Bcl-xL confers multi-drug resistance in several squamouscell carcinoma cell lines

T. Noutomi a,b, H. Chiba a,b, M. Itoh a,b, H. Toyota b,c, J. Mizuguchi b,c,*aDepartment of Oral & Maxillofacial-Surgery, Tokyo Medical University Hospital, 6-7-1 Nishi-Shinjuku, Shinjuku-ku, Tokyo 160-0023, Japan

bDepartment of Immunology, Tokyo Medical University, 6-1-1 Shinjuku, Shinjuku-ku, Tokyo 160-0022, JapancIntractable Disease Research Center, Tokyo Medical University, Tokyo, Japan

Received 13 July 2000; accepted 4 September 2000

Abstract

Carboplatin (CBDCA) alone or in combination with irradiation and other chemotherapeutic agents has been used for the treat-

ment of oral squamous carcinoma. However, there are some limitations for such therapy because of inherent or acquired resistanceto CBDCA. To gain some insights into the association of CBDCA resistance with Bcl-2 family level or p53 status, we establishedeight carcinoma cell lines, consisting of two resistant (MIT8, MIT16), two sensitive (MIT6, MIT7), and four intermediate lines. Allof the five cell lines with p53 mutation belonged to the resistant � intermediate group, whereas two of three other lines with wild-

type p53 were in the sensitive group. Interestingly, both of the two resistant cell lines showed elevated levels of Bcl-xL, almostdouble that of sensitive line (MIL5), whereas either Bcl-2 or Bax-a level did not correlate with the CBDCA-resistance. To furtherverify the association between the Bcl-xL level and the drug resistance, two transformants (xL-3, xL-6) overexpressing Bcl-xL in the

CBDCA-sensitive cell line MIT7 were established using the gene transfer method. Both clones showed resistance to multiplechemotherapeutic agents, including CBDCA, actinomycin D, etoposide, and mitomycin C. Moreover, MIT8 and MT16 also dis-played cross-resistance to these agents. These findings suggest that Bcl-xL may function as one of the key components conferring

multiple drug-resistance in squamous cell carcinomas. # 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Squamous cell carcinoma; Apoptosis; Carboplatin; Bcl-xL; Bax-a

1. Introduction

Carboplatin (CBDCA), a less nephrotoxic analog ofcis-diamminedichloroplatinum (cisplatin, CDDP), incombination with other chemotherapeutic drugs hasbeen used for the treatment of oral squamous cell car-cinoma (SCC) as well as small cell carcinoma of thelung, and ovarian carcinoma [1–3]. The cytotoxic effectof CBDCA is considered to stem largely from its inter-action with DNA through the formation of a covalentadduct [4]. The CBDCA-induced DNA lesion results inapoptosis induction [5–7].Apoptosis is thought to play a crucial role on devel-

opment, homeostasis, and elimination of unfavorablecells such as infected or transformed cells [8]. Theapoptotic process can be divided into three distinct

phases: initiation, effector, and degradation [9]. Theinitiation phase depends on the activation of individu-ally distinct signaling pathways. These pathways areintegrated into a common effector phase, followed by afinal degradation stage resulting in apoptosis. Bcl-2 isone of the regulatory components acting during theeffector phase. The bcl-2 gene, originally isolated atthe chromosomal breakpoint of t(14;18) -bearing B -celllymphomas [10], is localized in mitochondrial mem-brane, endoplastic reticulum, and nuclear membrane.Bcl-2 is a member of the growing family of apoptosis-regulating gene products which include both apoptosis-accelerating (Bax-a, Bad) and apoptosis-inhibiting proteins(Bcl-2, Bcl-xL, Mcl) [11–14]. Overexpression of Bcl-2 orBcl-xL has been shown to prevent the apoptosis inductioninduced by various stimuli including chemotherapeuticagents, irradiation, and growth factor withdrawal[14–16], whereas that of Bax results in increased drugsensitivity in several cell lines [17,18].p53 is a transcription factor that causes growth arrest

as well as apoptosis in several cell types [19]. Various

1368-8375/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.

PI I : S1368-8375(00 )00098-1

Oral Oncology 38 (2002) 41–48

www.elsevier.com/locate/oraloncology

* Corresponding author. Tel.: +81-3-3351-6141; fax: +81-3-3341-

2941.

