A

ib1oaCMCdt1t©

K

1

siowg

aP

RT

0d

Journal of Ethnopharmacology 108 (2006) 228–235

In vitro anti-proliferative and antioxidant studies onDevil’s Club Oplopanax horridus

Joseph Tai a,∗, Susan Cheung a, Stefanie Cheah a,Edwin Chan b, David Hasman b

a Department of Pathology and Pediatrics, Center for Complementary Medicine Research, BC Child andFamily Research Institute, University of British Columbia, Vancouver, BC, Canada

b Forensic Science Center, British Columbia Institute of Technology, Burnaby, BC, Canada

Received 11 April 2005; received in revised form 8 May 2006; accepted 12 May 2006Available online 26 May 2006

bstract

Devil’s Club, Oplopanax horridus (OH), is a widely used folk medicine in Alaska and British Columbia for treating a variety of ailmentsncluding arthritis, fever and diabetes. HPLC profiling shows that numerous compounds are present in the 70% ethanolic extract of OH dry rootark powder. OH extract inhibited K562, HL60, MCF7 and MDA-MB-468 cell growth with the 50% inhibition (IC50) estimated at 1/2700, 1/1700,/500 and 1/2500 dilutions, respectively. Non-cytotoxic concentrations (<IC50) of OH extract combined with non-cytotoxic concentrations (<IC50)f camptothecin (CAM) or paclitaxel (PTX) were tested on the tumor cell lines. Of the 19 combinations tested, 9 showed additive or synergisticnti-proliferative effect while the rest showed antagonistic effect. Combination of OH extract at 1/4000 and 1/16,000 dilutions with 0.1 �M ofAM produced additive, anti-proliferative effects on K562 cells, as did 1/2000 and 1/1000 dilutions of OH combined with 2.5 nM of PTX onCF7, and 1/4000 dilution of OH with 1 nM of PTX on MDA-MB-468 cells. Combination of 1/8000 and 1/4000 dilution of OH with 0.05 �M ofAM showed strong synergistic anti-proliferative effects on HL60 cells. At non-cytotoxic 1/4000 dilution, OH induced 13.1% of HL60 cells to

ifferentiate into granulocytes but had no effect on monocyte/macrophage differentiation. A cell free hydroxyl radical scavenging assay estimatedhat OH at 1/100, 1/10 and 1/5 dilutions showed activities equivalent to 2.7, 15.7 and 25.6 �M of Trolox, respectively. At non-cytotoxic 1/4000 and/2000 dilutions, OH significantly reduced nitric oxide production by lipopolysaccharide activated RAW 264.7 cells (p < 0.001). Our data showhat the ethanolic extract of OH has anti-proliferative on several cancer cell lines, and has strong antioxidant activity.

2006 Elsevier Ireland Ltd. All rights reserved.

ifferen

Ltiia

eywords: North American Native herb; Oplopanax horridus; Cytotoxicity; D

. Introduction

Devil’s Club, Oplopanax horridus (Sm.) Miq., is a deciduoushrub that grows in the Pacific Northwest with most abundancen Alaska and British Columbia. This shrub is a member

f the Araliaceae family thus is related in taxonomy to theell-known medicinals such as the Asian ginseng (Panaxinseng C.A. Meyer), American ginseng (Panax quinquefolius

Abbreviations: ABTS, 2,2′-azino-bis-3-ethylbenzothi-azoline-6-sulfoniccid; CAM, camptothecin; LPS, lipopolysaccharide; OH, Oplopanax horridus;TX, paclitaxel; TEAC, Trolox equivalent antioxidant capacity∗ Corresponding author at: Child and Family Research Institute,oom L306, 4480 Oak Street, Vancouver, BC, Canada V6H 3V4.el.: +1 604 875 2457; fax: +1 604 875 2371.

E-mail address: jtai@interchange.ubc.ca (J. Tai).

tCta1nn1(pc

378-8741/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.jep.2006.05.018

tiation; Tumor cell lines; Antioxidant

.) and eleuthero (Eleutherococcus senticosus Maxim. or Acan-hopanax senticosus, formerly known as Siberian ginseng). Thenner barks and roots are widely used as folk medicine by thendigenous people for treating a variety of ailments includingrthritis, colds, fever and diabetes. Limited available informa-ion in the literature showed that methanolic extracts of Devil’slub displayed anti-fungal and anti-viral activity. In addition,

he extracted polyynes from OH demonstrated anti-candida,ntibacterial, anti-mycobacterial activity (McCutcheon et al.,993, 1994, 1995; Kobaisy et al., 1997). A constituent in OH,erolidol, was shown to have inhibited azoxymethane-inducedeoplasia of the large bowel in male F344 rats (Wattenberg,

991). Despite its use as a folk medicine for cancer treatmentLantz et al., 2004), there are at present no reported studies ineer-reviewed literature of the in vitro effects of OH on cancerells. The purpose of this study was to investigate the in vitro

harm

aal

2

2

oHwrpc7t25fc

2

cAadtmaar((

2p

ptatfiw

euu(wCcpT

mepp25c

2c

b(Fdppetitom1eataiei

2

4RfHgMwcmltMilu

2

J. Tai et al. / Journal of Ethnop

ctivities of OH including the anti-proliferation, anti-oxidantnd differentiation inducing properties on several cancer cellines.

