dM

r f

in

a

Biochemical Pharmacology 87 (2014) 292302

Contents lists available at ScienceDirect

Biochemical Ph

jo u rn al h om epag e: ww w.els eviOntario Cancer Institute/Princess Margaret Hospital, University of Toronto, 610 University Avenue, Toronto, Ontario, Canada M5G 2M9

1. Introduction

Breast cancer remains the most commonly diagnosed cancerand the leading cause of cancer death among women [1]. Despiteadvances achieved in early detection as well as therapeuticstrategies with surgery, radiotherapy, and chemotherapy, theoutcome for breast cancer patients remains poor. A major obstacleto a more effective cure for this disease is attributed to the abilityof cancer cells to acquire resistance to chemotherapeutic agents,and, mostly, to multidrug resistance (MDR), a phenomenon inwhich tumors acquire cross chemoresistance to a number ofirrelevant chemotherapeutic agents after exposure to a single

particular cytotoxic drug [2]. Albeit the fact that the exactmechanism of MDR remains obscure, abundant evidence conrmsthat the majority of MDR are due to overexpression of proteinsbelonging to the ATP binding cassette (ABC) transporter super-family. Among these proteins, the mammalian P-glycoprotein (P-gp), which is encoded by the MDR1 (ABCB1) gene, is the most wellstudied for its ability to induce a MDR phenotype [3]. P-gp confersMDR by acting as an ATP-dependent pump, which activelyefuxes drugs out of cancer cells. In addition to regulating MDR,several studies demonstrated additional functions of P-gp, such aspromotion of tumor cell proliferation and inhibition of apoptosis,during cancer progression [410]. Furthermore, overexpression ofP-gp has been shown to be correlated with a more aggressivephenotype and with poor prognosis in many forms of malignan-cies [1118]. Accumulating evidence also suggested that P-gpdirectly promotes invasion and metastasis of cancer cells in vitro[1921]. Consistently, many studies observed that drug-resistantcancer cells always displayed a more invasive phenotype in

A R T I C L E I N F O

Article history:

Received 17 September 2013

Accepted 5 November 2013

Available online 15 November 2013

Keywords:

P-glycoprotein

Anxa2

MDR

Invasion

Breast cancer

Src

A B S T R A C T

Several recent studies have suggested that the acquisition of the multidrug resistance (MDR) phenotype

is associated with elevated invasion and metastasis of tumor cells. P-glycoprotein (P-gp), the major

determinant in the generation of the MDR phenotype, was reported to be correlated with a more

aggressive phenotype and poor prognosis in many forms of malignancies. However, a clear

understanding of the association is still lacking. We previously showed that Anxa2, a calcium-

dependent phospholipid-binding protein, interacts with P-gp and contributes to the invasiveness of

MDR breast cancer cells. In the present study, a strong positive correlation between MDR1 and Anxa2

mRNA expression in invasive breast cancer tissues during cancer progression was observed. In addition,

exposure to adriamycin signicantly enhanced motility in breast cancer cells and increased levels of P-gp

and Anxa2. Moreover, inhibition of P-gp activity, using selective P-gp modulators, was found to

signicantly inhibit the invasive capacity of MCF-7/ADR cells without affecting the interaction and co-

localization between P-gp and Anxa2. However, suppression of P-gp pump activity and knockdown of

MDR1 expression both disrupted adriamycin-induced Anxa2 phosphorylation. Interestingly, P-gp was

further demonstrated to interact with Src, a tyrosine kinase upstream of Anxa2. Taken together, our

results indicate that P-gp may promote the invasion of MDR breast cancer cells by modulating the

tyrosine phosphorylation of Anxa2. The interaction between Anxa2 and P-gp is possibly, at least in part,

responsible for the association between MDR and invasive potential in breast cancer cells.

2013 Elsevier Inc. All rights reserved.

* Corresponding author. Tel.: +86 022 23340123 6018;

fax: +86 022 23340123 6018.

E-mail address: niuruifang@tjmuch.com (R. Niu).1 These authors contributed equally to this work.

0006-2952/$ see front matter 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.bcp.2013.11.003P-glycoprotein associates with Anxa2 anmultidrug resistant breast cancer cells

Fei Zhang a,1, Haichang Zhang a,b,1, Zhiyong Wang a, Yi Yang a, Ruifang Niu a,*a Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Cente

Therapy, Ministry of Education, Key Laboratory of Cancer Prevention and Therapy, TianjbDepartment of Nuclear Medicine, Tianjin First Center Hospital, Tianjin 300192, PR Chinc promotes invasion in

an Yu c, Ran Tian a, Wei Ji a,

or Cancer, The Key Laboratory of Breast Cancer Prevention and

300060, PR China

armacology

er .c o m/lo cat e/bio c hem p har m

comparison with that of drug-sensitive parental cells [19,2226].Overall, these results suggest that a functional association existsbetween P-gp/MDR and tumor invasion/metastasis. Therefore, abetter understanding of the exact roles played by P-gp inregulating cancer cell invasion/metastasis and its underlyingmolecular mechanism(s) will be benecial for establishment ofnovel and more effective treatment options.

Anxa2 is a 36 kDa calcium-dependent phospholipid-bindingprotein and exists as either a monomer or a heterotetramericcomplex with other proteins such as S100A10 [27]. Anxa2 is

2. Materials and methods

2.1. Cell culture

Human breast cancer adriamycin-sensitive MCF-7 cells andtheir adriamycin-resistant counterpart MCF-7/ADR cells wereprovided by Dr. Zizheng Hou of Henry Ford Hospital, Detroit, MI,USA. Both cell lines were cultured in improved RPMI-1640 mediumcontaining 10% fetal bovine serum (FBS, Hyclone, Logan, UT, USA)and 1% penicillin/streptomycin (Gibco BRL, Rockville, MD). MCF-7/

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302 293implicated in several pivotal cellular processes, such as cellproliferation, cell adhesion, cell motility, actin rearrangement,angiogenesis, and metastasis [2830]. High expression level ofAnxa2 was observed in many cancer types and was related withcancer progression [2830]. Interestingly, increased expression ofAnxa2 was also reported in MDR cancer cells [3133]. Moreover,upregulation of Anxa2 was signicantly associated with rapidrecurrence of cancer in patients after chemotherapy [3436]. In aprevious study, we showed that knockdown of Anxa2 inhibitedinvasiveness of MDR MCF-7 breast cancer cells [31]. Theseobservations are in support of a possible association betweenAnxa2 and MDR cancer cell invasion.

The functional activity of Anxa2 is regulated by post-translational modication and proteinprotein interaction. Tyro-sine 23 phosphorylation of Anxa2 regulates actin reorganizationand cell adhesion [37,38]. Another study also showed thattyrosine phosphorylation of Anxa2 is required for invasioncapacity and epithelial-mesenchymal transition (EMT) of pancre-atic ductal adenocarcinoma [39]. Phosphorylation of Anxa2mediates glucocorticoid resistance in MLL-rearranged infantacute lymphoblastic leukemia. In addition, proteinproteininteraction also modulates its functions in cancer cells [30]. Forinstance, Anxa2 interacts with HAb18G/CD147 and promotesinvasion and migration of human hepatocellular carcinoma cells[40]. Moreover, Anxa2 interacts with phosphorylated STAT6 andcontributes to the metastasis of prostate cancer [41]. Recently, weidentied a novel interaction between Anxa2 and P-gp in MDRMCF-7 cells [42]. However, the functional role of the interactionbetween Anxa2 and P-gp in MDR cancer cell migration/invasion isnot fully understood.

