Biomaterials 35 (2014) 4333e4344

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Hyaluronic acid-chitosan nanoparticles for co-delivery of MiR-34a and doxorubicin in therapy against triple negative breast cancer Xiongwei Deng a, b,1, Minjun Cao b,1, Jiakun Zhang a, c, Kelei Hu a, b, Zhaoxia Yin b, Zhixiang Zhou b, Xiangqian Xiao b, Yishu Yang b, Wang Sheng b, *, Yan Wu a, *, Yi Zeng b a CAS Key Laboratory for Biomedical Effects of Nanomaterials and Nanosafety, National Center for Nanoscience and Technology of China, No. 11 Beiyitiao, Zhongguancun, Beijing 100190, PR China b College of Life Science and Bioengineering, Beijing University of Technology, No. 100 Pingleyuan, Beijing 100124, PR China c Key Laboratory of Pesticide Chemistry and Application, MOA, Institute of Plant Protection, CAAS, Beijing 100194, PR China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 January 2014 Accepted 4 February 2014 Available online 22 February 2014

Metastatic relapse, development of drug resistance in cancer cells and adverse side effects of chemotherapeutic agents are the major obstacles for effective chemotherapy against triple-negative breast cancer. To address these problems, miR-34a, a potent endogenous tumor suppressive molecule in breast cancer, was co-encapsulated with doxorubicin (DOX) into hyaluronic acid (HA)-chitosan (CS) nanoparticles (NPs) and simultaneously delivered into breast cancer cells for improved therapeutic effects of drug. DOX-miR-34a co-loaded HA-CS NPs were successfully prepared through ionotropic gelation method in water. In vitro and in vivo experiments showed that miR-34a and DOX can be efficiently encapsulated into HA-CS NPs and delivered into tumor cells or tumor tissues and enhance anti-tumor effects of DOX by suppressing the expression of non-pump resistance and anti-apoptosis proto-oncogene Bcl-2. In addition, intracellular restoration of miR-34a inhibited breast cancer cell migration via targeting Notch-1 signaling. The obtained data suggest that co-delivery of DOX and miR-34a could achieve synergistic effects on tumor suppression and nanosystem-based co-delivery of tumor suppressive miRNAs and chemotherapeutic agents may be a promising combined therapeutic strategy for enhanced anti-tumor therapy. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Nanoparticles Co-delivery MicroRNA-34a Doxorubicin Combined therapy

1. Introduction Breast cancer is the most common invasive malignancy in women worldwide and the second leading cause of cancer-related death in women after lung cancer [1]. Despite a significant progress in early diagnosis and treatment, resistance to conventional chemotherapeutics continuously poses a tremendous challenge to effective breast cancer therapy. In addition to metastasis, drug resistance is a major obstacle in the treatment of breast cancer which may potentially lead to tumor relapse and the failure of therapy [2]. Therefore, strategies for the overcoming of drug resistance could provide rational therapeutic approaches to increase chemotherapy efficacy and improve the clinical outcome of breast cancer patients.

* Corresponding authors. Tel.: þ86 10 82545614; fax: þ86 10 62656765. E-mail addresses: [email protected] (W. Sheng), [email protected] (Y. Wu). 1 Both authors contributed equally to this work. http://dx.doi.org/10.1016/j.biomaterials.2014.02.006 0142-9612/Ó 2014 Elsevier Ltd. All rights reserved.

MicroRNAs (miRNAs) are a class of small, endogenous noncoding RNAs that post-transcriptionally control the translation and stability of mRNAs [3]. Dysregulation of miRNAs has been shown in many cancers and recognized as a hallmark of cancer. MiRNAs can function as tumor suppressors or on-miRs during tumor initiation and progression [4]. The expression of tumor suppressive miRNAs is usually lower in tumoral cells. Restoration of synthetic mimics of tumor suppressive miRNAs is designed to recover the suppressive function of the endogenous miRNAs and have shown to induce cell apoptosis and block the proliferation of cancer cells [5,6]. MiRNA-based anti-cancer therapies are thus being under development, either alone or in combination with conventional anti-cancer agents, such as chemotherapeutic drugs, with the goal of improving survival and clinical outcome [7]. One single miRNA can target multiple genes due to the imperfect complementarity with target mRNA and simultaneously regulate different signaling pathways, implying that miRNA may potentially play a major regulatory role in coordinating cancerous signaling networks [8]. Importantly, multiple-targeting ability of tumor suppressive

4334

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

miRNAs is displaying an increasing interest for improved tumor treatment due to the complexity of cancer-related signaling networks [9]. Therefore, the therapeutic approach to restore homeostasis by altering miRNA expression has a great potential to be a more practical strategy than silencing individual genes by siRNAs. MicroRNA-34a (miR-34a) is a miRNA regulated by the p53 network at transcriptional level and has been shown to be remarkably downregulated in a variety of cancers [10]. Lower expression of miR-34a has been reported in triple negative and mesenchymal-type breast cancer cell lines. Exogenous expression of miR-34a in breast cancer cells induced cell apoptosis and inhibited cell proliferation and migration through targeting Bcl-2, CD44 and SIRT1 (silent information regulator 1), Rac1, Fra-1, Notch-1 and various cyclins [11e13]. A variety of viral carriers has been designed for miRNA delivery and has shown a high transfection efficiency over a broad range of cell types [14]. However, the safety concerns are currently perceived as hamper to the clinical application of viral vector-based therapy. Nanocarriers have been largely developed to encapsulate and deliver therapeutic agents, such as micelles and nanoparticles (NPs) formed with cationic polymers, polysaccharides, peptides and liposomes, and have attracted increasing attention in miRNA delivery due to their various advantages over the viral counterparts [15e18], including clinical potential and the ease of production. In addition to nanocarriers designed to deliver a single therapeutic agent, there are growing interests in developing multi-agent codelivery nanocarriers that can simultaneously incorporate and deliver multiple types of therapeutic payloads to disease sites in a targeted and controlled manner for combined therapy [19e21]. These co-delivery nanocarriers may potentially allow to create synergetic effects of different therapeutic approaches to eventually improve overall treatment outcomes [22].

Nowadays, polyionic nano-complexes (PICMs), which are composed of polycations and opposite polyanions, have demonstrated a great potential in biomedical and nano-biotechnological applications [23], including the controlled drug release, good biocompatibility and gene transfection [24,25]. From the point of view of nano-complexes formulation, polyionic nanoparticles have the advantage of being highly versatile and secure as a nanosystem which can be easily adapted to different charged cargos. Among numerous polyionic nano-complexes systems, hyaluronic acidchitosan nanoparticles (HA-CS NPs) have been broadly studied. Apart from its biocompatibility and biodegradability, the integrated HA backbone in itself is endowed with tumor-targeting property through specifically binding to CD44 molecule, an integral membrane glycoprotein over-expressed on the surface of various tumor cells, including MDA-MB-231 breast cancer cells [26], which makes it as an ideal polymer carrier for systemic drug delivery applications [27e29]. In our present studies, nano-complexes were prepared using hyaluronic acid and chitosan which simultaneously encapsulate positive charged Doxorubicin (DOX) and negative charged miR-34a mimics (Scheme 1). The particle size, surface zeta potential, morphology, DOX and miR-34a encapsulation efficiency, particle stability in various solutions and in vitro release were characterized. The synergetic effects and mechanisms of DOX and miR-34a in breast cancer therapy were further investigated by in vitro and in vivo models. 2. Materials and methods 2.1. Materials Doxorubicin hydrochloride salt (DOX) was purchased from Beijing HuaFeng Co. LTD (Beijing, China). Cell counting kit-8 was purchased from Dojindo Molecular Technologies (Tokyo, Japan). MicroRNA-34a (MiR-34a) and FAM-labeled miR-34a

Scheme 1. Schematic illustration of the construction of HA-CS-based nanosystem for the simultaneous co-delivery of DOX and miR-34a to MDA-MB-231 human breast cancer cells for enhanced anti-cancer effects.