. Materials and methods

.1. Plant material and preparation of the extract

The dry herb powder was prepared from the dry root barkf Oplopanax horridus (OH) supplied by Mary Joe Burgener ofaida Farm, Montana Creek, Alaska, USA. The sample receivedas a representative from a larger quantity of the herbal mate-

ial determined to be of medicinal quality by MJB who is aractitioner and supplier of this herb for many years. For cellulture studies, 0.5 g of the powder was extracted with 5 ml of0% ethanol for 2 h at 55 ◦C. This method has been in use inhe authors’ laboratory for extraction of herbs (Cheung and Tai,005; Blumenthal, 1998). The suspension was centrifuged at000 × g for 10 min and the supernatant was used as stock withurther dilution to 1/500, 1/1000, 1/2000, 1/4000 and 1/8000 forell culture studies.

.2. Chemicals and reagents

Unless specified, all chemicals were of reagent grade pur-hased from Sigma–Aldrich of Canada (Oakville, Ontario).nti-neoplastic agents camptothecin (CAM, Cat. #C9911)

nd paclitaxel (PTX, Cat. #T7402) were dissolved inimethyl sulfoxide (DMSO) as 1 mM stock and fur-her diluted to the desired concentrations with cultureedium. 6-Hydoxy-2,5,7,8-tetramethyl-chroman-2-carboxilic

cid (Trolox, Cat. #39,192-1), 2,2′-azinobis[3-ethylbenzothi-zoline-6-sulfonic acid] (ABTS, Cat. #A1888), lipopolysaccha-ide (LPS Cat. #L2630), 4-beta-phorbo-12-myristate-13-acetatePMA), dimethyl formamide (DMF), nitro-blue tetrazoliumNBT, Cat. #N6876).

.3. High performance liquid chromatography (HPLC)rofiling

The extraction of Devil’s Club dried root bark was accom-lished by adding 2.5 ml of a 70% solution of ethanol:watero 254 mg of ground tissue. This solution was then placed in

constant temperature bath, at 55 ◦C, for 2 h. Following cen-rifugation at 8000 rpm for 3 min the resulting supernatant wasltered through a 0.2 �m filter. An injection of 5 �l of this extractas used for chromatographic analysis.HPLC analysis was carried out using an Agilent 1100 HPLC

quipped with a binary pump, autosampler, thermostatted col-mn compartment, and a diode array detector. The mobile phasesed was: solvent A (H2O; 0.1% formic acid) and solvent BAcetonitrile; 0.1% formic acid). The flow rate was 0.9 ml/minith a column temperature of 40 ◦C. Using a Zorbax SB-

18 × 150 mm with 3.5 um particle size (pn 863953-902) thehromatographic gradient profile used was; 10% for 2 min, thenrogrammed to 24% in 18 min, and finally to 100% in 32 min.he diode array was set to collect all spectra as well as chro-

ic

acology 108 (2006) 228–235 229

atograms at wavelengths of 210, 270, and 330 nm. Additionalxperiments were carried out to determine optimum extractionrotocols for the root bark. Both ethanol and methanol wererepared in ranges of 25%, 50%, 75% and 100% in water and.5 ml was added to 250 mg of root bark. Both the normal 2 h5 ◦C extraction procedure and ultrasonic bath approaches werearried out.

.4. Antioxidant activity – Trolox equivalent antioxidantapacity (TEAC)

The test is based on the reduction of ABTS radical cationy antioxidants (Re et al., 1999; Schlesier et al., 2002). Trolox25 mM) was prepared in ethanol for use as a stock standard.resh working standards were prepared daily on dilution withistilled water. Stock solution of ABTS radical cation was pre-ared by mixing ABTS (7 mM) with 2.45 mM of potassiumersulfate in water. The mixture was kept for 12–24 h at ambi-nt temperature in the dark until the reaction was complete andhe absorbance became stable. To prepare an ABTS•+ work-ng solution, the ABTS•+ stock solution was diluted with watero an absorbance of 0.700 ± 0.02 at 734 nm. A stock solutionf Trolox (1 mM) was prepared with water. For the photo-etric assay, 1 ml of ABTS•+ working solution and 10 �l ofmM Trolox stock or 10 �l of a range of dilutions of OHthanolic extract were mixed for 45 s and measured immedi-tely after 1 min at 734 nm. The antioxidant activity of theest substances was calculated by determining the decrease inbsorbance using the following equation: %antioxidant activ-ty = {(E[ABTS•+] − E[standard])/E[ABTS•+]}× 100. Troloxquivalents were estimated by linear extrapolation of the antiox-dant activity from the Trolox standards.

