Journal of Controlled Release 153 (2011) 16–22

Contents lists available at ScienceDirect

Journal of Controlled Release

j ourna l homepage: www.e lsev ie r.com/ locate / jconre l

Tumor targeted quantum dot-mucin 1 aptamer-doxorubicin conjugate for imagingand treatment of cancer

Ronak Savla a, Oleh Taratula a, Olga Garbuzenko a, Tamara Minko a,b,⁎a Department of Pharmaceutics, Rutgers, The State University of New Jersey, Piscataway, NJ, 08854, USAb Cancer Institute of New Jersey, 195 Little Albany Street, New Brunswick, NJ 08903, USA

⁎ Corresponding author at: Department of PharmacPharmacy, Rutgers, The State University of New JerPiscataway, NJ 08854–8020. Tel.: +1 732 445 3831x21

E-mail address: minko@rci.rutgers.edu (T. Minko).

0168-3659/$ – see front matter © 2011 Elsevier B.V. Adoi:10.1016/j.jconrel.2011.02.015

a b s t r a c t

a r t i c l e i n f o

Article history:Received 11 October 2010Accepted 11 February 2011Available online 20 February 2011

Keywords:Quantum dotDoxorubicinAptamerDrug deliveryImagingNanomedicine

In this study, we report the design and delivery of a tumor-targeted, pH-responsive quantum dot-mucin1aptamer-doxorubicin (QD-MUC1-DOX) conjugate for the chemotherapy of ovarian cancer. To achieve activecancer targeting, QD was conjugated with a DNA aptamer specific for mutated MUC1 mucin overexpressed inmany cancer cells including ovarian carcinoma. DOXwas attached to QD via a pH-sensitive hydrazone bond inorder to provide the stability of the complex in systemic circulation and drug release in acidic environmentinside cancer cells. The data show that this bond is stable at neutral and slightly basic pH and undergoes rapidhydrolysis in mildly acidic pH. Confocal microscopy and in vivo imaging studies show that the developed QD-MUC1-DOX conjugate had higher cytotoxicity than free DOX in multidrug resistant cancer cells andpreferentially accumulated in ovarian tumor. Data obtained demonstrate a high potential of the proposedconjugate in treatment of multidrug resistant ovarian cancer.

eutics, Ernest Mario School ofsey, 160 Frelinghuysen Road,4; fax: +1 732 445 3134.

ll rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Ovarian cancer is one of the most common gynecologicalmalignancies in industrialized nations. Standard treatment of ovariancarcinoma involves aggressive surgery followed by chemotherapy.Micronodular and floating tumor colonies, which are spread withinthe peritoneal cavity, cannot be adequately treated by surgery orradiation and require extensive chemotherapy. Several approachesare being used for chemotherapy of ovarian cancer including thetreatment with platinum derivatives [1,2] and doxorubicin (DOX)delivered by different nanocarriers, such as liposomes (Doxil) [3–5],polymers [2,6], gold nanoparticles [7], mesoporous silica films[8], anddendrimers [9]. Although most patients respond effectively to initialtherapy, the efficiency of the treatment usually rapidly decreasesbecause of the development of cancer cellular resistance againstchemotherapy. The development of such resistance and highrecurrence rates of ovarian carcinoma represent a challenge for thetreatment of this disease. Ovarian cancer patients that do not respondto initial chemotherapy or relapse after achieving a response aregenerally incurable [10]. In order to enhance the drug uptake byresistant ovarian cancer cells, increase the efficiency of treatment andsubstantially limit adverse side effects of high dose chemotherapy, atargeting of anticancer drug(s) specifically to tumor cells can beemployed [11]. Consequently, the development of novel drug delivery

vehicles capable to deliver an anticancer drug specifically to ovariancancer cells and to effectively penetrate inside resistant tumor cellsreleasing their payload is an important task in enhancing thetreatment of ovarian carcinoma.

Quantum dots (QD) are newer nanoparticles that have garneredextensive investigation as potential drug delivery vehicles. QD re-present nanometer size semiconductor particles that are gainingpopularity as fluorescent markers and delivery vehicles. QD inher-ently possess numerous advantages over traditional fluorescent dyessuch as increased photostability, higher brightness, and narrowfluorescence spectra [12,13]. QD can be designed to have emissionpeaks at different wavelengths by adjusting their size. This customi-zation allows QD to be produced for specific functions such asFluorescence Resonance Energy Transfer (FRET), in which a specificemission wavelength can transfer energy to drug molecules orphotosensitizers. Consequently, FRET phenomenon can be potentiallyused for the development of novel therapeutic approaches for cancertreatment [14]. Furthermore, functionalized QD are versatile and canbe modified with biomolecules such as antibodies, small molecules,peptides, and aptamers [12,13]. Due to these properties, QD can beutilized to design multifunctional systems for simultaneous imagingand delivery of therapeutic agents, a task that represents a substantialchallenge for other types of nanocarriers. Recent reports showed thatQD can be used in photodynamic and radiation therapy [15], whichwould make QD a prospective therapeutic drug carrier since radiationtherapy is employed as an adjunct for the treatment of many types ofcancers. In addition, quenching of fluorescence of QD by a cargo due toFRET would allow for monitoring the attachment and release of QDpayload.