X. Deng et al. / Biomaterials 35 (2014) 4333e4344 (FAM-miR-34a) were obtained from RiboBio Co. Ltd (Guangzhou, China). Sodium hyaluronate (HA) with an average molecular weight of 170 kDa was obtained from Dongyuan Biochem. Co. Ltd. (Jiangsu, China). Chitosan hydrochloride salt (CS) with an average molecular weight of 110 kDa was purchased from Zhejiang Aoxing Biochem Co. Ltd. Tripolyphosphate (TPP) was obtained from SigmaeAldrich (St. Louis, MO, USA). RMPI 1640 medium and fetal bovine serum (FBS) were purchased from Gibco BRL (Grand Island, NY, USA). Penicillin and streptomycin were provided by SigmaeAldrich (St. Louis, MO, USA). All other chemicals and solvents were of the highest grade commercially available. The water used was of ultrapure grade and was supplied by a Milli-Q purification system (Millipore Co., Billerica, MA, USA). 2.2. Preparation of blank HA-CS NPs (BNPs), DOX-loaded HA-CS NPs (DNPs), miR34a-loaded HA-CS NPs (MNPs) and DOX-miR-34a co-loaded HA-CS NPs (CNPs) Given amounts of HA and CS were dissolved in nuclease-free water to produce the solutions with the concentrations of 1.25 mg/ml and 0.6125 mg/ml, respectively. The prepared solutions were filtered through a 0.22-mm syringe filter (Millipore, USA). Blank NPs were prepared by slowly mixing these two solutions in several different mass ratios by the ionotropic gelation techniques [27]. The mixtures were prepared with HA:CS mass ratios of 1:2, 1:1 and 2:1. Briefly, 50 ml of cross-linker (TPP) solution (0.5 mg/ml) were first mixed with 1 ml HA solution and the resulting HA-TPP solution was then gently mixed with 1, 2 and 4 ml of the prepared CS solution, respectively. After constant magnetic stirring for 10 min at room temperature, the mixtures were then incubated at room temperature for additional 20 min for the formation and stabilization of NPs. The NPs were further isolated by repeated low-speed centrifugation (3000 g, 10 min) in 20 ml of glycerol bed in eppendorf tubes. The supernatants were collected and the NPs were then dispersed in RNasefree water by shaking on a vortex mixer for 10 min. For the encapsulation of charged “drugs” in the HA-CS NPs, a primary ionic interaction between HA and CS occurs and subordinative electrostatic forces among charged “drugs”, anionic HA and cationic CS were the driving forces to obtain “drugs” encapsulated NPs [27,30]. To prepare DOX-loaded HA-CS nanoparticles (DNPs), 0.125 mg DOX was first added to the CS solution with the weight ratio of DOX to HA of 1:10, which was further slowly mixed with 1.0 ml of prepared TPP-HA solution. For the preparation of miR-34a-loaded nanoparticles (MNPs), 20 mg of miR34a mimics was dissolved in 100 ml nuclease-free water and added to 1.0 ml of TPPHA solution. The resulting mixture was then slowly mixed with the CS solution to produce MNPs. For the preparation of DOX-miR-34a co-loaded nanoparticles (CNPs), 0.125 mg DOX was added to the CS solution with the weight ratio of DOX to CS of 1:10 and 20 mg of miR-34a mimics were added to the 1.0 ml TPP-HA solution, both prepared solutions were mixed under constant magnetic stirring to form co-loaded NPs. MNPs, DNPs and CNPs were prepared in three HA:CS mass ratios of 1:2, 1:1 and 2:1, respectively and isolated by the methods as described above. The miR-34a sequence used in this study was: Sense-UUCUCCGAACGUGUCACGUdTdT. Antisense-ACGUGACACGUUCGGAGAAdTdT. 2.3. Physicochemical characterization of NPs in different formulations and their encapsulation efficiencies The average size and zeta potential of the BNPs, DNPs, MNPs and CNPs were determined by dynamic light scattering (DLS) using a ZetaSizer Nano series Nano-ZS (Malvern Instruments Ltd, Malvern, UK). Determinations were performed at 633 nm with a constant angle of 90 after the samples were appropriately diluted in distilled water. Each batch was analyzed in triplicate. Morphological examination of the BNPs, CNPs was performed by transmission electron microscopy (TEM) (EM-200CX, JEOL Ltd., Tokyo, Japan). The samples were stained with 1% (w/v) phosphotungstic acid for 10 s, immobilized on copper grids and dried overnight for viewing by TEM. To evaluate the encapsulation efficiency of DOX in DNPs, un-encapsulated DOX were recovered from the supernatant described in Section 2.2. The encapsulation efficiency was calculated as the follows: Encapsulation efficiency (%) ¼ (A  B)/ A  100%. A and B are the weights of total amount of added DOX and the DOX fraction recovered from the supernatant after the centrifugation, respectively. The amount of un-encapsulated DOX in supernatant was determined using an ultraviolet-visible (UVeVis) spectrofluorometer (lex 480 nm and lem 590 nm; Perkin Elmer, Fremont, CA, USA). The supernatant recovered from BNPs without DOX was used for baseline correction. The encapsulation efficiency of FAM-labeled miR-34a in MNPs was measured in the similar way than DOX using spectrofluorometer (lex 488 nm and lem 520 nm, which corresponds to the fluorescence measurement of FAM) by comparing the absorbance values of un-encapsulated FAM-miR-34a in the solution to that of the original solution of miR-34a. The coloaded amount of DOX and FAM-labeled miR-34a in the CNPs were quantitatively analyzed using the same methods as described above. 2.4. Gel electrophoresis and serum stability assay CNPs were analyzed by 2% agarose gel electrophoresis. The gels were prepared with 2% agarose in tris-acetate-ethylenediaminetetraacetic acid (EDTA) buffer containing 0.5 mg/ml GelREDÔ (Biotium, USA). For electrophoretic mobility shift

