increased ap-1 activity in drug resistant human breast cancer mcf-7 cells

Breast Cancer Research and Treatment53: 229–240, 1999.c© 1999Kluwer Academic Publishers. Printed in the Netherlands.

Report

Increased AP-1 activity in drug resistant human breast cancer MCF-7 cells

Phillip J. Daschner, Henry P. Ciolino, Cathie A. Plouzek, and Grace Chao YehCellular Defense and Carcinogenesis Section, Basic Research Laboratory, National Cancer Institute, NIH, [H.P.C.,C.A.P., G.C.Y.] Frederick, MD, USA; and Intramural Research Support Program, SAIC, NCI-FCRDC, [P.J.D.]Frederick, MD, USA

Key words:AP-1,MDR1, MCF-7 cells, multidrug resistance

Summary

The expression, DNA binding, and transactivating activity of activator protein 1 (AP-1) was examined in a series ofmultidrug resistant (MDR) MCF-7 human breast cancer cells that have increasing levels ofMDR1gene expression.We observed an increase in the amount of both c-jun and c-fos mRNA in cells with 12-, 65-, or 200-fold higherresistance to adriamycin when compared to drug-sensitive MCF-7 wild type (WT) cells. Electrophoretic mobilityshift assays (EMSA) demonstrated an increase in the DNA binding activity of an AP-1 complex in nuclear extractsfrom MDR MCF-7 cells when compared to extracts from WT cells. We observed a proportional increase in luciferaseexpression from a reporter vector containing consensus AP-1 binding sites in transiently transfected MDR cellswhen compared to WT cells, indicating that AP-1 mediated gene expression is increased in drug-resistant MCF-7cells. Since theMDR1promoter contains a putative AP-1 binding site, we used EMSA to examine AP-1 bindingactivity to an oligonucleotide probe that contained the relevantMDR1promoter sequences (−123 to−108). Nuclearextracts from resistant MCF-7 cells displayed an increased level of DNA binding of Jun/Jun dimers to the probe,indicating that AP-1 was capable of binding to this promoter site. A luciferase reporter construct containing triplicatecopies of theMDR1promoter sequence was expressed at higher levels in transiently transfected MDR cells whencompared to expression in WT cells. Co-transfection of WT cells with a c-jun expression vector and either of theAP-1 luciferase constructs demonstrated that c-jun could activate gene expression from both the consensus and theMDR1AP-1 sites in a dose dependent manner. In addition, RT-PCR and western blot analysis showed that levelsof MDR1 mRNA and Pgp were increased in c-jun transfected WT cells. Taken together, these data indicate thatincreased AP-1 activity may be an important mediator of MDR by regulating the expression ofMDR1.

Introduction

Overexpression of the 170 kD transmembrane proteinP-glycoprotein (Pgp) in tumor cells is associated withthe development of MDR to cancer chemotherapy [1,2]. Pgp is an energy-dependent drug efflux pump thatmaintains intracellular drug concentrations below cy-totoxic levels, thereby decreasing the cytotoxic effectsof a variety of chemotherapeutic agents [3], includinganthracyclines, vinca alkaloids, and epipodophyllo-toxins. Pgp is encoded by a multigene family thatincludes the human geneMDR1. Previous studies haveshown that Pgp levels can be increased both byMDR1gene amplification [4, 5] and by transcriptional ac-tivation [6–9]. The promoter and coding regions ofMDR1 have been cloned and sequenced [6], and al-

though several important promoter elements necessaryfor basal and inducible expression have been identi-fied and characterized [10–12, 17–22], the molecularmechanisms that controlMDR1expression remain un-clear. HumanMDR1 can be induced by exposure toheat shock, arsenite [13], retinoic acid, sodium butyrate[14], UV radiation [15], and chemotherapeutic drugs[16]. MDR1 gene expression is mediated by severaltrans-acting factors, including SP-1 [17], YB-1 [11,12], HSP [13], NF- IL6 [20], p53 [21, 22], and EGR1[23]. These studies suggest that expression ofMDR1isregulated at the transcriptional level by multiple factorsthat can be activated via stress-response pathways.

A well known mediator of stress response pathwaysis the transcription factor AP-1 [24], which is com-posed of dimers from the Jun and Fos protein families

230 PJ Daschner et al.

[25]. Several lines of evidence led us to investigatea possible role for AP-1 in the transcriptional regula-tion of MDR1. First, the promoter region ofMDR1contains a putative binding site for AP-1 [6] locatedwithin a region of the promoter that has been demon-strated to be necessary for gene induction [15, 26]. Inaddition, MDR human tumors and cell lines have beenfound to constitutively overexpress the component AP-1 proteins c-jun [27] and c-fos [28], and adriamycintreatment of a human T-cell leukemia line has beenshown to activate c-Jun N-terminal kinase [29], sug-gesting a possible role for AP-1 in the signal pathwayleading to MDR. These data are also consistent withreports thatMDR1 is induced through Ras/Raf medi-ated signal transduction pathways [30–32], which areknown to modulate AP-1 activity.

Although Pgp is constitutively expressed in a num-ber of human tissues [33], its normal function remainsunclear. We have reported that Pgp mediates the cel-lular efflux of several environmental carcinogens [34],and thus may be an important mechanism in preventingthe genotoxic effects of these compounds. Others [35]have demonstrated that Pgp in brain capillary endothe-lial cells forms an important part of the blood-brainbarrier. A better understanding of the factors regulat-ing Pgp expression is therefore important not only forimproving the efficacy of chemotherapy, but also forunderstanding its normal physiologic function.

In the present study we have examined the expres-sion levels, DNA binding activity, and transactivatingactivity of AP-1 in a series of drug-resistant MCF-7cells. Our results indicate that drug resistance in MCF-7 breast tumor cells is accompanied by increases inAP-1 activity, and suggest that this upregulation maybe important inMDR1expression in these cells.