17R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

Mucin 1 (MUC1) cell surface associated mucin is mutated andoverexpressed on the surface of many cancer cells [16]. Aberrantforms of MUC1 are overexpressed in more than 90% of late stageepithelial ovarian cancers and associated metastatic lesions [17].Consequently, specific targeting of drug delivery system to thismutated mucin has a potential to improve internalization of ananticancer drug by resistant ovarian cancer cells and limit adverseside effects of chemotherapy on healthy organs. The design andcharacterization of a DNA aptamer, an oligonucleotide, specific forMUC1 was previously reported [18]. The authors demonstrated thatthe aptamer had a higher binding affinity for extracellular mutatedmucin and lower immunogenicity than monoclonal antibodies.Therefore, it is logical to assume that the use of such aptamer as atargeting moiety and penetration enhancer specific to ovarian cancercells may improve the efficacy of treatment and imaging of bothprimary and metastatic ovarian tumors.

Besides the specific tumor targeting, a cancer specific drug deliverysystem should protect an anticancer drug from degradation during itsjourney in the systemic circulation and provide for an effective releaseof its payload inside the cancer cells. A pH-sensitive hydrazone bond isoften being used to conjugate a delivered substance with a vehicle toachieve required stability of payload in mildly basic or neutralextracellular environment and its release under acidic conditionsinside endo- or lysosomes after the internalization of an entire systemby cancer cells [19–23].

Based on the aforementioned consideration, we hypothesize that adrug delivery system that consists of (1) QD as a carrier, (2) DOX as ananticancer drug conjugated to QD via a pH-sensitive hydrazone bondand (3) aptamer targeted to mutated MUC1 mucin as a targetingmoiety/penetration enhancer will be exceptionally effective forimaging and treatment of drug resistant ovarian malignances. Suchsystem can form the basis for a multi-pronged attack on resistantovarian cancer and metastases by simultaneously fulfilling thefollowing tasks: (1) increasing penetration of anticancer drug intoovarian cancer cells; (2) preventing adverse side effects of chemo-therapy by targeting the entire system specifically to primary ovariancancer cells; (3) effective imaging of primary and metastatic cancercells; (4) protection of drug in systemic circulation and release insidecancer cells; and (5) monitoring drug release in targeted ovariancancer cells. The present study is aimed at testing this hypothesis.

2. Materials and methods

2.1. Synthesis of quantum dot-MUC1 aptamer conjugates

Carboxyl terminated quantum dots (QD-COOH) with an emissionpeak at 490 nm (eBioscience, Inc., San Diego, CA) were used as carriers.One nanomole of QD-COOH was added to 1225 μL of 10 mM boratebuffer (pH=7.4) and mixed well by stirring. After this, 200 μL of500 μM 5′-amine terminated MUC1 DNA aptamer [5′-GCA-GTT-GAT-CCT-TTG-GAT-ACC-CTG-G-3′] (Invitrogen, Carlsbad, CA) were added tothe reaction mixture under continued stirring. Further, 50 μL of 50 mM1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) (Sigma-Aldrich, St. Louis, MO) and 25 μL of 25 mM N-hydroxysuccinimide(NHS) (Sigma-Aldrich, St. Louis, MO) were added to the reactionmixture and incubated with gentle stirring for 2 h at room temperatureprotected from light. After the incubation, the mixture was centrifugedusing Microcon YM-30 Centrifugal Filter Unit (Millipore, Billerica, MA)with molecular weight cutoff of 30 kDa for 10 min at 13,000 rpm for 5buffer exchanges using 50 mM borate buffer (pH=8.3) and recon-stituted in 1 mL H2O.

2.2. Synthesis of quantum dot-MUC1 aptamer-doxorubicin

After the purification of the quantum dot-MUC1 conjugate bycentrifugation, the solution was concentrated to 500 μL. Afterwards,

500 μL anhydrous dimethyl sulfoxide (Acros Organics, Morris Plains,NJ), 50 μL of 50 mM EDC, 25 μL of 25 mM NHS, 40 μL hydrazinemonohydrate (Sigma-Aldrich, St. Louis, MO, 64% v/v) were added tothe reaction mixture and incubated for 3 h at room temperature withgentle stirring. After the incubation, the reaction solution was dilutedto 3 mL with H2O and dialyzed by the centrifugation as describedabove for 10 min at 13,000 rpm for 3 exchanges with H2O. Theretentate was concentrated to 500 μL using anhydrous dimethylsulfoxide (DMSO, Acros Organics, Morris Plains, NJ). Then, 100 μL ofdoxorubicin hydrochloride (10 mg/mL, Sigma-Aldrich, St. Louis, MO)and 600 μL anhydrous DMSOwere added and allowed to react for 48 hat room temperature under gentle stirring after adding a few drops ofacetic acid (Fisher Scientific, Pittsburg, PA) to act as a catalyst. At theend of the incubation, the reaction solution was diluted to 3 mL usingH2O, centrifuged as described above and concentrated to 1 mL usingH2O.