E-mail address: mizu@tokyo-med.ac.jp (J. Mizuguchi).

mutations in p53 gene are commonly found in humantumors. There are some arguments as to the correlationof p53 alterations with prognosis [20,21]. Moreover,some studies showed an association of p53 status withthe cisplatin-induced apoptosis [22,23], whereas othersdid not show this [24,25].One major obstacle for clinical effectiveness is that

tumor cells sometimes display inherent or acquiredresistance following treatment. Several distinct mechan-isms have been suggested for CBDCA resistanceincluding decreased drug accumulation, elevated glu-tathione level, and enhanced repair of damaged-DNA[26]. To examine the mechanisms of the CBDCA resis-tance we determined Bcl-2 family protein level andp53 status in eight SCC cell lines including CBDCA-sensitive and resistant lines and found that both of thetwo CBDCA-resistant lines contained elevated levelsof Bcl-xL.To further test the notion that the CBDCAresistance is associated with increased Bcl-xL levels,CBDCA sensitivity was examined in the transformantsoverexpressing Bcl-xL. Based on these findings, wemight conclude that Bcl-xL is a new member of compo-nents responsible for multiple drug resistance in SCC.

2. Materials and methods

2.1. Establishment of eight cell lines

The eight cell lines from tumor tissues obtained fromsix patients who had undergone surgical treatment forSCC at the Department of Oral and Maxillofacial-Surgery of Tokyo Medical University were establishedaccording to the method previously described [27].Clinical data from the patients on tumor location,metastasis, and TNM stage [28] are shown in Table 1.Cells were grown in monolayers in RPMI-1640 (com-plete RPMI-1640) containing 15% fetal bovine serum,50 mM 2 mercaptoethanol, 2 mM L-glutamine, and100 mg/ml kanamycin. At every passage, cells wereharvested as single cell suspensions using trypsin/EDTA (Life Technologies Inc., Grand Island, NY,USA).

2.2. Determination of cytotoxicity

The cells (5�103/well) in 96-well plates were culturedwith or without various concentrations of CBDCA(Bristol-Myers Squibb Company, Princeton, NJ, USA)and the number of cells were estimated by the procedureusing WST-8 assay kit (Dojindo Laboratories, Kuma-moto, Japan), which is based on colorimetric quantifi-cation of NADH and is used for cell viability assay [29].CBDCA-mediated cytotoxicity was assessed by the IC50

value, representing a concentration inhibiting growth by50% over a 2-day culture period, which was determinedby the formula of Chou and Talalay [30].

2.3. Assessment of apoptosis

Apoptosis was evaluated by flow cytometric analysisas previously described [7]. Briefly, the cells stimulatedunder the above conditions were fixed with 70% ethanol,stained with propidium iodide (PI, 36 mg/ml) (Sigma, St.Louis, MO, USA) and analyzed on a flow cytometer(FACSCalibur, Becton Dickinson Biosciences, San Jose,CA, USA) using CELL Quest software (Becton Dick-inson Biosciences). In some experiments, the cells werecultured with 270 mM CBDCA, 8 nM actinomycin D(CN Biosciences, Inc., San Diego, CA, USA) , 50 mMetoposide (Bristol-Myers Squibb Company), and 30mM mitomycin C (Kyowa Hakko Kogyo Co., Ltd,Tokyo, Japan). Following the incubation for 24 h (acti-nomycin D and mitomycin C) or 48 h (carboplatin andetoposide), the cell viability was determined by WST-8assay.

2.4. Flow cytometry

Indirect fluorescent staining was carried out as pre-viously described [31]. Briefly, the cells were stained withthe monoclonal antiobody (mAb) (W6/32, ATCC,Manassas, VA, USA) to major histocompatibilitycomplex (MHC) class I molecule, followed by incuba-tion with FITC-conjugated anti-mouse IgG Ab (CappelResearch Products, Durham, NC, USA). Flow cyto-metry was performed as described above.

Table 1

p53 mutations in eight squamous cell carcinoma (SCC) cell linesa

Patient No. (cell lines) Primary tumor Histology Metastasis TNM Exon Codon Mutation Amino–acid change

5 (MIL 5) Tongue MD LN T3N0Mx 7 245 GGC!AGC Gly!Ser

6 (MIT 6) Tongue WD – T2N1Mx Wild type

6 (MIL 6) Tongue WD LN 5 177–182 � � 18 amino acids

7 (MIT 7) Tongue WD – T3N2bMx Wild type

7 (MIL 7) Tongue WD LN Wild type

8 (MIT 8) Gingiva WD – T4N0Mx 6 213 CGA!TGA Arg!STOP

15 (MIT 15) Buccal mucosa PD – T3N2bMx 7 248 CGG!CAG Arg!Gln

16 (MIT 16) Floor of mouth PD – T3N0Mx 8 273 CGT!CAT Arg!His

a MD, Moderately differentiated SCC; LN, Lymph node; WD, Well differenciated SCC; PD, Poorly differenciated SCC.

