Mh

Sa

b

c

a

ARRA

KAMSC

I

aelaa

fl(hsocit

nC

0h

Phytomedicine 21 (2013) 39– 46

Contents lists available at ScienceDirect

Phytomedicine

j ourna l h o mepage: www.elsev ier .de /phymed

echanism of arctigenin-mediated specific cytotoxicity againstuman lung adenocarcinoma cell lines

iti Susanti a,b,c, Hironori Iwasakib, Masashi Inafukub, Naoyuki Tairaa,b, Hirosuke Okub,∗

United Graduate School of Agricultural Sciences, Kagoshima University, 1-21-24, Korimoto, Kagoshima 890-0065, JapanCenter of Molecular Biosciences, Tropical Biosphere Research Center, University of the Ryukyus, Nishihara, Okinawa 903-0213, JapanDepartment of Animal and Agricultural Sciences, Diponegoro University, Central Java, Indonesia

r t i c l e i n f o

rticle history:eceived 24 April 2013eceived in revised form 5 July 2013ccepted 2 August 2013

eywords:rctigenin (ARG)echanism

a b s t r a c t

The lignan arctigenin (ARG) from the herb Arctium lappa L. possesses anti-cancer activity, however themechanism of action of ARG has been found to vary among tissues and types of cancer cells. The currentstudy aims to gain insight into the ARG mediated mechanism of action involved in inhibiting proliferationand inducing apoptosis in lung adenocarcinoma cells. This study also delineates the cancer cell speci-ficity of ARG by comparison with its effects on various normal cell lines. ARG selectively arrested theproliferation of cancer cells at the G0/G1 phase through the down-regulation of NPAT protein expression.This down-regulation occurred via the suppression of either cyclin E/CDK2 or cyclin H/CDK7, while apo-

pecificytotoxicity

ptosis was induced through the modulation of the Akt-1-related signaling pathway. Furthermore, a GSHsynthase inhibitor specifically enhanced the cytotoxicity of ARG against cancer cells, suggesting that theintracellular GSH content was another factor influencing the susceptibility of cancer cells to ARG. Thesefindings suggest that specific cytotoxicity of ARG against lung cancer cells was explained by its selectivemodulation of the expression of NPAT, which is involved in histone biosynthesis. The cytotoxicity of ARGappeared to be dependent on the intracellular GSH level.

ntroduction

Lung cancer is a major worldwide health problem and hasccounted for approximately 16% of global cancer deaths (Pisanit al., 1999). There are 2 main types of lung cancer: small-cellung cancer (SCLC) and non-small-cell lung cancer (NSCLC). Lungdenocarcinoma, which belongs to the NSCLC group, constitutespproximately 75–85% of all lung cancers (Greenlee et al., 2001).

Chemotherapy has been established as a standard treatmentor NSCLC; however, despite the success of chemotherapy, it has aimited ability to improve a patient’s symptoms and quality of lifeTen Bokkel Huinink et al., 1999). While some lung cancer therapiesave shown promising results for the control of disease progres-ion, the toxicity of this therapeutics often restricts the completionf the recommended dose. Therefore, the identification of an anti-

ancer agent with high selectivity against NSCLC merits furthernvestigation and is preferable than searching for agents with highoxicity.

Abbreviations: ARG, arctigenin; GSH, glutathione; G0/G1, Gap 0/Gap 1; NPAT,uclear protein of the ataxia telangiectasia locus; CDK2, cyclin-dependent kinase 2;DK7, cyclin-dependent kinase 7; akt-1, alpha serine/threonine-protein kinase.∗ Corresponding author. Tel.: +81 98 895 8972; fax: +81 98 895 8944.

E-mail address: okuhiros@comb.u-ryukyu.ac.jp (H. Oku).

944-7113/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.phymed.2013.08.003

© 2013 Elsevier GmbH. All rights reserved.