2.3. Synthesis of quantum dot-doxorubicin conjugates

One nanomole of QD-COOH was added to 875 μL of anhydrousdimethyl sulfoxide, 50 μL of 50 mM EDC, 25 μL of 25 mM NHS, and40 μL of hydrazine monohydrate (64% v/v) and was allowed to reactfor 3 h at room temperature with gentle stirring and diluted to 3 mLusing H2O. The solution was dialyzed by the centrifugation asdescribed above for 10 min at 13,000 rpm for 3 exchanges of H2Oand concentrated to 500 μL. After this, 100 μL of DOX (10 mg/mL) and600 μL of anhydrous DMSOwere added to the solution; a few drops ofacetic acid were added as a catalyst and allowed to react for 48 h atroom temperature under gentle stirring. The reaction mixture wasdiluted to 3 mL with H2O, dialyzed by the centrifugation as describedabove for 3 exchanges of H2O and concentrated to 1 mL with H2O.

2.4. Determination of doxorubicin loading

After removal of free DOX by a centrifugation dialysis, theabsorbance of the QD-MUC1-DOX conjugate at 492 nmwasmeasuredusing a microplate reader (GENios, Tecan US, Inc., Durham, NC). QD-MUC1 solution with the same concentration of QD as in the QD-MUC1-DOX conjugate was used as a reference. DOX concentration inthe conjugate was calculated from a standard curve of DOX absor-bance created by measuring the absorbance of several predefinedDOX concentrations.

2.5. Hydrolytic release of doxorubicin

In order to determine the rate of DOX release from QD conjugates,the conjugate was incubated in an acidic environment. To this end, aquantum dot-doxorubicin conjugate was dissolved in Dulbecco'sphosphate buffered saline (DPBS, Lonza, Inc., Mapleton, IL) with pHadjusted to 5. At pre-determined time intervals, small samples weretaken from the bulk solution. Released free DOX was eliminated by acentrifugation dialysis using Microcon YM-30 Centrifugal Filter Unitwith molecular weight cutoff of 30 kDa for 10 min at 13,000 rpm for 3exchanges of H2O. The remaining concentration of DOX in the samplewas determined by measuring the absorbance at 492 nm using thestandard curve and the amount of DOX released from the samples wasdetermined by comparing DOX concentration in the sample with aninitial concentration in the intact conjugate.

2.6. Cell line

The A2780/AD multidrug resistant human ovarian carcinoma cellline was obtained from T. C. Hamilton (Fox Chase Cancer Center,Philadelphia, PA). Cells were cultured in RPMI 1640 medium (Sigma,St. Louis, MO) supplemented with 10% fetal bovine serum (FisherChemicals, Fairlawn, NJ) and 1.2 mL/100 mL penicillin–streptomycin

18 R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

(Sigma, St. Louis, MO). Cells were grown at 37 °C in a humidifiedatmosphere of 5% CO2 (v/v) in air. All experiments were performed oncells in the exponential growth phase.

2.7. Cytotoxicity assay

The cellular cytotoxicity of QD, MUC1 aptamer-QD, free DOX andall drug-QD-aptamer conjugates were accessed using a modified MTT(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)assay as previously described [24–26].

2.8. Cellular internalization and localization

Cellular uptake of non-modified QD, MUC1 aptamer-QD conjugateand QD-MUC1-DOX conjugates was studied by a confocal microscopy.To assess intracellular distribution of the substances, 12 optical sec-tions, known as a z-series, were scanned sequentially along the verti-cal (z) axis from the top to the bottom of the cell.

2.9. Animal tumor model, in vivo imaging and organ distribution

An animal model of human ovarian carcinoma xenografts wasused as previously described [27–30]. Briefly, A2780/AD multidrugresistant human ovarian carcinoma cells (2×106) were subcutane-ously transplanted into the flanks of female athymic nu/nu mice.When the tumors reached a size of about 0.2–0.3 cm3 (15–20 daysafter transplantation), mice were treated intravenously with non-modified QD and tumor-targeted by MUC1 aptamer QD conjugate(QD-MUC1). Animals were anesthetized with isoflurane using theXGI-8 Gas Anesthesia System (Xenogen, Alameda, CA). Fluorescenceof injected QD was visualized using IVIS imaging system (Xenogen,Alameda, CA) 24 h after the treatment. Visible light and fluorescenceimages were taken and overlaid using the manufactures software toobtain composite images.

2.10. Statistical analysis

Data obtained were analyzed using descriptive statistics, singlefactor analysis of variance (ANOVA) and presented as mean value±the standard deviation (SD) from four to eight independent mea-surements in separate experiments. The comparison among groupswas performed by the independent sample Student's t-tests. Thedifference between variants is considered significant if Pb0.05.

