Toxicology 249 (2008) 230–237

Contents lists available at ScienceDirect

Toxicology

journa l homepage: www.e lsev ier .com/ locate / tox ico l

Arecoline, a major alkaloid of areca nut, inhibits p53, represses DNA repair,and triggers DNA damage response in human epithelial cells

Yi-Shan Tsaia,b, Ka-Wo Leec, Jau-Ling Huangd, Yu-Sen Liua, Suh-Hang Hank Juob,e,Wen-Rei Kuoc, Jan-Gowth Changa,b, Chang-Shen Lina,b,∗, Yuh-Jyh Jonga,b,∗

a Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, No. 100, Shi-Chuan 1st Road, Kaohsiung 807, Taiwanb Center of Excellence for Environmental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwanc Department of Otolaryngology, Kaohsiung Medical University Hospital, Kaohsiung, Taiwand Department of Bioscience Technology, College of Health Science, Chang Jung Christian University, Tainan, Taiwan

e Department of Medical Genetics, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan

for Rponeat BQ

a nutg arealoidounddam. Thi

ATM,mage

a r t i c l e i n f o

Article history:Received 25 March 2008Received in revised form 12 May 2008Accepted 13 May 2008Available online 21 May 2008

Keywords:Areca nutArecolineDNA damage responseDNA repairp53

a b s t r a c t

The International AgencyAreca nut is the main comical studies have shown thhas been known that areclar mechanisms underlyinthat arecoline, a major alkp53 and DNA repair. We frylation, a sensitive DNAwere elicited by arecolinehyperphosphorylation ofing that a cellular DNA da

Oral cancer for arecoline-elicited DNA damtransactivation function. As a rerepair were repressed by arecolregulated in BQ-associated orarole in BQ-associated tumorige

1. Introduction

DNA repair machineries play a pivotal role in maintaininggenome integrity (Kolodner et al., 2002; Sancar et al., 2004).Deregulation of DNA repair can result in genomic instability(Hoeijmakers, 2001; Lengauer et al., 1998), which is a hallmark ofcancer cells (Hanahan and Weinberg, 2000). Upon DNA damage, theserine 139 of histone H2AX, a histone H2A variant, is quickly phos-phorylated (Rogakou et al., 1998). This phosphorylation is mediatedby ataxia telangiectasia mutated (ATM) kinase (Burma et al., 2001),then the serine 139-phosphorylated histone H2AX is denoted as �-H2AX (Rogakou et al., 1998). Meanwhile, Mre11, Rad50, and Nbs1are recruited to the damaged lesions to form MRN complex in an

∗ Corresponding authors at: Graduate Institute of Medicine, College of Medicine,Kaohsiung Medical University, No. 100, Shi-Chuan 1st Road, Kaohsiung 807, Taiwan.Tel.: +886 7 3121101x2019; fax: +886 7 3218309.

E-mail addresses: csl@kmu.edu.tw (C.-S. Lin), yjjong@kmu.edu.tw (Y.-J. Jong).

0300-483X/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.tox.2008.05.007

esearch on Cancer declared that areca nut was carcinogenic to human.nt of betel quid (BQ), which is commonly consumed in Asia. Epidemiolog-

chewing is a predominant risk factor for oral and pharyngeal cancers. Itis genotoxic to human epithelial cells. However, the molecular and cellu-ca nut-associated genotoxicity are not fully understood. Here we showedof areca nut, might contribute to oral carcinogenesis through inhibiting

, on the biological aspect, that arecoline could induce �-H2AX phospho-age marker, in KB, HEp-2, and 293 cells, suggesting that DNA damagess phenomenon was supported by the observations of arecoline-inducedNbs1, Chk1/2, p53, and Cdc25C, as well as G2/M cell cycle arrest, indicat-response was activated. To explore the possible mechanism accountingages, we found that arecoline could inhibit p53 by its expression andsult, the expression of p53-regulated p21WAF1 and the p53-activated DNAine. Finally, we showed that p53 mRNA transcripts were frequently down-l cancer, suggesting that arecoline-mediated p53 inhibition might play anesis.

© 2008 Elsevier Ireland Ltd. All rights reserved.

ATM-dependent process (Lim et al., 2000; Wu et al., 2000). ATMis, again, activated by the DNA lesions-MRN complex (Lee andPaull, 2004; Uziel et al., 2003), and then phosphorylates histoneH2AX and Nbs1 further (Burma et al., 2001; Lim et al., 2000; Wuet al., 2000). As a result, the DNA damage signal is amplified andis passed to the downstream effectors including Chk1, Chk2, p53,Cdc25, and many others (Lisby et al., 2004; Zhou and Elledge, 2000).The cell cycle checkpoint kinases Chk1 and Chk2 are phosphory-lated and activated by ATM. Subsequently, they phosphorylate thecell cycle regulators Cdc25 and p53, leading to an inactivation ofcyclin-dependent kinases and an arrest of cell cycle (Lisby et al.,2004; Zhou and Elledge, 2000). This arrest of cell cycle progressionis essential for cells to have more time to repair DNA damages.

In addition to the function of cell cycle regulation, the tumorsuppressor p53 also plays a central role in regulating DNA repair(Helton and Chen, 2007; Sengupta and Harris, 2005). It canmodulate almost all DNA repair pathways including both excisionand DNA strand break repair (Helton and Chen, 2007). We andothers have found that p53 can enhance UV-induced nucleotide

ology

Wildbad, Germany), then 20 �l of Stop & Glo Reagent (Promega, Madison, WI, USA)was added to the lysates to quench the firefly luciferase activity and activate theRenilla luciferase activity that was read using a microplate luminometer (Centro

Y.-S. Tsai et al. / Toxic

excision repair (Liu et al., 2005; Wang et al., 2003, 1995). It is, atleast in partly, contributed by the p53’s transcriptional activity(Adimoolam and Ford, 2002; Hwang et al., 1999; Liu et al., 2005).Therefore, p53-dependent transcriptional activity is important forregulation of nucleotide excision repair by p53 (Helton and Chen,2007).

