(CANCER RESEARCH 55, 4352-4360, October 1, 1995]

Increased Rate of Adenosine Triphosphate-dependent Etoposide (VP-16) Efflux in aMurine Leukemia Cell Line Overexpressing the Multidrug Resistance-associatedProtein (MRP) Gene1

Aurelio Lorico, Germana Rappa, Srinivasan Srimatkandada, Carlo V. Catapano, Daniel J. Fernandes,Joseph F. Germino, and Alan C. Sartorelli2

Department of Pharmacology and Developmental Therapeutics Program ¡A.L, G. R., S. S., A. C. S.] and Department of Internal Medicine (Oncology) [J. F. G.I, Cancer Center,Yale University School of Medicine, New Haven, Connecticut 06520, and Department of Experimental Oncology, Hollinga Oncology Center, Medical University of SouthCarolina, Charleston, South Carolina 29425 /C. V. C., D. J. F.¡

ABSTRACT

WEHI-3B/NOVO is a cloned murine leukemia cell line selected forresistance to novobiocin that is cross-resistant to the cytotoxic action ofetoposide (VP-16) and to a lesser extent to a variety of other topoisomeraseII (topo ID-reactive drugs. We have reported previously (Cancer Res. 52:2782-2790, 1992) that WEHI-3B/NOVO cells exhibit a pronounced decrease in VP-16 induced DNA-topo II cross-link formation compared tothe parental WEHI-3B/S cell line in intact cells, in the absence of a

significant difference in the IM unknotting activity of topo II assayed innuclear extracts. Because the pattern of cross-resistance was suggestive ofa topo H-mediated mechanism, we have ascertained whether a change intopo II can account for the multidrug-resistant phenotype of WEHI-3B/NOVO cells. No differences existed between WEHI-3B/S and WEHI-3B/

NOVO cells in topo II inKN A and protein levels, as well as in the amountof topo II associated with the nuclear matrix.

Neither sensitive nor resistant cells expressed detectable levels of theMDR1 gene; however, VP-16 accumulation in WEHI-3B/NOVO cells was3-4-fold less than that present in WEHI-3B/S cells, whereas doxorubicin

accumulation was the same in both cell lines. Over the first 60 s, nodifference existed in the rate of uptake of VP-16 between parental andresistant cells; however, beyond the first 60 s of incubation, | 'll|\ l'-ld

accumulated to a greater extent in parental sensitive cells. Thus, anincreased rate of efflux of VP-16 was responsible for the lower steady-stateconcentration of the drug in resistant cells. The efflux Km for VP-16 inWEHI-3B/NOVO cells was 254.7 MMand the Vmax was 10.4 pmol/s/107

cells. In the presence of the inhibitors of energy metabolism, sodium azideand deoxyglucose, the efflux of VP-16 was markedly inhibited; readdition

of glucose restored the original efflux rate.Northern blot analyses using the human 10.1 probe for the 3'-terminal

region of the multidrug-resistance protein (MRP) cDNA revealed amKNA species of approximately 6 kb in WEHI-3B/NOVO cells but not inWEHI-3B/S cells. Overexpression was associated with amplification of thecognate gene. To ascertain whether the overexpressed gene in WEHI-3B/

NOVO cells was the murine MRP or a different member of the samesuperfamily of ATP-binding ABC cassette transporters, a 341-bp MRP

cDNA probe was generated from a murine genomic library. The murineprobe spanned a region corresponding to nucleotides 4367-4708 of the

human cDNA sequence, with which it shared 86% nucleotide sequencehomology. Using this probe, Northern blot analyses demonstrated thatWEHI-3B/NOVO cells overexpressed the MRP gene relative to WEHI-

3B/S cells. It has been shown clearly by others that transfection of thehuman MRP cDNA is sufficient to induce VP-16 resistance in several

tumor cell lines. Thus, although the characteristics of the efflux mechanism in WEHI-3B/NOVO cells differ from that reported for the MRP, itis conceivable that Overexpression of the murine MRP gene in WEHI-3B/NOVO cells is responsible for the increased rate of VP-16 efflux that

Received 8/16/94; accepted 8/3/95.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicale this fact.

1This research was supported in part by United States Public Health Service Grant

CA-44597 (D. J. F.). C. V. C. is the recipient of a Berti Fellowship from ChampionIndustries, Winston-Salem, NC.

2 To whom requests for reprints should be addressed, at Department of Pharmacology,

Yale University School of Medicine, 333 Cedar Street, New Haven, CT 06520.

results in decreased accumulation of drug, decreased formation of potentially lethal topo II-DNA covalent complexes and, ultimately, reduced

cytotoxicity.

INTRODUCTION

The epipodophyllotoxins VP-163 and VM-26 are among the most

useful of the available antineoplastic agents, with clinical activitybeing demonstrated both in hematological malignancies and in solidtumors (1). The anticancer activity of these agents is presumably dueto their ability to stabilize the topo II enzyme intermediate covalentlybound to DNA in a step that precedes DNA religation (2). Thismechanism of action is shared by other antitumor agents such as theanthracyclines, aminoacridines, and ellipticines (2-4). As is the case

for most chemotherapeutic agents, the efficacy of the epipodophyllotoxins in cancer therapy is limited by the occurrence of drug resistancein the targeted tumor cell population. The cellular insensitivity todrugs that act at the level of topo II is usually expressed as MDR (5).In several cell lines, MDR has been related to the Overexpression ofthe mdrl gene that codes for a membrane-associated glycoprotein, theP-glycoprotein, which causes a decrease in the intracellular accumu

lation of a variety of cytotoxic drugs, including those that inhibit topoII (5, 6).

Several cell lines display an atypical pattern of MDR; the predominant mechanism of resistance appears to be due to qualitative and/orquantitative changes in topo II rather than to alterations in drug export(7-19). The clinical relevance of these mechanisms of multidrugresistance ¡sstill uncertain. Recently, other transport-related but P-glycoprotein-independent mechanisms of resistance have been reported (20, 21), expecially for VP-16, which appears to have a loweraffinity for the P-glycoprotein compared to other MDR 1-related drugs

(22). Cole et al. (20) have identified the MRP gene, which encodes aMr 190,000 membrane-bound glycoprotein (23). This gene, which isoverexpressed in several non-P-glycoprotein multidrug-resistant celllines (23-25), has been cloned and the nucleotide sequence determined (20). Because it shares sequence homology with ATP-binding

cassette transmembrane transporters, it has been included in thissuperfamily (20), which also includes the mdrl gene. The amino acididentity between the MRP and the P-glycoprotein, however, is only

15%. A higher homology exists between MRP and other members ofthis superfamily, including the Leishmania cystic fibrosis transmembrane conductance regulator ItpgpA, which confers resistance to ar-senite, and Y-CF1, which confers resistance to cadmium in yeast.Transfection of human MRP cDNA was sufficient to confer a multi-drug-resistant phenotype in several tumor cell lines (26, 27).

Because the antineoplastic agents frequently used in the selection of

'The abbreviations used are: VP-16, eloposide; VM-26, teniposidc; topo II, topoisomerase II; WEH1-3B/S, drug-sensitive subclone of WEHI-3B D* monomyelocyticleukemia; WEHI-3B/NOVO, drug-resistant subclone of WEHI-3B D ' monomyelocytic

leukemia; MDR. multidrug resistance; GAPDH, glyceraldehyde 3-phosphatc dehydrogen-ase; MRP, multidrug resistance-associated protein.

4352

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

MDR cell lines (i.e., VP-16 and doxorubicin) target topo II, areextruded from cells by the P-glycoprotein, and appear to be substrates

for the MRP protein, many cell lines selected for resistance to theseagents exhibit at least two different alterations capable of conferringinsensitivity (10, 11, 16, 25). Such multiple changes complicate thedetailed analysis of individual mechanisms in resistant cell lines.

Novobiocin is a topo Il-targeted drug that is not extruded from cellsthrough the P-glycoprotein efflux system (28); for this reason, weused this antibiotic to develop and characterize a novobiocin-resistantsubline of the WEHI-3B murine monomyelocytic leukemia (29). Incontrast to a relatively low level of resistance to novobiocin (1.7-fold),WEHI-3B/NOVO cells exhibit a high level of resistance to VP-16(11-fold). Although WEHI-3B/NOVO cells displayed cross-resistanceto a wide variety of anti-topo II inhibitors, namely VP-16, VM-26,doxorubicin, aclacinomycin A, m-AMSA, and elsamicin, and al

though the resistant phenotype seemed to result from a decrease in thedrug-induced formation of covalent complexes between topo II and

DNA (29), no differences in topo II enzymatic activity were detectedbetween resistant and sensitive sublines. Furthermore, WEHI-3B/NOVO cells were not cross-resistant to vincristine (29), which is agood substrate for the P-glycoprotein.

