ORIGINAL ARTICLE

Overexpression of cMOAT (MRP2/ABCC2) Is Associated withDecreased Formation of Platinum-DNA Adducts and DecreasedG2-Arrest in Melanoma Cells Resistant to Cisplatin

Bernd Liedert,nVerena Materna, Dirk Schadendorf,w Jˇrgen Thomale,n and Hermann Lage1Institute of Pathology, ChariteŁ Campus Mitte, Humboldt University Berlin, Berlin, Germany; n1Institute of Cell Biology, University of Essen, Germany,wSkin Cancer Unit, German Cancer Research Center (DKFZ), and University Heidelberg, Clinic of Dermatology Mannheim, Mannheim, Germany

Resistance to various anti-neoplastic agents is a com-mon observation in clinical management of melanoma.The biologic mechanisms conferring these di¡erentdrug-resistant phenotypes, including resistance againstthe commonly used anti-cancer drug cisplatin, areunclear. In order to elucidate the role of the mem-brane adenosine triphosphate binding cassette-transpor-ter cMOAT (canalicular multispeci¢c anion transporter)(MRP2/ABCC2) in cisplatin resistance of melanoma,the expression of this protein was analyzed in the plati-num drug-resistant cell line MeWo CIS 1. Cisplatin-resistant melanoma cells showed a distinct over-expression of cMOATon mRNA and protein level. Thisobservation was accompanied by a reduced formationof platinum-induced intrastrand cross-links in the nu-clear DNA measured by an immunocytologic assay.

This decrease in DNA platination was accompanied byan accelerated re-entry into the cell cycle after the typi-cal cisplatin-induced G2 arrest, and a resistance to un-dergo apoptosis. Kinetics of formation and eliminationof platinum-DNA adducts suggest that the DNA repaircapacity for Pt-d(GpG) adducts was not elevated in pla-tinum drug-resistant melanoma cells. The decrease inplatinum-DNA adduct formation in cisplatin-resistantmelanoma cells was rather a re£ection of the protectingactivity of the transporter cMOAT. In conclusion, thefunctional inhibition of cMOAT might be a promisingstrategy in the reversal of resistance to platinum-basedanti-cancer drugs in human melanoma. Key words:ABCC2/cisplatin/cMOAT/drug resistance/melanoma/MeWo/MRP2. J Invest Dermatol 121:172 ^176, 2003

Melanoma showed rapidly increasing incidenceand mortality rates, and is characterized by ahigh risk for metastasis. A successful curativetreatment is possible only by an opportune, ade-quate surgical resection. The bene¢t of both ad-

juvant and palliative systemic chemotherapy is not unequivocallyestablished. Therapeutic regimens based on dacarbazine are themost commonly applied strategies for the treatment of dissemi-nated melanoma; however, only a minority of patients obtainlong-lasting response rates. Various response rates have been re-ported for polychemotherapy treatment protocols, frequently con-taining cisplatin [cis-dichlorodiamine-platinum (II)]. Althoughcisplatin is one of the most active and widely applied anti-neoplastic agents, partial response in melanoma is observedonly for about 10% of the patients. Chemotherapeutic strategiesare extremely ine¡ective and unsatisfactory for treating melanomadue to the pronounced drug-resistant characteristics of thismalignancy. Data from in vitro studies suggest the presence of in-trinsic biologic mechanisms conferring this high drug resistance(Schadendorf et al, 1994), whereby the precise cellular mechanisms

have not been identi¢ed (Helmbach et al, 2001). Despite the un-satisfactory response rates, anti-cancer drug-based chemotherapywill continue to be the ¢rst line treatment of disseminated mela-noma, although there is an urgent need for improvement. Onestrategy to achieve this, is to identify the molecular mechanismsconferring drug resistance in melanoma, and, based on thisknowledge, to develop new treatment modalities to improve re-sponse rates in melanoma patients.In order to investigate drug resistance in human melanoma, we

have previously established and analyzed a panel of various mela-noma-derived cell variants exhibiting di¡erent drug resistancepro¢les (Kern et al, 1997; Lage et al, 1999, 2000, 2001; Nesslinget al, 1999; Grottke et al, 2000; Rˇnger et al, 2000; Sinha et al,2000; Christmann et al, 2001; Helmbach et al, 2002). One of thesecell models, MeWo CIS 1, exhibiting a cisplatin-resistant pheno-type, was established by continuous exposure of the melanomacell line MeWo to cisplatin.Although only about 1% of the intracellular cisplatin reacts