The areca nut (without tobacco) is recognized as a group I car-cinogen to human by the International Agency for Research onCancer (IARC), World Health Organization in 2004 (IARC, 2004).It is estimated that hundreds millions of Asians have a habit ofbetel quid (with areca nut) chewing in their daily life. Epidemiolog-ical studies indicate that the use of areca nut is closely associatedwith oral and pharyngeal cancers (Ko et al., 1995; Lee et al., 2005).However, the molecular mechanism underlying the carcinogenic-ity of areca nut is poorly understood. Extracts of areca nut hasbeen shown cytotoxic and genotoxic to human buccal epithelialcells (Sundqvist et al., 1989). This may be correlated to its abil-ity to increase DNA strand breaks, micronucleus formation, genemutation, and to elevate chromosomal aberrations (Panigrahi andRao, 1982; Stich et al., 1981). But there is still a lack of molecu-lar investigation regarding to the effects of areca nut on regulatingDNA repair and genome stability genes, which maintain genomefidelity. Arecoline is a major alkaloid of areca nut and is proba-bly the key substance with such genotoxic effect (Panigrahi andRao, 1982; Shirname et al., 1984; Stich et al., 1981). In addition,it has been shown that arecoline can induce cell cycle arrest atG2/M stage (Chang et al., 2001, 2004; Lee et al., 2006), which isa consequence as well as a hallmark of cells with DNA damages.To further clarify whether arecoline could elicit DNA damages ina cell, we used a sensitive biological DNA damage marker, the �-H2AX, to demonstrate that arecoline could induce DNA damages inhuman epithelial cells. We also showed that the ATM-dependentDNA damage signaling was induced by arecoline and the cell cyclewas arrested at G2/M phase. In addition, we found that arecolinecould inhibit p53, both on expression and on transactivation func-tion. This inhibition played an important role in arecoline-mediatedsuppression of DNA repair. Finally, we showed that the expressionof p53 mRNA was frequently down-regulated in BQ-associated oralcancer.

2. Materials and methods

2.1. Cell culture and arecoline treatment

The KB cells (initial passage number was 367) were purchased from the Biore-

source Collection and Research Center, Taiwan. HEp-2 cells (initial passage numberwas 273) were obtained from the Virology Section, Department of Pathology andLaboratory Medicine, Taipei Veterans General Hospital, Taiwan. The 293 and H1299cells were gifts of Dr. MR Chen (Graduate Institute of Microbiology, National TaiwanUniversity, Taiwan) and Dr. YS Lin (Institute of Biomedical Science, Academic Sinica,Taiwan), respectively. Cells were grown in Dulbecco’s modified Eagle’s medium(HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (Invitrogen,Carlsbad, CA), and were incubated at 37 ◦C with saturating humidity and 5% CO2. Allexperiments were performed within 6 weeks after thawing of the cells from the ini-tial stocks. To investigate the effect of arecoline (Sigma–Aldrich, St. Louis, MO, USA),various concentrations [0.3 mM in most experiments, which is the concentrationusually observed in the saliva of betel quid chewer (Nair et al., 1985)] of arecolinewere added into culture medium for 24 h (or indicated time points), then cells wereharvested and analyzed. Arecoline was dissolved in sterile distilled water as 100 mMstock and was diluted to working concentration using culture medium.

2.2. Indirect immunofluorescent staining of �-H2AX

To detect �-H2AX phosphorylation, cells were fixed with ice-coldmethanol/acetone (1:1) for 10 min, stained with anti-phospho-H2AX anti-body (JBW301, Upstate, Lake Placid, NY) at a 1:2000 dilution, then visualized byRhodamine RedTM-X-conjugated secondary antibody (Jackson-ImmunoResearchLaboratories, West Grove, PA, USA). The cells were counterstained with 1 �g/mlof 4,6-diamidino-2-phenylindole (DAPI) for 5 min at room temperature and theimages were taken using a digital camera (CoolSNAP ES, Photometrics, Tucson, AZ,

249 (2008) 230–237 231

USA) connected to an upright fluorescent microscope (Axioplan 2 Imaging MOT,Carl Zeiss, Germany).

2.3. Western blot analyses

Cells were harvested in RIPA lysis buffer [50 mM Tris–HCl (pH 8.0), 150 mMNaCl, 0.5% (w/v) sodium deoxycholate, 1% (v/v) Nonidet P-40, 0.1% (w/v) SDS, 1 mMDTT, 50 mM NaF, 1 mM Na3VO4, 10 mM �-glycerol phosphate, 1 mM EGTA] con-taining a cocktail of protease inhibitors (Roche, Mannheim, Germany) on ice for30 min. Protein lysates were cleared by centrifugation and protein concentrationswere determined using Bio-Rad protein assay kit. Equivalent protein lysates wereadded in Laemmli buffer, boiled, separated by SDS–PAGE, and then transferred topolyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). ThePVDF membranes were blocked in 5% nonfat milk/TBS-T (20 mM Tris–HCl, pH 7.4,150 mM NaCl, 0.1% Tween 20) for 1 h, followed by TBS-T washes and an overnight(4 ◦C) or 1-h (room temperature) incubation with primary antibody. This step wasfollowed by three TBS-T washes and by the incubation with secondary antibodyfor 1 h at room temperature. After 3 TBS-T washes, the protein was visualized byenhanced chemiluminescence (Millipore, Bedford, MA, USA) and autoradiography.For ATM (∼370 kDa), the 3–8% of NuPAGE Tris-acetate gels (Invitrogen, Carlsbad,CA, USA) were used for proper separation of the proteins. Others were carriedout by regular SDS–PAGE. A mouse monoclonal antibody against �-actin (AC-15,Sigma–Aldrich, St. Louis, MO, USA) or GAPDH (6C5, Biodesign, Saco, ME, USA) wasused as an internal control. The antibodies against ATM (2C1), Chk1 (FL-476), Chk2(H-300), p53 (DO-1), Cdc25C (C-20), Cdc25C-pS216, Cyclin B1 (GNS1) were obtainedfrom Santa Cruz (Santa Cruz, CA, USA); Chk1-pS317 and -pS345, Chk2-pS19 and -pT68, p53-pS15, Nbs1-pS343, Nbs1, and p21WAF1 antibodies were purchased fromCell Signaling (Danvers, MA, USA); the ATM-pS1981 antibody was from Rockland(Gilbertsville, PA, USA).

