T

Ma Db D

a

ArReReAcAv

KeAAAABiBuCoCy

ADdptrNdran

03do

Journal of Ethnopharmacology 119 (2008) 673–679

Contents lists available at ScienceDirect

Journal of Ethnopharmacology

journa l homepage: www.e lsev ier .com/ locate / je thpharm

he in vitro biological activity of selected South African Commiphora species

.P. Paraskevaa, S.F. van Vuurena, R.L. van Zyla, H. Davidsa, A.M. Viljoenb,∗

epartment of Pharmacy and Pharmacology, Faculty of Health Sciences, University of the Witwatersrand, 7 York Road, Parktown 2193, South Africaepartment of Pharmaceutical Sciences, Tshwane University of Technology, Private Bag X680, Pretoria 0001, South Africa

r t i c l e i n f o

ticle history:ceived 5 May 2008ceived in revised form 17 June 2008cepted 23 June 2008ailable online 2 July 2008

ywords:nticancernti-inflammatoryntimicrobialnti-oxidantological activityrseraceaemmiphoratotoxicity

a b s t r a c t

Ten South African Commiphora (Burseraceae) species were investigated to validate their use in traditionalhealing rites. The leaf and stem extracts of each species were analysed for the anti-oxidant (ABTS andDPPH assays), antimicrobial (MIC and death kinetic assays), anti-inflammatory (5-LOX assay), anticancer(SRB assay) properties, as well as the cytotoxic effects (tetrazolium-based assay). The best anti-oxidantactivity (ABTS assay) was observed for the stem extracts of Commiphora tenuipetiolata IC50 = 5.10 �g/ml),Commiphora neglecta (IC50 = 7.28 �g/ml) and Commiphora mollis (IC50 = 8.82 �g/ml). Extracts generallyexhibited poor anti-oxidant activity in the DPPH assay, with the exception of Commiphora schimperi(stem), Commiphora neglecta (stem), Commiphora tenuipetiolata (stem and leaf), and Commiphora edulis(stem), with IC50 values ranging between 7.31 and 10.81 �g/ml. The stem extracts exhibited moderate togood 5-LOX inhibitory activity with Commiphora pyracanthoides (stem) displaying the greatest inhibitoryeffect (IC50 = 27.86 ± 4.45 �g/ml). For the antimicrobial (MIC) assay, a greater selectivity was exhibitedby the extracts against the Gram-positive bacteria (0.01–8.00 mg/ml) and the yeasts (0.25–8.00 mg/ml)than against the Gram-negative bacteria (1.00–8.00 mg/ml). Using death kinetic studies (time–kill stud-ies), the rate at which Commiphora marlothii (stem) kills Staphylococcus aureus over a 24 h period wasdetermined. Mostly, a concentration-dependent antibacterial activity was observed beginning after ca.30 min. All concentrations exhibited antibacterial activity, with complete bactericidal effect achieved bythe 24th hour. The most active Commiphora species against the HT-29 cells (SRB anticancer assay) wereCommiphora glandulosa (leaf and stem) and Commiphora marlothii (leaf). The MCF-7 cells (SRB anticancerassay) exhibited the highest sensitivity to indigenous Commiphora species, with Commiphora edulis (leafand stem), Commiphora glandulosa (leaf and stem), Commiphora marlothii (leaf), Commiphora pyracan-thoides (leaf and stem), Commiphora schimperi (stem), and Commiphora viminea (stem) all possessing apercentage inhibition greater than 80% at 100 �g/ml. Commiphora glandulosa (leaf and stem) and Com-miphora pyracanthoides (leaf and stem) were the two most active species against the SF-268 cells (SRB

anticancer assay), with IC50 values ranging between 68.55 ± 2.01 and 71.45 ± 1.24 �g/ml. The majority ofthe Commiphora extracts were linvestigated in the MTT assay.

1

fo(otai

Abbreviations: ABTS, 2,2′-azino-bis(3-ethyl-benzthiazoline-6-sulfonic acid);TCC, American type culture collection; CFU, colony forming units; DMEM,ulbecco’s Modified Eagle’s Medium; DMSO, dimethyl sulfoxide; DPPH, 2,2-iphenyl-1-picrylhydrazyl; HPLC, high performance liquid chromatography; INT,-iodonitrotetrazolium; 5-LOX, 5-lipoxygenase; MIC, minimum inhibitory concen-

ation; MTT, 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-2H-tetrazolium bromide;aCl, sodium chloride; NCTC, national culture type collection; NDGA, nordihy-oguaiaretic acid; SRB, antiproliferative sulforhodamine B; TEAC, Trolox equivalentti-oxidant capacity; TCA, cold trichloroacetic acid; TSA, tryptone soya agar.∗ Corresponding author. Tel.: +27 12 3826360; fax: +27 12 3826243.

E-mail address: viljoenam@tut.ac.za (A.M. Viljoen).

u(epl

78-8741/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.i:10.1016/j.jep.2008.06.029

argely non-cytotoxic against Graham human kidney epithelial cells when

© 2008 Elsevier Ireland Ltd. All rights reserved.

. Introduction

The Burseraceae family consists of approximately 700 speciesrom 18 genera of which Commiphora is one. The name Commiphorariginates from the Greek words kommi (meaning ‘gum’) and pheromeaning ‘to bear’). The majority of the species yield a fragrantleo-gum-resin following damage to the bark (Steyn, 2003). Ofhe more than 200 species of Commiphora native to the season-lly dry tropics of Africa, Arabia and India, about 40 species occurn southern Africa. Traditionally, Commiphora species have been

sed in southern Africa for the treatment of colds (stem), feverstem), malaria (stem), typhoid (fruit), wound healing (resinousxudates), as an antiseptic (resinous exudates), snake and scor-ion bites, tumours (bark, resin, leaf), stomach aches (bark, resin,