Materials and methods

Cell lines

The MDR cells derived from the human breast car-cinoma cell line MCF-7 have been described [34].Drug-sensitive WT and multidrug resistant subclones(R12, R65, R200) were maintained in RMPI mediumcontaining 2 mM glutamine and 10% fetal bovineserum (Life Technologies, Rockville, MD) at 37◦C, ina humid atmosphere containing 5% CO2. Cytotoxicityvalues for drugs tested were determined using the sul-forhodamine B staining assay [36]. For all experiments,MDR cells were used in the fifth to 15th drug-freepassage.

Northern blot analysis

Total RNA was extracted from cell cultures by themethod of Chirgwin [37] and messenger RNA iso-lated using the FastTrack kit (Invitrogen, San Diego,CA) according to the manufacturer’s instructions. RNAwas separated by electrophoresis in a 1.2% denatur-ing agarose gel containing 2.2 M formaldehyde andtransferred onto Nytran membranes (Schleicher andSchuell, Keene, NH) by electroblotting, and was cross-linked before being probed with cDNA labeled by ran-dom primer extension with [32P]dCTP (New EnglandNuclear, Boston, MA). Probes used for hybridizationincludedMDR1 (a generous gift from M. Gottesman,38), c-jun, and c-fos(a generous gift from M. Birrer, 39,40). Hybridization was carried out overnight at 42◦Cin 50% formamide, 8× Denhardt’s buffer containing100mg/ml salmon sperm DNA. Blots were washed for30 min at a final stringency of 50◦C in 0.5× SSC/0.1%SDS.

Plasmid constructs

Two synthetic oligonucleotides and their complemen-tary sequences (Bioserve Biotechnologies, Laurel,MD) were annealed, endonuclease digested, and G-50 sephadex purified (Boehringer Mannheim, Pis-cataway, NJ) to generate DNA fragments that wereligated into an SstI/XhoI-digested pGL3 promoterluciferase vector (Promega, Madison, WI). All re-combinant DNA manipulations were performed us-ing standard techniques [41]. The inserted oligonu-cleotides contained either the sequence from position−123 to −108 of the MDR1 promoter region [9],(5′-GCTAGAGCTCGA(GCATTCAGTCAATCCG)3CTCGAGCTCGAGAGCG-3′) or an AP-1 consensussite [42], (5′-GCTAGAGCTCGAGCT(CATTGACTC-ATCCGA)3GCTCGAGCG)-3′). Plasmid constructscontaining oligonucleotide inserts with the actualMDR1 promoter sequences were designated pLucA,while constructs containing inserts with the AP-1 con-sensus binding sequence were designated pLucC. Allconstruct sequences were confirmed by the dideoxychain termination method [43].

Transient transfection assay

DNA transfections were performed using Lipofec-tamine reagent (Life Technologies, Rockville, MD)according to the manufacturer’s directions. Cells wereplated at a concentration of 4× 105 per well intosix-well plates and cultured overnight before beingtransfected with 1mg of the pGL3 reporter construct

Increased AP-1 in drug resistance231

and 15 ng of pCMVbgal (Clonetech Labs, PaloAlto, CA) to control for transfection efficiencies. Co-transfections were performed with the addition of 2mgpCMV-jun DNA (a generous gift from M. Birrer, 39),to the transfection medium, unless otherwise indicated.Transfected cells were harvested in reporter lysis buffer(Promega, Madison, WI) 48 h after addition of trans-fection mixture and lysates were cleared of cellulardebris by a brief centrifugation. Lysates were assayedfor luciferase activity using the luciferase assay reagent(Promega, Madison, WI) and forbgal activity using theONPG method [44]. Total protein content of lysateswas measured using Bradford reagent [45] (BioRad,Hercules, CA), with bovine serum albumin as standard.

Preparation of nuclear extracts

Nuclear extracts from MCF-7 cells grown in mono-layers were prepared using the method of Dignamet al. [46] with modifications. Briefly, cells wereharvested with trypsin-EDTA, washed with PBS, andpelleted by centrifugation at 1,000 g for 10 min at4◦C. Cell pellets were suspended in 3 pellet volumesof hypotonic buffer [10 mM HEPES-KOH (pH 7.9),10 mM KCl, 1.5 mM MgCl2, 0.2 mM phenylmethyl-sulfonylfluoride (PMSF), 0.5 mM DTT, 1mg/ml le-upeptin] and placed on ice for 10 min. Cells werelysed for 30 s at 15,000 rpm in a PT 3000 Polytron(Brinkman, Westbury, NY) homogenizer. Nuclei werepelleted by centrifugation at 3,000 g for 15 min, at 4◦Cand resuspended in one-half pellet volume of low saltbuffer [25% glycerol, 20 mM HEPES-KOH, 20 mMKCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.2 mM PMSF,0.5 mM DTT, 1mg/ml leupeptin]. An equal volume ofhigh salt buffer [25% glycerol, 20 mM HEPES-KOH,1.4 M KCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.2 mMPMSF, 0.5 mM DTT, 1mg/ml leupeptin] was addeddropwise to the gently stirred suspension and the nu-clei extracted on ice for 30 min. Extracts were clarifiedby centrifugation at 16,000 g for 30 min, 4◦C, and su-pernatants dialyzed overnight [20% glycerol, 20 mMHEPES-KOH, 100 mM KCl, 0.25 mM EDTA, 0.2 mMPMSF, 0.5 mM DTT] before aliquots were quick frozenin liquid nitrogen and stored at−70◦C.