4335

analysis, samples were incubated at room temperature for 15 min, 10% glycerine and 5 ml 2% SDS were then added to sample, respectively. Gel electrophoresis was carried out at 110 V for 10 min and the gel was subsequently photographed using Alpha Innotech gel imager system. Serum degradation assays were performed as fellows: Samples of naked miR34a in aqueous solution and the CNPs were respectively mixed with fresh serum (Gibco, USA) in a 1:1 ratio to give 50% serum concentration. The mixtures were then incubated at 37  C for the indicated times. Aliquots (15 ml) from each mixture samples were picked up and mixed with 5 ml 2% SDS and 2.5 ml 10% glycerine. The mixtures were further loaded onto a 2% agarose gel containing 0.5 mg/ml GelREDÔ (Biotium, USA). Gel electrophoresis was carried out at 110 V for 10 min and the gel was subsequently photographed using Alpha Innotech gel imager system. 2.5. The stability of CNPs and the controlled release of DOX and miR-34a from CNPs in vitro CNPs were suspended in deionized water, phosphate-buffered saline (PBS) of pH 7.4 and RMPI 1640 with 10% FBS, respectively, under mild stirring at 37  C. The CNPs sizes were determined at 0, 1 and 2 days to examine the stability of CNPs at different liquid environments by dynamic light scattering (DLS). To investigate the release kinetics of DOX and FAM-labeled miR-34a from CNPs, the CNPs were suspended in 2 ml of deionized water and sealed in two separate dialysis tubes with molecular weight cutoff of 1000 Da for DOX and 50,000 Da for miR-34a, respectively. The dialysis tubes were then immersed in 10 ml of PBS buffer dialysis solution at 37  C under moderate shaking using a water-bathing constant temperature vibrator. In addition, DOX release from DNPs and FAMmiR-34a release from MNPs were measured in the same manner for comparison. The released DOX or miRNA in the dialysis buffer were monitored at indicated time intervals over 10 days. The concentration of released DOX or FAM-miR-34a was determined by measuring the fluorescent density of DOX (lex 480 nm and lem 590 nm) and FAM (lex 488 nm and lem 520 nm) using a hybrid multi-mode microplate reader (Tecan, Durham, USA). 2.6. Cell culture and cellular uptake Human breast cancer cells MDA-MB-231 were cultured in RPMI-1640 media supplemented with 10% FBS (Gibco-BRL) and incubated at 37  C in a humidified atmosphere with 5% CO2. Cell uptake of CNPs carrying FAM-labeled miR-34a (green fluorescence) and DOX (red fluorescence) was evaluated by using confocal laser scanning microscopy (CLSM, Carl Zeiss, Boston, MA, USA) and flow cytometry (AttuneÒ Acoustic Focusing Cytometer, Life Technologies, USA). For confocal microscope analysis, MDA-MB-231 cells were seeded onto a borosilicate chambered cover glass (Nunc, USA) at a density of 1  105 cells per well at 37  C and cultured for 12 h before transfection. The cells were then incubated with BNPs and CNPs (100 nM FAM-miR-34a) in 1 ml serum-free culture media for 4 h. Subsequently, the cells were washed twice with Dulbecco’s phosphate buffered saline (DPBS) to remove the remaining NPs and fixed with 4% formaldehyde. Cell nuclei were stained with bluefluorescent DAPI (40 ,6-diamidino-2-phenylindole) (Beyotime, China) at room temperature for 15 min. Confocal fluorescence microscope images were acquired with a Nikon Ti-e microscope equipped with an UltraVIEW Vox confocal attachment (Perkin Elmer, Fremont, CA, USA). A quantitative determination of cellular uptake of CNPs by FACS analysis was performed as follows. Briefly, MDA-MB-231 cells were seeded in a 6-well plate (5  105 cells per well) and incubated with CNPs as described above. After removal of the free NPs, the cells were harvested and suspended in 1 ml PBS containing 1% PFA (paraformaldehyde) for fluorescence analysis by flow cytometry using a FACSAria Cell Sorter (Becton Dickinson BD). All data of the mean fluorescence signal were obtained from a population of 10,000 cells. Cells without NPs treatment and treated with FAM-miR-34a alone were used as negative control groups. 100 nM of FAMlabeled miR-34a were used for transfection. For competitive inhibition studies, HA was employed as inhibitor. The cells were first cultured in 1 ml of serum-free medium containing HA polymer (10 mg/ml) and then incubated with CNPs for 4 h. Cell uptake of CNPs was analyzed by using confocal microscopy and flow cytometry as described above. 100 nM of FAM-miR-34a were used for uptake analysis. 2.7. Cell cytotoxicity assay Human breast cancer cells MDA-MB-231 were seeded at a density of 2  104 cells per well in 96-well plates and incubated at 37  C with 5% CO2 for 12 h before evaluation. The culture media were removed and then replaced by 200 ml of RPMI 1640 supplemented with 10% fetal bovine serum containing various equivalent concentrations of free DOX, BNPs, MNPs, DNPs, and CNPs for 24 h or 48 h, respectively. 100 nM of miRNA were constantly used, while DOX was loaded with 3 different drug concentrations 0.1, 0.5 and 2.5 mM, which correspond to low, moderate and high doses of DOX. Control cells were cultured in the same conditions without nanoparticles treatment. Cell viability was determined at indicated time points using CCK-8 assay according to manufacturer’s instructions (Dojindo, Japan). The percentage of the viable cells was calculated using the following formula: Viability % ¼ A  B/C  B  100%, where A represents the absorbance of test, B represents the

4336

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

absorbance of blank (medium), C represents the absorbance of control (cells). Absorbance at 450 nm was detected with TECAN Infinite M200 microplate reader (Tecan, Durham, USA). To calculate 50% inhibitory concentration (IC50), the cells in 96-well plates were treated with free DOX and DOX-loaded nanoparticles (DNPs and CNPs) at various concentrations ranging from 0.01 to 10 mM. 100 nM of miR-34a were constantly used. IC50 was calculated based on the dose of DOX yielding 50% cell death compared to the untreated control. All experiments were conducted in triplicate. 2.8. Cell apoptosis AnnexinV/propidium iodide (PI) apoptosis detection kit (BD, USA) was used for apoptosis assessment. Cells treated with free DOX and NPs in different formulations containing the equivalent DOX concentration of 0.1 mM were assessed. 100 nM of miR34a in MNPs and CNPs were used. At 48 h post treatment, cells were harvested and processed by using AnnexinV/propidium iodide (PI) apoptosis detection kit in compliance with the manufacturer’s protocol and subjected to flow cytometric analysis (FACSCalibur, BD, USA). Each experiment was performed in triplicate. 2.9. Quantitative real-time PCR Quantitative Real-Time PCR (qRT-PCR) was applied to evaluate the expression level of miR-34a and gene Bcl-2. The concentration of miR-34a used was fixed at 100 nM. Total RNA was extracted from cultured cells or frozen tumor tissues using Trizol reagent (Invitrogen, USA) following the protocols of the manufacturer. For miRNA expression analysis, 1 mg of total RNA was applied for reverse transcription to synthesize cDNA of miRNA using TaqmanÒ MicroRNA Reverse Transcription Kit (Applied Biosystems). QRT-PCR analysis was performed using TaqmanÒ Universal Master Mix II, No AmpEraseÒ UNG (2) (Invitrogen, USA) and special RT-primer designed for miR-34a (TaqmanÒ assay ID 000426, Invitrogen, USA) according to the manufactuer’s instructions. 25 ml of the reverse transcription mix was amplified by PCR with the following program: 95  C for 10 min, 55  C for 2 min, 72  C for 2 min and 12 cycles at 95  C for 15 s and 60  C for 4 min. U6 was used as an endogenous control for normalization. Each sample was analyzed in triplicate. Primer sequences used were as follows: MiR-34a: 50 -AACAACCAGCTAAGACACTGCCA-30 U6: 50 -GCTTCGGCAGCACATATACTAAAAT-30 For gene expression analysis, 1 mg of total RNA was reverse transcribed in a final volume of 20 ml to synthesize first-strand cDNA using ImProm-IIÔ reverse transcription system (Promegra, USA) according to the instructions of manufacturer. Gene expression was detected by qRT-PCR using BrilliantÒ II SYBRÒ green qPCR master mix (Stratagene, USA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an endogenous control to normalize expression. 25 ml PCR reactions in triplicate were carried out by an initial denaturation at 95  C for 10 min followed by 40 cycles, each consisting of 30 s at 95  C, 30 s at 58  C, 30 s at 72  C and then 1 cycle for melting curve consists 1 min at 95  C, 30 s at 55  C, 30 s at 95  C. Bcl2 primer sequences used were as follows: Forward: 50 -CAT GCTGGG GCC GTA CAG-30 Reverse: 50 -GAA CCG GCA CCTGCA CAC-30 2.10. Western blot assay Cultured cells were harvested and lyzed with RIPA lysis buffer by incubation on ice for 10 min. After centrifugation at 12,000 g for 10 min, the supernatants were collected and the concentrations of proteins were measured using Bradford’s reagent (Bio-Rad laboratories, USA). The protein samples were denatured by boiling for 10 min and loaded onto SDS-PAGE gel for electrophoresis. The proteins were then incubated in the blocking solution (5% non-fat dried milk) at room temperature for 1 h after transferred onto PVDF membranes (Millipore, USA). The anti-Bcl-2 and anti-Notch-1 antibody (Abcam, USA) was added into blocking solution and incubated at 4  C overnight. The membranes were subsequently incubated with the secondary goat anti-mouse antibody conjugated with fluorescent dyes: IRDye 800 CW (KPL, USA) for 45 min. Protein expression was normalized against GAPDH expression (RD, USA). Blotting images were acquired with the Odyssey infrared imaging system (Li-COR Biosciences, USA) and analyzed by the software provided by the manufacturer. 2.11. In vitro transwell migration assay MDA-MB-231 cells were seeded at a density of 6  105 cells per well in 6-well plates and incubated for 12 h before experiments. The cultured cells were incubated with free DOX, BNPs, DNPs, MNPs and CNPs (different formulations containing 0.1 mM of DOX or 100 nM miR-34a), respectively, in 1 ml of serum-free culture media for 4 h. All treated cells were then washed twice with PBS to remove the remaining NPs and cultured in RPMI-1640 medium with 10% FBS for additional 24 h. The migration assay was conducted using transwell insert chambers (Corning, USA). 1  105 NPs-transfected cells were harvested and plated into the upper chamber in serum-free medium. Medium containing 20% FBS in the lower chamber served as chemoattractant. After incubation for 24 h at 37  C in a 5% CO2 humidified incubator, cells in the upper chambers were removed by wiping with a cotton swab and cells migrated to the lower surface of filter were fixed in 70% ethanol for 30 min and