eaf), diseases of the gall bladder (bark), chest ailments (roots) and

6 hnoph

s2atmtetipiw

smCaiw

2

2

iaama

2

2ia

2

twaedtpwDtgiiwtepvVtcs

t

bstat75saealw

2

2

m(AbiCtBopDAmsm(w4

(ovT(dCwr

2

MaaCttiFNca

74 M.P. Paraskeva et al. / Journal of Et

kin infections (resin) (Kokwaro, 1976; Lemenih et al., 2003; Steyn,003). Commiphora species have been a source of several novelnd bio-active compounds. Previous studies on this genus includehe isolation of four active compounds, namely mansumbinone,

ansumbinoic acid, picropolygamain, lignan-1(methoxy-1,2,3,4-etrahydropolygamain) from Commiphora kua var. kua. Also, thextract of stem bark yielded three labile C22 octanordammareneriterpenes. The dihydroflavonol glucoside, phellamurin has beensolated from Commiphora africana and seven dammarene triter-enes from the stem bark of Commiphora dalzielli have been

solated. Pentacyclic triterpene with anti-inflammatory activityas isolated from Commiphora merkeri (Hanus et al., 2005).

The botanical diversity of this genus in South Africa warrants atudy to provide scientific evidence for the traditional use of Com-iphora species in African healing rites. Thus, 10 South African

ommiphora species were investigated. In this study, the leafnd stem extracts of each species were analysed for their anti-nflammatory, anti-oxidant, antimicrobial, anticancer activities as

ell as the cytotoxic effects.

. Materials and methods

.1. Collection of plant material

Aerial parts (leaves and stems) of 10 Commiphora species weredentified and collected at random from natural populations withinselected geographical region in the Limpopo Province. The leavesnd stems were air-dried and ground separately. Voucher speci-ens were prepared and deposited at the Department of Pharmacy

nd Pharmacology, University of the Witwatersrand.

.2. Preparation of extracts

Extracts were prepared over a period of 6 h (three extractions ofh each). The leaves and stems (10 g of each) underwent extraction

n a conical flask using chloroform:methanol (1:1) in a water batht 40 ◦C.

.3. Determination of anti-oxidant activity

Two assays, viz. the 2,2-diphenyl-1-picrylhydrazyl (DPPH) andhe 2,2′-azino-bis(3-ethyl-benzthiazoline-6-sulfonic acid) (ABTS)ere employed for the determination of potential anti-oxidant

ctivity. The ABTS anti-oxidant assay is also known as the Troloxquivalent anti-oxidant capacity (TEAC) assay. The DPPH assay, asescribed by Shimada et al. (1992), was employed to determinehe radical scavenging activity of the plant extracts. Aliquots oflant extract dissolved in dimethyl sulfoxide (DMSO, Saarchem)ere plated out in triplicate in a 96-well microtiter plate. ThePPH solution (Fluka) was added to alternating columns of the

est samples and high performance liquid chromatography (HPLC)rade methanol (Ultrafine Limited) for control of test samples,n the remaining columns. The plate was shaken for 2 min andncubated for 30 min in the dark. The percentage decolourisationas obtained spectrophotometrically at 550 nm using the Labsys-

ems Multiskan RC microtiter plate reader, linked to a computerquipped with GENESIS® software. Percentage decolourisation waslotted against the concentration of the sample and the IC50alues were determined using Enzfitter® version 1.05 software.itamin C (l-ascorbic acid) and TroloxTM (6-hydroxy-2,5,7,8-

etramethylchromon-2-carboxylic acid) were used as positiveontrols. At least three independent tests were performed for eachample.

The quenching of the ABTS radical cation results in the evalua-ion of the radical scavenging activity. This assay was first reported

cptaa

armacology 119 (2008) 673–679

y Miller et al. (1993) and Rice-Evans (1994). Stock solutions anderial dilutions of the plant extracts were prepared in DMSO. Theotal scavenging capacity of the extracts was quantified through theddition of ABTS+ (Sigma–Aldrich) to the plant extract. The solu-ions were heated to 37 ◦C for 4 min and the absorbance read at34 nm on a spectrophotometer (Milton Roy Spectronic GENESYS). All assays were done in triplicate. The percentage decolouri-ation was calculated relative to the control. TroloxTM was useds the positive control and ethanol as the negative control. Thextent of inhibition of the absorbance of the ABTS+ was plotted asfunction of the concentration to determine the TroloxTM equiva-

ent anti-oxidant capacity. The IC50 values and standard deviationsere determined.

.4. Determination of antimicrobial activity

.4.1. Minimum inhibitory concentration assayThe minimum inhibitory concentration (MIC) values were deter-

ined and modified from the microtiter plate dilution methodEloff, 1998) on two Gram-positive bacteria (Staphylococcus aureusTCC 6358 and Bacillus cereus ATCC 11778), two Gram-negativeacteria (Klebsiella pneumoniae NCTC 9633 and Pseudomonas aerug-

nosa ATCC 9027) and two yeasts (Candida albicans ATCC 10231 andryptococcus neoformans ATCC 90112). The reference cultures (withhe exception of Candida albicans, obtained from the South Africanureau of Standards) were obtained from the National Health Lab-ratory Services, Johannesburg. Stock solutions of the respectivelant extracts were prepared by dissolving dry plant extract inMSO, as a result of the insolubility of certain extracts in acetone.liquots of the stock solution were transferred aseptically into aicrotiter plate, and serial dilutions (1:1) of the plant extract with

terile water were carried out several times. An equal volume oficrobial culture at a concentration of 1 × 108 colony forming units

CFU)/ml, grown in tryptone soya broth (Oxiod) was added to eachell prior to being incubated at 37 ◦C. Overnight incubation and8 h incubation followed for bacteria for the yeasts, respectively.

After incubation, a 0.4 mg/ml solution of p-iodonitrotetrazoliumINT, Sigma–Aldrich) was added to each well (40 �l) as an indicatorf microbial growth. The plates were incubated at 25 ◦C and the MICalues visually determined after 6 h (bacteria) and 24 h (yeasts).he concentration that inhibited bacterial/yeast growth completelythe first clear well) was taken as the MIC value. MIC values wereetermined at least in duplicate and repeated to confirm activity.iprofloxacin (Sigma–Aldrich) and amphotericin B (Sigma–Aldrich)ere used as the positive controls for the bacterial and yeast strains,

espectively. Negative controls contained only DMSO.

