[CANCER RESEARCH 59, 831–842, February 15, 1999 ... · proliferation and cancer progression. The...

[CANCER RESEARCH 59, 831–842, February 15, 1999]

Peripheral-Type Benzodiazepine Receptor (PBR) in Human Breast Cancer:Correlation of Breast Cancer Cell Aggressive Phenotype with PBRExpression, Nuclear Localization, and PBR-mediated CellProliferation and Nuclear Transport of Cholesterol1

Matthew Hardwick, Djamil Fertikh, Martine Culty, Hua Li, Branislav Vidic, and Vassilios Papadopoulos 2, 3

Departments of Cell Biology [M. H., D. F., M. C., H. L., B. V., V. P.] and Pharmacology [V. P.], and the Lombardi Cancer Center [M. C., V. P.], Georgetown University MedicalCenter, Washington, DC 20007

ABSTRACT

Aberrant cell proliferation and increased invasive and metastatic be-havior are hallmarks of the advancement of breast cancer. Numerousstudies implicate a role for cholesterol in the mechanisms underlying cellproliferation and cancer progression. The peripheral-type benzodiazepinereceptor (PBR) is an Mr 18,000 protein primarily localized to the mito-chondria. PBR mediates cholesterol transport across the mitochondrialmembranes in steroidogenic cells. A role for PBR in the regulation oftumor cell proliferation has also been shown. In this study, we examinedthe expression, characteristics, localization, and function of PBR in abattery of human breast cancer cell lines differing in their invasive andchemotactic potential as well as in several human tissue biopsies. Expres-sion of PBR ligand binding and mRNA was dramatically increased in thehighly aggressive cell lines, such as MDA-231, relative to nonaggressivecell lines, such as MCF-7. PBR was also found to be expressed at highlevels in aggressive metastatic human breast tumor biopsies comparedwith normal breast tissues. Subcellular localization with both antibodiesand a fluorescent PBR drug ligand revealed that PBR from the MDA-231cell line as well as from aggressive metastatic human breast tumor biop-sies localized primarily in and around the nucleus. This localization is indirect contrast to the largely cytoplasmic localization seen in MCF-7 cells,normal breast tissue, and to the typical mitochondrial localization seen inmouse tumor Leydig cells. Pharmacological characterization of the recep-tor and partial nucleotide sequencing of PBR cDNA revealed that theMDA-231 PBR is similar, although not identical, to previously describedPBR. Addition of high affinity PBR drug ligands to MDA-231 cellsincreased the incorporation of bromodeoxyuridine into the cells in adose-dependent manner, suggesting a role for PBR in the regulation ofMDA-231 cell proliferation. Cholesterol uptake into isolated MDA-231nuclei was found to be 30% greater than into MCF-7 nuclei. High-affinityPBR drug ligands regulated the levels of cholesterol present in MDA-231nuclei but not in MCF-7. In addition, the PBR-dependent MDA-231 cellproliferation was found to highly correlate (r 5 20.99) with the PBR-mediated changes in nuclear membrane cholesterol levels. In conclusion,these data suggest that PBR expression, nuclear localization, and PBR-mediated cholesterol transport into the nucleus are involved in humanbreast cancer cell proliferation and aggressive phenotype expression, thusparticipating in the advancement of the disease.

INTRODUCTION

The PBR4 was originally discovered as an alternative binding sitefor the benzodiazepine diazepam (Valium; Ref. 1). PBR is a multim-eric receptor composed of theMr 18,000 receptor protein and theMr

34,000 voltage-dependent anion channel protein. TheMr 18,000 sub-unit is thought to be responsible for the binding of isoquinolines,whereas both subunits are required for the binding of benzodiazepines(2). Expression of PBR has been found in every tissue examined;however, it is most abundant in steroid-producing tissues such as theadrenals, testes, ovaries, placenta, and brain, where it is primarilylocalized in the outer mitochondrial membrane (3). In steroid-produc-ing tissues, PBR is known to regulate the transport of cholesterol fromthe outer to the inner mitochondrial membrane, the rate-determiningstep in steroid biosynthesis (4). Although the primary function of PBRis the regulation of steroidogenesis, its existence in nonsteroidogenictissues as well as in other cellular compartments including the plasmamembrane (3) suggests that there may be other roles for PBR.

In the early 1980s it was shown that diazepam (Valium), a ligandfor both the central (GABAA) and PBR receptors, induced murineFriend erythroleukaemia (MEL) cell differentiation and inhibited 3T3cell proliferation at micromolar concentrations (5). A series of 15benzodiazepines, including diazepam, also inhibited thymoma cellproliferation at micromolar concentrations (6) and further suggestedthat this effect was mediated by PBR rather than the GABAA receptor.Stimulation of cell proliferation was shown to occur when gliomacells were incubated with nanomolar concentrations of Ro5-4864 andPK 11195, both high-affinity PBR ligands (7). In this same study, itwas shown that Ro5-4864 and PK 11195 inhibited glioma cell pro-liferation at micromolar concentrations. Similar results were obtainedin testicular Leydig tumor cells (8). The effect of PK 11195, anexclusive ligand of PBR, provided unequivocal evidence that theeffects seen were mediated exclusively by PBR. Inhibition of cellproliferation by micromolar concentrations of Ro5-4864 and PK11195 have also been reported in astrocytes and V79 Chinese Hamsterlung cells (9, 10). In addition, micromolar concentrations of PBRligands have also been shown to inhibit growth factor-induced cellproliferation in both astrocytes and lymphoma cells (11, 12). Consid-ering the abundance of PBR in tumors (6, 7, 9–12) and the proposedcorrelation between PBR expression and human brain tumor aggres-sion (13, 14), we examined PBR expression and, if present, itsfunction in breast cancer.

Breast cancer is the most common neoplasm and the leading causeof cancer-related deaths for women in most developing countries (15).The American Cancer Society estimates that in 1998 nearly 180,000new cases of invasive breast cancer will be diagnosed among women,with over 43,000 deaths, in the United States alone. Two fundamental

Received 8/5/98; accepted 12/15/98.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported in part by Grant ES-07747 from the National Institute ofEnvironmental Health Sciences and SPORE Grant P50-CA-58185 from the NationalCancer Institute, NIH. This work was partially presented at the 88th Annual Meeting ofthe American Association for Cancer Research (San Diego, CA) and the Xth InternationalMeeting on Hormonal Steroids (Quebec City, Canada).

2 Supported by a Research Career Development Award HD-01031 from the NationalInstitute of Child Health and Human Development, NIH.

3 To whom requests for reprints should be addressed, at Department of Cell Biology,Georgetown University Medical Center, 3900 Reservoir Road, Washington, DC 20007.Phone: (202) 687-8991; Fax: (202) 687-1823; E-mail: [email protected].

4 The abbreviations used are: PBR, peripheral-type benzodiazepine receptor; hPBR,human PBR; FBS, fetal bovine serum; BrdUrd, 5-bromo-29-deoxyuridine; HRP, horse-radish peroxidase; DBI, diazepam binding inhibitor; SREBP, sterol regulatory elementbinding protein; SSC, sodium chloride/sodium citrate.

831

Research. on October 1, 2020. © 1999 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

questions in breast cancer research must be answered: (a) what are thechanges in cellular and molecular functions that account for thedevelopment and progression of breast cancer? and (b) how caninvestigators use what is known about the genetic and cellular changesin breast cancer patients to improve detection, diagnosis, prevention,treatment and follow-up care?

Tumor progression is a multistep process by which normal cellsgradually acquire more malignant phenotypes, including the ability toinvade tissues and form distal metastases, the primary cause of mor-tality in breast cancer. During this process, the “aberrant” expressionof a number of gene products may be the cause or the result oftumorigenesis. Considering that the first step of tumor progression iscell proliferation, a fundamental biological phenomenon common tonormal and “aberrant” development, it can be proposed that tumori-genesis and malignancy are related to the proliferative potential oftumoral cells.