In the present study, we investigated the role of P-gp and Anxa2in drug resistance-induced migration and invasiveness of breastcancer cells. We provided evidence that there exists a functionalassociation between the MDR phenotype and the enhancedinvasive potential of breast cancer cells. A strong positivecorrelation between P-gp and Anxa2 mRNA expression wasrevealed in invasive breast cancer tissues during cancer progres-sion. We also showed concurrent elevated expression of P-gp andAnxa2 in drug-treated breast cancer cells along with enhanced cellmotility. Furthermore, P-gp pump activity was able to mediateadriamycin-induced tyrosine phosphorylation of Anxa2 in MDRcancer cells and contribute to MDR cancer cell migration andinvasion. Our ndings suggest a novel regulatory mechanismemployed by P-gp in promoting MDR cancer cell migration andinvasion.

Table 1PCR primers and PCR reaction conditions.

Gene Primer sequence

Anxa2 Sense: 50 GGAGATGACTGAAGCCCGACA 30

Antisense: 50 GAAGGCCAGGCAATGCTTAGG 30

MDR1 Sense: 50 TCCTGGAGCGGTTCTACGAC 30

Antisense: 50 ATGGGCTCCTGGGACACGAT 30

b-Actin Sense: 50 CAGAGCAAGAGAGGCATCC 30

Antisense: 50 CTGGGGTGTTGAAGGTCTC 30ADR cells were cultured in the presence of 0.5 mM adriamycin tomaintain the MDR phenotype while being grown in a drug-freemedium for at least 1 week prior to the experiment. P-gp stableknockdown MCF-7/ADR and control cells were obtained in ourprevious study [42], and routinely cultured in RPMI-1640 mediumin the presence of 150 mg/mL hygromycin B (BD Biosciences, SanJose, CA, USA).

2.2. RNA extraction, reverse transcription and real-time PCR

All tissue specimens including 20 cases of ductal carcinoma insitu, 100 cases of invasive ductal carcinoma, and 50 cases ofadjacent normal tissues (partially matched) were collected frompatients diagnosed with breast cancer in Tianjin MedicalUniversity Cancer Institute and Hospital (PR China) betweenOctober 2003 and November 2008. The patients age at diagnosiswas from 30 years to 81 years old, with a median age of 49. Allpatients received surgical therapy, and all tissue specimens werepathologically conrmed by using hematoxylineosin staining.Tissue samples were excised from surgical resected specimens,weighted and immediately snap-frozen in liquid nitrogen andstored at 80 8C until use. All cancer tissues selected in the presentstudy consisted of at least 70% cancer cells, and normal tissuescontaining more than 50% epithelial cells were used. This studywas approved by the Medical Ethics Committee of Tianjin MedicalUniversity and all patients provided written informed consent.RNA extraction was performed using Trizol reagent (Invitrogen,Carlsbad, CA, USA), and 1 mg of total RNA was reverse-transcribedinto cDNA using an M-MLV RT kit (Takara Bio, Dalian, P.R. China)according to the manufacturers instructions. Quantitative real-time PCR was performed using the SYBR Premix Ex Taq (Takara Bio,Dalian, PR China). PCR primers and reaction conditions aredescribed in Table 1. The relative expression levels of MDR1 andAnxa2 mRNA were determined using deltadelta CT method, andb-actin was utilized as reference gene to normalize geneexpression.

2.3. Drug exposure and wound healing assay

Human breast cancer MCF-7 cells were cultured in 6 cm dishesand continuously exposed to approximately 5% of IC50 doses ofadriamycin (0.15 mM) for 6 weeks, during which the medium waschanged every 3 d, and cells were subcultured to 8090%conuence. IC50 was determined using MTT assay as describedpreviously [31]. After exposure to the drugs for 1, 2, 4, and 6 weeks,

Annealing temperature (8C) Product length (bp)

60 176

60 118

60 217

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302294wound healing assay was performed to examine the changes in cellmigration ability compared with WT MCF-7 cells. In brief, cellswere cultured to 100% conuence in 3.5 cm dishes and were thenwounded using a 10 mL pipette tips, washed with PBS to removecell debris, and incubated in a medium containing 2% FBS at 37 8Cfor 0, 4, 8, 12, 24, and 36 h. The wound areas were photographed,and migration distance was quantied under a light microscope.

2.4. In vitro cell invasion assay

Cell invasion assay was performed using Boyden chamber assayas described previously [31]. Briey, cells were trypsinized,washed with PBS, suspended and adjusted to a concentration of5 105 cells/mL in FBS-free DMEM medium. Subsequently, 200 mLof cells were loaded into the upper chamber containing 8 mmporous membranes pre-coated with matrigel. The lower chamberwas lled with 10% FBS-containing medium. After incubation at37 8C for 1236 h, the invaded cells through the membrane werexed, stained, and counted using a light microscope at 400 (high-powered eld). All assays were performed in triplicate.

2.5. Western blotting and co-immunoprecipitation assays

Western blotting was performed as described previously [31].In brief, cells cultured in 6 cm dishes were washed with PBS andlysed in 400 mL of 1 SDS cell lysis buffer. The cell lysates werequantied with a BCA protein assay kit. Then, 1040 mg of totalprotein was separated by SDS-PAGE and transferred onto a PVDFmembrane. After blocking the membrane with 5% non-fat milk/TBST at room temperature for 1 h, the membranes were incubatedwith corresponding primary antibodies overnight at 4 8C. Theantibodies and dilutions are as follows: P-gp (H-241, 1:500, SantaCruz Biotechnology, CA, USA), Anxa2 (C10, 1:5000, Santa CruzBiotechnology, CA, USA), p-Anxa2 (1:200, Santa Cruz Biotechnolo-gy, CA, USA), and b-actin (1:5000, Santa Cruz Biotechnology, CA,USA). Then, the membrane was washed three times with TBST andincubated with HRP-linked secondary antibodies, followed bydetection with an ECL kit (Millipore, Billerica, MA, USA) accordingto the manufacturers instructions. b-Actin was used as an internalcontrol. For co-immunoprecipitation assay, MCF-7/ADR cells wereseeded in 10 cm dishes and cultured to 90% conuence. Then, thecells were washed three times with serum-free medium andtreated with or without P-gp inhibitors in medium containing 0.5%FBS for 24 h. After washing the dishes three times with ice-coldPBS, cells were lysed in ice-cold cell lysis buffer (40 mM Tris,150 mM NaCl, 1% Triton X-100, 50 mM NaF, 5 mM Na3VO4, 2 mMEDTA, and protease inhibitor cocktail) on ice for 30 min. The celllysates were transferred to microcentrifuge tubes and centrifugedat 12,000 g for 15 min at 4 8C. The supernatants were transferredto another tube, and the protein concentration was quantiedusing a BCA protein assay kit. After preclearing the protein lysatesby adding 50 mL of protein A-agarose beads (Invitrogen, Carlsbad,CA, USA), the supernatants were then immunoprecipitated with2 mg anti-P-gp antibodies by incubating the antigen-antibodycomplex overnight at 4 8C. The immunocomplex was capturedwith protein A-agarose beads, washed with cell lysis buffer, andfurther analyzed by Western blotting as described above. Normalrabbit IgG (Santa Cruz Biotechnology, CA, USA) was applied as anegative control.