Increasing attention has been paid to the pharmaceutical valueand the biological activity of herbal plant medicines over thelast decade. Numerous bioactive compounds isolated from herbalplants exhibit selective toxicity toward tumorigenic tissues, andthus display cancer-targeting properties (Quasney et al., 2001). Thelignans compose one group of chemical compounds found in plants,and include arctigenin (ARG), a dibenzylbutyrolactone lignan iso-lated from Arctium lappa L. ARG has been reported to possess manyimportant pharmacological or biological properties, includinganti-oxidative, anti-tumorigenic, and anti-inflammatory activities(Awale et al., 2006; Matsumoto et al., 2006). Our previous studyshowed that ARG specifically inhibited the proliferation of lungadenocarcinoma (A549) cells, which might consequently lead tothe induction of apoptosis (Susanti et al., 2012). Although the anti-cancer activity of ARG appears to be promising, the mechanism ofits action is not yet fully understood. Several studies have reportedthat ARG blocks the unfolded protein response and activates Akt inglucose-deprived solid tumors (Kim et al., 2010; Awale et al., 2006).However, the mechanism proposed for the glucose-deprived can-cer cells may not necessarily be applicable to our current study. Itis known that the inhibition of cell proliferation is involved in the

anti-cancer mechanism of ARG; however, the signal transductionpathways leading to cell cycle arrest are entirely unknown. ARG hasbeen shown to induce cell cycle arrest at the G0/G1 phase in gastriccancer cells by modulation of regulatory proteins such as Rb and

4 medic

ctptGbo

iotovc

M

I

w

C

3lBma

C

piwu(Pto

C

itsbswowcRApif

W

Sc

0 S. Susanti et al. / Phyto

yclin D1 (Jeong et al., 2011), and at the G2/M phase in colorec-al cancer cells via the regulation of the Wnt/�-catenin signalingathway (Yoo et al., 2010). Furthermore, ARG was detoxified by glu-athione (GSH) in liver cells, indicating that the expression level ofSH synthase may be an additional factor controlling cell suscepti-ility to this agent (Moritani et al., 1996). Therefore, the mechanismf action of ARG is likely to vary among tissues and cancer cell types.

The current study aims to gain insight into the mechanismsnvolved in ARG-mediated inhibition of proliferation and inductionf apoptosis in lung adenocarcinoma. Furthermore, this study aimso delineate the specificity of ARG via comparative studies on vari-us normal cell lines. The results of this investigation will providealuable new information on the usefulness of ARG as a promisinghemotherapeutic agent with few adverse effects.

aterials and methods

solation and purification of ARG

The isolation and purification of ARG from A. lappa L. extractere performed as described previously in Susanti et al. (2012).

ell cultures

Various human normal embryo fibroblast (OUMS-36, OUMS-6T-2F, and OUMS-36T-5F) and lung adenocarcinoma (A549) cell

ines were purchased from the Japanese Cancer Research Resourcesank (JCRB, Ibaraki, Japan). Cells were cultured in DMEM supple-ented with 10% FBS (Fetal Bovine Serum) in a humidified 5% CO2

tmosphere at 37 ◦C.

ell viability assay

Cells suspended in DMEM were seeded at 1 × 104 cells (100 �l)er well in 96-well plates and pre-incubated overnight in a humid-

fied atmosphere of 5% CO2 at 37 ◦C. After a 24-h incubationith varying concentrations of ARG, cell viability was determinedsing MTS assay kits according to the manufacturer’s instructionsCell Titer 96® AQueous non-radioactive cell proliferation assay,romega Co., Madison, USA). All experiments were performed inriplicate, and cell viability was expressed as the relative viabilityf the treated cells compared to untreated cells (control).

ell cycle analysis

Cells (3 × 105) were seeded in 60-mm culture dishes and pre-ncubated for 24 h at 37 ◦C. Cells were washed with PBS beforehe replacement of the medium, and then cultured in DMEMupplemented with or without ARG (6 �g/ml). After a 24 h incu-ation, cells were harvested using ICT AccutaseTM cell detachmentolution (Innovative Cell Technologies, Inc, San Diego, CA, USA),ashed with PBS, fixed with 70% cold ethanol, and incubated

vernight at 4 ◦C. Ethanol was removed by decantation, and cellsere re-suspended in PBS on ice for 10 min. Cell suspensions were

entrifuged at 1000 rpm for 10 min, re-suspended in a 250 U/mlNase solution, and incubated for 20 min at room temperature.fter 200 �g/ml propidium iodide solutions was added, the cell sus-ensions were transferred and filtered through a 35 mM nylon filter

nto flow cytometer tubes. Data acquisition and analysis were per-ormed by a FACS Calibur flow cytometer system (BD Biosciences).