.4.2. Death kinetic assayBased on the preliminary promising results obtained from the

IC determination, Commiphora marlothii (stem) was identifieds a suitable candidate for the inactivation broth death kineticssay, as described by Lattaoui and Tantaoui-Elaraki (1994) andhristoph et al. (2001). Staphylococcus aureus (ATCC 6538) was cul-ured overnight on tryptone soya agar (TSA) at 37 ◦C, after whichhe resulting colonies were removed from the agar and used tonoculate a sterile 0.9% sodium chloride (NaCl, Labchem) solution.rom this cell suspension, serial dilutions were prepared in 0.9%aCl in order to obtain a suspension with an appropriate colonyount of 1 × 106 CFU/ml. This was obtained by performing dilutionsnd counting viable colonies in order to extrapolate the bacterial

ell concentration in the inoculum suspension. Solutions were pre-ared (extract dissolved in DMSO, broth and bacterial suspension)o yield final plant extract concentrations of 0.125, 0.25, 0.5, 0.75nd 1.0% (w/v), and were stabilised at 37 ◦C in a shaking water bath,fter which bacterial inoculum was added. A sample of each incu-

hnoph

b1twfqraqteAs

2

ea(etl0p1tibaietvw(

2

eocnl

tlicwveTtmAifwamTfi5

dtadpsstvtmA

2

btckmhlpa

emiiee(fsffAA1aTtsg

3

3

OalwlCpt

M.P. Paraskeva et al. / Journal of Et

ated test solution was removed at standard time intervals of 0, 5,5, 30, 60, 120 and 240 min as well as at 8, 24 and 48 h, and addedo inactivation broth (0.1%, w/v peptone, 5%, w/v lecithin and 5%,/v yeast extract). All samples were analysed in duplicate, except

or those taken at 48 h, which were taken for the sole purposes ofualitative evaluation of re-growth. Four serial dilutions, from theesulting solutions, were prepared in 0.9% NaCl solution, of whichliquots were spread out onto TSA plates. The plates were subse-uently incubated at 37 ◦C for 24 h. After the incubation period,he colonies on each of the plates were counted and the resultsxpressed in a log10 reduction time–kill plot of CFU/ml versus time.control was included to evaluate staphylococcal growth using the

ame broth formulation with solvent in the absence of plant extract.

.5. Anti-inflammatory activity

To investigate the anti-inflammatory activity of the plantxtracts, the in vitro 5-lipoxygenase (5-LOX) assay was performed,s determined by Sircar et al. (1983) and later modified by Evans1987). The standard assay mixture contained 10 �l of the plantxtract dissolved in DMSO and Tween® 20 (Merck). The addi-ion of 0.1 M potassium phosphate buffer (pH 6.3, 25 ◦C), 100 �Minoleic acid (Fluka), 100 units isolated 5-LOX enzyme diluted with.1 M cold potassium phosphate buffer constitutes each test sam-le run. Increases in absorbance were recorded at 234 nm for0 min, using a UV–VIS Analytikjena Specord 40 spectrophotome-er, equipped with Winaspect® software. The percentage enzymenhibition attributable to each of the extracts was then determinedy comparison with the negative control, comprising of DMSOnd Tween® 20 in the absence of plant extract. Nordihydrogua-aretic acid (NDGA) was used as the positive control. The percentagenzyme inhibition exhibiting anti-inflammatory activity was plot-ed against the concentration of plant extract (�g/ml). The IC50alues (concentration at which 50% of the enzyme was inhibited)ere determined from the dose–response curves using Enzfitter®

version 1.05) software. The assay was undertaken in triplicate.

.6. The anticancer activity

The antiproliferative sulforhodamine B (SRB) assay wasmployed according to Monks et al. (1991) to assess the abilityf extracts to inhibit the in vitro growth of three human cancerell lines, namely the colon adenocarcinoma (HT-29), breast ade-ocarcinoma (MCF-7), and the neuronal glioblastoma cancer cell

ine (SF-268).The HT-29, MCF-7 and SF-268 cell lines were obtained from

he National Cancer Institute (USA). The MCF-7 and SF-268 cellines were maintained with RPMI-1640 and HT-29 cells were grownn Dulbecco’s Modified Eagle’s Medium (DMEM, Highveld Biologi-al). Stock solutions (50 mg/ml methanol) of the respective extractsere used to prepare plant samples in experimental medium, and

arious concentrations (100, 50, 25, 12.5 and 6.25 �g/ml) of thextracts were plated out in 96-well microtiter plates in triplicate.he following controls were prepared: (i) methanol in experimen-al media as a negative control, (ii) plant extract with experimental

edia and no cells, (iii) positive control: 5′-fluorouracil (Fluka).liquots of cell suspension (200 �l; 150,000 cells/ml) were seeded

nto a 96-well microtitre plate. The plates were incubated at 37 ◦Cor 24 h to facilitate the attachment of the cells to the bottom of theells. No cells were seeded into the blank wells; instead media was

dded. Plant extract and serial twofold dilutions with cell cultureedium was added to the wells already containing cell suspension.

he plates were incubated for a further 48 h at 37 ◦C. The cells werexed to the bottom of the well by layering the medium with cold0% (w/v) (50 �l) trichloroacetic acid (TCA, Saarchem) prepared in

C(

tm

armacology 119 (2008) 673–679 675

istilled water. The plates were incubated at 4 ◦C for 1 h, after whichhe supernatant was washed from the wells. SRB (0.4%, w/v) wasdded and then washed with 1% acetic acid to remove any unboundye. Trisma base (pH 10.5, 10 mM) was added to all wells and thelate was shaken at 960 rpm for 3 min on a microplate reader (Lab-ystems iEMS Reader MF) equipped with the Ascent® version 2.4oftware program. The absorbance was then read at 492 nm wherehe colour intensity of each well corresponded to the number ofiable cells. The percentage inhibition of cells was calculated usinghe Enzfitter® version 1.05 software. The IC50 values were deter-

ined from the log sigmoid dose–response profile for each sample.ssays were undertaken in triplicate.