2.4. Cell cycle analysis by flow cytometry

For cell cycle analysis, cells that were treated with or without arecoline weretrypsinized and fixed with ice-cold 100% methanol for 30 min, then stained with10 �g/ml of propidium iodide. The samples were applied to FACScan and analyzedusing Cell Quest software (BD Biosciences, Franklin Lakes, NJ, USA).

2.5. Analyses of DNA repair using host cell reactivation (HCR) assay

The HCR assay was conducted as described previously (Liu et al., 2005). Briefly,the purified firefly luciferase reporter plasmid (pCMV-Luc), either damaged with2000 J/m2 of UV light or mock treated, together with an undamaged Renilla luciferasereporter pRL-CMV (Promega, Madison, WI, USA) as an internal control were co-transfected into cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). At24 h post-transfection, cells were lysed in 100 �l lysis buffer (0.1 M HEPES, pH 7.8,1% Triton X-100, 1 mM CaCl2 and 1 mM MgCl2) and assayed for firefly luciferaseactivity using 40 �l aliquots of the lysates. First, 20 �l of Luciferase Assay ReagentI (Promega, Madison, WI, USA) was added to the lysates for measuring the fire-fly luciferase activity by a microplate luminometer (Centro LB 960, Berthold, Bad

LB 960, Berthold, Bad Wildbad, Germany). The firefly luciferase activity of eachsample was normalized to the Renilla luciferase activity. The DNA repair activity,which was represented by the fold of HCR, was calculated by dividing the repairconversion of arecoline-treated cells by that of vehicle-treated cells. The repair con-version was calculated by dividing the normalized firefly luciferase activity fromcells transfected with UV-irradiated pCMV-Luc by that of non-irradiated pCMV-Luctansfectants. Data from at least three independent experiments were averaged tocalculate the mean and the standard deviation. To study the effect of arecoline onp53-regulated DNA repair, the wild-type p53 expression plasmids pCSL5 (Lin et al.,2000) or the vector control, together with an internal control pRL-CMV, were co-transfected into the p53-null H1299 cells, then arecoline was added to the cells for24 h. The resulting HCR activity of the pCSL5-transfectants was compared with thoseof vector control.

2.6. Luciferase activity assays for p53-regulated promoters

The p53-regulated promoter-luciferase reporter plasmids, p3PREc-Luc (contains3 copies of consensus p53-binding sites), and the p21WAF1 promoter-luciferase con-struct, p21-Luc (contains two p53-binding sites), were described in our previouspaper (Lin et al., 2000). To examine the effect of arecoline on p53’s transactiva-tion function, p3PREc-Luc or p21-Luc, together with an internal control pRL-CMV,were co-transfected into cells using Lipofectamine 2000 (Invitrogen, Carlsbad,CA, USA), then the cells were treated with various doses of arecoline for 24 hand harvested. The dual-luciferase assays were performed as described in HCRassay.

ology

232 Y.-S. Tsai et al. / Toxic

2.7. Clinical samples

The 13-paired oral cancer specimens were collected from the Department ofOtolaryngology, Kaohsiung Medical University Hospital (KMUH). These specimenswere collected after obtaining patients’ informed consents that were approved bythe Institutional Review Board. Specimens were snap frozen in liquid nitrogen andwere stored at −80 ◦C until use.

2.8. RNA isolation and real-time quantitative reverse transcription-polymerasechain reaction (RT-PCR)

The method for quantitatively analyzing mRNA expression was as described(Chiou et al., 2007). Briefly, total RNA was isolated by Tri-reagent (Sigma–Aldrich,St. Louis, MO, USA). Reverse transcription was made using one microgram of total

RNA and the High-Capacity cDNA Archive Kit (Applied BioSystems, Foster City, CA,USA), according to the manufacturer’s instruction. The resulting cDNA was 20 �l, andwas subsequently diluted to a total volume of 100 �l. A 2 �l of diluted cDNA wasused as templates for real-time quantitative PCR (Q-PCR). Q-PCR was conductedby PowerSYBR Green reagent (Applied BioSystems, Foster City, CA, USA) in 20 �land ABI Prism 7500 Sequence Detection System instrument (Applied BioSystems,Foster City, CA, USA). Cycling conditions were 50 ◦C for 2 min and 95 ◦C for 10 minfollowed by 50 cycles at 95 ◦C for 15 s and 60 ◦C for 1 min. Dissociation curve wasadded at the final step to ensure that only the specific target was amplified. Atleast three independent analyses were performed for each sample. The PCR primerswere designed using the web-based ProbeFinder software (Roche Applied Science,Mannheim, Germany). The primer sequences used were: p53-F AGGCCTTGGAACT-CAAGGAT, p53-R CCCTTTTTGGACTTCAGGTG; p21-F TCACTGTCTTGTACCCTTGTGC;p21-R GGCGTTTGGAGTGGTAGAAA; GAPDH-F AGCCACATCGCTCAGACAC, GAPDH-RGCCCAATACGACCAAATCC. Differential RNA expressions between various sampleswere calculated by 2−��CT method (Livak and Schmittgen, 2001) and using GAPDHas an internal control.

3. Results

3.1. Arecoline induces �-H2AX phosphorylation

To evaluate whether arecoline could induce DNA damage atmolecular level, we used �-H2AX as a marker to examine DNA

Fig. 1. Arecoline induces �-H2AX phosphorylation and formation of DNA damage foci.24 h, then immunofluorescent assays were conducted using antibody against the phosp4,6-diamidino-2-phenylindole. (C and D) Western blot analyses of �-H2AX phosphorylacells were treated with various doses of arecoline for 24 h (C), or with 0.3 mM of arecolanalyses.