.5. Cell lines and culture conditions

Human mammary adenocarcinoma MCF7 and MDA-MB-68, human promyelomonocytic HL60 cells, and murineAW264.7 macrophage/monocyte cell line were purchased

rom the American Type Culture Collection (Rockville, MD).uman leukemia cell line K562 was a gift from Dr. A.J. Tin-le, BCs research Institute for Children’s and Women’s Health.DA-MB-468 cells were cultured in L15 medium, HL60 cellsere cultured in Iscove’s medium, MCF7, K562 and RAW264.7

ells were cultured in Dulbecco’s minimal essential (DMEM)edium supplemented with 10% fetal bovine serum, 2 mM

-glutamine, and 50 �g/ml gentamycin. The cultures were main-ained in a humidified 5% CO2 incubator at 37 ◦C except for

DA-MB-468, which was maintained in an incubator with airntake only. Cells were sub-cultured every 3–4 days to maintainogarithmic growth and were allowed to grow for 24 h beforese.

.6. Cell proliferation assay

For testing, tumor cells were cultured in 96-well plates. Start-ng cell numbers were 2.5 × 104 cells per well for K562, 104

ells for MCF7 and MDA-MB-468. For HL60 cells, the start-

2 harm

iocsdecdbtcmcaftrt

w(ioa

C

w(tb5pdmaahuoaacv

2

2r

fpNFpid

sauiwe

2b

octmp

2p

cmi14tcbiw1

3

3

Fwbtaacme

3

fcs

30 J. Tai et al. / Journal of Ethnop

ng cell number was 105 cells per well in 24-well plates inrder to coordinate with the cell differentiation assay. After theells have stabilized overnight, triplicate (duplicate for HL60)amples of cells were treated with culture medium containingifferent concentrations of OH or in combination with differ-nt concentrations of anti-neoplastic agent CAM or PTX. Cellounting was set at 2 days for K562, 3 days for HL60 and 4ays for MCF7 and MDA-MB-468. Cell proliferation definedy increase in cell numbers was determined by hemocytome-er counting of viable cells by Trypan blue dye exclusion. Cellounts in samples treated with the test compounds were nor-alized to percent of control and the means and SEMs were

alculated from at least three independent experiments (Tai etl., 2004). IC50 (inhibitor concentration yielding 50% inhibition)or OH extract, CAM, PTX alone and OH extract in combina-ion with CAM or PTX on the cell lines was estimated by linearegression from a plot of percent inhibition against the concen-rations of OH, CAM and PTX.

The interaction between OH and the anti-neoplastic agentsas expressed as combination index (CI) based on median effect

Chou and Talalay, 1981; Menendez et al., 2005). This methodnvolved plotting dose–effect curves for each agent (OH, PTXr CAM) and for multiply diluted fixed ratio combination ofgents. A combination index is determined with the equation:

I = (D)a/(Dx)a + (D)b/(Dx)b + α[(D)a(D)b/(Dx)a(Dx)b]

here (Dx)a is the dose of agent a required to produce 50% effectIC50) alone, and (D)a is the dose of agent a required to producehe same 50% effect in combination with a fixed dose of agent. Similarly (Dx)b is the dose of agent b required to produce the0% effect when added alone and (D)b is the dose of agent b toroduce the same 50% effect when in combination with a fixedose of agent a. If the agents are mutually exclusive (e.g. similarode of action), then α is 0 (i.e. CI is sum of two terms); if the

gents are mutually non-exclusive (e.g. independent mode ofction), α is 1 (i.e. CI is the sum of three terms). For simplicity,owever, the third term of the Chou and Talalay equation issually omitted, and, thus, the mutually exclusive assumptionr classic isobologram is indicated. The combination index isdditive if it is equal or close to 1, synergistic if it is <1 andntagonistic if it is >1. For this study here, a CI of 1 ± 0.1 isonsidered additive, a CI value below 0.9 is synergy and a CIalue above 1.1 is considered sub-additive or ‘antagonistic’.

.7. Cell differentiation assay

.7.1. Differentiation into granulocyte lineage by NBTeduction assay

Duplicate 50 �l aliquots of HL60 cells treated with the dif-erent dilutions OH were incubated with an equal volume ofhenol red free RPMI containing 200 ng/ml of PMA and 0.2%BT at 37 ◦C for 45 min in a flat bottom 96-well microtiter plate.

or each preparation, 200 cells were examined to determine theroportion of cells containing blue–black formazan granules,ndicative of the ability of HL60 to generate superoxide anionuring an induced respiratory burst. In addition, cytocentrifuge

m1aw

acology 108 (2006) 228–235

mears of control and treated cells were stained with Giemsand analyzed by microscopy to allow the observation of gran-locytic features, i.e. multilobular nucleus, prominent cellularndentation. Positive control of differentiation to granulocytesas achieved by the addition of 100 mM DMF. The results were

xpressed as a percentage of positive cells (Suh et al., 1995).