Electro mobility shift assays (EMSA)

Nuclear extracts were assayed for DNA binding pro-teins using the double stranded DNA oligonucleotideprobes described in Figure 3. Oligonucleotides were[32P]dCTP labeled by a fill-in reaction with Klenowenzyme and purified from unincorporated nucleotidesby centrifugation through a G-25 sephadex column

(Boehringer Mannheim, Piscataway, NJ). The stan-dard binding reaction contained 15 mM Tris (pH 7.9),50 mM KCl, 5% glycerol, 1 mM MgCl2, 0.5 mMDTT, 0.25 mM EDTA, 0.5mg poly (dI-dC), and 10mgnuclear extract in a final volume of 20ml. The re-action mixture was preincubated for 10 min at roomtemperature prior to the addition of 50,000 cpm of la-beled probe (∼1 ng) and then further incubated 20 minat room temperature before DNA-protein complexeswere electrophoresed on a 4% polyacrylamide gel in25 mM Tris/glycine (pH 8.3). Gels were dried on 3mMWhatman paper and visualized by autoradiography.For competition assays, homologous or heterologousoligonucleotides were added to the binding reaction5 min before the addition of labeled probe.

Membrane preparations

Crude membrane fragments were prepared from con-fluent cell cultures in 75 cm2 tissue culture flasks(Costar, Cambridge, MA). Cells were sonicated in asucrose buffer (250 mM sucrose, 10 mM Tris-HCl (pH7.4), 0.1 mM PMSF) and centrifuged at 432,000 g for15 min at 4◦C. The membrane pellet was resuspendedin 100ml of 50 mM Tris-HCl (pH 7.4) with proteaseinhibitors (30mM leupeptin, 1mg/ml pepstatin A,100 U/ml Trasylol).

Western immunoblot analysis

Equal amounts of cell lysate or crude membranepreparation were separated on an 8% Tris-glycine-SDS polyacrylamide gel (Novex, San Diego, CA)for 1.5 h at 20 mA. Gels were electroblotted ontonitrocellulose and proteins were identified using mon-oclonal antibody C219 (Signet, Dedham, CA) for Pgp,or polyclonal antibody sc-45 (Santa Cruz Biotech-nology, Santa Cruz, CA) for c-jun. Detection wasperformed using horseradish peroxidase conjugatedsecond antibody with ECL reagents (Amersham, Ar-lington Heights, IL) according to the manufacturer’sinstructions.

RT-PCR for MDR1, c-jun, c-fos, andglyceraldehyde-3-phosphate dehydrogenase(G-3PDH)

cDNA templates were synthesized from 10mg of totalRNA using an RT-PCR kit (Clontech, Palo Alto, CA)as instructed. PCR was carried out using 5ml of cDNAmixed with 10× PCR buffer containing 50 mM MgCl2(Perkin–Elmer, Foster City, CA), 0.2mM primiersfor MDR1[47], c-jun, c-fos, or G-3PDH (Clontech),


1.5mCi [32P]-dATP, 200mM of nucleotides (dATP,dTTP, dCTP, dGTP), and 1.0 U AmpliTaq DNA poly-merase (Perkin–Elmer). Hot start was performed bypre-mixing AmpliTaq with anti-Taq antibody (Clon-tech) as instructed. Total volume of reaction mixturewas 25ml. Following PCR, the samples were subjectedto electrophoresis on 10% acrylamide TBE gels for1 h at 200 V. The gels were then dried and the resultsvisualized and quantitated with a GS-363 MolecularImager (BioRad, Hercules, CA).

Results

Overexpression of AP-1 components c-jun and c-fosin multidrug resistant MCF-7 cells

We examined the mRNA levels of the AP-1 componentproteins c-jun and c-fos in a series of adriamycin re-sistant human breast cancer MCF-7 cells. These cellsdisplayed increased expression of Pgp and increasedresistance against adriamycin, vinblastine, and taxol(Table 1), confirming their MDR phenotype. The mul-tidrug resistant MCF-7 cells were designated accordingto their relative resistance to adriamycin. Thus, R12,R65, and R200 cells were 12-, 65-, and 200-fold moreresistant to adriamycin than WT cells as measured bycell growth cytotoxicity assays. We have previously re-ported that the resistant MCF-7 cells display increasedPgp function, as measured by increased cellular effluxof Pgp substrates [48]. In agreement with previousfindings in other cell lines [5, 7, 14] we observedthat in multidrug resistant MCF-7 cells, increased drug

Table 1. IC50 (mM) and relative MDR1/Pgp densitometryvalues in MCF-7 cell lines

Treatment WT R12 R65 R200

IC50

Adriamycin 0.004 0.050 0.258 0.880Taxol 0.010 nd 0.500 1.500Vinblastine 0.006 0.150 0.500 0.500

Relative levels

Pgp 1.0 6.0 13.5 28.0MDR1mRNA 1.0 3.0 9.9 19.0MDR1gene 1.0 1.9 3.1 3.9

IC50 values were calculated from growth inhibition assaysof cell cultures exposed to increasing concentrations ofdrug. Relative levels of Pgp,MDR1 mRNA, andMDR1gene copy number, were determined by densitometryof western, northern, and southern blots, respectively.nd = not determined.

Figure 1. Analysis of c-jun and c-fos mRNA levels fromdrug-sensitive and drug-resistant MCF-7 cell lines. A, Repre-sentative RT-PCR for c-fosand c-jun. RT-PCR was performed ontotal cell RNA and normalized with G-3PDH. B, Densitometryvalues from RT-PCR analysis.n = 4 ± SE.

resistance correlated with increased levels ofMDR1mRNA and Pgp protein. Although Southern blot anal-ysis revealed a modest increase (4-fold) inMDR1genecopy number in highly drug resistant cells, this in-crease is not enough to account for the relatively largeincreases in mRNA (19-fold) and Pgp (28-fold) thatwe observed (Table 1). Using RT-PCR analysis, wefurther observed that increased levels of both c-jun andc-fostranscripts in the MDR cells correlated well withincreased resistance (Figure 1).

