Caveolin-1 Tyrosine Phosphorylation Enhances Paclitaxel-mediated Cytotoxicity

Caveolin-1 Tyrosine Phosphorylation EnhancesPaclitaxel-mediated Cytotoxicity*

Received for publication, September 14, 2006, and in revised form, December 5, 2006 Published, JBC Papers in Press, December 26, 2006, DOI 10.1074/jbc.M608857200

Ayesha N. Shajahan‡, Aifen Wang‡, Markus Decker‡, Richard D. Minshall§, Minetta C. Liu‡, and Robert Clarke‡¶1

From the ‡Department of Oncology, Lombardi Comprehensive Cancer Center, and ¶Departments of Physiology and Biophysics,Georgetown University, College of Medicine, Washington, D. C. 20057 and §Departments of Pharmacologyand Anesthesiology, University of Illinois at Chicago, Chicago, Illinois 60612

Caveolin-1 (CAV1), a highly conservedmembrane-associatedprotein, is a putative regulator of cellular transformation. CAV1is localized in the plasmalemma, secretory vesicles, Golgi, mito-chondria, and endoplasmic reticulummembrane and associateswith the microtubule cytoskeleton. Taxanes such as paclitaxel(Taxol) are potent anti-tumor agents that repress the dynamicinstability of microtubules and arrest cells in the G2/M phase.Src phosphorylation of Tyr-14 on CAV1 regulates its cellularlocalization and function. We report that phosphorylation ofCAV1 on Tyr-14 regulates paclitaxel-mediated apoptosis inMCF-7 breast cancer cells. Befitting its role as a multitaskingmolecule, we show that CAV1 sensitizes cells to apoptosis byregulating cell cycle progression and activation of the apoptoticsignaling molecules BCL2, p53, and p21. We demonstrate thatphosphorylated CAV1 triggers apoptosis by inactivating BCL2and increasing mitochondrial permeability more efficientlythan non-phosphorylated CAV1. Furthermore, expression ofp21, which correlates with taxane sensitivity, is regulated byCAV1 phosphorylation in a p53-dependent manner. Collec-tively, our findings underscore the importance of CAV1 phos-phorylation in apoptosis and suggest that events that negateCAV1 tyrosine phosphorylation may contribute to anti-micro-tubule drug resistance.

Caveolin-1 (CAV1)2 is a 21–24-kDa protein and the proto-type of a family of integral membrane proteins that associatewith specific cholesterol- and sphingolipid-rich domains toform the structural foundation of membrane invaginationscalled caveolae. Caveolae act as sites of signal transduction invarious cell types (1). CAV1 is thought to regulate the activity of

proteins such as Src kinases, epidermal growth factor tyrosinekinase, Her2/neu (ErbB2) kinase, ERK (extracellular signal-reg-ulated kinase), H-Ras, endothelial nitric-oxide synthase, and Gproteins (1, 2) involved in survival pathways. In human breasttumors, CAV1 levels inversely correlate with tumor size (3),and CAV1 expression reduces the growth of mouse mammarytumors and their spontaneous metastasis to lung and bone (4).However, in breast cancer cell culture models, CAV1 is down-regulated in non-invasive human breast cancer cells but up-regulated in cells with an invasive phenotype (5–7).Taxanes are potent anti-tumor agents that function by bind-

ing to the � subunits of tubulin and repressing the dynamicinstability of spindles (8, 9), activities that lead to cell cyclearrest in the G2/M phase (10). Taxanes such as paclitaxel(Taxol) or docetaxel (Taxotere) are routinely used in the first-line treatment of metastatic breast, lung, ovarian, and digestivecancers (11). In primary breast cancer, inclusion of taxane inadjuvant chemotherapy reduces the relative risk of recurrenceand improves overall survival (12). Acquired resistance throughcellular adaptations or mutations in neoplastic cells remains amajor problem in chemotherapy. Although taxanes are sub-strates for ABC transporters, other resistance mechanisms areclearly important (13). Therefore, it is important to improveour understanding of the mechanisms of drug responsivenessand to identify better predictors of drug efficacy.CAV1 is essential for the formation andmovement of caveo-

lae through the cytoplasm along microtubule tracks, and it islocalized in the microtubule-organizing center or peri-centro-somal region in Chinese hamster ovary cells (14). These find-ings suggest an essential relationship between CAV1 andmicrotubules or microtubule-associated proteins and theirfunction. Treatment with a cytostatic dose of paclitaxel blockslung cancer cells inG2/Mand causes an up-regulation ofCAV1,implicating CAV1 in taxane-mediated cell death and perhapsdrug resistance (15–18). However, the role of CAV1 function incell death remains unclear. In macrophages, induction of apo-ptosis by different apoptotic agents such as simvastatin andcamptothecin leads to a large increase in CAV1 expression. Asan early event, this increase in CAV1 is independent of caspaseactivation or DNA fragmentation but is associated with theplasma membrane translocation of phosphatidylserine (19).Originally identified as a substrate for v-Src (20), CAV1 is

phosphorylated on Tyr-14 by c-Src (21). Mounting evidencesuggests that phosphorylated CAV1 regulates caveolae forma-tion and function (21–25). The precise role of Src kinase intaxane-mediated cytotoxicity is unclear. Src can increase Taxo-

* This work was supported by Public Health Service Grant R01-CA096483 andDept. of Defense Grant BC030280 (to R. C.) and Susan G. Komen BreastCancer Foundation Postdoctoral Fellowship PDF0600477 (to A. N. S.).Technical services were provided by the Flow Cytometry and Cell Sorting,Tissue Culture Core and Microscopy and Imaging Shared Resources Centerfunded through Public Health Service Award 1P30-CA-51008 (LombardiComprehensive Cancer Center support grant). The costs of publication ofthis article were defrayed in part by the payment of page charges. Thisarticle must therefore be hereby marked “advertisement” in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

1 To whom correspondence should be addressed: Dept. of Oncology, Lom-bardi Comprehensive Cancer Center, Georgetown University, College ofMedicine, 3970 Reservoir Rd. N.W., NRB W405, Washington, D. C. 20057.Tel.: 202-687-7237; Fax: 202-687-7505; E-mail: [email protected].