249 (2008) 230–237

damage foci in human epithelial cells. We first treated the oral can-cer cell line KB with or without 0.3 mM, the concentration usuallyfound in saliva of betel quid chewers (Nair et al., 1985), of arecol-ine for 24 h, and an immunofluorescent assay was performed usinganti-�-H2AX antibody. As shown in Fig. 1A, DNA damage foci wereclearly observed in the arecoline-treated cells, but it was not thecase in control cells. Similar results were seen in the arecoline-treated HEp-2 laryngeal carcinoma cell line (Fig. 1B) and 293 humanembryonic kidney cells (data not shown). To further character-ize the effect of arecoline on DNA damage, we treated KB and293 cells with various doses of arecoline and examined �-H2AXphosphorylation using Western blots. Fig. 1C shows that �-H2AX

phosphorylation is correlated well to arecoline doses in KB and 293cells. The kinetics studies demonstrated that �-H2AX phosphory-lation was gradually accumulated by arecoline treatment (Fig. 1D).We also found that whole areca nut extract (1000 �g/ml, but not200 �g/ml) could induce �-H2AX phosphorylation in KB cells (datanot shown). These data indicated that arecoline could induce DNAdamages in human epithelial cells by dose- and time-dependentmanners.

3.2. Arecoline triggers the ATM-dependent signal pathway andinduces G2/M arrest

Since ATM is the major kinase that can phosphorylate H2AXand various downstream targets, such as Chk1/2, p53, and Cdc25,it plays a central role in activating DNA damage response (Zhouand Elledge, 2000). For this reason, we examined whether areco-line could activate ATM kinase by using an antibody specificallyrecognizing the phosphoserine 1981 of ATM, which could reflectits kinase activity (Bakkenist and Kastan, 2003). Fig. 2A shows thatarecoline increases ATM phosphorylation in KB, HEp-2, and 293cells in a dose-dependent manner. This indicated that arecoline

KB (A) and HEp-2 (B) cells were treated with 0.3 mM of arecoline or vehicle forhoserine 139 of histone H2AX to display DNA damage foci. Arec., arecoline; DAPI,tion demonstrated the time- and dose-dependent effects of arecoline. KB and 293ine for different periods (D), then the cell lysates were prepared for Western blot

Y.-S. Tsai et al. / Toxicology 249 (2008) 230–237 233

Fig. 2. Arecoline triggers the ATM-dependent DNA damage response and causes cell cycle arrest. (A) Arecoline induces ATM phosphorylation at serine 1981 in a dose-dependent manner. KB, HEp-2, and 293 cells were treated with various doses of arecoline for 24 h, then were harvested for Western blot analyses by using 3–8% of NuPAGEgels. The phosphorylation status of serine 1981 of ATM (ATM-p) was detected by a rabbit anti-ATM phosphoserine 1981 antibody. The total ATM (ATM) and �-actin wereused as loading controls. (B) Hyperphosphorylation of ATM downstream substrates in arecoline-treated KB and HEp-2 cells. Cells were treated with 0.3 mM of arecoline orvehicle for 24 h, then were harvested for Western blot analyses using antibodies against various ATM targets. (C) Arecoline induces inhibitory phosphorylation of Cdc25Cat serine 216 and accumulation of cyclin B1. KB and HEp-2 cells were treated as described in (B), and then Western blots were performed. The hyperphosphorylation formof Cdc25C (indicating by arrowhead) could also be detected using an antibody against total Cdc25C. GAPDH was used as an internal control. (D) Arecoline causes G2/M cellcycle arrest. Cells were treated with 0.3 mM of arecoline for 0, 12, and 24 h, and then harvested for flow cytometric analyses. The cells with DNA contents of 2N and 4N areindicated by arrowheads.

ology 249 (2008) 230–237

234 Y.-S. Tsai et al. / Toxic

could activate ATM kinase activity. We next investigated whetherthe ATM downstream targets were hyperphosphorylated uponarecoline treatment. As shown in Fig. 2B, the serines 317 and 345of Chk1 kinase; the serine 19 and threonine 68 of Chk2 kinase;the serine 15 of p53; and the serine 343 of Nbs1 were hyperphos-phorylated in arecoline-treated KB and HEp-2 cells. These resultssuggested that arecoline could activate ATM-dependent DNA dam-age response, which usually resulted in cell cycle arrest at G2/Mphase. In this regard, we examined the effects of arecoline on thekey regulators that control G2/M progression. We found that areco-line treatment could induce a hyperphosphorylation of Cdc25C atserine 216 and cause an accumulation of cyclin B1 (Fig. 2C), bothindicated that these cells were at G2/M phase. Flow cytometricanalyses also demonstrated that arecoline treatment led to a cellcycle arrest at G2/M stage (Fig. 2D). Taken together, these dataindicated that arecoline could trigger the ATM-dependent signalpathway that resulted in �-H2AX phosphorylation and cell cyclearrest at G2/M.