.7. Cytotoxicity

The 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-2H-tetrazoliumromide (MTT) cellular viability assay (Mosmann, 1983) was usedo determine the cytotoxicity of the stem and leaf extracts at con-entrations 200, 150, 100 and 20 �g/ml on transformed humanidney epithelium (Graham) cells. These cells were continuouslyaintained in HAM F10 culture medium, incubated at 37 ◦C in a

umidified atmosphere of 5% CO2. Aliquots (180 �l of 0.25 mil-ion cells/ml) were seeded into the 96-well microtiter plates. Thelates were then incubated at 37 ◦C in 5% CO2 for 6 h to facilitatettachment of the cells to the bottom of the wells.

Stock solutions (50 mg/ml) in methanol of the respectivextracts were used to prepare plant samples in experimentaledium and various concentrations of the extracts were plated out

n 96 well microtiter plates. The controls comprised of: (i) methanoln experimental media (negative control), (ii) plant extract withxperimental media in the absence of cells (colour control), (iii)xperimental media in absence of both plant extract and cellsblank control), and (iv) quinine (Fluka). The plates were incubatedor 44 h, after which 12 mM MTT solution in phosphate-bufferedaline (pH 7.4), was added to each well before being incubatedor a further 4 h. The supernatant was discarded, DMSO addedor dissolution and measurement of the formed formazan crystals.

Labsystems iEMS Reader MF microplate reader, equipped withscent® version 2.4 software was employed, agitating the plates at020 rpm for 4 min, while measuring and recording absorbance’s attest wavelength of 540 nm and a reference wavelength of 690 nm.he results were expressed in terms of percentage cellular viabilityaking the relevant controls into account. The IC50 value for eachample was determined from a log sigmoid dose–response curveenerated by Enzfitter®. The assay was performed in triplicate.

. Results and discussion

.1. Anti-oxidant activity

The radical scavenging potential is summarised in Table 1.verall, the Commiphora extracts displayed good anti-oxidantctivity with 11 of the 18 (61.1%) extracts possessing IC50 valuesess than 20 �g/ml. The best anti-oxidant activity (ABTS assay)as observed for the stem extracts of Commiphora tenuipetio-

ata IC50 = 5.10 �g/ml), Commiphora neglecta (IC50 = 7.28 �g/ml) andommiphora mollis (IC50 = 8.82 �g/ml). Extracts generally exhibitedoor anti-oxidant activity in the DPPH assay, with the excep-ion of Commiphora schimperi (stem), Commiphora neglecta (stem),

ommiphora tenuipetiolata (stem and leaf), and Commiphora edulisstem), with IC50 values ranging between 7.31 and 10.81 �g/ml.

Stem extracts produced much greater anti-oxidant activity thanhe leaf extracts. With the exception of Commiphora africana, Com-iphora tenuipetiolata and Commiphora viminea (DPPH assay) and

676M

.P.Paraskevaet

al./JournalofEthnopharmacology

119(2008)

673–679

Table 1In vitro anti-oxidant, antimicrobial and anti-inflammatory properties of selected Commiphora extracts

Species Authority and vouchernumber

Anti-oxidant activity (�g/ml) Anti-inflammatoryactivity IC50 (�g/ml)

Antimicrobial activity MIC (mg/ml)

ABTS IC50 DPPH IC50 PseudomonasaeruginosaATCC 9027

KlebsiellapneumoniaeNTCC 9633

Bacillus cereusATCC 11778

Staphylococcusaureus ATCC 6538

Candida albicansATCC 10231

CryptococcusneoformansATCC 90112

Commiphora africana (leaf)(A. Rich.) Engl. AV1080

29.64 ± 3.81 43.00 ± 1.37 n.d. 8.00 4.00 4.00 8.00 2.00 0.25Commiphora africana (stem) 12.97 ± 1.23 39.44 ± 1.70 n.d. 8.00 4.00 4.00 4.00 2.00 4.00Commiphora edulis (leaf)

(Klotzsch) Engl. AV1089n.d. 59.70 ± 1.97 n.d. 8.00 2.00 4.00 8.00 4.00 2.00

Commiphora edulis (stem) 23.75 ± 3.51 10.59 ± 0.50 55.61 ± 1.25 8.00 n.a. 4.00 4.00 4.00 2.00Commiphora glandulosa (leaf)

Schinz AV108812.19 ± 0.16 41.39 ± 1.73 50.54 ± 14.70 8.00 2.00 0.01 4.00 4.00 0.50

Commiphora glandulosa (stem) 10.69 ± 1.47 27.27 ± 0.15 66.16 ± 3.61 8.00 4.00 0.01 1.00 1.00 1.00Commiphora marlothii (leaf)

Engl. AV108317.66 ± 0.75 66.81 ± 0.43 n.d. 8.00 2.00 2.00 1.00 2.00 1.00

Commiphora marlothii (stem) 15.67 ± 1.79 32.16 ± 1.72 n.d. 8.00 4.00 0.31 1.00 1.00 1.00Commiphora mollis (leaf)

(Oliv.) Engl. AV108260.11 ± 8.73 89.95 ± 0.04 n.d. 8.00 2.00 8.00 8.00 1.00 8.00

Commiphora mollis (stem) 8.82 ± 0.72 22.17 ± 0.33 53.98 ± 1.59 8.00 4.00 2.00 4.00 2.00 1.00Commiphora neglecta (leaf)

I.Verd AV1085n.d. 98.61 ± 1.97 n.d. 8.00 4.00 4.00 8.00 2.00 2.00

Commiphora neglecta (stem) 7.28 ± 0.29 10.36 ± 1.89 61.65 ± 8.98 8.00 4.00 4.00 8.00 2.00 0.50Commiphora pyracanthoides