estern blot analysis

Proteins were isolated using PRO-PREPTM Protein Extractionolution (iNtRON Biotechnology, Korea) from the lysate of 2.5 × 106

ells that had been treated with 6 �g/ml ARG for 24 h. The protein

ine 21 (2013) 39– 46

concentration was measured using Quant-iTTM Protein Assay Kits(Invitrogen, USA) according to the manufacturer’s instructions. Pro-tein samples were dissolved in an equal amount of sample buffersolution (EzApply, Atta, Osaka, Japan). Proteins (10 �g) were sepa-rated on 12.5% SDS–PAGE gels (e-PAGEL, E-R12.5L, ATTO, Tokyo,Japan) run at 15–20 mA in running buffer (EzRun, ATTO, Tokyo,Japan). The proteins were transferred to PVDF membranes by usingthe iBlotTM Dry Blotting System (Invitrogen, Tokyo, Japan). ThePVDF membranes were treated with blocking buffer (Blocking One,Nacalai Tesque, Kyoto, Japan) for 1 h and washed with TBS (1×PBS in 0.1% Tween-20). Subsequently, the membranes were incu-bated with primary antibody (�-actin, CDK2, p-CDK, CDK7, CyclinE, Cyclin H, Rb, P-Rb, and NPAT) with gentle shaking overnight.After washing, the membranes were incubated with an HRP-linkedanti-rabbit IgG secondary antibody for 1 h and then washed again.The detection of target proteins was performed using a lumines-cent image analyzer (ImageQuant LAS 4000 mini, GE Healthcare,Uppsala, Sweden) and Amersham ECL Advance Western BlottingDetection Kits (GE Healthcare, Buckinghamshire, UK). The relativeintensities (%) of protein bands were measured using Image J soft-ware version 1.47d.

Total RNA isolation and quantitative real-time PCR analysis

Total RNA was extracted from either untreated or treatedcells (1 × 106) by using the AquaPure RNA Isolation Kit (Bio-RadLaboratories) according to the manufacturer’s instructions. Thequality of the isolated RNA was checked using MultiNA ReagentKits on a microchip electrophoresis system (MultiNA-Biotech,Shimadzu, Japan) according to the manufacturer’s instructions.Total RNA (2 �g) was reverse transcribed to produce cDNA byusing High Capacity RNA-to-cDNA Kits according to the manu-facturer’s instructions (Applied Biosystems, USA). cDNA samples(20 �l aliquots) were stored at −20 ◦C. For real-time PCR analy-sis, the cDNAs were amplified in a StepOnePlus Real-Time PCRsystem (Applied Biosystems, CA, USA) by using Fast SYBR GreenMaster Mix (Applied Biosystems, CA, USA), according to the man-ufacturer’s instructions. Table 1 lists the sequences of the primersthat were used for quantitative real-time RT-PCR analysis. All anal-yses were performed in triplicate, and the gene expression levelswere normalized to the housekeeping gene ACTB. Fold changes ingene expression were calculated on the basis of the standard curve,which was constructed using the calibration data produced by theStepOne software.

Effect of GSH depletion on ARG cytotoxicity

The effect of GSH depletion on the cytotoxicity of ARG wasstudied using the GSH synthase inhibitor l-buthionine-(S,R)-sulfoximine (BSO). Cells suspended in DMEM were seeded at1 × 104 cells (100 �l) per well in 96-well plates and pre-incubatedin a humidified 5% CO2 atmosphere at 37 ◦C for 10 h. The cells wereincubated with 50 �M of BSO for 14 h, and subsequently treatedwith varying concentrations of ARG for 24 h. Cell viability was thendetermined using MTS assay kits according to the manufacturer’sinstructions (Cell Titer 96® AQueous non-radioactive cell prolif-eration assay, Promega Co., Madison, USA). All experiments wereperformed in triplicate, and the cell cytotoxicity was expressed asthe relative viability of treated cells compared to that of untreatedcontrol cells. In the case of BSO pretreatment, control cells weretreated with BSO only. The ED50 of ARG with or without BSO pre-treatment was estimated by fitting the following formula to the

titration curve:

y = ˇ3 + ˇ4

{1 + exp(ˇ1 + ˇ2x)}

S. Susanti et al. / Phytomedicine 21 (2013) 39– 46 41

Table 1Primer sequences used for quantitative real time RT-PCR.

Gene Description Primer sequence Product Accession no.