with nuclear DNA, this is presumably the critical event in cispla-tin-mediated cytotoxicity (Gonzalez et al, 2001). The two majoradducts are Pt-d(GpG) and Pt-d(ApG) intrastrand cross-links re-presenting about 90% of the DNA platination. The other lesionsinclude Pt-(dG) monoadducts, Pt-d(GpNpG) intrastrand cross-links, and Pt(dG)2 interstrand cross-links (Fichtinger-Schepmanet al, 1985). The formation of these platinium-DNA adducts per semay not be su⁄cient to cause cell death, whereby the exact cas-cade of the downstream events leading to cell death is not yetclear. It is generally accepted, however, that the formation of1Both institutes contributed equally to this study.

Address correspondence and reprint requests to: PD Dr H. Lage, Insti-tute of Pathology, ChariteŁ Campus Mitte, Humboldt University Berlin,Schumannstr. 20/21, D-10117 Berlin, Germany. Email: hermann.lage@charite.de

Manuscript received November 14, 2002; revised February 11, 2003;accepted for publication February 27, 2003

0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc.

172

platinum-DNA adducts and subsequent triggering of cellular sig-nal transduction pathways leading to apoptosis may be the pri-mary cytotoxic mechanism of cisplatin (Fisher, 1994).In vitro studies demonstrated that cellular mechanisms causing

resistance to cisplatin can be associated with detoxi¢cation bymethallothioneins (Kelley et al, 1988), increased glutathione (g-glutamylcysteinyl glycine) level as well as increased activity ofglutathione-S-transferases (Zhang et al, 1998), increased DNA re-pair (Masuda et al, 1988), decreased activity of the DNA mismatchrepair system (Lage and Dietel, 1999), reduced apoptotic response(Perego et al, 1996), or enhanced expression of the ABC (adeno-sine triphosphate binding cassette)-transporter cMOAT (canalicu-lar multispeci¢c anion transporter) (Taniguchi et al, 1996), alsodesignated as MRP2 orABCC2. All these studies, however, wereperformed using carcinoma cells derived from various nonecto-dermal-derived cells, but not melanoma-derived cells.

MATERIALS AND METHODS

Cell culture Establishment and cell culture conditions of the cisplatin-resistant subline MeWo CIS 1, derived from the drug-sensitive humanmelanoma cell line MeWo (Fogh et al, 1978), was described earlier (Kernet al, 1997; Lage et al, 2001).

Cell proliferation assay Chemoresistance was tested using aproliferation assay as described previously (Lage et al, 2001).

Northern blot Expression of the cMOAT encoding mRNA wasdetermined by northern blot analysis applying standard proceduresas described previously (Lage et al, 2001). A 453 bp cMOAT-speci¢c hybridization probe was generated by reverse transcription^polymerasechain reaction (reverse transcription^PCR) using cMOAT-speci¢c oligo-nucleotide primers designed after alignment with published sequences(GenBank accession number NM 000392): cMOAT-forward, 50 -GGAACA ATT GTA GAG AAA GGA TC-30; and cMOAT-reverse, 50 -CACAAA CGC AAG GAT GAT GAA GAA-30. As control for equal RNAloading the membranes were rehybridized with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe, prepared as describedin the next subsection (‘Quantitative reverse transcription^PCR’).

Quantitative reverse transcription^PCR For quantitative reversetranscription^PCR, a real-time reverse transcription^PCR was carriedout using a LightCycler instrument (Roche Diagnostics, Mannheim,Germany). Oligodeoxynucleotide primers used for ampli¢cation werecMOAT-forward and cMOAT-reverse (see above) and as controlGAPDH-forward, 50 -CAC CGT CAA GGC TGA GAA C-30; andGAPDH-reverse, 50 -ACC ACT GAC ACG TTG GCA G-30 yielding anexpected ampli¢cation product of 556 bp speci¢c for the house keepinggene encoding GAPDH (GenBank accession number NM 002046).