.7.2. Differentiation into macrophage/monocyte lineagey non-specific esterase (NSE) activity assay

Standard NSE staining (Collins et al., 1980) was performedn duplicate microscope slides from samples prepared by aytocentrifuge. Two hundred cells were examined to determinehe proportion of positive cells (brownish red cytoplasm) under

icroscope. Cells treated with 50 ng/ml of PMA were used asositive control.

.8. Inhibition of LPS-activated nitric oxide (NO)roduction by RAW264.7 cells

RAW 264.7 cells were cultured in 24 well plates (1 × 106

ells/ml/well, NUNC culture plates) with phenol red free RPMIedium supplemented with 5% FBS. On the following day vary-

ng dilutions of OH were added alone, or in combination with�g/ml LPS, in fresh medium to the RAW264.7 cells. After8 h of incubation at 37 ◦C, the amount of NO released intohe culture supernatant was determined with Greiss reagent andompared to a sodium nitrite standard of known concentrationy a spectrophotometer at 550 nm. Cell proliferation and viabil-ty at the end of experiment was determined by staining the cellsith neutral red dye at 50 �g/ml for 1 h (Wadsworth and Koop,999; Miyamoto et al., 2002).

. Results

.1. HPLC profile

The resulting chromatogram at 210 nm is presented inig. 1A. It is clear there are approximately 15 large peaks, asell as many small minor peaks, with the peak at 6.192 mineing the major peak. As an example, the UV–vis spectra ofhe peak at 6.192 min is given in Fig. 1B. Both the ultrasonicnd heating approach resulted in almost identical profiles. Inddition, the use of methanol versus ethanol did not alter thehromatographic profiles. It was also determined that approxi-ately 75% ethanol or methanol led to optimum recoveries of

xtracted compounds.

.2. Anti-proliferative activity

Fig. 2 and Table 1 show that IC50 of OH extract has dif-erential anti-proliferative activity on the breast and leukemiaell lines tested. Among these cell lines, K562 is the mostusceptible to OH while MCF7 the least. IC50 is esti-

ated at 1/(2700 ± 200), 1/(1700 ± 300), 1/(500 ± 80), and

/(2500 ± 280) dilution, respectively, for K562, HL60, MCF7nd MDA-MB-468 cells. IC50 of CAM on K562 and HL60as 0.29 ± 0.02 and 0.25 ± 0.04 �M, respectively, and that

J. Tai et al. / Journal of Ethnopharmacology 108 (2006) 228–235 231

10 nm

o5

eaM

dawwd

aHw0l1a0

Fig. 1. (A) HPLC profile of Oplopanax horridus extract at 2

f PTX on MCF7 and MDA-MB-468 was 5.29 ± 0.39 and.34 ± 0.25 nM, respectively.

Combination index (CI) analysis showed that OH had syn-rgistic anti-proliferative effect with CAM on HL60, and haddditive anti-proliferative effects on K562, MCF7 and MDA-B-468 cells at some dilutions.Table 1 shows that for K562 cell, OH at 1/16,000 and 1/4000

ilution combined with 0.1 �M of CAM was additive in their

nti-proliferative effect with CI at 1.05 and 1.05, respectively,hereas 1/8000 dilution of OH combined with 0.1 �M of CAMas antagonistic with CI = 1.13. A 1/16,000, 1/8000 and 1/4000ilution of OH combined with 0.05 �M of CAM were also

aadp

. (B) UV–vis spectra of peak at 6.192 min as shown in (A).

ntagonistic with CI at 1.43, 1.51 and 1.43, respectively. For theL60 cells, combination of OH at 1/8000 and 1/4000 dilutionsith both 0.05 and 0.1 �M of CAM had a CI of 0.64 and 0.57,.35 and 0.29, respectively, showing strong synergistic antipro-iferative activity on HL60 cells. For MCF7 cells, 1/2000 and/1000 dilution of OH combined with 2.5 nM of PTX produceddditive anti-proliferative effect with CI estimated at 0.98 and.92, respectively. Combination of lower concentration of OH

t 1/4000 dilution with 2.5 nM of PTX was antagonistic withCI of 1.32. For MDA-MB-468 cells, combination of 1/4000

ilution of OH and 1 nM PTX was the only dosage combinationroduced an additive anti-proliferative effect with CI of 1.06.