Increased AP-1 binding activity in nuclear extracts ofmultidrug resistant MCF-7 cells

To determine whether the increased c-jun and c-fosex-pression seen in the MDR cells could have a functionaleffect on AP-1 binding activity, we used EMSA to de-termine the DNA binding capacity of nuclear extractsfrom WT and drug-resistant MCF-7 cells. An oligonu-cleotide probe containing sequences corresponding toa consensus AP-1 binding site [42] was incubated withnuclear extracts from WT, R12, R65, and R200 cells.The probe formed a shifted complex that was identi-cal in mobility to a complex formed from incubations


Figure 2. EMSA comparing AP-1 binding activities of WT andMDR MCF-7 cells. Nuclear extracts from WT, R12, R65, andR200 cell lines were incubated with a consensus AP-1 probe.Recombinant human c-Jun (rhJun), was used as a positive con-trol. The MDR cells demonstrated increased AP-1 binding of theJun/Junhomodimer to the probe.

with recombinant human c-Jun (Figure 2) and that wasspecifically competed by cold probe or by antibod-ies to c-Jun (not shown). A positive correlation wasobserved between increased drug resistance and in-creased binding activity of Jun/Jun homodimers to theprobe. The extracts from highly resistant MCF-7 cells(R200) demonstrated an approximately 8-fold increasein Jun/Jun binding activity to the probe when comparedto extracts from the drug-sensitive WT cells. The bind-ing activities of the other observed complexes, whichwere not identified, did not change significantly in ex-

tracts from drug-resistant cells and therefore do notappear to be associated with increased drug resistance.

Increased AP-1 binding to a partial MDR1 promotersequence in MDR cells

Since sequence differences in the core and flankingregions of AP-1 sites are known to affect the bindingaffinities of different AP-1 forms [42], we wanted todetermine if the near-consensus AP-1 site located inthe MDR1 proximal promoter (Figure 3) was func-tional with regard to AP-1 binding. Results of EMSAusing an oligonucleotide probe containing the actualsequences from theMDR1promoter(−123 to−108)clearly demonstrated that nuclear extracts from MDRMCF-7 cells had increased binding activity of Jun/Junhomodimers to the probe when compared to WT cells(Figure 4A). These results were consistent with theEMSA data obtained using the consensus AP-1 probe(Figure 2) and with the increased levels of c-Juntranscripts seen in MDR cells (Figure 1).

To determine the specificity of the binding activityobserved in the EMSA, binding reactions with nuclearextracts from R65 cells were performed in the pres-ence of up to 40-fold molar excess of unlabeled probe.Results showed that the labeled probe was competedaway from the shifted complexes, while a 40-fold molarexcess of a nonspecific oligonucleotide did not com-pete the probe from the complex (Figure 4B). Takentogether, the data demonstrated that AP-1 complexescould recognize and bind specificMDR1proximal pro-moter sequences and suggested that binding of theJun/Jun form of AP-1 may be required to enhance genetranscription.

Functional analysis of increased AP-1 bindingactivity in MDR MCF-7 cells

Transient transfection assays using AP-1 reporter con-structs (Figure 5) were employed to determine whetherthe increased AP-1 binding activity observed in theMDR cells could affect AP-1 mediated gene expres-sion. We observed that luciferase expression was sig-nificantly increased in MDR cells transfected with theconstruct that contained the AP-1 consensus bindingsite (pLucC) when compared to expression levels intransfected WT cells (Figure 6A). Luciferase expres-sion from the parent vector (pGL3) did not changesignificantly when transfected into the different celllines, indicating that AP-1 mediated transcription isspecifically activated in the MDR cells, but not in WTcells. To determine whether theMDR1proximal pro-moter sequences that contain the AP-1 binding site


Figure 3. Schematic of the humanMDR1proximal promoter. TheMDR1cDNA sequence is represented by the solid line, with the majortranscription start site assigned+1. Cis-acting consensus binding sites are indicated by symbols with the acronym for the binding factoror sequence above. The lower portion of the figure shows an expanded view of the AP-1/CAAT box sequence that was used as an EMSAprobe (Actual), and is aligned with the consensus AP-1 binding oligonucleotide probe (Consensus). The AP-1 binding site is highlightedand the single base mismatch within the site is indicated by an asterisk.

could affect gene expression in drug-resistant cells,pLucA (Figure 5) was transfected into MCF-7 WT andMDR cells and the relative luciferase activity was de-termined in each cell type. Expression from the partialMDR1 promoter was significantly higher in the mul-tidrug resistant cells (about 2.5-fold) than in the WTcells, while basal expression from the parent vectorremained unchanged in all transfected cell lines (Fig-ure 6B). These results suggested that the endogenousMDR1promoter may be responsive to AP-1 activation.

Since we had observed that MDR MCF-7 cellsoverexpressed c-jun (Figure 1) and carried increasedJun/Jun DNA binding capacity (Figures 2 and 4), weexamined whether overexpression of exogenous c-juncould activate expression from AP-1 responsive pro-moter constructs. WT MCF-7 cells were co-transfectedwith either of our luciferase reporter constructs andthe c-jun expression vector pCMV-jun [39]. As inthe endogenously c-jun overexpressing multidrug re-sistant cells, we found that overexpression of c-junin WT cells resulted in a dose-dependent increase inthe transcriptional activation from both theMDR1par-tial promoter-containing construct (pLucA) and theAP-1 consensus construct (pLucC) (Figure 7). Theseresults confirmed that theMDR1partial promoter con-struct was responsive to increased levels of c-jun andimplicated AP-1 as a possible mediator ofMDR1overexpression in drug resistant human breast cancercells.