Fas Fas (TNF receptor superfamily, member 6) 5′-AAGAATGGTGTCAATGAAGCCA-3′ (forward primer, F) 103 NM 152871.25′-GAAGTTGATGCCAATTACGAAGC-3′ (reverse primer, R)

Bcl-2 B-cell lymphoma 2 5′-TTCTACGACAGCAAATTGCCC-3′ (forward primer, F) 102 NM 004049.35′-TTGCCTTATCCATTCTCCTGTGT-3′ (reverse primer, R)

Bax BCL2-associated X protein 5′-TCTGACGGCAACTTCAACTGG-3′ (forward primer, F) 115 NM 138764.45′-AGCCCATGATGGTTCTGATCA-3′ (reverse primer, R)

Bad BCL2-associated agonist of cell death 5′-CAGTGACCTTCGCTCCACATC-3′ (forward primer, F) 123 NM 004322.35′-AAGGAGACAGCACGGATCCTC-3′ (reverse primer, R)

Akt1 v-akt murine thymoma viral oncogene homolog 1 5′-AGAGAAGCCACGCTGTCCTCT-3′ (forward primer, F) 102 NM 001014432.15′-CCGCAGGATAGTTTTCTTCCC-3′ (reverse primer, R)

AIF Mitochondrial apoptosis inducing factor 5′-AGTAGTTTGCCCACAGTTGGTGTT-3′ (forward primer, F) 101 NM AL049703.15′-TCACTCTCTGATCGGATACCAGTTC-3′ (reverse primer, R)

Casp 8 Caspase 8, apoptosis-related cysteine peptidase 5′-GATATATCCCGGATGAGGCTGAC-3′ (forward primer, F) 101 NM 001080124.15′-TGACTGGATGTACCAGGTTCCC-3′ (reverse primer, R)

Casp 3 Caspase 3, apoptosis-related cysteine peptidase 5′-GCCTGAGCAGAGACATGACTCA-3′ (forward primer, F) 106 NM 004346.35′-TCATCCACACATACCAGTGCG-3′ (reverse primer, R)

Casp 9 Caspase 9, apoptosis-related cysteine peptidase 5′-TGACTTGTGTCCCATGATCCC-3′ (forward primer, F) 108 NM 001229.35′-AATGTACAGGACAGCCTCACAGC-3′ (reverse primer, R)

AGCGCACGC

wsfu

S

tt

R

S

oct2

Fpv

ACTB Actin, beta 5′-TCACCG5′-TAATGT

here y is the cell viability; x is the concentration of the testubstance in the medium; and ˇ1−�4 is the constant. Fitting theormula after transformation to the linear model was performedsing a simple regression analysis program (R Version 2.11.1).

tatistical analysis

The data are expressed as means ± standard deviation (SD). Sta-istical significance between pairs of means was evaluated usinghe Student’s t test (p < 0.05 or p < 0.01).

esults

pecific cytotoxicity of ARG in various cell lines

Fig. 1 shows the effect of ARG concentration on the cell viability

f human lung cancer (A549) and human normal (OUMS-36T-2F)ells. OUMS-36 and OUMS-36T-5F human normal cells were foundo be resistant to ARG, whereas A549 cancer cells and OUMS-36T-F normal cells were susceptible. Among the cells studied, the A549

ig. 1. The effect of arctigenin on the viability of various human cell lines. Data areresented as means ± SD of experiments performed in triplicate. *p < 0.05; **p < 0.01ersus the untreated cell (control).

CGGCT-3 (forward primer, F) 60 NM 001101.3ACGATTTCCC-3′ (reverse primer, R)

cancer cells were found to be most susceptible to ARG treatment,reinforcing the concept of cancer-specific ARG cytotoxicity.

The effect of ARG on the cell cycle in various cell lines

The flow cytometric histogram in Fig. 2 illustrates that ARGtreatment increased the proportion of cells in the G0/G1 phase from55.8% (control) to 81.4% in A549 cells, and from 48.5% (control)to 74.8% in OUMS-36T-2F normal cells. This result indicates thatARG (6 �g/ml) arrested the cell cycle of both cell types at the G0/G1phase. In contrast, there were no significant changes in the propor-tion of cells in the G0/G1 phase for the 2 other normal cell lines(OUMS-36 and OUMS-36T-5F).

The effect of ARG on the expression of cell cycle regulatory proteins

ARG has been shown to inhibit the proliferation of A549 lungcancer cells without any adverse effect on normal cells (Susantiet al., 2012). In order to explore the molecular mechanism ofARG-mediated cell cycle arrest, the expression of several cell cycleregulatory proteins was analyzed by western blotting.