2 The abbreviations used are: CAV1, caveolin-1; EV, empty vector; IMEM,improved minimal essential medium; FBS, fetal bovine serum; GAPDH,glyceraldehyde phosphate dehydrogenase; wt, wild-type; ANOVA, analy-sis of variance; siRNA, small interfering RNA.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 282, NO. 8, pp. 5934 –5943, February 23, 2007© 2007 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

5934 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 8 • FEBRUARY 23, 2007

at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

tere sensitivity by mediating downstream apoptotic eventsthrough BCL2 phosphorylation in v-Src-transformed humangall bladder epithelial cells (26). However, in human ovariancancer cells, an inhibition of Src activity increases paclitaxel-induced cytotoxicity (27). Thus, the precise role of CAV1, andparticularly that of phospho-CAV1(Y14), in affecting breastcancer cell responsiveness to taxanes is unknown.The current study was undertaken to determine whether

CAV1 is involved in the cytotoxic and proapoptotic actions ofpaclitaxel in MCF-7 human breast adenocarcinoma cells. Weoverexpressed wild type (wt), a phosphorylation-defectiveCAV1mutant (Y14F), or empty vector (EV) inMCF-7 cells thatnormally express low levels of endogenousCAV1.The effects ofwtCAV1, Y14F, or EV expression on cell growth, apoptosis, andp53/p21 transcription in response to low dose (10 nM) pacli-taxel were analyzed. This study provides novel insights into thefunction of CAV1 in paclitaxel sensitivity and the role of phos-phorylation onTyr-14 inCAV1-mediated effects onmitochon-drial permeability and apoptosis.

EXPERIMENTAL PROCEDURES

Cell Culture and Reagents—MCF-7 cells (obtained from theLCCC Tissue Culture Shared Resource) were cultured inimproved minimal essential medium (IMEM; Biofluids, Rock-ville, MD) supplemented with 5% fetal bovine serum (FBS).Cells were maintained in a humidified atmosphere at 37 °C and95% air/5% CO2. Paclitaxel was obtained from Sigma and wasdissolved in ethanol (which was used as the vehicle control).PP2 was purchased from Calbiochem. All other reagents wereobtained from Sigma unless otherwise indicated.Generation of Stable Cell Lines—MCF-7 cells were grown to

50–60% confluence and transfected with either human wtCAV1 or Tyr-143 Phe-14 phosphorylation-deficient mutant(Y14F) in pcDNA6 or EV using the FuGENE 6 transfectionreagent (Roche Applied Sciences). Medium was replaced 24 hlater with complete growthmedium, and the cells were allowedto grow for 5 days. Complete growth medium containing blas-ticidin S HCl (Invitrogen) (10 �g/ml) was used for stable selec-tion. For all experiments, pooled populations of stable cell lineswere used.Western Blot Analyses—To determine the effects of pacli-

taxel on CAV1 protein expression, cells were treated with vehi-cle or 10 nM paclitaxel in FBS-IMEM for 24 h. Controls weretreatedwith vehicle alone (0.02% v/v ethanol). ForWestern blotanalysis, cells were lysed for 30min at 4 °C in lysis buffer (50mMTris-HCl, pH 7.5, containing 150 mM NaCl, 1 mM EDTA, 0.5%sodium deoxycholate, 1% Igepal CA-630, 0.1% SDS, 1 mMNa3VO4, 44 �g/ml phenylmethylsulfonyl fluoride) supple-mented with Complete Mini protease inhibitor mixture tabletsand 1mM sodium orthovanadate phosphatase inhibitor (RocheApplied Science). Total protein was quantified using the bicin-choninic acid assay (Pierce). Whole cell lysate (20–50 �g) wasresolved by SDS-PAGE. The following primary antibodies wereused for immunoblotting: monoclonal antibody against phos-pho-CAV1(Y14) and polyclonal antibody against CAV1 (BDBiosciences); monoclonal cleaved poly(ADP-ribose) polymer-ase, polyclonal phospho-BCL2(S70), polyclonal p53, polyclonalSrc, and phospho-Src(pY418) antibodies (Cell Signaling, Dan-

vers, MA); and monoclonal BCL2 (Stressgen Corp., Ann Arbor,MA); monoclonal �-tubulin (Sigma); monoclonal p21 (Calbio-chem). Equal protein loading of gels was confirmed by immuno-staining with a glyceraldehyde-3-phosphate dehydrogenase(GAPDH) antibody (Santa Cruz Biotechnology, Santa Cruz, CA).Immunostaining and Confocal Microscopy—Cells grown on

coverslips were washed with phosphate-buffered saline and incu-bated at least 3 h in serum-free andphenol red-freemedium.Cellswere then fixed, permeabilized, and incubated with primary anti-body. Fluorophore conjugates and4�,6-diamidino-2-phenylindoledihydrochloride (DAPI) were obtained from Molecular Probes,Inc. (Eugene, OR). Where appropriate, DAPI was added tovisualize the nucleus, and non-confocal DAPI images wereacquired using Hg lamp excitation and a UV filter set. Con-focal microscopy was performed using an Olympus IX-70confocal microscope with 405-, 488-, and 543-nm excitationlasers. Fluorescence emission was separately detected foreach fluorophore in optical sections �1 �m in thickness(pinhole set to achieve 1 Airy unit).Transcriptional Reporter Assays—Cells were transfected

with 0.4 �g of p53 (Panomics, Fremont, CA) and p21 luciferasereporter plasmid (a gift from Dr. Jane Trepel, National Insti-tutes of Health) and 0.1 �g of pCMV-Renilla (Promega, Madi-son,WI) perwell using the FuGENE6 transfection reagent. Thenext day, cells were treated with 10 nM paclitaxel for 24 h. Acti-vation of the luciferase constructs wasmeasured using theDualLuciferase assay kit (Promega). Luciferase values were normal-ized to Renilla luminescence. Three independent experimentswere performed in quadruplicate. Data are presented as themean � S.E. for all experiments.Cell Proliferation Assays—Cells were seeded at a density of