stained with 0.2% crystal violet for 10 min. Cell migration was scored by counting five random fields per filter under a light microscope. Cell migration rate was calculated by formula as follows: Relative rate of migration (%) ¼ migrating cells with treatment/migrating cells without treatment  100. The transwell assay was carried out in triplicate and repeated three times. 2.12. In vivo anti-tumor experiments Female athymic nude Balb/c mice were purchased from Vital River Laboratory Animal Technology Co., Ltd (Beijing, China). All of the animals were kept at pathogen-free condition. All animal studies were conducted in accordance with the experimental protocols involving animal study approved by the Institutional Animal Care and Use Committee of Peking University. To establish the tumor model, mice were subcutaneously injected in the right axilla with 5  106 MDA-MB-231 cells suspended in 100 ml of PBS. The mice were randomly assigned to five groups (n ¼ 5 for each group) and injected through the tail vein with PBS saline, DOX, DNPs, MNPs and CNPs (DOX of 5 mg/kg and miR-34a of 2 mg/kg), respectively, while the tumor volumes reached around 0.1 cm3 at the time point of 14 days post inoculation. Mice were administrated through the tail vein every 3 days for 5 times. Animal weight and tumor volume were measured and calculated once every two days for 20 days. The animals were then sacrificed and individual tumor volume (V) was calculated using the formula: V ¼ (L  W2)/2, wherein length (L) is the longest diameter and width (W) is the shortest diameter perpendicular to length. 2.13. Statistical analysis The mean  SD was determined for each treatment group. Comparison of each group was evaluated by one-way analysis of variance (ANOVA). The differences were considered significant for *p < 0.05.

3. Results and discussion 3.1. Preparation and characterization of NPs in different formulations MiRNAs are identified as potent post-transcriptional regulators in gene expression. It has been shown that miRNAs play a critical role in cancer pathogenesis and some of them may function as tumor suppressors [31]. Unlike the better known siRNA-mediated mRNA silencing designated to specifically knock down a single gene, a miRNA may simultaneously targets multiple different genes. Thus, miRNA-based therapy has a great potential to be a more powerful tool in tumor treatment by simultaneously modulating multiple genes which are involved in distinct tumor-related signaling networks [32]. MiR-34a is one of the most defined tumor suppressor miRNA in a variety of tumors, including breast cancer. Over-expression of miR-34a in breast cancer cells induces cell apoptosis and inhibits cell migration and tumor growth in vivo [33]. Doxorubicin is the most common chemotherapeutic agent used for breast cancer treatment, but its clinical application is limited due to severe side effects and drug resistance. In the present studies, doxorubicin and miR-34a were co-encapsulated into HA-CS NPs to investigate their synergistic effects in breast cancer suppression. Scheme 1 depicts the preparation of DOX and miR-34a co-encapsulated HA-CS NPs (CNPs) and the potential to realize an enhanced anti-cancer chemotherapeutic efficacy after intracellular internalization of CNPs. The mechanism of formation of HA-CS NPs via ionotropic gelation technique and incorporation of charged molecules in the particles have been illustrated previously [27]. In our preparation, DOX and miR-34a were co-encapsulated into HA-CS NPs through a facile self-assembly approach, in which electrostatic interaction of positively charged CS with oppositely charged HA and TPP is the primary driving force for nanostructure formation by a self-assemble process. In addition, the interaction between HA and CS through hydrogen bonds and other intermolecular forces as well as the electrostatic interaction between positively charged DOX and negatively charged miR-34a can also contribute to the self-assembly of nanocomplexes and drug encapsulation [34].

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

DLS analysis was further performed to evaluate the influence of individual encapsulation or co-encapsulation of DOX and miR-34a on the average hydrodynamic diameter and surface charge of NPs composed of different weight ratios of HA and CS. As shown in Table 1, the size and zeta potential of the NPs were dependent on not only the composition of NPs, but also the ionic pharmaceutical agents encapsulated. The obtained results illustrated that the increase in HA weight ratio is accompanied by an increase of particle size and a decrease of zeta potential. Shift of zeta potential values was observed when the weight ratio of HA to CS increased to 2:1, indicating that NP surface was now covered by excess of HA bearing a negative charge. The average size of MNPs is similar to that of BNPs, indicating that the influence of the incorporation of miR-34a on the particle size is limited. In contrast, significant increase in size was observed in DNPs and CNPs, suggesting that such an increase was correlative to the incorporation of DOX in the formulations. Encapsulation efficiency of “pharmaceutical agents” in NPs is an important factor to be considered to evaluate the therapeutic efficacy of nanocarriers by in vitro and in vivo models. Therefore, the loading efficiency of NPs with different compositions was subsequently investigated. As shown in Table 1, the highest encapsulation efficiency of DOX (44.2%) was observed in the NPs consisting of the weight ratio of HA to CS of 2:1, indicating that encapsulating efficiency of positively charged DOX correlated with the increase in the amount of negatively charged HA in NPs. Relative high miR-34a encapsulation efficiency (more than 80%) was observed in all of the tested experiments, indicating that miRNA can be encapsulated into the HA-CS-based NPs with high encapsulation efficiency. However, the increase in miR-34a encapsulation efficiency is also dependent on the increasing amount of the CS in the NPs, suggesting a high affinity of the positively charged CS for the nucleotides. It is worth mentioning that the encapsulation efficiency of DOX and miR-34a in CNPs were both higher than that in DNPs and MNPs, this phenomenon may be contributed by an additional interaction of positively charged DOX with negatively charged miR-34a. The NPs with the weight ratio of HA to CS of 2:1 bear negative zeta potential and higher disposition of the HA on the surface of NPs may achieve not only enhanced stability of particles in serum due to the negligible nonspecific interaction with serum components, but also HA-directed tumor-specific targeting ability [35,36]. Additionally, this ratio produced the highest DOX and miR-34a co-

Table 1 Characterization of the various NPs. Formation HA:CS Particles Zeta potential Encapsulation Encapsulation group weight size (nm)a (mV)a of DOX (%)b of miR-34a (%)b ratio BNPs DNPs MNPs CNPs

1:2

171 192 173 193

   

15 7 8 9

31 32 30 32

   

3 4 2 3

e 27.6  1.8 e 30.3  2.1

e e 94% 98%

BNPs DNPs MNPs CNPs

1:1

165 182 167 185

   

10 3 2 1

19 22 21 22

   

2 3 2 2

e 32.3  4 e 36.2  3

e e 90% 93%

BNPs DNPs MNPs CNPs

2:1

189 212 190 214

   

8 4 6 3

33 30 33 33

   

2 3 2 2

e 44.2  2 e 48.3  3

e e 81% 91%

a Determined by dynamic light scattering (DLS). The NPs were prepared by directly dissolving in distilled water at a concentration of 0.5 mg/ml, followed by 5 min sonication. b Effects of formulation parameters on encapsulating efficiencies of DOX and miR34a. Results means  SD (n ¼ 3).