In this report, we tested the hypothesis that the PBR is: (a) part ofthe changes in cellular and molecular functions that account for thedevelopment and progression of breast cancer; and (b) could be usedin the future as a tool to improve detection, diagnosis, prevention, andtreatment in breast cancer patients. Using a battery of well-character-ized breast cancer cell lines differing in their invasive and metastaticabilities (16, 17), we demonstrate that expression of PBR correlateswith the expression of breast cancer cell aggressive phenotype. Weconfirmed our findings in human biopsies from normal breast tissueand aggressive metastatic breast tumors. In agreement with the well-documented function of PBR in steroid synthesizing tissues, control ofcholesterol transport into mitochondria (4), we identified that PBR inaggressive breast tumor cells regulates cell proliferation and choles-terol transport into the nucleus. Statistical analysis reveals that thesetwo functions are highly correlated in aggressive breast tumor cells.

MATERIALS AND METHODS

Cell Culture. Human breast cancer cell lines (BT549, HS-578-T, MCF-7,ADR, MDA-231, MDA-435, MDA-468, T47D, and ZR-75-1) were obtainedfrom the Lombardi Cancer Center, Georgetown University Medical Center.MA-10 mouse Leydig tumor cells were a gift from Dr. Mario Ascoli (Univer-sity of Iowa) and were maintained in Waymouth’s MB752/1 medium supple-mented with 15% horse serum as described previously (18). All cell lines werecultured on polystyrene culture dishes (Corning, Corning, NY) and grown inDMEM supplemented with 10% FBS.

Radioligand Binding Assays.Cells were scraped from 150-mm culturedishes into 5 ml PBS, dispersed by trituration, and centrifuged at 5003 g for15 min. Cell pellets were resuspended in PBS and assayed for protein con-centration. [3H]PK 11195 binding studies on 50mg of protein from cellsuspensions were performed as described previously (2). Scatchard plots wereanalyzed by the LIGAND program (19). Specific binding of [3H]PK 11195(2.0 nM) to MDA-231 cells was measured in the presence or absence of theindicated concentrations of competing PBR ligands as described previously(2). IC50 estimation was performed using the LIGAND program (19).

Protein Measurement. Protein levels were measured by the Bradfordmethod (20) using the Bio-Rad Protein Assay kit (Bio-Rad Laboratories,Hercules, CA) with BSA as a standard.

Transmission Electron Microscopy. MDA-231, ADR, and MCF-7 cellscultured on 25-cm2 culture dishes were first washed with PBS for 5 min threetimes. The cells were then fixed with a solution of 1% paraformaldehyde, 2%glutaraldehyde, and 0.1M PBS for 15 min at room temperature and thenwashed three times with PBS. The cells were then embedded in Epon-aralditeand further processed as described previously (4).

RNA (Northern) Analysis. The levels of PBR mRNA from MDA-231,MCF-7, and ADR cells were compared by Northern blot analysis. Totalcellular RNA was isolated from cells grown on 150-mm culture dishes by theaddition of 4.5 ml RNAzol B (Tel-Test, Inc., Friendswood, TX) and 0.45 mlof chloroform. After vigorous shaking and centrifugation at 90003 g for 30min, the aqueous phase was transferred to a fresh tube and mixed 1:1 with

isopropanol (v:v), stored at220°C for 2 h, and centrifuged at 90003 g for 30min. The RNA pellet was then washed with 75% ethanol and centrifuged75003 g for 8 min. The pellet was then air dried and resuspended in formazol.RNA concentrations and purity were determined at 260/280 nm. Twentymg oftotal RNA from each cell line were run on 1% agarose gels containing 134-morpholinepropanesulfonic acid and 5.3% formaldehyde. Gels were thentransferred overnight to nylon membranes (S&S Nytran; Schleicher & Schuell,Keene, NH; Ref. 21). A 0.2-kb hPBR cDNA fragment (derived from thepCMV5-PBR plasmid vector containing the full-length hPBR kindly given byDr. Jerome Strauss, University of Pennsylvania, Philadelphia, PA) was radio-labeled with [a-32P]dCTP using a random primers DNA labeling system (LifeTechnologies, Inc., Gaithersburg, MD). The filter was first prehybridizedovernight at 68°C in 63 SSC, 0.5% SDS, and 100mg/ml denatured, frag-mented, salmon sperm DNA. After hybridization, the membrane was washedtwice with 23 SSC, 0.5% SDS for 10 min, once with 0.23 SSC, 0.5% SDSfor 30–60 min at room temperature, and once with 0.23 SSC, 0.5% SDS for 30min at 60°C. Autoradiography was performed by exposing the blots to X-OMATAR film (Kodak, Rochester, NY) at270°C for 4–48 h. Quantification of PBRmRNA was carried out using the SigmaGel software (Jandel Scientific, SanRafael, CA).

Partial cDNA Sequencing. PBR cDNAs were prepared from total MDA-231 and MCF-7 RNA using the Perkin-Elmer RT-PCR kit (Branchburg, NJ).PCR was performed on cDNAs using primers designed from the known humansequence (22). Labeling of PCR products was performed using the ABIPRISM Dye Terminator Cycle Sequencing Ready Reaction kit (Perkin-Elmer).Labeled PCR products were sequenced by the Lombardi Sequencing CoreFacility (Georgetown University Medical Center, Washington, DC).

Fluorescent Microscopy with the Fluorescent PBR Ligand Compound4. MA-10, MDA-231, MCF-7, and ADR cells were grown on glass coverslipsas described previously (23). Cells were then washed twice with sterile PBSand incubated for 45 min with 1mM compound 4, a fluorescent derivative ofthe PBR ligand FGIN-27, with or without a competing PBR ligand, FGIN-27,at a concentration of 100mM. After the incubation period, the cells werewashed with PBS and examined by fluorescent microscopy using an OlympusBH-2 fluorescence microscope.

Immunocytochemistry of MDA-231 Cells. MDA-231 cells were culturedovernight on eight-chambered SuperCell Culture Slides (Fisher Scientific,Pittsburgh, PA) at a concentration of;50,000 cells/chamber. Cells were thenfixed in 70% ethanol for 15 min at 4°C. After washing three times in distilledH20 for 2 min each, the fixed cells were incubated overnight at 4°C with eitheranti-PBR (24) or anti-DBI, a generous gift from Dr. A. S. Brown (Prince ofWales Hospital, Randwick, Australia); polyclonal antisera concentrations wereof 1:100, 1:200, 1:500, or 1:1000. The slides were washed three times in PBSfor 2 min each. Slides were then incubated at room temperature for 1 h withHRP-coupled goat anti-rabbit secondary antibody diluted 1:2000 in PBSsupplemented with 10% calf serum. After washing the slides three times inPBS for 2 min each, fresh H2O2 diluted 1:1000 with 3-amino-9-ethyl carbazolewas added, and slides were incubated for 1 h at37°C. The slides were thenrinsed in distilled H20 before mounting with Crystal/Mount.

Nuclear Transport of [3H]Cholesterol. Nuclei were isolated from MDA-231 and MCF-7 as described by Elangoet al. (25). Isolated nuclei wereresuspended in 1 ml of ice-cold PBS. [3H]Cholesterol uptake in MDA-231 andMCF-7 nuclei was examined in the presence of 6.7 nM (1, 2)[3H]cholesterol(50.0 Ci/mmol) and 3mg of nuclear protein for 60 min at 37°C (26). Sampleswere then centrifuged at 5003 g for 30 min, and pellets were washed in 500ml of ice-cold PBS. After a second centrifugation at 5003 g for 30 min, 200ml of 1.0 N NaOH was then added to the pellets and incubated overnight at37°C. After incubation, 200ml of 1.0N HCl was added, and samples weremixed vigorously. Three ml of scintillation cocktail were then added, andradioactivity was measured by liquid scintillation spectroscopy using a Wallac1409 Liquid Scintillation Counter.

BrdUrd Cell Proliferation Assays. MDA-231 cells were plated on 96-well plates at a concentration of;10,000 cells/well (24-h incubation) or;5,000 cells/well (48-h incubation) in DMEM supplemented with 0.1% FBS.The cells were then incubated in either 0.1% or 10% FBS with variousconcentrations of PK 11195 for either 24 or 48 h. Differences in cell prolif-eration were analyzed by measuring the amount of BrdUrd incorporationdetermined by the BrdUrd ELISA (Boehringer Mannheim, Indianapolis, IN).

