Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy

Acta Biomaterialia xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Full length article

Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy Lianjiang Tan a,⇑, Ran Huang b, Xiaoqiang Li c, Shuiping Liu c, Yu-Mei Shen a,⇑ a

Key Laboratory of Systems Biomedicine (Ministry of Education), Shanghai Center for Systems Biomedicine, Shanghai Jiao Tong University, Shanghai 200240, China State Key Laboratory of Microbial Metabolism and School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai 200240, China c Key Laboratory of Eco-Textiles, Ministry of Education and College of Textile & Clothing, Jiangnan University, Wuxi 214122, China b

a r t i c l e

i n f o

Article history: Received 13 February 2017 Received in revised form 20 April 2017 Accepted 8 May 2017 Available online xxxx Keywords: Controllable release Nitric oxide Doxorubicin Antitumor Synergistic therapy

a b s t r a c t NaYF4:Yb,Er upconversion nanoparticles (UCNPs) capped with long-chain carboxylic acid were synthesized and then conjugated with chitosan (CS) in the aid of N-hydroxysuccinimide. The resultant nanocompound was integrated with doxorubicin (DOX) and Roussin’s black salt (RBS), a photosensitive nitric oxide (NO) donor to produce stimuli-responsive UCNPs(DOX) CS-RBS nanospheres as nanocarriers for controllable drug delivery. On the one hand, the encapsulated UCNPs can efficiently absorb NIR photons and convert them into visible photons to trigger NO release. On the other hand, the entrapped DOX can be released at lowered pH from the swollen nanospheres caused by stretched oleoyl-CS chains under acidic conditions. The UCNPs(DOX)@CS-RBS nanospheres exhibit great therapeutic efficacy, which is attributable to the combination of NO and DOX releases based on NO dose-dependent mechanisms. This study highlights the controllable release of NO and DOX from the same nanocarriers and the synergistic therapeutic effect on tumors, which could give new insights into improving cancer nanotherapeutics. Statement of Significance In this paper, core-shell structured UCNPs(DOX) CS-RBS nanospheres have been designed and synthesized via a step-by-step procedure. The stimuli-responsive UCNPs(DOX) CS-RBS nanospheres act as nanocarriers for controllable drug delivery towards cancer therapy. The encapsulated UCNPs can efficiently absorb NIR photons and convert them into visible light to trigger NO release. Meanwhile, the entrapped DOX can be released from the swollen nanospheres caused by stretched oleoyl-CS chains at lowered pH typical of intracellular environment. Synergistic cancer therapy will be achieved through the combination of NO and DOX releases based on NO dose-dependent mechanisms. This study provides new drug nanocarriers with high antitumor efficacy for synergistic cancer therapy. Ó 2017 Published by Elsevier Ltd on behalf of Acta Materialia Inc.

1. Introduction Nitric oxide (NO) is an endogenously generated short-lived free radical that acts as a bioregulatory molecule in the body [1]. It plays a pivotal role in a variety of physiological and pathological processes [2]. NO has been reported as being able to modulate various cancer-related events such as cell cycle, apoptosis, angiogenesis, metastasis and invasion [3]. It has been suggested that NO has tumoricidal effects at appropriate concentrations and timing [4–6]. The likely mechanisms proposed for the anticancer ⇑ Corresponding authors. E-mail addresses: [email protected] (L. Tan), [email protected] (Y.-M. Shen).

properties of NO include suppression of DNA synthesis, proapoptotic modulator by activating caspase family proteases, expression alteration of apoptosis-associated proteins, etc. [7–9]. Moreover, it has been found that the resistance of cancer cells to chemotherapeutics may be reversed by NO by means of reducing Pglycoprotein (P-gp) expression levels [10,11]. Considering the potential of NO as a cancer therapeutic agent, developing controllable NO donors for targeted NO delivery attracts extensive interest. Photoexcited NO delivery has unique advantages in that it allows precise temporal and spatial control of NO release with few influences from the surroundings [12,13]. By tuning the photoexcitation signal, dose-controllable NO release to specific physiological targets could be achieved. Compared with visible light and

http://dx.doi.org/10.1016/j.actbio.2017.05.019 1742-7061/Ó 2017 Published by Elsevier Ltd on behalf of Acta Materialia Inc.