3.3. Arecoline suppresses DNA repair

Arecoline has been shown to increase micronuclei formation,gene mutations, and chromosome aberrations in mammalian cells(Panigrahi and Rao, 1982; Shirname et al., 1984; Stich et al., 1981).Along with our findings that arecoline could induce �-H2AX phos-phorylation (Fig. 1) and activate the ATM-dependent DNA damageresponses (Fig. 2), it indicated that more severe DNA damages werepresent in arecoline-treated cells. This might be a result of an inhi-bition of DNA repair by arecoline. To examine this possibility, weconducted the host cell reactivation assays (Liu et al., 2005) toevaluate the effect of arecoline on DNA repair. In this assay, anUV-damaged luciferase reporter plasmid (pCMV-Luc) was trans-fected into HEp-2 cells, then the cells were treated with arecolineor vehicle for 24 h. If the UV-damaged pCMV-Luc was repaired bythe host cells, then luciferase could be expressed. Therefore, theresulting luciferase activity could reflect the DNA repair activity ofthe recipient host cells. Fig. 3A shows a lower luciferase activityin arecoline-treated cells than in vehicle control. This repressionof luciferase activity correlated to the doses of arecoline (Fig. 3A),suggesting that arecoline could repress DNA repair. Since we andothers have demonstrated that p53 played an important role inregulating DNA repair (Liu et al., 2005; Wang et al., 2003, 1995), wewondered whether arecoline had any effect on p53-regulated DNArepair. We showed previously that ectopically expressed p53 could

stimulate HCR activity in the p53-null H1299 cells (Liu et al., 2005),here we found that the p53-stimulated HCR activity in H1299 cellswas repressed by arecoline treatment (Fig. 3B). However, the HCRactivity was not repressed apparently in the absence of endoge-nous p53 (Fig. 3B), suggesting that arecoline repressed DNA repairthrough, at least in partly, inhibiting p53.

3.4. Arecoline inhibits the expression and transactivationfunction of p53

Regarding the effect of arecoline on inhibiting p53, we notedthat the p53 protein expression was repressed in arecoline-treatedKB and HEp-2 cells (Fig. 2B). Such repression of p53 protein couldalso be found in whole areca nut extract (1000 �g/ml)-treated KBcells (data not shown). To further verify this finding, we examinedwhether arecoline could inhibit p53 mRNA expression by real-time quantitative RT-PCR (qRT-PCR). Fig. 4A shows that p53 mRNAexpression was down-regulated in arecoline-treated KB, HEp-2,and 293 cells. These results indicated that arecoline could repressp53 expression. We next investigated whether p53’s transactiva-tion function was affected by arecoline. To this aim, a luciferase

Fig. 3. Host cell reactivation (HCR) assays for arecoline’s effect on inhibiting UV-induced DNA repair. (A) HEp-2 cells were transfected with a UV-damaged luciferasereporter plasmid (pCMV-Luc) and treated with various doses of arecoline for 24 h,then were harvested for luciferase assay. The capacity of DNA repair (HCR activity)was represented by luciferase activity (see Section 2). (B) The p53-activated HCRactivity is repressed by arecoline. The UV-damaged pCMV-Luc was co-transfectedwith or without the wild-type p53 expression plasmid into the p53-null H1299 lungcancer cells, and treated with 0.3 mM of arecoline or vehicle for 24 h. The cells werethen harvested for luciferase activity assays. HCR activity could be activated by theectopic expression of p53 in H1299 cells (comparing column 1 with 3). Additionof arecoline could reduce this activation (columns 3 and 4), but had no apparentinhibition in the absence of p53 (columns 1 and 2). The results presented representthe mean and standard deviation from at least three independent experiments.

reporter containing 3 consensus p53-binding sites (Lin et al., 2000)was transfected into HEp-2 and 293 cells, then the cells weretreated with 0.3 mM of arecoline for 24 h and luciferase assays wereperformed. The results showed that arecoline could inhibit p53’stransactivation function both in HEp-2 and in 293 cells (Fig. 4B).In addition, the p53 regulated p21WAF1 promoter, which containstwo p53-binding sites, was also repressed by arecoline (Fig. 5A).This resulted in down-regulation of p21WAF1 mRNA and proteinexpressions (Fig. 5B and C). Taken together, these data indicatedthat arecoline could inhibit the expressions of p53, as well as thep53 target genes.

3.5. The expression of p53 mRNA is down-regulated in the betelquid-associated oral cancer specimens

To evaluate whether the repression of p53 by arecoline foundin vitro had any implication on betel quid-associated malignancies,we first examined the p53 mRNA expression in 13-paired oral can-cer specimens using qRT-PCR. Among them, as shown in Fig. 6, 10specimens were obtained from BQ-chewer (solid bar), and 3 were

Y.-S. Tsai et al. / Toxicology 249 (2008) 230–237 235

Fig. 5. Arecoline inhibits the promoter activity, mRNA, and protein expressions ofp21WAF1, a p53-regulated gene. (A) Luciferase activity assays of the p21WAF1 pro-moter (p21-Luc) affected by arecoline. HEp-2 cells were transfected with p21-Lucand treated as in Fig. 4B. (B) Real-time quantitative RT-PCR analyses of 21WAF1 mRNAexpression. KB, HEp-2, and 293 cells were treated as in Fig. 4A. Both results werepresented as a mean and standard deviation from at least three independent exper-iments. (C) Western blot analyses of p21WAF1 protein expression. KB, HEp-2, and293 cells were treated with 0.3 mM of arecoline or vehicle for 24 h, then cells wereharvested for Western blot analyses.

tive DNA damage marker, �-H2AX phosphorylation, to demonstrateon biological aspect that DNA damages, indeed, took place uponarecoline treatment (Fig. 1). This arecoline-induced �-H2AX phos-

Fig. 4. Arecoline inhibits p53 mRNA expression and its transactivating function.(A) Real-time quantitative RT-PCR (qRT-PCR) analyses of p53 mRNA expression. KB,HEp-2, and 293 cells were treated with 0.3 mM of arecoline (solid bar) or vehi-cle (open bar) for 24 h, then cells were harvested for RNA isolation and qRT-PCR.The GAPDH transcripts were used as an internal control to calibrate the p53 mRNAexpression for each sample. And the relative p53/GAPDH expression was set to 1for vehicle controls. *P < 0.05 when compares with vehicle-treated cells (Student’st-test). (B) Luciferase activity assays of the reporter plasmid (p3PREc-Luc) contain-

ing only 3 p53-binding sites and a TATA box affected by arecoline. HEp-2 and 293cells were transfected with p3PREc-Luc and treated with various doses of arecolinefor 24 h, then the cells were harvested for luciferase assays. The luciferase activitiesfrom vehicle-treated cells were set to 1. All of these results were presented as themean and standard deviation from at least three independent experiments. *P < 0.05when compares with vehicle-treated cells (Student’s t-test).

from non-chewer (open bar). The results showed that p53 mRNAwas frequently down-regulated in tumor tissues when comparedwith their normal counterparts (Fig. 6). These data suggested thatsuppression of p53 might play an important role in tumorigenesisof BQ-associated oral cancer.