42 T. Noutomi et al. / Oral Oncology 38 (2002) 41–48

2.5. Single-strand conformation polymorphism (SSCP)analysis

High-molecular-weight genomic DNA was extractedfrom tumor cells with the method of proteinase K diges-tion and phenol-chloroform purification [32]. SSCPanalysis was done according to a modified version of apreviously described method [33]. Polymerase chainreaction (PCR) was carried out using ExpandTM HighFidelity PCR System (Roche Diagnostics GmbH,Mannheim, Germany) in a final volume of 10 ml con-taining 100 ng genomic DNA and 20 p mol of each pri-mer, 200 mM dNTPs, 10 mM Tris–HCl (pH8.3), and 2.6units DNA polymerase with the following conditions: 35cycles of denaturation (94�C for 30 s), annealing(60�C for 60 s for reactions amplifying exons 5, 6, and 8;70�C for 60 s for exon 7), and extension (72�C for 60 s) inan automated thermocycler (Takara, Otsu, Japan). TheSSCP primes for p53 exons 5 through 8 were given asfollows: exon 5 (forward, 50-TTCCTCTTCCTACAGT-ACTCC-30; reverse, 50-GCCCCAGCTGCT CACCAT-CGC-30); exon 6 (forward, 50-CACTGATTGCTCTTA-GGTCTG-30; reverse, 50-AGTTGCAAACCAGACCT-CAGG-30); exon 7 (forward, 50-CCAAGG CGCACTG-GCCTCATC-30; reverse, 50-TCAGCGGCAAGCAGA-GGCTGG-30); exon 8 (forward, 50-CCTATCCTG-AG-TAGTGGTAAT-30; reverse, 50-GTCCTGCTT GC-TT-ACCTCGCT-30). Two microliters of PCR products werediluted 1:50 in a stop solution (95% formamide, 20 mMEDTA). Samples were denatured at 80�C for 5 min,cooled on ice, and immediately loaded onto a nonredu-cing 5% polyacrylamide gel containing 5% (v/v) gly-cerol. The gels were analyzed on an automated laserfluorescence DNA sequencer ALF express (AmershamPharmacia Biotech, Uppsala, Sweden) using Allele linksoftware (Amersham Pharmacia Biotech).

2.6. Direct DNA sequencing analysis for p53 mutation

Samples displaying an altered electrophoretic mobilityby the above SSCP analysis were selected for DNAsequencing. PCR was performed using each primer.The PCR products were purified by a High Pure PCRProduct Purification kit (Roche Diagnostics GmbH)and subjected to direct sequencing with ABI PRISMTM

310 Genetic Analyzer (PE Applied Biosystems, FosterCity, CA, USA) using ABI PRISM BigDyeTM Termi-nator Cycle Sequencing Ready Reaction kit (PEApplied Biosystems).

2.7. Reverse transcription (RT)-PCR and Bcl-xLexpression plasmid

The cDNA for Bcl-xL was cloned by RT-PCRmethod. Briefly, total RNA was isolated from Daudicells with Tri Reagent (Molecular Research Center,

Inc., Cincinatti, OH, USA) according to the manufac-turer’s instruction. RNA was reverse transcribed usingRNA PCR Kit (AMV) (Takara, Tokyo, Japan) forPCR with the supplied oligo dT-adaptor primer using athermal program of 42�C for 30 min, 99�C for 5 min,and 5�C for 5 min. PCR reactions were done with KODpolymerase (TOYOBO, Tokyo, Japan) using the fol-lowing primers: forward, 50-TCCGAATTCAATGTCT-CAGAG CAACCGGGA-30; reverse, 50-ATAGTTTA-GCGGCCGCTCATTTCC GACTGA AGAG-30, gen-erating fragments of 745 bp. The PCR profile was 30cycles at 98�C 15 s, 65�C 2 s, 74�C 30 s. The sequenceof the PCR products was confirmed to be homologousto the sequence reported previously [34]. The bcl-xLcDNA was subcloned at EcoRI and Not I into theexpression vector pMKIT Neo, which was generouslyprovided by Dr. K. Maruyama (Tokyo Medical andDental University, Tokyo).

2.8. Transfections and generations of stable MIT7 celllines overexpressing Bcl-xL

MIT7 cells were transfected with 5 mg of eitherpMKIT-Neo-HA-Bcl-xL or pMKIT-Neo-HA emptyvector (Neo) using Gene Porter (Gene Therapy Systems,Inc., San Diego, CA), following the manufacturer’sinstruction. Transfectants (3�105 cells/well) were cul-tured in six-well plates for 48 h and then selected in thecomplete RPMI-1640 medium with G418 (1000 mg/ml)(Life Technologies). Several individual clones wereobtained by limiting dilution. The level of HA-Bcl-xLprotein was determined byWestern blotting, as describedbelow.