3. Results

Carboxyl-terminated QD were conjugated with MUC1 aptamerand DOX (Fig. 1). The MUC1 aptamer was modified with a 5′- primaryamine which was conjugated to the carboxyl groups on the surface ofQD forming an amide bond. Conjugation of aptamer onto QD wasconfirmed using gel electrophoresis (results not shown). Approxi-

Fig. 1. Synthesis of quantum dot-MUC1 aptame

mately five aptamers per quantum dot were attached. On the nextstep, DOX was loaded onto the QD-aptamer conjugate. In order toprovide the release of DOX in the acidic environment inside cancercells, the drug was attached to the QD using an acid labile hydrazonebond [31]. DOX absorbance was used to calculate loading the drugonto QD. Using a standard concentration curve, we calculated thatapproximately 46 DOXmolecules were bound per one QDmolecule inthe QD-MUC1-DOX conjugate.

The absorbance of DOX and fluorescence emission of used QDweremeasured. As can be seen from Fig. 2A, normalized DOX absorbanceand QD emission spectra overlap. This creates precondition for socalled Fluorescence Resonance Energy Transfer (FRET) between QDand DOX and for quenching of QD fluorescence. A direct measurementof fluorescence of naked QD and QD-DOX conjugates under the360 nm excitation (Fig. 2B) confirms quenching of QD fluorescence byFRET mechanism. The release of DOX from QD-DOX conjugate in anacidic environment was studied. A direct measurement showed thatabout 35% of bound DOX was released from QD-DOX conjugate afterfive hours of incubation in an acidic environment (Fig. 2C).

Cellular uptake studies were performed and the results wereevaluated using confocal microscopy. Aminimal uptake of unmodifiedquantum dots by A2780/AD cells was found (Fig. 3). Modification ofQD with tumor targeting moiety/penetration enhancer MUC1 apta-mer led to the more substantial uptake of QD-MUC1 conjugate byresistant cancer cells. Similarly, strong fluorescent signal was regis-tered inside cancer cells after incubation with QD-MUC1-DOX conju-gate. Furthermore, the fluorescence intensity of targeted conjugatecontaining DOX was even more pronounced when compared withQD-MUC1 conjugate. This fact clearly not only shows enhanced up-take of tumor targeted conjugates but also confirms DOX release fromthe conjugate inside cancer cells, separation of QD and DOX, restora-tion of QD fluorescence as a result of stopping fluorescence quenchingby FRET after the separation and doubling of fluorescence intensity bysimultaneous fluorescence of QD and DOX.

The cytotoxicity of unmodified and modified quantum dots wereanalyzed by modified MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-nyl tetrasodium bromide) assay in DOX-resistant A2780-AD humanovarian cancer cells. The results show no apparent cytotoxicity ofunmodified QD and MUC1 aptamer-QD conjugate at tested concen-trations (Fig. 4). The measurement of IC50 dose of free DOX resulted inthe relatively high value (~4 μM) comparable to that measured morethan 10 years ago in the same cell line [25]. This in turn shows thatA2780/AD human ovarian carcinoma cells did not lose their resistanceto doxorubicin. Surprisingly, we did not find a statistically significantincrease in cytotoxicity after conjugation of low molecular weightDOX to high molecular weight QD. Targeting QD-DOX conjugatesspecifically to cancer cells led to the statistically significant increase intoxicity and decrease in IC50 value. These data correlate well with theresults of cellular uptake study.

In vivo organ distribution of targeted and non-targeted QDwas carried out in nude mice model of tumor xenografts. Multidrug

r doxorubicin (QD-MUC1-DOX) conjugate.

0

5

10

15

20

25

400 450 500 550 600

0

5

10

15

20

25

30

35

40

0 2 4 6

Wavelength, nm

Time, h

Flu

ore

scen

ce, a

. u.

DO

X r

elea

sed

, %

B

C

QD

QD-DOX

0

0.4

0.8

1.2

400 450 500 550 600

Wavelength, nm

No

rm. i

nte

nsi

ty, r

el. u

nit

s

A

QD emission

DOX absorption

Fig. 2. Characterization of non-modified quantum dots (QD)-doxorubicin (DOX)conjugates. (A) Overlap of QD fluorescence (blue curve) and DOX absorbance (redcurve); (B) quenching of QD fluorescence by DOX; (C) release of DOX from QD-DOXconjugate in an acidic environment. Means±S.D. are shown.

Fig. 3. Intracellular localization of non-modified quantum dots (QD), cancer celltargeted by MUC1 aptamer QD (QD-MUC1), and MUC1 aptamer QD-DOX (QD-MUC1-DOX) conjugates in multidrug resistant A2780/AD human ovarian carcinoma cells.Representative confocal microscope images of cells incubated for 24 h with substancesindicated (z-series, from the top of the cell to the bottom).