ARG (6 �g/ml) lowered the level of regulatory proteins involvedin the G1/S checkpoint signaling, including cyclin H, CDK7, cyclinE, CDK2, p-CDK, and NPAT (Fig. 3). The effects of ARG appeared todepend on the malignancy of the cells (Fig. 3). In A549 lung cancercells, ARG significantly decreased the concentration of the cyclinH, CDK7, cyclin E, CDK2, p-CDK, and NPAT proteins. Similarly, ARGtreatment lowered the concentration of these proteins in OUMS-36T-2F normal cells, except for cyclin H. In contrast, no effect wasobserved in the 2 other types of normal cells (OUMS-36 and OUMS-36T-5F). On the other hand, ARG increased the concentration of p-CDK in both OUMS-36 and OUMS-36T-5F normal cells, and cyclin Ein OUMS-36 cells only. These data are consistent with our previous

observation that ARG specifically arrested the cell cycle of cancercells at the G0/G1 phase. Furthermore, no significant difference wasobserved in the concentration of Rb and P-Rb proteins betweenARG-treated and untreated cells.

42 S. Susanti et al. / Phytomedicine 21 (2013) 39– 46

ous hu

T

acpeAg

rpABctcg

eca

T

lnracdBo

Fig. 2. The effect of arctigenin on the cell cycle of vari

he effect of ARG on the expression of apoptosis-related genes

ARG has been reported to induce apoptosis through caspase-3ctivation in A549 lung cancer cells (Susanti et al., 2012). In theurrent study, we aimed to gain more insight into the signalingathway of apoptosis in ARG-treated cells. Thus, we analyzed thexpression of caspase-3 and other pro-apoptosis (Fas, Bax, Bad,IF, Caspase-8, and Caspase-9) and anti-apoptosis (Bcl-2 and Akt-1)enes.

Fig. 4 shows the effect of ARG on the expression of apoptosis-elated genes. ARG treatment up-regulated the expression of thero-apoptosis genes in A549 lung cancer cells, except for Bad gene.RG also up-regulated the expression of the anti-apoptosis genecl-2 and down-regulated another anti-apoptosis gene Akt-1. Inontrast to these observations in A549 cells, ARG had no effect onhe anti-apoptotic genes in OUMS-36 and OUMS-36T-5F normalells and largely decreased the expression of some pro-apoptoticenes.

Furthermore, ARG treatment had no differential effect on thexpression of apoptosis-related genes in OUMS-36T-2F normalells, as ARG treatment significantly up-regulated both pro- andnti-apoptosis genes in this cell type.

he effect of GSH depletion on the cytotoxicity of ARG

Our previous study noted that the responses of normal cellines (OUMS-36, OUMS-36T-5F, and OUMS-36T-2F) to ARG wereot necessarily consistent. OUMS-36 and OUMS-36T-5F cells wereesistant to ARG, whereas OUMS-36T-2F cells were sensitive to ARGnd showed a response that was comparable to that of A549 lung

ancer cells. The clarification of the mechanism underlying theseifferent responses to ARG is highly important. Thus, we employedSO, a glutathione synthase inhibitor agent, to investigate the effectf glutathione depletion on the sensitivity of various cell lines to

man cell lines performed by flow cytometry analysis.

ARG treatment. The ED50 of ARG was significantly reduced by BSOpre-treatment in A549 lung cancer cells and OUMS-36T-2F normalcells, whereas no changes in ED50 were observed in OUMS-36 andOUMS-36T-5F normal cells (Fig. 5).

Discussion

We screened 365 Kampo Medicine and found the occurrence oflung cancer-specific cytotoxicity in 9 plant species. Among them, A.lappa L. was found to contain the cancer-specific agent ARG (Susantiet al., 2012), the selectivity of which was reinforced in this study(Fig. 1).