4337

encapsulation efficiency. Thus, the NPs with this ratio were selected for subsequent studies. The morphology of BNPs and CNPs was observed using TEM (Fig. 1A and B). The dried NPs retained their spherical shapes with good dispersion in aqueous solution. All NPs had a narrow particle size distribution with the average size of (150e180). The DLS (Fig. 1C and D) reflects the hydrodynamic diameter of NPs swelling in aqueous solution, whereas TEM reflects the diameter of dried NPs. Therefore, the diameter determined by TEM was little smaller than that measured by DLS. Compared to BNPs, co-encapsulation of DOX and miR-34a in CNPs did not have any significant influence on the morphology of particles. The encapsulating miR-34a in CNPs was also reflected by the results from gel electrophoretic assay (Fig. 1 E). Compared to naked miR-34a (Lane 4), the migration of miR-34a into the gel was nearly completely prevented by the CNPs-encapsulation (Lane 1) and reappeared after treatment of CNPs with SDS prior to electrophoresis (Lane 2), indicating that miR-34a was indeed encapsulated in HA-CS NPs via electrostatic interaction and can be released from CNPs after treatment with SDS. 3.2. Serum stability of miR-34a in CNPs, the stability of CNPs in various solvents, and in vitro release of DOX and miR-34a from CNPs We first examined the stability of nanoparticle-encapsulated miR-34a in serum. The obtained data showed that the miR-34a encapsulated in CNPs were protected from degradation for up to 24 h compared to the naked miR-34a (Fig. 1F), implying that HA-CS NPs-encapsulated miR-34a were stable and miR-34a were protected from degradation in serum and suitable for in vivo application. The stability of NPs is closely dependent on the surface property of nanoparticles. The stability of CNPs was further determined under conditions similar to those used for routine cell culture. CNPs were re-suspended in various solvents, including deionized water, PBS of pH 7.4 and RMPI 1640 with 10% FBS. As shown in Fig. 1G, no aggregation of nanoparticles was observed in all tested solutions and all of the NPs showed similar sizes in a range of 200e240 nm for up to 3 days. Although a slight increase in the diameter of CNPs was observed when suspended in cell culture media. The size of CNPs can be consistently maintained at 230 nm for three days, indicating the DOX and miR-34a co-loading CNPs are stable and suitable for further analysis and even clinical application. The in vitro release profile of DOX and miR-34a from CNPs was investigated at 37  C and in PBS of pH 5.5 and pH 7.4, respectively. As shown in Fig. 2, approximately 38% and 28% of DOX can be released from NPs within the first 24 h at pH 5.5 and pH 7.4, respectively, which were followed by continuous slow release. In contrast, the release of miR-34a from CNPs was slower than DOX, only around 20% and 16% of miRNA can be released within the first 24 h at pH 5.5 and pH 7.4, respectively. However, both DOX and miR-34a released from the CNPs exhibited a sustained release behavior. The faster release of DOX may be attributed by its small molecular weight, which allows better diffusion from the CNPs, while the high net negative charge and high molecular weight of miR-34a mimics may potentially result in a strong electrostatic interaction of miR-34a mimics with CS, leading to a slower release of miR-34a than DOX. Additionally, the calculated results also revealed that the release of both DOX and miR-34a at pH 5.5 was faster than at pH 7.4, suggesting that acidic condition increases the release efficiency of HA-CS-based nanoparticles compared with neutral condition, which are consistent with previous reports [27,28]. Acid condition was preferentially found in tumor environment and lysosomal environment. Thus, enhanced DOX and miR34a release in pH 5.5 would be helpful for tumor therapy.

4338

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

Fig. 1. TEM images, DLS histograms, electrophoretic mobility and stability of nanoparticles. TEM images of nanoparticles. A: BNPs; B: CNPs. Scale bar: 200 nm; DLS histograms of nanoparticles. C: BNPs; D: CNPs; E: Electrophoretic mobility shift assay was performed to determine the amount of miR-34a encapsulated in CNPs; F: The stability of free and nanoparticle-encapsulated miR-34a was monitored under the presence of serum at 37  C. G: Post-formulation stability of nanoparticles in different solvents as judged by their average hydrodynamic diameters at 37  C.

3.3. HA mediated targeted co-delivery of DOX and miR-34a in breast cancer cells Many tumor cells over-express HA-binding receptors such as CD44, and consequently, HA disposition on the surface of NPs may result in active targeting ability [37]. To monitor HA-CS NPs-mediated intracellular co-delivery of DOX and miR-34a, miR-34a mimics was labeled with FAM fluorescent dye in order to visualize the cell uptake of miR-34a compared with red fluorescence of DOX. BNPs were used as negative control. As shown in Fig. 3A, after 4 h of

Fig. 2. In vitro DOX and miR-34a release profiles from CNPs in PBS at pH 5.5 and pH 7.4 at 37  C, respectively.

incubation with CNPs, significant fluorescent signals of DOX (red (in the web version)) and FAM-labeled miR-34a (green (in the web version)) were present in the perinuclear regions of the cytoplasm. Overlapping of red and green fluorescence of cells treated with DOX and FAM-miR-34a NPs generated yellow (in the web version) stains in the merged image. In contrast, no fluorescent signals were observed in cytoplasm when incubated with BNPs. The nucleus was stained in blue (in the web version) with DAPI in all of the cells. The data suggest that DOX and miR-34a can be effectively co-delivered into breast cancer cells by HA-CS based nanoparticles. It has been broadly reported that the coating of HA on the nanoparticles surface would facilitate the cell uptake of particles via interacting with CD44 receptor through endocytic pathway. We further evaluate the mechanism of cellular uptake of CNPs. The cells were pre-treated by adding excess of HA in culture medium for 3 h before incubation with CNPs. The intensity of intracellular fluorescence of DOX and FAM-labeled miR-34a was remarkably weakened, indicating that the internalization of CNPs into cells was reduced by the pretreatment of free HA. The data suggest that selective cell targeting and cell uptake of HA-CS NPs are mediated by HA-dependent pathway. The delivery efficiency was further evaluated by flow cytometry based on the labeled fluorescent signals of FAM-miR-34a.100 nM FAMmiR-34a in NPs was used. BNPs treated cells were used as negative control. Data in Fig. 3B showed that CNPs are capable of delivering miR-34a into breast cancer cells with high efficiency, up to 80% of cells were positive for fluorescence. In contrast, no significant cell uptake of miR-34a was obtained under its native form, indicating that naked miRNA mimics are unfavorable for efficient cell uptake. The percentage of fluorescence-positive cells was remarkably diminished after HA pretreatment, also indicating that HA pretreatment could significantly reduce the intracellular uptake of miR-34a by CNPs. The data obtained above are correlated to the mechanism of HAcoated nanocarriers in drug or gene delivery and demonstrated that

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

4339

Fig. 3. Intracellular uptake of nanoparticles was monitored by confocal microscope and flow cytometry. A: Laser confocal microscopic images of MDA-MB-231 cells incubated with BNPs, CNPs, and CNPs with the pretreatment of 10 mg/ml of HA. B: MiR-34a was labeled by FAM. FAM-positive cells were quantitatively determined by flow cytometry. Mean data are in the inset shown with standard error bars, n ¼ 3; (Dose: 100 nM FAM-miR-34a).

