Reversal of multidrug resistance phenotype in human breast cancer cells

Reversal of multidrug resistance phenotype in human breast cancer cells using doxorubicin-liposome –microbubble complexes assisted by ultrasoundZhiting Deng a ,Fei Yan a ,⁎,Qiaofeng Jin a ,Fei Li a ,Junru Wu b ,Xin Liu a ,Hairong Zheng a ,⁎aPaul uterbur Research Center for Biomedical Imaging,Institute of Biomedical and Health Engineering,Shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences,Shenzhen 518055,China bDepartment of Physics,University of Vermont,Burlington,VT 05405,USAa b s t r a c ta r t i c l e i n f o Article history:Received 2July 2013Accepted 19November 2013Available online 25November 2013Keywords:Doxorubicin-liposome –microbubble complexesMultidrug resistance Ultrasound Doxorubicin Drug deliveryThe circumvention of multidrug resistance (MDR)plays a critically important role in the success of chemother-apy.The aim of this work is to investigate the effectiveness and possible mechanisms of the reversal of MDR phe-notype in human breast cancer cells by using doxorubicin-liposome –microbubble complexes (DLMC)assisted by ultrasound (US).DLMC is fabricated through conjugating doxorubicin (DOX)-liposome (DL)to the surface of microbubbles (MBs)via the biotin –avidin linkage.The resulting drug-loaded complexes are then characterized and incubated with MCF-7/ADR human breast cancer cells and followed by US exposure.Our results show the more rapid cellular uptake,evident enhancement of nuclear accumulation and less drug ef flux in the resistant cells treated by DLMC +US than those treated by DL,DL +verapamil under the same US treatment or DLMC without US.The enhanced drug delivery and cellular uptake also associated with the increase of cytotoxicity against MCF-7/ADR cells,lower MCF-7/ADR cell viability and higher apoptotic cells.Mechanism investigations further disclose a signi ficant increase of reactive oxygen species (ROS)level,enhanced DNA damage and obvious reduction of P-glycoprotein expression in the resistant cells treated with DLMC +US compared with the control cases of cells treated by DLMC,DL +US or DL +verapamil +US.In conclusion,our study demonstrates that DLMC in combination with US may provide an effective delivery of drug to sensitize cells to circumvent MDR and to enhance the therapeutic index of the chemotherapy.©2013Elsevier B.V.All rights reserved.1.IntroductionIt is known that multidrug resistance (MDR)is one of the major obsta-cles to the successful cancer chemotherapy.Many of the initially respon-sive tumors relapse to develop resistance to multiple anticancer agents [1].Some tumors such as metastatic breast cancer show MDR even in the first treatment and become insensitive to a new drug [2].In general,MDR can be mediated by a number of mechanisms.A leading means is through the usage of ATP-binding cassette (ABC)transporters to actively transport anticancer drugs across biological membranes,preventing drugs from reaching resistant cancer cells before attaining the targets [3].It has been reported that most drug transporter proteins are located in the plasma membrane of cells and contribute to the major form of MDR phenotype [4].There has been intense search for compounds which can act to reverse MDR phenotype not only in cultured cells,but also in animal models [5–7].Various drug delivery systems are also engineered toovercome drug extrusion by ef flux transporters [8,9].Liposome –microbubble complexes (LMC)assisted by ultrasound (US)developed by Kheirolomoom et al.have become promising drug delivery systems due to its non-invasive nature and the associated drug release proce-dure could be triggered and controlled by US [10].Using sonoporation (a phenomenon that US increases cell membrane permeability),cargo-loaded liposome –microbubble complexes under US exposure can transiently perforate the cell membrane and thus facilitate trans-membrane transport of drugs/gene into cells [11,12].There have