19R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

resistant human ovarian cancer cells, A2780/AD, were implantedsubcutaneously above the right hind leg of the mice and allowed togrow. Unmodified quantum dots and tumor targeted quantum dots(QD-MUC1) were injected intraperitoneally. The whole body fluores-cent images taken by IVIS system one hour after injection show higheraccumulation of the targeted conjugate (QD-MUC1) in the tumor andlower accumulation in other studied organs when compared with theunmodified quantum dot (Fig. 5).

4. Discussion

In the present study we constructed and evaluated in vitro andin vivo non-targeted and tumor targeted quantum dot-doxorubicinconjugates. DOX was attached to the QD using an acid labile hydra-zone bond [31]. The hydrazone bond is being widely used for theattachment of doxorubicin to structurally and chemically differenttypes of delivery vehicles [31,32]. However, the specific type of

hydrazone bond utilized may have dramatic effects on the doxoru-bicin activity. Lee et al. [32] studied acyl hydrazone linkages andhydrazone carboxylate linkages. The hydrazone carboxylate linkagesreleased a derivative of doxorubicin due to an intramolecular cycli-zation reaction. This derivative did not retain the activity of DOX. Inaddition, the hydrazone carboxylate linkage was not sufficientlystable at physiological pH. The authors discovered that in the case ofacyl hydrazone bonds, DOX is released without any chemical modi-fications and that the linkage is stable at neutral pH. Based on thesefindings we used an acyl hydrazone bond to conjugate DOX to quan-tum dots. Our results show a relatively higher doxorubicin loadingefficiency of QD via this bond. It was found that around 46 moleculesof DOX were bound per one QD molecule. Using molecular weight of

0

20

40

60

80

100

120

0.01 0.1 1 10

Via

bili

ty, %

0

1

2

3

4

5

IC50

, µM

Incr

ease

in c

yto

toxi

city

DOX QD-DOX QD-MUC1-DOX

*†

QD

DOX

QD-DOX

QD-MUC1-DOX

QD-MUC1

Fig. 4. Cytotoxicity of different formulations of quantum dots (QD). Multidrug resistantA2780/AD human ovarian carcinoma cell were incubated with the indicatedformulations. Upper panel: cellular viability of formulations with and without DOX.Bottom panel: IC50 doses of formulations that contain DOX. Means±SD are shown.*Pb0.05 when compared with free DOX. †Pb0.05 when compared with free QD-DOX.

20 R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

QD and DOX reported by the manufacturers (~150 kDa and ~580 Da,respectively), one can calculate the loading level equal to 17 wt.%. Thisdemonstrates a relatively high loading capacity of quantum dotscomparable with the loading capacity of other carriers. For instance,Prabaharan et al. [7] used gold nanoparticles to deliver doxorubicin

Fig. 5. Organ and tumor content of non-modified quantum dots (QD) and tumor-targetedovarian carcinoma cells were transplanted subcutaneously into the flanks of nudemice. QD aother organs was analyzed in live anesthetized animals 24 h after injection using IVIS imagireflecting the lowest intensity and red color reflecting the highest intensity. After measuringexcised and their fluorescence was registered and processed by the imaging system.

and reported a DOX loading level of 17 wt.%. A loading capacity of14 wt.% was achieved using superparamagnetic iron oxide nanopar-ticles [33]. Using N-(2-hydroxypropyl)methacrylamide (HPMA)copolymers, Etrych et al. [31] achieved 12 wt.% loading capacity ofDOX bound to the polymer via an acid labile hydrazone bond.

The results of fluorescence studies demonstrated quenching of QDfluorescence by FRET between doxorubicin and QD. Fluorescencemeasurements revealed that QD used in the present study absorbslight with a wavelength of 390 nm and emits at 490 nm. DOX absor-bance spectrum showed that DOX can absorb light with this wave-length. Due to the proximity of DOX and QD in a conjugate, thefluorescence signal of QD is absorbed by DOX, which then can emit asignal at 560 nm. Based on the measured fluorescence spectrum, it isclear that attachment of doxorubicin quenches the QD fluorescencesignal. According to the literature data, eight DOX molecules arerequired to fully quench the fluorescence of one quantum dot [12].Based on DOX loading registered in the present study, synthesizedQD-DOX conjugates exceed this requirement and demonstratedcomplete quenching of QD fluorescence. It is known that when themolecules that emit and absorb light are separated beyond 10 nm,FRET no longer can occur [34]. Consequently, the release of DOX fromthe conjugates and its diffusion beyond 10 nm from QD shoulddiscontinue FRET and one would be able to register an increase in QDfluorescence and probably a certain decrease in DOX fluorescence.This would allow monitoring of DOX release in vitro and in-vivo usingfluorescent imaging techniques. This in turn increases the significanceof the constructed QD-DOX conjugates for both mechanistic studiesand monitoring QD-DOX accumulation and internalization in thetargeted sites during chemotherapy.