In all non-P-glycoprotein-expressing MDR cell lines reported thusfar, resistance has been associated with a reduction in drug-inducedtopo II-DNA covalent complexes or, upon protein denaturation, indrug-induced topo II-mediated DNA cleavage. Recent studies suggest

that the functional topo II in proliferating cells is the form that isincorporated into the salt-insoluble nuclear matrix (30-32). Thus, theeffects of both m-AMSA and VM-26 in CEM leukemia cells have

been found to be localized in the nuclear matrix (33); moreover, in aVM-26-resistant cell line, insensitivity was related to the decreased

activity of nuclear matrix topo II (34). In the present report, we havedetermined the potential role of topo II, in particular the importance ofits association with the nuclear matrix and the potential role of theP-glycoprotein and the MRP genes in the MDR phenotype of WEHI-

3B/NOVO cells.

MATERIALS AND METHODS

Cell Growth Characteristics. WEH1-3B/S and WEHI-3B/NOVO cellclones were maintained in suspension culture in McCoy's 5A medium

(GIBCO, Grand Island, NY) supplemented with 15% fetal bovine serum at37°Cin an atmosphere of 5% CO2 in air as described previously (29).

Immunoblotting. WEHI-3B/S and WEHI-3B/NOVO cells in exponential

growth were washed with cold PBS containing a mixture of five proteaseinhibitors (1 mM phenylmethylsulfonyl fluoride, 10 fig/ml of soybean trypsininhibitor, 1 mM benzamidine, 2 fig/ml of aprotinin, and 50 /¿g/mlof leupeptin).The cell pellet was then lysed with 2% SDS in PBS containing the fiveprotease inhibitors at 68°Cfor 5 min, loaded onto a 7.5% polyacrylamide gel,

and subjected to electrophoresis. After electrotransfer onto a Biotrace nitrocellulose membrane (Gelman Sciences, Ann Arbor, MI), blots were blockedwith 1% BSA in 100 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 0.05% Tween

20 to prevent nonspecific binding. The membrane was then incubated for 3 hat 20°Cwith the primary antibody. Bound immunoglobulins were detected

using an alkaline phosphatase-conjugated secondary antibody. Washings between the binding of antibodies were performed with 100 mM Tris-HCl (pH

7.5), 150 mM NaCl, and 0.05% Tween 20. Visualization of immunoreactivepeptides was accomplished by incubation with /3-nitrotetrazolium blue and5-bromo-4-chloro-3-indolyl phosphate (Kirkegaard & Perry Laboratories, Inc.,

Gaithersburg, MD). Because topo II is a nuclear protein, comparisons betweentumor samples were accomplished by subjecting whole cell extracts containingequal amounts of DNA per sample to Western blotting. However, becauseWEHI-3B/S and WEHI-3B/NOVO cells in exponential growth have the same

mean DNA and protein contents, equal quantities of total protein were loadedonto the gels. A Coomassie blue-stained companion gel was always run

simultaneously with the Western blot to check for equal loading and general

proteolysis. The relative amounts of specific proteins present on Western blotsof WEHI-3B/S and WEHI-3B/NOVO cells were quantified using a Visage

2000 image analysis system (Biolmage, Ann Arbor, Ml).

Isolation of the Nuclear Matrix. For the preparation of cell membranes,nuclei were centrifuged at 1,000 x g for 10 min at 4°C,and the supernatants

were layered over 5 ml of a buffer containing 35% sucrose, 0.01 M Tris-HCl(pH 7.5), and 1 mM EDTA and centrifuged at 16.300 X g for 30 min at 4°C.

The interface was diluted 1:5 with 0.25 M sucrose containing 0.01 M Tris-HCl(pH 7.5) and subsequently centrifuged at 100,000 X g for 45 min at 4°C.The

pellet was resuspended in 0.25 M sucrose containing 0.01 MTris-HCl (pH 7.5),passed several times through a 25-gauge needle, and used immediately orstored at -70°C for later use.

Nuclear matrices were prepared by a modification of the procedure described earlier (35). Briefly, 8 X IO7 cells in exponential growth were har

vested and washed once in serum-free medium. The cell pellets were resus

pended in homogenization buffer (10 mM Tris-HCl (pH 7)-2 mM MgCl,-l mMphenylmethylsulfonyl fluoride-10 /ig/ml of soybean trypsin inhibitor-1 mMbenzamidine-2 fig/ml of aprotinin-50 /xg/ml of leupeptin) and allowed to swell

for 15 min on ice. The cells were then disrupted by 25 strokes in a Douncehomogenizer. The cell lysates were layered over a solution of 45% sucrose(w/v) at 4°Cand centrifuged at 1900 X g for 30 min at 4°C.The nuclear pellets

were resuspended in 1 ml of low salt buffer (10 mM Tris-HCl (pH 7)-l mMMgCl2-H) mM NaCl-1 mM phenylmethylsulfonyl fluoride-10 /xg/rnl of soybean trypsin inhibitor-1 mM benzamidine-2 fig/ml of aprotinin-50 /xg/ml ofleupeptin). One ml of high salt buffer [10 mM Tris-HCl (pH 7)-0.2 mMMgCl2-l mM phenylmethylsulfonyl fluoride-10 fAg/ml of soybean trypsininhibitor-1 mM benzamidine-2 /xg/ml of aprotinin-50 fig/ml of leupeptin,

containing either 2 M NaCl or 0.5 M ammonium sulfate] was then added overa period of l h with gentle shaking. After an additional 30 min of incubationon ice, DNase I (2000 units) was added, and the samples were incubated for 30min at 37°C.The resulting pellets were resuspended in 10 ml of high salt buffer

containing 1.0 M NaCl or 0.25 M ammonium sulfate, centrifuged at 7000 X gfor 15 min at 4°C,and then washed an additional time with 10 ml of low saltbuffer at 4°C.The final nuclear matrix pellets were resuspended in 50 mM

Tris-HCl (pH 7.5), 10% glycerol, 1 mM DTT, and the above listed protease

inhibitors.Transport of ['HJVP-16. The cellular fluxes of radiolabeled VP-16 were

measured according to methodology described earlier (28). The viability ofcells after exposure to investigational drugs was always assessed by theexclusion of trypan blue. At the indicated times after the onset of drugtreatment, 100 >xlof cell suspension were added to a tube containing 10 ml ofice-cold PBS, and the cells were collected by centrifugation. An aliquot of the

supernatant was saved for measurement of extracellular radioactivity, and thepellet was resuspended in 60 fil of ice-cold PBS and placed in an "oil stop"

tube consisting of a 400-^1 Eppendorf microfuge tube containing 125 fi\ of oil

(16% Fisher 0121 light paraffin oil and 84% Dow Corning 550 silicon fluid,with a final specific gravity of 1.04 g/ml) layered over 30 /nl of 15% trichlo-

roacetic acid and immediately centrifuged for 30 s at 10,000 g using aBeckman model B microfuge. Microfuge tubes were cut through the oil layer,and the lower parts placed in glass mini scintillation vials. After the additionof 200 fi\ of distilled water, vials were vortexed vigorously, 5 ml of scintillation cocktail were added, and radioactivity therein was measured by scintillation spectrometry. Zero-time binding, measured by adding radiolabeledVP-16 to cells prechilled on ice, was always less than 1%. The intracellularvolume of WEHI-3B cells was calculated for untreated cells and for cellsexposed to different concentrations of drugs using 'HjO to determine the totalwater space and [MC]inulin to calculate extracellular space (36).

Construction and Screening of Genomic DNA Library. Genomic DNAwas extracted from WEHI-3B/NOVO cells by standard procedures (35), subsequently cleaved by BamHl, and ligated to BomHI-digested ZAP Express

vector (Stratagene, La Jolla, CA). Packaging and amplification of the librarywere accomplished by standard procedures (35). Total plaques (2 X IO6) were

screened under conditions of low stringency of hybridization with the human10.1 MRP cDNA probe. The inserts from the positive plaques were recoveredby PCR using T3 and T7 as primers. The inserts were sequenced by the dyetermination method (35), and the sequence information was compared to thesequence of human MRP cDNA by computer analysis using the University ofWisconsin Genetics software package to determine the degree of homology.