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

2

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

ultraviolet (UV), near-infrared (NIR) light is preferable for photogeneration of NO as it has great penetration depth in and negligible damage to tissues [14]. Researchers have devoted their efforts to the development of NIR-triggered NO delivery systems [15–17]. In their NO-generating nanovehicles, upconversion nanoparticles (UCNPs) were used as photoactive centers that harvest NIR photons and upconvert them into visible light. The incorporated Roussin’s black salt (RBS) absorbs the visible photons and releases NO via photochemical reaction [18]. As a cationic polysaccharide, chitosan (CS) has been extensively used for various biomedical applications [19–21]. The pKa of CS is 6.0–6.5 in aqueous media [22], and the charged state and physiochemical properties of CS are significantly influenced by the ambient environmental pH [23]. CS has been found to form dissociated precipitates in aqueous phase at physiological pH of 7.4 due to rapid local aggregation of CS polymeric chains [24]. Sung’s group fabricated a comblike associating polyelectrolyte by conjugating a hydrophobic palmitoyl group onto the free amine groups of CS [23,25]. Through balancing charge repulsion and hydrophobic interaction, the chain conformation of the associating polyelectrolyte can be controlled simply by adjusting the environmental pH within a narrow range. Inspired by their work, we have intended to design a pH-sensitive nanostructure composed of a UCNP core and a CS shell for drug delivery purposes. This coreshell nanostructure could be realized by conjugating the organic ligands capping the UCNP with the amine groups of CS. In the current work, we report preparation of CS-encapsulating NaYF4:Yb,Er UCNPs with entrapped doxorubicin (DOX) and attached RBS for synergistic cancer therapy. Oleic acid-capping NaYF4:Yb,Er UCNPs were synthesized, reacted with Nhydroxysuccinimide and conjugated with CS at some of the amino sites. Coexisting with DOX in aqueous solution, the hydrophobic oleoyl groups tended to form local aggregates and entrap DOX by hydrophobic interaction, producing UCNPs(DOX) CS nanospheres spontaneously. The core-shell structured nanospheres were further conjugated with the RBS [NH4][Fe4S3(NO)7] via electrostatic interaction (Fig. 1a). In the resultant UCNPs(DOX)@CS-RBS nanospheres, the UCNPs act as antennae to receive NIR photons and convert them into visible photons, which sensitize the attached RBS to trigger NO release. On the other hand, the oleoyl-CS chains are sensitive to environmental pH changes. At pH
7.0, the DOX is entrapped in the nanospheres owing to the strong hydrophobic interaction between the oleoyl groups. At a low pH, the protonated amine groups increase the charge repulsion between the oleoyl-CS chains. As a result, the entrapped DOX is released as the chains expanded (Fig. 1b). The developed UCNPs(DOX) CS-RBS nanospheres in this work are aimed at synergistic tumor therapy through dose-controllable NO generation in combination with pH-responsive antitumor drug release.

2. Experimental 2.1. Materials NaOH (>98%), NH4F (99%), YCl36H2O (99.99%), YbCl3 (99.99%), ErCl3 (99.99%), 1-octadecene (98%), oleic acid (99%), Nhydroxysuccinimide (NHS, 99%), fetal bovine serum (FBS), Dulbecco’s modified Eagle’s medium (DMEM), 3-(4,5-dimethyl-thia zol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), doxorubicin hydrochloride (DOXHCl, >98%) and N,N0 -dicyclohexylcarbodiimide (DCC, 98%) were purchased from Sigma-Aldrich. Chitosan (CS, Mw = 35000, degree of deacetylation = 88.7%, pKa = 6.09) was purchased from Sinopharm Chemical Reagent Co., Ltd, China, and used without further purification. Roussin’s Black Salt (RBS, [NH4][Fe4S3(NO)7]) was prepared following the procedure reported by