The data obtained in the present study confirmed an efficientrelease of DOX from QD-DOX conjugates when nearly 35% of loadeddrug was released after five hours of incubation of conjugates in vitroin an acidic environment. This release rate can be considered asrelatively high when compared with that previously measured for asimilar type of hydrazone bond [35]. In that study, approximately 45%of the DOX loaded into polymeric micelles and bound by pH-sensitive

by MUC1 aptamer QD (QD-MUC1) conjugate. Multidrug resistant A2780/AD humannd QD-MUC1were injected intravenously intomice. The distribution of QD in tumor andng system. The intensity of fluorescence is expressed by different colors with blue colorthe fluorescence in the entire animal, mice were euthanized and tumor and organs were

21R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

hydrazone bonds was released after 24 h of incubation of micelles inacidic conditions. Several factors can potentially be responsible forthe registered high rate of DOX release from constructed QD-DOXconjugates. In our system, DOX was attached on the periphery of theconjugates. This may lead to the more favorable exposure of thehydrazone bond to the acidic environment when compared withmicelles used in the cited work [35]. Second, the micelles used by theauthors were much larger than QD employed in the present study.This may have also hindered the exposure to the acidic environment.It would be premature to transfer these results onto intracellularrelease of DOX from the conjugates because the conditions in theendosomes and tumor microenvironment are different from simple invitro incubation settings. Nonetheless, the results provide a founda-tion for an effective delivery of DOX by QD to the tumor and releaseinside cancer cells. It should also be stressed that the handling andprocessing of the sample for measurements may have had an insig-nificant impact on the results because the hydrazone linkage is quitestable in neutral pH and should not have been disturbed by the shortcentrifugation process. Moreover, our in vitro and in vivo data clearlyconfirmed that DOX delivered by tumor targeted QD-MUC1-DOXconjugates accumulated predominately in the tumor and effectivelyreleased inside cancer cells. These results also allow us to hypothesizethat the release of DOX from the conjugatesmost probably occurred inan acidic environment inside lysosomes after the internalization ofconjugates by endocytosis and fusion of lysosomes with conjugate-containing endosomes.

Due to the presence in QD cadmium and selenium, which haveshown to be toxic when accumulated, toxicity of QD remains a primaryconcern in their clinical applications. The results show no apparentcytotoxicity of unmodified QD at tested concentrations. The presence ofthe zinc sulfide shell and polymer or lipid coating dramatically reducedexposure to the toxic quantum dot core [36]. Currently there are nostudies evaluating long term toxicities of quantum dots. Studies onshort term toxicities (up to 72 h of exposure) have yielded differentresults. The majority of the studies show that CdSe/ZnS quantum dotshave no toxic effects [36]. Although, the low toxicity of drug carriers isnot very important for their use in cancer chemotherapy when thecarrier is conjugatedwith highly toxic anticancer drug, our data showedthat QD employed in the present study were not toxic in the con-centrations used. This forms a basis for their possible applications forthe delivery of relatively low toxic drugs for clinical use other thanchemotherapy.

Based on our previous data on HPMA copolymer-bound DOX thatshowed significantly (more than six-folds) lower cytotoxicity of highmolecular weight conjugated DOX when compared with lowmolecular weight free DOX [25], we expected a substantial increasein the IC50 dose (decrease in cytotoxicity) for QD-DOX conjugates.Moreover, because QD have substantially higher molecular weight(~150 KDa for QD versus ~20 KDa for the studied polymeric conju-gate) the expected decrease in the toxicity could be even more pro-nounced. However, no statistically significant difference was found inthe present study between the IC50 doses for free DOX and QD-DOXconjugates. Such discrepancy between our expectations and thereality confirms higher penetration ability of non-targeted QD-DOXconjugates in multidrug resistant cancer cells and effective DOXrelease in the intracellular environment. It also suggests that QD-DOXconjugates were internalized by multidrug resistant cancer cells via apathway different from diffusion (probably by endocytosis) allowingto overcome existing drug efflux pumps overexpressed in thesecells [25,37]. Targeting of QD-DOX conjugates to cancer cells by MUC1aptamer further increased their cytotoxicity showing a remarkablepotential of tumor targeted QD as delivery vehicles for anticancerdrugs.

A comparison of in vivo images reflecting the biodistribution ofnon-targeted and tumor targeted QD-DOX conjugates show that bothpassive and active tumor targeting mechanisms [11] are involved in

quantum accumulation within the tumor. The Enhanced Permeationand Retention (EPR) effect is likely the phenomenon responsible forpassive targeting of QD-DOX conjugates to tumor. The active targetingwas achieved by incorporating the MUC1 specific aptamer into QD-DOX conjugates. Although this aptamer is not well validated, it doesenhance the accumulation of QD within the tumor and cancer cells asevidenced by in vivo IVIS and in vitro confocal imaging. The presentexperimental data clearly showed that MUC1 aptamer as a tumor-targeting moiety is responsible for a preferential accumulation of QD-MUC1-DOX conjugates in the ovarian tumor, their effective internal-ization by cancer cells and release of the drug inside the cells.