2.6. Immunouorescence confocal microscopy analysis

Cells were seeded in 12-well plates containing glass coverslipsat a density of 5 104 cells/mL per well and cultured for 24 h.Then, the cells were washed three times with serum-free mediumand incubated in 0.5% FBS-containing medium in the presence orabsence of P-gp inhibitors for another 24 h. After washing with PBSthree times, cells were xed with 4% paraformaldehyde at roomtemperature for 10 min, permeabilized in freshly prepared 0.1%saponin in PBS for 10 min, blocked in 3% BSA/PBS at roomtemperature for 1 h, and incubated with anti-P-gp (1:100) andanti-Anxa2 (1:200) antibodies in a humid chamber at 4 8Covernight. After repeated washing with PBS, cells were stainedwith Alexa Fluor 488 (anti-rabbit) and Alexa Fluor 546 (anti-mouse) conjugated secondary antibodies at room temperature for1 h, followed by mounting the coverslips with Mowiol-based anti-fading medium. Finally, the subcellular localization of Anxa2 andP-gp in MCF-7/ADR cells were visualized under a laser scanningconfocal microscope (Leica TCS SP5).

2.7. Flow cytometry assay

Fluorescent dye rhodamine 123 (Rh-123) is a substrate for P-gp.P-gp activity can be assessed by measuring the efux of Rh-123. Todetermine whether the P-gp modulators are efcient in inhibitingthe activity of P-gp, Rh123 efux assay was performed using owcytometry. Cells were cultured in 10 cm dishes at a density of1 106 cells/mL and grown to 90% conuence. Cells were thenharvested, washed twice with RPMI-1640 medium, and adjusted toa concentration of 2 106/mL cells. Then, 5 mg/mL of Rh-123 wasadded, and cells were further incubated for 30 min at 37 8C. Afterthe extracellular-free dye was removed, cells were transferred andcultured in dye-free medium or in medium containing differentconcentrations of verapamil (VRP, SigmaAldrich, St. Louis, MO,USA) (0, 1, 5, and 30 mg/mL) or triuiperazine (TFP, SigmaAldrich,St. Louis, MO, USA) (0, 1, 5, and 15 mg/mL) for 10 min at 37 8C. Afterwashing twice with ice-cold PBS, cells were analyzed by owcytometry (Beckman Coulter EPICS). The assay was performed intriplicate.

2.8. Statistical analysis

All data were analyzed by the Graphpad Prim 5.02 software.The expression levels of MDR1 and Anxa2 mRNA were notnormally distributed. The comparison of mRNA expression levelsamong different groups was performed using the Wilcoxon test(for two groups) or the KruskalWallis test followed by Dunnstest (for three groups). The correlation between the expressionlevels of Anxa2 and P-gp was evaluated by using the Pearsoncorrelation test. For cell migration or invasion assay, quantita-tive data were presented as mean SD. The comparisons betweendifferent groups were performed using the one-way ANOVA. Pvalues less than 0.05 (two-tailed) were considered as statisticallysignicant.

3. Results

3.1. Elevated expression of MDR1 mRNA correlates with lymph node

metastasis of invasive ductal carcinoma

We previously reported that knockdown of MDR1 expression inMDR breast cancer cells inhibited cell migration and invasionability [42]. In the present study, in order to determine whether P-gp/MDR1 expression is associated with breast cancer progression,we analyzed and compared the expression levels of MDR1 mRNAin 30 cases of invasive breast cancer tissues and matching normaltissues by quantitative RT-PCR. We observed a relative increasedexpression of MDR1 mRNA in invasive breast cancer tissues incomparison with matching normal tissues (Fig. 1A, P = 0.0034).Next, we analyzed the expression levels of MDR1 mRNA in 50 casesof normal breast tissues and 120 cases of primary breast cancertissues including ductal carcinoma in situ and invasive ductal

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302 295carcinoma. Fig. 1B shows that the relative expression levels ofMDR1 mRNA remained at a very low level in normal breast tissues(1.000 1.5618, n = 50). The expression of MDR1 mRNA waselevated in ductal carcinoma in situ (2.555 1.5047, n = 20,P = 0.0011) and invasive ductal carcinoma (3.180 9.7625, n = 100,P < 0.0001). Furthermore, high level of MDR1 mRNA expression isassociated with lymph node metastasis in cancer patients (Fig. 1C,Table 2), suggesting a pathological relevance of MDR1 expression inbreast cancer progression (P < 0.0001).

Fig. 1. Quantitative real time PCR analysis of expression of MDR1 and Anxa2 mRNA in breadjacent normal tissues (Normal). (A) Expression of MDR1 mRNA was signicantly eleva

tissues. (B) Expression of MDR1 mRNA was signicantly elevated in DCIS (P = 0.0011) and

of MDR1 mRNA in IDC tissues associated with lymph node metastasis (LNM) of breas

carcinoma was higher than that of matching normal tissues (P = 0.0155). (E) Expression

normal breast (P = 0.0001). (F) High level of Anxa2 mRNA in IDC tissues associated with

expression of MDR1 mRNA positively correlated with Anxa2 mRNA expression in IDC t

different groups were performed using Wilcoxon test (for two groups) or KruskalWallis

were considered as statistically signicant.3.2. Elevated expression of Anxa2 mRNA correlates with lymph node

metastasis of invasive ductal carcinoma

Recently, Anxa2 upregulation was shown to promote theinvasion ability of MCF-7 breast cancer cell [43]. However, thedetailed role of Anxa2 in breast cancer progression remainsunclear. Hence, we investigate the expression pattern of Anxa2mRNA in breast cancer tissues of various stages. As shown inFig. 1D, expression of Anxa2 mRNA in invasive ductal carcinoma

ast tissues from invasive ductal carcinoma (IDC), ductal carcinoma in situ (DCIS) and

ted in IDC (P < 0.0034) breast cancer tissues in comparison with matching normal

IDC (P < 0.0001) breast cancer tissues compared with normal tissues. (C) High level

t cancer patients (P < 0.0001). (D) Expression of Anxa2 mRNA in invasive ductal

of Anxa2 mRNA was elevated in IDC tissues compared with tissues from DCIS and

lymph node metastasis (LNM) of breast cancer patients (P < 0.0001). (G) Increased

issues (r = 0.7463, P < 0.0001). The comparisons of mRNA expression levels among

test followed by Dunns test (for three groups). P values less than 0.05 (two-tailed)

an

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302296was higher than that of matching normal tissues (P = 0.0155).Although no signicant difference in Anxa2 mRNA expression wasdetected between ductal carcinoma in situ and normal breasttissues, the expression of Anxa2 mRNA in invasive ductalcarcinoma was signicantly higher than that of ductal carcinomain situ (Fig. 1E, P = 0.0001) and normal breast tissues (Fig. 1E,P < 0.0001). Further analysis showed that elevated expression ofAnxa2 correlated with lymph node metastasis of breast cancerpatients (Fig. 1F, Table 2). Interestingly, a signicant positivecorrelation was found between overexpression of MDR1 andAnxa2 (Fig. 1G), indicative of a possible functional linkage of thesetwo genes during breast cancer progression (r = 0.7463,P < 0.0001).