4. Discussion

It has been shown that arecoline exhibited genotoxic activityby elevating micronucleus in mice erythrocytes, and by increasingmutagenicity in V79 Chinese hamster cells as well as in bacteria(Shirname et al., 1984, 1983). However, no significant amounts ofDNA strand breaks could be observed in arecoline-treated humanbuccal epithelial cells (Sundqvist et al., 1989). Here, we used a sensi-

phorylation was dose- and time-dependent, indicating the intrinsicgenotoxicity of arecoline in human epithelial cells. Besides �-H2AX

Fig. 6. The p53 mRNA transcripts are frequently down-regulated in betel quid (BQ)chewing-associated oral cancer. The relative expressions of p53 mRNA from 13-paired oral cancer tumor tissues to their normal counterparts were analyzed usingreal-time quantitative RT-PCR. The expression of GAPDH mRNA was used as aninternal control to calibrate the p53 mRNA expression for each sample. Solid bar,specimens obtained from BQ-chewer; Open bar, specimens from non-chewer.

ology

236 Y.-S. Tsai et al. / Toxic

phosphorylation, we also showed that arecoline could induce theATM-dependent DNA damage signaling, by which Nbs1, Chk1/2,and p53 were hyperphosphorylated and the cell cycle was arrestedat G2/M phase in arecoline-treated KB, HEp-2, and 293 cells (Fig. 2).These results first demonstrated that arecoline could trigger a func-tional DNA damage response in a cell, which led to an activation ofcell cycle checkpoint and resulted in cell cycle arrest at G2/M. There-fore, our finding of arecoline-induced DNA damage response couldprovide a molecular explanation for the previous observations ofarecoline-induced cyclin B1 expression and G2/M cell cycle arrestin KB cells (Chang et al., 2001, 2004; Lee et al., 2006).

There were two possible mechanisms that might correspond toarecoline-induced �-H2AX phosphorylation. First, arecoline mighttarget at DNA and result in DNA damages. It is known thatarecoline can be metabolized to nitroso-compounds, such as 3-(N-nitrosomethylamino)propionitrile (NMPN) in saliva of betel quidchewers (Prokopczyk et al., 1987; Wenke and Hoffmann, 1983).NMPN is a potent carcinogen that can methylate and cyanoethylateDNA in rats (Prokopczyk et al., 1988, 1987; Wenke et al., 1984), andeventually results in DNA damages. Second, arecoline might repressDNA repair through inhibiting DNA repair machineries. As a result,both endogenous and exogenous borne DNA damages could not befully repaired. It is not known previously whether arecoline has anyeffect on regulating DNA repair. Here we used host cell reactivation,a functional assay for DNA repair, to evaluate the role of arecolinein the repair of a reporter plasmid damaged by UV. Our data clearlyshowed that the repair of UV-damaged plasmid was repressed inHEp-2 cells treated with arecoline (Fig. 3A). To our knowledge, thisis the first demonstration of arecoline’s effect on inhibiting DNArepair functionally. This repression of DNA repair was associatedwith arecoline-mediated inhibition of p53, since arecoline did notinhibit HCR activity in the H1299 cells that was absent of wild-typep53; and when the p53 expression was restored in H1299, arecol-ine became capable of repressing DNA repair (Fig. 3B). These resultsimplied that p53 might be the major target of arecoline that causeda suppression of UV-induced DNA repair. However, the involvementof other DNA repair genes that were responsive to UV-induced DNAdamages and were affected by arecoline could not be excluded.

The inhibition of p53 by arecoline was further confirmed bythe findings of arecoline-mediated down-regulation of p53 andp21WAF1 expressions (Figs. 4A and 5), as well as the repressionof p53’s transactivation activity (Fig. 4B). Although the serine 15of p53 was hyperphosphorylated, which was as a result of ATMkinase activation (Banin et al., 1998), upon arecoline treatment,

the overall reduction of p53 expression still caused suppressionof its associated DNA repair (Fig. 3B) and transactivation activ-ities (Fig. 4B). Most importantly, we found that the p53 mRNAwas frequently down-regulated in BQ-associated oral cancer spec-imens (Fig. 6). Our preliminary data also showed that p53 mRNAexpression was repressed in pharyngeal cancer (unpublished data).These observations implied that down-regulation of p53, possiblyby arecoline and/or other ingredients of betel quid, should playa critical role in the tumorigenesis of BQ-associated malignan-cies. Besides arecoline-mediated repression, the genome guardianactivity of p53 can also be inhibited through tobacco usage, genemutation, or human papillomavirus infection of oral epithelial cells(Blons and Laurent-Puig, 2003; Gillison et al., 2000; Pfeifer et al.,2002). These factors may be the reasons accounting for reduced p53mRNA found in the two patients (cases 4 and 13) with no BQ chew-ing, and are needed to be examined further. Therefore, it seems thatinactivation of p53, through multiple pathways, may be essentialfor the formation of oral cancer.

Although we have demonstrated that arecoline could inhibitp53 expression and its function, we did not know the molecu-lar mechanism underlying this inhibition so far. It is possible that

249 (2008) 230–237

arecoline could affect p53 expression by repressing its promoteractivity, since we found that the p53 mRNA levels were reducedin arecoline-treated KB, HEp-2, and 293 cells (Fig. 4A). The p53promoter activity might be repressed through certain transcrip-tion factor(s) that was affected by arecoline; or through epigeneticmodification on the promoter. In this regard, we have found thatarecoline could regulate expressions of the genes involving inmethylation and acetylation of DNA or protein (unpublished data),by which it might contribute to arecoline-mediated inhibition ofp53 mRNA expression. The exact molecular mechanism accountingfor arecoline-induced p53 down-regulation is still under investiga-tion.