1–2� 104 cells/well in 24-well plates. For Src inactivation, cellswere incubated with 10 �M PP2 for 2–3 h before adding pacli-taxel. For small interfering RNA (siRNA)-mediated knockdown of CAV1, cells were transiently transfected with CAV1siRNA for 48 h before adding paclitaxel. To assess paclitaxel-induced growth inhibition, cells were treated with 10 nM pacli-taxel (in FBS-IMEM) for 24 h. Cells were then trypsinized,resuspended in phosphate-buffered saline, and counted using aZ1 Single Coulter Counter (Beckman Coulter, Miami, FL). Atleast three independent experiments were done in sextuplicate.Datawere normalized to vehicle-treated cells and are presentedas the mean � S.E. from a representative experiment.Cell Cycle and Apoptosis Assays—Following treatment of

cells with 10 nM paclitaxel in FBS-IMEM for 24 h, cells werefixed in 70% ethanol for 20 min at 4 °C. Cell cycle distributionwas measured by fluorescence-activated cell sorting in theLombardi Comprehensive Cancer Center Flow CytometryShared Resource facility. Annexin V and propidium iodidestaining was done using an Annexin V-fluorescein isothiocya-nate kit (Trevigen, Gaithersburg, MD). Mitochondrial perme-ability was detected using the ApoAlert mitochondrial mem-brane sensor kit (Clontech, Mountain View, CA).Quantitative Real-time Polymerase Chain Reaction—Primer

for CAV1 (Hs00971716_m1) and the housekeeping gene ribo-somal protein, large, P0 (RPLPO) (Hs99999902_m1) was pur-chased fromApplied Biosystems (Foster City, CA).MCF-7 cellswere allowed to grow to 60–70% confluence in 75-cm2 flasks

Caveolin-1 Regulates Paclitaxel Sensitivity

FEBRUARY 23, 2007 • VOLUME 282 • NUMBER 8 JOURNAL OF BIOLOGICAL CHEMISTRY 5935

at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

and were then treated with either different concentrations ofpaclitaxel (0, 10, 100 nM) for 24 h or 10 nM paclitaxel for 0, 24,and 48 h. RNAwas extracted using the TRIzol reagent (Invitro-gen), cleaned using the RNeasy kit (Qiagen, Valencia, CA), andanalyzed by the Agilent Bioanalyzer 2100 (Santa Clara, CA).About 1 �g of DNase I (Invitrogen)-treated RNA was reverse-transcribed with SuperScript II reverse transcriptase (Invitro-gen) using Oligo(dT)16 (Applied Biosystems). Real-time abso-lute quantitative PCR for each cDNA sample and a standardcurve were established using TaqMan PCR mastermix in thepresence ofCAV1primer or the internal control RPLP0primer.Reactions (10 �l) were run in triplicate in 384-well plates on anABI Prism 7900 HT sequence detection system using the pro-tocol suggested by themanufacturer. The ratio of CAV1 induc-tionwas estimated in comparisonwith RPLPO expression; datapresented are the mean � S.E.

Transfection of siRNA—Cells were plated in 12- or 24-wellplates in complete medium and allowed to grow to 50% conflu-ence. Approximately, 100 nM p21 siRNA, p53 siRNA (Cell Sig-naling), CAV1 siRNA (Dharmacon, Lafayette, CO), or theirrespective control siRNA were transfected using the TransIT-siQUEST (Mirus,Madison,WI) transfection reagent accordingto the manufacturer’s protocol. At 24 h, 10 nM paclitaxel orvehicle was added to the siRNA-transfected cells. Cells werelysed at 48 h post-transfection and subjected to Western blotanalysis or cell proliferation assay as described above.Statistical Analyses—Statistical analyses were performed

using the Sigmastat software package (Jandel Scientific, SPSS,Chicago, IL). Where appropriate, protein expression, cellgrowth, and apoptosis were compared using Student’s t test orANOVAwith a post hoc t test formultiple comparisons.Whereseveral groups were compared with the same control, we used

FIGURE 1. Paclitaxel induces CAV1 expression in MCF-7 cells. A, quantitative real-time PCR assay was used to measure CAV1 mRNA expression levelsin MCF-7 cells following treatment with 10 nM paclitaxel for 24 and 48 h. Data presented are mean � S.E. of three determinations in which expressionlevels were determined as a ratio of CAV1:PRLPO (a housekeeping gene). ANOVA, p � 0.004; *, p � 0.05 versus MCF-7 at 0 h. B, Western blot analysis ofCAV1 protein expression following treatment with 10 nM paclitaxel for 0 (untreated), 24, or 48 h; representative immunoblot of CAV1 and GAPDH(loading control). C, CAV1 mRNA expression levels in MCF-7 cells following treatment with 0 (vehicle), 10, or 100 nM paclitaxel for 24 h. ANOVA, p � 0.001;*, p � 0.05 versus MCF-7 with 0 nM paclitaxel. D, CAV1 protein levels detected by Western blotting after treatment with 0 (vehicle), 10, or 100 nM paclitaxelfor 24 h.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

Dunnett’s test. Differences were considered significant at p �0.05; all tests were two-tailed.

RESULTS

Exposure to Taxane Increased CAV1 Expression in MCF-7Cells—CAV1 expression is very low but remains detectablein MCF-7 human breast cancer cells. To evaluate the effectof paclitaxel on CAV1 expression, MCF-7 cells were grownto 50–70% confluence and treated with 10 nM paclitaxel for0, 24, or 48 h or with vehicle alone, 10 nM, or 100 nM pacli-taxel for 24 h. Real-time PCR showed a significant increase inCAV1 expression in MCF-7 cells within 24 or 48 h followingpaclitaxel treatment compared with control (untreated) (Fig.1A, p � 0.004, one-way ANOVA). Western blot analysisshowed a corresponding increase in CAV1 protein within 24or 48 h (Fig. 1B). Treatment of MCF-7 cells with 10 and 100nM for 24 h showed a significant increase in both CAV1transcription (Fig. 1C, p � 0.001, one-way ANOVA) and

protein expression (Fig. 1D) compared with controls (vehi-cle). Thus, in MCF-7 cells, a significant induction of CAV1occurs within 24 h following treatment with 10 nM paclitaxel.The induction of CAV1 following paclitaxel treatment sug-gests a role for CAV1 in drug responsiveness. Moreover,attenuating levels of endogenous CAV1 in MCF-7 cells withsiRNA correlates with reduction in paclitaxel-induced inhi-bition of cell growth (Fig. 2, A and B).Phosphorylation of CAV1 on Tyr-14 by Src is known to reg-

ulate CAV1 functions (21–25). To show whether Src inactiva-tion reduces paclitaxel-induced inhibition of cell growth, weused PP2 to inhibit Src activation as detected by phosphoryla-tion on Tyr-418 by Western blot analysis (Fig. 2C). WhereasCAV1 protein expression increases following treatment ofMCF-7 cells with 10 nMpaclitaxel, phospho-CAV1 (p-CAV1) isundetectable using our Western blot analysis protocol (for upto 50�g of protein loading/gel). However, incubation ofMCF-7cells with 10�MPP2 in addition to 10 nMpaclitaxel significantly