4353

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-lf, RESISTANCE

Preparation of the Murine MRP Probe. One /xg of total RNA, isolatedby the method of Chomczynski and Sacchi (37) from WEHI-3B/NOVO cells,was reverse transcribed in 20 /xl of RT buffer (10 mM Tris-HCl (pH 8.3)-50mM K.CI-5 mM MgCl2-l mM each of dATP, dCTP, dGTP, and dTTP) contain

ing 2 unii . /i I of RNase inhibitor (Promega, Madison, WI), 0.003 A2IMrandomhexanucleotides (Boehringer Mannheim, Indianapolis, IN), and 0.4 unit//al ofMoloney murine leukemia virus reverse transcriptase (Perkin Elmer Cetus,Norwalk, CT). After incubation for 10 min at 25°Cand for 15 min at 42°C,thereaction was terminated by incubation for 5 min at 99°C.A 341-bp fragment

of murine MRP cDNA was subsequently amplified from the resulting cDNAmixture by PCR with the following specific primers: 5:-primer, TGC-AGT-CAG-TGT-CCT-GAT-GA; 3'-primer, TTC-TCC-TTT-GTC-CAG-GAC. The

fragment was cloned in the pCR vector (Invitrogen, San Diego, CA), cut withEcfjRl, separated on a 2% agarose gel, and recovered and purified from the gelby the Gene Clean II kit procedure (Bio 101, Inc., Vista, CA). The fidelity ofthe murine MRP sequence was confirmed by DNA sequence analysis.

Northern Blotting. Northern blotting was performed using standard techniques (35). The human topo II cDNA probe pBS-hTop2 (38) and the human

topo I DNA probe (39) were gifts from Dr. J. C. Wang (Harvard University,Boston, MA). The human MRP probe, a 1.0-kb EroRI cDNA fragment (20),was donated by Dr. S. Cole (Queen's University, Kingston, Canada). The

human MDR, cDNA probe, pHaMDA,/A (40), was a gift from Dr. M. G.Gottesman (National Cancer Institute, Bethesda, MD). Radiolabeling of theseprobes was performed using the Amersham Multiprime labeling system (Am-ersham, Arlington Heights, IL). The blots were hybridized at 42°Covernight

with the radiolabeled probe in 50% formamide. After washings, the membranes were exposed to X-ray film at —70°Cwith intensifying screens.

Southern Blotting. Genomic DNA (15 /xg) from WEHI-3B/S and WEHI-

3B/NOVO cells was completely digested with EcoRI, HinM\, Bam\\\, Pstl, orB.c/II. subjected to electrophoresis on a 0.7% agarose gel, and blotted onto anitrocellulose membrane. The DNA was hybridized at 42°Covernight in 50%

formamide with a 1.0-kb EcoRI cDNA fragment of the human MRP cDNA, orwith the pBS-hTop2 probe, which were labeled by random priming with[a-3;;P|dCTP. After washing, the membranes were exposed to X-ray film at-70°C with intensifying screens.

RESULTS

Patterns of Cross-Resistance. We have reported previously that,in addition to an approximately 2-fold level of resistance to novobio-cin, the WEHI-3B/NOVO subline exhibited 7- and 11-fold levels ofcross-resistance to the topo II-targeted drugs VM-26 and VP-16,respectively, and about 2-fold resistance to the intercalating topoII-reactive drugs doxorubicin, m-AMSA, and aclacinomycin A,whereas sensitivity to the cytotoxic action of the non-topo II-acting

agents camptothecin and vincristine was not altered (29). We havenow extended the analysis of the pattern of cross-resistance to includethe antimetabolite 1-ß-D-arabinofuranosylcytosine, the VP-16 analogue W-68 (the substitution of the glycosidic moiety of which by anarylamino group has made it more cytotoxic toward cells that over-express the P-glycoprotein while leaving its topo Il-inhibitory activityunchanged; Ref. 41), sodium arsenite (for which MRP-overexpressing

cells appear to be resistant), and the antimitotic agent colchicine.When WEHI-3B/S cells were continously exposed for 72 h to 1-ß-D-arabinofuranosylcytosine, W-68, sodium arsenite, or colchicine, the

IC5(,s were 0.3 /J.M,0.03 /¿M,2 fiM, and 2 /tg/ml, respectively. ForWEHI-3B/NOVO cells, the IC5()swere 0.3 /XM,0.06 /XM,2.15 JAM,and2 /tg/ml for 1-ß-D-arabinofuranosylcytosine, W-68, sodium arsenite,

and colchicine, respectively. Therefore, whereas the sensitivity ofWEH1-3B/NOVO cells to 1-ß-D-arabinofuranosylcytosine, sodiumarsenite, and colchicine was unchanged compared to WEHI-3B/Scells, a 2-fold level of cross-resistance to W-68 was observed.

Expression of MDR,. We have shown that WEHI-3B/NOVOcells are not cross-resistant to the MDR-related drugs vincristine andcolchicine (29), which are substrates for the P-glycoprotein. Furthermore, the P-glycoprotein was not detectable on the surface of WEHI-

3B/S and WEHI-3B/NOVO cells by flow cytometric analyses ofimmunostained cells (28). To obtain further evidence that the P-glycoprotein was not expressed in WEHI-3B cells, both WEHI-3B/Sand WEHI-3B/NOVO cells were analyzed for the expression of

MDR, mRNA. Northern blotting, using a human cDNA probe knownto cross-hybridize well with murine MDRlu and MDR,,, (40), did not

detect MDR¡mRNA in either cell line (results not shown).Band Depletion Immunoblotting. Because the resistant pheno-

type of the WEHI-3B/NOVO leukemic clone appears to result from adecrease in the drug-induced formation of DNA-protein cross-links,as measured by the potassium-SDS precipitation assay (29), we haveused a band-depletion immunoblotting assay to verify that the formation of VP-16-stabilized covalent complexes between the M, 170,000

isoform of topo II and DNA was decreased in the resistant sublinecompared to parental cells. Several groups have shown that topoi-

somerases can be prevented from entering polyacrylamide gels bytheir drug-induced covalent binding to DNA (42, 43). When intactWEH1-3B/S cells were exposed to different concentrations of VP-16for 30 min at 37°Cand then lysed with an SDS-containing solution,

the endogenous complexes between the Mr 170,000 isoform of topo IIand DNA, stabilized at the time of the addition of SDS, were co-valently linked, and the DNA-bound Mr \ 70,000 form of topo II waspartially prevented from entering the gel. However, VP-16 treatmentonly minimally depleted the bands from WEHI-3B/NOVO cells.

Thus, densitometric analyses showed that 50.6% of the M, 170,000topo II band was depleted in WEHI-3B/S cells by a 30-min period ofexposure to 50 /J.MVP-16, whereas no depletion occurred in WEHI-

3B/NOVO cells (Table 1). The same experimental approach was usedto analyze the effects of VP-16 on the M, 180,000 isoform of topo II.No band depletions were observed up to 100 JJ.MVP-16 in eitherWEHI-3B/S or WEHI-3B/NOVO cells (data not shown), presumablyreflecting a lower affinity of VP-16 for the M, 180,000 form of the

enzyme (44).Expression of Topo II. In several non-P-glycoprotein-mediated

multidrug-resistant cell lines, drug insensitivity has been associated

with a decreased level and/or activity of topo II. Measurement of thecatalytic activity of topo II by assaying the ATP-dependent unknotting

of P4 DNA did not demonstrate differences in enzyme activity between the two cell lines (29). Similar levels of the Mr 170,000 topo IIwere observed in cell lysates from WEHI-3B/S and WEHI-3B/NOVO

cells (Fig. 1, left panel) using monoclonal antibodies raised against asynthetic peptide from the human M, 170,000 topo II sequence (45).An antiserum raised in rabbits against the native purified p 170 form oftopo II from the murine P388 leukemia (44, 46) also gave analogousresults (data not shown). To determine whether the equivalent levelsof topo II in the sensitive and resistant clones corresponded to anequal expression of the topo II gene, Northern analyses were con-

Table 1 Interference with the capacity' of immunoreaclive Mr 170.000 topo II preparedfrom WEHI-3B/S and WEHI-3B/NOVO cell lysates after treatment of the intact cells

HïlhVP-16 from entering a polyacrylamide get11

Cell line VP-16 concentration(fiM)WEHI-3B/SWEHI-3B/NOVO0255002550Percent ofcontrol10074.749.410098.3100

" Cells (10' cells/ml) in logarithmic growth were incubated with 25 or 50 JAMVP-16for 30 min at 37°C.Cells were washed, lysed as described in "Materials and Methods."

and 25 mg of lysate protein were subjected to 7.5% SDS-PAGE, after which the proteinswere immunohlotted using an antiserum against the M, 170,000 isoform of topo II. Themembranes were analyzed by densitometry using a Visage 2000 image analysis system.Controls were normalized to 100%. Results are the mean of two separate experiments inwhich variability was less than 10%.