832

PBR IN HUMAN BREAST CANCER



Determination of DNA Content with Hoechst 33258.MDA-231 cellswere plated on six-well plates at a concentration of;150,000 cells/well inDMEM supplemented with 10% FBS. After 24 h, cells were washed two timeswith PBS, and the medium was changed to DMEM supplemented with 0.1%FBS. After an additional 24 h, cells were treated with either 0, 10210, 1028,1026, and 1024 M PK 11195 for 48 h. After incubation, cells were againwashed two times with PBS and then solubilized with 200ml of 0.2% SDS inETN buffer (10 mM EDTA, 10 mM Tris-HCl, and 100 mM NaCl, pH 7.0) for15 min at 37°C. Determination of DNA content was accomplished with theHoechst 33258 dye as described previously (27).

Analysis of Human Biopsies.All human biopsies were obtained from theLombardi Cancer Center Tumor Bank and the Department of Pathology at theGeorgetown University Medical Center. Biopsies were analyzed and classifiedby the pathologists. Tumor biopsies were sectioned and then placed on glassslides. The slides were then heated at 56°C for 30 min. For specimens not inpolyester wax, the slides were treated with xylene two times for 5 min each atroom temperature. All slides were treated two times with 100% ethanol for 2min each at room temperature and then two times with 95% ethanol for 2 mineach at room temperature. After rinsing in running distilled H2O for 2 min, theslides were then treated with microwave fix (0.01M citrate buffer, pH 6.0) inthe microwave for 15 min at half-power and then 10 min at full power. Theslides were then left to cool for 2 h at room temperature. For immunohisto-chemistry with anti-PBR primary antibodies, tissue sections were treated witha 30% H2O2:methanol mixture (1:9 ratio) for 5 min at room temperature toneutralize endogenous peroxidase activity and then washed well with PBS.Primary antibody in 10% calf serum in PBS was added to sections at aconcentration of 1:250 at room temperature for 1 h. Secondary antibodyreactions using either HRP or FITC were carried out at concentrations of1:500, as described above. For PBR fluorescent labeling with compound 4(23), sections were treated with 13 1027 M compound 4 for 2 h at roomtemperature in a dark room and then viewed on an Olympus BH-2 fluorescentmicroscope.

Statistical Analysis. Comparison of multiple means was performed withInStat’s one-way ANOVA (GraphPad, Inc., San Diego, CA). AllF statisticsand Ps for one-way ANOVAs are provided in the text. Comparison ofindividual drug treatments to the control treatments was performed withSigmaPlot’s unpairedt test (Jandel Scientific). AllPs for unpairedt tests areprovided in the text.

RESULTS

Increased Expression of the PBR Correlates with IncreasedAggressive Phenotype in Human Breast Cancer Cell Lines.Initialstudies performed in MCF-7 human breast cancer cells indicated

extremely limited expression of PBR (Fig. 1). The data generatedwere insufficient for further Scatchard analysis. Before dismissing ouroriginal hypothesis, we decided to examine expression of the receptorin eight other human breast cancer cell lines using the PBR-specific,high-affinity ligand PK 11195. The results from these experimentsindicated that those cell lines with a more invasive and chemotacticpotential, such as HS-578-T and MDA-231, display dramaticallyincreased levels of PBR binding relative to nonaggressive cell linessuch as ZR-75-1, T47D, and MCF-7 (Table 1 and Fig. 1). Further-more, the MCF-7 Adriamycin-resistant derivative cell line, ADR,which expresses medium invasive and chemotactic potential as wellas intermediate levels of CD44, expressed;20–40% PBR bindingrelative to the MDA-231 cell line (Table 1). Scatchard analysis ofPBR binding in the MDA-231 and ADR cell lines further shows eachto have aBmax of 8.7 6 1.4 and 1.36 0.23 pmol/mg protein,respectively (Table 2). As mentioned previously, despite obtainingspecific PK 11195 binding in MCF-7 cells, the low binding levelswere inadequate for estimating binding characteristics using Scat-chard plot analysis (Table 2).

RNA (Northern) blot analysis was performed to determine whetherthe differences shown in PBR binding between the cell lines reflect

Fig. 1. The specific PBR binding characteristics of various human breast cancer celllines. PK 11195-specific binding was determined using increasing concentrations ofcellular protein for 50mg of protein from each of the indicated cell lines described inTable 1.Ps were determined using the unpairedt test provided by SigmaPlot software(Jandel Scientific).ppp, P , 0.05; pp, P , 0.01; NS, not significant.Bars,SE.

Table 1 Comparison of invasive characteristics of human breast cancer cell lines toPBR expression

The various characteristics of the human breast cancer cell lines described below arefrom Azzamet al. (16) and Cultyet al. (17). The invasive and chemotaxis assays weredetermined by quantifying the migration of cells in Boyden chamber assays usingfibroblast-conditioned medium as the chemoattractant. Chemoinvasion was studied usingpolycarbonate filters coated with a uniform layer of Matrigel constituting a barrier that thecells had to degrade to reach the filters and migrate through them. To determine thechemotactic behavior of the cells, the filters were coated with a thin layer of collagen IVthat promotes cell attachment and allows the free migration of the cells toward thegradient of fibroblast-conditioned medium. The relative amounts of PBR were determinedwith binding assays in which increasing concentrations of cellular protein were incubatedwith a constant level of [3H]PK 11195 (6 nM). Nonspecific binding was determined in thepresence of cold PK 11195. CD44 levels are relative among the cell lines listed.

Cell lineEstrogenreceptora Vimentina Invasionb Chemotaxisb CD44 PBRb

ZR-75-1 1 2 1 1 2 2T47D 1 2 1 1 1 1MCF-7 1 2 11 11 1 1MDA-435 2 1 111 11 111 1ADR 2 1 111 111 11 11BT549 2 1 11111 11111 1111 111MDA-468 2 6 1 111 1111 1111HS-578-T 2 6 11111 1111 11111 11111MDA-231 2 1 11111 11111 111 11111

a The presence or absence of estrogen receptor and vimentin are indicated by eithera 1 or 2, respectively.

b Invasion, chemotactic, and PBR binding were graded as the percentage of MDA-231values:2, not detectable;1, 0–20%;11, 20–40%;111, 40–60%;1111, 60–80%; and11111, .80%.

Table 2 PBR-binding characteristics of MDA-231, ADR, and MCF-7 cells

Scatchard analysis of MDA-231, ADR, and MCF-7 cells (50mg) was performed asdescribed previously (2). Unpairedt test comparison of the meanBmaxs for MDA-321 andADR PBRs gaveP 0.0062, suggesting that the differences between the twoBmaxs aresignificant. Similar comparison of the meanKDs of these cell lines also suggests signif-icant differences in their binding affinity for PK 11195 (P 5 0.017).

Cell line

PK 11195

KD (nM)Bmax

(pmol/mg protein)

MDA-231a 7.86 1.8b 8.76 1.4b

ADR 1.96 0.47b 1.36 0.23b

MCF-7 NDc NDc

a Quantification of Ro5-4864 binding sites (Bmax) and affinity (KD) through Scatchardanalysis could not be determined despite low levels of specific binding and significantdisplacement of PK 11195 by Ro5-4864 in MDA-231 cells.

b Errors are SE.c ND, not detectable. Scatchard analysis of the binding data could not be performed,

although low levels of specific binding could be seen, indicating the presence of PBR.

833




differential expression of PBR mRNA. As shown in Fig. 2, MDA-231cells express;20-fold more PBR mRNA than MCF-7 cells. Thisresult fits with the correlation between PBR expression and increasedaggressive behavior between these cell lines. The amount of PBRmRNA expressed in the ADR cell line does not conform to this,however. In fact, ADR cells express as much PBR mRNA as MDA-231 cells (Fig. 2). This seemingly anomalous result will be discussedlater.