2.9. Western blot analysis

Western blotting was done as previously described[27]. The mAbs or polyclonal Abs used here wereobtained from the following: rabbit anti-human Bcl-xLAbs (Transduction Laboratories, Lexington, KY,USA); anti-human Bcl-2 mAb and anti-Bax-a mAb(Medical and Biological Laboratories, Nagoya, Japan);rabbit anti-actin Ab (Sigma); horse-radish-peroxidase(HRP)-labeled goat anti-mouse IgG Fc Ab (CappelResearch Products); HRP-labeled goat anti-rabbit IgGFc Ab (EY Laboratories Inc., San Mateo, CA, USA).The density of each band was measured using a densit-ometer (Image Master DTS, Amersham PharmaciaBiotech).

2.10. Statistical analysis

Data were expressed as means�S.D. for each group.Statistical significance was determined by Student’s t-test, and a difference of P<0.05 was considered toindicate statistical significance.

T. Noutomi et al. / Oral Oncology 38 (2002) 41–48 43

3. Results

3.1. Sensitivity to CBDCA among eight cell linesderived from patients with SCC

Eight cell lines were established from six patients withprimary and/or metastatic tumors: five lines werederived from primary and three from metastases (Table1). The cells (5�103/well) from eight cell lines were cul-tured in 96-well plates with various concentrations ofCBDCA for 48 h and cell survival was assessed byWST-8 assay. The IC50 value was determined for eachcell line. The eight cell lines were arbitrarily classifiedinto three groups (sensitive, intermediate, and resistant),based on the sensitivity to CBDCA exposure. As shownin Fig. 1, the CBDCA-sensitive group with IC50 value

130 mM and 99 mM included MIT6 and MIL7, whileboth MIT8 and MIT16 showed considerable levels ofresistance (resistant group) with IC50 values of 594 mMand 447 mM, respectively. The other four cell lines(MIL5, MIL6, MIL7, and MIT15) displayed an inter-mediate level of resistance ranging from 164 mM to199 mM (intermediate group).

3.2. Induction of apoptosis to CBDCA in sensitive celllines, but not resistant cell lines

The cell growth is thought to be regulated by the bal-ance among proliferation, survival, and apoptosis.Induction of an apoptosis, evaluated by PI stainingmethod, was investigated in both CBDCA-sensitive and-resistant cell lines following CBDCA stimulation. Aconsiderable proportion of CBDCA-stimulated cellsfrom the sensitive group (MIT6, MIL7) underwentapoptosis (67, 69%, respectively vs. 5% with mediumalone) at 48 h, while only a minor portion of the cellsfrom the resistant group (MIT8, MIT16) resulted inapoptosis (20, 38%, respectively) (Fig. 2). The CBDCA-stimulated cells from the intermediate group displayedapoptosis between those values. These results suggestthat the apoptosis appears to represent the CBDCA-mediated cytotoxicity in SCC cells.

3.3. p53 mutations in SCC cell lines

It has been suggested that p53 mutation is associatedwith development and progression of tumor [21]. PCR-SSCP was done on the eight SCC cell lines. All tumorsrevealing aberrant migrating SSCP bands had p53mutations, as determined by direct sequencing analysis.Mutations and/or deletion were found in five cell lines,consisting of four lines derived from primary lesionsand one from a metastatic lesion, among the eight lines

Fig. 2. Carboplatin (CBDCA) stimulation induces a greater proportion of apoptosis against CBDCA-sensitive lines (MIT6, MIT7) than in resistant

lines (MIT8, MIT16). The cells cultured with or without 270 mM CBDCA for 48 h were assessed for apoptosis induction by the PI staining method.

Fig. 1. Carboplatin (CBDCA)-mediated cytotoxicity against eight

SCC cell lines. Cells (5�103/well) from eight cell lines were cultured

with various concentrations of CBDCA for 48 h and cell viability was

determined by WST-8 assay as described in Section 2. Percent viability

was expressed as ‘‘% of control’’ by the formula: (% viability with

CBDCA/% viability with medium alone) �100.

44 T. Noutomi et al. / Oral Oncology 38 (2002) 41–48

(Table 1). In one case No. 7, no mutation was found incell lines established from either the primary (MIT7) orits metastastatic lesion (MIL7). MIT6 from a primarytumor contained wt p53, whereas MIL6 from a meta-static lymph node from the same patient had 18 bpdeletion ranging from codon 177 to 182. Of the five celllines with p53 mutation or deletion, two lines (MIT8 andMIT16) were classified in the resistant group and three(MIL5, MIL6, and MIT15) were in the intermediategroup. Two of three lines with wt p53 were in the sen-sitive group and one was in the intermediate group.These findings suggest that p53 mutations per se are notapparently associated with the CBDCA resistance.