In summary, this study presented a mechanism to elucidate thegenotoxicity of arecoline. That arecoline could repress p53’s expres-sion and transactivation function, as well as inhibiting UV-inducedDNA repair. As a result, it might cause elevated DNA damages andgene alterations in human epithelial cells.

Conflict of interest

None.

Acknowledgements

We thank Ms. Yu-Ting Chang and Cathy Yang for critical read-ing of the manuscript. This study was supported by grants fromKaohsiung Medical University (QC094004, Q096019, Q097027) andKaohsiung Medical University Hospital (5D-27).

References

Adimoolam, S., Ford, J.M., 2002. p53 and DNA damage-inducible expression ofthe xeroderma pigmentosum group C gene. Proc. Natl. Acad. Sci. U.S.A. 99,12985–12990.

Bakkenist, C.J., Kastan, M.B., 2003. DNA damage activates ATM through intermolec-ular autophosphorylation and dimer dissociation. Nature 421, 499–506.

Banin, S., Moyal, L., Shieh, S., Taya, Y., Anderson, C.W., Chessa, L., Smorodinsky, N.I.,Prives, C., Reiss, Y., Shiloh, Y., Ziv, Y., 1998. Enhanced phosphorylation of p53 byATM in response to DNA damage. Science 281, 1674–1677.

Blons, H., Laurent-Puig, P., 2003. TP53 and head and neck neoplasms. Hum. Mutat.21, 252–257.

Burma, S., Chen, B.P., Murphy, M., Kurimasa, A., Chen, D.J., 2001. ATM phosphory-lates histone H2AX in response to DNA double-strand breaks. J. Biol. Chem. 276,42462–42467.

Chang, M.C., Ho, Y.S., Lee, P.H., Chan, C.P., Lee, J.J., Hahn, L.J., Wang, Y.J., Jeng, J.H., 2001.Areca nut extract and arecoline induced the cell cycle arrest but not apoptosisof cultured oral KB epithelial cells: association of glutathione, reactive oxygen

species and mitochondrial membrane potential. Carcinogenesis 22, 1527–1535.

Chang, M.C., Wu, H.L., Lee, J.J., Lee, P.H., Chang, H.H., Hahn, L.J., Lin, B.R., Chen, Y.J.,Jeng, J.H., 2004. The induction of prostaglandin E2 production, interleukin-6production, cell cycle arrest, and cytotoxicity in primary oral keratinocytes andKB cancer cells by areca nut ingredients is differentially regulated by MEK/ERKactivation. J. Biol. Chem. 279, 50676–50683.

Chiou, S.S., Huang, J.L., Tsai, Y.S., Chen, T.F., Lee, K.W., Juo, S.H., Jong, Y.J., Hung,C.H., Chang, T.T., Lin, C.S., 2007. Elevated mRNA transcripts of non-homologousend-joining genes in pediatric acute lymphoblastic leukemia. Leukemia 21,2061–2064.

Gillison, M.L., Koch, W.M., Capone, R.B., Spafford, M., Westra, W.H., Wu, L., Zahurak,M.L., Daniel, R.W., Viglione, M., Symer, D.E., Shah, K.V., Sidransky, D., 2000. Evi-dence for a causal association between human papillomavirus and a subset ofhead and neck cancers. J. Natl. Cancer Inst. 92, 709–720.

Hanahan, D., Weinberg, R.A., 2000. The hallmarks of cancer. Cell 100, 57–70.Helton, E.S., Chen, X., 2007. p53 modulation of the DNA damage response. J. Cell.

Biochem. 100, 883–896.Hoeijmakers, J.H., 2001. Genome maintenance mechanisms for preventing cancer.

Nature 411, 366–374.Hwang, B.J., Ford, J.M., Hanawalt, P.C., Chu, G., 1999. Expression of the p48 xeroderma

pigmentosum gene is p53-dependent and is involved in global genomic repair.Proc. Natl. Acad. Sci. U.S.A. 96, 424–428.

IARC, 2004. Betel-quid and areca-nut chewing and some areca-nut-derivednitrosamines—summary of data reported and evaluation. IARC Monographs onthe Evaluation of Carcinogenic Risks to Humans 85, 1–334.

Ko, Y.C., Huang, Y.L., Lee, C.H., Chen, M.J., Lin, L.M., Tsai, C.C., 1995. Betel quid chewing,cigarette smoking and alcohol consumption related to oral cancer in Taiwan. J.Oral Pathol. Med. 24, 450–453.

ology

Y.-S. Tsai et al. / Toxic

Kolodner, R.D., Putnam, C.D., Myung, K., 2002. Maintenance of genome stability inSaccharomyces cerevisiae. Science 297, 552–557.

Lee, J.H., Paull, T.T., 2004. Direct activation of the ATM protein kinase by theMre11/Rad50/Nbs1 complex. Science 304, 93–96.

Lee, K.W., Kuo, W.R., Tsai, S.M., Wu, D.C., Wang, W.M., Fang, F.M., Chiang, F.Y., Ho, K.Y.,Wang, L.F., Tai, C.F., Kao, E.L., Chou, S.H., Lee, C.H., Chai, C.Y., Ko, Y.C., 2005. Dif-ferent impact from betel quid, alcohol and cigarette: risk factors for pharyngealand laryngeal cancer. Int. J. Cancer 117, 831–836.

Lee, P.H., Chang, M.C., Chang, W.H., Wang, T.M., Wang, Y.J., Hahn, L.J., Ho, Y.S., Lin, C.Y.,Jeng, J.H., 2006. Prolonged exposure to arecoline arrested human KB epithelialcell growth: regulatory mechanisms of cell cycle and apoptosis. Toxicology 220,81–89.

Lengauer, C., Kinzler, K.W., Vogelstein, B., 1998. Genetic instabilities in human can-cers. Nature 396, 643–649.