FIGURE 2. CAV1 is required for paclitaxel-induced cell growth inhibition. A, paclitaxel-induced up-regulation of CAV1 expression was inhibited with siRNAin MCF-7 cells. MCF-7 cells were transfected with either control or CAV1 siRNA for 48 h. Next, paclitaxel was added to the transfection medium to attain a finalconcentration of 10 nM; cells were incubated with paclitaxel for 24 h before Western blot analysis or cell proliferation assay. B, attenuation of CAV1 expressionin MCF-7 cells correlated with reduction in paclitaxel-induced inhibition of cell growth. *, p � 0.05 versus control siRNA with vehicle alone. C, incubation ofMCF-7 cells with 10 �M PP2 inhibited Src activation as detected by Western blot analysis of phosphor-Src(Y418). Although the level of CAV1 protein expressionincreased following treatment of MCF-7 cells with 10 nM paclitaxel, phospho-CAV1(Y14) was undetectable using our Western blot analysis protocol (for up to50 �g of protein loading/gel). Incubation of MCF-7 cells with 10 �M PP2 in addition to 10 nM paclitaxel significantly reduced cell growth inhibition effectof paclitaxel compared with paclitaxel alone (panel D, p � 0.05). *, p � 0.05 versus MCF-7 with paclitaxel alone.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

reduces the cell growth inhibition effect of paclitaxel comparedwith paclitaxel alone (Fig. 2D, p � 0.05).wtCAV1 Expression Enhanced Paclitaxel-induced Growth

Inhibition and Cell Cycle Arrest at G2/M—To establish thefunctional relevance of CAV1 in paclitaxel sensitivity, we gen-eratedMCF-7 cell lines that stably express either the full-lengthwild-type CAV1 (MCF-7/wtCAV1), Tyr-14 3 Phe phospho-rylation-deficient (MCF-7/Y14F) cells, or empty vector (MCF-7/EV) (Fig. 3A). Expression levels of CAV1 in MCF-7/wtCAV1and MCF-7/Y14F were measured by Western blot analysisusing specific antibodies for CAV1 or p-CAV1. CAV1� con-tains residues 1–178, whereas CAV1� contains residues32–178. Because Tyr-14 is the principal substrate for Srckinase, only CAV1� undergoes tyrosine phosphorylation (28).CAV1 is thought to interact with Src and inhibit its activation(2). Although not fully inactive, Src kinase activity is decreasedin untreated MCF-7/wtCAV1 cells compared with MCF-7/EVand MCF-7/Y14F cells (Fig. 3A).Expression of CAV1 in MCF-7 cells can reduce the rate of

proliferation (29). Comparison of growth curves in basal

medium shows a significant decrease in the rate of cell prolif-eration for MCF-7/wtCAV1 cells (p � 0.05) at day 4 throughday 6 compared with that for MCF-7/EV cells. In contrast, thedifference in rates of proliferation between MCF-7/EV andMCF-7/Y14F cells is not significant (Fig. 3B). Cells were treatedwith paclitaxel to determine whether CAV1 tyrosine phospho-rylation affects growth inhibition by the drug. MCF-7 cellsexpressing wtCAV1 are more sensitive than cells expressingeither EV or Y14F (Fig. 3C). These data suggest that CAV1phosphorylation on Tyr-14 plays a key role in paclitaxel-in-duced growth inhibition. Furthermore, the effects of CAV1 onpaclitaxel sensitivity are not simply a consequence of changes inthe rate of proliferation; cells that have a higher rate of prolif-eration are generally more sensitive to cell cycle-specific cyto-toxic drugs like taxanes (30). Additionally, we did not see anydifference in the partitioning of caveolar and intercellular com-partments following subcellular fractionation (data not shown)between wtCAV1 or the Y14F mutant-expressing cell undercontrol conditions or following paclitaxel treatment for 24 h.Thus, in terms of paclitaxel sensitivity, caveolar localization is

FIGURE 3. wtCAV1 increases sensitivity to paclitaxel in MCF-7 cells. Cells were transfected with pcDNA6 plasmid vector containing wild type (wt), phos-phorylation-deficient mutant (Y14F), or empty vector (EV) and stably selected. A, Western blot analysis of whole cell lysates from the MCF-7/EV, MCF-7/wtCAV1,and MCF-7/Y14F. Using specific antibodies, phospho-Src(Y418), total Src, phospho-CAV1(Y14), and total CAV1 were detected. GAPDH blot was used as theloading control. Note CAV1 phosphorylation on Tyr-14 was detected only in MCF-7/wtCAV1 cells. B, expression of wtCAV1, but not Y14F, decreased the growthrate of MCF-7 cells. Cells were seeded in quadruplicate in FBS-IMEM and trypsinized and counted at the indicated number of days. *, p � 0.05 versus MCF-7/EVat Days 4, 5, and 6 by Student’s t test. C, expression of wtCAV1 increased sensitivity of breast cancer cells to paclitaxel. Cells were cultured in quadruplicate withthe indicated concentrations of paclitaxel for 48 h. Data points are the mean of relative proliferation, bars � S.E. (n � 3).