3.4. Bcl-xL level is elevated in CBDCA resistant celllines MIT8 and MIT16

Induction of apoptosis is regulated at least in part bythe balance between the apoptosis-accelerating products(Bax and Bad) and apoptosis-inhibiting ones (Bcl-2 andBcl-xL) [11,13,14]. The levels of Bcl-2 family proteinsfrom the eight cell lines were determined by Westernblotting. Both of CBDCA-resistant cell lines MIT8 andMIT16 showed an increased level of Bcl-xL (1.9, 2.1),whereas the sensitive lines displayed low-to-intermediatelevels of Bcl-xL (1.0–1.6) (Fig. 3). Although the level ofboth anti-apoptotic protein Bcl-2 and pro-apoptoticprotein Bax-a considerably varied among cell lines, thereappeared to be no relationship between the protein leveland the drug sensitivity. These findings indicate that the

anti-apoptotic molecule Bcl-xL has certain associationwith CBDCA-resistance in SCC cell lines.

3.5. Overexpression of Bcl-xL confers the resistance tomultiple drugs in MIT7 cells

To further examine the participation of Bcl-xL in drug-resistance, the CBDCA-sensitive cell line MIT7 over-expressing Bcl-xL was established. MIT7 cells weretransfected with a HA-Bcl-xL expression vector pMKIT-Neo-Bcl-xL or control Neo alone and then selected withG418. Multiple independent clones expressing a highlevel of transgene-encoded HA-Bcl-xL (p32-Bcl-xL) inaddition to endogenous p31-Bcl-xL were isolated byWestern blotting using anti-Bcl-xL Ab. Two representa-tive clones (xL-3, xL-6) with levels of MHC class I mole-cule nearly identical to that of parent MIT7 cells wereselected for further study (Fig. 4). Controls consisted ofMIT7 cells transfected with control vector alone (Neo-1).We examined the sensitivity of the clones xL-3 and xL-

6 as well as the CBDCA-resistant cell lines MIT8 andMIT16 to chemotherapeutic drugs by WST-8 assay.

Fig. 3. Bcl-xL level is elevated in Carboplatin-resistant lines MIT8 and

MIT16. The level of Bcl-2 family proteins Bcl-xL, Bcl-2, and Bax-a in

SCC cell lines was determined by Western blotting. The level of each

protein was determined by densitometry and normalized by actin

levels. The relative protein level in each cell line was expressed in

relation to that in MIL5.

Fig. 4. Establishment of the cell lines overexpressing Bcl-xL through

the gene transfer method. MIT7 cells were transfected with pMKIT-

Neo-Bcl-xL or control Neo. Transfectants were selected with complete

RPMI-1640 medium containing G418, followed by limiting dilution to

obtain individual clones. (a) The level of Bcl-xL protein from indivi-

dual clones, as determined by Western blotting. (b) Membrane density

of major histocompatibility complex class I molecule, as determined

by flow cytometry using anti-class I mAb.

T. Noutomi et al. / Oral Oncology 38 (2002) 41–48 45

Both the MIT7 and Neo-1 cells were sensitive to themultiple drugs including CBDCA, etoposide, mitomy-cin C, and actinomycin D, whereas the resistant celllines MIT8 and MIT16 showed cross-resistance to theother drugs (Fig. 5). Interestingly, the cell lines xL-3 andxL-6 also displayed the multiple drug resistance. Takentogether, these results suggest Bcl-xL might contributeto the resistance to multiple drugs such as CBDCA,actinomycin D, etoposide, and mitomycin C in SCC celllines.

4. Discussion

Although CBDCA has been widely used for thetreatment of SCC as well as other tumor types [1–3],CBDCA treatment is hampered by the appearance ofdrug-resistant cells during treatment. Several mechan-isms for the CBDCA resistance have been proposed:first, tumor cells contain elevated level of mdr-1 geneproduct, P-glycoprotein, which is a transporter respon-sible for drug efflux; second, the drug is inactivated byglutathione; third, DNA repair is enhanced [26]. Bcl-2family proteins have recently proved to play a crucialrole in apoptosis induction [11,14]. In the present studywe examined whether Bcl-2 family gene products and/or p53 status are implicated in the resistance to chemo-therapeutic drugs including CBDCA, etoposide, actino-mycin-D, and mitomycin C in SCC cell lines.