Western blot Western blot experiments were performed as describedpreviously (Lage and Dietel, 2002). The antibodies used were a mousemonoclonal antibody (MoAb) directed against cMOAT (clone M2I-4;Sanbio, Uden, the Netherlands), and as control for equivalent proteinloading, a mouse MoAb directed against actin (Chemicon, Temecula,California; no. MAB 1501R). Protein^antibody complexes were visualizedby chemoluminescence (ECL system, Amersham Aylesbury, UK).

Cell cycle analyses by £ow cytometry For cell-cycle analysis inexponentially growing cell cultures, melanoma cells were exposed to 5 mgcisplatin per mL. After 24 h, 48 h, or 72 h cells were ¢xed in ice-cold 75%ethanol and stored at ^201C. After washing cells were resuspended in PBScontaining 200 mg DNase-free RNase A per mL, 50 units per mL RNAseT1 and 20 mg propidium iodide per mL and incubated for 30 min at 251C.Analysis was performed on a FACScan £ow cytometer (Calibur 750;Becton Dickinson, MountainView, California) using Cellquest software.

Detection of platinum-DNA adducts Cisplatin-induced Pt-d(GpG)intrastrand cross-links in the nuclear DNA of single cells were visualizedand measured by an immunocytologic assay using adduct-speci¢c MoAb asdescribed previously (Buschfort-Papewalis et al, 2002).

RESULTS AND DISCUSSION

Human melanoma is well known for its high resistance to manyanti-neoplastic agents used for therapy. In this study, the cellularmechanisms conferring resistance to the commonly used anti-cancer drug cisplatin were analyzed in the cisplatin-resistant mel-anoma cell line MeWo CIS 1. Previous studies have demonstratedthat this cisplatin-resistant phenotype was accompanied by: (1) ade¢cient DNA mismatch repair pathway (Lage et al, 1999); (2) amodulation of DNA repair activities (Rˇnger et al, 2000); and(3) an apoptosis de¢ciency (Helmbach et al, 2002).The fundamen-tal biologic principle causing the cisplatin-resistant phenotype inMeWo CIS 1, however, remained unclear.Earlier studies demonstrated that a cisplatin-resistant pheno-

type could be associated with the overexpression of the ABC-transporter cMOAT. In these studies carcinoma-derived cells withacquired cisplatin resistance (Taniguchi et al, 1996; Kool et al, 1997)or kidney cells transfected with a cMOAT encoding cDNA (Cui

Figure1. (A) Northern blot analysis of cMOAT mRNA expressionin human melanoma cells. RNA was hybridized with a 32P-labeledcMOAT-speci¢c cDNA. As control, the blots were probed with a cDNAprobe of GAPDH. (B) Relative cMOATmRNA expression level after nor-malization against GAPDH expression level determined by quantitativereal time reverse transcription^PCR using a LightCycler instrument. Cis-platin-resistant MeWo CIS 1 cells showed a 3.7-fold enhanced expression ofthe cMOATmRNA. (C) Western blot analysis using a mouse MoAb M2I-4directed against cMOAT. As control, a mouse MoAb directed against actinwas used.

cMOAT IN MELANOMA CELLS 173VOL. 121, NO. 1 JULY 2003

et al, 1999; Kawabe et al, 1999) were used. The conclusiveness ofthese studies, however, was hampered by the lack of functionalassays. Thus, here the involvement of cMOAT in platinum resis-tance of human melanoma was investigated.Expression analyses revealed that cMOAT is overexpressed in

the cisplatin-resistant melanoma cell MeWo CIS 1 on the mRNAlevel (Fig 1A,B) as well as on the protein level (Fig 1C). No ex-pression of alternative drug resistance-associated ABC-transpor-ters, such as MDR1/P-glycoprotein, MRP1, or BCRP wasfound in MeWo CIS 1 (data not shown); neither on mRNA levelas measured by reverse transcription^PCR and northern blot, noron protein level as determined by western blot. For examinationwhether the cross-resistant phenotype of MeWo CIS 1 is in accor-dance with the typical cMOAT-speci¢c cross-resistance pattern(Cui et al, 1999), the cytotoxicities of di¡erent anti-cancer drugswere determined (Table I). As expected for a cMOAT-mediatedcross-resistance, the cisplatin-resistant cell variant exhibited ahigh cross-resistance to carboplatin, whereas no or only a weakcross-resistance could be observed against the remaining drugs.Flow cytometry-based cell cycle analysis revealed a typical cis-