(leaf)Engl. AV1081

51.44 ± 0.27 29.32 ± 5.22 n.d. 8.00 1.00 2.00 2.00 1.00 0.25

Commiphora pyracanthoides(stem)

18.68 ± 8.84 19.02 ± 0.12 27.86 ± 4.45 8.00 n.a. 0.04 1.00 0.50 1.00

Commiphora schimperi (leaf)(O.Berg) Engl. AV1084

25.25 ± 1.74 55.30 ± 3.73 76.22 ± 4.84 8.00 4.00 4.00 8.00 1.00 0.50Commiphora schimperi (stem) 11.22 ± 3.61 7.31 ± 0.14 58.38 ± 13.88 8.00 4.00 2.00 4.00 1.00 1.00Commiphora tenuipetiolata

(leaf)Engl. AV1087

17.47 ± 1.30 10.81 ± 0.56 n.d. 8.00 2.00 2.00 2.00 0.50 1.00

Commiphora tenuipetiolata(stem)

5.10 ± 0.66 10.75 ± 0.36 53.58 ± 10.44 8.00 2.00 2.00 4.00 0.50 1.00

Commiphora viminea (leaf)Burtt Davy AV1086

45.89 ± 0.79 78.49 ± 3.46 n.d. 8.00 2.00 2.00 8.00 1.00 0.25Commiphora viminea (stem) 26.30 ± 0.23 84.01 ± 7.07 62.97 ± 11.64 8.00 4.00 0.23 1.00 1.00 0.25Controls 5.41 ± 0.51a 4.18 ± 0.56b;

7.03 ± 1.16a4.95 ± 0.07c 2.5 × 10−3d 2.5 × 10−3d 3.13 × 10−4d 2.5 × 10−3d 1.25 × 10−3e 1.25 × 10−3e

n.d.: not determined due to insufficient plant material; n.a.: considered not active, as MIC value of test sample is equivalent to DMSO control.a TroloxTM.b Vitamin C.c NDGA.d Ciprofloxacin.e Amphotericin B.

hnopharmacology 119 (2008) 673–679 677

CwrnaaliOoentwctmtdtcoaim(poatatae(sAtrsaa

3

aablecClaSm(inlavir

Fc

toP

mrtca(

3

e6ple5m(

3

oTvtT5Pad

somAiflta

M.P. Paraskeva et al. / Journal of Et

ommiphora glandulosa and Commiphora marlothii (ABTS assay)hich demonstrated analogous anti-oxidant activity. Pronounced

adical scavenging activity has been reported in plants with phe-olic moieties, the presence of which is common in naturalnti-oxidants. These phenolic moieties include substances suchs tannins, flavonoids, tocopherols and catecheses. Tannins are, ateast in part, responsible for the strong free radical scavenging activ-ties working synergistically with other anti-oxidant substances.rganic acids and protein hydrolysates may additionally act as anti-xidants (Dapkevicius et al., 1998). The phenomenon that the stemxtracts demonstrated substantial radical scavenging capacity isot completely surprising, in view of the high polyphenolic con-ent. Most extracts displayed greater activity in the ABTS assay,ith the exception of Commiphora edulis (stem), Commiphora pyra-

anthoides (leaves), Commiphora schimperi (stem) and Commiphoraenuipetiolata (leaves). The DPPH and ABTS assays have the same

echanism of action, but, in most cases, the results obtained fromhe ABTS assay are higher than those from DPPH assay. It has beenocumented that results reported for the ABTS assay do not onlyake into account the activity of the parent compound, but also theontribution of reaction products and other individual compoundsn the activity, which is not the case in the DPPH assay (Arts etl., 2003). In the ABTS assay, the total amount of radical scavengeds measured over a period of time, while most anti-oxidant assays

easure the rate at which a radical is scavenged by an anti-oxidantArts et al., 2003). The total amount of ABTS+ scavenged by a com-ound correlates with the biological activity in a selected groupf flavonoids (Arts et al., 2003). It was also determined by Lee etl. (2003) that the total anti-oxidant capacity of the samples beingested in the ABTS and DPPH assays correlated with the phenolicnd flavonoid content. However, the determined values indicatedhat the DPPH assay underestimates the anti-oxidant capacity bypproximately 30%, when compared to the ABTS assay. This differ-nce has previously been reported by Kim et al. (2002) and Arnao2000), and may be attributed to the absorbance interruption at thepecified wavelength by other compounds in the DPPH assay. TheBTS assay is also a very sensitive assay requiring only a short reac-

ion time of approximately 4 min (versus the 30 min reaction timeequired for DPPH), and can be used in both organic and aqueousolvent systems (Lee et al., 2003), i.e. the application of the ABTSssay is for both hydrophilic and lipophilic compounds (Mathewnd Abraham, 2006).

.2. Antimicrobial activity

The crude stem and leaf extracts exhibited antimicrobialctivity mostly against the Gram-positive bacteria as well asgainst the yeasts (Table 1). The most promising activity wasy both the stem and leaf extract of Commiphora glandu-

osa against Bacillus cereus (MIC = 0.01 mg/ml). Three other stemxtracts demonstrated a strong inhibitory effect against Bacillusereus, these being Commiphora pyracanthoides (MIC = 0.04 mg/ml),ommiphora viminea (MIC = 0.23 mg/ml) and Commiphora mar-

othii (MIC = 0.31 mg/ml). Commiphora species displaying promisingctivity (MIC = 1.00 mg/ml) against the Gram-positive bacteriumtaphylococcus aureus were Commiphora glandulosa (stem), Com-iphora marlothii (stem and leaf), Commiphora pyracanthoides

stem) and Commiphora viminea (stem). Greater inhibitory activ-ty was mostly observed against the yeasts especially Commiphoraeoformans with MIC values ranging from 0.25 mg/ml (in the

eaf extracts of Commiphora africana, Commiphora pyracanthoidesnd Commiphora viminea and the stem extract of Commiphoraiminea) to 8.00 mg/ml in the leaf extract of Commiphora mollis. Thenhibitory activity of Commiphora spp. against Commiphora albicansanged from 0.50 to 4.00 mg/ml. In general, the Gram-negative bac-

sCta

ig. 1. Time–kill efficacy of Commiphora marlothii (stem) extract against Staphylo-occus aureus (ATTC 6538).

eria displayed the least sensitivity towards the extracts, and allf the plant extracts exhibited poor and unvaried activity againstseudomonas aeruginosa.