Among eight cell lines derived from patients withSCC, five cell lines were from primary tumors and threewere from metastatic regional lymph nodes. Approxi-mately 70% of the CBDCA-sensitive cells becameapoptotic upon stimulation with CBDCA, whereas theproportion in the resistant cell lines (MIT8, MIT16) wasless than 30%, suggesting that CBDCA-mediated cyto-toxicity is related to the induction of apoptosis [7,26].Bcl-2 family proteins Bcl-2/Bcl-xL act as apoptosis

inhibitor [11,13,35]. Both cell lines which showedresistance to CBDCA stimulation, as assessed by cel-lular cytotoxicity using WST-8 assay, had elevatedlevels of Bcl-xL compared with CBDCA-sensitive celllines (MIT6 and MIT7). The MIT7 cells overexpressingBcl-xL through gene-mediated transfer displayed multi-drug resistance, further suggesting that Bcl-xL pro-vides drug resistance. In agreement with our observa-tion, Bcl-2 family proteins Bcl-2 and Bcl-xL conferedresistance to several chemotherapeutic agents, includingcisplatin, in several cell lines [15,16,36,37]. However, theBcl-2 level did not correlate with the CBDCA-mediatedcytotoxicity in the eight cell lines we tested. Theseobservations might reflect the notion that regulation ofBcl-xL expression is different from that of Bcl-2 in vari-ous tumors, and may depend on stage of differentiationand cell lineage.Bax-a favors apoptosis induction, unlike Bcl-2/Bcl-xL

[11,13,14]. The cellswith overexpression ofBax-a throughgenetic manipulation become susceptible to apoptosis

Fig. 5. Both Carboplatin (CBDCA)-resistant lines and Bcl-xL-overexpressing clones display multidrug resistance. The cells were cultured with 270

mM CBDCA (a), 8 nM actinomycin D (b), 50 mM etoposide (c), and 30 mM mitomycin C (d). After incubation for 24 h (b & d) or 48 h (a & c), the

cell viability was determined by WST-8 assay as described in Section 2. The viability of both the resistant lines and the xL clones in response to

various drugs significantly differ from that in control (MIT7 & Neo).

46 T. Noutomi et al. / Oral Oncology 38 (2002) 41–48

induction in response to chemotherapeutic drugs [17,18]. However, both of the cell lines with considerablyelevated level of Bax-a were not sensitive to CBDCAexposure (Figs. 1 and 3), suggesting that Bax-a level perse is not associated with the drug sensitivity in SCClines. Moreover, the ratio of Bax-a to Bcl-2/Bcl-xL isnot critical for drug sensitivity.The role of p53 mutation in CBDCA-mediated apop-

tosis is controversial [22–25]. Almost half of tumorscontain a variety of mutations in the p53 gene [38],which causes growth arrest as well as apoptosis . Thegenetic alterations obtained in this study were inthe region with frequent mutation (amino acids 106–292), corresponding to DNA binding domain of p53[38]. Two out of three wt p53 cell lines were in theCBDCA-sensitive group and the other (MIL7) was inthe intermediate group. The other five cell lines withmutant p53 were classified into intermediate or resistantgroups, suggesting that p53 status per se has no rela-tionship with the CBDCA-resistance in SCC cell lines.However, it should be noted that both the two cell lineswith enhanced levels of Bcl-xL had p53 mutations: onemutation was located in codon 273 of exon 8, which isfrequently mutated in various tumors [38–40]. It could,therefore, be possible that elevated Bcl-xL somehowcombines with mutated p53, resulting in drug resistance.Anti-sense oligonucleotides — stretches of single-

stranded DNA that are complementary to selectedmRNA (sense strand) — are capable of preventing geneexpression and hold promise as tools in modulating geneexpression in a variety of conditions, especially tumors[41]. Indeed, Bcl-2 anti-sense oligonucleotide therapyfor lymphoma with elevated Bcl-2 level is in a phase Iclinical study 1999 [42]. A similar approach might beeffective for the treatment of SCC with elevated Bcl-xL.Bcl-xL appears to be one of the critical components

implicated in multiple drug resistance, based on the fol-lowing observations: (1) two cell lines from CBDCAresistant group had enhanced level of Bcl-xL comparedwith the sensitive group; and (2) CBDCA-sensitive lineMIT7 was genetically manipulated to become resistantto multiple drugs through Bcl-xL gene transfer. Thesefindings provide some insights for the understanding ofthe mechanisms of CBDCA resistance and would prob-ably be valuable in the treatment of SCC.

Acknowledgements

This work was supported by Grant-in-Aid for Scien-tific Research (C, No. 12671838) and the IntractableDisease Research Center of Tokyo Medical University.We thank Dr. K. Maruyama (Tokyo Medical andDental University, Tokyo) for the pMKIT Neo expres-sion vector and Prof. J. Patrick Barron (Tokyo MedicalUniversity) for reading the manuscript.

References

[1] Niitani H, Kobayashi K. Cisplatin/Carboplatin therapy in non-

small cell lung cancer. Oncology 1992;49(Suppl 1):51–6.