Regulation of MDR1 expression by c-jun in MCF-7wild type cells

To determine whether increased c-junexpression couldenhanceMDR1expression, we transfected WT MCF-7

cells with pCMV-jun and examined bothMDR1 ex-pression and Pgp production. Using RT-PCR analysisof total cellular RNA, we observed a marked increasein the level ofMDR1 mRNA in the cells transfectedwith c-jun when compared to the mock-transfected orcontrol cells (Figure 8A). We also observed a cor-responding increase in Pgp protein levels in c-juntransfected cells by using western immunoblot anal-ysis (Figures 8B and 8C). Transfection of R65 cellswith pCMV-jun did not result in a detectable increasein MDR1expression or Pgp production, probably dueto the high Pgp levels already exhibited in these cells(data not shown). These results demonstrated that c-Juncould enhanceMDR1expression in WT human breastcancer MCF-7 cells.

Discussion

The complex molecular mechanisms underlying thetranscriptional regulation ofMDR1 in multidrug re-sistant tumor cells allow for its activation in responseto a wide variety of environmental stimuli [13–16].Several transcription factors have been demonstratedto mediateMDR1expression in tumor cells [17–22]. Itis therefore likely that several separate signal pathwaysconverge to activateMDR1 and that several coordi-nately regulated transcription factors, perhaps actingsynergistically, are needed for high level expression ofMDR1.

Since data from several previous reports are con-sistent with the hypothesis that the transcription factorAP-1 may be important in the development of MDR, weinvestigated the role of AP-1 in the regulation ofMDR1expression in a series of multidrug resistant human


Figure 4. A, EMSA comparing AP-1 binding activities of WT and MDR MCF-7 cells. The indicated nuclear extracts were incubatedwith anMDR1partial promoter probe. B, EMSA of R65 nuclear protien binding. Lane 1; recombinant human c-Jun. Lane 2; R65 nuclearextract. Lanes 3–5; R65 nuclear extract with increasing molar excess of unlabeled probe. Lane 6; 40-fold molar excess of unlabelednonspecific (NS) oligonucleotide. * indicates unidentified complexes that are specifically competed by unlabeled probe.

breast cancer MCF-7 cells. We observed constitutiveoverexpression ofMDR1, c-Jun, and c-fos transcriptsin the multidrug resistant cells when compared to thedrug-sensitive parental MCF-7 cells. Our data thussupport previous reports [28, 54] correlating increasedlevels of AP-1 component proteins with increased drugresistance.

Other studies have shown that AP-1 regulatesMDR1homologs in both mouse and hamster cells [52,53] from promoter sequences that are highly conserved

in the human gene. Moreover, constitutive overexpres-sion of both c-fos [28] and c-Jun [54] and increasedAP-1 binding activity [55] have been observed in sev-eral human multidrug resistant cell lines, while reducedAP-1 binding activity has been associated with in-creased drug sensitivity in others [56]. In addition,protein kinase activators upstream from AP-1 in thestress response signal transduction pathway have beenfound to increaseMDR1 promoter activity [31, 57],and TPA treatment, a known activator of AP-1, has


Figure 5. Structure of luciferase reporter constructs used in transient transfection assays. Restrction sites and the SV40 early promoterregion of the parent vector (pGL3) are labeled, and the inserted AP-1 binding sites of the test constructs are indicated by circles. Theplasmid construct containing copies of the actualMDR1promoter region was designated pLucA. The plasmid construct containing copiesof the consensus AP-1 binding site was designated pLucC.

been shown to increaseMDR1 expression in severalhuman tumor cell lines [30], although this increase isat least partly mediated by EGR1 [23].

EMSA experiments performed using either con-sensus AP-1 binding sites or actualMDR1 promotersequences(−123 to −108) as probes demonstratedthat the elevated levels of component AP-1 proteinsfound in MDR MCF-7 cells resulted in the increasedbinding activity of Jun/Jun homodimers and the bind-ing of other unidentified site-specific complexes. Theseresults suggest that AP-1 may be important in theregulation ofMDR1 in human breast cancer MCF-7cells. The existence of an AP-1 binding site within thehumanMDR1 promoter was identified soon after thesequence was cloned [6]. A detailed examination ofthis promoter sequence (Figure 3) revealed overlap-ping AP-1 and CAAT box binding elements. Similarsequence overlaps of AP-1 with other binding siteshave been found in other promoter regions, such as theosteocalcin gene [49]. This sequence structure suggestsa possible mechanism for modulating transactivationfrom the site by coordinate occupancy competition [50]between different AP-1 forms and CAAT box enhancerbinding proteins. The presence of other specific com-plexes binding to our promoter probes (Figures 2 and 4)supports this possibility, and experiments to determinethe identity and competitive interactions of these com-

plexes are currently under investigation. Alternatively,the ability of AP-1 to activateMDR1expression couldbe regulated by a separate transcriptional repressor sitein drug-sensitive cells that becomes de-repressed indrug resistant cells [51].

Transfection assays usingMDR1 promoter dele-tion constructs that remove the putative AP-1 site haveyielded contrary results, either increasing transcrip-tional activity 4–5 fold (17) or decreasing activity 4–5fold (26). These data suggest that the−123 to−108region of theMDR1promoter can act as either a tran-scriptional repressor or activator, depending upon whatfactor(s) occupies the site. We offer evidence that AP-1 binding to this region enhancesMDR1 transcriptionin MCF-7 cells. The results from our transient trans-fection assays demonstrate that the−123 to −108promoter region is responsive to increased levels ofc-Jun from either endogenous (resistant cells) or ex-ogenous (co-transfected) sources (Figures 6 and 7).Although the transient early induction of c-jun in drug-sensitive cells in response to stress, such as cytotoxicdrug treatment, would be expected to involve a differentpathway than that associated with the establishmentof long-term drug resistance, our data indicate thatconstitutive activation of AP-1 may be a key factorcontributing to elevated levels ofMDR1expression indrug-resistant cells. We observed increased transcrip-


Figure 6. A, Enhanced AP-1 mediated expression indrug-resistant MCF-7 cells. WT and MDR MCF-7 cellswere transfected with the consensus AP-1 promoter constructpLucC and lysates assayed for luciferase activity. Transfectionswere performed in triplicate and data representing the meanof three determinations±SE is given. Luciferase activity wasnormalized toβ-gal activity to control for transfection efficiency.B, Enhanced luciferase expression in drug-resistant MCF-7 celllines transfected withMDR1 partial promoter constructs. WTand MDR MCF-7 cells were transfected with the partialMDR1promoter construct pLucA and lysates assayed for luciferaseactivity as described above.