Previous studies demonstrated that, in most tissues, PBR isprimarily localized to the mitochondria (3). To rule out the possi-bility that the differences between aggressive and nonaggressivehuman breast cancer cell lines are not due to differences in mito-chondrial content, morphometry analysis was performed on trans-mission electron micrographs of two of the extreme representativecell lines, MDA-231 and MCF-7 (data not shown). Numerous

morphological differences were found between the two cell lines,including differences in vacuole content and the presence of mys-terious dark bodies, which may reflect their differences in meta-bolic activity. Upon first observation, the micrographs appear toreveal that MDA-231 cells contain more mitochondria than MCF-7cells. Morphometric analysis, however, indicates that the largerMCF-7 mitochondria cover the same surface area/cell in the mi-crographs as do the MDA-231 mitochondria.

To further characterize the differences between these human breastcancer cell lines, subcellular localization was carried out using com-pound 4, the fluorescent derivative of FGIN-27, a specific PBR ligand(23). PBR has been shown previously to localize primarily to the outermitochondrial membrane in MA-10 mouse tumor Leydig cells, thecell line used to characterize the only known function of PBR (3). InMA-10 cells, compound 4 fluorescent labeling is localized to thecytoplasm, presumably to the mitochondria (Fig. 3A). Similar toMA-10 cells, PBR is localized almost exclusively to the cytoplasm inMCF-7 cells (Fig. 3B). Strikingly however, PBR localizes primarily tothe nucleus in MDA-231 cells (Fig. 3,C and D). This fluorescenceindicates localization to either the nucleus (Fig. 3C) or the perinuclearenvelope (Fig. 3D). Although not shown, in ADR cells PBR localizeschiefly to the cytoplasm, although nuclear fluorescence is also seen.Displacement of fluorescent labeling by 100mM of FGIN-27 indicatesthat compound 4 binding is specific for PBR (Fig. 3,E and F).Scatchard analysis of [3H]PK 11195 binding to nuclei isolated fromMDA-231 cells revealed aKD of 10.3 6 8.4 nM and a Bmax of6.9 6 4.8 pmol/mg nuclear protein. Furthermore, anti-PBR immuno-staining of MDA-231 cells supports the nuclear localization of thereceptor seen with compound 4 (Fig. 4,A andB).

PBR Expression and Localization Correlates with IncreasedAggressive Phenotype in Human Breast Tumor Biopsies.Beforecharacterization of the receptor in the MDA-231 cells, we examinedwhether the differences seen in the expression and localization of PBRin human breast cancer cell lines reflects real differences found inhuman breast cancer or an artifact of the cell culture system. To thisend, we examined the expression and localization of PBR in normalbreast tissue procured by reductive mammoplasty and metastaticaggressive breast tumor biopsies. Fig. 5A shows HRP staining of thePBR antiserum used to detect theMr 18,000 molecular weight proteinin normal breast tissue. This section was counterstained with hema-toxylin. A light immunostaining for PBR can be seen. The lowstaining is better seen in Fig. 5B in which the hematoxylin counter-stain was omitted. These data were confirmed using a FITC-labeledsecondary antibody (Fig. 5C). It should be noted that PBR appears tobe localized in the ductal epithelium. Fig. 5D is the phase contrastimage of the Fig. 5C. The level of expression and localization of theMr 18,000 molecular weight PBR protein was further examined innormal human breast tissue using compound 4 (Fig. 5E). Incubationof the sections with unlabeled PK 11195 displaced the fluorescentlabeling seen with compound 4 (Fig. 5F). Similarly, low levels ofcytoplasmic expression were obtained in specimens fromin situ andnoninvasive breast tumors (data not shown). These data were repli-cated in multiple sections taken from specimens from three separateindividuals.

PBR expression and localization was then examined in a seriesof biopsies taken from aggressive metastatic breast tumors. Fig. 5Gshows HRP detection of the PBR protein in tumor cells, wherestaining was more pronounced in the nucleus. Omission of thehematoxylin counterstain further enhanced PBR nuclear staining(Fig. 5H). These data were further confirmed using the FITClabeled secondary antibody (Fig. 5I), where robust perinuclear andnuclear staining is evident. Fig. 5J is the phase contrast image ofthe Fig. 5I. The level of expression and localization of the PBR

Fig. 2. Expression of PBR mRNA in MDA-231, ADR, and MCF-7 cell lines. TotalRNA was isolated from MDA-231, ADR, and MCF-7 cells and loaded onto a 1%formaldehyde gel at a concentration of 10mg/lane. Northern blots were incubated with32P-labeled hPBR probe and exposed to X-OMAT Kodak film.Top,28S and 18S rRNAvisualized by ethidium bromide staining.Middle,autoradiogram of the blot. PBR migratesat 0.9 kb.Bottom,relative intensity of the PBR mRNA/28S rRNA. The autoradiogram andPBR mRNA quantitation represent one of three independent experiments.

834




protein was further examined in the tumor biopsies using com-pound 4 (Fig. 5K). A high level of perinuclear and nuclear stainingwas seen, thus confirming the data obtained using the FITC andHRP antibody labeling procedures. Incubation of the sections withunlabeled PK 11195 displaced the fluorescent labeling seen withcompound 4 (Fig. 5L). These data were replicated in multiplesections of specimens taken from three separate individuals.

PBR Found in the MDA-231 Human Breast Cancer Cell Line IsSimilar to PBR Found in Other Human Tissues. Given the numer-ous differences between both the expression and localization of PBRin MDA-231 cells and the other human breast cancer cell linesstudied, as well as reports published previously, it became importantto determine whether we were dealing with the same receptor. Thefirst step toward this end was to establish a pharmacological profilefor the MDA-231 PBR. Displacement of [3H]PK 11195 with increas-

ing concentrations of the indicated PBR ligands is similar, althoughnot identical, to the pharmacological profiles reported previously forhuman PBR (Table 3; Refs. 28–31). Specifically, whereas the affin-ities for the benzodiazepines Ro5-4864 and diazepam are similar tofindings reported previously, the affinities for the isoquinoline enan-tiomers, (-) PK 14067 and (1) PK 14068, were roughly 100-foldlower, suggesting that this receptor has distinct binding characteristicsrelative to the previous studies. It should be noted, however, that theseprevious studies are rather incomplete in their assessment of theligand binding characteristics of PBR and, therefore, should be usedonly to a limited extent in the comparison of human PBR character-istics from tissue to tissue.

Next we obtained partial PBR cDNA sequences for both MDA-231 and MCF-7 PBR. The obtained nucleotide sequences revealedtwo point mutations resulting in the replacement of alanine 147

Fig. 3. Subcellular localization of PBR using the fluorescentPBR ligand compound 4. MA-10 (A), MCF-7 (B), and MDA-231 (CandD) cells were cultured on coverslips and incubatedwith compound 4 (1mM) for 45 min at 37°C. MDA-231 cellswere incubated with compound 4 (1mM) for 45 min at 37°C inthe presence of 100mM of FGIN-27 (E), the nonfluorescentPBR ligand used to develop compound 4 (23). At the end of theincubation time, the cells were washed, and PBR was localizedby fluorescence microscopy.F, phase-contrast of the imageshown inE.

835




with a threonine and histidine 162 with an arginine in both MDA-231 and MCF-7 (Fig. 6). Given that these mutations occur in bothcell lines, it is unclear what role they play in cancer pathogenesis.Despite many efforts, sequences could not be obtained for the

region immediately surrounding the translation start sight. Theregion 39 to the start sight may provide key evidence for thedifferential localization (cytoplasmicversusnuclear) of PBR be-tween these two cell lines.

Fig. 4. PBR and DBI immunocytochemistry in MDA-231 and MCF-7 cells. InA andB, cultured MDA-231 cells were probed with anti-PBR primary antiserum at a concentrationof 1:500;A, 3300;B, 3600.C, MDA-231 cells incubated in the absence of primary antibody;3300. InD andE, cultured MDA-231 cells were probed with anti-DBI primary antiserumat a concentration of 1:500;3400. InF, cultured MCF-7 cells were probed with anti-PBR primary antiserum at a concentration of 1:500;3400. InG, cultured MCF-7 cells were probedwith anti-DBI primary antiserum at a concentration of 1:500;3400.H, MCF-7 cells incubated in the absence of primary antibody;3400. PBR and DBI antiserum localization wasrevealed using HRP-coupled secondary antibody at a concentration of 1:2000.