been several subsequent reports of ef ficient drug delivery in vitro by LMC and US-controlled drug delivery.Successful delivery of the anticancer drug doxorubicin (DOX)by US exposure of LMC has been reported in a melanoma cell culture model [13].Our previous study also showed a signi ficant enhancement of antitumor ef ficacy using paclitaxel-liposome –microbubble complexes combined with US in 4T1-bearing mice [14].However,to the best of our knowledge,there have been few reports of application of LMC as drug vehicles for the treatment of drug resistant cancer cells.More importantly,the corresponding mechanisms to reverse MDR phenotype remain elusive.We have investigated experimentally the reversal of MDR phenotype through US-triggered drug delivery of DOX-liposome –microbubbleJournal of Controlled Release 174(2014)109–116⁎Corresponding authors.Tel.:+8675586392244;fax:+8675596382299.E-mail addresses:fei.yan@ (F.Yan),hr.zheng@ (H.Zheng).0168-3659/$–see front matter ©2013Elsevier B.V.All rights reserved./10.1016/j.jconrel.2013.11.018Contents lists available at ScienceDirectJournal of Controlled Releasej o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m/l o c a t e /j c o n r e lcomplexes(DLMC)and anti-tumor activities in the doxorubicin-resistant MCF-7cell line(MCF-7/ADR).The detailed study of the underlying mechanisms by which this drug delivery system can overcome MDR is also performed.2.Materials and methods2.1.Materials1,2distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(poly-ethylene glycol)-2000](DSPE-PEG2000),1,2-distearoyl-sn-glycero-3-phosphatidylcholine(DSPC),dipalmitoylphosphatidylcholine (DPPC)and1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[biotinyl(polyethylene Glycol)2000](DSPE-PEG2000-Biotin)were purchased from Avanti Polar Lipids Inc.(Alabaster,AL,USA). Perfluoropropane(C3F8)was purchased from Huahe new-technology development company(Tianjin,China).All other reagents were of analytical grade.Doxorubicin hydrochloride(DOX,N98%),verapamil hydrochloride(≧99.0%),monoclonal anti-β-actin-peroxidase antibody, and bovine serum albumin(BSA),avidin,and4′,6-diamidino-2-phenylindole(DAPI)were obtained from Sigma-Aldrich(St.Louis,MO, USA).RPMI1640and penicillin/streptomycin were purchased from HyClone Inc.(Logan,UT,USA).Fetal bovine serum(FBS)was obtained from GIBCO(Grand Island,NY,USA).Cell Counting Kit-8(CCK-8)was purchased from Dojindo Laboratories(Tokyo,Japan),Glycoprotein P Monoclonal Antibody(C219)and ECL Western Blotting Substrate were purchased from Thermo Pierce,and Histone H2A.x Phospho(pS139) rabbit monoclonal antibody was purchased from Epitomics(Burlingame, CA,USA).2.2.Preparation of biotinylated DOX-liposomes(DL)The biotinylated DOX-liposome formulation was composed of DPPC: cholesterol:DSPE-PEG-biotin in the molar ratio of60:40:5[13].The organic solvents were removed under a nitrogenflow until a thin lipid film was formed,which was further dried for over2h under vacuum. The lipidfilm was hydrated at60°C with a250mM(NH4)2SO4buffer (pH5.4)and the suspension was extruded through a polycarbonate membrane of200nm using a mini-extruder(Avanti Polar Lipids, Alabaster,AL),then the extra liposomal ammonium sulfate was re-placed by PBS(137mM NaCl,2.7mM KCl,10mM Na2HPO4,2mM KH2PO4,pH7.4)overnight in a dialysis bag(MWCO3500).Subsequent-ly,a doxorubicin solution in PBS(1mg/ml)was added to the empty li-posomes and incubated at65°C for4h.Finally,the liposomes were passed through a Sephadex column(Sephadex G-50,Sigma-Aldrich) equilibrated with PBS to remove traces of unencapsulated doxorubicin.The encapsulation efficiencies(EE)of DOX were calculated using the formula:EE%(W i/W total)×100%.W i is the measured amount of DOX in the liposome suspensions after passing over the Sephadex G-50 column,and W total is the measured amount of DOX in the liposome suspensions before passing over the Sephadex G-50column.The DOX concentration was determined byfluorescence intensity measurements (λex=485nm,λem=550nm).The average diameter of the bio-tinylated DOX-liposomes was determined by dynamic light scattering (Zetasizer Nano ZS,Malvern Instrument,UK).2.3.Preparation of MBs and DOX-liposome–microbubblecomplexes(DLMC)The DSPC:DSPE-PEG2000:DSPE-PEG2000-biotin with molar ratios (9:0.5:0.5)were blended in chloroform and the solvent was removed under nitrogenflow at room temperature.Residual chloroform was fur-ther eliminated by evaporation under a vacuum for at least2h.The dried phospholipid blends were hydrated with a given buffer consisting of0.1M Tris(pH7.4):glycerol:propylene glycol(80:10:10by volume). Then air in the vial was exchanged with perfluoropropane(C3F8).Microbubbles(MBs)were obtained by mechanically vibrating the admix-ture for45s.After MBs were washed with PBS solution three times to remove excess unincorporated lipids by centrifuge at400g,50μg of avidin per108MBs was then added to the washed MB dispersion.Follow-ing15min of incubation at room temperature,the MBs were washed three times to remove unreacted avidin,and incubated at room temper-ature with biotinylated DOX-liposomes for another15min.Free DOX-liposomes were removed through washing with PBS.Morphologic characteristics of DOX-liposome–microbubble com-plexes(DLMC)were determined under afluorescent microscope (Leica DMI3000B,Wetzlar,Germany).Particle size,size distribution and concentration of MBs were determined using an optical particle counter with a0.5μm diameter detection limit(Accusizer780;Particle Sizing Systems,Santa Barbara,CA,USA).2.4.Cell cultureThe doxorubicin resistant MCF-7cell line,MCF-7/ADR was pur-chased from Cancer Institute and Hospital,Chinese Academy of Medical Sciences(Beijing,China).The cells were cultured in RPMI1640contain-ing10%fetal bovine serum(FBS),100U/ml penicillin,100μg/ml strep-tomycin,and2mM L-glutamine.To maintain the drug-resistant phenotype,MCF-7/ADR cells were cultured in the presence of1μg/ml doxorubicin and passaged for1week in a drug-free medium before the experiment.The cells were grown in a humidified5%CO2incubator at37°C.2.5.Determination of DOX uptake and retention in DOX-resistant cellsFor cellular uptake study,5×105MCF-7/ADR cells were seeded in 24-well plates and incubated overnight to allow cell adhesion.Unless otherwise stated,cells were immediately subjected to US exposure after addition of DL,DL+verapamil(10μM)or DLMC,and only DLMC without US treatment,wherein the concentrations of DOX were kept at10μg/ml.The ultrasound radiation was performed for15s by moving a20mm US probe E1609(Valpey Fisher Inc,Hopkinton,MA, USA)over the cell culture plate at the following settings:1MHz,20% duty cycle,and an US intensity of1.65W/cm2,with an US peak intensity of0.35MPa.After that,these treated cells were incubated for further15 or30min.Then cells were rinsed to remove uninternalized DOX.To quantitatively determine the cellular DOX,the treated cells were col-lected and lysed with RAPI cell lysis buffer(Beyotime,China),and the DOX concentration in the cell lysates was detected in a microplate read-er(Synergy™4,BioTek,VT,USA)at excitation and emission wavelength of485/550nm;the results were normalized to total cellular protein content of the cells,which was determined using a BCA protein assay kit(Beyotime,China).For drug efflux studies,cells were treated with DLMC,DL+US, DL+verapamil+US or DLMC+US as described above.After incuba-tion for4h,cells were rinsed and incubated with fresh cell culture medium for30,90,120min,respectively.At each time point,the intra-cellular DOX concentration was measured by a microplate reader according to the same method.The percentage of DOX efflux was calcu-lated using thefluorescence values from cells lysed at t=0as the100% DOX content in the cells.2.6.Confocal microscopy to determine cellular distribution and retention of DOXA laser confocal scanning microscope was used to evaluate the cellu-lar uptake and intracellular distribution of DOX in MCF-7/ADR