platin-triggered G2 arrest (Chu, 1994) in drug-sensitive melanomacells as well as in cisplatin-resistant melanoma cells (Fig 2). Aftertreatment with a low cisplatin concentration (5 mg per mL) bothcell variants were able to re-enter the cell cycle, but with adistinct delay in MeWo cells (Fig 2A,B). Exposure to a higherdose of cisplatin (20 mg per mL) resulted in a complete G2 arrestin MeWo cells even after 72 h with clear signs of apoptosis (Fig2C), whereas the resistant MeWo CIS 1 cells were able to re-enterthe cell cycle (Fig 2D). The changes in cell cycle distribution of

MeWo CIS 1 cells after high dosage cisplatin (20 mg per mL) werecomparable with the e¡ect observed in the parental MeWo cellsexposed a 4-fold lower dose of the drug, which is in completeconcordance with the relative resistance factor of 3.7 (Table I).To elucidate whether this decreased inclination to undergo cis-

platin-induced cell cycle arrest and cell death is caused by a de-creased platinum-DNA adduct formation, by more e⁄cientadduct repair, or by augmented tolerance to DNA damage, thekinetics of formation and elimination of nuclear Pt-d(GpG) in-trastrand cross-links, the major product of cisplatin-inducedDNA adducts were measured. In the ¢rst hours after cisplatin

Figure 2. Cell cycle analyses of drug-sensitive (MeWo) and cisplatin-resistant (MeWo CIS 1) melanoma cells after propidium iodide stainingusing £ow cytometry. Analyses were performed at 0 h, 24 h, 48 h, and 72 h after cisplatin treatment. (A) Exposure of MeWo to 5 mg cisplatin per mL:t¼ 0 h, 31.4% in SþG2 phase; t¼ 24 h after cisplatin treatment, 80.0% in SþG2 phase; t¼ 48 h after cisplatin treatment, 67.2% in SþG2 phase; t¼ 72 hafter cisplatin treatment, 56.1% in SþG2 phase. (B) Exposure of MeWo CIS 1^5 mg cisplatin per mL: t¼ 0 h, 30.9% in SþG2 phase; t¼ 24 h after cisplatintreatment, 58.3% in SþG2 phase; t¼ 48 h after cisplatin treatment, 40.5% in SþG2 phase; t¼ 72 h after cisplatin treatment, 29.6% in SþG2 phase. (C)Exposure of MeWo to 20 mg cisplatin per mL: t¼ 0 h, 31,4% in SþG2 phase; t¼ 24 h after cisplatin treatment, 87,4% in SþG2 phase; t¼ 48 h aftercisplatin treatment, 92.2% in SþG2 phase; t¼ 72 h after cisplatin treatment, 98,7% in SþG2 phase. (D) Exposure of MeWo CIS 1^20 mg cisplatin permL: t¼ 0 h, 30,9% in SþG2 phase; t¼ 24 h after cisplatin treatment, 79.2% in SþG2 phase; t¼ 48 h after cisplatin treatment, 66,6% in SþG2 phase;t¼ 72 h after cisplatin treatment, 63.7% in SþG2 phase.

Table I. Cross resistance pattern of cisplatin-resistantmelanoma cells

Cell Lines

MeWo MeWo CIS 1Drugs IC50

1 IC501 - RR2 � p-value3

cisplatin 5.5 20.3^3.7^o 0.001carboplatin 55.3 158.4^2.9^o 0.001daunoblastin 0.0121 0.0162^1.3^o 0.01etoposide 0.1957 0.3139^1.6^o 0.05vincristine 0.000793 0.000778^1.0^n.s.4

1IC50, inhibitory concentration (mM).2RR, relative resistance (x-fold).3p-value, calculated by Student’s T-test.4n.s., no signi¢cant di¡erence.