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

unlikely to account for any differences between MCF-7 cellsexpressing wtCAV1 and Y14F.Taxanes stabilize microtubules and block sensitive cells in

the G2/M cell cycle phase (31). To determine whether CAV1expression alters G2/M cell cycle arrest following treatmentwith paclitaxel, MCF7-EV, MCF7-wtCAV1, or MCF-7/Y14Fcells were treated with either vehicle alone or 10 nM paclitaxelfor 24 h prior to fluorescence-activated cell sorter analysis ofcell cycle distribution (Fig. 4A). MCF-7/Y14F cells exhibited asignificant increase (p � 0.05) in the proportion of cells in theS-phase relative toMCF-7/EV. Following treatment with 10 nMpaclitaxel, the percentage of cells arrested in the G2/M phasewas significantly higher in MCF7-wtCAV1 compared withMCF-7/EV (p � 0.05). In MCF-7/Y14F cells, the percentage ofcells in theG2/Mphasewas comparablewith that inMCF-7/EVcells (Fig. 4B). Thus, the taxane-induced decrease in cell prolif-eration in MCF7-wtCAV1 cells is likely to be a consequence ofincreased cell cycle arrest in G2/M phase. To assess cellular

morphology in vehicle (control) or drug-treated cells, MCF-7/EV,MCF-7/wtCAV1, andMCF-7/Y14F cells were treated witheither vehicle or 10 nM paclitaxel for 24 h. Cells were fixed andpermeabilized and co-stained for both �-tubulin and CAV1. Inall three cell lines, vehicle-treated cells displayed an organizedtubulin network that excludes the nucleus and extendsthroughout the cytoplasm (Fig. 4C). After 24 h of paclitaxeltreatment, microtubule disruption wasmore distinct inMCF-7cells expressing wtCAV1 compared with MCF-7 cells express-ing EV or Y14F, as evident from clusters of microtubules orasters in the respective cells (Fig. 4D).Increased Apoptosis in MCF-7/wtCAV1-expressing Cells in

Response to Paclitaxel—Paclitaxel can induce apoptosis insome breast epithelial cells (32). To determine whether CAV1expression affects apoptosis, we treated MCF-7/EV, MCF-7/wtCAV1, andMCF-7/Y14F cells with 10 nM paclitaxel for 24 h.Apoptosis was detected by flow cytometry after staining forfluorescein isothiocyanate-conjugated Annexin V and for pro-

FIGURE 4. wtCAV1 increases the rate of cell cycle arrest in mitosis following paclitaxel treatment. Cells were treated with vehicle (A) or 10 nM paclitaxel (B)in FBS-IMEM for 24 h before cell cycle analysis. Columns, mean for three independent experiments (% total cells); bars, � S.E. *, p � 0.05 versus MCF-7/EV byStudent’s t test. Paclitaxel increased the number of cells arrested in mitosis in the presence of wtCAV1. Cells were treated with vehicle (C) or with 10 nM paclitaxel(D), fixed, permeabilized, and stained with anti-�-tubulin (green) and CAV1 (red) antibody. Confocal images showed very low levels of CAV1 in EV-expressingcells compared with wtCAV1- or Y14F-expressing cells. Cells with wtCAV1 showed a dramatic increase in cells that were arrested in mitosis at 24 h followingtreatment with 10 nM paclitaxel as visualized by clustering of �-tubulin. Scale bar, 10 �m.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

pidium iodide. At 10 nM paclitaxel for 24 h, only 1% of theMCF-7/EV or MCF-7/Y14F cells underwent apoptosis com-pared with 3% in MCF-7/wtCAV1 cells (Fig. 5A, p � 0.05).Additionally, cells were analyzed for apoptosis byWestern blotanalysis of anti-poly(ADP-ribose) polymerase cleavage to a85–89-kDa fragment (Fig. 5B). Paclitaxel-induced poly(ADP-ribose) polymerase cleavage appeared in MCF-7/wtCAV1within 24 h following drug treatment and was more pro-nounced at 48 h relative to MCF-7/EV or MCF-7/Y14F cells.Although the poly(ADP-ribose) polymerase antibody(according to the manufacturer) should detect only a singleband for cleaved poly(ADP-ribose) polymerase, we detecteda nonspecific higher molecular band at �100–110 kDa thatdid not accurately reflect levels of full-length poly(ADP-ri-bose) polymerase in our experiments. These findings suggestthat tyrosine phosphorylation of CAV1 accelerates apopto-sis in response to paclitaxel treatment and/or could be essen-tial to signaling pathway(s) that are required for the induc-tion of apoptosis.Microtubule disruption following treatment with taxanes

increases phosphorylation of BCL2(S70) in the G2/M phase ofthe cell cycle, abrogating the normal anti-apoptotic function ofBCL2 and initiating an apoptotic program in cycling cancercells (33). To determine whether activation of BCL2 phospho-rylation correlates with the degree of apoptosis, Western blotanalyses were done with whole cell lysate cells following treat-ment with vehicle or 10 nM paclitaxel for 24 or 48 h. At 24 h, wedetected a significant increase in the level of BCL2(S70) phos-

phorylation in MCF-7/wtCAV1cells in comparisonwithMCF-7/EVcells (p � 0.05). In cells expressingY14FCAV1, the level of BCL2(S70)phosphorylation remains compara-ble with cells expressing EV (Fig.6A). Furthermore, measurement ofmitochondrial permeability, whichis tightly associated with the releaseof pro-death molecules into thecytoplasm, is significantly increasedin MCF-7/wtCAV1 cells treatedwith paclitaxel in comparison tocells treated with vehicle alone (Fig.6B, p � 0.05).wtCAV1 Regulates p53-depen-

dent p21 Induction following Treat-ment with Paclitaxel—Low dose(�10 nM) paclitaxel treatment canincrease p21 synthesis through ap53-dependent pathway (34, 35),and CAV1 and p53 can induceeach other in fibroblasts (36). Inhi-bition of p53 expression withsiRNA in MCF-7/wtCAV1 cellsprevents paclitaxel-induced p21protein expression (Fig. 7, A–C).Thus, in wtCAV1-expressing cells,induction of p21 following pacli-taxel treatment is p53-dependent.

To determine how expression of CAV1 interferes with p53activity, we transiently expressedMCF-7/EV,MCF-7/wtCAV1,and MCF-7/Y14F with a p53-responsive promoter-reporterconstruct. Following 24 h of treatment with paclitaxel, p53-responsive element activation is significantly decreased inMCF-7/wtCAV1-expressing cells relative to vehicle-treatedcells (Fig. 7D, p � 0.05); this activation remains unchangedin MCF-7/EV andMCF-7/Y14F cells relative to their respec-tive controls.MCF-7 cells express a functional p53, and treatment with

paclitaxel enhances p53 expression and function as seen in anup-regulation of p21 and changes in cell cycle progression andapoptosis. Induction of p21 confers resistance to the cytotoxiceffects of taxanes inMCF-7 cells (37). Transient transfection ofa full-length p21 promoter-reporter construct showed that p21promoter activity in MCF-7/wtCAV1 cells corresponds withp53-responsive element activation under control or drug treat-ment conditions. In MCF-7/EV and MCF-7/Y14F cells, p21promoter activity following vehicle or drug treatment remainsunchanged, as also occurswith p53 transcriptional activation inthese cells (Fig. 7E).