3.3. Adriamycin induces increased migration of breast cancer cells

Table 2Correlation of P-gp and Anxa2 mRNA expression with clinicopathologic parameters

Parameters/markers Number P-gp

Mean SD Menopausal

Pre-menopausal 69 3.960 11.5428 Post-menopausal 31 1.445 2.8283

Lymph node status

Negative 52 0.575 0.3351 Positive 48 6.117 13.7242

Clinical stage

I 31 1.833 5.5285 IIa, IIb 57 3.196 11.6251 IIIIV 12 6.578 8.2879

ER status

Negative 47 3.952 13.3985 Positive 53 2.495 4.6502

PR status

Negative 44 4.488 13.8499 Positive 56 2.152 4.3748

HER2 status

Negative 64 3.519 11.3702 Positive 36 2.577 6.0242 Development of MDR was reported to be related to increasedinvasiveness of cancer cells [4]. To study the possible impacts ofadriamycin on breast cancer cell migration, MCF-7 cells wereexposed to adriamycin for six consecutive weeks, and theirmigration ability was examined. Fig. 2A shows that treatment ofMCF-7 cells with adriamycin led to a signicant increase of cellmigration, as measured by the wound healing assay (P < 0.0001).Likewise, transwell-based assay also showed enhanced cellinvasion ability (Fig. 2B, P < 0.0001). It has been documented thattreatment with chemotherapeutic drugs could increase theexpression of MDR-associated proteins such as P-gp [44]. Inagreement with previous reports, our results showed thatadriamycin treatment induced upregulation of P-gp in MCF-7cells at both protein and mRNA levels (Fig. 2C and D, P < 0.0001).Interestingly, adriamycin also induced a signicant increase inAnxa2 expression (Fig. 2C and D, P < 0.0001). Anxa2 is proved to becritical for migration of many types of cancer cells [30]. Theseresults indicate that elevated expression of Anxa2 may contributeto the enhanced cell migration induced by adriamycin.

3.4. Inhibition of p-gp activity impaired invasion of MDR breast

cancer cells

To further test if inhibition of P-gp activity can inuence theinvasion of MCF-7/ADR cells, we treated the cells with VRP and TFP,two well-known inhibitors of P-gp [45,46], and examined theireffects on cell migration ability. As measured by the efux of Rh-123 from MCF-7/ADR cells, both TFP and VRP markedly reducedthe efux rate of Rh-123 in MCF-7/ADR cells in a dose-dependentmanner (Fig. 3B), indicating the inhibition of P-gp activity. Thetranswell assay showed that noncytotoxic concentrations of bothTFP and VRP can signicantly retard the invasion of MCF-7/ADRcells in a dose-dependent manner (Fig. 3C and D, P < 0.0001).Moreover, VRP and TFP treatment did not alter the total level ofAnxa2 or P-gp (Fig. 3A), supporting an idea that decrease in cellmotility is not due to the reduction of protein content. Together,these data suggest that P-gp activity may contribute to MDR cancercell invasion.

3.5. P-gp inhibition has no signicant effect on the interaction

between P-gp and Anxa2

d other biomarkers.

Anxa2

P value Mean SD P value

0.2353 16.012 37.9937 0.17066.267 14.4844

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302 297apparently induced tyrosine phosphorylation of Anxa2 in adose- and time-dependent manner. Considering that adriamy-cin is a substrate of P-gp, the potential role of P-gp in drug-induced phosphorylation of Anxa2 was postulated and theeffect of P-gp inhibition on adriamycin-induced Anxa2 phos-phorylation was examined. As shown in Fig. 5C, VRP or TFPtreatment signicantly suppressed the adriamycin-inducedtyrosine phosphorylation of Anxa2. Knockdown of MDR1 inMDR MCF-7 cells using small intereference RNA was found tosignicantly inhibit adriamycin-induced tyrosine phosphory-lation of Anxa2 in comparison with control cells (Fig. 5D). Inaddition, immunoprecipitation from MCF-7/ADR and MDR1knockdown cells using anti-Anxa2 antibody was performedand then immunoblotting was conducted in parallel with anti-Anxa2 and anti-phosphortyrosine antibodies (p-Tyr, 4G10,Millipore, Billerica, MA, USA). Fig. 5E shows the result of P-gpknockdown or inhibition by VRP-suppressed tyrosine phos-phorylation of Anxa2. P-gp was further conrmed as necessaryfor adriamycin-induced tyrosine phosphorylation of Anxa2.

Fig. 2. Chronic exposure of MCF-7 cells to adriamycin induced a signicant increase of cwith adriamycin for consecutive six weeks, the wound areas were photographed and

mean SD of triplicates, P < 0.0001. (B) Adriamycin exposure led to an increase of invasion aa concentration of 5 105 cells/mL. Then, 200 mL of cells were loaded into the upper chammedium. After an incubation for 24 h, the invaded cells were counted in ve randomly chose

in the right panel. Data as mean SD, P < 0.0001 versus MCF-7 cells. (C) Western blottindifferent times, the protein level of Anxa2 and P-gp in MCF-7 cells was signicantly elevate

treated with adriamycin for 1 week. (D) The expression of Anxa2 and P-gp mRNA in adriamy

time PCR analysis, P < 0.0001. All statistical analysis was performed by a one-way ANOVA3.7. P-gp mediates Anxa2 phosphorylation through interaction with

src

Anxa2 is a substrate for Src kinase, which interacts andphosphorylates Anxa2 at tyrosine 23 both in vivo and in vitro [5052]. Given that P-gp interacts and co-localizes with Anxa2 in MDRMCF-7 cells, we hypothesized that P-gp may mediate Anxa2tyrosine phosphorylation through its interaction with Src kinase.To test this possibility, we performed co-immunoprecipitationassay by utilizing anti-P-gp antibodies. Fig. 6 shows theimmunoprecipitation with anti-P-gp co-precipitated Src andAnxa2 in cell lysates from MDR MCF-7 cells. Consistently, Srcalso co-precipitated Anxa2 and P-gp, which further conrmed thephysical interaction between P-gp and Src.

4. Discussion

MDR is a major obstacle in the effective treatment of cancer. P-gp is known to be a key player in the mediation of MDR in cancer. In

ell migration and invasion ability. (A) Wound healing assay of MCF-7 cells exposed

relative migration distance was quantied and plotted in the right panel. Data as

bility of breast cancer cells. Cells were trypsinized, suspended in serum-free medium at

ber pre-coated with matrigel. The lower chamber was lled with 10% FBS-containing

n areas under a microscopy, and the average of triplicate experiments was summarized

g analysis of Anxa2 and P-gp expression in MCF-7 cells exposed with adriamycin for

d after exposure to adriamycin for 4 or 6 weeks; P < 0.0001 versus MCF-7 cells or cells

cin treated MCF-7 cells was signicantly increased as determined by quantitative real-

followed by Tukeys multiple comparison test.