174 LIEDERT ETAL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

exposure, in both melanoma variants the detected nuclearamounts of Pt-d(GpG) adducts corresponded to the relationshipof the ongoing formation of platinum-DNA intrastrand cross-links and the initiating adduct elimination by activated cellularDNA repair (Fig 3). After a time period of 8 h, the peak of nu-clear platinum-DNA adduct content was reached and thenamounts of adducts started to drop, ¢rst caused predominantlyby prevailing DNA repair and later supported by ‘‘adduct dilu-tion’’due to ongoing cell proliferation.At each time point after cisplatin exposure, the nuclear levels of

platinum-DNA adducts were lower in drug-resistant cells com-pared with the cisplatin-sensitive cell line (Fig 3). MeWo CIS 1cells did not show a faster or more e⁄cient removal of Pt-d(GpG) cross-links in comparison with the parental MeWo cellline, indicating that in both cell variants similar DNA repair ac-tivities were triggered by these lesions. The fact that adduct levelswere already 2-fold lower in MeWoCIS 1 cells as early as 2 h aftercisplatin exposure (t0) clearly suggests a decreased adduct forma-tion in these cells resulting in more convenient conditions forproliferation, reduced cell cycle e¡ects, and less induction ofapoptosis. This situation is re£ected by the early re-entry ofMeWo CIS 1 cells into the cell cycle following G2 arrest and isaccompanied by a drug-resistant phenotype as measured in thecell proliferation assay (Table I). Both measures, peak cross-linkformation and integrated ‘‘adducts under curve’’ values stronglysuggest a signi¢cantly reduced DNA platination by cisplatin inMeWo CIS 1. To induce an adduct burden comparable with thatin the sensitive parental line MeWo, cisplatin-resistant cells had tobe exposed to a four times higher dose of cisplatin (20 mg permL) (Fig 3).Altogether, the data imply, that the biologic mechanism lead-

ing to the decreased drug sensitivity of MeWo CIS 1 cells actsvery early after cisplatin exposure. This instantly acting resistancemechanism might be represented by the ABC-transportercMOAT that can extrude platinum-based compounds from cells

directly after in£ux of the drug. Thus, cMOATwas identi¢ed asthe most decisive cellular mechanism to prevent the formation ofcisplatin-induced DNA lesions, and therewith to circumvent theactivation of signal transduction pathways leading to cell cycle ar-rest and apoptosis in MeWo CIS 1 cells.DNA-repair analyses from a previous study (Rˇnger et al,

2000) and the data from this study imply that neither an increasein the activity of the nucleotide excision repair pathway nor inthe base excision repair pathway is involved in the platinum-re-sistant phenotype of MeWo CIS 1.The decreased mismatch repairpathway in MeWo CIS 1 (Lage et al, 1999) has most likely no in-£uence on the elimination of platinum-DNA adducts. The mis-match repair system rather participates in the recognition of theseadducts, and as a result, trigger pathways that end in programmedcell death (Lage and Dietel, 1999). Lower levels of mismatch re-pair components therefore may lead to a reduction of apoptoticsignals and therewith to an elevated survival of drug-treated tu-mor cells.A decreased activation of distinct apoptotic pathways has been

reported for the cisplatin-resistant melanoma cells. After treat-ment with cisplatin (15 mg per mL) these cells showed a reductionin caspase 9-like activity, in cytochrome c release, and in poly(adenosine diphosphate ribose) polymerase cleavage (Helmbachet al, 2002). In the light of the ¢ndings in this study, however,these di¡erences are most likely due to the reduced DNA-adductformation in MeWo CIS 1 cells.In summary, the data from this study taken together with pre-

viously published results indicate that the cisplatin-resistant phe-notype of MeWo CIS 1 cells is mediated predominantly by thedecreased formation of platinum-DNA adducts due to the over-expression of cMOAT. A 4-fold higher dose of cisplatin is neces-sary in MeWo CIS 1 compared with MeWo cells to achieve: (1)the same cytotoxic e¡ects; (2) a similar degree of cell cycle distur-bance; and (3) the same burden of DNA-platination.The data obtained in this study may be important for clinical

management of melanoma as they demonstrate that the cruciale¡ect for the mediation of the cytotoxic e¡ects of platinum-basedanti-neoplastic agents is the formation of platinum-DNA ad-ducts. Thus, the prevention of a decreased platinum-DNA adductformation in cisplatin-resistant melanoma cells is the most pro-mising strategy to overcome cisplatin resistance. All consecutivecellular events are merely a direct consequence of the platinum-DNA adduct formation. Accordingly, the inhibition of the bio-logic activity of cMOAT should be the main goal in overcomingresistance to platinum-containing anti-cancer drugs in human mela-noma. Moreover, cMOAT appears as an excellent candidate factorfor the diagnosis of clinical platinum resistance in melanoma.