836




A Functional Role for PBR in Human Breast Cancer. Previousstudies from this laboratory have shown that PBR plays a key role insteroidogenesis by mediating the translocation of cholesterol from theouter to the inner mitochondrial membrane (18). More recently, wehave shown that PBR mediates cholesterol uptake even in nonmito-chondrial membranes (26). To test whether PBR plays a similar rolein MDA-231 nuclear membranes, intact nuclei were isolated fromboth MDA-231 and MCF-7 cells. Isolated nuclei were incubated with10 nM [3H]cholesterol in the absence or presence of increasing con-centrations of PK 11195 (Fig. 7A). In the absence of PK 11195,MDA-231 nuclear membranes demonstrated the ability to take up30% more cholesterol relative to MCF-7 nuclei. In MDA-231 nuclei,addition of 1028 to 1026

M PK 11195 resulted in a decrease (25–30%;P 5 0.012 and 0.006, respectively) in the amount of cholesterolretained by the nuclear membrane to levels relative to control treat-ment. This change corresponds to 0.52 pmol of cholesterol/mg ofnuclear protein. MCF-7 nuclei failed to respond to the PK 11195 doseresponse (F52.204,P 5 0.0625; Fig. 7A). This change in cholesterollevels may reflect either reduced cholesterol uptake or increasedrelease of nuclear membrane cholesterol. Considering our previousstudies showing that perturbation of PBR by specific ligands inducesthe transport of cholesterol through bacterial membranes (26, 32), wefavor the second hypothesis.

Numerous studies performed in the early 1980s showed that Ro5-4864 and PK 11195, specific PBR ligands, regulate cell proliferationin a number of cancer models (5–12). Using the BromodeoxyuridineCell Proliferation ELISA (Boehringer Mannheim), we examined theeffects of PK 11195 on MDA-231 cell proliferation (Fig. 7B). After24 h, low nanomolar PK 11195 (10210 and 1029

M) showed no effecton MDA-231 cell proliferation. However, 1028

M PK 11195 stimu-lated MDA-231 cell proliferation between 20 and 25%, an increasesimilar to earlier reports (7). Stimulation of MDA-231 cell prolifera-tion was maximal (40%) at 1025

M PK 11195. After 48 h, thedose-response curve shifted to the left (data not shown). Cell prolif-eration was stimulated 40% by 1028

M PK 11195, although nostimulation was seen at any of the micromolar concentrations. InMCF-7 cells, on the other hand, after 48 h there was no stimulation ofcell proliferation at any of the PK 11195 concentrations tested (datanot shown). PK 11195 (1024

M), however, did result in a 40%decrease in MCF-7 cell proliferation relative to no treatment at thistime point (data not shown). We believe the 40% decrease in MCF-7cell proliferation seen with 1024

M PK 11195 is meaningless given theobvious nonspecific effects of any drug at this concentration.

Although it is generally assumed that the incorporation of BrdUrdinto DNA is the result of cell division, it is also possible that theresults presented in this study are due to repair of damaged DNA.However, using the Hoechst 33258 DNA assay, we saw increases intotal MDA-231 DNA content in response to PK 11195, similar to theresults presented above (data not shown). These data strongly suggestthat the increases in BrdUrd incorporation into MDA-231 cells reflectincreases in MDA-231 cell division and not DNA damage.

Change in MDA-231 Nuclear Cholesterol Levels Correlateswith an Increase in Cell Proliferation. We have shown that PK11195 reduced the amount of radiolabeled cholesterol retained inMDA-231 nuclear membranes at nanomolar and low micromolarconcentrations. We have also shown that PK 11195 also stimulatesMDA-231 cell proliferation at these concentrations. We were theninterested in ascertaining whether the regulation of nuclear cholesteroltransport correlates with the PBR-mediated regulation of cell prolif-eration in these cells. To determine such a relationship, all of theMDA-231 cholesterol data for 0, 10210, 1028, and 1026

M PK 11195was plotted against the proliferation data at the same PK 11195concentrations. A regression line for all points gave a coefficient of

correlation (r) of 20.99 (Fig. 7C). The data corresponding to 1024M

PK 11195 was omitted from regression analysis due to nonphysiologi-cal and potentially toxic effects of this concentration. It should benoted, however, that the inclusion of this data still resulted in anr of20.75.

MDA-231 Cells Express DBI, the Endogenous PBR Ligand.Given the ability of exogenous PBR ligands to regulate nuclearcholesterol transport and cell proliferation in MDA-231 cells, we thenexamined whether MDA-231 cells express the polypeptide DBI, theendogenous PBR ligand. DBI has been shown to stimulate synthesisof steroids in adrenocortical, Leydig, and glial cells through specificactivation of PBR (3). DBI expression in cells has been reported atlevels comparable with those needed for PBR activation, furthersupporting its role as an endogenous PBR ligand (33). The presenceof a PBR-specific ligand such as DBI in an aggressive human breastcancer cell line would support the hypothesis that PBR is involved inthe advancement of human breast cancer. Indeed, immunocytochem-istry of MDA-231 cells with anti-DBI antiserum reveals that this cellline possesses DBI (Fig. 4,D andE). Interestingly, the pattern of DBIimmunostaining was shown to be similar (nuclear and perinuclear) tothat of PBR fluorescence and immunostaining. MCF-7 cells displayedDBI immunostaining in a similar pattern to that seen in MDA-231(Fig. 4G). However, DBI immunostaining appears to be less robust inMCF-7 relative to MDA-231.

DISCUSSION

In this report, we examined the role of PBR in human breast usinga series of human breast cancer cell lines and human breast tumorbiopsies. Through the course of this study, we describe a strongcorrelation between the expression of PBR ligand binding activity andthe invasive and chemotactic potential, as well as the expression of thebreast cancer marker CD44, among the cell lines. In addition, weshow that PBR is expressed at much higher levels in aggressivemetastatic human breast tumor biopsies compared with normal breasttissue and to noninvasive breast tumors. Furthermore, we show thatPBR is differentially localized between highly aggressive and nonag-gressive cell lines and between normal and aggressive metastatichuman breast tissue. Characterization of breast cancer PBR revealsthat it is similar, although not identical, to PBR studied in other humantissues. Two point mutations were found that result in the replacementof an alanine residue at position 147 with a threonine residue and ahistidine residue at position 162 with an arginine residue.

Functionally, we find that PBR is responsible for the increasedcholesterol transport into the nuclei of a highly aggressive cell line,MDA-231, relative to a nonaggressive cell line, MCF-7 (P 5 0.0009and 0.0625, respectively). Furthermore, we find that PBR regulatescell proliferation of MDA-231 (P , 0.0001) and, moreover, that thisregulation is strongly linked to the ability of PBR to regulate choles-terol transport into MDA-231 nuclei (r 5 20.99). The fact thatnanomolar and low micromolar concentrations, and not high micro-molar concentrations, of PK 11195 are responsible for both of theseactions indicates that these events are the result of specific interactionsbetween the drug ligands used and PBR. The lack of significanteffects of the PBR ligand PK 11195 on MCF-7 nuclear cholesteroltransport and cell proliferation suggests that PBR’s regulation of theseevents occurs only in more aggressive breast cancer cells, furthersupporting a role for PBR in breast tumor progression. Although thefunctional data presented in this paper were elicited by an exogenousPBR ligand, the identification of the endogenous PBR ligand, DBI, inMDA-231 cells around the nucleus further supports a physiologicalrole for the function of PBR described herein.