4354

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

kDa R R

Fig. 1. Western blot analyses of topo II fromWEHI-3B/NOVO and WEHI-3B/S cells. Equivalent amounts of protein from cell lysates (a-d) orfrom nuclear matrices (e-h) prepared from WEHI-3B/NOVO (K) and WEHI-3B/S (S) cells were subjected to 7.5% SDS-PAGE and transferred to Im

mobilem filters. Filters were incubated with a mAbraised against a synthetic peptidc from the humanMj 170,000 topo II sequence, and then with phos-phatase-conjugated goat antirabbit IgG as detailedin "Materials and Methods." a and c, 25 fig of total

protein; b and d-h, 50 pig of total protein, e and/,

nuclear matrices extracted with ammonium sulfate;g and A, nuclear matrices extracted with NaCl.Molecular size standards are shown on the left.Similar results were obtained in several replicateexperiments. kDa, molecular weight in thousands.

205-

R

e

S

f

R

g

116-

80-

ducted with human topo II cDNA. Consistent with previous reports(38, 39), the topo Il-specific probe hybridized to an RNA species ofapproximately 6.2 kb (Fig. 2A), whereas the topo I-specific probe

hybridized to an RNA species of approximately 4.1 kb (Fig. 2B). Bothtopo I and II mRNA levels were very similar in the two cell lines. Inthese studies, y-actin was used as an internal standard.

Nuclear Matrix-associated Topo II. One of our laboratories has

shown previously that the amount of immunoreactive topo II in thenuclear matrices of multidrug-resistant CEM/VM-1 cells was decreased more than 3-fold relative to that of parental CEM cells,

whereas no significant difference occurred in the amount of enzymepresent in the nonmatrix fraction of nuclei from these cell lines.Therefore, we have ascertained whether this was also the case forWEHI-3B/NOVO cells. In Fig. 1, right panel, is shown a represent

ative Western blot that was obtained after electrophoresis of 50 fig ofmatrix protein isolated from each cell line with either a NaCl- or anammonium sulfate-containing solution. Quantitation of the relative

amounts of immunoreactive topo II in 7 immunoblots of 5 differentmatrix preparations was obtained by scanning the positive films of theWestern blots using a Visage 2000 scanner; these measurementsdemonstrated that the amounts of immunoreactive topo II in thenuclear matrices of WEHI-3B/NOVO cells was equivalent to thatpresent in WEHI-3B/S cells.

Drug Transport Studies. Recently, P-glycoprotein-independenttransport-related mechanisms of resistance have been proposed (21,47), especially for VP-16, which appears to have a lower affinity forthe P-glycoprotein than that of other MDR1-related drugs (22). Thefew reports of the membrane transport of VP-16 suggest that the drug

enters cells by passive diffusion and is actively extruded by theP-glycoprotein in cells in which this protein is expressed (48, 49). To

ascertain whether a difference in drug transport might be responsiblefor the resistant phenotype of WEHI-3B/NOVO cells, we measuredthe cellular uptake of VP-16, demonstrating that in WEHI-3B/NOVOcells the steady-state levels of [3H]VP-16 were 3-4 times lower than

in parental WEHI-3B/S cells (Fig. 3a). In contrast, no difference wasobserved between WEHI-3B/S and WEHI-3B/NOVO cells in the

accumulation of doxorubicin, as measured by flow cytometric analysis of cells exposed to doxorubicin for 2 h (Fig. 3b). Furthermore, nodifference occurred between parental and resistant sublines in theaccumulation of the antimetabolite [3H]5-fluoro-2'-deoxyuridine

(data not shown).In both sensitive and resistant WEHI-3B cells, intracellular VP-16

binding did not appear to play an important role in the accumulationof the epipodophyllotoxin because the noneffluxable pool of VP-16

was found to be extremely small (50); therefore, differences in thebinding of this agent to cellular components do not appear to play arole in the phenomenon of resistance to VP-16 in WEHI-3B/NOVOcells. Measurement of the bidirectional fluxes of [3H]VP-16 were

conducted to ascertain whether a decrease in the rate of uptake or anincrease in the rate of efflux, or both, were responsible for thedecrease in the steady-state accumulation of VP-16. No differenceexisted in the rate of uptake of VP-16 between parental and resistant

cells over the first 60 s (data not shown). Beyond the first 60 s ofincubation, however, [3H]VP-16 began to accumulate to a greater

extent in parental cells, reaching a plateau at 20 min (Fig. 4). Thesefindings implied that a change in the rate of efflux of VP-16 wasresponsible for the marked decrease in the steady-state levels of thisagent in WEHI-3B/NOVO cells. To determine whether this wasindeed the case, WEHI-3B/S and WEHI-3B/NOVO cells were loadedto the same intracellular concentration of [3H]VP-16 by incubation for

30 min with 3.3 and 10 p,M VP-16, respectively. Cells were thenwashed free of excess drug at 0°Cand resuspended in VP-16-freemedium at 37°C.The rate of efflux of VP-16 in WEHI-3B/NOVO

cells was considerably faster than in WEHI-3B/S cells (Fig. 5),indicating that in WEH1-3B/NOVO cells a membrane protein responsible for the export of VP-16 was operative.

Although the P-glycoprotein is not expressed in WEHI-3B/NOVOcells, the possibility that an alternative ATP-dependent pump analogous to the P-glycoprotein was present in WEHI-3B/NOVO cells wasexamined. To determine whether VP-16 transport in WEHI-3B/

NOVO cells was ATP dependent, we inhibited cellular energy metabolism by incubation of cells in medium without glucose thatcontained 10 ITIMsodium azide and 1 mg/ml of 2-deoxy-o-glucose.

The intracellular concentration of ATP was measured by HPLC on aWhatman Partisil 10 SAX anion-exchange column and was found to

be decreased by greater than 90% after incubation for 20 min in thepresence of sodium azide. Under these conditions of energy deprivation, the rates of VP-16 efflux from WEHI-3B/NOVO and WEHI-3B/S cells were measured (Fig. 5). Cells were loaded with radiola-beled VP-16 for 30 min by exposure to the epipodophyllotoxin in thepresence of sodium azide and 2-deoxy-o-glucose, subsequentlywashed with ice-cold PBS, and resuspended in drug-free medium at37°Cin the presence or absence of sodium azide and 2-deoxy-D-

glucose. At various intervals of time thereafter, the efflux of VP-16

4355

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

A.

Topo II-

28S-

O

O2 COm OSco co

x xLU LU

••

§oZ C/Dm mco coi iX XLU LU<: £

18S-

—y-Actin

B.

VP-16 in WEHI-3B/NOVO cells was found to be 254.7 /XM,whereasthe Vmax was 10.4 pmol/s/107 cells (Fig. 6).

Many well known ATP-dependent membrane carriers, including

active nucleoside transporters, require the presence of extracellularsodium for their function (36). To ascertain whether the transport ofVP-16 in WEHI-3B/NOVO cells was sodium dependent, VP-16 ac

cumulation was measured in the absence of extracellular sodium afterwashing cells twice with Na +-free HBSS containing 0.15 M choline

chloride as the counterion instead of 0.15 MNaCl, and conducting themeasurements in HBSS containing choline chloride. This procedure issufficient to reduce the extracellular Na + concentration to less than

5% of its normal concentration (36). Under these conditions, thesteady-state concentrations of [3H]VP-16 were 1.3 ± 0.16

(mean ±SD) JU.Min the presence of sodium and 1.1 ±0.07 JIM insodium-depleted medium (difference not statistically significant byStudent's t test), indicating that the VP-16 transporter operative in

WEHI-3B/NOVO cells is not sodium dependent.Expression of MRP. The probe 10.1, complementary to a 1-kb

region of the 3' end of the human MRP cDNA, hybridized to an

mRNA species of approximately 6 kb present in WEHI-3B/NOVOcells but not in parental WEHI-3B/S cells (Fig. 7A). Southern blot

analyses were used to ascertain whether the selective expression ofthis gene in the resistant cell line was attributable to amplification ofthe MRP gene (Fig. 8). Genomic DNA (15 /j.g) from WEHI-3B/S andWEHI-3B/NOVO cells was completely digested with £coRI,Hindlll,

BtiinHl, Pxtl, or Bglll. After electrophoresis through a 0.7% agarosegel and blotting, the DNA was hybridized with probe 10.1 or with thepBS-hTop2 probe. The probes were subsequently stripped from the

28S-TopoI~ •

18S-—y-Actin

Fig. 2. Northern Wot analyses of (opo I and II mRNAs from WEHI-3B/NOVO andWEHI-3B/S cells. Twenty /ig'of total RNA from WEHI-3B/NOVO and WEH1-3B/S cells

were hybridized with cDNAs for topo II (A), topo l (B), and y-actin as an internal control.

