We are very grateful to Dr. Nazek (Pasteur Institute, Tunis) forroviding us with microorganisms. The ointment formulated usinghe crude methanolic extract from Punica granatum L. peels is pro-ected by a Tunisian patent (N◦7260).

eferences

hmed, R., Ifzal, S.M., Saifuddin, A., Nazeer, M., 1995. Studies on Punica granatum: iso-lation and identification of some constituents from the seeds of Punica granatum.Pak. J. Pharma. Sci. 8, 69–72.

liyeva, E., Umur, S., Zafer, E., Acigoz, G., 2004. The effect of polylactide membraneson the levels of reactive oxygen species in periodontal flaps during wound heal-ing. Biomaterials 25, 4633–4638.

slam, M.N., Lansky, E.P., Varani, J., 2006. Pomegranate as a cosmeceutical source:pomegranate fractions promote proliferation and procollagen synthesis andinhibit matrix metalloproteinase-1 production in human skin cells. J. Ethnophar-macol. 103, 311–318.

viram, M., Dornfeld, L., Rosenblat, M., Volkova, N., Kaplan, M., Coleman, R., Hayek, T.,Presser, D., Fuhrman, B., 2000. Pomegranate juice consumption reduces oxida-tive stress, atherogenic modifications to LDL, and platelet aggregation: studies

in humans and in atherosclerotic apolipoprotein E-deficient mice. Am. J. Clin.Nutr. 71, 1062–1076.

alick, M., Cox, P.A., 1996. Plants, People and Culture: The Science of Ethnobotany.Scientific American Library, W.H. Freeman and Company, New York.

eanes, S.R., Dang, C., Soo, C., Ting, K., 2003. The Phases of Cutaneous Wound Healing.Expert Reviews in Molecular Medicine. Cambridge University Press, UK.

cine 18 (2011) 976– 984

Burton, K., 1956. A study of the conditions and mechanism of the diphenylaminereaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J.62, 315–323.

Cedra, B., Ceron, J.J., Tomas-Barberan, F.A., Espin, J.C., 2003. Repeated oral adminis-tration oh high doses of the pomegranate ellagitannin punicalagin to rats for 37days is not toxic. J. Agric. Food Chem. 51 (11), 3493–3501.

Chithra, P., Sajithlal, G.B., Chandrakasan, G., 1998. Influence of Aloe vera on collagencharacteristics in healing dermal wounds in rats. Mol. Cell. Biochem. 181, 71–76.

Du, C.T., Wang, P.L., Francis, F.J., 1975. Anthocyanins of pomegranate (Punica grana-tum). J. Food Sci. 40, 417–422.

García-Alonso, M., Pascual-Teresa, S., Santos-Buelga, C., Rivas-Gonzalo, J.C., 2004.Evaluation of the antioxidant properties of fruits. Food Chem. 84, 13–18.

Harborne, J.B., 1967. The Anthocyanin Pigments. In Comparative Biochemistry of theFlavonoids. Academic Press, New York, pp 1-30.

Harmam, D., 2001. Aging: overview. Ann. N. Y. Acad. Sci. 928, 1–8.Heinrich, M., Gibbons, S., 2001. Ethnopharmacology in drug discovery: an analysis

of its role and potential contribution. J. Pharm. Pharmacol. 53, 425–430.Lansky, E.P., Newman, R.A., 2007. Punica granatum (pomegranate) and its potential

for prevention and treatment of inflammation and cancer. J. Ethnopharmacol.109, 177–206.

Li, Y.F., Guo, C.J., Yang, J.J., Wei, J.Y., Xu, J., Cheng, S., 2006. Evaluation of antioxidantproperties of pomegranate peel extract in comparison with pomegranate pulpextract. Food Chem. 96, 254–260.

Lin, C.C., Hsu, Y.F., Lin, T.C., Hsu, H.Y., 2001. Antioxidant and hepatoprotective effectsof punicalagin and punicalin on acetaminophen-induced liver damage in rats.Phytother. Res. 15, 206–212.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.T., 1951. Protein measurementwith the folin phenol reagent. J. Biol. Chem. 193, 265–268.