232 J. Tai et al. / Journal of Ethnopharmacology 108 (2006) 228–235

Fig. 2. Effect of Oplopanax horridus (OH) extract on the proliferation of K562, HL60, MCF7 and MDA-MB-468 cells. Cell culture condition and method ofassessment are as described in Section 2. Results are normalized and presented as percent of control (mean ± S.E.M.) of at least three independent experiments. (�)Represents a dose–response curve of OH extract, diluted as indicated in (A–D). The other curves in each figure represent the same concentrations of OH extract plusa the figd 1 �Ma

AoP1

TIl

fixed concentration of a chemotherapeutic drug CAM or PTX as indicated inilutions, e.g. in panel A, K562 cells: *below data point OH1/8000 and CAM 0.t 0.1 �M.

ll other dosage combination of 1/16,000 and 1/8000 dilutionf OH with 1 nM and 1/16000, 1/8000 and 1/4000 with 2.5 nMTX had antagonistic anti-proliferative effect with CI of 1.46,.5, 1.91, 1.96 and 1.51, respectively.

3

t

able 1C50 and combination index (C.I.) of combined anti-proliferative effect of OH extracines

Cell line IC50

OH dilutionb CAM (�M), mean ± S.E.M. PTX (nM), me

K5621/(2700 ± 200) 0.29 ± 0.02 –– – –– – –

HL601/(1700 ± 300) 0.25 ± 0.04 –– – –

MCF71/(500 ± 80) – 5.29 ± 0.39– – –– – –

MDAMB4681/(2500 ± 280) – 5.34 ± 0.25

a C.I. calculated according to Chou and Talalay.b IC50 in dilutions: 1/(mean ± S.E.M.); n ≥ 9.

ure. Plot area: *p < 0.05 compared to corresponding therapeutic agent and OHrepresents p < 0.05 when data were compared to OH 1/8000 dilution and CAM

.3. Effects on differentiation of HL60 cells

HL60 cells were treated with non-cytotoxic dilutions of OHo assess its ability to induce cell differentiation. Reduction of

t and camptothecin (CAM) or paclitaxel (PTX) on different human tumor cell

OH dilution Combination index (C.I.)a

an ± S.E.M. CAM PTX

0.05 �M 0.1 �M 1 nM 2.5 nM

1/16000 1.43 1.05 – –1/8000 1.51 1.13 – –1/4000 1.43 1.05 – –

1/8000 0.64 0.35 – –1/4000 0.57 0.29 – –

1/4000 – – – 1.321/2000 – – – 0.981/1000 – – – 0.92

1/16000 – – 1.46 1.911/8000 – – 1.50 1.961/4000 – – 1.06 1.51

J. Tai et al. / Journal of Ethnopharmacology 108 (2006) 228–235 233

Fig. 3. Effects of Devil’s Club extract (OH, Oplopanax horridus) on the differ-entiation of HL60 leukemia cells. A 1 ml of HL60 (1 × 105 ml−1) was culturein each well of 24-well culture plate. After overnight culture, fresh media con-taining 1/8000, 1/4000, 1/2000 and 1/1000 dilutions of OH were added thecells were cultured for additional 3 days for nitroblue tetrazolium (NBT) andnon-specific esterase (NSE) assay. Panel A shows their effect on HL60 cell dif-ferentiation along the granulocyte lineage by NBT. Panel B shows their effectoDc

NNailCcC1cwcc

3

Aae

Fig. 4. Antioxidant properties of Oplopanax horridus extract (OH). Panel Ashows the ABTS hydroxyl radical scavenging activity of various dilutions ofOH extract in the Trolox equivalent antioxidant capacity assay. Panel B showsthe effect of OH on nitric oxide (NO) production in mouse macrophage RAW264.7 cell line. RAW 264.7 cells were treated with different dilutions of OH aloneor in combination with LPS (1 �g/ml) for 48 h. The culture supernatants werecollected and analyzed for NO content represented as nitrite levels with GriessrOb

tcT

(bi1dradtb

4

n HL60 cell differentiation along the monocyte/macrophage lineage by NSE.MF (100 mM) and PMA (50 ng/ml) are inducers of granulocyte and mono-

yte/macrophage differentiation, respectively. *p < 0.05 compared to control.

BT was used as a marker of granulocyte differentiation, andSE staining as a marker of monocyte/macrophage differenti-

tion. Fig. 3A shows that OH addition alone at 1/4000 dilutionnduced 13.1% of HL60 cells to differentiate along the granu-ocyte lineage compared to 6.1% in control samples (p = 0.04).AM added alone at 0.05, 0.1 and 0.25 �M slight, but statisti-ally insignificant granulocyte-inducing capacity on HL60 cells.ombination of 0.05 or 0.1 �M of CAM with 1/8000, 1/4000 and/2000 dilutions of OH did not enhance granulocyte-inducingapacities. Fig. 3B shows that NSE staining of HL60 cells treatedith 1/8000, 1/4000 and 1/2000 dilutions of OH alone or in

ombination with 0.05 or 0.1 �M of CAM did not have mono-yte/macrophage inducing capacity over the untreated controls.