5. Conclusions

In this paper, we have demonstrated the design and functionalityof tumor targeted quantum dot-based anticancer drug deliverysystem. The system demonstrated the ability to preferentially targetovarian cancer cells, efficiently released doxorubicin in acidic pH, andhad higher toxicity in ovarian cancer cells when compared with freedoxorubicin. We also showed that quantum dots have potential ascarriers of drugs and as in vivo imaging agents. Furthermore, the dataconfirmed that the MUC1 aptamer may be an effective alternative toconventional targeting agents such as peptides and antibodies.

References

[1] R.C. Bast Jr., B. Hennessy, G.B. Mills, The biology of ovarian cancer: newopportunities for translation, Nat. Rev. Cancer 9 (2009) 415–428.

[2] M. Markman, M.A. Bookman, Second-line treatment of ovarian cancer, Oncologist5 (2000) 26–35.

[3] D.J. Porche, Liposomal doxorubicin (Doxil), J. Assoc. Nurses AIDS Care 7 (1996)55–59.

[4] I. Steppan, D. Reimer, U. Sevelda, H. Ulmer, C. Marth, A.G. Zeimet, Treatment ofrecurrent platinum-resistant ovarian cancer with pegylated liposomal doxorubi-cin—an evaluation of the therapeutic index with special emphasis on cardiactoxicity, Chemotherapy 55 (2009) 391–398.

[5] T. Tejada-Berges, C.O. Granai, M. Gordinier, W. Gajewski, Caelyx/Doxil for thetreatment of metastatic ovarian and breast cancer, Expert Rev. Anticancer Ther. 2(2002) 143–150.

[6] A. Malugin, P. Kopeckova, J. Kopecek, Liberation of doxorubicin from HPMAcopolymer conjugate is essential for the induction of cell cycle arrest and nuclearfragmentation in ovarian carcinoma cells, J. Control. Release 124 (2007) 6–10.

[7] M. Prabaharan, J.J. Grailer, S. Pilla, D.A. Steeber, S. Gong, Gold nanoparticles with amonolayer of doxorubicin-conjugated amphiphilic block copolymer for tumor-targeted drug delivery, Biomaterials 30 (2009) 6065–6075.

[8] T. Lebold, C. Jung, J. Michaelis, C. Brauchle, Nanostructured silicamaterials as drug-delivery systems for doxorubicin: single molecule and cellular studies, Nano Lett.9 (2009) 2877–2883.

[9] P.S. Lai, P.J. Lou, C.L. Peng, C.L. Pai, W.N. Yen, M.Y. Huang, T.H. Young, M.J. Shieh,Doxorubicin delivery by polyamidoamine dendrimer conjugation and photo-chemical internalization for cancer therapy, J. Control. Release 122 (2007) 39–46.

[10] G. Oskay-Ozcelik, J. Sehouli, Pros and cons for systemic therapy in recurrentovarian cancer, Anticancer Res. 29 (2009) 2831–2836.

[11] T. Minko, S.S. Dharap, R.I. Pakunlu, Y. Wang, Molecular targeting of drug deliverysystems to cancer, Curr. Drug Targets 5 (2004) 389–406.

[12] V. Bagalkot, L. Zhang, E. Levy-Nissenbaum, S. Jon, P.W. Kantoff, R. Langer, O.C.Farokhzad, Quantum dot-aptamer conjugates for synchronous cancer imaging,therapy, and sensing of drug delivery based on bi-fluorescence resonance energytransfer, Nano Lett. 7 (2007) 3065–3070.

[13] A.M. Smith, H. Duan, A.M. Mohs, S. Nie, Bioconjugated quantum dots for in vivomolecular and cellular imaging, Adv. Drug Deliv. Rev. 60 (2008) 1226–1240.

[14] R. Bakalova, H. Ohba, Z. Zhelev, M. Ishikawa, Y. Baba, Quantum dots asphotosensitizers? Nat. Biotechnol. 22 (2004) 1360–1361.

[15] P. Juzenas, W. Chen, Y.P. Sun, M.A. Coelho, R. Generalov, N. Generalova, I.L.Christensen, Quantum dots and nanoparticles for photodynamic and radiationtherapies of cancer, Adv. Drug Deliv. Rev. 60 (2008) 1600–1614.

[16] M.A. Hollingsworth, B.J. Swanson, Mucins in cancer: protection and control of thecell surface, Nat. Rev. Cancer 4 (2004) 45–60.

[17] L. Wang, J. Ma, F. Liu, Q. Yu, G. Chu, A.C. Perkins, Y. Li, Expression of MUC1 inprimary and metastatic human epithelial ovarian cancer and its therapeuticsignificance, Gynecol. Oncol. 105 (2007) 695–702.

[18] C.S. Ferreira, C.S. Matthews, S. Missailidis, DNA aptamers that bind to MUC1tumour marker: design and characterization of MUC1-binding single-strandedDNA aptamers, Tumour Biol. 27 (2006) 289–301.