The expression of PBR protein levels in the cell lines and human837




Fig. 5. PBR expression in normal human breast and aggressive metastatic tumor tissues. InA, paraffin-embedded sections of normal breast tissue were immunostained with ananti-PBR antiserum at 1:500 dilution and counterstained with hematoxylin as described in “Materials and Methods”. InB, the hematoxylin counterstain was omitted to better determinewhether the nucleus of the cells contained immunoreactive PBR protein.C, localization of immunoreactive PBR protein using a FITC-coupled secondary antibody.D, phase contrastmicroscopy of the same tissue area.E, detection of PBR ligand binding activity using the fluorescent PBR ligand compound 4. A filter was used to enhance the detection of low

838




fluorescence levels.F, displacement of the fluorescence with 1000-fold excess of the competitive ligand PK 11195. InG, paraffin-embedded sections of aggressivemetastatic human breast carcinoma tissue were immunostained with an anti-PBR antiserum at 1:500 dilution and counterstained with hematoxylin as describedpreviously (HRP staining). InH, the hematoxylin counterstain was omitted to examine whether the nucleus of the cells contained immunoreactive PBR protein.I, localization of immunoreactive PBR protein in aggressive metastatic human breast carcinoma tissue using the FITC-coupled secondary antibody.J, phasecontrast microscopy of the same tissue area.K, detection of PBR ligand binding activity in aggressive metastatic human breast carcinoma tissue using thefluorescent compound 4. A filter was used to enhance the detection of low fluorescence levels.L, displacement of the fluorescence with 1000-fold excess of thecompetitive ligand PK 11195.A–H, J,andL, 375; I andK, 3150.

839




tissues studied in this paper mirrors that seen in other human cancerstudies. Cornuet al. (13) have shown that PBR site densities are asmuch as 12-fold higher in high-grade astrocytomas and glioblastomasrelative to normal brain tissue. A study by Miettinenet al. (14) also

indicates that PBR is highly up-regulated in high-grade human astro-cytic tumors relative to low-grade tumors. Furthermore, a positronemission tomography study by Pappataet al. (34) revealed thatbinding of PK 11195, the PBR-specific ligand used throughout thepresent study, is 2-fold greater in glioblastomas than in normal humangray matter. Our data support these previous studies by showing thatPBR binding in MDA-231 cells is;7-fold higher than the mildlyaggressive ADR cell line and immeasurably greater than in the non-aggressive MCF-7 cell line.

At the mRNA level, however, this correlation does not appear to beas tight. Whereas MDA-231 cells express 17–20-fold higher PBRmRNA than MCF-7 cells, PBR mRNA expression is as much in theADR cell line as in MDA-231 cells. This result appears to be anom-alous and is difficult to explain because little is known about theregulation of PBR expression. Considering that ADR cells apparentlylocalize PBR to the cytoplasm and the nuclear envelope, increasedtranscription of PBR mRNA may represent a transitional phase be-tween the nonaggressive state and a more aggressive state in thecontext of the human breast cancer cell lines examined in this paper.Alternatively, the low PBR binding levels seen in ADR cells mayimplicate differences in translational regulation of PBR expressionbetween MDA-231 and ADR cells. Neither possibility can be ruledout until further characterization of PBR expression is performed.

Partial sequence analysis revealed two point mutations in bothMDA-231 and MCF-7 cells that result in the replacement of alanine147 with a threonine residue and histidine 162 with an arginineresidue. Molecular modeling of the receptor indicates that the firstresidue lies within the cholesterol entry region of the receptor (28, 32).

Fig. 6. Partial cDNA sequence for MDA-231 andMCF-7 human breast cancer PBR. The partial PBRnucleotide sequences obtained for both MDA-231 andMCF-7 cells were compared with the published humanPBR sequence. The sequences reveal three point mu-tations common to both MDA-231 and MCF-7 PBRcDNA and a single point mutation unique to MDA-231. Deduction of the amino acid sequence revealedthat only two of the common point mutations alter theamino acid sequence (Ala 1473Thr and His1623Arg), whereas the others were silent.

Table 3 Pharmacological characterization of MDA-231 PBR: Comparison of varioushuman PBR pharmacological profiles

Current literature provides several partial profiles of inhibition of PK 11195 binding byvarious ligands in human brain, renal cortex, lymphocytes, and gliomas as indicated byeither the IC50s or theKIs for each of the indicated PBR ligands. Specific binding of[3H]PK 11195 (2 nM) to MDA-231 human breast cancer cells was measured in thepresence of various concentrations of the indicated competing ligands. For competingligands FGIN-27, Ro5-4864, (2) PK 14067, and (1) PK 14068, concentrations rangedfrom 10210 to 1024 M. For competing ligands, diazepam, flunitrazepam, and clonazepamconcentrations ranged from 1027 to 1024 M. These concentration ranges cover theexpected affinity range of each ligand for PBR. IC50 estimation was performed using theLIGAND program (42).KI estimation was obtained with the equation:KI 5 [IC50]/(11[ligand]/KD), where [ligand]5 concentration of PK 11195 andKD 5 dissociationconstant for PK 11195 as presented in Table 1.

Ligand

IC50 (nM) KI (nM)

BrainaRenal

cortexb MDA-231 MDA-231 Lymphocytesc Gliomad

FGIN-27 1.94 1.54Ro5-4864 138 822 528 419 6.7 84.9Diazepam 1,780 1,410 360(2) PK 14067 23.3 3,940 3,130Flunitrazepam 58,800 46,700 848Clonazepam 119,000 94,400 .10,000(1) PK 14068 1,422 186,000 148,000

a From Changet al. (28).b From Broaddus and Bennett (29).c From Alexanderet al. (30).d From Olsonet al. (31).

840




Presently, it is not apparent whether this mutation has a resultingphenotype. It appears that it does not alter the ability of cholesterol tomove through PBR because cholesterol is incorporated into MDA-231nuclei. The fact that it is present in both the MDA-231, a highlyaggressive breast cancer cell line, and in MCF-7, a nonaggressive cellline, suggests that this mutation may represent an early event in theprogression of this disease. The second mutation (His 1623Arg) is aconserved mutation, given that both histidine and arginine are basicamino acids, and likely has no significant effect on PBR structure andfunction.

PBR is primarily targeted to the outer mitochondrial membrane intissues in which it is expressed in great abundance (3). However, it hasalso been found in other cellular organelles such as the plasmamembrane as well as the peroxisome (3, 35). The lack of a distinctmitochondrial target sequence and the largely hydrophobic nature ofPBR make it feasible that PBR could exist in a variety of membranes.Differential localization of PBR may also be possible through theexistence of chaperone proteins and PBR-associated proteins that maydirect PBR to the membranes of specific organelles and may influencethe functioning of PBR (36). The significance of such differentiallocalization, however, has not been investigated and is presentlyunknown. It will be necessary to distinguish whether the nuclearlocalization of PBR in MDA-231 cells is the result of a specific aminoacid sequence present in the yet undetermined NH2 terminus of theprotein or the shuttling of PBR to the nucleus via association withanother protein.

The data presented in this paper suggest that nuclear PBR isresponsible for regulating movement of cholesterol into the nuclearmembrane and that this regulation is related to its modulation of cellproliferation. Cholesterol is a major lipid component of every mem-brane and influences the degree of membrane fluidity (37). Choles-terol has also been shown to play a certain but controversial role in theadvancement of a variety of pathologies including breast cancer(38–40). Furthermore, reports on animal dietary, cellular, and en-zyme-specific studies implicate a role for cellular cholesterol in theregulation of cell proliferation (38). Cholesterol has been shown to

tightly regulate the activity of the SREBPs found in the nuclearmembrane and the endoplasmic reticulum (41). In the presence ofexcessive cholesterol, premature SREBP is not fully cleaved, andtherefore, the mature form is not released and cannot enter the nucleusto carry out transcriptional activation (41). SREBPs are responsiblefor the transcriptional regulation of the enzymes involved in thecholesterol biosynthetic pathway as well as the enzymes involved infatty acid synthesis and uptake (41). One possible outcome of con-centrating cellular cholesterol to the nuclear membrane may be toinhibit the activation of nuclear membrane SREBPs. Conversely, weshow that PBR ligands reduce the presence of cholesterol in MDA-231 nuclear membranes. As mentioned before, we have reason tobelieve that these data are due to release of cholesterol into thenucleus. What role cholesterol might play in the nucleus and whetherthe effect we show on cell proliferation is direct or indirect is com-pletely open to debate at this point. With the tight correlation betweennuclear transport of cholesterol in MDA-231 and the regulation ofMDA-231 cell proliferation by PBR, the SREBP pathway may shedsome light as to how PBR is regulating cell proliferation in these cellsand should be the target of future research in this area. However, otherpossibilities such as changes in nuclear membrane fluidity induced byalterations of membrane cholesterol content and the direct effects ofcholesterol on gene transcription must also be pursued to definitivelyanswer these questions.