Topo //. topo II (approximately 6.2 kh) mRNA; Topo I, topo I (approximately 4.1 kh)mRNA; 2KS and IKS, positions of rRNA species. Similar results were obtained in severalreplicate experiments.

Fig. 3. Steady-stale cellular levels of ['HJVP-16 (A) and doxorubicin (B) in WEHI-

3B/S (O) and WEHI-3B/NOVO (D) cells exposed to various extracellular concentrationsof drug. Exposure to ['H]VP-16 was for 30 min. which resulted in the attainment of

steady-state levels of VP-16, and intracellular radioactivity was measured as described in"Materials and Methods." Exposure to doxorubicin was for 2 h; doxorubicin fluorescence

was measured by flow cylometry. and the results expressed in arbitrary units of fluorescence. Points, mean of two separate experiments in which the variability was less than10%.

was terminated by the addition of ice-cold PBS, and the quantity ofradiolabeled VP-16 retained by the cells was measured. The rate ofefflux of VP-16 was markedly decreased in energy depleted WEHI-3B/NOVO cells. In contrast, in WEHI-3B/S cells, a much smallerdecrease in the rate of VP-16 efflux was observed under conditions of

energy deprivation (Fig. 5). Moreover, under these conditions ofenergy deprivation, the decrease in the steady-state levels of VP-16 inWEHI-3B/NOVO cells was restored to the levels found in WEH1/

3B/S cells. Subsequently, the addition of sufficient glucose to theincubation medium for 15 min to restore the intracellular levels ofATP reestablished the accumulation defect in WEHI-3B/NOVO cells(data not shown). These observations suggest that the efflux of VP-16occurs through an ATP-dependent membrane transporter that is at

least partially responsible for the resistant phenotype. Moreover, thistransporter appears to be present in both parental and resistant clones,with a much higher activity in the resistant clone. The efflux Km for

|

10

Time (min)

Fig. 4. Time course of the influx of VP-16 by WEHI-3B/NOVO (D) and WEHI-3B/S(O) cells. Cells were incubated with 4 (¿M|'H]VP-16 as described in "Materials andMethods," and intracellular radioactivity was measured at different times thereafter.

Points, mean of 2-3 experiments in which the variability was less than 10%.

4356

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OK VP-1 h RESISTANCE

WEIII.JB/NOVO

2 4 6 8 10

Time (min)

0246

Time (min)

Fig. 5. Effects of the inhibition of cellular energy metabolism on the efflux of VP-16by WEHI-3B/NOVO and WEH1-3B/S cells. Cells were incubated for 30 min lo achievethe same intraccllular level of VP-16 by incubation with 3.3 (WEHI-3B/S) or 1Ü(WEHI-3B/NOVO) U.M|'H|VP-16. Cells were washed free of extracellular VP-16 by 3 washingsat 4°C,inii.iivlliil.li radioactivity was measured, and cells were rcsuspended in VP-16-freemedium at 37°C.Over the next 10 min, the levels of intraccllular radioactivity were

measured. In some cases, cells were incubated in HBSS with (•)or without (O) 10 HIMsodium azide and I mg/ml of 2-deoxy-u-glucose for 15 min al 37°C.['H]VP-I6 (4 JIM)

was then added for 30 min. Cells were subsequently washed free of extracellular drug at4°C.Intraccllular radioactivity was measured, and the cells were resuspended in VP-16free medium at 37°Cin the presence or absence of sodium azide and 2-deoxy-n-glucose.

The levels of intraccllular radioactivity were then measured. Pinnls, mean of 2-4

independent experiments.

filters, which were then rehybridized with a GAPDH probe as aninternal control. A comparison of the intensity of the bands, normalized for the GAPDH signal, revealed that the gene recognized by theanti-MRP probe was amplified 2.5 ± 0.3-fold (mean ± SE) inWEHI-3B/NOVO cells compared to parental WEHI-3B/S cells,

whereas no change was detected in the copy number of the topo IIgene. For the restriction enzymes used, no qualitative differences werenoted in the hybridization pattern between resistant and parental cells(Fig. 8).

To ascertain whether the overexpressed gene in WEHI-3B/NOVOcells was the murine MRP or a different member of the same super-family of ATP-binding ABC cassette transporters, we adopted thefollowing strategy. A BamHI-digested genomic library was constructed from WEHI-3B/NOVO cells in the ZAP Express vector, and

the library was screened with the human 10.1 MRP cDNA probeunder conditions of low stringency of hybridization. The screening ofthe WEHI-3B/NOVO genomic library resulted in the identification of

two plaques. The insert size in both cases was 1.7 kb, correspondingto the size of the lower band obtained by Southern analysis withßamHI-digested WEHI-3B/NOVO DNA (Fig. 8). Partial sequence

analysis revealed that the two inserts were identical. Two regions of100 and 150 nucleotides were 80-85% homologous to a correspond

ing region of the terminal part of the human MRP cDNA and 100%homologous to the murine MRP cDNA.4 A 341-bp fragment of the

murine MRP cDNA was then amplified by reverse transcriptase-PCRfrom the total RNA of WEHI-3B/NOVO cells, using as specific

primers two oligonucleotides based on the nucleotide sequence of themurine regions of homology with the human MRP cDNA. The 341-bpfragment was cloned as described in "Materials and Methods," se-

quenced by the dye termination method (35) and used as a probe inNorthern blot hybridizations. The probe spanned a region corresponding to nucleotides 4367-4708 of the human cDNA sequence, with

which it shared 86% nucleotide sequence homology. A representativeexperiment shown in Fig. IB illustrates that the murine MRP gene wasoverexpressed in WEHI-3B/NOVO cells compared to the parental

strain.

DISCUSSION

Many cell lines, in addition to the WEHI-3B/NOVO cell linedescribed in this report, that have a non-P-glycoprotein-mediatedMDR phenotype have been characterized recently (7-25). In these cell

lines, resistance was generally induced by exposure of cells to compounds that are mutagenic in mammalian cells (i.e., VP-16, VM-26,doxorubicin, m-AMSA. or an ellipticine; Refs. 51-54). Thus, mostnon-P-glycoprotein-overexpressing MDR cell lines described thus far

appear to have mixed phenotypes with more than one of the followingchanges: (a) a low level and/or activity of topo II; (b) an increasedexpression of MRP; (r) an increased content of glutathione and/orglutathione S-transferase activity; and/or (d) a change in the level ofdrug-metabolizing enzymes. In contrast, WEHI-3B/NOVO cells were

developed by exposure to novobiocin, which is presumed to benonmutagenic (55). Therefore, the risk of genotypic changes unrelatedto the resistant phenotype, or of mixed mechanisms of resistanceshould be very low. In fact, novobiocin-sensitive and novobiocin-resistant clones of WEHI-3B displayed the same growth rate, cell

cycle distribution, cell size, DNA and protein contents, cloning efficiency, and intracellular ATP levels.

In virtually all non-P-glycoprotein MDR cell lines, a decrease in theformation of drug-induced topo II-DNA covalent complexes or DNA

cleavage has been reported, generally with a good correlation beingexhibited between the degree of resistance to the formation of thesecomplexes and the level of insensitivity to the cytotoxic properties ofthe drug under study. We have reported previously that WEHI-3B/NOVO cells exhibit a pronounced decrease in VP-16 induced DNA-topo II cross-links in intact cells, compared to parental sensitiveWEHI-3B/S cells (29). In the present report, we extend these findingsby showing that the amount of VP-16-stabilized covalent complexes

between the Mr 170,000 isoform of topo II and chromatin DNA is lessin WEHI-3B/NOVO compared to WEHI-3B/S cells. Although theseobservations, together with the pattern of cross-resistance that wasexhibited only by topo II-targeted drugs, were suggestive of an alter

ation in topo II as the predominant cause of resistance, we did notdetect any difference in the P4 unknotting activity of topo II fromWEHI-3B/S and WEHI-3B/NOVO cells (29). Furthermore, theamount of VP-16-induced cleavable complexes in nuclear extracts

from the two cell lines was equivalent, as were the ATP requirementsof the topo II extracted from the two cell lines (29). We now reportthat WEHI-3B/S and WEHI-3B/NOVO cells have equal intracellularlevels of the enzyme, as well as the same steady-state levels of topo

II mRNA.Fernandos et al. (34) have reported that the level and activity of

topo II in isolated nuclear matrices from VM-26-resistant CEM/VM-1

leukemia cells were decreased relative to the sensitive parental line,

6-

e 4-

100 200 3i

VP-16(LtM)

400

4 R. G. Deeley, personal communication.

Fig. 6. Initial rate of efflux of [-'HJVP-16 in WEHI-3B/NOVO cells. Ceils wereincubated for 30 min with |'H|VP-16. washed free of extracellular VP-16 by 3 washingsat 4°C,and resuspended in VP-16-free medium at 37°C.Over the next 60 s, the levels of

intracellular radioactivity were measured. Points, mean of 2-4 independent experiments.