cells. Briefly,MCF-7/ADR cells were seeded on sterile round-shaped12mm coverslips in24-well plates to70%confluence,then treated with DLMC,DL+US,DL+verapamil+US or DLMC+US containing 10μg/ml of DOX.Control experiments were performed by adding blank medium.The cells were incubated for another4h,and then110Z.Deng et al./Journal of Controlled Release174(2014)109–116were rinsed with PBS for three times.Subsequently,the cells werefixed with4%paraformaldehyde solution for30min at room temperature, and each sample was mounted on a glass slide using mounting medium containing DAPI.The cells were then examined using a confocal laser scanning microscope(Leica TCS SP5,Wetzlar,Germany)to localize the DOX relative to the cells.The same procedures were used to observe the intracellular reten-tion of DOX,except that after4h incubation,cells were washed by PBS to remove uninternalized DOX,and incubated with fresh cell culture medium for another24h.2.7.Cytotoxicity assayMCF-7/ADR cells were seeded in24-well plates for overnight to allow cell adhesion.Cells were incubated with unloaded MBs,DL, DL+verapamil or DLMC,containing5or10μg/ml offinal DOX concentrations,respectively.Then ultrasound radiation was applied for these cells as stated above.DLMC containing5or10μg/ml offinal DOX concentrations(without US treatment)were also investigated as a control.After24h treatment,the cells were rinsed twice and replaced with fresh culture medium.Cell viability was determined by the Cell Counting Kit-8kit(Dojindo,Japan)by measuring the absorbance at 450nm using a multimode plate reader(Synergy™4,BioTek,VT,USA).2.8.Reactive oxygen species(ROS)detectionThe intracellular ROS generation of cells was detected using2′,7′-dichlorofluorescein-diacetate(DCFH-DA)according to the previous re-port[15].Briefly,MCF-7/ADR cells were treated with DLMC,DL+US, DL+verapamil+US or DLMC+US.2h later,dichlorofluorescein diacetate(DCFH-DA,Beyotime)at afinal concentration of10μM was added and incubated for1h at37°C.After that,the cells were lysed and the lysates were centrifuged at10,000×g at4°C for5min.The su-pernatant was transferred to96-well black plates and measured using a multimode plate reader at the excitation wavelength of488nm and emission wavelength of525nm(Synergy™4,BioTek,VT,USA).Relative fluorescence units(RFUs)of samples were calculated and normalized to the untreated cells.2.9.Immunofluorescence staining forγ-H2AXMCF-7/ADR cells were treated with DL,DL+verapamil or DLMC in combination with ultrasound,or only DLMC(without US treatment). Thefinal DOX doses were kept at10μg/ml concentrations for each group.At8h after treatment,the cells werefixed in4%formaldehyde for10min,permeabilized with1%Triton X-100for30min and then blocked with2%BSA for30min.Then,the cells were incubated with Histone H2A.x Phospho(pS139)antibody(Epitomics,USA)(1:50dilu-tion)for overnight at4°C,followed by incubating with Alexa Fluor 488-labeled goat anti-rabbit IgG(Beyotime,China)(1:200dilution) for2h at room temperature.After washed with PBS,the cells were stained with4′,6-diamidino-2-phenylindole(DAPI)for15min.Fluores-cent images were acquired using a confocal laser scanning microscope (Leica TCS SP5,Wetzlar,Germany)equipped with63×oil-immersion lenses.The relative fraction ofγH2AX+was then determined using the NIH imaging software Image J.2.10.Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling(TUNEL)assayThe in-situ fragmented genomic DNA was detected by using terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling(TUNEL) kit(Roche)according to the manufacturer's protocol.In brief,the treated MCF-7/ADR cells with DLMC,DL+US,DL+verapamil+US,or DLMC+US at the samefinal DOX concentration(10μg/ml)were fur-ther cultivated for24h at37°C.After the cells werefixed,permeabilized and incubated with100μl TUNEL reaction mixture,DAPI was used to stain nuclei for15min.Thereafter,these cells were visualized by a Leica confocal laser microscope.The apoptotic cells were quantified by