.4. Antioxidant activities

OH was tested for its antioxidant activities with cell freeBTS OH radical scavenging assay, and inhibition of LPS

ctivated NO production by RAW 264.7 cells. In the Troloxquivalent antioxidant capacity (TEAC) assay, it was estimated

snD

eagent. Data represent mean ± S.E.M. of at least three independent experiments.H at 1/4000, 1/2000 and 1/1000 dilutions significantly reduced NO productiony LPS-activated RAW 264.7 cells (#p < 0.001).

hat a 1/100, 1/10 and 1/5 dilution of OH has hydroxyl radi-al scavenging activity equivalent to 2.7, 15.7 and 25.6 �M ofrolox, respectively (Fig. 4A).

Fig. 4B showed that OH effectively inhibited nitric oxideNO) production by LPS-activated RAW 264.7 cells. While theasal level of NO represented by nitrite in the culture mediums less than 2 �M in the non-stimulated RAW 264.7 cells. OH at/4000, 1/2000 and 1/1000 dilutions did not stimulate NO pro-uction in RAW 264.7 cells. At these dilutions, OH significantlyeduced NO production by LPS activated RAW 264.7 cells indose dependent manner (p < 0.001). In addition, OH at theseilutions were neither cytotoxic to the RAW 264.7 cells, nor didhey affect cell viability in the LPS activated RAW 264.7 cellsy neutral red viability staining.

. Discussion

Devil’s Club is a deciduous shrub often described as the mostignificant plant, both medically and spiritually, to the indige-ous people in British Columbia and Alaska. Medicinal use ofevil’s Club was cited as early as 1842 by Eduardo Blaschke,

2 harm

traatrsnSAoCc2

ptcaictbaccDt2gteaeiacpG

bn1taawcpnovpiMtn

ac

ca(rtmoii

(aopaShNgtcco

r(drd2asa

aEpsHtitw

A

rp

34 J. Tai et al. / Journal of Ethnop

he chief physician for the Russian American Company, whoeported the use of Devil’s Club ash as a treatment for soresmong the Tlingit people. Recent studies on the phytochemicalnd biological properties of Devil’s Club confirm its widespreadraditional use for a variety of ailments. Due to its botanicalelationship to the well-known medicinal ginsengs (Panax gin-eng, Panax quinquefolius), Devil’s Club has been marketed asutraceutical under the misleading, and now illegal in the Unitedtates, common names of “Alaskan ginseng”, “wild armoredlaskan ginseng” and “Pacific ginseng”. Such marketing reliesn the purported phytochemical similarities between Devil’slub and Panax spp., a presumption that is not supported byurrent phytochemical research (Wills and Lipsey, 1999; Awang,003; Lantz et al., 2004).

Normally comparison of chromatoghraphic profiles wouldrovide some clues as to the variety of compounds present inhe extract. Unfortunately there are relatively few articles wherehromatographic profiles are presented. In an article by Gruber etl. (2004) the identification of stigmasterol and trans-nerolidoln a similar extract was carried out. But comparison with thehromatographic profile presented by Gruber is difficult due tohe differences in gradient and mobile phase. An investigationy Kobaisy et al. (1997) identified five polyynes associated withnti-mycobacterial activity as well as falcarindiol. However, nohromatographic profiles were presented. Apart from these arti-les, it appears, no other compounds have been identified inevil’s Club. In the profile presented in Fig. 1A, UV–vis spec-

ra indicate the presence of phenolic acids eluting from 2 to0 min. As an example the spectra of the peak at 6.192 min isive in Fig. 1B. Furthermore, by carrying out pH variations ofhe mobile phase additional evidence of this is provided. Peaksluting from 20 min on are in general neutral compounds suchs the ones observed in Gruber’s and Kobaisy’s articles (Grubert al., 2004; Kobaisy et al., 1997). For chromatographic compar-son there are numerous articles investigating phenolics (Lee etl., 2004; Maatta-Riihinen et al., 2004). These articles allowedhromatographic comparisons, and added to the evidence of theresence of these compounds. Currently there are LC/MS andC/MS investigations on going to confirm these conclusions.Teas made from the root bark of Devil’s Club had been used

y the Alutiig, Gitxsan, Haida, Tlingit and Tsimshian indige-ous people for treating cancer (Lantz et al., 2004). As early as944, Stuhr and Henry performed a preliminary chemical inves-igation into the constituents of various extracts of Devil’s Clubnd found that alkaloids and gallic acid were absent and oleicnd unsaturated fatty acids, saponins, glycerides and tanninsere present in the extracts (Smith, 1983). These phytochemi-

als are known to have anti-proliferative and chemopreventiveroperties. In an in vivo study, a constituent of Devil’s Club,erolidol, is shown to inhibit azoxymethane-induced neoplasiaf the large bowel in male F344 rats (Wattenberg, 1991). Our initro study here shows that Devil’s Club extract has strong anti-roliferative effect against several human tumor cell lines. IC50

s <1/1500 dilution on K562 and HL60 leukemia and MDA-

B-468 breast tumor cell lines, and at 1/500 on MCF7 breastumor cell line. Although the effective anti-proliferative compo-ent(s) in the Devil’s Club extract is yet to be identified, its potent