For the time–kill study (Fig. 1) the efficacy of Commiphoraarlothii (stem) extract against Staphylococcus aureus displayed a

eduction in CFU/ml after ca. 30 min of exposure of all concentra-ions tested. A complete bactericidal effect was achieved for alloncentrations with concentrations 0.5, 0.75 and 1% having cidalctivity within 4 h. Furthermore, no re-growth occurred after 48 hnot shown in Fig. 1), except at a concentration of 0.13% (w/v).

.3. Anti-inflammatory activity

In the 5-LOX assay, all the stem extracts displayed good to mod-rate enzyme inhibition. IC50 values ranged from 27.86 ± 4.45 to6.16 ± 3.61 �g/ml; with Commiphora pyracanthoides (stem) dis-laying the best activity (27.86 ± 4.45 �g/ml) which is sixfold

ess effective than nordihydroguaiaretic acid. In contrast, the leafxtracts of eight species displayed minimal inhibition of the-LOX enzyme with IC50 values being reported only for Com-iphora schimperi (76.22 ± 4.84 �g/ml) and Commiphora glandulosa

50.54 ± 14.70 �g/ml).

.4. Anticancer activity

The inhibition of cancer cell proliferation and cellular viabilityf the extracts against three human cancer cell lines is detailed inable 2. The percentage cell growth inhibition at 100 �g/ml and IC50alues for those plant extracts that showed more than 80% inhibi-ion of cells (at 100 �g/ml) was evaluated for all cell lines tested.his would require that the inhibitory concentration at which0% of the cells are inhibited (IC50) be 30 �g/ml (Suffiness andezzuto, 1990). For those Commiphora species exhibiting anticancerctivity, the inhibitory effect was observed to be concentration-ependent.

The most promising activity against the HT-29 cells was pre-ented by Commiphora glandulosa (leaf and stem), with IC50 valuesf 52.34 ± 1.95 and 57.89 ± 2.04 �g/ml, respectively, and Com-iphora marlothii (leaf), with an IC50 value of 72.12 ± 3.14 �g/ml.t 100 �g/ml, the leaf extracts exhibited a slightly higher

nhibitory effect. However, comparing the results to that of 5′-uorouracil (IC50 = 7.00 ± 2.20 �g/ml), the inhibitory activity ofhe most active extract against this cell line was sevenfold lessctive.

In general, the leaf extracts were more active than thetem extracts against the HT-29 cell line, with the exception ofommiphora africana, Commiphora edulis and Commiphora pyracan-hoides, while the stem extracts were observed to be more activegainst the MCF-7 and SF-268 cell lines, with the exception of

678 M.P. Paraskeva et al. / Journal of Ethnoph

Tab

le2

Invi

tro

anti

can

cer

acti

vity

and

cyto

toxi

city

ofse

lect

edCo

mm

ipho

raex

trac

ts

Spec

ies

%C

ellg

row

thin

hib

itio

n(C

GI)

and

the

IC50

valu

esin

the

SRB

anti

can

cer

assa

yC

ytot

oxic

ity

IC50

valu

e(�

g/m

l)

%C

GIo

fHT-

29at

100

�g/

ml

IC50

valu

eH

T-29

(�g/

ml)

%C

GIo

fMC

F-7

at10

0�

g/m

lIC

50va

lue

MC

F-7

(�g/

ml)

%C

GIo

fSF-

268

at10

0�

g/m

lIC

50va

lue

SF-2

68

(�g/

ml)

Com

mip

hora

afri

cana

(ste

m)

29.0

1±

3.42

n.d

.53

.00

±4.

82n

.d.

0.04

±3.

43n

.d.

>20

0.0

Com

mip

hora

afri

cana

(lea

f)0.

01±

0.92

n.d

.31

.10

±3.

75n

.d.

0.02

±4.

71n

.d.

>20

0.0

Com

mip

hora

edul

is(s

tem

)36

.26

±2.

54n

.d.

80.5

0±

4.6

867

.85

±5.

4624

.32

±4.

69n

.d.

194.

0±

8.4

8Co

mm

ipho

raed

ulis

(lea

f)33

.18

±1.

69n

.d.

82.1

0±

1.99

50.3

4±

4.25

47.5

3±

3.67

n.d

.99

.5±

0.71

Com

mip

hora

glan

dulo

sa(s

tem

)86

.41

±1.

9357

.89

±2.

0495

.80

±3.

0924

.28

±0.

1886

.20

±3.

3570

.32

±2.

4530

.5±

3.54

Com

mip

hora

glan

dulo

sa(l

eaf)

90.2

5±

0.42

52.3

4±

1.95

89.2

0±

4.37

39.5

8±

2.11

83.2

1±

3.60

71.4

5±

1.24

106.

5±

3.53

Com

mip

hora

mar

loth

ii(s

tem

)59

.13

±1.

83n

.d.

77.4

0±

2.09

n.d

.0.

05±

2.58

n.d

.12

3.0

±4.

24Co

mm

ipho

ram

arlo

thii

(lea

f)82

.48

±3.

4572

.12

±3.

1484

.20

±2.

7263

.27

±1.

5135

.43

±4.

83n

.d.

97.5

±0.

71Co

mm

ipho

ram

ollis

(ste

m)

29.3

8±

1.90

n.d

.74

.30

±1.

64

n.d

.57

.40

±6.

65n

.d.

172.

0±

1.41

Com

mip

hora

mol

lis(l

eaf)

42.9

7±

4.59

n.d

.60

.80

±3.

75n

.d.

0.01

±3.