Figure 7. Enhanced luciferase expression in WT MCF-7 cellsco-transfected with c-junand AP-1 reporter construct. Luciferaseactivity was determined as described in Figure 6, with appropriateamounts of pCMV vector DNA added to co-transfections to keepthe total amount of transfecting DNA equal. Background valuesfrom the parent vector, pGL3, were subtracted from each datapoint to correct for nonspecific expression.

tional activity fromMDR1partial promoter constructsin the drug-resistant cell lines and in WT cells tran-siently co-transfected with a c-Jun expression vector.While these results suggest that AP-1 binding sitepresent in the proximal promoter region ofMDR1canserve as a transcriptional enhancer element, we cannoteliminate the possibility that AP-1 binding to sequencesmore distal to theMDR1 transcription start site couldalso modulate gene expression.

Our transfection results contrast with those ob-tained by Chin et al. [21], in which co-transfectionof an MDR1CAT reporter construct with c-jun didnot cause increased expression. Several factors mayaccount for these differences, including different celllines, different reporter vectors, and different lengthpromoter sequences, the latter of which may have in-cluded repressor elements thought to be contained inthe MDR1 promoter region [17, 51]. Our relativelyshortMDR1promoter construct was designed to elim-inate any enhancer or repressor effects that could beattributed to DNA binding sites that flank the AP-1region of the natural promoter. Thus, our constructallowed us to observe luciferase expression that wascontrolled specifically by the AP-1 site.

The ability of human tumor cells to acquire mul-tidrug resistance has led others to suggest that activatedoncogenes may help regulateMDR1 expression [21].Our results from c-jun transfected WT cells demon-strate that c-Jun can activate expression ofMDR1andincrease levels of Pgp, indicating that expression ofMDR1 may be coordinately regulated with cellular


Figure 8. RNA and protein analysis of c-jun transfected WT MCF-7 cells. A, RT-PCR analysis ofMDR1transcript levels from transfectedMCF-7 cells. Total RNA from the indicated cell treatments was subjected to RT-PCR and separated on an 8% acrylamide gel. Lane 1;untreated MCF-7 cells. Lane 2; cells mock transfected with pCMV vector. Lane 3; cells transfected with pCMV-jun. Relative levels ofMDR1 were normalized to G-3PDH levels for all samples. B, Western blot analysis of Pgp protein levels in c-jun transfected WT cells.40mg of a crude membrane fraction was separated by SDS-PAGE and detected by the ECL method. C, Western blot of c-jun proteinlevels in c-jun transfected WT cells. 10mg of a cytosolic fraction was separated by SDS-PAGE and detected by the ECL method.


events triggered by the activation of oncogenes suchas c-jun. Taken together, this data offers evidencethat AP-1 activity levels affectMDR1 expression inmultidrug-resistant human tumor cells. These resultsprovide new information on the regulatory mechanismsinvolved in MDR1 expression and may provide newstrategies for MDR reversal in human breast cancercells.

Acknowledgements

This project has been funded in whole or in partwith Federal funds from the National Cancer Institute,National Institutes of Health, under Contract No. NO1-CO-56000. The content of this publication does notnecessarily reflect the views or policies of the Depart-ment of Health and Human Services, nor does mentionof trade names, commercial products, or organizationsimply endorsement by the U.S. Government.

References

1. Ueda K, Cardarelli C, Gottesman MM, Pastan I: Expressionof a full length cDNA for the humanMDR1 gene confersresistance to colchicine, doxorubicin and vinblastine. ProcNatl Acad Sci USA 84: 3004–3008, 1987

2. Pastan I, Gottesman MM, Udea K, Lovelace E, RutherfordAV, Willingham MC: A retrovirus carrying anMDR1cDNAconfers multidrug resistance and polarized expression of P-glycoprotein in MDCK cells. Proc Natl Acad Sci USA 85:4486–4490, 1988

3. Goldstein LJ, Galski H, Fojo A, Willingham M, Lai SL,Gazdar A, Pirker R, Green A, Crist W, Brodeur GM, LieberM, Cossman J, Gottesman MM, Pastan I: Expression of amultidrug resistance gene in human cancers. J Natl CancerInst 81: 116–124, 1989

4. Riordan JR, Deuchars K, Kartner N, Alon N, Trent J, LingV: Amplification of P-glycoprotein genes in multidrug-resistant mammalian cell lines. Nature 316: 817–819,1985

5. Shen DW, Fojo AT, Roninson IB, Chin JE, Richert N, PastanI, Gottesman MM: Human multidrug-resistant cell lines:increasedMDR1expression can precede gene amplification.Science 232: 643–645, 1986

6. Udea K, Pastan I, Gottesman MM: Isolation and sequence ofthe promoter region of the human multidrug resistance (P-glycoprotein) gene. J Biol Chem 262: 17432–17436, 1987

7. Burt RK, Thorgeirsson SS: Coinduction ofMDR1multidrug-resistance and cytochrome P-450 genes in ratliver by xenobiotics. J Natl Cancer Inst 80: 1343–1386,1988

8. Cornwell MM: The human multidrug resistance gene: se-quences upstream and downstream of the initiation siteinfluence transcription. Cell Growth Differ 1: 607–615,1990

9. Kohno K, Tanimura H, Sato S, Nakayama Y, Makino Y,Wada M, Fojo AT, Kuwano M: Cellular control of humanmultidrug resistance 1 (MDR1) gene expression in absenceand presence of gene amplification in human cancer cells.J Biol Chem 269: 20503–20508, 1994