This work was supported by grants of the Deutsche Forschungsgemeinschaft (DFG)to H.L. (LA 1039/1-3) and to J.T. (FOR 236/2).We like to thank Beate Karow andBirgit Schaefer for skillful technical assistance.

REFERENCES

Buschfort-Papewalis C, Moritz T, Liedert B, Thomale J: Down-regulation of DNArepair in human CD34(þ ) progenitor cells corresponds to increased drug sen-sitivity and apoptotic response. Blood 100:845^853, 2002

Christmann M, Pick M, Lage H, Schadendorf D, Kaina B: Acquired resistance ofmelanoma cells to the antineoplastic agent fotemustine is caused by reactiva-tion of the DNA repair gene MGMT. Int J Cancer 92:123^129, 2001

Chu G: Cellular responses to cisplatin. The roles of DNA-binding proteins andDNA repair. J Biol Chem 269:787^790, 1994

Cui Y, K˛nig J, Buchholz JK, Spring H, Leier I, Keppler D: Drug resistance andATP-dependent conjugate transport mediated by the apical multidrug resis-tance protein, MRP2, permanently expressed in human and canine cells. MolPharmacol 55:929^937, 1999

Fichtinger-Schepman AM, van der Veer JL, den Hartog JH, Lohman PH, Reedijk J:Adducts of the antitumor drug cis-diamminedichloroplatinum(II) with DNA.Formation, identi¢cation, and quantitation.. Biochemistry 24:707^713, 1985

Fisher DE: Apoptosis in cancer therapy: Crossing the threshold. Cell 78:539^542, 1994

Figure 3. Time kinetics of cisplatin-induced Pt-d(GpG) adduct for-mation in the drug-sensitive melanoma line MeWo and its cisplatin-resistant subline MeWo CIS 1 as measured by an immunocytologicassay. Cisplatin was added at t¼ ^ 2 h. The cells were exposed to 5 or 20mg per mL cisplatin for 2 h and nuclear Pt-d(GpG) adduct content wasdetermined after di¡erent time periods by £uorescence microscopy usingthe rat anti-Pt-d(GpG) adduct MoAb ‘‘18G10’’and a £uorescein isothiocya-nate-based double sandwich staining procedure. Signals were ampli¢edand recorded by a dual mode CCD camera, and fed into a four parameterimage analysis program. Displayed values represent the mean, the verticalerror bars SD.

cMOAT IN MELANOMACELLS 175VOL. 121, NO. 1 JULY 2003

Fogh J, Bean MA, Brˇggen J, et al: Comparison of a human tumor cell line beforeand after growth in the nude mouse. In: Fogh J, Giovanella B (eds).The NudeMouse in Experimental and Clinical Research. New York: Academic Press, 1978;p 215^234

Gonzalez VM, Fuertes MA, Alonso C, Perez JM: Is cisplatin-induced cell death al-ways produced by apoptosis? Mol Pharmacol 59:657^663, 2001

Grottke C, Mantwill K, Dietel M, Schadendorf D, Lage H: Identi¢cation of di¡er-entially expressed genes in human melanoma cells with acquired resistance tovarious antineoplastic drugs. Int J Cancer 88:535^546, 2000

Helmbach H, Kern MA, Rossmann E, Renz K, Kissel C, Gschwendt B, SchadendorfD: Drug resistance towards etoposide and cisplatin in human melanoma cellsis associated with drug-dependent apoptosis de¢ciency. J Invest Dermatol118:923^932, 2002

Helmbach H, Rossmann E, Kern MA, Schadendorf D: Drug resistance in humanmelanoma. Int J Cancer 93:617^622, 2001