R

A

acology 108 (2006) 228–235

nti-proliferative effects seen here would support the traditionallaims of Devil’s Club usefulness for treating cancers.

Since our data here show that extracts of Devil’s Club inombination with non-toxic levels of conventional cancer ther-peutic drugs have some additive proliferative effect on breastMCF7 and MDA-MB-468) and K562 leukemia cell lines andemarkable synergistic effects on HL60 cells, it might be a poten-ial candidate herb for use in adjunct cancer drug therapy. The

echanism by which OH extract enhanced the therapeutic effectf camptothecin and paclitaxel is still unknown. Further researchs needed to delineate the herb’s effective components and theirn vitro and in vivo pharmacological efficacy.

Terminal differentiation of human promyelocytic leukemiaHL60) cells can be induced by a variety of chemical agentsnd this process can be monitored readily by the generationf mature monocyte/macrophages and/or granulocytes. Com-ounds known to be efficacious cancer chemopreventive agentsre potent inducers of HL60 differentiation (Kawaii et al., 1999).ome agents that showed toxicity at high concentrations canave differentiation inducing activities at lower concentrations.on-cytotoxic concentrations of Devil’s Club show only mildranulocyte inducing capacity on HL60 cells, and it is not effec-ive in inducing monocyte/macrophage differentiation on theseells. In addition, combining non-toxic doses of OH extract withamptothecin did not add to the granulocyte inducing capacityf camptothecin.

Nitric oxide is a major effector molecule of host immuneesponse, effective against viral infection including retrovirusesTorre et al., 2002). Our study shows that Devil’s Club extractemonstrates significant antioxidant activities in hydroxyl freeadical scavenging activity by TEAC assay as well as in dose-ependent inhibition of NO production by LPS activated RAW64.7 cells. Although the in vivo and in vitro mechanism ofction by Devil’s Club are still lacking, these observations wouldupport the traditional use of Devil’s Club as a tonic and forrthritis and rheumatism treatment.

In conclusion, this study surveyed the anti-proliferative andntioxidant activities of Devil’s Club (Oplopanax horridus).thanol extract of Devil’s Club shows significant in vitro anti-roliferative activities on several human tumor cell lines. Ithows a mild granulocyte differentiation-inducing property onL60 cells. It also demonstrates significant antioxidant activi-

ies in the cell free TEAC assay as well as in the in vitro NOnhibition assay in RAW 264.7 cells. Further research is requiredo elucidate this widely used medicinal herb of the Pacific North-est.

cknowledgements

This work was supported by the Lotte and John Hecht Memo-ial Foundation. We would like to thank Mary Joe Burgener forroviding us Devil’s Club herbal material.

eferences

wang, D.V.C., 2003. What in the name of Panax are those other ‘Ginsengs”?HerbalGram 57, 35–44.

harm

B

C

C

C

G

K

K

L

L

M

M

M

M

M

M

R

S

S

S

T

T

W

W

J. Tai et al. / Journal of Ethnop

lumenthal, M. (Ed.), 1998. The Complete German Commission E Mono-graphs: Therapeutic Guide to Herbal Medicines. American Botanical Coun-cil (Publ.), pp. 575–579.

heung, S., Tai, J., 2005. In vitro studies of the dry fruit of Chinese fan palmLivistona chinensis. Oncology Reports 14, 1331–1336.

hou, T.-C., Talalay, P., 1981. Generalized equations for the analysis of inhi-bitions of Machaelis–Menton and higher-order kinetic systems with two ormore mutually exclusive and nonexclusive inhibitors. European Journal ofBiochemistry 115, 207–216.

ollins, R.D., Cousar, J.B., Russell, W.G., Glick, A.D., 1980. Diagnosis of neo-plasm of the immune system. In: Rose, N.R., Friedman, H. (Eds.), Manualof Clinical Immunology, 2nd ed. American Society for Microbiology, Wash-ington, DC, pp. 98–99.

ruber, J.W., Kittipongpatana, N., Bloxton II, J.D., Der Marderosian, A., Schae-fer, F.T., Gibbs, R., 2004. High-performance liquid chromatography andthin-layer chromatography assays for Devil’s Club (Oplopanax horridus).Journal of Chromatographic Science 42, 196–199.