Lim, D.S., Kim, S.T., Xu, B., Maser, R.S., Lin, J., Petrini, J.H., Kastan, M.B., 2000.ATM phosphorylates p95/nbs1 in an S-phase checkpoint pathway. Nature 404,613–617.

Lin, C.S., Kuo, H.H., Chen, J.Y., Yang, C.S., Wang, W.B., 2000. Epstein-Barr virus nuclearantigen 2 retards cell growth, induces p21WAF1 expression, and modulates p53activity post-translationally. J. Mol. Biol. 303, 7–23.

Lisby, M., Barlow, J.H., Burgess, R.C., Rothstein, R., 2004. Choreography of the DNAdamage response: spatiotemporal relationships among checkpoint and repairproteins. Cell 118, 699–713.

Liu, M.T., Chang, Y.T., Chen, S.C., Chuang, Y.C., Chen, Y.R., Lin, C.S., Chen, J.Y., 2005.Epstein-Barr virus latent membrane protein 1 represses p53-mediated DNArepair and transcriptional activity. Oncogene 24, 2635–2646.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 25,402–408.

Nair, J., Ohshima, H., Friesen, M., Croisy, A., Bhide, S.V., Bartsch, H., 1985. Tobacco-specific and betel nut-specific N-nitroso compounds: occurrence in saliva andurine of betel quid chewers and formation in vitro by nitrosation of betel quid.Carcinogenesis 6, 295–303.

Panigrahi, G.B., Rao, A.R., 1982. Chromosome-breaking ability of arecoline, a majorbetel-nut alkaloid, in mouse bone-marrow cells in vivo. Mutat. Res. 103,197–204.

Pfeifer, G.P., Denissenko, M.F., Olivier, M., Tretyakova, N., Hecht, S.S., Hainaut, P.,2002. Tobacco smoke carcinogens, DNA damage and p53 mutations in smoking-associated cancers. Oncogene 21, 7435–7451.

Rogakou, E.P., Pilch, D.R., Orr, A.H., Ivanova, V.S., Bonner, W.M., 1998. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol.Chem. 273, 5858–5868.

249 (2008) 230–237 237

Prokopczyk, B., Bertinato, P., Hoffmann, D., 1988. Cyanoethylation of DNA in vivo by3-(methylnitrosamino)propionitrile, an Areca-derived carcinogen. Cancer Res.48, 6780–6784.

Prokopczyk, B., Rivenson, A., Bertinato, P., Brunnemann, K.D., Hoffmann, D., 1987.3-(Methylnitrosamino)propionitrile: occurrence in saliva of betel quid chewers,carcinogenicity, and DNA methylation in F344 rats. Cancer Res. 47,467–471.

Sancar, A., Lindsey-Boltz, L.A., Unsal-Kacmaz, K., Linn, S., 2004. Molecular mecha-nisms of mammalian DNA repair and the DNA damage checkpoints. Annu. Rev.Biochem. 73, 39–85.

Sengupta, S., Harris, C.C., 2005. p53: traffic cop at the crossroads of DNA repair andrecombination. Nat. Rev. Mol. Cell Biol. 6, 44–55.

Shirname, L.P., Menon, M.M., Bhide, S.V., 1984. Mutagenicity of betel quid and itsingredients using mammalian test systems. Carcinogenesis 5, 501–503.

Shirname, L.P., Menon, M.M., Nair, J., Bhide, S.V., 1983. Correlation of mutagenicity

and tumorigenicity of betel quid and its ingredients. Nutr. Cancer 5, 87–91.

Stich, H.F., Stich, W., Lam, P.P., 1981. Potentiation of genotoxicity by concurrentapplication of compounds found in betel quid: arecoline, eugenol, quercetin,chlorogenic acid and Mn2+. Mutat. Res. 90, 355–363.

Sundqvist, K., Liu, Y., Nair, J., Bartsch, H., Arvidson, K., Grafstrom, R.C., 1989. Cytotoxicand genotoxic effects of areca nut-related compounds in cultured human buccalepithelial cells. Cancer Res. 49, 5294–5298.

Uziel, T., Lerenthal, Y., Moyal, L., Andegeko, Y., Mittelman, L., Shiloh, Y., 2003. Require-ment of the MRN complex for ATM activation by DNA damage. EMBO J. 22,5612–5621.

Wang, Q.E., Zhu, Q., Wani, M.A., Wani, G., Chen, J., Wani, A.A., 2003. Tumor suppressorp53 dependent recruitment of nucleotide excision repair factors XPC and TFIIHto DNA damage. DNA Repair (Amst.) 2, 483–499.

Wang, X.W., Yeh, H., Schaeffer, L., Roy, R., Moncollin, V., Egly, J.M., Wang, Z., Freidberg,E.C., Evans, M.K., Taffe, B.G., Bohr, V.A., Weeda, G., Hoeijmakers, J.H., Forrester, K.,Harris, C.C., 1995. p53 modulation of TFIIH-associated nucleotide excision repairactivity. Nat. Genet. 10, 188–195.

Wenke, G., Hoffmann, D., 1983. A study of betel quid carcinogenesis. 1. On the invitro N-nitrosation of arecoline. Carcinogenesis 4, 169–172.

Wenke, G., Rivenson, A., Hoffmann, D., 1984. A study of betel quid carcinogene-sis. 3. 3-(Methylnitrosamino)-propionitrile, a powerful carcinogen in F344 rats.Carcinogenesis 5, 1137–1140.

Wu, X., Ranganathan, V., Weisman, D.S., Heine, W.F., Ciccone, D.N., O’Neill, T.B., Crick,K.E., Pierce, K.A., Lane, W.S., Rathbun, G., Livingston, D.M., Weaver, D.T., 2000.ATM phosphorylation of Nijmegen breakage syndrome protein is required in aDNA damage response. Nature 405, 477–482.

Zhou, B.B., Elledge, S.J., 2000. The DNA damage response: putting checkpoints inperspective. Nature 408, 433–439.