Nasr, C.B., Ayed, N., Metche, M., 1996. Quantitative determination of the polypheno-lic content of pomegranate peel. Zeitschrzfi fur Lebensmittel Unterschung UndForschung 203, 374–378.

Navarro, V., Villarreal, M.L., Rojas, G., Lozoya, X., 1996. Antimicrobial evaluation ofsome plants used in Mexican traditional medicine for the treatment of infectiousdiseases. J. Ethnopharmacol. 53, 143–147.

NCCLS (National Committee for Clinical Laboratory Standards), 1997. Referencemethod for broth dilution antifungal susceptibility testing of yeasts. Approvedstandard M27-A. Wayne, PA, USA, pp. 1–29.

NCCLS (National Committee for Clinical Laboratory Standards), 1999. Performancestandards for antimicrobial disk susceptibility testing. 9th International Supple-ment. M100-S9.

Okonogi, S., Durangrat, C., Anuchpreeda, A., Tachakittiorungrod, S., Chowwanapoon-pohn, S., 2007. Comparison of antioxidant capacities and cytotoxicities of certainfruit peels. Food Chem. 103, 839–846.

Pratt, D.E., 1980. Natural antioxidants of soybean and other oil-seeds. In: Simic, M.G.,Karel, M. (Eds.), Autoxidation in Food and Biological Systems. Plenum Press, NewYork, pp. 283–292.

Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Rice-Evans, C., 1999.Antioxidant activity applying an improved ABTS radical cation decolorizingassay. Free Radic. Biol. Med. 26, 1231–1235.

Reddy, M.K., Gupta, S.K., Jacob, M.R., Khan, S.I., Ferreira, D., 2007. Antioxidant, anti-malarial and antimicrobial activities of tannin-rich fractions, ellagitannins andphenolic acids from Punica granatum L. Planta Med. 73, 461–467.

Ricci, D., Giamperi, L., Bucchini, A., Fraternale, D., 2006. Antioxidant activity of Punicagranatum fruits. Fitoterapia 77, 310–312.

Schneider, W.C., 1957. Nucleic acids and derivatives. Method Enzymol. 3, 680–684.Shukla, A., Rasik, A.M., Jain, G.K., Shankar, R., Kulshrestha, D.K., Dhawan, B.N., 1999.

In vitro and in vivo wound healing activity of asiaticoside isolated from Centellaasiatica. J. Ethnopharmacol. 65, 1–11.

Suguna, L., Sivakumar, P., Chandrakasan, G., 1996. Effects of Centella asiatica extracton dermal wound healing in rats. Indian J. Exp. Biol. 34, 1208–1211.

Tanaka, T., Nonaka, G.I., Nishioka, I., 1986. Tannins and related compounds. XL.Revision of the structures of punicalin and punicalagin, and isolation and char-acterization of 2-galloylpunicalin from the bark of Punica granatum L. Chem.Pharm. Bull. 34, 650–655.

Tsuda, T., Shiga, K., Ohshima, K., Kawakishi, S., Osawa, T., 1996. Inhibition of lipidperoxidation and the active oxygen radical scavenging effect of anthocyaninpigment isolated from Phaseolus vulgaris L. Biochem. Pharmacol. 52, 1033–1037.

Turkoglu, A., Duru, M.E., Mercan, N., Kivrak, I., Gezer, K., 2007. Antioxidant andantimicrobial activities of Laetiporus sulphureus (Bull.) Murrill. Food Chem. 101,267–273.

Udupa, A.I., Kulkarni, D.R., Udupa, S.L., 1995. Effect of Tridax procumbens extracts onwound healing. Int. J. Pharm. 33, 37–40.

Villegas, L.F., Fernandez, I.D., Maldonado, H., Torres, R., Zavaleta, A., Vaisberg, A.J.,Hammond, G.B., 1997. Evaluation of the wound healing activity of selected

traditional medicinal plants from Perù. J. Ethnopharmacol. 55, 193–200.

Woessner Jr., J.R., 1961. The determination of hydroxyproline in tissues and pro-tein samples containing small proportions of this amino acid. Arch. Biochem.Biophys. 93, 440–447.

Zhou, K., Yu, L., 2004. Effects of extraction solvent on wheat bran antioxidant activityestimation. LWT-Food Sci. Technol. 37, 717–721.