determining the percentage of TUNEL-positive(greenfluorescence) cells from six randomly chosenfields at×400magnification.2.11.Western blotting analysisTotal proteins of the treated cells were extracted using an RIPA lysis buffer(Beyotime,China)containing Protease inhibitor Cocktail Set I (Merck,Darmstadt,Germany).Proteins were separated in8%precast SDS-PAGE gels and transferred onto the nitrocellulose membrane.The membrane was further incubated with primary P-glycoprotein mAb (1:50dilution;C219,Pierce)and the secondary HRP conjugated anti-mouse IgG(1:1000dilution;Santa Cruz,CA).As a control,the mouse anti-β-actin-HRP antibody(1:20,000dilution;Sigma,St.Louis,MO) was used.The immunoblots were visualized by an enhanced chemilu-minescence(ECL)detection system(Pierce,IL,USA)and images were acquired by a Kodak Image Station4000MM PRO Digital Imaging Sys-tem(Eastman Kodak Company,Rochester,NY,USA).Protein abundance was quantified by Image J software.2.12.Statistical analysesThe mean±SD was determined for each treatment group.Statisti-cal analysis was performed using a Student's t-test.The differences were considered significant for*p b0.05,and very significant for**p b0.01.3.Results3.1.Characterization of DL and DLMCFig.1a presented typical images of DLMC observed under afluo-rescent microscope and its corresponding morphological character-istics under a light microscope.Due to the presence of DL,the surface of the DLMC appeared red under afluorescent microscope, indicating the successful conjugation of the biotinylated DL onto the surface of microbubbles.The average diameter of DL was 101.8nm and the polydensity index(PDI)was0.12(Fig.1b).About 73.0%of DL,with94.6±1.6%of DOX encapsulation efficiency,was attached to the biotinylated MBs,resulting in a slightly lager mean size of DLMC(2.09±0.21μm),compared with the unloaded MBs (1.64±0.14μm).Typical size distributions of plain MBs and DLMC were shown in Fig.1c.3.2.Rapid intracellular uptake and nuclear accumulation of DOXAs shown in Fig.2a,most DOX was distributed in the cytosol in MCF-7/ADR cells when treated by DLMC and DL+US.The treat-ment of DL along with10μM verapamil(thefirst compound found to be able to reverse MDR,and showed inhibition of adriamycin efflux [5,16]),improved the effectiveness of DOX nuclear accumulation under the same US exposure compared with DL+US treatment.By contrast, a significant larger quantity of DOX nuclear accumulation was observed in MCF-7/ADR cells after exposure to DLMC+US.DOX was mainly distributed in the nucleoplasm and the perinuclear region.Fig.2b showed that DLMC+US promoted rapid intracellular uptake of DOX into MCF-7/ADR cells,compared with the cases of DLMC, DL+US,and DL+verapamil+US.After15min incubation the intracellular drug concentration in the resistant cells treated with DLMC+US was1.46-fold,1.45-fold,and1.38-fold higher than that with DLMC(p b0.05),DL+US(p b0.05)and DL+verapamil+US (p b0.05).Interestingly,despite the presence of verapamil,the intracel-lular drug concentration was not significantly improved in a short time (15min).A longer incubation time(30min)of cells after DLMC+US treatment showed a similar trend,indicating1.29-fold,1.27-fold,and111Z.Deng et al./Journal of Controlled Release174(2014)109–1161.18-fold higher than that with DLMC (p b 0.01),DL +US (p b 0.01)and DL +verapamil +US cases.(p b 0.05).3.3.Cellular retention of DOX in MCF-7/ADR cellsNext,we examined whether DLMC +US could help drug nuclear retention in MCF-7/ADR cells using the confocal laser scanning mi-croscopy.Signi ficantly apparent nuclei localization was found in MCF-7/ADR cells treated with DLMC +US,compared with DLMC,DL +US and DL +verapamil +US (Fig.3a).DOX ef flux rates in MCF-7/ADR cells were shown in Fig.3b,indicating that the intracellular DOX level decreased rapidly in MCF-7/ADR cells treated with DL +US.Only 46.7±2.3%or 47.6±3.1%DOX was retained after 2h in the cells treated with DL +US or with DLMC (without US),respectively.About 