Seyferth et al. [26]. The RBS was stored in the dark and inert atmosphere. Thiolated oleic acid was prepared in our own lab. Millipore water was used in all experiments. All other chemicals and solvents were provided by Sinopharm Chemical Reagent Co., Ltd and used as received without purification (analytically pure). 2.2. Synthesis of NaYF4:Yb,Er UCNPs NaYF4:Yb,Er UCNPs were synthesized according to the following procedure: 1.56 mmol of anhydrous YCl3, 0.40 mmol of YbCl3 and 0.04 mmol of ErCl3 were added to 12 mL of oleic acid and 15 mL of 1-octadecene under stirring. The mixture was heated to 150 °C and held for 30 min under vacuum to remove oxygen and water before cooling down to room temperature. Then 5 mmol of NaOH and 8 mmol of NH4F were added and the resultant solution was stirred at room temperature for 30 min. Thereafter, the solution was transferred to an autoclave and heated to 300 °C in argon atmosphere for 1 h and cooled down naturally to room temperature. Precipitates were obtained by adding excessive ethanol and centrifuging at 6830g for 15 min. Subsequently, a mixture of assynthesized NaYF4:Yb,Er UCNPs, 15 mL of thiolated oleic acid, 20 mL of 1-octadecene and 20 mmol of ethanol were stirred at room temperature for 32 h for ligand exchange. The oleic acidcapping NaYF4:Yb,Er UCNPs were obtained by centrifugation at 20,490g for 20 min, washing with ethanol, and dialysis against water. 2.3. Preparation of UCNPs(DOX)@CS-RBS nanospheres The as-synthesized UCNPs were mixed with 35 mmol of NHS in 20 mL of anhydrous dimethyl formamide (DMF), to which 70 mmol of DCC was added slowly and the mixture was allowed to react under stirring for 24 h at room temperature in nitrogen atmosphere. Subsequently, the mixture was filtered, washed by ethyl ether thoroughly and rotation-evaporated to obtain UCNPs capped by oleic acid N-hydroxysuccinimide ester. The modified UCNPs were dispersed in ethanol, which was then added dropwise to 20 mL of a CS (0.2 g)/aqueous acetic acid (1 wt%) solution at 95 °C to react for 36 h under stirring. The resultant mixture was cooled down to room temperature and precipitated by adding acetone and adjusting pH to 9.0. The precipitates were filtered, washed with acetone for three times, air-dried and redispersed in aqueous acetic acid. The degree of substitution on CS was 11.1 ± 0.2% (n = 5) based on ninhydrin assay [24]. Thereafter, DOXHCl was added into the above aqueous acetic acid solution of UCNPs@CS at room temperature. The preset weight ratio of DOX to UCNPs@CS was 1:10. The resultant solution was stirred in the open air for 10 min, whose pH was then adjusted to 7.4. After stirring for another 30 min, the solution was dialyzed against water (pH 7.4) to remove free DOX. At room temperature, 50 mg of the obtained UCNPs(DOX)@CS was redispersed in 20 mL of aqueous acetic acid in a brown vial, where 20 mL of RBS aqueous solution (0.5 mg/mL) was added and the mixture was stirred for 10 min in the dark under nitrogen protection. Then the pH of the mixture was adjusted to 7.4, and stirring was continued for another 5 h. The resultant UCNPs (DOX)@CS-RBS nanospheres were obtained by centrifugation at 6830g for 10 min, washed thoroughly with ethanol for three times, and dried under vacuum for 3 h. Loading content (LC) and loading efficiency (LE) of DOX were calculated using the equations as follows:

LC DOX ¼ ðweight of loaded DOX=weight of nanospheresÞ  100% LEDOX ¼ ðweight of loaded DOX=weight of DOX in feedÞ  100%

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

3

Fig. 1. (a) Schematic synthesis procedure of UCNPs(DOX)@CS-RBS. (b) Scheme of pH-responsive DOX release and NIR-triggered NO release from UCNPs(DOX) CS-RBS.