10. Madden MJ, Morrow CS, Nakagawa M, Goldsmith ME,Fairchild CF, Cowan KH: Identification of 5′ and 3′ se-quences involved in the regulation of transcription of thehumanMDR1genein vivo. J Biol Chem, 268: 8290–8297,1993

11. Goldsmith ME, Madden MJ, Morrow CS, Cowan KH: A Y-box consensus sequence is required for basal expression ofthe human multidrug resistance (MDR1) gene. J Biol Chem268: 5856–5860, 1993

12. Ohga T, Koike K, Ono M, Makino Y, Itagaki Y, TanimotoM, Kuwano M, Kohno K: Role of the human Y box-bindingprotein YB-1 in cellular sensitivity to the DNA-damagingagents cisplatin, mitomycin C, and ultraviolet light. CancerRes 56: 4224–4228, 1996

13. Chin KV, Tanaka S, Darlington G, Pastan I, Gottesman MM:Heat shock and arsenite increase expression of the multidrugresistance (MDR1) gene in human carcinoma cells. J BiolChem 265: 221–226, 1990

14. Bates SE, Mickley LA, Chen YN, Richert N, Rudick J,Biedler JL, Fojo AT: Expression of a drug resistance genein human reuroblastoma cell lines: modulation by retinoicacid-induced differentiation. Mol Cell Biol 9: 4337–4344,1989

15. Uchiumi T, Dohno K, Tanimura H, Matsuo K, Sato S,Uchida Y, Kuwano M: Enhanced expression of the hu-man multidrug resistance 1 gene in response to UV lightirradiation. Cell Growth Differ 4: 147–157, 1993

16. Kohno K, Sato S, Takano H, Matsuo K, Kuwano M: The di-rect activation of human multidrug resistance gene (MDR1)by anticancer agents. Biochem Biophys Res Commun 165:1415–1421, 1989

17. Cornwell MM, Smith DE: SP-1 activates theMDR1 pro-moter through one of two distinct GC-rich regions that mod-ulate promoter activity. J Biol Chem 268: 19505–19511,1993

18. Glazer RI, Rohlff C: Transcriptional regulation of multidrugresistance in breast cancer. Breast Cancer Res Treat 31: 263–271, 1994

19. Kohno K, Sato S, Uchiumi T, Takano H, Kato S, Kuwano M:Tissue-specific enhancer of the human multidrug resistance(MDR1) gene. J Biol Chem 265: 19690–19696, 1990

20. Combates NJ, Rzepka RW, Chen YP, Cohen D: NF-IL6, amember of the C/EBP family of transcription factors, bindsand trans-activates the humanMDR1gene promoter. J BiolChem 269: 29715–29719, 1994

21. Chin KV, Ueda K, Pastan I, Gottesman MM: Modulation ofactivity of the promoter of the humanMDR1gene by Rasand p53. Science 255: 459–462, 1992

22. Goldsmith ME, Gudas JM, Schneider E, Cowan KH: Wildtype p53 stimulates expression from the human multidrugresistance promoter in a p53-negative cell line. J Biol Chem270: 1894–1898, 1995

23. McCoy C, Smith DE, Cornwell MM: 12-O-tetradecan-oylphorbol-13-acetate activation of theMDR1promoter ismediated by EGR1. Mol Cell Biol 15: 6100–6108, 1995

24. Angel P, Imagawa M, Chiu R, Stein B, Imbra RJ, RahmsdorfHJ, Jonat C, Herrlich P, Karin M: Phorbol ester-induciblegenes contain a common cis element recognized by TPA-modulated trans-acting factor. Cell 49: 729–739, 1987

25. Angel P, Karin M: The role of Jun, Fos and the AP-1complex in cell proliferation and transformation. BiochimBiophys Acta 1072: 129–157, 1991

26. Ogura M, Takatori T, Sugimoto Y, Tsuruo T: Identificationand characterization of three DNA-binding proteins on the


promoter of the humanMDR1 gene in drug-sensitive andresistant cells. Jpn J Cancer Res 82: 1151–1159, 1991

27. Ritke MK, Bergoltz VV, Allan WP, Yalowich JC: Increasedc- jun/AP-1 levels in etoposide-resistant human leukemiaK562 cells. Biochem Pharmacol 48: 525–533, 1994

28. Bhushan A, Abramson R, Chiu JF, Tritton TR: Expressionof c- fosin human and murine multidrug-resistant cells. MolPharmacol 42: 69–74, 1992

29. Yu R, Shtil AA, Tan TH, Roninson IB, Kong AT:Adriamycin activates c-Jun N-terminal kinase in humanleukemia cells: a relevance to apoptosis. Cancer Lett 107:73–81, 1996

30. Chaudhary PM, Roninson IB: Activation ofMDR1 (P-glycoprotein) gene expression in human cells by proteinkinase-C agonists. Oncol Res 4: 281–290, 1992

31. Kim SH, Lee SH, Kwak NH, Kang CD, Chung BS: Effectsof the activated Raf protein kinase on the human multidrugresistance 1 (MDR1) gene promoter. Cancer Lett 98: 199–205, 1996

32. Burt RK, Garfield S, Johnson K, Thorgeirsson SS: Trans-formation of rat liver epithelial cells with v-H-ras or v-rafcauses expression ofMDR1, glutathione-S-transferase-Pand increased resistance to cytotoxic chemicals. Carcino-genesis 9: 2329–2332, 1988

33. Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, PastanI, Willingham MC: Cellular localization of the multidrugresistance gene product P-glycoprotein in normal humantissues. Proc Natl Acad Sci USA 84: 7735–7738, 1987

34. Yeh GC, Lopaczynska J, Poore CM, Phang JM: Anew functional role for P-glycoprotein: efflux pump forbenzo(a)pyrene in human breast cancer MCF-7 cells. Can-cer Res 52: 6692–6695, 1992