61.4±2.0%intracellular retention of DOX was detected in the cells treated with DL +verapamil +US.In contrast,the ef flux ratio of DOX was signi ficantly lower when the cells were treated with DLMC +US,achieving 73.8±1.8%DOX retained in the resis-tant cells (p b 0.01vs DL +US).The result suggested there was a signi ficant slower ef flux rate of drugs when cells were treated with DLMC +US.3.4.Enhanced cell cytotoxicity to MCF-7/ADR cellsWe also explored whether DLMC +US would increase the drug cyto-toxicity against DOX-resistant cells.As shown in Fig.4,DLMC +US signi ficantly enhanced the cytotoxic effects of DOX in these resistant cancer cells.The viability of cells treated with DLMC +US containing 5μg/ml of DOX was signi ficantly lower,achieving 58.3±1.8%,compared to DL +US (85.5±1.4%,p b 0.01),DL +verapamil +US (78.6±2.4%,p b 0.05),and DLMC (85.1±5.0%,p b 0.01).Using higher DOX dose (10μg/ml)would result in more signi ficant cell cytotoxicity.The viability of cells treated with DLMC +US achieved 52.1±2.5%,with a signi ficantly lower cell viability than that of cells treated with DL +US (80.7±2.0%,p b 0.01),DL +verapamil +US (69.5±2.9%,p b 0.05),and DLMC (80.1±4.9%,p b 0.01).The cells treated with unloaded MBs +US (not loaded with DOX-liposomes)remained about 84.6±4.0%of viability after insonation.3.5.Increased ROS levels in MCF-7/ADR cellsThe ROS activities in MCF-7/ADR cells were indicated by DCF fluores-cence intensity.As demonstrated in Fig.5,cells treated with DLMC +US caused the most signi ficant change of ROS level with 70.1%increase in ROS production.By contrast,there were 47.3%,29.8%,19.4%,and 30.0%increase in the MCF-7/ADR cells treated with DL +verapamil +US,DL +US,unloaded MBs +US or DLMC without US,respectively (Fig.5).No obvious ROS generation was found when US was applied alone (without MBs).3.6.DNA damage in MCF-7/ADR cellsγ-H2AX formation is an early chromatin modi fication during apo-ptosis when DNA double-stranded breaks (DSBs)are introduced,which can be detected by nuclear foci through immuno fluorescence staining [17].As shown in Fig.6,a pan-nuclear γH2AX +distribution was exhibited when incubating the cells for 8h after treatment with DLMC +US (Fig.6a).The relative fraction of γH2AX +was more signif-icant (68.5±4.8%)in MCF-7/ADR cells treated with DLMC +US than that of DL +US (34.8±2.7%,p b 0.01),DL +verapamil +US (46.6±3.4%,p b 0.01),and DLMC (33.6±3.0%,p b 0.01)(Fig.6b and c).3.7.Apoptosis of MCF-7/ADR induced by DLMC and USTo figure out the mechanisms of enhanced cytotoxicity,we detected an in situ cell-death using terminal-deoxynucleotidyl transferase mediat-ed nick end labeling (TUNEL).As shown in Fig.7a,the cells incubated with DLMC and DL +US showed weak fluorescence,suggesting that there were a few apoptotic cells.The treatment of DL +verapamil +US showed an increased apoptotic population in MCF-7/ADR cells.In contrast,cells treated with DLMC +US of the same DOX concentration exhibited much stronger fluorescence in most of the cells,indicating the considerably enhanced cell apoptosis.Quantitative analysis showed that there was a much higher apoptotic cell ratio in the cells treated with DLMC +US (83.3±4.8%),compared to cases with DL +US (47.6±3.5%,p b 0.01),DL +verapamil +US (57.5±5.6%,p b 0.01)and DLMC (48.5±4.6%,p b 0.01),respectively (Fig.7b).ThecontrolFig.1.Characterization of DOX liposomes (DL)and DOX-liposome –microbubble complexes (DLMC).(a)Fluorescent micrograph of DLMC using the fluorescence of the bound doxorubicin and corresponding transmission image (scale bar =20μm).(b)Size distribution of DL.