[19] Y. Bae, A.W. Alani, N.C. Rockich, T.S. Lai, G.S. Kwon, Mixed pH-Sensitive PolymericMicelles for Combination Drug Delivery, Pharm Res 27 (2010) 2421–2432.

22 R. Savla et al. / Journal of Controlled Release 153 (2011) 16–22

[20] Q. Yuan, W.A. Yeudall, H. Yang, PEGylated polyamidoamine dendrimers with bis-aryl hydrazone linkages for enhanced gene delivery, Biomacromolecules 11(2010) 1940–1947.

[21] A.A. Kale, V.P. Torchilin, Environment-responsive multifunctional liposomes,Methods Mol Biol 605 (2010) 213–242.

[22] S. Aryal, C.M. Hu, L. Zhang, Polymer–cisplatin conjugate nanoparticles for acid-responsive drug delivery, ACS Nano 4 (2009) 251–258.

[23] S.S. Banerjee, D.H. Chen, Multifunctional pH-sensitive magnetic nanoparticles forsimultaneous imaging, sensing and targeted intracellular anticancer drugdelivery, Nanotechnology 19 (2008) 505104.

[24] S. Jayant, J.J. Khandare, Y.Wang, A.P. Singh, N. Vorsa, T. Minko, Targeted sialic acid-doxorubicin prodrugs for intracellular delivery and cancer treatment, Pharm. Res.24 (2007) 2120–2130.

[25] T. Minko, P. Kopeckova, V. Pozharov, J. Kopecek, HPMA copolymer boundadriamycin overcomes MDR1 gene encoded resistance in a human ovariancarcinoma cell line, J. Control. Release 54 (1998) 223–233.

[26] R.I. Pakunlu, Y. Wang, W. Tsao, V. Pozharov, T.J. Cook, T. Minko, Enhancement ofthe efficacy of chemotherapy for lung cancer by simultaneous suppression ofmultidrug resistance and antiapoptotic cellular defense: novel multicomponentdelivery system, Cancer Res. 64 (2004) 6214–6224.

[27] P. Chandna, M. Saad, Y.Wang, E. Ber, J. Khandare, A.A. Vetcher, V.A. Soldatenkov, T.Minko, Targeted proapoptotic anticancer drug delivery system, Mol. Pharm. 4(2007) 668–678.

[28] S.S. Dharap, Y. Wang, P. Chandna, J.J. Khandare, B. Qiu, S. Gunaseelan, P.J. Sinko, S.Stein, A. Farmanfarmaian, T. Minko, Tumor-specific targeting of an anticancerdrug delivery system by LHRH peptide, Proc. Natl Acad. Sci. USA 102 (2005)12962–12967.

[29] R.I. Pakunlu, Y. Wang, M. Saad, J.J. Khandare, V. Starovoytov, T. Minko, In vitro andin vivo intracellular liposomal delivery of antisense oligonucleotides andanticancer drug, J. Control. Release 114 (2006) 153–162.

[30] M. Saad, O.B. Garbuzenko, E. Ber, P. Chandna, J.J. Khandare, V.P. Pozharov, T.Minko, Receptor targeted polymers, dendrimers, liposomes: which nanocarrier isthe most efficient for tumor-specific treatment and imaging? J. Control. Release130 (2008) 107–114.

[31] T. Etrych, M. Jelinkova, B. Rihova, K. Ulbrich, New HPMA copolymers containingdoxorubicin bound via pH-sensitive linkage: synthesis and preliminary in vitroand in vivo biological properties, J. Control. Release 73 (2001) 89–102.

[32] C.C. Lee, E.R. Gillies, M.E. Fox, S.J. Guillaudeu, J.M. Frechet, E.E. Dy, F.C. Szoka, Asingle dose of doxorubicin-functionalized bow-tie dendrimer cures mice bearingC-26 colon carcinomas, Proc. Natl Acad. Sci. USA 103 (2006) 16649–16654.

[33] E. Munnier, S. Cohen-Jonathan, C. Linassier, L. Douziech-Eyrolles, H. Marchais, M.Souce, K. Herve, P. Dubois, I. Chourpa, Novel method of doxorubicin-SPIONreversible association for magnetic drug targeting, Int. J. Pharm. 363 (2008)170–176.

[34] E.A. Jares-Erijman, T.M. Jovin, FRET imaging, Nat. Biotechnol. 21 (2003) 1387–1395.[35] M. Hruby, C. Konak, K. Ulbrich, Polymeric micellar pH-sensitive drug delivery

system for doxorubicin, J. Control. Release 103 (2005) 137–148.[36] J.L. Pelley, A.S. Daar, M.A. Saner, State of academic knowledge on toxicity and

biological fate of quantum dots, Toxicol. Sci. 112 (2009) 276–296.[37] T. Minko, P. Kopeckova, J. Kopecek, Comparison of the anticancer effect of free and

HPMA copolymer-bound adriamycin in human ovarian carcinoma cells, Pharm.Res. 16 (1999) 986–996.