97n

.d.

>20

0.0

Com

mip

hora

negl

ecta

(ste

m)

14.3

8±

2.77

n.d

.76

.80

±3.

41n

.d.

28.9

1±

3.75

n.d

.>2

00.

0Co

mm

ipho

rane

glec

ta(l

eaf)

18.8

8±

2.66

n.d

.61

.45

±6.

98n

.d.

0.01

±4.

03n

.d.

111.

5±

4.95

Com

mip

hora

pyra

cant

hoid

es(s

tem

)77

.46

±2.

62n

.d.

80.6

5±

2.23

20.6

0±

0.73

80.0

0±

3.77

69.3

6±

1.6

410

1.5

±0.

71Co

mm

ipho

rapy

raca

ntho

ides

(lea

f)36

.57

±2.

23n

.d.

86.5

0±

1.37

35.5

1±

3.57

94.8

5±

1.92

68.

55±

2.01

104.

0±

7.07

Com

mip

hora

schi

mpe

ri(s

tem

)0.

02±

3.47

n.d

.88

.30

±1.

8883

.14

±5.

100.

02±

3.59

n.d

.13

6.5

±0.

71Co

mm

ipho

rasc

him

peri

(lea

f)7.

43±

1.37

n.d

.24

.90

±5.

43n

.d.

0.05

±3.

72n

.d.

>20

0.0

Com

mip

hora

tenu

ipet

iola

ta(s

tem

)10

.63

±2.

45n

.d.

71.0

0±

0.70

n.d

.36

.72

±6.

02n

.d.

>20

0.0

Com

mip

hora

tenu

ipet

iola

ta(l

eaf)

15.4

2±

1.39

n.d

.28

.26

±3.

93n

.d.

4.73

±2.

83n

.d.

>20

0.0

Com

mip

hora

vim

inea

(ste

m)

48.

00

±2.

03n

.d.

88.9

0±

4.92

96.0

2±

0.42

21.7

6±

4.58

n.d

.>2

00.

0Co

mm

ipho

ravi

min

ea(l

eaf)

53.7

9±

4.82

n.d

.59

.20

±5.

44

n.d

.5.

63±

4.37

n.d

.14

1.5

±7.

78C

ontr

ols

96.7

6±

1.19

a7.

00

±2.

20a

98.8

1±

0.93

a1.

11±

0.31

a20

.70

±0.

86a

n.d

.13

6.1

±4.

06b

n.d

.:n

otd

eter

min

edd

ue

toin

suffi

cien

tp

lan

tm

ater

ial;

a5′ -

flu

orou

raci

l;bqu

inin

e.

Cc

aatttttgmtts(dw

ndtbraoa

3

htDptMekvtlalt

4

Ctpsooupgiai

mmc

armacology 119 (2008) 673–679

ommiphora edulis, Commiphora marlothii and Commiphora pyra-anthoides (Table 2).

The extracts of each of the species were analysed using HPLC,nd it was determined that the leaf extracts contain varyingmounts of flavonoids (Paraskeva, 2007). Flavonoids are knowno have anticancer activity, and may to some degree contributeo the observed activity. Flavonoids are potent antiprolifera-ive agents, in which the C2–C3 double bond and the lack ofhe 6-hydroxyl group are important structural requirements forheir cytostatic effects (Rusak et al., 2005). The percentage cellrowth inhibition observed at the highest concentration for Com-iphora mollis (leaf) was observed to be 42.97 ± 4.59% against

he HT-29 cells. This leaf extract showed similar flavonoid pat-erns to Commiphora africana and Commiphora schimperi, andimilarities in their HPLC chromatograms. Commiphora africana%CGI = 0.02 ± 4.71) and Commiphora schimperi (%CGI = 7.43 ± 1.37)emonstrated a decreased inhibitory effect against this cell line;hich differed from that of Commiphora mollis.

The MCF-7 cells exhibited the highest sensitivity to indige-ous Commiphora species. Commiphora africana (leaf and stem)isplayed cancer cell specificity against the MCF-7 cell line, wherehe percentage cell growth inhibition at 100 �g/ml was found toe 53.00 ± 4.82 and 31.00 ± 3.75% for the stem and leaf extracts,espectively. No cell inhibition was observed in SF-268 cells, as wells no cell inhibition by the leaf extract against HT-29 cell line, withnly slight inhibitory activity by the stem extract (29.00 ± 3.42%)gainst this cell line.

.5. Cytotoxicity

The toxicity of the extracts (Table 2) against the transformeduman kidney epithelium cells, in the MTT assay versus its cyto-oxicity against cancer cell lines in the SRB assay may be compared.ata from several hundred agents screened in a MTT assay, inarallel with the SRB assay, indicated that under similar experimen-al conditions and within the applied limits of data analyses, the

TT and SRB assays generally yielded similar results (Rubinsteint al., 1990; Monks et al., 1991). Cytotoxicity against the humanidney epithelium cells was minimal, with the percentage celliability being far greater than the percentage cell viability ofhe cancer cell lines, with the exception of Commiphora glandu-osa (stem). This indicates the possibility of a selective anticancerctivity, but further investigations against a larger number of cellines and in vivo studies would need to be conducted to confirmhis.

. Conclusion

Although a plethora of research has been reported on exoticommiphora species, this study represents the first account onhe in vitro pharmacological activity of the South African counter-arts. Biological investigations indicated that certain Commiphorapecies displayed promising activity in the anti-inflammatory, anti-xidant, antimicrobial and anticancer assays. The biological activitybserved in Commiphora species provides a scientific basis for these of the plants in traditional medicines. Commiphora species haveroven to be relatively safe, with the exception of Commiphoralandulosa (stem), at the concentrations tested. While the toxic-ty against the cancer cell lines was more prevalent, cytotoxicitygainst the Graham cells was minimal. This indicates that the activ-

ty is selective.