awaii, S., Tomono, Y., Katase, E., Ogawa, K., Yano, M., 1999. Effect ofcitrus flavonoids on HL60 cell differentiation. Anticancer Research 19,1261–1270.

obaisy, M., Abramowski, Z., Lerner, L., Saxena, G., Hancock, R.E.W., Tow-ers, G.H.N., Doxsee, D., Stokes, R.W., 1997. Antimycobacterial polyynesof Devil’s Club (Oplopanax horridus), a North American native medicinalplant. Journal of Natural Products 60, 1210–1213.

antz, T.C., Swerhun, K., Turner, N.J., 2004. Devil’s Club (Oplopanax hor-ridus): an ethnobotanical review. HerbalGram 62, 33–48.

ee, J., Finn, C.E., Wrolstad, R.E., 2004. Comparison of anthocyanin pigmentand other phenolic compounds of Vaccinium membranaceum and Vacciniumovatum native to the Pacific Northwest of North America. Journal of Agri-cultural and Food Chemistry 52, 7039–7044.

aatta-Riihinen, K.R., Kamal-Eldrin, A., Mattila, P.H., Gonzalez-Paramas,A.M., Torronen, A.R., 2004. Distribution and contents of phenolic com-pounds in eighteen Scandinavian berry species. Journal of Agricultural andFood Chemistry 52, 4477–4486.

cCutcheon, A.R., Ellis, E., Hancock, R., Towers, G., 1993. Antibiotic screen-

ing of medicinal plants of the British Columbian native people. Journal ofEthnopharmacology 37, 213–223.

cCutcheon, A.R., Ellis, E., Hancock, R., Towers, G., 1994. Antifungal screen-ing of medicinal plants of the British Columbian native people. Journal ofEthnopharmacology 44, 157–169.

W

acology 108 (2006) 228–235 235

cCutcheon, A.R., Roberts, T.E., Gibbons, E., Ellis, S.M., Babiuk, L.A., Han-cock, R., Towers, N., 1995. Antiviral screening of British Columbian medic-inal plants. Journal of Ethnopharmacology 49, 101–110.

enendez, J.A., Vellon, L., Colomer, R., Lupu, R., 2005. Pharmacologicaland small interference RNA-mediated inhibition of breast cancer-associatedfatty acid synthase (oncogenic antigen-519) synergistically enhances Taxol(Paclitaxel)-induced cytotoxicity. International Journal of Cancer 115,19–35.

iyamoto, M., Hashimoto, K., Minagawa, K., Satoh, K., Komatsu, N., Fujimaki,M., Nakashima, H., Yokote, Y., Akahani, K., Guputa, M., Sarma, D.N.K.,Sakagami, H., 2002. Effect of poly-herbal formula on NO productionby LPS-stimulated macrophage-like cells. Anticancer Research 22, 3293–3302.

e, R., Pellegrini, N., Proteggente, A., Pannela, A., Yang, M., Rice-Evans, C.,1999. Antioxidant activity applying an improved ABTS radical cation decol-orization assay. Free Radical Biology and Medicine 26, 1231–1237.

chlesier, K., Harwat, M., Bohm, V., Bitsch, R., 2002. Assessment of antioxi-dant activity by using different in vitro methods. Free Radical Research 36,177–187.

mith, G.W., 1983. Arctic pharmacognosia II. Devil’s Club, Oplopanax hor-ridus. Journal of Ethnopharmacology 7, 313–320.

uh, N., Luyengi, L., Fong, H.H.S., Kingshorn, A.D., Pezzuto, J.M.,1995. Discovery of natural product chemopreventative agents utilizingHL60 cell differentiation as a model. Anticancer Research 15, 233–240.

ai, J., Cheung, S., Chan, E., Hasman, D., 2004. In vitro culture studies of Suther-landia frutescens on human tumor cell lines. Journal of Ethnopharmacology93, 9–19.

orre, D., Puglisse, A., Speranza, F., 2002. Role of nitric oxide inHIV-1 infection: friend or foe? Lancet Infectious Disease 2, 273–280.

adsworth, T.L., Koop, D.R., 1999. Effects of wine polyphenolics quercetinand resveratrol on pro-inflammatory cytokine expression in RAW 264.7macrophages. Biochemical Pharmacology 57, 941–949.

attenberg, L.W., 1991. Inhibition of azoxymethane neoplasia of the bowel by 3-

hydroxy-3,7,11-trimethyl-1,6 10-dodecatriene (nerolidol). Carcinogenesis12, 151–152.

ills, R.M., Lipsey, R.G., 1999. An economic strategy to develop non-timberforest products and services in British Columbia. Final Report. Victoria,B.C.: Forest Renewal B.C.; Project No. PA97538-ORE.