We measured the level of p53 and p21 protein expression inMCF-7/EV, MCF-7/wtCAV1, and MCF-7/Y14F cells. Within24 and 48 h of drug treatment, p53 and p21 protein levelsincreased in all the three cell lines (Fig. 8, A and B). However,the increase in MCF-7/wtCAV1 cells was significantly lower atboth 24 and 48 h following drug treatment when comparedwith MCF-7/EV or MCF7-Y14FCAV1 cells (p � 0.05). CAV1

FIGURE 5. wtCAV1 enhances apoptosis in MCF-7 cells following paclitaxel treatment. A, at 24 h, apoptosisin 10 nM paclitaxel- or ethanol-treated cells was detected by flow cytometry using fluorescein isothiocyanate-conjugated Annexin V staining. Columns, mean for three independent experiments (% total cells); bars, � S.E.*, p � 0.05 versus control-treated MCF-7/wtCAV1 cells by Student’s t test. B, cells expressing wtCAV1 showedincreased poly(ADP-ribose) polymerase cleavage at 24 and 48 h subsequent to 10 nM paclitaxel treatment.GAPDH was used as a loading control.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

protein expression in both MCF-7/wtCAV1- and MCF-7/Y14F-expressing cells was increased at 48 h following drugtreatment. p53 can initiate both cell cycle arrest and apoptosis,and an inhibition of p53-dependent p21 induction may drivecells toward apoptosis (38). These results suggest that in MCF-7/wtCAV1 cells, phosphorylated CAV1 on Tyr-14 regulatesinduction of p53 and p21 following treatment with paclitaxel

and is key in initiating the proapoptotic signal upstream of bothp53 and p21.

DISCUSSION

CAV1 is a versatile protein that has received increased atten-tion in cancer because of its role in cell survival (39). CAV1proteins form the foundation of cholesterol-rich membrane

FIGURE 6. Paclitaxel treatment of MCF-7/wtCAV1 cells increases BCL2(S70) phosphorylation. A, cells were treated with 10 nM paclitaxel for 24 or 48 h andphosophorylated-BCL2(S70) was detected by immunoblot analysis using a phospho-specific antibody (upper panel). Lower panels, immunoblot analysis of totalBCL2 and GAPDH (loading control). B, mitochondrial permeability in MCF-7/wtCAV1 cells is significantly increased compared with control within 24 h followingpaclitaxel treatment (10 nM). *, p � 0.05 versus control-treated MCF-7/wtCAV1 cells by Student’s t test.

FIGURE 7. p53 is required for paclitaxel-induced expression of p21 in MCF-7/EV (A), MCF-7/wtCAV1 (B), and MCF-7/Y14F (C) cells. Cells were transfectedwith control, p21, or p53 siRNA for 48 h before treatment with 10 nM paclitaxel for 24 h. Induction of p53 and p21 was detected by immunoblot analysisusing specific antibodies (top and middle panels). Lower panels, GAPDH immunoblot was used as a loading control. Expression of wtCAV1 and notY14F-CAV1 regulates p53-dependent up-regulation of p21. Activation of p53-responsive element (D) and p21 promoter (E) in MCF-7/EV, MCF-7/wtCAV1, or MCF-7/Y14FCAV1 cells by luciferase assay. Experiments were done in quadruplicate. The results were normalized to Renilla luminescenceand are presented as relative luciferase activity to respective cells under control conditions. *, p � 0.05 versus control-treated MCF-7/wtCAV1 cells byStudent’s t test.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

microdomains called caveolae, within which CAV1 regulatesthe functions of various signaling proteins (1). Among themultiple cellular events that are regulated by CAV1, regula-tion of tumorigenesis remains controversial. Alteration ormodification of CAV1 expression is common in some trans-formed cells, and understanding the signaling pathwaysaffected by CAV1 may provide novel insight into new cancertherapies. In the present study, we show that treatment with 10nM paclitaxel increased CAV1 protein expression in MCF-7breast cancer cells within 24 h; MCF-7 cells express very lowlevels of CAV1 under basal conditions. Down-regulation ofCAV1 expression with siRNA correlated with decreased pacli-taxel-induced growth inhibition. Regulation of CAV1 signalinghas been correlated with Src-mediated tyrosine phosphoryla-tion at Tyr-14 (21–25, 40, 41). We show that preventing Srcactivation with PP2 reduced paclitaxel-induced growth inhibi-tion. We found that expression of wtCAV1, but not the phos-phorylation-deficient mutant Y14F, augmented sensitivity topaclitaxel in MCF-7 cells and that tyrosine phosphorylation atTyr-14 is essential inmediating themolecular events leading toapoptosis.Our results show that expression of wtCAV1, but not Y14F,

increases the inhibition of cell growth following treatment withpaclitaxel. The inhibition of cell proliferation was associatedwith an increase in the blockade of cell cycle progression in cellsharboringwtCAV1, and this associationmay explain the reduc-tion in cell growth following paclitaxel treatment. Subsequentto treatmentwith taxanes, cells arrest in theG2/M-phase due tomicrotubule stabilization, an event followed by the onset ofapoptosis. Thus, our observations imply an interplay betweencytoskeletal integrity and nuclear transcriptional events.MCF-7 cells express very low levels of endogenous CAV1 andexhibit both a time- and dose-dependent up-regulation ofCAV1 in response to paclitaxel treatment. Because microtu-bules are involved in the distribution of CAV1 between theGolgi/endoplasmic reticulum system and caveolae (42), inter-ruption of CAV1 delivery by microtubule stabilization follow-ing paclitaxel treatment could trigger the cell to increase CAV1expression. The spatial requirement for CAV1 in taxane-in-

duced cell cycle arrest and subsequent cell death is currentlybeing investigated in our laboratory.Accumulating evidence suggests a possible role for CAV1 in