HA-CS-based NPs can efficiently co-deliver DOX and miR-34a to MDAMB-231 cells via HA receptor mediated endocytosis pathway. The HACS NPs-mediated intracellular restoration of tumor suppressor miR34a was subsequently evaluated by quantitative real-time PCR (polymerase chain reaction). Untreated cells were used as negative control. As shown in Fig. 4A, over 17,000-fold increases in miR-34a were observed in cells treated with MNPs and CNPs compared with untreated cells as well as the cells treated with BNPs and DNPs. In addition, real-time PCR evaluation showed that HA pretreatment could significantly reduce intracellular restoration of miR-34a delivered by MNPs and CNPs, which are consistent with the results obtained above using confocal microscopy and flow cytometry. 3.4. Suppression of anti-apoptosis Bcl-2 gene expression by miR34a restoration Doxorubicin is a commonly used chemotherapeutic agent that promotes DNA damage and induces apoptosis of cancer cells.

Resistance to this agent is a major cause of cancer relapse and consequent chemotherapeutic failure for cancer patients. The resistance to DOX can be mediated by various mechanisms. In addition to enhanced drug efflux from cells mediated by up-regulated expression of P-glycoprotein, over-expression of anti-apoptotic genes in cancer cells has been considered as nonpump resistance mechanism. Bcl-2 has been identified as a potent regulator of apoptosis. This proto-oncogene inhibits cell apoptosis by preventing the permeability of mitochondrial. Bcl-2 has been found to be frequently overexpressed in various cancers and its anti-apoptotic role is tightly associated with its expression levels [38,39]. The over-expression of Bcl-2 renders cancer cells resistance to DOX-induced apoptosis. Previous studies showed that siRNA targeting on Bcl-2 and antisense Bcl-2 oligonucleotide increased the susceptibility of cancer cells to DOX-induced apoptosis and co-delivery of DOX and Bcl-2 siRNA by nanoparticles enhances chemotherapy efficacy [40,41]. Bcl-2 has been previously identified as a promising downstream target of miR-34a (Fig. 4B) [11]. In the present studies, the

4340

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

Fig. 4. Intracellular delivery of miR-34a down-regulates Bcl-2 expression. A: Intracellular uptake of miR-34a was detected by real-time PCR. B: Putative miR-34a target site within 30 -UTR of Bcl-2 mRNA. C: The expression of Bcl-2 was quantitatively analyzed by real-time PCR after treatment with different components. Mean values were shown with standard error bars (n ¼ 3). D: Bcl-2 expression was evaluated by Western blot.

expression of Bcl-2 was evaluated at both transcription and protein levels after delivering miR-34a in breast cancer cells by HA-CS NPs. Untreated cells were used as negative control. The expression level of Bcl-2 in BNPs- and DNPs-treated cells was similar to that of control cells, suggesting that HA-CS NPs and DOX had no regulatory effects on the expression of Bcl-2. Remarkable down-regulation of Bcl-2 expression was observed in the cells treated by MNPs and CNPs, which were both capable of efficiently delivering miR-34a into cells. Modest down-regulation of Bcl-2 expression was observed in the cells received pretreatment with HA, which may partially suppress MNPs and CNPs-mediated intracellular delivery of miR-34a, indicating that the expression of Bcl-2 is controlled by miR-34a and NPs-mediated intracellular delivery of miR-34a may effectively suppress Bcl-2 expression (Fig. 4C and D). The data are consistent with previously reports [18]. The expression of Bcl-2 was equally evaluated by western-blot upon treatment of the cells with HA-CS NPs in different formulations, which resulted in the same pattern as observed by real-time analysis. Obvious lower expression of Bcl-2 was observed in cells treated with MNPs and CNPs than that of untreated cells as well as cells treated with BNPs and DNPs, confirming that HA-CS NPs-mediated intracellular delivery of miR-34a could efficiently down-regulate the expression of Bcl-2 at both transcription and protein level. 3.5. In vitro cytotoxic efficacy and apoptosis assay Previous studies reported that miRNAs have synergistic antitumor effects with conventional chemotherapy [33,42]. To

investigate whether the HA-CS NPs-mediated intracellular codelivery of miR-34a and DOX can effectively increase the chemotherapeutic efficacy of DOX by efficiently silencing Bcl-2 expression, the in vitro cytotoxicity of NPs in different formulations were evaluated by CCK8 assay. Fig. 5 showed the in vitro viability of MDA-MB231 cells after 24 h and 48 h incubation with BNPs, free DOX, DNPs, MNPs and CNPs at the DOX concentration of 0.1 mM (A), 0.5 mM (B) and 2.5 mM (C). No obvious cell death was noticed in BNP-treated cells, indicating a low cytotoxicity of HA-CS NPs. The cells treated with MNPs showed reduced cell viability, suggesting that intracellular restoration of miR-34a can trigger the apoptosis of breast cancer cells, which are consistent with the tumor suppressor functions of miR-34a. Free DOX, DNPs and CNPs exhibited remarkable cytotoxicity to MDA-MB-231 cells and showed a reverse correlation with respect to the applied doses of DOX at tested time points in all experiments. DNPs displayed slightly higher cytotoxicity than free DOX, although HA in DNPs can mediate the higher accumulation of DOX inside the cell by targeting ability, the sustained release of DOX in DNPs may exhibit delayed cytotoxicity, thus these two factors resulted in comparable cytotoxicity of free DOX and DNPs, rather than obvious difference [43]. However, significant decreases in cell viability were observed in the cell samples treated with CNPs compared to those treated with DNPs or free DOX at identical dose level, suggesting that CNPs has an increased cytotoxic activity to breast cancer cells than DNPs and free DOX at the same dose level and the increase in cytotoxic activity of CNPs is mediated by the additional co-incorporation of miR-34a in NPs with DOX. In our experiments, the combination treatment exhibited greater

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

4341

Fig. 5. In vitro cell viability of MDA-MB-231 cells treated with different components and IC 50 measurement. Cell viability was quantitatively determined using CCK-8 Kit upon incubated with different concentrations of DOX for the indicated times. A: 0.1 mM; B: 0.5 mM; C: 2.5 mM. D: Cell viability was quantitatively measured using CCK-8 Kit upon incubation with different concentrations of free DOX and nanocomplexes. E: IC 50 of free DOX and nanocomplexes were calculated. Data were shown as the mean of triplicate measurements with standard error bars (n ¼ 3).

inhibition on cell proliferation than the individual treatments using either miR-34a or DOX. Based on these measurements, the IC50 doses (dose that kills 50% of cells) of the different formulations were shown in (Fig. 5D and E), and we can rank the killing efficiency as follows: CNPs > DNPs > free DOX. In addition, the IC50 dose of free DOX was determined to be 0.55 mM and that of CNPs (100 nM miR-

34a) was calculated to be 0.12 mM, which was approximately 4.58 times lower than the IC50 of free DOX. We thereafter investigated whether synergistic effects of miR34a and DOX in cytotoxicity were mediated by increased apoptosis induced by CNPs-mediated intracellular delivery of miR34a. Cell apoptosis was subsequently monitored by flow cytometry

Fig. 6. Cell apoptotic assay. Cell apoptosis was analyzed by flow cytometry using annexin V-FITC kit and PI staining. A: Normally cultured cells used as control; B: BNPs-treated cells; C: free DOX-treated cells; D: MNPs-treated cells; E: DNPs-treated cells; F: CNPs-treated cells.