35. van Asperen J, Schinkel AH, Beijnen JH, Nooijen WJ, BorstP, van Tellingen O: Altered pharmacokinetics of vinblastinein Mdr1a P-glycoprotein-deficient mice. J Natl Cancer Inst88: 994–999, 1996

36. Skehan P, Storeng R, Scudiero D, Monks A, McMahon J,Vistica D, Warren JT, Bokesch H, Kenney S, Boyd MR:New colormetric cytotoxicity assay for anticancer-drugscreening. J Natl Cancer Inst 82: 1107–1112, 1990

37. Berger SL, Chirgwin JM: Isolation of RNA. MethodsEnzymol 180: 3–13, 1989

38. Chen CJ, Chin JE, Ueda K, Clark DP, Pastan I, GottesmanMM, Roninson IB: Internal duplication and homology withbacterial transport proteins in theMDR1 (P-glycoprotein)gene from multidrug resistant human cells. Cell 47: 381–389, 1986

39. Alani R, Brown P, Binetruy B, Dosaka H, Rosenberg RK,Angel P, Karin M, Birrer M: The transactivating domain ofthe c-Jun proto-oncoprotein is required for cotransformationof rat embryo cells. Mol Cell Biol 11: 6286–6295, 1991

40. Brown PH, Chen TK, Birrer MJ: Mechanism of action of adominant-negative mutant of c-jun. Oncogene 8: 791–799,1993

41. Sambrook J, Fritsch EF, Maniatis TE (eds): MolecularCloning: A Laboratory Manual. Cold Spring Harbor, NY;Cold Spring Harbor Laboratory, 1989

42. Ryseck RP, Bravo R: c-Jun, Jun B, and Jun D differ in theirbinding affinities to AP-1 and CRE consensus sequences:effects ofFosproteins. Oncogene 6: 533–542, 1990

43. Sanger F, Nicklen S, Coulson AR: DNA sequencing withchain-terminating inhibitors. Proc Natl Acad Sci 74: 5463–5467, 1977

44. Rosenthal N: Identification of regulatory elements of clonedgenes with functional assays. Methods Enzymol 152: 704,1987

45. Bradford MM: A rapid and sensitive method for the quanti-tation of microgram quantities of protein utilizing the prin-ciple of protein-dye binding. Anal Biochem 72: 248–254,1976

46. Dignam JD, Lebowitz RM, Roeder RG: Accurate transcrip-tion initiation by RNA polymerase II in a soluble extractfrom isolated mammalian nuclei. Nucleic Acids Res 11:1475–1489, 1983

47. Oda Y, Rosek I, Radig K, Wagemann W, Mittler U, Roess-ner A: Expression ofMDR1/P-glycoprotein and multidrugresistance-associated protein in childhood solid tumors.Virchows Arch 430: 99–105, 1997.

48. Phang JM, Poore MC, Lopaczynska J, Yeh GC: Flavanol-stimulated efflux of 7,12-dimethylbenzanthracene in mul-tidrug resistant breast cancer cells. Cancer Res 53: 5977–5981, 1993

49. Owen TA, Bortell R, Yocum SA, Smock S, Zhang M, AbateC, Shalhoub V, Aronin N, Wright KL, VanWijnen AJ, SteinJL, Curran T, Lian JB, Stein GS: Coordinate occupancy ofAP-1 sites in the vitamin D- responsive and CCAAT boxelements by Fos–Jun in the osteocalcin gene; model forphenotype suppression of transcription. Proc Natl Acad Sci87: 9990–9994, 1990

50. Masquiller D, Sassone-Corsi P: Transcriptional cross-talk:nuclear factors CREM and CREB bind AP-1 sites and in-hibit activation by c-jun. J Biol Chem 267: 22460–22466,1992

51. Gant TW, Silverman JA, Thorgeirsson SS: Regulationof P-glycoprotein gene expression in hepatocyte culturesand liver cells by a trans-acting transcriptional repressor.Nucleic Acids Res 20: 2841–2846, 1992

52. Yu L, Cohen D, Piekarz RL, Horwitz SB: Three distinctnuclear protein binding sites in the promoter of the murinemultidrug resistanceMDR1b gene. J Biol Chem 268: 7520–7526, 1993

53. Teeter LD, Eckersberg T, Tsai Y, Kuo MT: Analysis of theChinese hamster P-glycoprotein/multidrug resistance genepgp 1 reveals that the AP-1 site is essential for full promoteractivity. Cell Growth Differ 2: 429–437, 1991

54. Ritke MK, Yalowich JC: Altered gene expresion in humanleukemia K562 cells selected for resistance to etoposide.Biochem Pharmacol 46: 2007–2020, 1993

55. Chen TK, Smith LM, Gebhardt DK, Birrer MJ, Brown PH:Activation and inhibition of the AP-1 complex in humanbreast cancer cells. Mol Carcinogen 15: 215–226, 1996

56. Rohlff C, Safa B, Rahman A, Cho-Chung YS, KleckerRW, Glazer RI: Reversal of resistance to Adriamycin by 8-Cl-cAMP in adriamycin resistant HL-60 leukemia cells isassociated with reduction of type I cAMP-dependent proteinkinase and cAMP response element binding protein DNAbinding activity. Mol Pharmacol 43: 372–379, 1993

57. Cornwell MM, Smith DE: A signal transduction pathwayfor activation of theMDR1 promoter involves the proto-oncogene c-raf kinase. J Biol Chem 268: 15347–15350,1993

Address for offprints and correspondence:Grace Chao Yeh,Cellular Defense and Carcinogenesis Section, BRL, Bldg. 560,Rm. 12-91, NCI-FCRDC, Frederick, MD 21702, USA;Tel:301-846-5369;Fax: 301-846-6093

increased ap-1 activity in drug resistant human breast cancer mcf-7 cells

Documents