Our current study furthermore showed that ARG selectivelyinduced apoptosis via the up-regulation of pro-apoptosis genes(Bax, Fas, Caspase-3, 8, and 9) and the down-regulation of an anti-apoptosis gene (Akt-1) (Fig. 4). However, the pro-apoptosis geneBad and anti-apoptosis gene Bcl-2 were down- and up-regulated,respectively, by ARG in A549 cancer cells. Bad activity has beenshown to be regulated by phosphorylation via the action of Akt-1. Bad binds with Bcl-2/Bcl-XL in the cytosol and represses theiranti-apoptotic activity. Phosphorylation of Bad by Akt-1 promotesits dissociation from the complex and relocates Bcl-2/Bcl-XL to themitochondria, thereby inactivating the apoptotic process (Albertset al., 2002). Thus, the induction of apoptosis by ARG may bemediated via the suppression of Akt-1. However, some of thegene expression profiles in A549 cancer cells appeared to argueagainst this explanation; ARG increased the expression of the anti-apoptotic gene Bcl-2 and decreased the expression of the apoptoticgene Bad. These data indicate that the role of Akt-1 signaling mighthave limited significance in the induction of apoptosis by ARG. Fur-

ther protein profile studies are needed to determine which signaltransduction pathways are involved in ARG-induced apoptosis.

In addition to its role in apoptosis induction, Akt is known toplay a role in the control of the cell cycle (Fig. 6). Under various

S. Susanti et al. / Phytomedicine 21 (2013) 39– 46 43

Fig. 3. The effect of arctigenin on the expression of cell cycle regulatory proteins of various human cell lines performed by Western Blot analysis. Values are expressed asthe intensity ratio of treated to untreated control cells. Data are presented as means ± SD (n = 3). *p < 0.05; **p < 0.01 versus the untreated cells (control).

44 S. Susanti et al. / Phytomedicine 21 (2013) 39– 46

F us hue esentt

caealApaAt(

ct

Fi*

ig. 4. The effect of arctigenin on the expression of apoptosis-related genes in varioxpressed as fold changes after normalization to the housekeeping gene. Data are prhe untreated cells (control).

ircumstances, the activation of Akt overcomes the G1 cell cyclerrest, thus enabling cell survival and proliferation (Ramaswamyt al., 1999). Inhibition of PI3K/Akt signaling has been shown torrest the cell cycle at the G1 phase and induce apoptosis in Hodgkinymphoma (Georgakis et al., 2005). Similarly, the suppression ofkt-1 in the ERBB-2/Akt signaling pathway can inhibit E2F-1 (arotein with a crucial role in controlling cell cycle progression)nd induce G0/G1 cell cycle arrest (Guo et al., 2010). Thus, theRG-mediated decrease in Akt-1 expression may contribute in part

o the cell cycle arrest at the G0/G1 phase observed in this study

Fig. 2).

Cell cycle arrest and apoptosis are closely linked in mammalianells. The sensitivity of cells to certain types of apoptosis induc-ion often depends on the cell cycle, and inhibition of the cell cycle

ig. 5. The effect of BSO treatment on the cytotoxicity (ED50) of arctigenin (ARG)n various human cell lines. Data are presented as means ± SD (n = 3). *p < 0.05;*p < 0.01 versus the ARG-treated cells.

man cell lines. Descriptions of the genes analyzed are shown in Table 1. Values areed as means ± SD of experiments performed in triplicate. *p < 0.05; **p < 0.01 versus

has been considered as a target for cancer treatment (Sherr, 1994;Kornberg and Lorch, 1999). A previous study reported that ARGblocked progression of the cell cycle from the G1 phase to the Sphase by down-regulation of P-Rb in human gastric cancer (Jeonget al., 2011). Similarly, ARG increased the proportion of cells in theG1 phase of the cell cycle in A549 cancer cells and induced G1/Sarrest. The investigation of cell cycle regulators showed that ARGdecreased the levels of cyclin H, CDK-7, cyclin E, CDK-2, P-CDK, andNPAT proteins and had no effect on Rb and P-Rb proteins (Fig. 3). Itis therefore likely that ARG arrests the cell cycle of lung cancer cellsvia a mechanism that is different from that for of human gastriccancer.

A previous study showed that in human gastric cancer, ARGarrested the cell cycle at the G0/G1 phase thorough the phosphor-ylation state of Rb protein, probably via the down-regulation ofcyclin E/CDK2 or cyclin D/CDK4 (Jeong et al., 2011). ARG has beenshown to arrest the cell cycle of human colorectal cancer cells inthe G2/M phase by the modulation of the Wnt/�-catenin signalingpathway (Yoo et al., 2010). These observations collectively indicatethe variation in the cell cycle arrest point and relevant signalingpathways in cancer cells treated with ARG. It is noteworthy thatARG decreased the level of cyclin H/CDK-7. In addition to cyclinbinding, CDK activity is also regulated by phosphorylation of con-served threonine and tyrosine residues. Thus, the full activationof CDK-2 requires phosphorylation of threonine 160 mediated bya cyclin H/CDK-7 complex termed CDK activating kinase (CAK).Our observation provided the first indication that ARG affected all

cell cycle phases through the down-regulation of a cyclin H/CDK-7complex (Fig. 6).