It is possible that the relationship between increased expression andnuclear localization of PBR with both aggressive breast cancer celllines and malignant breast biopsies may be due to differences inmetabolism and cellular activity between the cell lines and tissuesstudied. However, the functionality of PBR in the MDA-231 cell line,i.e., the ability to regulate both nuclear cholesterol uptake and cellproliferation, as well as the strong correlation between these twoseemingly separate events, however, suggests that PBR is indeedplaying a role in the progression of breast malignancies. The presenceand localization of the putative endogenous PBR ligand, DBI, inMDA-231 cells further suggests the likelihood that PBR is fullyfunctional in these cells.

Fig. 7. PK 11195 regulated MDA-231 nuclear cholesterol transport and cell proliferation are highly correlated. InA, transport of [3H]cholesterol into nuclei isolated from MDA-231and MCF-7 cells was measured in response to varying doses of PK 11195. Data are expressed as percentage of cholesterol uptake into MCF-7 nuclei in the absence of any PK 11195.One hundred % cholesterol uptake corresponds to 1.75 pmol/mg protein. Data points represent the means of five (MDA-231) or four (MCF-7) independent experiments carried outin quadruplicate;bars,SE. One-way ANOVA indicates that treatment of isolated MDA-231 nuclei but not MCF-7 nuclei with PK 11195 induces significant changes in cholesterolcontent (F5 4.74,P 5 0.0009;F 5 2.204,P 5 0.0625, respectively; GraphPad, Inc.). InB, MDA-231 cells grown in 96-well plates were washed with PBS and cultured in mediasupplemented with 0.1% FBS 24 h before any treatment. The indicated concentrations of PK 11195 were added to the cells cultured in DMEM supplemented with 0.1% FBS andincubated for 24 h at 37°C. Four h before the end of incubation, BrdUrd was added to each well. Incorporation of BrdUrd was measured at 450 nm (reference, 700 nm). Data pointsrepresent the means of three independent experiments carried out in quadruplicate;bars,SE. One-way ANOVA indicates that MDA-231 cell proliferation was significantly altered bytreatment with PK 11195 at both the 24- and 48-h time points (F 5 20.339,P , 0.0001;F 5 23.126,P , 0.0001, respectively; GraphPad, Inc.). Comparison of each PK 11195 treatmentto control was performed with the unpairedt test provided by the SigmaPlot software (ppp, P , 0.05; Jandel Scientific).C, the means of all data points for 0,210, 28, and26 M

PK 11195 from the previously described cell proliferation assay were plotted against the corresponding means from the cholesterol uptake assay described previously. A regressionline drawn for all plotted data gives a coefficient of correlation of20.99. Numeric values in parentheses indicate the number data points taken for each mean;bars,SE.

841




Malignant breast tumors are primarily characterized by aberrantcell proliferation, tumor invasion, and metastasis. Several molecularand cellular mechanisms have been proposed to account for thesephenomena, and a number of prognostic indicators have been identi-fied. Although these markers have been useful in helping cliniciansdevelop prognoses, they have failed to provide adequate informationabout the mechanisms responsible for tumor malignancy so thateffective anticancer therapies may be developed. Given the datapresented in this report, we believe that PBR is a major component ofthe progression of breast cancer. Although a great deal more needs tobe learned about PBR and its ability to regulate cell proliferation andcholesterol movement, we believe this work is a major step in under-standing this disease. Our present work with both human breast cancercell lines and human breast tissue, combined with the availability ofradiolabeled and fluorescent PBR ligands, may be useful in thediagnosis and prognosis of the disease. Moreover, a great number ofPBR ligands are known, including benzodiazepines and isoquinolinecarboxamides, the PBR-binding and pharmacological characteristicsof which are well documented. Some of these ligands may havepotential as future anticancer therapies.

REFERENCES

1. Braestrup, C., and Squires, R. F. Specific benzodiazepine receptors in rat braincharacterized by high-affinity [3H]diazepam binding. Proc. Natl. Acad. Sci. USA,74:3805–3809, 1977.

2. Garnier, M., Dimchev, A. B., Boujrad, N., Price, J. M., Musto, N. A., and Papado-poulos, V. In vitro reconstitution of a functional peripheral-type benzodiazepinereceptor from mouse Leydig tumor cells. Mol. Pharmacol.,45: 201–211, 1994.

3. Papadopoulos, V. Peripheral-type benzodiazepine/diazepam binding inhibitor recep-tor: biological role in steroidogenic cell function. Endocr. Rev.,14: 222–240, 1993.

4. Papadopoulos, V., Amri, H., Li, H., Boujrad, N., Vidic, B., and Garnier, M. Targeteddisruption of the peripheral-type benzodiazepine receptor gene inhibits steroidogen-esis in the R2C Leydig tumor cell line. J. Biol. Chem.,272: 32129–32135, 1997.

5. Clarke, G. D., and Ryan, P. J. Tranquilizers can block mitogenesis in 3T3 cells andinduce differentiation in Friend cells. Nature (Lond.),287: 160–161, 1980.

6. Wang, J. K. T., Morgan, J. I., and Spector, S. Benzodiazepines that bind at peripheralsites inhibit cell proliferation. Proc. Natl. Acad. Sci. USA,81: 753–756, 1984.

7. Ikezaki, K., and Black, K. L. Stimulation of cell growth and DNA synthesis byperipheral benzodiazepine. Cancer Lett.,49: 115–120, 1990.

8. Garnier, M., Boujrad, N., Oke, B. O., Brown, A. S., Riond, J., Ferrara, P., Shoyab, M.,Suarez-Quian, C. A., and Papadopoulos, V. Diazepam binding inhibitor is a paracrine/autocrine regulator of Leydig cell proliferation and steroidogenesis: action viaperipheral-type benzodiazepine receptor and independent mechanisms. Endocrinol-ogy, 132: 444–458, 1993.

9. Bruce, J. H., Ramirez, A. M., Lin, L., Oracion, A., Agarwal, R. P., and Norenberg,M. D. Peripheral-type benzodiazepines inhibit proliferation of astrocytes in culture.Brain Res.,564: 167–170, 1991.

10. Camins, A., Diez-Fernandez, C., Pujadas, E., Camarasa, J., and Escubedo, E. A newaspect of the antiproliferative action of peripheral-type benzodiazepine receptorligands. Eur. J. Pharmacol.,272: 289–292, 1995.

11. Laird, H. E., II, Gerrish, K. E., Duerson, K. C., Putnam, C. W., and Haddock-Russell,D. Peripheral benzodiazepine binding sites in Nb 2 node lymphoma cells: effects onprolactin-stimulated proliferation and ornithine decarboxylase activity. Eur. J. Phar-macol.,171: 25–35, 1989.

12. Neary, J. T., Jorgensen, S. L., Oracion, A., Bruce, J. H., and Norenberg, M. D.Inhibition of growth factor-induced DNA synthesis in astrocytes by ligands ofperipheral-type benzodiazepine receptors. Brain Res.,675: 27–30, 1995.

13. Cornu, P., Benavides, J., Scatton, B., Hauw, J. J., and Philippon, J. Increase inperipheral-type benzodiazepine binding site densities in different types of humanbrain tumors. A quantitative autoradiography study. Acta Neurochir.,119: 146–152,1992.

14. Miettinen, H., Kononen, J., Haapasalo, H., Helen, P., Sallinen, P., Harjuntausta, T.,Helin, H., and Alho, H. Expression of peripheral-type benzodiazepine receptor anddiazepam binding inhibitor in human astrocytomas: relationship to cell proliferation.Cancer Res.,55: 2691–2695, 1995.