[2] Stiff PJ, McKenzie RS, Alberts DS, Sosman JA, Dolan JR, Rad

N, McCloskey T. Phase I clinical and pharmacokinetic study of

high-dose mitoxantrone combined with carboplatin, cyclopho-

sphamide, and autologous bone marrow rescue: high response

rate for refractory ovarian carcinoma. Journal of Clinical Oncol-

ogy 1994;12:176–83.

[3] Zamboglou N, Schnabel T, Kolotas C, Achterrath W, Strehl H,

Dalhauser S, Vogt HG, Lenaz L, Schmitt G. Carboplatin and

radiotherapy in the treatment of head and neck cancer: six years’

experience. Seminars in Oncology 1994;21:45–53.

[4] Zamble DB, Lippard SJ. Cisplatin and DNA repair in cancer

chemotherapy. Trends in Biochemical Siences 1995;20:435–9.

[5] Barry MA, Behnke CA, Eastman A. Activation of program-

med cell death (apoptosis) by cisplatin, other anticancer drugs,

toxins and hyperthemia. Biochemical Pharmacology 1990;40:

2353–62.

[6] Felice VD, Lauricella M, Giuliano M, Emanuele S, Vento R,

Tesoriere G. The apoptotic effects of cisplatin and carboplatin in

retinoblastoma Y79 cells. International Journal of Oncology

1998;13:225–32.

[7] Yanase N, Takada E, Yoshihama I, Ikegami H, Mizuguchi J.

Participation of Bax-a in IFN-amediated apoptosis in Daudi B

lymphoma cells. Journal of Interferon and Cytokine Research

1998;18:855–61.

[8] Raff MC. Social controls on cell survival and cell death. Nature

1992;356:397–400.

[9] Thompson CB. Apoptosis in the pathogenesis and treatment of

disease. Science 1995;267:1456–62.

[10] Tsujimoto Y, Gorham J, Cossman J, Jaffe E, Croce CM. The

t(14;18) chromosome translocations involved in B-cell neoplasms

result from mistakes in VDJ joining. Science 1985;229:1390–3.

[11] Kroemer G. The proto-oncogene Bcl-2 and its role in regulating

apoptosis. Nature Medicine 1997;3:614–20.

[12] Adams JM, Cory S. The Bcl-2 protein family: arbiters of cell

survival. Science 1998;281:1322–6.

[13] Chao DT, Korsmeyer SJ. Bcl-2 family: regulators of cell death.

Annual Review of Immunology 1998;16:395–419.

[14] Reed JC. Bcl-2 family proteins. Oncology 1998;17:3225–36.

[15] Miyashita T, Reed JC. bcl-2 gene transfer increases relative

resistance of S49.1 and WEHI7.2 lymphoid cells to cell death and

DNA fragmentation induced by glucocorticoid and multiple

chemotherapeutic drugs. Cancer Research 1992;52:5407–11.

[16] Simonian PL, Grillot DAM, Nunez G. Bcl-2 and Bcl-xL can dif-

ferentially block chemotherapy-induced cell death. Blood

1997;90:1208–16.

[17] Wagener C, Bargou RC, Daniel PT, Bommert K, Mapara MY,

Royer HD, Dorken B. Induction of the death-promoting gene

bax-asensitizes cultured breast-cancer cells to drug-induced

apoptosis. International Journal of Cancer 1996;67:138–41.

[18] Guo B, Cao S, Toth K, Azrak RG, Rustum YM. Overexpression

of Bax enhances antitumor activity of chemotherapeutic agents in

human head and neck squamous cell carcinoma. Clinical Cancer

Research 2000;718-724:718–24.

[19] Levine AJ. p53, the cellular gate keeper for growth and division.

Cell 1997;88:323–31.

[20] Nylander K, Stenling R, Gustafsson H, Zackrisson B, Roos G.

p53 expression and cell proliferation in squamous cell carcinomas

of head and neck. Cancer 1995;75:87–93.

[21] Hirvikoski P, Kumpulaninen E, Virtaniemi J, Johansson R,

Haapasalo H, Marin S, Halonen P, Helin H, Raitiola H, Pukan-

der J, Kellokumpu-Lehtinen P, Kosma V-M. p53 expression and

cell proliferation as prognostic factors in laryngeal squamous cell

carconoma. Journal of Clinical Oncology 1997;15:3111–20.

T. Noutomi et al. / Oral Oncology 38 (2002) 41–48 47

[22] Gallagher WM, Cairnney M, Schott B, Roninson IB, Brown R.

Identification of p53 genetic suppressor elements which confer

resistance to cisplatin. Oncogene 1997;14:185–93.