the initiation of cell death. For example, CAV1 is up-regulatedin A549 lung cancer cells in response to paclitaxel (15), andthe induction of apoptosis by several agents can significantlyincrease CAV1 expression in some cells (18). Paclitaxel-in-duced apoptosis inMCF-7 cells, as shownbyAnnexinV stainingand poly(ADP-ribose) polymerase cleavage, significantlyincreased in the presence of wtCAV1. Phosphorylation/inacti-vation of BCL2, as induced by paclitaxel (43, 44), is associatedwith G2/M cell cycle arrest (45). Increased expression of phos-pho-BCL2(S70) and mitochondrial permeability in MCF-7/wt-CAV1 cells when comparedwithMCF-7/Y14F cells imply a rolefor CAV1 phosphorylation in affecting the mitochondrial-dependent, intrinsic apoptosis pathway in paclitaxel-inducedcell death.Clinical studies have shown that taxane-containing regimens

are associatedwith a statistically significant improvement in over-all survival compared with nontaxane-containing regimens; how-ever, these regimens are associated with leukopenia and neuro-toxicity (46). Considering the heterogeneity among individuals,it is essential to discovermolecular biomarkers in breast cancercells that could help improve the benefits of taxanes by increas-ing the efficacy-to-toxicity ratio. Tumors that show the highestdegree ofG2/Mphase arrest, BCL2phosphorylation, and loworabsent p21 after drug treatment should exhibit sensitivity topaclitaxel. BCL2, an anti-apoptotic protein, is transcriptionallyregulated by factors such as p53 (47). Paclitaxel induces p21 inboth p53-wild type and p53-null cells (48). Although CAV1may be a target in p53-dependent signaling (5, 49), the signifi-cance of this interaction in apoptosis is unknown. Inhibition ofp53 expression in MCF-7 cells expressing either wtCAV1,Y14F, or EVprevented the induction of p21 following paclitaxeltreatment (Fig. 7,A–C). Thus, in cells with a functional p53, thecontribution of CAV1 to paclitaxel-mediated cytotoxicityappears to be p53-dependent.A significant decrease in p53- and p21-luciferase reporter

construct activation was observed within 24 h following treat-

FIGURE 8. Paclitaxel-induced increase in p21 expression is regulated by CAV1 Tyr-14 phosphorylation. A, differential induction of p21 in MCF-7 cellsexpressing EV, wtCAV1, or Y14F was determined by Western blot analysis following treatment of cells with 10 nM paclitaxel for 0, 24, or 48 h. Total protein wasanalyzed for p53, p21, phospho-CAV1, and total CAV1. GAPDH was used as a loading control. B, graphical representation of p21:GAPDH ratio from threeindependent experiments. ANOVA, p � 0.005; *, p � 0.05 versus MCF-7/EV.



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

ment with paclitaxel in MCF-7/wtCAV1 cells (Fig. 7, D and E).In contrast to its role in cell cycle arrest, p21 can increase cellsurvival following paclitaxel treatment inMCF-7 cells and inhi-bition of p21 can increase paclitaxel cytotoxicity inMCF-7 cells(37). Thus, expression of wtCAV1 inMCF-7 cells may increasethe rate of apoptosis inMCF-7/wtCAV1 cells by down-regulat-ing p21 expression. To our knowledge, this is the first reportshowing that phosphorylation of CAV1 at Tyr-14 sensitizescells to BCL2-mediated apoptosis in a p53/p21-dependentmanner.It is likely that sensitivity as well as resistance to taxanes is

dependent on the function of a panel of genes rather than asingle gene or pathway.Our data suggest thatCAV1 could be anessential component of signal transduction in taxane-mediatedcell death. Up-regulation of CAV1 protein following treatmentwith paclitaxel has prompted investigators to implicate it as amarker for cytotoxicity and cell death in lung cancer cells (15).However, it is unclear how CAV1 regulates cellular events atthe onset of apoptosis. Expression of a phosphorylation-defi-cient Y14F mutant failed to augment sensitivity to paclitaxelwhen compared with cells expressing wtCAV1 in MCF-7 cells.This observation emphasizes the role of events that regulatephosphorylation of CAV1 in taxane sensitivity. Furthermore,CAV1 acts by regulating paclitaxel-induced apoptosis in aman-ner that is dependent upon p53/p21 signaling and suppressionof BCL2. Thus, this study provides direct experimental evi-dence of the role of CAV1 and its phosphorylation at Tyr-14 inaffecting sensitivity to paclitaxel in p53-wild type breast cancercells. Further analyses of the involvement of CAV1-mediatedsignaling in the taxane responsiveness in breast cancer cells areunderway in our laboratory and may prove to be important forpredicting tumor response to chemotherapy with these drugs.

REFERENCES1. Okamoto, T., Schlegel, A., Scherer, P. E., and Lisanti, M. P. (1998) J. Biol.

Chem. 273, 5419–54222. Li, S., Couet, J., and Lisanti, M. P. (1996) J. Biol. Chem. 271, 29182–291903. Sagara, Y., Mimori, K., Yoshinaga, K., Tanaka, F., Nishida, K., Ohno, S.,

Inoue, H., and Mori, M. (2004) Br. J. Cancer 91, 959–9654. Sloan, E. K., Stanley, K. L., and Anderson, R. L. (2004) Oncogene 23,

7893–78975. Lee, S.W., Reimer, C. L., Oh, P., Campbell, D. B., and Schnitzer, J. E. (1998)

Oncogene 16, 1391–13976. Zajchowski, D. A., Bartholdi, M. F., Gong, Y., Webster, L., Liu, H. L.,

Munishkin, A., Beauheim, C., Harvey, S., Ethier, S. P., and Johnson, P. H.(2001) Cancer Res. 61, 5168–5178

7. Xie, Z., Zeng, X., Waldman, T., and Glazer, R. I. (2003) Cancer Res. 63,5370–5375

8. Manfredi, J. J., and Horwitz, S. B. (1984) Pharmacol. Ther. 25, 83–1259. Yvon, A. M., Wadsworth, P., and Jordan, M. A. (1999)Mol. Biol. Cell. 10,

947–95910. Dumontet, C., and Sikic, B. (1999) J. Clin. Oncol. 17, 1061–107011. Gligorov, J., and Lotz, J. P. (2004) Oncologist 9, 3–812. Clavarezza, M., Del Mastro, L., and Venturini, M. (2006) Ann Oncol. 17,

(Suppl. 7) vii22–vii2613. Trock, B. J., Leonessa, F., and Clarke, R. (1997) J. Natl. Cancer Inst. 89,

917–93114. Mundy, D. I., Machleidt, T., Ying, Y. S., Anderson, R. G., and Bloom, G. S.