(c)Size distribution of DLMC and non-loaded microbubbles (MBs).DOX,doxorubicin.112Z.Deng et al./Journal of Controlled Release 174(2014)109–116cells (unloaded MBs +US)had no visible green fluorescence,which sug-gested the absence of apoptotic cells (data not shown).3.8.Down regulation of P-glycoprotein protein levelIn order to examine whether P-glycoprotein participates in the sen-sitization of MCF-7/ADR cells,the expression level of P-glycoprotein was detected by western blotting and quanti fied by Image J software.As the results shown in Fig.8,cells treated with DLMC,DL +US showed no obvious decrease in P-glycoprotein level,compared with the control of MCF-7/ADR cells (without any treatment).US with unloaded MBs also did not affect its expression level either.As expectation,verapamil sensitizes drug resistant cancer cells to chemotherapy by inhibiting P-glycoprotein [18].Treatment with DL +verapamil +US decreased about 59.5%of P-glycoprotein level in these cells.While 44.4%decrease of P-glycoprotein expression was found in MCF-7/ADR cells treated with DLMC +US.4.Discussion and conclusionIn this study,we found that DLMC +US,as a novel drug delivery system,could effectively reverse multidrug resistance phenotype in MCF-7/ADR cells.As schematically shown in Fig.9,DLMC +US expo-sure could rapidly release the free drug and locally disrupt the cell membrane through sonoporation.Sonoporation can generate transient pores on the cell membrane allowing entry of extracellular drugs (such as free DOX and DL)into the cytoplasm [19];it is more effective in comparison with the treat-ment using DL,whose intracellular delivery is mainly dependent on endocytotic pathway.In an endocytotic pathway,it is reported that some kinds of liposome can be delivered by the endosome into the lyso-some.Then,the liposome can provoke endosome destabilization,which results in drug liberation into the cell cytoplasm [20].A.L.Seynhaeve also demonstrated the slow intracellular release of doxorubicin and its sequestering in lysosomes from DOX liposomes,resulting limited delivery to the nucleus [21].Free DOX in cytosol may gain access to the nuclei due to its small molecular weight (579.98),though some of them may be pumped out of the cells via P-glycoprotein.By contrast,a DL can't gain access to the nuclei because its size is much larger than the nuclear pore (about 10nm).Indeed,it has been elucidated that the success of intracellular delivery and subsequent subcellular localization of drug by sonoporation are dependent on the targeted molecular size [22].In our study,we discover that US promotes rapid intracellular uptake of DOX resulting much more nuclear accumulation of DOX when the cells are treated with DLMC +US (Fig.2);it is plausible that sonoporation caused by inertial acoustic cavitation improves the cell membrane per-meability by inducing transient pores on the cell membrane,which makes it possible that the released free DOX or DL may enter into the resistant cells.The free DOX after entering the cells may accumulate in the nucleus because of its high af finity for DNA [23].In thetreatmentFig.2.Intracellular DOX uptake and nuclear accumulation in MCF-7/ADR cells.(a)Confocal microscopy images of intracellular DOX distribution in MCF-7/ADR cells at 4h after DLMC,DL +US,DL +verapamil +US,and DLMC +US treatment.The nucleus is stained with DAPI (blue),and DOX is indicated by red fluorescence (scale bar =10μm).(b)DOX intracellular concentration in MCF-7/ADR post DLMC,DL +US,DL +verapamil +US,DLMC +US treatment.*p b 0.05,**p b 0.01.ver:verapamil.113Z.Deng et al./Journal of Controlled Release 174(2014)109–116using DL +US,sonoporation might not occur due to the absence of MBs.Thus,intracellular delivery of DL needs to undergo the direct or transfer-protein-mediated exchange of lipid components with the cell membrane or be subjected to a speci fic or nonspeci fic endocytosis,by which a signif-icant longer time is needed [21].Furthermore,the small size of DOX makes it easier to cross the nuclear pore,which is not accessible for the DL case.Since DOX is known to interact with DNA by intercalation in the nucleus,thus intranuclear drug concentration is crucial for it to be ef-fective to inhibit the growth of MCF-7/ADR cells.Indeed,such cellular characteristic of DOX was also observed by others [24].Notably,our