[23] Song K, Cowan KH, Sinha BK. In vivo studies of adenovirus-

mediated p53 gene therapy for cis-platinum-resistant human

ovarian tumor xenografts. Oncology Research 1999;11:153–9.

[24] Minagawa Y, Kigawa J, Itamochi H, Kanamori Y, Shimada M,

Takahashi M, Terakawa N. Cisplatin-resistant HeLa cells are

resistant to apoptosis via p53-dependent and -independent path-

ways. Japanese Journal of Cancer Research 1999;90:1373–9.

[25] Burger H, Nooter K, Boersma AW, Kortland CJ, Stoter G. Lack

of correlation between cisplatin-induced apoptosis, p53 status

and expression of Bcl-2 family proteins in testicular germ cell

tumor cell lines. International Journal of Cancer 1997;73:592–9.

[26] Eastman A. Mechanisms of resistance to cisplatin. Kluwer Aca-

demic Publishers: Boston, 1991.

[27] Itoh M, Chiba H, Noutomi T, Takada E, Mizuguchi J. Cleavage

of Bax-a and Bcl-xL during carboplatin-mediated apoptosis in

squamous cell carcinoma cell line. Oral Oncology 2000;36:277–85.

[28] Hermanek P, Hunter RVP, Sobin LH, Wagner G, Wittekind C.

TNM Atals. 4th ed. Springer, 1997.

[29] Ishiyama M, Miyazono Y, Sasamoto K, Ohkura Y, Ueno K. A

highly water-soluble disulfonated tetrazolium salt as a chromo-

genic indicator for NADH as well as cell viability. Talanta

1997;44:1299–305.

[30] Chou T-C, Talalay P. Quantitative analysis of dose-effect rela-

tionships: the combined effects of multiple drugs or enzyme inhi-

bitors. Advances in Enzyme Regulation 1984;22:27–55.

[31] Mizuguchi J, Yamanashi Y, Ehara K, Tamura T, Nariuchi H,

Gyotoku Y, Fukazawa H, Uehara Y, Yamamoto T. Tyrosine

protein kinase is involved in anti-IgM-mediated signaling in

BAL17 B lymphoma cells. The Journal of Immunology 1992;148:

689–94.

[32] Sato T, Tanigami A, Yamakawa K, Akiyama F, Kasumi F, Saka-

moto G, Nakamura Y. Allelotype of breast cancer: cumulative

allele losses promote tumor progression in primary breast cancer.

Cancer Recearch 1990;50:7184–9.

[33] Orita M, Suzuki Y, Sekiya T, Hayashi K. Rapid and sensitive

detection of point mutations and DNA polymorphisms using the

polymerase chain reaction. Genomics 1989;5:874–9.

[34] Boise LH, Gonzalez-Garcia M, Postema CE, Mao X, Nunez G,

Thompson CB. bcl-x, a bcl-2-related gene that functions as a

dominant regulator of apoptotic cell death. Cell 1993;74:597–608.

[35] Reed JC. Mechanism of apoptosis avoidance in cancer. Current

Opinion in Oncology 1999;11:68–75.

[36] Kojima H, Endo K, Moriyama H, Tanaka Y, Alnemri ES, Sla-

pak CA, Teicher B, Kufe D, Datta R. Abrogation of mitochon-

drial cytochrome c release and caspase-3 activation in acquired

multidrug resistance. The Journal of Biological Chemistry 1998;

273:16647–50.

[37] Decaudin D, Geley S, Hirsch T, Castedo M, Marchetti P,

Macho A, Kofler R, Kroemer G. Bcl-2 and Bcl-XL antagonize

the mitochondrial dysfunction preceeding nuclear apoptosis

induced by chemotherapeutic agents. Cancer Research 1997;57:

62–7.

[38] Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 muta-

tions in human cancers. Science 1991;253:49–53.

[39] Deimling Av, Eibl RH, Ohgaki H, Louis DN, Ammon Kv,

Petersen I, Kleihues P, Chung RY, Wiestler OD, Seizinger

BR. p53 mutations are associated with 17p allelic loss in grade

II and grade III astrocytoma. Cancer Research 1992;52:

2987–90.

[40] Oda T, Tsuda H, Scarpa A, Sakamoto M, Hirohashi S. Mutation

pattern of the p53 gene as a diagnostic marker for multiple

hepatocellular carcinoma. Cancer Research 1992;52:3674–8.

[41] Stein CA, Cheng YC. Antisense oligonucleotides as therapeutic

agents — Is the bullet really magical? Science 1993;261:1004–12.

[42] Cotter FE, Waters J, Cunningham D. Human Bcl-2 antisense

therapy for lymphomas. Biochemica et Biophysica ACTA

1999;1489:97–106.

48 T. Noutomi et al. / Oral Oncology 38 (2002) 41–48