4342

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

(Fig. 6). Cells maintained in standard culture medium were used as negative control. Flow cytometry analysis demonstrated that few apoptotic cells were detected in control cell samples and no increase in apoptotic cells was observed in the cells treated with empty HA-CS NPs, also indicating that the applied NPs are lowcytotoxic and not able to induce cell apoptosis. Significant increase in cell apoptosis was found in cells treated by MNPs compared to that of control cells and BNPs-treated cells, suggesting that miR-34a is able to trigger cell apoptosis upon HA-CS NPsmediated intracellular delivery. Furthermore, we found that free DOX and DNPS were both capable of inducing cell apoptosis, which are consistent with previous reports that DOX kills cancer cells through triggering cell apoptosis [44]. However, the highest percentage of cell apoptosis was obtained in cells treated with CNPs, suggesting that miR-34a and DOX have synergistic effects on cell apoptosis and the co-delivery of miR-34 with DOX enhances DOXmediated cell apoptosis, which in turn increases chemotherapeutic efficacy of DOX. 3.6. HA-CS NPs-mediated miR-34a delivery inhibits cancer cell migration In breast cancer, despite impressive advances in screening and adjuvant treatment, metastasis-related recurrence is still common and responsible for the majority of breast cancer-associated mortality. Inhibition of migration-related signaling has been thus emerging as an effective strategy for cancer treatment by

preventing tumor metastasis. We were therefore interested in investigating the regulatory effects of miR-34a and DOX coencapsulated CNPs on cells migration. Untreated cells were used as control. The obtained results in Fig. 7AeF showed that, compared to control, no obvious inhibition in cell migration was observed in cells treated with BNPs, free DOX and DNPs, respectively, suggesting that BNPs, free DOX and DNPs had no remarkable regulatory effects on cell migration. In contrast, significant inhibition in the migration of breast cancer cells was observed in cells treated by MNPs and CNPs, which are both NPs with the incorporation of miR-34a, indicating that the inhibition of cell migration is mediated by intracellular delivery of miR-34a. Recent studies pointed out the critical roles of Notch signaling in human breast cancer invasiveness and metastasis [45], The expression of Notch-1 was then examined in the cells treated with NPs in different formulations by Western-blot. Obvious decrease in the expression of Notch-1 was especially observed in the miR-34adelivered groups (Fig. 7H). Furthermore, a slight reduction in the migrated cell numbers of was observed in cells treated with CNPs, which may be potentially contributed by not only the intracellular delivery miR-34a that down-regulated the expression of Notch-1, but also the increased killing effects of the co-delivery of DOX and miR-34a. The results suggest that, compared to DNPs, CNPs display a dual anti-tumor activity by simultaneously inducing tumor cell death through apoptosis-related pathway and inhibiting tumor cell migration which may potentially prevent metastasisrelated death in clinic.

Fig. 7. MiR-34a inhibits the migration of breast cancer cell through targeting Notch-1 signaling. Photomicrographs of transwell filters show that the cells across an 8-mm pore size membrane through migration. A: normal control cells; B: BNPs-treated cells; C: Free DOX-treated cells; D: DNPs-treated cells; E: MNPs-treated cells; F: CNPs-treated cells. Scale bar: 50 mm; G: Quantitative determination of cell migration was performed in triplicate. Results are mean  SD, *p < 0.05. H: Notch-1 protein expression was evaluated by western blot analysis.

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

3.7. In vivo anti-tumor efficacy Subcutaneous tumor model of breast cancer was further used to evaluate the anti-tumor efficacy of nanocarriers (MNPs, DNPs and CNPs) or free DOX in vivo. The anti-tumor efficacy of CNPs was superior to that of all other treatments (Fig. 8A, C and D), suggesting that CNPs had a remarkable advanced activity over MNPs, DNPs and free DOX in suppressing MDA-MB-231 solid tumors. No body weight loss was observed in all of treated mice (Fig. 8B), indicating that no significant toxicity was observed in all bound or unbound drugs within experimental periods. The tumor volumes were evaluated at time of sacrifice. Interestingly, the growth of breast tumors was delayed in MNPs-treated mice, indicating that HA-CS NPs-mediated miR-34a delivery has anti-tumor effects in vivo. Effective intratumoral delivery of miR-34a was especially detected in MNPs and CNPs-treated mice (Fig. 8E), which subsequently led to significant suppression of Bcl-2 expression in xenograft tumors (Fig. 8F). H&E staining was used to examine the histological feature of tumor induced by different components. As shown in Fig. 8G, tumor sections from saline-treated groups were densely cellular, whereas the sections from nanocarriers and free DOX-treated groups are less cellular compared to the controls, indicating that administration of “drug”-loaded nanocarriers and free DOX induced cell apoptosis in tumor tissues. However, the fewest tumor cells were observed in CNPs-treated groups. The

4343

obtained results indicated that miR-34a delivered by HA-CS NPs offers a great therapeutic potential against breast cancers in vivo and significantly improved therapeutic efficacy was observed in mice treated with the co-delivery of DOX and miR-34a. 4. Conclusions In this study, we reported a successful application of HA-CS as a dual nanocarrier system to simultaneously deliver chemotherapeutic drug DOX and tumor suppressive miRNA-34a into triple negative breast cancer cells for improved efficacy of chemotherapy. Our data showed that the co-delivery of miR-34a with DOX was not only able to substantially enhance anti-tumor activity of DOX by silencing Bcl-2 expression, but also suppress breast cancer cells migration via targeting Notch-1 signaling, suggesting that codelivery of DOX and miR-34a could achieve synergistic effects on tumor suppression. Furthermore, nanocarrier-mediated delivery of DOX and miR-34a can not only result in an effective reduction of drug resistance and side effects of DOX, but also enhance the therapeutic outcome of DOX. Owing to the advantages of the co-delivery approach for the simultaneous inhibition of tumor growth and migration, co-delivery of miRNAs and chemotherapeutic drugs by nano-system demonstrates a great potential as combined therapeutic strategy in anti-cancer treatment.

Fig. 8. Administration of nanocomplexes inhibited tumor growth and induced tumor cell apoptosis through silencing Bcl-2 expression in vivo. A: Tumor volumes were measured at indicated time-points and shown as average of duplicate measurements with standard error bars (n ¼ 5 for each group). B: The body weights of mice were measured at indicated time-points and shown as average of duplicate measurements with standard error bars (n ¼ 5 for each group). C: The weights of tumors were calculated at time of sacrifice and mean data of three experiments were shown with standard error bars. D: Representative tumor tissues were extracted from mice and photographed. E: Intratumoral delivery of miR-34a was examined by real-time PCR. The mean values of three independent experiments were shown with standard error bars. F: The expression of Bcl-2 was detected by Western-blot. G: Representative data of H&E staining of tumor tissues.