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302298addition to being a drug pump, several studies have uncoveredmany other functions of P-gp during tumor progression, such ascell proliferation [710], cell cycle progression [7], apopto-sis[8,9,53,54], cell migration and invasion[19,21,42,55]. Over-expression of MDR1/P-gp was commonly reported to be correlatedwith poor prognosis of cancer patients [1118,56]. In vitro, MDRcancer cells often displayed enhanced invasion/metastatic poten-tials compared with their parental cells [19,22,23,25,31,5761].However, the detailed role of P-gp in promoting cancer cellmigration/invasion is not well understood. Herein, we demon-strated a novel mechanism employed by P-gp to promote MDRcancer cell migration and invasion. We presented evidences that P-gp modulates tyrosine phosphorylation of Anxa2 in MDR cancercells. We and other groups have shown that Anxa2 is critical formany types of cancer cell invasion and metastasis[31,40,43,47,49,6267]. These data indicated that P-gp-mediatedphosphorylation of Anxa2 plays an important role in MDR cancercell migration and invasion.

Anxa2 is well known to be a substrate of protein tyrosine kinaseSrc and a mediator of the plasminogen activator system. Intensivestudies indicated overexpression of Anxa2 in cancers of pancreas,colon, breast, liver, brain, kidney, and acute promyelocyticleukemia [16]. In the present study, increased expression of Anxa2mRNA was observed in invasive cancer tissues compared withtissues from normal breast and carcinoma in situ. High level ofAnxa2 was in a signicant correlation with lymph node metastasis

Fig. 3. Inhibition of P-gp activity impaired invasion of MDR breast cancer cells. (A) P-gp inin the presence or absence of TFP and VRP for 24 h. Then cells were lysed and total cel

reduced the efux rate of Rh-123 in MCF-7/ADR cells in a dose-dependent manner. (C a

dose-dependent manner. For cell invasion assay, the cell suspension at a density of 5

different concentration of P-gp inhibitors. After 24 h incubation, the number of invad

statistical analysis was performed by a one-way ANOVA, P < 0.0001 versus correspondingin breast cancer patients, suggesting that Anxa2 may be associatedwith breast cancer progression. In addition, elevated expression ofMDR1 mRNA at later stages of breast cancer was found to beassociated with lymph node metastasis in cancer patients.Interestingly, a strong positive correlation between P-gp andAnxa2 overexpression in breast cancer tissues was discovered.Coincidentally, elevated expression of Anxa2 was also observed inmultidrug resistant cancer cells [3133] as well as in drug treatedcancer cells [68,69], Moreover, high level of Anxa2 has beenreported to be related with poor response to chemotherapy incancer patients [3436]. These data raised a possible associationbetween P-gp and Anxa2 in breast cancer aggravation. Our in vitromodel analysis also found concurrent upregulation of P-gp andAnxa2 in drug-treated breast cancer cells. More importantly,chronic exposure to anticancer drugs was capable to enhancecancer cell migration and invasion, which further support the ideaof a functional association between P-gp and Anxa2 during cancerprogression.

The phenomenon that the elevated level of P-gp promotes cellmigration attracted our attention. Indeed, previous studies haveshown that treatment cells with P-gp substrate enhanced cellmigration ability [20,44,57], whereas exposure of MDR hepatomacells to the P-gp inhibitor PSC833 decreased their migrationcapability [26]. In the present study, we found that inhibition of P-gp using VRP or TFP signicantly decreased cell migration, which isin an agreement with our previous results that P-gp knockdown

hibitors treatment did not affect the total level of Anxa2 or P-gp. Cells were cultured

l lysate was further analyzed by Western blotting. (B) Both TFP and VRP markedly

nd D) Both TFP and VRP signicantly reduced the invasion of MCF-7/ADR cells in a

105 cells/mL was added into the transwell inserts in the absence or presence ofe cells were counted and plotted. Data was presented as mean SD of triplicates,

not-treated cells.

Fig. 4. Inhibition of P-gp activity has no signicant effect on the interaction between P-gp and Anxa2. (A) Confocal immunouorescence microscopy analysis showed that bothTFP and VRP treatment did not affect the colocalization between P-gp and Anxa2 at the cell membrane of MCF-7/ADR cells. (B) Co-immunoprecipitation assay showed that P-

gp still interacts with Anxa2 in the presence of TFP or VRP.

Fig. 5. Activity of P-gp is necessary for adriamycin-induced tyrosine phosphorylation of Anxa2. (A) Western blotting analysis of adriamycin-induced phosphorylation ofAnxa2 in total cell lysates from MCF-7/ADR cells. Cells were starved for 24 h and then exposed with different concentration of adriamycin for 30 min, FBS was used as positive

control. (B) Western blotting analysis of adriamycin-induced phosphorylation of Anxa2 in MCF-7/ADR cells treated with 0.5 mM adriamycin for different time points, 10% FBSwas used as positive control. (C) Western blotting analysis of adriamycin-induced phosphorylation of Anxa2 in MCF-7/ADR cells in the presence or absence of TFP or VRP. (D)

knockdown of P-gp impaired adriamycin induced phosphorylation of Anxa2 in MCF-7/ADR cells. (E) Co-immunoprecipitation assay conrmed that P-gp knockdown or

inhibition by VRP suppressed tyrosine phosphorylation of Anxa2.

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302 299

l

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302300inhibited MDR breast cancer cell migration. These results indicatethat P-gp may be directly involved in MDR cancer cell migrationand invasion. As our earlier work showed that P-gp interacts withAnxa2 in MDR MCF-7 cells [42], proteinprotein interaction mayplay a role in P-gp mediated cell migration. However, despite thefact that the membrane distribution of these two proteins waschanged after treatment with VRP or TFP, inhibitors of P-gp cannotdisturb the interaction between Anxa2 and P-gp, indicating that P-gp mediates cell migration through other unknown mechanisms.