The modulation of signaling pathways by ARG appeared toconverge into the regulation of the NPAT final effector proteins to

S. Susanti et al. / Phytomedicine 21 (2013) 39– 46 45

Fig. 6. Schematic representation of the mechanism of action of arctigenin to arrest the cell cycle of human adenocarcinoma (A549) cells at the G0/G1 phase.

f the d

cbepitN

Fig. 7. Schematic depiction o

ontrol the cell proliferation process. NPAT is involved in histoneiosynthesis, and the inhibition of NPAT expression impedesxpression of all histone subtypes (Zhao et al., 2000). The phos-

horylation of NPAT by cyclin E/CDK-2 has been shown to lead to

ts association with histone gene promoters. Therefore, it is likelyhat inhibition of cyclin E/CDK2 led to the down-regulation ofPAT observed in this study (Fig. 3).

etoxification of ARG by GSH.

A previous study reported that the cytotoxicity of A. lappa L.extract is reduced by intracellular GSH (Moritani et al., 1996).In the present study, the inclusion of a GSH synthase inhibitor

enhanced the sensitivity of A549 and OUMS-36-2F cells to ARG(Fig. 5). This finding suggested that the GSH content of cancer cellsmay also determine their susceptibility to ARG. It is suggested thatGSH sequestered ARG from the system through the conjugation

4 medic

rtGa

acset3Bmt2oOt3cmgtra

A

AR

R

A

A

6 S. Susanti et al. / Phyto

eaction (illustrated in Fig. 7) and consequently reduced its cyto-oxicity. Thus, in addition to cell cycle control, the expression ofSH synthase may become of a new target for the development ofnticancer agents.

Our previous study, among others, suggested that the anticancerctivity of ARG is quite cell-specific; the cytotoxicity of ARG affectsancer cells, while normal cells are resistant. However, the currenttudy also showed variation in the cytotoxic effects of ARG on differ-nt types of normal cells. OUMS-3T-2F normal cells were sensitiveo ARG, as was the case for A549 cancer cells, whereas OUMS-6 and OUMS-36T-5F normal cells were resistant to this agent.oth OUMS-3T-2F and OUMS-3T-5F were generated via transfor-ation of OUMS-36 primary cells with the hTRT gene (human

elomerase reverse transcriptase gene in pMX) (Kouchi and Namba,000). Restriction Fragment Length Polymorphism (RFLP) analysisr (short tandem repeat (STR)-PCR) of the genomic DNA found thatUMS-3T-2F only differs from other OUMS-36 series cell lines in

he length of the vWA microsatellite markers; the STR 20 in OUMS-T-2F was replaced with STR 21 in OUMS-3T-5F. Therefore, it islear that the difference in the genomic DNA sequence close to thisarker was responsible for the susceptibility to ARG, and investi-

ation of this notion may shed light on the mechanism underlyinghe cancer-specific cytotoxicity of ARG. Furthermore, continuedesearch into this topic will contribute to the development of novelnticancer agents with fewer adverse effects.

cknowledgments

We are grateful to the council of the United Graduate School ofgricultural Sciences for the funding for this project through theendai-Student Supporting Program (2012).

eferences

lberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K., Walter, P., 2002. Figure 15-60:

BAD phosphorylation by Akt. Mol. Biol. Cell, New York, Garland Science, ISBN0-8153-3218-1.

wale, S., Lu, J., Kalauni, S.K., Kurashima, Y., Tezuka, Y., Kadota, S., Esumi, H., 2006.Identification of arctigenin as an antitumor agent having the ability to eliminatethe tolerance of cancer cells to nutrient starvation. Cancer Res. 66, 1751–1757.

ine 21 (2013) 39– 46

Georgakis, G.V., Li, Y., Rassidakis, G.Z., Medeiros, L.J., Mills, G.B., Younes, A., 2005.Inhibition of the phosphatidylinositol-3 kinase/Akt promotes G1 cell cycle arrestand apoptosis in Hodgkin lymphoma. Br. J. Haematol. 132, 503–511.

Greenlee, R.T., Hill-Harmon, M.B., Murray, T., Thun, M., 2001. Cancer statistic. CancerJ. Clin. 51, 15–36.