4357

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

A.

Fig. 7. Northern blot analyses of MRP mRNAfrom WEHI-3B/NOVO and WEHI-3B/S cells. A,

15 (Lanes I and 3) or 30 (Lanes 2 and 4) (¿goftotal RNA were hybridized with a 1.0-kh EcoRI

fragment of the human MRP cDNA. B. 15 ng oftotal RNA were hybridized with a 341-bp probefrom the murine MRP cDNA. GAPDH was used asan internal control. Similar results were obtained inseveral replicate experiments.

WEHI-3B/NOVO WEHI-3B/S

12 34

WEHI-3B/NOVO WEHI-3B/S

1

6.0 kb-

GAPDH-

6.0kb-

GAPDH-

whereas the amount of extractable nuclear topo II was unchanged. Itwas suggested that a mutation in topo II may have altered the association of this enzyme with the newly synthesized DNA of the nuclearmatrix, leading to a decrease in the interaction between topo II-targeted drugs and matrix-bound enzyme. Our finding that WEHI-3B/NOVO cells are highly resistant to the formation of covalent complexes induced by VP-16 (29) in relatively newly synthesized DNA(30 min labeling in the presence of VP-16) suggested that in thesecells, as in CEM/VM-1 cells, an alteration in the association of topoII with the nuclear matrix was the basis of the resistant phenotype.Preparations of nuclear matrices from WEHI-3B/S and WEHI-3B/NOVO cells contained equal amounts of immunoreactive topo II. Thestate of phosphorylation of topo II has been shown to be associatedwith drug resistance, with phosphorylation of topo II resulting in a

reduction in the levels of VP-16-induced topo II-DNA covalent complexes (56-57). Moreover, a mutation in the topo II a gene in theabsence of an enzymatic defect has been reported to be associatedwith VP-16 resistance in the human FEM-X melanoma cell line (58).However, our observation that cleavable complex formation is essentially the same in nuclear extracts prepared from WEHI-3B/S andWEHI-3B/NOVO cells (29), rules against the presence of differencesin the state of phosphorylation or mutations of the enzyme in the twocell lines. Therefore, the two most studied mechanisms of multidrugresistance (i.e., overexpression of the P-glycoprotein and alteration ofthe activity of topo II) are most likely not operative in WEHI-3B/NOVO cells. We have also found that neither the parental nor theresistant WEHI-3B clones express detectable levels of the pi 10 protein (data not shown), which has also been implicated in non-P-

Fig. 8. Southern blot analyses of genomic DNAfrom WEH1-3B/NOVO (A) and WEHI-3B/S (S)cells digested with EroRl (£),Hindlll (H), BamHl(B), Pul (/>), or Bgl\\ (Bit). The DNA was hybridized with the pBS-hTop2 anti-topo II probe (0) orwith the I.O-kb EcoRI fragment of the human MRPcDNA (fc) as described in "Materials and Methods." Molecular size standards are shown on theleft. Exposure was for 12 days at —70°Cusing an

intensifying screen.

aR S

kb

23-

9.4-

6.5-

4.3-...

bS R

2.3-

l

P B E H P B

4358

B B H H Bg Bg E

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

glycoprotein-mediated multidrug resistance (21). Moreover, WEHI-3B/NOVO cells are not cross-resistant to vincristine, which is one ofthe agents to which pllO-overexpressing cell lines exhibit cross-

resistance, and which was used as the selecting agent for the resistantcell line to which an antibody to the M, 110,000 protein was developed.

The possibility that an alteration in drug transport was responsiblefor the resistant phenotype of WEHI-3B/NOVO cells was then inves

tigated. We reported previously (29) that only a small decrease in theaccumulation of [3H]VP-16 was detectable in WEHI-3B/NOVO cells

compared to the parental strain. Subsequently, however, we havefound that a high zero-time binding of [3H]VP-16 associated with the"cold" method (50) used interfered with the assay. The methodology

that we have used in the present study to measure the intracellulartransport of [3H]VP-16 in WEHI-3B/S and WEHI-3B/NOVO cells

combines the oil stop (36) with the cold method (50), a procedure thatreduces the zero-time binding of [3H]VP-16 to less than 1% of its

steady-state concentration, allowing accurate measurement of thistransport activity in WEHI-3B/NOVO cells, which accumulate lowlevels of [3H]VP-16. We have found that an increase in the efflux rate

of VP-16 in WEHI-3B/NOVO cells was responsible for a 3-4-folddecrease in the steady-state levels of VP-16 in these resistant cellscompared to the parental sensitive subline. The efflux of VP-16appears to be mediated by an energy-dependent, non-sodium-dependent saturable transporter with a relatively low affinity for VP-16, as

evidenced by a relatively high Km.Recently, by analyzing the intracellular accumulation of daunoru-

bicin in MRP-transfected cells, Zaman et al. (59) have concluded thatMRP is a plasma membrane drug-efflux pump. Northern blot analysesusing both human and murine probes for the 3'-terminal portion of

MRP cDNA revealed an mRNA species of approximately 6 kb inWEHI-3B/NOVO cells but not in the parental WEHI-3B/S cell line.

Overexpression of this gene is at least partially due to amplification ofthe cognate gene. To prove that this was indeed the murine MRP, wescreened a genomic DNA library from WEHI-3B/NOVO cells with

the human MRP probe and isolated a clone containing two regions of100-150 nucleotides with 80% nucleotide homology to the humanMRP cDNA and 100% homology to the murine MRP cDNA.4 On the

basis of this sequence information, we amplified by reverse tran-scriptase-PCR a 341-bp fragment from total RNA of WEHI-3B/

NOVO cells, cloned the fragment in an appropriate vector, and usedit as a probe in Northern blotting analyses, confirming that the MRPgene was indeed overexpressed in WEHI-3B/NOVO cells, relative toparental WEHI-3B/S cells.

Our findings that WEHI-3B/NOVO cells are not cross-resistant

to vincristine, colchicine, and arsenite, and that the low level ofresistance to doxorubicin is not associated with an alteration in the¡ntracellular accumulation of this anthracycline, differ from thepattern of cross-resistance reported for most human MRP-overex-

pressing cell lines, as well as for tumor cell lines transfected withthe human MRP cDNA, which become resistant to Vinca alkaloids,colchicine, anthracyclines, and the epipodophyllotoxin derivativesVP-16 and VM-26.

Two explanations can be considered for this discrepancy: (a) it ispossible that the native murine MRP gene preferentially encodes forepipodophyllotoxin, rather than Vinca alkaloid and anthracycline,resistance. A lack of significant cross-resistance to doxorubicin hasactually been reported by Slapak et al. (60) in vincristine-selected

murine erythroleukemia cells that overexpress the MRP gene. Moreover, because the level of primary resistance is low in WEHI-3B/NOVO cells, cross-resistance to vinblastine and colchicine may only

be detectable at higher levels of primary resistance and/or after higherlevels of MRP expression; or (b) another gene(s), possibly coampli-

fied with the MRP gene, might be responsible for the primary resistance to VP-16 and VM-26 and for the low level of primary resistanceto novobiocin. Preliminary studies from our laboratory on an MRP-

transfected cell line suggest that novobiocin is not a substrate for theMRP pump,5 strengthening the possibility that another not yet dis

covered gene is at least partially responsible for the pattern of resistance of WEHI-3B/NOVO cells.

In conclusion, we have shown that the predominant mechanism ofresistance to VP-16 in WEHI-3B/NOVO leukemia cells consists of anincrease in the rate of ATP-dependent VP-16 efflux, which results ina lower intracellular accumulation of VP-16, that in turn results in alower number of potentially lethal VP-16 stabilized covalent com

plexes between the Mr 170,000 isoform of topo II and DNA, and thatthis phenotype is associated with the overexpression of the MRP gene.Transfection of full-length murine MRP cDNA in the parental WEHI-

3B/S clone, currently ongoing in our laboratory, should clarifywhether the overexpression of the murine MRP gene is able to confera multidrug-resistant phenotype analogous to that of WEHI-3B/

NOVO cells.

ACKNOWLEDGMENTS

The authors thank Drs. F. Drake and G. Aslaldi-Ricotli for the anti-lopo II

antibodies. Dr. S. P. C. Cole for the MRP probe, Dr. R. E. Handschumacher forhelpful discussions, Dr. R. G. Deeley for sharing with us data on the murinesequence of the MRP cDNA. and V. Vellucci for technical advice.