Anxa2 phosphorylation is essential for cancer cell metastasis[3740,70]. Evidence that drug exposure induces increase in cellmotility led to the investigation of the potential relationshipbetween drug exposure and phosphorylation of Anxa2. Aninteresting nding is that adriamycin induced the upregulation ofAnxa2 phosphorylation in a dose- and time-dependent manner.Moreover, this effect can be inhibited by P-gp knockdown usingsmall interference RNA or suppression with P-gp inhibitors. Theseresults indicate that ATPase activity of P-gp regulates adriamycin-induced phosphorylation of Anxa2. Although the mechanism behindhow P-gp modulates drug-induced Anxa2 phosphorylation remainsunclear, the fact that P-gp transduces signaling is not surprising.Previous studies have shown that P-gp inhibited TNF or Fas inducedcaspase-3/8 activation in an ATP-dependent manner [5,6]. Similarly,the P-gp-mediated efux of platelet-activating factor contributed tothe activation of G protein-coupled receptor PAFR [4]. In addition,drug-induced activation of P-gp can transactivate PI3Kinase [71].These results indicate that a mechanism may exist by which P-gpinuences signal pathways. In most cases, Anxa2 is phosphorylated

Fig. 6. P-gp interacts with Src kinase and Anxa2. (A and B) MCF-7/ADR cells wereimmunoblotted with corresponding antibodies.at tyrosine 23 by protein tyrosine kinase Src [5052]. A possiblemechanism by which P-gp can inuence Anxa2 phosphorylation isthrough Src kinase. In this study, an interaction between P-gp andSrc kinase in MDR MCF-7 breast cancer cells using reciprocal co-immunoprecipitation assay was identied (Fig. 5). Although theexact role of P-gp in Src induced Anxa2 phosphorylation was notdelineated, the interaction between P-gp and Src may probablyaugment a necessary interaction between Src and Anxa2, therebyfacilitating the phosphorylation of Anxa2 by Src. In addition, weshowed that knockdown of Anxa2 was able to suppress cell invasionability without affecting P-gp activity [31], indicating that Anxa2might function downstream of P-gp and Src. Taken together, ourndings suggest a novel regulatory mechanism of P-gp-induced cellmigration.

In summary, we demonstrated concomitant upregulation of P-gp and Anxa2 in invasive breast cancer tissues and in breast cancercells that were continuously exposed to adriamycin. We alsoshowed a functional association between P-gp and Anxa2 duringbreast cancer progression. P-gp may promote the migration andinvasion of MDR breast cancer cells through interaction andmodulating the tyrosine phosphorylation of Anxa2. This processmight also involve the interaction between Src and P-gp. Ourndings provide novel insights into the mechanism underlying thepotential links between MDR phenotype and tumor invasion andmetastasis.

Acknowledgments

This research was supported by grants from the NationalNatural Science Foundation of China (No. 81071731, 81001188,and 81372844), Tianjin Municipal Science and TechnologyCommission (No. 12JCQNJC07000 and 12JCZDJC24500), Chang-jiang Scholars and Innovative Research Team (IRT1076), 863project (2012AA020206-5), and Tianjin Higher Education Scienceand Technology Development Project (20100120).

References

[1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics.CA Cancer J Clin 2011;61:6990.

[2] Beck WT, Cirtain MC, Danks MK, Felsted RL, Safa AR, Wolverton JS, et al.Pharmacological, molecular, and cytogenetic analysis of atypical multidrug-resistant human leukemic cells. Cancer Res 1987;47:545560.

[3] Bosch I, Croop J. P-glycoprotein multidrug resistance and cancer. BiochimBiophys Acta 1996;1288:F3754.

[4] Fletcher JI, Haber M, Henderson MJ, Norris MD. ABC transporters in cancer:more than just drug efux pumps. Nat Rev Cancer 2010;10:14756.

[5] Smyth MJ, Krasovskis E, Sutton VR, Johnstone RW. The drug efux protein, P-glycoprotein, additionally protects drug-resistant tumor cells from multiple

ysed, and immunoprecipitated with antibodies against P-gp (A) or Src (B), thenforms of caspase-dependent apoptosis. Proc Natl Acad Sci U S A1998;95:70249.

[6] Ruei AA, Tainton KM, Darcy PK, Smyth MJ, Johnstone RW. P-glycoproteininhibits caspase-8 activation but not formation of the death inducing signalcomplex (disc) following Fas ligation. Cell Death Differ 2002;9:126672.

[7] Katoh SY, Ueno M, Takakura N. Involvement of MDR1 function in proliferationof tumour cells. J Biochem 2008;143:51724.

[8] Rocco A, Compare D, Liguori E, Cianone A, Pirozzi G, Tirino V, et al. MDR1-P-glycoprotein behaves as an oncofetal protein that promotes cell survival ingastric cancer cells. Lab Invest 2012;92:140718.

[9] Guenova ML, Balatzenko GN, Nikolova VR, Spassov BV, Konstantinov SM. Ananti-apoptotic pattern correlates with multidrug resistance in acute myeloidleukemia patients: a comparative study of active caspase-3, cleaved PARPs,Bcl-2, Survivin and MDR1 gene. Hematology 2010;15:13543.

[10] Van Brussel JP, Jan Van Steenbrugge G, Van Krimpen C, Bogdanowicz JF, VanDer Kwast TH, Schroder FH, et al. Expression of multidrug resistance relatedproteins and proliferative activity is increased in advanced clinical prostatecancer. J Urol 2001;165:1305.

[11] Schneider J, Gonzalez-Roces S, Pollan M, Lucas R, Tejerina A, Martin M, et al.Expression of LRP and MDR1 in locally advanced breast cancer predictsaxillary node invasion at the time of rescue mastectomy after inductionchemotherapy. Breast Cancer Res 2001;3:18391.

[12] Zochbauer-Muller S, Filipits M, Rudas M, Brunner R, Krajnik G, Suchomel R,et al. P-glycoprotein and MRP1 expression in axillary lymph node metastasesof breast cancer patients. Anticancer Res 2001;21:11924.

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302 301[13] Leonessa F, Clarke R. ATP binding cassette transporters and drug resistance inbreast cancer. Endocr Relat Cancer 2003;10:4373.

[14] Raguz S, Tamburo De Bella M, Tripuraneni G, Slade MJ, Higgins CF, CoombesRC, et al. Activation of the MDR1 upstream promoter in breast carcinoma as asurrogate for metastatic invasion. Clin Cancer Res 2004;10:277683.

[15] Mignogna C, Staibano S, Altieri V, De Rosa G, Pannone G, Santoro A, et al.Prognostic signicance of multidrug-resistance protein (MDR-1) in renal clearcell carcinomas: a ve year follow-up analysis. BMC Cancer 2006;6:293.

[16] Vander Borght S, Komuta M, Libbrecht L, Katoonizadeh A, Aerts R, Dymar-kowski S, et al. Expression of multidrug resistance-associated protein 1 inhepatocellular carcinoma is associated with a more aggressive tumour phe-notype and may reect a progenitor cell origin. Liver Int 2008;28:137080.

[17] Chen H, Hao J, Wang L, Li Y. Coexpression of invasive markers (uPA, CD44) andmultiple drug-resistance proteins (MDR1, MRP2) is correlated with epithelialovarian cancer progression. Br J Cancer 2009;101:43240.

[18] Hoffmann AC, Wild P, Leicht C, Bertz S, Danenberg KD, Danenberg PV, et al.MDR1 and ERCC1 expression predict outcome of patients with locally ad-vanced bladder cancer receiving adjuvant chemotherapy. Neoplasia2010;12:62836.

[19] Colone M, Calcabrini A, Toccacieli L, Bozzuto G, Stringaro A, Gentile M, et al.The multidrug transporter P-glycoprotein: a mediator of melanoma invasion?J Invest Dermatol 2008;128:95771.

[20] Li QQ, Wang WJ, Xu JD, Cao XX, Chen Q, Yang JM, et al. Up-regulation of CD147and matrix metalloproteinase-2, -9 induced by P-glycoprotein substrates inmultidrug resistant breast cancer cells. Cancer Sci 2007;98:176774.