(2002) J. Cell Sci. 115, 4327–4339

15. Roussel, E., Belanger, M. M., and Couet, J. (2004) Anticancer Drugs 15,961–967

16. Yang, C. P., Galbiati, F., Volonte, D., Horwitz, S. B., and Lisanti, M. P.(1998) FEBS Lett. 439, 368–372

17. Couet, J., Belanger, M.M., Roussel, E., and Drolet, M. C. (2001)Adv. DrugDeliv. Rev. 49, 223–235

18. Belanger, M. M., Roussel, E., and Couet, J. (2003) Anticancer Drugs 14,281–287

19. Gargalovic, P., and Dory, L. (2003) J. Lipid Res. 44, 1622–163220. Glenney, J. R., Jr., and Zokas, L. (1989) J. Cell Biol. 108, 2401–240821. Lee, H., Volonte, D., Galbiati, F., Iyengar, P., Lublin, D.M., Bregman, D. B.,

Wilson, M. T., Campos-Gonzalez, R., Bouzahzah, B., Pestell, R. G.,Scherer, P. E., and Lisanti, M. P. (2000)Mol. Endocrinol. 14, 1750–1775

22. Ko, Y. G., Liu, P., Pathak, R. K., Craig, L. C., and Anderson, R. G. (1998)J. Cell. Biochem. 71, 524–535

23. Shajahan, A. N., Tiruppathi, C., Smrcka, A. V., Malik, A. B., andMinshall,R. D. (2004) J. Biol. Chem. 279, 48055–48062

24. Labrecque, L., Nyalendo, C., Langlois, S., Durocher, Y., Roghi, C.,Murphy,G., Gingras, D., and Beliveau, R. (2004) J. Biol. Chem. 279, 52132–52140

25. Kiss, A. L., Turi, A., Mullner, N., Kovacs, E., Botos, E., and Greger, A.(2005)Mol. Cell. Endocrinol. 245, 128–137

26. Boudny, V., and Nakano, S. (2002) Br. J. Cancer. 86, 463–46927. Chen, T., Pengetnze, Y., and Taylor, C. C. (2005) Mol. Cancer Ther. 4,

217–22428. Li, S., Seitz, R., and Lisanti, M. P. (1996) J. Biol. Chem. 271, 3863–386829. Fiucci, G., Ravid, D., Reich, R., and Liscovitch, M. (2002) Oncogene 21,

2365–237530. Matsuoka, H., Furusawa, M., Tomoda, H., and Seo, Y. (1994) Anticancer

Res. 14, 163–16731. Schiff, P. B., and Horwitz, S. B. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,

1561–156532. Yeung, T. K., Germond, C., Chen, X., and Wang, Z. (1999) Biochem.

Biophys. Res. Commu. 263, 398–40433. Haldar, S., Basu, A., and Croce, C. M. (1997) Cancer Res. 57, 229–23334. Heliez, C., Baricault, L., Barboule, N., and Valette, A. (2003)Oncogene 22,

3260–326835. Panno,M. L., Giordano, F.,Mastroianni, F.,Morelli, C., Brunelli, E., Palma,

M. G., Pellegrino, M., Aquila, S., Miglietta, A., Mauro, L., Bonofiglio, D.,and Ando, S. (2006) FEBS Lett. 580, 2371–2380

36. Galbiati, F., Volonte, D., Liu, J., Capozza, F., Frank, P. G., Zhu, L., Pestell,R. G., and Lisanti, M. P. (2001)Mol. Biol. Cell 12, 2229–2244

37. Barboule, N., Chadebech, P., Baldin, V., Vidal, S., and Valette, A. (1997)Oncogene 15, 2867–2875

38. Yu, J., and Zhang, L. (2005) Biochem. Biophys. Res. Commun. 331,851–858

39. Williams, T.M., and Lisanti, M. P. (2005)Am. J. Physiol. 288,C494–C50640. Kim, Y. N., Dam, P., and Bertics, P. J. (2002) Exp. Cell Res. 280, 134–14741. Orlichenko, L., Huang, B., Krueger, E., and McNiven, M. A. (2006) J. Biol.

Chem. 281, 4570–457942. Conrad, P. A., Smart, E. J., Ying, Y.-S., Anderson, R. G. W., and Bloom,

G. S. (1995) J. Cell Biol. 131, 1421–143343. Blagosklonny,M. V., Schulte, T.W., Nguyen, P., Mimnaugh, E. G., Trepel,

J., and Neckers, L. (1995) Cancer Res. 55, 4623–462644. Srivastava, R. K., Mi, Q. S., Hardwick, J. M., and Longo, D. L. (1999) Proc.

Natl. Acad. Sci. U. S. A. 96, 3775–378045. Ling, Y. E., Tornos, C., and Perez-Soler, R. (1998) J. Biol. Chem. 273,

18984–1899146. Ghersi, D., Wilcken, N., and Simes, R. J. (2005) Br. J. Cancer. 93, 293–30147. Decary, S., Decesse, J. T., Ogryzko, V., Reed, J. C., Naguibneva, I., Harel-

Bellan, A., and Cremisi, C. E. (2002)Mol. Cell. Biol. 22, 7877–788848. Blagosklonny, M. V., Giannakakou, P., el-Deiry, W. S., Kingston, D. G.,

Higgs, P. I., Neckers, L., and Fojo, T. (1997) Cancer Res. 57, 130–13549. Ravid, D., Maor, S., Werner, H., and Liscovitch, M. (2005) Oncogene 24,

1338–1347



at Georgetow

n Univ M

edical Center on F

ebruary 20, 2007 w

ww

.jbc.orgD

ownloaded from

http://www.jbc.org

Documents

Caveolin-1 Tyrosine Phosphorylation Enhances Paclitaxel-mediated Cytotoxicity