The amount of DOX left in the supernatant after centrifugation was determined by measuring the UV–vis absorbance at 485 nm, and the loading of DOX was obtained accordingly. To determine the LC of RBS, UV–vis absorbance of the supernatant after centrifugation at 430 nm was measured. The RBS amount in the supernatant was obtained based on the standard curves of RBS at varied concentrations (Fig. 1S, SI). The LC and LE of RBS were calculated using the equations as follows:

LC RBS ¼ ðweight of assayed RBS=weight of nanospheresÞ  100% LERBS ¼ ðweight of assayed RBS=weight of RBS in feedÞ  100%

2.4. Characterization Fourier transform infrared (FTIR) spectroscopy was performed on a NEXUS 670 FT-IR&Ramen spectrometer (Thermo Nicolet, US). X-ray diffraction (XRD) pattern was recorded by a D/max2200/PC X-ray diffractometer (Rigaku, Japan). Morphology observation and selected area electron diffraction (SAED) were carried out using a JEM-2100 transmission electron microscope (TEM, JEOL, Japan). A small amount of sample solution was dropped on

a carbon-coated copper grid, which was then freeze-dried in vacuum at -50 °C before testing. Size distribution and zeta potential of samples were determined by a Nano ZS90 particle size and zeta potential analyzer (Malvern, UK) based on dynamic light scattering (DLS). Thermogravimetry analysis (TGA) was conducted on a Q5000IR thermogravimetric analyzer (TA, US), with the samples heated under nitrogen flow from 25 °C to 800 °C at a rate of 20 °C/min. UCL and UCL excitation spectra were recorded with a Fluorolog-3 fluorescence spectrophotometer (Horiba Jovin Yvon, France) equipped with an adjustable continuous wave laser (300–1600 nm, 0–10 W). The excitation wavelength was tuned at 980 nm. UCL lifetime was measured on the Fluorolog-3 fluorescence spectrophotometer using the 980 nm laser with pulse duration of 5 ls. The Y concentration in different parts of the mice was determined on an X series inductively coupled plasma mass spectrometer (ICP-MS) (Thermo Elemental, UK). 2.5. Measurement of DOX release 2 mL of UCNPs(DOX)@CS-RBS/PBS solutions (1 mg/mL) was transferred into a dialysis tube with a molecular weight cut off of 1000 Da, which was then immersed in a beaker filled with 50 mL of PBS buffer at varied pH values (5, 6 and 7.4) to be measured at

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

4

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

37 °C. At predetermined time intervals, 5 mL of external solution was withdrawn and analyzed using a Synergy 2 Multi-Mode Reader (BioTek, US). The beaker was immediately refilled with 5 mL of fresh PBS of the same composition and pH for the next sampling. The cumulative release of DOX was determined by measuring the fluorescence intensity at 580 nm under excitation at 485 nm. The experiments were performed in triplicate. The cumulative DOX release is calculated according to the following equation [27]:

Er ¼

Ve

Pn1 l

Ci þ V o Cn mdrug

where Er is the total cumulative release% of DOX; Ve is the replacement of PBS volume (5 mL); V0 is the total volume of PBS (50 mL); Ci is DOX concentration of the i-th replacement liquid (mg/mL) (determined by fluorescence measurement); Cn is DOX concentration of the last replacement liquid (mg/mL); mdrug is the total amount of DOX encapsulated (mg). 2.6. Measurement of NO release Quantitative measurement of the released NO molecules from UCNPs@CS-RBS and UCNPs(DOX)@CS-RBS was carried out by a TBR 4100/1025 free radical analyzer equipped with an ISO-NOP sensor (WPI Ltd., US). The NO detection protocols were described in our previous work [28]. 2.7. Intracellular drug release CT26 cells (mouse colonic cancer cells) and MCF-7/ADR cells (drug-resistant human breast cancer cells), provided by Institute of Biochemistry and Cell Biology, Chinese Academy of Science (CAS), were cultured in Dulbecco’s Modified Eagle Medium (DMEM) with 10% FBS at 37 °C with 5% CO2. To monitor DOX release and NIR-triggered NO generation in cancer cells, 0.5 mL of CT26 cell suspension was transferred to an eight-well Lab-Tek II chamber slide (Nalge Nunc, Napevillem, IL), followed by removing the culture medium and adding a solution of UCNPs(DOX)@CSRBS nanospheres (10 lg/mL) along with 1 lg of Cu-2-{2-chloro-6hydroxy-5-[(2-methyl-quinolin -8-ylamino)-methyl]-3-oxo-3H-xa nthen-9-yl}-benzoic acid (CuFL), a NO fluorescent probe [29]. The cells were incubated for 4 h for cellular uptake of the nanospheres and the NO probes. At predetermined time points, the medium was aspirated from the wells, and the cells were rinsed with fresh culture medium for three times. The cell fluorescence was observed by a BX51TF fluorescence microscope (Olympus, Japan).

2.9. Expression of P-gp CT26 cells were incubated with 50 lM UCNPs(DOX) CS-RBS nanospheres for 4 h. Then the cells were irradiated by the 980 nm laser with the power density of 0.7 W/cm2 for different time periods. After that, the cells were washed twice with PBS, fixed in 4% paraformaldehyde, and then examined using immunofluorescence stains to quantify the expression of P-gp. The antibody used herein was anti-Pgp antibody (Abcam #ab129450, USA). The stained cells were analyzed by a flow cytometry (BD FACS Calibur, US). 2.10. Western blot analysis CT26 cells were seeded in 6-well plates at a density of 6  105 cells per well in 1.5 mL of complete DMEM and incubated for 24 h to grow. Then the cells were treated with DOX, UCNPs(DOX) @CS, UCNPs@CS-RBS and UCNPs(DOX) CS-RBS respectively at the same concentration of 50 lg/mL for 4 h. Then the cells were irradiated by the 980 nm laser for 5 min or 30 min (0.7 W/cm2) and further incubated for 24 h. Untreated CT26 cells were used as a control. Thereafter, the cells were washed with PBS and lysed in an ice-cold lysis buffer. The cell lysates (30 lg/well) were separated by 12% sodium dodecyl sulfate-polyacrylamide gels and transferred onto polyvinylidene fluoride (PVDF) membranes. The loaded membranes were incubated in 5% skimmed milk dissolved in Tris buffered saline containing 0.05% Tween-20 (TBST) at 37 °C and detected with relevant antibodies against b-actin (loading control), caspase-3 and P-gp at 1:1000 dilution. Then the membranes were washed and incubated with horseradish peroxidase (HRP)labeled anti-rabbit immunoglobulin-G (IgG) at 1:5000 dilution. Protein bands were analyzed using a ChemiDoc MP Imaging System (Bio-Rad, US). 2.11. Calcein AM/propidium iodide staining The cultured CT26 cells were transferred to a 24-well plate at a density of 42000 cells per well, into which fresh medium containing different drug formulations (at an equivalent DOX concentration of 1 lg/mL and an equivalent RBS concentration of 1.4 lg/ mL) were added. After incubation for 4 h, the cells treated with UCNPs@CS-RBS and UCNPs(DOX)@CS-RBS nanospheres were exposed to 980 nm-laser irradiation for 5 min or 30 min (0.7 W/ cm2). Then the cells were further incubated for 24 h and stained with calcein AM and propidium iodide, which denotes living cells and dead cells, respectively. The cells were observed by the BX51TF fluorescence microscope.