Kawabe T, Chen ZS,Wada M, Uchiumi T, Ono M, Akiyama S, Kuwano M: En-hanced transport of anticancer agents and leukotriene C4 by the human cana-licular multispeci¢c organic anion transporter (cMOAT/MRP2). FEBS Lett456:327^331, 1999

Kelley SL, Basu A,Teicher BA, Hacker MP, Hamer DH, Lazo JS: Overexpression ofmetallothionein confers resistance to anticancer drugs. Science 241:1813^1815,1988

Kern MA, Helmbach H, Artuc M, Karmann D, Jurgovsky K, Schadendorf D: Hu-man melanoma cell lines selected in vitro displaying various levels of drugresistance against cisplatin, fotemustine, vindesine or etoposide: Modulationof proto-oncogene expression. Anticancer Res 17:4359^4370, 1997

Kool M, de Haas M, Sche¡er GL, et al: Analysis of expression of cMOAT (MRP2),MRP3, MRP4, and MRP5, homologues of the multidrug resistance-asso-ciated protein gene (MRP1), in human cancer cell lines. Cancer Res 57:3537^3547, 1997

Lage H, Dietel M: Involvement of the DNA mismatch repair system in antineoplas-tic drug resistance. J Cancer Res Clin Oncol 125:156^165, 1999

Lage H, Dietel M: Multiple mechanisms confer di¡erent drug-resistant phenotypesin pancreatic carcinoma cells. J Cancer Res Clin Oncol 128:349^357, 2002

Lage H, Christmann M, Kern MA, Dietel M, Pick M, Kaina B, Schadendorf D:Expression of DNA repair proteins hMSH2, hMSH6, hMLH1, O6-methyl-guanine-DNA metyltransfease and N-methylpurine-DNA glycosylase in mel-anoma cells with acquired drug resistance. Int J Cancer 80:744^750, 1999

Lage H, Helmbach H, Dietel M, Schadendorf D: Modulation of DNA topoisome-rase II activity and expression in melanoma cells with acquired drug resistance.Br J Cancer 82:488^491, 2000

Lage H, Helmbach H, Grottke C, Dietel M, Schadendorf D: DFNA5 (ICERE-1)contributes to acquired etoposide resistance in melanoma cells. FEBS Lett494:54^59, 2001

Masuda H, Ozols RF, Lai GM, Fojo A, Rothenberg M, Hamilton TC: IncreasedDNA repair as a mechanism of acquired resistance to cis-diamminedichloro-platinum(II) in human ovarian cancer cell lines. Cancer Res 48:5713^5716, 1988

Nessling M, Kern MA, Schadendorf D, Lichter P: Association of genomic imbal-ances with resistance to therapeutic drugs in human melanoma cell lines. Cyto-genet Cell Genet 87:286^290, 1999

Perego P, Giarola M, Righetti SC, et al: Association between cisplatin resistance andmutation of p53 gene and reduced bax expression in ovarian carcinoma cellsystems. Cancer Res 56:556^562, 1996

Rˇnger TM, Emmert S, Schadendorf D, Diem C, Epe B, Hellfritsch D: Alterationsof DNA repair in melanoma cell lines resistant to cisplatin, fotemustine, oretoposide. J Invest Dermatol 114:34^39, 2000

Schadendorf D,Worm M, Algermissen B, Kohlmus CM, Czarnetzki BM: Chemo-sensitivity testing of human malignant melanoma. A retrospective analysis ofclinical response and in vitro drug sensitivity. Cancer 73:103^108, 1994

Sinha P, Kohl S, Fischer J, et al: Identi¢cation of novel proteins associated with thedevelopment of chemoresistance in malignant melanoma using two-dimen-sional electrophoresis. Electrophoresis 21:3048^3057, 2000

Taniguchi K, Wada M, Kohno K, et al: A human canalicular multispeci¢c organicanion transporter (cMOAT) gene is overexpressed in cisplatin-resistant humancancer cell lines with decreased drug accumulation. Cancer Res 56:4124^4129,1996

Zhang K, Mack P,Wong KP: Glutathione-related mechanisms in cellular resistanceto anticancer drugs. Int J Oncol 12:871^882, 1998

176 LIEDERT ETAL THE JOURNAL OF INVESTIGATIVE DERMATOLOGY