4344

X. Deng et al. / Biomaterials 35 (2014) 4333e4344

Acknowledgments This work was supported by grants from MOST 973 program (2010CB934004) and NSFC (81272453), from National High Technology Research and Development Program of China (863 Program, No. 2007AA021107), Program for New Century Excellent Talents in University (No. 39015001200801). References [1] Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24(14):2137e50. [2] Coley HM. Mechanisms and strategies to overcome chemotherapy resistance in metastatic breast cancer. Cancer Treat Rev 2008;34(4):378e90. [3] Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004;116(2):281e97. [4] Sun W, Julie Li Y-S, Huang H-D, Shyy JY, Chien S. microRNA: a master regulator of cellular processes for bioengineering systems. Annu Rev Biomed Eng 2010;12:1e27. [5] Fasanaro P, Greco S, Ivan M, Capogrossi MC, Martelli F. microRNA: emerging therapeutic targets in acute ischemic diseases. Pharmacol Ther 2010;125(1): 92e104. [6] Bader AG, Brown D, Winkler M. The promise of microRNA replacement therapy. Cancer Res 2010;70(18):7027e30. [7] Tong A, Nemunaitis J. Modulation of miRNA activity in human cancer: a new paradigm for cancer gene therapy? Cancer Gene Ther 2008;15(6):341e55. [8] Huntzinger E, Izaurralde E. Gene silencing by microRNAs: contributions of translational repression and mRNA decay. Nat Rev Genet 2011;12(2):99e110. [9] Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144(5):646e74. [10] Chang T-C, Wentzel EA, Kent OA, Ramachandran K, Mullendore M, Lee KH, et al. Transactivation of miR-34a by p53 broadly influences gene expression and promotes apoptosis. Mol Cell 2007;26(5):745e52. [11] Hermeking H. The miR-34 family in cancer and apoptosis. Cell Death Differ 2009;17(2):193e9. [12] Liu C, Kelnar K, Liu B, Chen X, Calhoun-Davis T, Li H, et al. The microRNA miR34a inhibits prostate cancer stem cells and metastasis by directly repressing CD44. Nat Med 2007;17(2):211e5. [13] Ji Q, Hao X, Zhang M, Tang W, Yang M, Li L, et al. MicroRNA miR-34 inhibits human pancreatic cancer tumor-initiating cells. PLoS One 2009;4(8):e6816. [14] Christensen M, Larsen LA, Kauppinen S, Schratt G. Recombinant adenoassociated virus-mediated microRNA delivery into the postnatal mouse brain reveals a role for miR-134 in dendritogenesis in vivo. Front Neural Circuits 2009;3(16). [15] Suh JS, Lee JY, Choi YS, Chong PC, Park YJ. Peptide-mediated intracellular delivery of miRNA-29b for osteogenic stem cell differentiation. Biomaterials 2013;34(17):4347e59. [16] Trang P, Wiggins JF, Daige CL, Cho C, Omotola M, Brown D, et al. Systemic delivery of tumor suppressor microRNA mimics using a neutral lipid emulsion inhibits lung tumors in mice. Mol Ther 2011;19(6):1116e22. [17] Cao M, Deng X, Su S, Zhang F, Xiao X, Hu Q, et al. Protamine sulfateenanodiamond hybrid nanoparticles as a vector for MiR-203 restoration in esophageal carcinoma cells. Nanoscale 2013;5(24):12120e5. [18] Hu QL, Jiang QY, Jin X, Shen J, Wang K, Li YB, et al. Cationic microRNAdelivering nanovectors with bifunctional peptides for efficient treatment of PANC-1 xenograft model. Biomaterials 2013;34(9):2265e76. [19] Cheng D, Cao N, Chen J, Yu X, Shuai X. Multifunctional nanocarrier mediated co-delivery of doxorubicin and siRNA for synergistic enhancement of glioma apoptosis in rat. Biomaterials 2012;33(4):1170e9. [20] Wang H, Zhao Y, Wu Y, Hu Y-l, Nan K, Nie GJ, et al. Enhanced anti-tumor efficacy by co-delivery of doxorubicin and paclitaxel with amphiphilic methoxy PEG-PLGA copolymer nanoparticles. Biomaterials 2011;32(32): 8281e90. [21] Hu C-MJ, Aryal S, Zhang L. Nanoparticle-assisted combination therapies for effective cancer treatment. Ther Deliv 2010;1(2):323e34.

[22] Creixell M. Peppas NA.Co-delivery of siRNA and therapeutic agents using nanocarriers to overcome cancer resistance. Nano Today 2012;7(4):367e79. [23] Luo Y, Bernshaw NJ, Lu Z-R, Kopecek J, Prestwich GD. Targeted delivery of doxorubicin by HPMA copolymer-hyaluronan bioconjugates. Pharm Res 2002;19(4):396e402. [24] Bayó-Puxan N, Dufresne M-H, Felber AE, Castagner B, Leroux J-C. Preparation of polyion complex micelles from poly (ethylene glycol) -polyions. J Control Release 2011;156(2):118e27. [25] Suma T, Miyata K, Ishii T, Uchida S, Uchida H, Itaka K, et al. Enhanced stability and gene silencing ability of siRNA-loaded polyion complexes formulated from polyaspartamide derivatives with a repetitive array of amino groups in the side chain. Biomaterials 2012;33(9):2770e9. [26] Aruffo A, Stamenkovic I, Melnick M, Underhill CB, Seed B. CD44 is the principal cell surface receptor for hyaluronate. Cell 1990;61(7):1303e13. [27] de la Fuente M, Seijo B, Alonso MJ. Novel hyaluronic acid-chitosan nanoparticles for ocular gene therapy. Invest Ophthalmol Vis Sci 2008;49(5):2016e24. [28] Duceppe N, Tabrizian M. Factors influencing the transfection efficiency of ultra low molecular weight chitosan/hyaluronic acid nanoparticles. Biomaterials 2009;30(13):2625e31. [29] Al-Qadi S, Alatorre-Meda M, Taboada P, Remuñán-López C, Zaghloul EM. Chitosan-hyaluronic acid nanoparticles for gene silencing: the role of hyaluronic acid on the nanoparticles’ formation and activity. Colloids Surf B Biointerfaces 2013;103(1):615e23. [30] Oyarzun-Ampuero FA, Goycoolea FM, Torres D, Alonso MJ. A new drug nanocarrier consisting of polyarginine and hyaluronic acid. Eur J Pharm Biopharm 2011;79(2):54e7. [31] Ladeiro Y, Couchy G, Balabaud C, Bioulac-Sage P, Pelletier L, Rebouissou S, et al. MicroRNA profiling in hepatocellular tumors is associated with clinical features and oncogene/tumor suppressor gene mutations. Hepatology 2008;47(6):1955e63. [32] Pereira DM, Rodrigues PM, Borralho PM, Rodrigues CM. Delivering the promise of miRNA cancer therapeutics. Drug Discov Today 2013;18(5e6): 282e9. [33] Li L, Yuan L, Luo J, Gao J, Guo J, Xie X, et al. MiR-34a inhibits proliferation and migration of breast cancer through down-regulation of Bcl-2 and SIRT1. Clin Exp Med; 2012:1e9. [34] De la Fuente M, Seijo B, Alonso M. Design of novel polysaccharidic nanostructures for gene delivery. Nanotechnology 2008;19(7):075105. [35] Park K, Lee M-Y, Kim KS, Hahn SK. Target specific tumor treatment by VEGF siRNA complexed with reducible polyethyleneimineehyaluronic acid conjugate. Biomaterials 2010;31(19):5258e65. [36] Oh EJ, Park K, Kim KS, Kim J, Yang J-A, Lee MY, et al. Target specific and longacting delivery of protein, peptide, and nucleotide therapeutics using hyaluronic acid derivatives. J Control Release 2010;141(1):2e12. [37] Choi KY, Yoon HY, Kim J-H, Bae SM, Park R-W, Kang YM, et al. Smart nanocarrier based on PEGylated hyaluronic acid for cancer therapy. ACS Nano 2011;5(11):8591e9. [38] Jabr-Milane LS, van Vlerken LE, Yadav S, Amiji MM. Multi-functional nanocarriers to overcome tumor drug resistance. Cancer Treat Rev 2008;34(7): 592e602. [39] Youle RJ, Strasser A. The BCL-2 protein family: opposing activities that mediate cell death. Nat Rev Mol Cell Biol 2008;9(1):47e59. [40] Zhang L, Lu Z, Zhao Q, Huang J, Shen H, Zhang Z, et al. Enhanced chemotherapy efficacy by sequential delivery of siRNA and anticancer drugs using PEI-grafted graphene oxide. Small 2011;7(4):460e4. [41] Emi M, Kim R, Tanabe K, Uchida Y, Toge T. Targeted therapy against Bcl-2related proteins in breast cancer cells. Breast Cancer Res 2005;7(6):R940e52. [42] Kim S-J, Oh J-S, Shin J-Y, Lee K-D, Sung KW, Nam SJ, et al. Development of microRNA-145 for therapeutic application in breast cancer. J Control Release 2011;155(3):427e34. [43] Yoo HS, Lee KH, Oh JE, Park TG. In vitro and in vivo anti-tumor activities of nanoparticles based on doxorubicinePLGA conjugates. J Control Release 2000;68(3):419e31. [44] Kalyanaraman B, Joseph J, Kalivendi S, Wang S, Konorev E, Kotamraju S, et al. Doxorubicin-induced apoptosis: implications in cardiotoxicity. Mol Cell Biochem 2002;37:119e24. [45] Parr C, Watkins G, Jiang W. The possible correlation of Notch-1 and Notch-2 with clinical outcome and tumour clinicopathological parameters in human breast cancer. Int J Mol Med 2004;14(5):779e86.