[21] Miletti-Gonzalez KE, Chen S, Muthukumaran N, Saglimbeni GN, Wu X, Yang J,et al. The CD44 receptor interacts with P-glycoprotein to promote cell migra-tion and invasion in cancer. Cancer Res 2005;65:66607.

[22] Thompson EW, Paik S, Brunner N, Sommers CL, Zugmaier G, Clarke R, et al.Association of increased basement membrane invasiveness with absence ofestrogen receptor and expression of vimentin in human breast cancer celllines. J Cell Physiol 1992;150:53444.

[23] dit Faute MA, Laurent L, Ploton D, Poupon MF, Jardillier JC, Bobichon H.Distinctive alterations of invasiveness, drug resistance and cellcell organiza-tion in 3D-cultures of MCF-7, a human breast cancer cell line, and its multidrugresistant variant. Clin Exp Metastasis 2002;19:1618.

[24] Yang JM, Xu Z, Wu H, Zhu H, Wu X, Hait WN. Overexpression of extracellularmatrix metalloproteinase inducer in multidrug resistant cancer cells. MolCancer Res 2003;1:4207.

[25] Staroselsky AN, Mahlin T, Savion N, Klein O, Nordenberg J, Donin N, et al.Metastatic potential and multidrug resistance correlation in the B16 melano-ma system. J Exp Ther Oncol 1996;1:2519.

[26] Bjornland K, Lehne G, Johansen HT, Fodstad O, Rugstad HE, Aasen AO, et al.Human hepatoma cells rich in P-glycoprotein display enhanced in vitroinvasive properties compared to P-glycoprotein-poor hepatoma cells. OncolRes 1998;10:25562.

[27] Gerke V, Creutz CE, Moss SE. Annexins: linking Ca2+ signalling to membranedynamics. Nat Rev Mol Cell Biol 2005;6:44961.

[28] Bharadwaj A, Bydoun M, Holloway R, Waisman D. Annexin A2 heterotetramer:structure and function. Int J Mol Sci 2013;14:6259305.

[29] Zhang X, Liu S, Guo C, Zong J, Sun MZ. The association of annexin A2 andcancers. Clin Transl Oncol 2012;14:63440.

[30] Lokman NA, Ween MP, Oehler MK, Ricciardelli C. The role of annexin A2 intumorigenesis and cancer progression. Cancer Microenviron 2011;4:199208.

[31] Zhang F, Zhang L, Zhang B, Wei X, Yang Y, Qi RZ, et al. Anxa2 plays a critical rolein enhanced invasiveness of the multidrug resistant human breast cancer cells.J Proteome Res 2009;8:50417.

[32] Peng X, Gong F, Xie G, Zhao Y, Tang M, Yu L, et al. A proteomic investigation intoadriamycin chemo-resistance of human leukemia K562 cells. Mol Cell Bio-chem 2011;351:23341.

[33] Cole SP, Pinkoski MJ, Bhardwaj G, Deeley RG. Elevated expression of annexin II(lipocortin II, p36) in a multidrug resistant small cell lung cancer cell line. Br JCancer 1992;65:498502.

[34] Chuthapisith S, Bean BE, Cowley G, Eremin JM, Samphao S, Layeld R, et al.Annexins in human breast cancer: possible predictors of pathological responseto neoadjuvant chemotherapy. Eur J Cancer 2009;45:127481.

[35] Takano S, Togawa A, Yoshitomi H, Shida T, Kimura F, Shimizu H, et al. AnnexinII overexpression predicts rapid recurrence after surgery in pancreatic cancerpatients undergoing gemcitabine-adjuvant chemotherapy. Ann Surg Oncol2008;15:315768.

[36] Kagawa S, Takano S, Yoshitomi H, Kimura F, Satoh M, Shimizu H, et al. Akt/mTOR signaling pathway is crucial for gemcitabine resistance induced byAnnexin II in pancreatic cancer cells. J Surg Res 2012;178:75867.

[37] Rescher U, Ludwig C, Konietzko V, Kharitonenkov A, Gerke V. Tyrosine phos-phorylation of annexin A2 regulates Rho-mediated actin rearrangement andcell adhesion. J Cell Sci 2008;121:217785.

[38] de Graauw M, Tijdens I, Smeets MB, Hensbergen PJ, Deelder AM, van de WaterB. Annexin A2 phosphorylation mediates cell scattering and branching mor-phogenesis via colin Activation. Mol Cell Biol 2008;28:102940.

[39] Zheng L, Foley K, Huang L, Leubner A, Mo G, Olino K, et al. Tyrosine 23phosphorylation-dependent cell-surface localization of annexin A2 is requiredfor invasion and metastases of pancreatic cancer. PLoS ONE 2011;6:e19390.

[40] Zhao P, Zhang W, Tang J, Ma X, Dai J, Li Y, et al. Annexin II promotes invasionand migration of human hepatocellular carcinoma cells in vitro via its inter-action with HAb18G/CD147. Cancer Sci 2010;101:38795.[41] Das S, Shetty P, Valapala M, Dasgupta S, Gryczynski Z, Vishwanatha JK.Signal transducer and activator of transcription 6 (STAT6) is a novelinteractor of annexin A2 in prostate cancer cells. Biochemistry2010;49:221626.

[42] Zhang HC, Zhang F, Wu B, Han JH, Ji W, Zhou Y, et al. Identication of theinteraction between P-glycoprotein and Anxa2 in multidrug-resistant humanbreast cancer cells. Cancer Biology & Medicine 2012;9:99104.

[43] Wu B, Zhang F, Yu M, Zhao P, Ji W, Zhang H, et al. Up-regulation of Anxa2 genepromotes proliferation and invasion of breast cancer MCF-7 cells. Cell Prolif2012;45:18998.

[44] Liang Y, ODriscoll L, McDonnell S, Doolan P, Oglesby I, Duffy K, et al. Enhancedin vitro invasiveness and drug resistance with altered gene expression pat-terns in a human lung carcinoma cell line after pulse selection with anticancerdrugs. Int J Cancer 2004;111:48493.

[45] Schroder LE, Blumenstein BA, Flanigan RL, Borst JR, David Crawford E. Phase IIevaluation of doxorubicin/vinblastine combined with inhibitors triuopera-zine/verapamil of P-glycoprotein in patients with advanced renal carcinoma.Urologic Oncology 1997;3:948.

[46] Murren JR, Durivage HJ, Buzaid AC, Reiss M, Flynn SD, Carter D, et al. Triuo-perazine as a modulator of multidrug resistance in refractory breast cancer.Cancer Chemother Pharmacol 1996;38:6570.

[47] Sharma MR, Koltowski L, Ownbey RT, Tuszynski GP, Sharma MC. Angiogene-sis-associated protein annexin II in breast cancer: selective expression ininvasive breast cancer and contribution to tumor invasion and progression.Exp Mol Pathol 2006;81:14656.

[48] Inokuchi J, Narula N, Yee DS, Skarecky DW, Lau A, Ornstein DK, et al. AnnexinA2 positively contributes to the malignant phenotype and secretion of IL-6 inDU145 prostate cancer cells. Int J Cancer 2009;124:6874.