REFERENCES1. O'Dwyer, P. J., Lcyland-Jones, B.. Alonso. M. T.. Marsoni, S., and Wines. R. E. Drug

therapy. Etoposide (VP-16-213). Current status of an active anticancer drug. N. Engl.J. Med., 312: 692-700. 1985.

2. Liu. L. F. DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem.. 58:351-375, 1989.

3. Kohn, K. W.. Pommier, Y., Kerrigan, D., Markovitz. J., and Covey, J. M. Topoisomerase II as a large! of anlicancer drug action in mammalian cells. Nail. CancerInst. Monogr., 4: 61-71, 1987.

4. Drlica, K., and Franco. R. J. Inhibition of DNA topoisomerases. Biochemistry. 27:2253-2259, 1988.

5. Beck, W. T. The cell biology of multiple drug resistance. Biochem. Pharmacol., 36:2879-2887, 1987.

6. Ling, V. Genetic basis of drug resistance in mammalian cells. In: N. Bruchovsk^ «nt!J. H. Goldie (eds.). Drugs and Hormone Resistance in Neoplasia, pp. 1-20. Boca

Raton, FL: CRC Press, 1982.7. Deffie. A. M.. Batra, J. K., and Goldenberg, G. J. Direct correlation between DNA

topoisomerase II activity and cytotoxicity in Adriamycin-sensitivc and -resistant P388leukemia cell lines. Cancer Res., 49: 58-62, 1989.

8. Danks, M. K., Schmidt, C. A., Cirtain, M. C., Suttle, D. P., and Beck, W. T. Alleredcatalytic activity of and DNA cleavage by DNA topoisomerase II from humanleukemic cells selected for resistance to VM-26. Biochemistry, 27: 8861-8869, 1988.

9. De Isabella, P.. Capranico, G., Binaschi, M.. Tinelli. S., and Zunino. F. Evidence ofDNA topoisomerase II-dependent mechanisms of multidrug resistance in P38K leukemia cells. Mol. Pharmacol., 37: 11-16, 1990.

10. De Jong, S., Zijlska. J. G., de Vries, E. G. E., and Mulder. N. H. Reduced DNAtopoisomerase II activity and drug induced DNA cleavage activity in an adriamycinresistant human small cell lung carcinoma cell line. Cancer Res., 50: 304-309, 1990.

11. Spiridonidis, C. A., Chatterjee, S.. Petzold, S. J., and Berger. N. A. TopoisomeraseII-dcpendenl and independent mechanisms of eloposide resistance in Chinese hamstercell lines. Cancer Res., 49: 644-650, 1989.

12. Pommier, Y., Kerrigan, D., Schwartz, R. E.. Swack, J. A., and McCurdy, A. AlteredDNA topoisomerase II activity in Chinese hamster cells resistant to topoisomerase IIinhibitors. Cancer Res., 46: 3075-3081, 1986.

13. Long, B. H., Wang, L., Lorico, A., Wang, R. C. C., Brattain, M. G., and Casazza,A. M. Mechanisms of resistance to etoposidc (VP-16) and tcniposide (VM-26) in

acquired resistant human colon and lung carcinoma cell lines. Cancer Res., 51:5275-5284, 1991.

14. Per, S. R., Mattern, M. R., Mirabelli, C. K., Drake, F. H., Johnson, R. K., and Crooke,S. T. Characterization of a subline of P388 leukemia resistant to amsacrine: evidenceof altered topoisomerase II function. Mol. Pharmacol., 32: 17-25, 1987.

15. Zwclling, L. A.. Hinds, M., Chan, D., Mayes, J., Sie, K. L., Parker, E.. Silberman. L.,Radcliffe, A., Beran, M.. and Blick, M. Characterization of an amsacrine-resistant lineof human leukemia cells. Evidence for a drug-resistant form of topoisomerase II. J.Biol. Chem., 264: 16411-16420, 1989.

5 G. Rappa, A. Lorico. and A. C. Sartorelli. unpublished report.

4359

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

MECHANISM OF VP-16 RESISTANCE

16. Ferguson. P. J.. Fisher, M. H., Stephenson, J., Li, D.. Zhou, B., and Cheng, Y.Combined modalities of resistance in etoposide-resislant human KB cell lines. CancerRes., 48: 5956-5964, 1988.

17. Politi, P. M.. and Sinha, B. K. Role of differential drug uptake, efflux, and binding ofetoposide in sensitive and resistant human tumor cell lines: implications for themechanisms of drug resistance. Mol. Pharmacol., 35: 271-278, 1989.

18. Tan, K. B., Mattern, M. R., Eng, W. K., McCabe, F. L., and Johnson, R. K.Nonproductive rearrangement of DNA topoisomerase I and II genes: correlation withresistance to topoisomerase inhibitors. J. Nati. Cancer Inst., 81: 1732-1735. 1989.

19. Nitiss, J. L., Liu, Y-X., and Hsiung, Y. A temperature sensitive topoisomerase IIalíeleconfers temperature dependent drug resistance on amsacrine and etoposide: agenetic system for determining the targets of topoisomerase II inhibitors. Cancer Res.,53: 89-93, 1993.

20. Cole, S. P. C, Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E., Almquist,K. C., Stewart, A. J., Kurz, E. U., Duncan, A. M. V., and Deeley, R. G. Overexpres-sion of a transporter gene in a multidrug-resistant human lung cancer cell line.Science (Washington DC). 258: 1650-1654, 1992.

21. Scheper, R. J., Broxterman, H. J., Scheffer, G. L.. Kaaijk, P., Dalton, W. S., vanHeijningen, T. H. M., van Kalken, C. K., Slovak, M. L., de Vries, E. G. E., van derValk, P., Meijer, C. J. L. M., and Pinedo, H. M. Overexpression of a M, 110,000vesicular protein in non-P-glycoprotein-mediated multidrug resistance. Cancer Res.,53: 1475-1479, 1993.

22. Politi. P. M., Arnold, S. T., Felsted, R. L., and Sinha, B. K. P-Glycoprotein-indepcndent mechanism of resistance to VP-16 in multidrug-resistant tumor cell lines:pharmacokinetic and photoaffinity labeling studies. Mol. Pharmacol., 37: 790-796,

1990.23. Krishnamachary, N., and Center, M. S. The MRP gene associated with non-P-

glycoprotein multidrug resistance encodes a 190-kDa membrane bound glycoprotein.Cancer Res., 53: 3658-3661, 1993.

24. Barrand, M. A.. Heppell-Parton, A. C., Wright. R. A., Rabbitts, P. H., and Twenty-man, P. R. A 190-kilodalton protein overexpressed in non-P-glycoprotein-containingmultidrug-resistant cells and its relationship to the MRP gene. J. Nati. Cancer Inst.,86: 110-117, 1994.

25. Schneider, E.. Horton, J. K., Yang, C-H., Nakagawa. M., and Cowan, K. H. Multidrugresistance-associated protein gene Overexpression and reduced drug sensitivity oftopoisomerase II in a human breast carcinoma MCF7 cell line selected for etoposideresistance. Cancer Res., 54: 152-158, 1994.

26. Grant, C. E., Hipfner, D. R., Valdimarrson, G., Almquist, K. C., Cole, S. P. C., andDeeley, R. G. Overexpression of multidrug resistance-associated protein (MRP)increases resistance to natural product drugs. Cancer Res., 54: 357-361, 1994.

27. Kruh, G. D., Chan. A., Myers, K., Gaughan, K.. Miki, T., and Aaronson, S. A.Expression complementary DNA library transfer establishes mrp as a multidrugresistance gene. Cancer Res., 54: 1649-1652, 1994.

28. Rappa, G., Lorico, A., and Sartorelli, A. C. Reversal of etoposide resistance in nonP-glycoprotein expressing multidrug resistant tumor cell lines by novobiocin. CancerRes., 53: 5487-5493, 1993.

29. Rappa, G., Lorico, A., and Sartorelli, A. C. Development and characterization of aWEHI-3B D ' monomyelocytic leukemia cell line resistant to novobiocin and cross-

resistant to other topoisomerase Il-targeted drugs. Cancer Res., 52: 2782-2790, 1992.

30. DiNardo, S, Voelkel, K., and Sternglanz. R. DNA topoisomerase II mutant ofSaccharomvces cereviaiae'. topoisomerase II is required for segregation of daughter

molecules at the termination of DNA replication. Proc. Nati. Acad. Sci. USA, 81:2616-2620, 1984.

31. Gasser, S. M., and Lacmmli. U. K. The organization of chromatin loops: characterization of a scaffold altachment site. EMBO J., 5: 511-518, 1986.

32. Wu, H-Y., Shyy, S., Wang, J. C., and Liu, L. F. Transcription generates positively andnegatively supercoiled domains in the template. Cell, 53: 433-440, 1988.