15. Lippman, M. E. The development of biological therapies for breast cancer. Science(Washington DC),259: 631–632, 1993.

16. Azzam, H. S., Arand, G., Lippman, M. E., and Thompson, E. W. Association ofMMP-2 activation potential with metastatic progression in human breast cancer celllines independent of MMP-2 production. J. Natl. Cancer Inst.,85: 1758–1764, 1993.

17. Culty, M., Shizari, M., Thompson, E. W., and Underhill, C. B. Binding and degra-dation of hyaluronan by human breast cancer cell lines expressing different forms ofCD44. Correlation with invasive potential. J. Cell. Physiol.,160: 275–286, 1994.

18. Papadopoulos, V., Mukhin, A. G., Costa, E., and Krueger, K. E. The Peripheral-typebenzodiazepine receptor is functionally linked to Leydig cell steroidogenesis. J. Biol.Chem.,265: 3772–3779, 1990.

19. Munson, P. J., and Rodbard, D. LIGAND. A versatile computerized approach forcharacterization of ligand binding systems. Anal. Biochem.,107: 220–239, 1980.

20. Bradford, M. M. A rapid and sensitive method for the quantification of microgramquantities of protein using the principle of protein-dye binding. Anal. Biochem.,72:248–254, 1976.

21. Maniatis, T., Fritsch, E. F., and Sambrook, J. Molecular Cloning: A LaboratoryManual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989.

22. Riond, J., Mattei, M. G., Kaghad, M., Dumont, X., Guillemot, J. C., LeFur, G., Caput,D., and Ferrara, P. Molecular cloning and chromosomal localization of a humanperipheral-type benzodiazepine receptor. Eur. J. Biochem.,195: 305–311, 1991.

23. Kozikowski, A. P., Kotoula, M., Ma, D., Boujrad, N., Tuckmantel, W., and Papado-poulos, V. Synthesis and biology of a 7-nitro-2,1,3-benzoxadiazol-4-yl derivative of2-phenylindole-3-acetamide: a fluorescent probe for the peripheral-type benzodiaz-epine receptor. J. Med. Chem.,40: 2435–2439, 1997.

24. Amri, H., Ogwuebu, S. O., Boujrad, N., Drieu, K., and Papadopoulos, V.In vivoregulation of peripheral-type benzodiazepine receptor and glucocorticoid synthesis byGinkgo bilobaextract EGb 761 and isolated ginkgolides. Endocrinology,137: 5707–5718, 1996.

25. Elango, N., Vivekananda, J., Strong, R., and Katz, M. S. Nuclei isolation from bonecells for nuclear run-on assays. BioTechniques,23: 422–424, 1997.

26. Li, H., and Papadopoulos, V. Peripheral-type benzodiazepine receptor function incholesterol transport: identification of a cholesterol recognition/interaction amino acidsequence and consensus pattern. Endocrinology,139: 4991–4997, 1998.

27. Green, B., and Leake, R. E. Steroid hormones: a practical approach. Oxford: IRLPress Limited, 1987.

28. Chang, Y. J., McCabe, R. T., Rennert, H., Budarf, M. L., Sayegh, R., Emanuel, B. S.,Skolnick, P., and Strauss, J. F. The human “peripheral-type” benzodiazepine receptor:regional mapping of the gene and characterization of the receptor expressed fromcDNA. DNA Cell Biol., 11: 471–480, 1992.

29. Broaddus, W. C., and Bennett, J. P. Peripheral-type benzodiazepine receptors inhuman glioblastomas: pharmacologic characterization and photoaffinity labeling ofligand recognition site. Brain Res.,518: 199–208, 1990.

30. Alexander, B. E., Roller, E., and Klotz, U. Characterization of peripheral-typebenzodiazepine binding sites on human lymphocytes and lymphoma cell lines andtheir role in cell growth. Biochem. Pharmacol.,44: 269–274, 1992.

31. Olson, J. M., Junck, L., Young, A. B., Penney, J. B., and Mancini, W. R. Isoquinolineand peripheral-type benzodiazepine binding in gliomas. Implications for DiagnosticImaging. Cancer Res.,48: 5837–5841, 1988.

32. Culty, M., Li, H., Boujrad, N., Amri, H., Vidic, B., Bernassau, J. M., Reversat, J. L.,and Papadopoulos, V.In vitro studies on the role of the peripheral-type benzodiaz-epine receptor in steroidogenesis.J. Steroid Biochem. Mol. Biol., in press, 1999.

33. Papadopoulos, V., Berkovich, A., Krueger, K. E., Costa, E., and Guidotti, A. Diaz-epam binding inhibitor (DBI) and its processing products stimulate mitochondrialsteroid biosynthesis via an interaction with mitochondrial benzodiazepine receptors.Endocrinology,129: 1481–1488, 1991.

34. Pappata, S., Cornu, P., Samson, Y., Prenant, C., Benavides, J., Scatton, B., Crouzel,C., Hauw, J. J., and Syrota, A. PET study of carbon-11-PK 11195 binding toperipheral type benzodiazepine sites in glioblastoma: a case report. J. Nucl. Med.,32:1608–1610, 1991.

35. Woods, M. J., Zisterer, D. M., and Williams, D. C. Two cellular and subcellularlocations for the peripheral-type benzodiazepine receptor in rat liver. Biochem.Pharmacol.,51: 1283–1292, 1996.

36. Papadopoulos, V. Structure and function of the peripheral-type benzodiazepine re-ceptor in steroidogenic cells. Proc. Soc. Exp. Biol. Med.,217: 130–142, 1998.

37. Oke, B. O., Suarez-Quian, C. A., Riond, J., Ferrara, P., and Papadopoulos, V. Cellsurface localization of the peripheral-type benzodiazepine receptor (PBR) in adrenalcortex. Mol. Cell. Endocrinol.,87: R1–R6, 1992.

38. Coleman, P. S., Chen, L. C., and Sepp-Lorenzino, L. Cholesterol metabolism andtumor cell proliferation.In: R. Bittman (ed.), Cholesterol: Its Functions and Metab-olism in Biology and Medicine, pp. 363–435. New York: Plenum Publishing Corp.,1997.

39. Subbaiah, P. V., Liu, M., and Witt, T. R. Impaired cholesterol esterification in theplasma in patients with breast cancer. Lipids,32: 157–162, 1997.

40. Kokoglu, E., Karaarslan, I., Karaarslan, H. M., and Baloglu, H. Alterations of serumlipids and lipoproteins in breast cancer. Cancer Lett.,82: 175–178, 1994.

41. Brown, M. S., and Goldstein, J. L. The SREBP pathway: regulation of cholesterolmetabolism by proteolysis of a membrane-bound transcription factor. Cell,89:331–340, 1997.

42. Albini, A., Iwamoto, Y., Kleinman, H. K., Martin, G. R., Aaronson, S. A., Kozlowski,J. M., and McEwan, R. N. A rapidin vitro assay for quantitation of the invasivepotential of tumor cells. Cancer Res.,47: 3239–3245, 1987.

842




1999;59:831-842. Cancer Res Matthew Hardwick, Djamil Fertikh, Martine Culty, et al. CholesterolPBR-mediated Cell Proliferation and Nuclear Transport of Phenotype with PBR Expression, Nuclear Localization, andBreast Cancer: Correlation of Breast Cancer Cell Aggressive Peripheral-Type Benzodiazepine Receptor (PBR) in Human

Updated version

http://cancerres.aacrjournals.org/content/59/4/831

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/59/4/831.full#ref-list-1

This article cites 38 articles, 10 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/59/4/831.full#related-urls

This article has been cited by 28 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected] 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/59/4/831To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/59/4/831.full#ref-list-1

http://cancerres.aacrjournals.org/content/59/4/831.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



[CANCER RESEARCH 59, 831–842, February 15, 1999 ... · proliferation and cancer progression. The...

Documents

Transcript of [CANCER RESEARCH 59, 831–842, February 15, 1999 ... · proliferation and cancer progression. The...