data also showed an obvious cellular retention of DOX in MCF-7/ADR cells treated with DLMC +US (Fig.3),attributing,to some degree,to more nuclear accumulation of DOX in MCF-7/ADR cells.It is worthwhile to point out that DLMC +US treatment enhances signi ficant cytotoxicity against MCF-7/ADR cells,indicating that DLMC +US can increase the potency of DOX to reverse MDR pheno-type in these cells.The following reasons may account for the advantage of DLMC over DL nanoparticles.First,the higher cellular uptake and faster intracellular delivery of the drugs are shown in Fig.2.Second,reduction of P-glycoprotein expression may cause the difference,as is showed in Fig.8.Moreover,our studies have also demonstrated much more apoptotic MCF-7/ADR cells treated by DLMC +US,which further con firms that the reduced cell viability may result from the apoptotic induction of these resistant cells (Fig.7).In this study,a signi ficant increase of intracellular radical production after exposure to ultrasound is demonstrated (Fig.5).In fact,the gener-ation of reactive oxygen species in cells due to acoustic cavitation is con-sidered to be a possible reason for cell killing [25,26].Besides the DNA intercalating function,DOX has been also proposed to induce oxidative stress in tumor cells,leading to apoptosis [27,28].As the amount of in-tracellular DOX increased (Fig.2),the enhancement of ROS generation in MCF-7/ADR cells is observed in this study (Fig.5).Our results indicate that DLMC may produce a similar mechanism as a vehicle for delivery of drugs,indicating that the generation of reactive oxygen species during the insonation mediates the generation of DNA single-strand breaks [29].γ-H2AX formation is an early chromatin modi fication following initiation of DNA fragmentation during apoptosis [7].It is worth highlighting that more pronounced γ-H2AX foci staining in resistant cancer cells are observed post-DLMC +US treatment for 8h (Fig.6a,b and c),indicating a more dramatic DNA damage in these resistant cells.One possible explanation for these observations is that the more drugs accumulated in the nuclei.DOX is involved in DNA damage through topoisomerase II inhibition and free radical generation [27,30].The improved generation of reactive oxygen species during the insonation may be another reason to mediate the generation of DNA single-strand breaks [29].Overexpression of the transmembrane drug ef flux pump P-glycoprotein is one of the major mechanisms by which cancer cells develop multidrug resistance against natural product antican-cer drugs including DOX.The published data suggest that P-gp,acting as a hydrophobic vacuum cleaner,removes drugs from the cell membrane and cytoplasm and transports them to the external medium [31].This may make it possible for many cancer cells to be resis-tant with a broad range of structurally and functionally distinct anticancer agents.Evidences have demonstrated that reduction of P-glycoprotein expression can prevent drugs from pumping out and reduce theirrateFig.3.Cellular retention of DOX in MCF-7/ADR cells.(a)Confocal microscopic image of DOX retention in MCF-7/ADR cells after DLMC,DL +US,DL +verapamil +US and DLMC +US treatment.The images were taken after cells were cultured in fresh medium for another 24h (scale bar =50μm).(b)Cellular retention of DOX in MCF-7/ADR cells at various times after DLMC,DL +US,DL +verapamil +US,and DLMC +US treatment,**p b0.01.Fig.4.Cell cytotoxicity of DOX in MCF-7/ADR cells after DLMC without US,DL +US,DL +verapamil +US,and DLMC +US treatment.The cell viability was measured after cells were cultured in fresh medium for another 24h by CCK-8assay.*p b 0.05,**p b0.01.Fig.5.Generation of the reactive oxygen species (ROS)in MCF-7/ADR cells was measured after treatment with DLMC,DL +US,DL +verapamil +US or DLMC +US.The results were shown as relative fluorescence units (RFUs).*p b 0.05,**p b 0.01.114Z.Deng et al./Journal of Controlled Release 174(2014)109–116。