33. Fernandes, D. J., Smith-Nanni. C., Faff, M. T., and Neff, T. A. Effects of anti-leukemia agents on nuclear matrix-bound DNA replication in CCRF-CEM leukemiacells. Cancer Res., 48: 1850-1855, 1988.

34. Fernandes. D. J., Danks, M. K., and Beck, W. T. Decreased nuclear matrix DNAtopoisomerase II in human leukemia cells resistant to VM-26 and m-AMSA. Biochemistry, 29: 4235-4241, 1990.

35. Sambrook, J., Frisch, E. F., and Maniatis, T. Molecular Cloning: A LaboratoryManual, Ed. 2, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989.

36. Darnowski, J. W.. Holdridge. C., and Handschumacher, R. E. Concentrative uridinetransport by murine splenocytes: kinetics, substrate specificity, and sodium dependency. Cancer Res., 47: 2614-2619, 1987.

37. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156-

159, 1987.38. Tsai-Pflugfelder, M., Liu, L. F., Liu, A. A., Tewey, K. M., Whang-Peng, J., Knutsen,

T., Huebner, K.. Croce. C. M., and Wang, J. C. Cloning and sequencing of cDNAencoding human DNA topoisomerase II and localization of the gene to chromosomeregion 17q21-22. Proc. Nati. Acad. Sci. USA, 85: 7177-7181, 1988.

39. Juan, C. C., Hwang, J., Liu, A. A.. Whang-Peng, J., Knutsen, T., Huebner, K., Croce,C. M., Zhang, H., Wang, J. C., and Liu, L. F. Human DNA topoisomerase I isencoded by a single-copy gene that maps to chromosome region 20ql2-13.2. Proc.Nati. Acad. Sci. USA, «5:8910-8913, 1988.

40. Ueda, K., Cardarelli, C., Gottesman, M. M., and Pastan, I. Expression of a full-lengthcDNA for the human "MDR1" gene confers resistance to colchicine, doxorubicin, and

vincristine. Proc. Nati. Acad. Sci. USA, 84: 3004-3008, 1987.

41. Chang, J. Y., Han, S. F., Liu, S. Y., Wang, Z. Q., Lee, K. H., and Cheng, Y. C. Effectof 4'-ß-arylamino derivatives of 4'-O-demethyl epipodophyllotoxin on human DNA

topoisomerase II. tubulin polymerization, KB cells, and their resistant variants.Cancer Res., 51: 1755-1759, 1991.

42. Hsiang, Y. H., and Liu, L. F. Identification of mammalian DNA topoisomerase I asan intracellular target of the anticancer drug camptothecin. Cancer Res., 48: 1722-1726, 1988.

43. Hwang, J., Shyy, S., Chen, A. Y., Juan, C. C., and Whang-Peng, J. Studies oftopoisomerase-specific antitumor drugs in human lymphocytes using rabbit antiseraagainst recombinant human topoisomerase II polypeptide. Cancer Res., 49: 958-962,1989.

44. Drake, F. H., Hofmann, G. A., Bartus, H. F., Mattern, M. R., Crooke, S. T., andMirabelli, C. K. Biochemical and pharmacological properties of pl70 and pl80 formsof topoisomerase II. Biochemistry, 28: 8154-8160, 1989.

45. Negri, C., Chiesa. R., Cerino, A., Bestagno, M., Sala, C., Zini, N., Maraldi, N. M., andAstaldi-Ricotti, G. C. B. Monoclonal antibodies to human DNA topoisomerase I andthe two isoforms of DNA topoisomerase II: 170- and 180-kDa isozymes. Exp. CellRes., 200: 452-459, 1992.

46. Chung, T. D. Y., Drake, F. H., Tan, K. B.. Per, S. R., Crooke, S. T., and Mirabelli,C. K. Characterization and immunological identification of cDNA clones encodingtwo human DNA topoisomerase II isozymes. Proc. Nati. Acad. Sci. USA, 86:9431-9435, 1989.

47. Deeley, R. G., Grant, C. E., Almquist, K. C., Hipfner, D. R.. Valdimarsson, G., Loe,D. W., Callen, D. F., KUSS,B., and Cole, S. P. C. Multidrug resistance mediated bythe 190 kDa MRP, a novel member of the ATP-binding cassette transporter super-family. Proc. Am. Assoc. Cancer Res., 35: 698, 1994.

48. Allen, L. M. Comparison of uptake and binding of two epipodophyllotoxin gluco-pyranosides, 4'-demethyl epipodophyllotoxin thenylidenc-ß-D-glucoside and 4'-dem-

ethyl epipodophyllotoxin etnylidene-ß-D-glucoside, in the L1210 leukemia cell. Cancer Res., 38: 2549-2554, 1978.

49. Seeber, S., Osieka, R., Schmidt, C. G., Achterrath, W., and Crooke, S. T. In vivoresistance towards anthracyclines, etoposide, and CK-diamminedichloroplatinum(II).Cancer Res., 42: 4719-4725, 1982.

50. Lorico, A., Rappa, G., and Sartorelli, A. C. Novobiocin-induced accumulation ofetoposide (VP-16) in WEHI-3B D * leukemia cells. Int. J. Cancer, 52: 903-909,1992.

51. Gupta, R. S., Bromke, A., Bryant, D. W., Gupta, R., Singh, B., and McCalla, D. R.Etoposide (VP16) and teniposide (VM26): novel anticancer drugs, strongly muta-genie in mammalian but not prokaryotic test systems. Environ. Mutagen., 9 (Suppl.8): 43, 1987.

52. Moore, M. M., Brock, K. H., Doerr, C. L., and DeMarini, D. M. Mutagenesis ofL5178Y/TK+/-3.7.2C mouse lymphoma cells by the clastogen ellipticine. Environ.Mutagen., 9: 161-170, 1987.

53. Gupta, R. Tests for the genotoxicity of m-AMSA, etoposide, teniposide and ellipticinein Neurospora crossa. Mutât.Res., 240: 47-58, 1990.

54. Clements, J., and Vogel, E. W. Somatic recombination and mutation assays inDrosophila: a comparison of the response of two different strains to four mutagens.Mutât.Res., 209: 1-5, 1988.

55. Takahashi, K., Kaneko, I., Nishi, Y., and Inui, N. Chromatid aberrations in anovobiocin-resistant subline of Chinese hamster V79 cells after exposure to novobiocin. Mutât.Res., 174: 279-284, 1986.

56. DeVore. R. F.. Corbett, A. H., and Osheroff, N. Phosphorylation of topoisomerase IIby casein kinase II and protein kinase C: effects on enzyme-mediated DNA cleavage/religation and sensitivity to the antineoplastic drugs etoposide and 4'-(9-acridinyl-

amino)methane-sulfon-m-anisidide. Cancer Res., 52: 2156-2161, 1992.57. Takano, H., Kohno, K., Ono, M., Uchida, Y., and Kuwano, M. Increased phospho-

rylation of DNA topoisomerase II in etoposide-resistant mutants of human cancer KBcells. Cancer Res., 51: 3951-3957, 1991.

58. Campain, J. A., Gottesman, M. M., and Pastan, I. A novel mutant topoisomerase II apresent in VP-16-resistant human melanoma cell lines has a deletion of alanine 429.Biochemistry, 33: 11327-11332, 1994.

59. Zaman. G. J. R., Flens, M. J., van Leusden, M. R., de Haas, M., Mulder, H. S.,Lankelma, J.. Pinedo, H. M., Scheper, R. J.. Baas, F., Broxterman, H. J., and Borst,P. The human multidrug resistance-associated protein MRP is a plasma membranedrug-efflux pump. Proc. Nati. Acad. Sci. USA, 91: 8822-8826, 1994.

60. Slapak, C. A., Fracasso, P. M., Marteli, R. L., "-ppmeyer, D. L., Lecerf, J-M., and

Levy, S. B. Overexpression of the multidrug resistance-associated protein (MRP)gene in vincristine but not doxorubicin-selected multidrug-resistant murine erythro-leukemia cells. Cancer Res., 54: 5607-5613, 1994.

4360

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1995;55:4352-4360. Cancer Res   Aurelio Lorico, Germana Rappa, Srinivasan Srimatkandada, et al.  

) GeneMRPOverexpressing the Multidrug Resistance-associated Protein (Etoposide (VP-16) Efflux in a Murine Leukemia Cell Line Increased Rate of Adenosine Triphosphate-dependent

  Updated version

  http://cancerres.aacrjournals.org/content/55/19/4352

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/55/19/4352To request permission to re-use all or part of this article, use this link

on March 27, 2020. © 1995 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from