[49] Mohammad HS, Kurokohchi K, Yoneyama H, Tokuda M, Morishita A, Jian G,et al. Annexin A2 expression and phosphorylation are up-regulated in hepa-tocellular carcinoma. Int J Oncol 2008;33:115763.

[50] Gerke V, Weber K. Identity of p36K phosphorylated upon Rous sarcoma virustransformation with a protein puried from brush borders; calcium-depen-dent binding to non-erythroid spectrin and F-actin. EMBO J 1984;3:22733.

[51] Gould KL, Woodgett JR, Isacke CM, Hunter T. The protein-tyrosine kinasesubstrate p36 is also a substrate for protein kinase C in vitro and in vivo. MolCell Biol 1986;6:273844.

[52] Spijkers-Hagelstein JA, Mimoso Pinhancos S, Schneider P, Pieters R, Stam RW.Src kinase-induced phosphorylation of annexin A2 mediates glucocorticoidresistance in MLL-rearranged infant acute lymphoblastic leukemia. Leukemia2013;27:106371.

[53] Mantovani I, Cappellini A, Tazzari PL, Papa V, Cocco L, Martelli AM. Caspase-dependent cleavage of 170-kDa P-glycoprotein during apoptosis of human T-lymphoblastoid CEM cells. J Cell Physiol 2006;207:83644.

[54] Tainton KM, Smyth MJ, Jackson JT, Tanner JE, Cerruti L, Jane SM, et al.Mutational analysis of P-glycoprotein: suppression of caspase activation inthe absence of ATP-dependent drug efux. Cell Death Differ 2004;11:102837.

[55] Barakat S, Turcotte S, Demeule M, Lachambre MP, Regina A, Baggetto LG, et al.Regulation of brain endothelial cells migration and angiogenesis by P-glyco-protein/caveolin-1 interaction. Biochem Biophys Res Commun 2008;372:4406.

[56] Surowiak P, Materna V, Matkowski R, Szczuraszek K, Kornafel J, Wojnar A, et al.Relationship between the expression of cyclooxygenase 2 and MDR1/P-gly-coprotein in invasive breast cancers and their prognostic signicance. BreastCancer Res 2005;7:R86270.

[57] Liang Y, Meleady P, Cleary I, McDonnell S, Connolly L, Clynes M. Selection withmelphalan or paclitaxel (Taxol) yields variants with different patterns ofmultidrug resistance, integrin expression and in vitro invasiveness. Eur JCancer 2001;37:104152.

[58] Lopes EC, Ernst G, Aulicino P, Vanzulli S, Garcia M, Alvarez E, et al. Dissimilarinvasive and metastatic behavior of vincristine and doxorubicin-resistant celllines derived from a murine T cell lymphoid leukemia. Clin Exp Metastasis2002;19:28390.

[59] Kotchetkov R, Cinatl J, Blaheta R, Vogel JU, Karaskova J, Squire J, et al.Development of resistance to vincristine and doxorubicin in neuroblastomaalters malignant properties and induces additional karyotype changes: apreclinical model. Int J Cancer 2003;104:3643.

[60] Hiscox S, Jiang WG, Obermeier K, Taylor K, Morgan L, Burmi R, et al. Tamoxifenresistance in MCF7 cells promotes EMT-like behaviour and involves modula-tion of beta-catenin phosphorylation. Int J Cancer 2006;118:290301.

[61] Li QQ, Wang WJ, Xu JD, Cao XX, Chen Q, Yang JM, et al. Involvement of CD147 inregulation of multidrug resistance to P-gp substrate drugs and in vitro inva-sion in breast cancer cells. Cancer Sci 2007;98:10649.

[62] Tatenhorst L, Rescher U, Gerke V, Paulus W. Knockdown of annexin 2 decreasesmigration of human glioma cells in vitro. Neuropathol Appl Neurobiol2006;32:2717.

[63] Mussunoor S, Murray GI. The role of annexins in tumour development andprogression. J Pathol 2008;216:13140.

[64] Shiozawa Y, Havens AM, Jung Y, Ziegler AM, Pedersen EA, Wang J, et al.Annexin II/annexin II receptor axis regulates adhesion, migration, homing,and growth of prostate cancer. J Cell Biochem 2008;105:37080.

[65] Sharma M, Ownbey RT, Sharma MC. Breast cancer cell surface annexin IIinduces cell migration and neoangiogenesis via tPA dependent plasmin gen-eration. Exp Mol Pathol 2010;88:27886.

[66] Wang YQ, Zhang F, Tian R, Ji W, Zhou Y, Sun XM, et al. Tyrosine 23 phos-phorylation of annexin A2 promotes proliferation, invasion, and Stat3 phos-phorylation in the nucleus of human breast cancer SK-BR-3 cells. CancerBiology & Medicine 2012;9:24853.

[67] Rankin CR, Hilgarth RS, Leoni G, Kwon M, Den Beste KA, Parkos CA, et al.Annexin A2 regulates beta1 integrin internalization and intestinal epithelialcell migration. J Biol Chem 2013;288:1522939.

[68] Maxwell PJ, Longley DB, Latif T, Boyer J, Allen W, Lynch M, et al. Identicationof 5-uorouracil-inducible target genes using cDNA microarray proling.Cancer Res 2003;63:46026.

[69] Boyer J, Maxwell PJ, Longley DB, Johnston PG. 5-Fluorouracil: identication ofnovel downstream mediators of tumour response. Anticancer Res2004;24:41723.

[70] Zhao P, Zhang W, Wang SJ, Yu XL, Tang J, Huang W, et al. HAb18G/CD147promotes cell motility by regulating annexin II-activated RhoA and Rac1signaling pathways in hepatocellular carcinoma cells. Hepatology 2011;54:201224.

[71] Yang JM, Vassil A, Hait WN. Involvement of phosphatidylinositol-3-kinase inmembrane rufing induced by P-glycoprotein substrates in multidrug-resis-tant carcinoma cells. Biochem Pharmacol 2002;63:95966.

F. Zhang et al. / Biochemical Pharmacology 87 (2014) 292302302

P-glycoprotein associates with Anxa2 and promotes invasion in multidrug resistant breast cancer cellsIntroductionMaterials and methodsCell cultureRNA extraction, reverse transcription and real-time PCRDrug exposure and wound healing assayIn vitro cell invasion assayWestern blotting and co-immunoprecipitation assaysImmunofluorescence confocal microscopy analysisFlow cytometry assayStatistical analysis

ResultsElevated expression of MDR1 mRNA correlates with lymph node metastasis of invasive ductal carcinomaElevated expression of Anxa2 mRNA correlates with lymph node metastasis of invasive ductal carcinomaAdriamycin induces increased migration of breast cancer cellsInhibition of p-gp activity impaired invasion of MDR breast cancer cellsP-gp inhibition has no significant effect on the interaction between P-gp and Anxa2Activity of P-gp affects tyrosine phosphorylation of Anxa2P-gp mediates Anxa2 phosphorylation through interaction with src

DiscussionAcknowledgmentsReferences