It is important to correct the misguided belief that herbaledicines do not cause adverse effects, and even though thereay be no evidence of cytotoxicity in vitro, the possibility of in vivo

ytotoxicity cannot be excluded.

hnoph

A

Fpi

R

A

A

C

D

E

E

H

K

K

L

L

L

M

M

M

M

P

R

R

R

S

M.P. Paraskeva et al. / Journal of Et

cknowledgements

We acknowledge and are indebted to the National Researchoundation (Indigenous Knowledge Systems) for the financial sup-ort. We acknowledge the assistance of Mr. M. Steyn in the

dentification of plant material.

eferences

rnao, M.B., 2000. Some methodological problems in the determination of antioxi-dant activity using chromogen radicals: a practical case. Trends in Food Scienceand Technology 11, 419–423.

rts, M.J.T.J., Dallinga, J.S., Voss, H.P., Haenen, G.R.M.M., Bast, A., 2003. A criticalappraisal of the use of the anti-oxidant capacity (TEAC) assay in defining optimalanti-oxidant structures. Food Chemistry 80, 409–414.

hristoph, F., Stahl-Biskup, E., Kaulfers, P.-M., 2001. Death kinetics of Staphylococcusaureus exposed to commercial tea tree oils. Journal of Essential Oil Research 13,98–102.

apkevicius, A., Venskutanis, R., Van Beek, T.A., Linssen, J.P.H., 1998. Anti-oxidativeactivity of extracts obtained by different isolation procedures from some aro-matic herbs grown in Lithuania. Journal of the Science of Food and Agriculture77, 140–146.

loff, J.N., 1998. A sensitive and quick microplate method to determine the mini-mal inhibitory concentration of plant extracts for bacteria. Planta Medica 64,711–713.

vans, A.T., 1987. Actions of cannabis constituents on enzymes of arachido-nate metabolism anti-inflammatory potential. Biochemical Pharmacology 36,2035–2037.

anus, L.O., Rezanka, T., Dembitsky, V.M., Moussaieff, A., 2005. Myrrh—CommiphoraChemistry. Biomedical Papers of the Medical Faculty of the University of thePalacky Olomouc, Czechoslovakia 149, 3–28.

im, D.-O., Lee, K.W., Lee, H.J., Lee, C.Y., 2002. Vitamin C equivalent anti-oxidant(VCEAC) of phenolic phytochemicals. Journal of Agricultural and Food Chemistry

50, 3713–3717.

okwaro, O., 1976. Medicinal Plants of East Africa. East African Literature Bureau,Kampala, Nairobi, Dar as Salaam.

attaoui, N., Tantaoui-Elaraki, A., 1994. Comparative kinetics of microbial destructionby the essential oils of Thymus broussonettii, T. zygis and T. satureioides. Journalof Essential Oil Research 6, 165–171.

S

SS

armacology 119 (2008) 673–679 679

ee, K.W., Kim, Y.J., Lee, H.J., Lee, C.Y., 2003. Cocoa has more phenolic phytochemicalsand a higher anti-oxidant capacity than teas and red wine. Journal of Agriculturaland Food Chemistry 51, 7292–7295.

emenih, M., Abebe, T., Olsson, M., 2003. Gum and resin resources from some Acacia,Boswellia and Commiphora species and their economic contributions in Liban,south-east Ethiopia. Journal of Arid Environments 55, 465–482.

athew, S., Abraham, T.E., 2006. Studies on the anti-oxidant activities of cinna-mon (Cinnamomum verum) bark extracts, through various in vitro models. FoodChemistry 94, 520–528.

iller, N.J., Rice-Evans, C.A., Davies, M.J., Gopinathan, V., Milner, A., 1993. A novelmethod for measuring anti-oxidant capacity and its application to monitoringthe anti-oxidant status in premature neonates. Clinical Science 84, 407–412.

onks, A., Scudiero, D., Skehan, P., Shoemaker, R., Paull, K.D., Vistica, D., Hose, C.,Langley, J., Cronise, P., Vaigro-Wolff, A., Gary-Goodrich, M., Campbell, H., Mayo,J., Boyd, M.R., 1991. Feasibility of a high-flux anticancer drug screen using adiverse panel of cultured human tumour cell lines. Journal of the National CancerInstitute 183, 757–766.

osmann, T., 1983. Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. Journal of ImmunologicalMethods 65, 55–63.

araskeva, M.P., 2007. A Phytochemical and Pharmacological Study of Ten Com-miphora Species Indigenous to South Africa. MSc Dissertation, University of theWitwatersrand, South Africa.

ice-Evans, C.A., 1994. Formation of free radicals and mechanisms of action in nor-mal biochemical processes and pathological states. In: Free Radical Damage andits Control. Elsevier Science BV, Amsterdam, The Netherlands.

ubinstein, L.V., Shoemaker, R.H., Paull, K.D., Simon, R.M., Tosini, S., Skehan, P.,Scudiero, D.A., Monks, A., Boyd, M.R., 1990. Comparison of in vitro anticancer-drug-screening data generated with a tetrazolium assay versus a protein assayagainst a diverse panel of human tumour cell lines. Journal of the National CancerInstitute 82, 1113–1117.

usak, G., Gutzeit, H.O., Müller, J.L., 2005. Structurally related flavonoids with antiox-idative properties differentially affect cell-cycle progression and apoptosis ofhuman acute leukemia cells. Nutrition Research 25, 141–153.

himada, K., Fujikawa, K., Yahara, K., Nakamura, T., 1992. Anti-oxidative properties ofxanthan on the auto-oxidation of soybean oil in cyclodextrin emulsion. Journalof Agricultural and Food Chemistry 40, 945–948.

ircar, J.C., Shwender, C.F., Johnson, E.A., 1983. Soybean lipoxygenase inhibition bynonsteroidal anti-inflammatory drugs. Prostaglandins 25, 393–396.

teyn, M., 2003. Southern Africa Commiphora: United Litho South Africa.uffiness, M., Pezzuto, J.M., 1990. Assays related to cancer drug discovery. In:

Hostettmann, K. (Ed.), Methods in Plant Biochemistry: Assays for Bio-Activity.Academic Press, London.