Guo, X., Guo, L., Ji, J., Zhang, J., Zhang, J., Chen, X., Cai, Q., Li, J., Gu, Q., Liu, B., Zhu, Z.,Yu, Y., 2010. miRNA-331-3p directly targets E2F1 and induces growth arrest inhuman gastric cancer. Biochem. Biophys. Res. Commun. 398, 1–6.

Jeong, J.B., Hong, S.C., Jeong, H.J., Koo, J.S., 2011. Arctigenin induces cell cycle arrestby blocking the phosphorylation of Rb via the modulation of cell cycle reg-ulatory proteins in human gastric cancer cells. Int. Immunopharmacol. 11,1573–1577.

Kim, Y.J., Hwang, J.H., Cha, M., Yoon, M.Y., Son, E.S., Tomida, A., Ko, B., Song, S.W.,Shin-Ya, K., Hwang, Y., Park, H.R., 2010. Arctigenin blocks the unfolded proteinresponse and shows therapeutic antitumor activity. J. Cell Physiol. 224, 33–40.

Kouchi, H., Namba, M., 2000. Normal human fibroblast cell lines transfected withthe hTRT gene-OUMS36T cell line series. Tiss. Cult. Res. Commun. 19, 203–204.

Kornberg, R.D., Lorch, Y., 1999. Twenty-five years of the nucleosome, fundamentalparticle of the eukaryotic chromosome. Cell 98, 285–294.

Matsumoto, T., Hosono-Nishiyama, K., Yamada, H., 2006. Anti-proliferative andapoptotic effects of butyrolactone lignans from Arctium lappa on leukemic cells.Planta Med. 72, 276–278.

Moritani, S., Nomura, M., Takeda, Y., Miyamoto, K.I., 1996. Cytotoxic components ofBardanae Fructus (Goboshi). Biol. Pharm. Bull. 19 (11), 1515–1517.

Pisani, P., Parkin, D.M., Bray, F., Ferlay, J., 1999. Estimates of the worldwide mortalityfrom 25 cancers in 1990. Int. J. Cancer 83 (18), 29.

Quasney, M.E., Carte, L.C., Oxford, C., Watkins, S.M., Gershwin, M.E., German, J.B.,2001. Inhibition of proliferation and induction of apoptosis in SNU-1 human gas-tric cancer cells by the plant sulfolipid, sulfoquino-vosyldiacylglycerol. J. Nutr.Biochem. 12, 310–315.

Ramaswamy, S., Nakamura, N., Vazquez, F., Batt, D.B., Perera, S., Roberts, T.M., Sellers,W.R., 1999. Regulation of G1 progression by the PTEN tumor suppressor proteinis linked to inhibition of the phosphatidylinositol 3-kinase/Akt pathway. Proc.Natl. Acad. Sci. U.S.A. 96 (5), 2110–2115.

Susanti, S., Iwasaki, H., Itokazu, Y., Nago, M., Taira, N., Saitoh, S., Oku, H., 2012. Tumorspecific cytotoxicity of arctigenin isolated from herbal plant Arctium lappa L. J.Nat. Med., http://dx.doi.org/10 1007/s11418-012-0628-0.

Sherr, C.J., 1994. G1 phase progression: cycling on cue. Cell 79, 551–555.Ten Bokkel Huinink, W.W., Bergman, B., Cheimaissani, A., Dornoff, W., Drings,

P., Kellokumpu-Lehtinen, P.L., Liippo, K., Mattson, K., von Pawel, J., Ricci, S.,Sederholm, C., Stahel, R.A., Wagenius, G., Walree, N.V., Manegold, C., 1999.Single-agent gemcitabine: an active and better tolerated alternative to standartdcisplatin-based chemotheraphy in locally advanced or metastatic non-small celllung cancer. Lung Cancer 26, 85–94.

Yoo, J.H., Lee, H.J., Kang, K., Jho, E.H., Kim, C.Y., Baturen, D., Tunsag, J., Nho, C.W.,

2010. Lignans inhibit cell growth via regulation of Wnt/�-catenin signaling. FoodChem. Toxicol. 48, 2247–2252.

Zhao, J., Kennedy, B.K., Lawrence, B.D., Barbie, D.A., Matera, A.G., Fletcher, J.A.,Harlow, E., 2000. NPAT links cyclin E-Cdk2 to the regulation of replication-dependent histone gene transcription. Genes Dev. 14, 2283–2297.