2.8. MTT assay

2.12. Cell apoptosis assay

Cytotoxicity of free DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS and UCNPs(DOX) CS-RBS nanospheres under various conditions was assessed by MTT viability assay. CT26 cells or MCF-7/ADR cells were seeded in 96-well culture plates at a density of 4800 cells/ well and incubated at 37 °C for 24 h to attach the cells. The culture medium in each well was then replaced by a fresh medium containing DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS and UCNPs(DOX) @CS-RBS at different concentrations. After incubation for 4 h, the cells containing UCNPs CS-RBS and UCNPs(DOX) CS-RBS nanospheres were irradiated by the 980 nm laser for 5 min or 30 min with a power density of 0.7 W/cm2. After additional incubation for 24 h, the cell viability was determined following the protocols reported in our previous work [28]. For the cytotoxicity of UCNPs@CS and UCNPs@CS-RBS without NIR irradiation, the MTT assay was performed for the CT26 cells with incubation time of 24 h and 48 h, respectively.

CT26 cells were seeded in 6-well plates at a density of 6  105 cells per well in 1.5 mL of complete DMEM and cultured for 24 h for attachment. The cells were treated with UCNPs(DOX)@CS, UCNPs@CS-RBS plus 30 min irradiation and UCNPs(DOX) CS-RBS plus 30 min irradiation respectively at a concentration of 50 lg/ mL for 24 h. CT26 cells untreated were used as a control. Quantitative measurement of apoptosis and necrosis was conducted on the flow cytometry (BD FACS Calibur, US). The treated cells were collected and washed for three times with ice-cold PBS, stained with fluorescein isothiocyanate (FITC)-Annexin V and propidium iodide (PI). 2.13. Animal tumor model and histology analysis A mouse CT26 tumor model was used to evaluate in vivo tumor inhibition by UCNPs(DOX) CS-RBS and other drug formulations

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

for comparison. All experiments were performed according to approved protocols by the Animal Ethics Committee of Shanghai Jiao Tong University. Seven weeks old female BALB/c mice (18 g) were provided by Institute of Biochemistry and Cell Biology, CAS. CT26 cells were washed twice in PBS and re-suspended in sterile normal saline. The cell suspension containing 1  105 CT26 cells was injected subcutaneously in the flank region of the mice to achieve tumor inoculation. The tumors were allowed to grow to a volume of 100 mm3, which was estimated with the reported method [30]. The CT26 tumor-bearing mice were randomly divided into five groups. The mice in groups 1–3 were treated with tail vein injections of PBS, DOX (dose of 0.44 mg/kg body weight) and UCNPs(DOX)@CS (19.4 mg/kg), respectively. The mice in groups 4 and 5 were injected intravenously with UCNPs@CS-RBS (19.2 mg/kg) and UCNPs(DOX)@CS-RBS (20 mg/kg), followed by 980 nm-laser irradiation (0.7 W/cm2, for 30 min) at 24 h postinjection. The treatments for the five groups of mice were repeated every other day for a total of 12 days. The tumor volumes and body weights of each treatment modality were measured every other day. At the end of the treatments, the mice were anesthetized and sacrificed, from which tumors were harvested and cut into small pieces, fixed in 10% formalin, paraffin-blocked, and sectioned for histopathological analysis via H&E staining. The tissue sections were imaged using an Olympus Fluorescence Microscope (Olympus BX61, Japan). 2.14. Statistical analysis Student’s t-test was used to evaluate the statistical significance of differences between treatments. Values were represented as mean ± standard deviation and the data were considered as statistically significant at P < 0.05 or P < 0.01. 3. Results and discussion 3.1. Morphology and structure As antennae that collect NIR energy and convert it into visible light, NaYF4:Yb,Er UCNPs were first synthesized and characterized. The chemical composition of as-synthesized NaYF4:Yb,Er UCNPs was confirmed by the EDX pattern (Fig. 2S, SI). Morphology of the NaYF4:Yb,Er UCNPs and UCNPs(DOX) CS-RBS nanospheres was observed by TEM. Fig. 2a shows monodisperse NaYF4:Yb,Er UCNPs with a spherical profile. The average diameter of the UCNPs was 27.4 nm, determined through measuring 100 nanoparticles by TEM. The high-resolution TEM image of the UCNPs in Fig. 2b indicates two groups of lattice fringes with an interplanar spacing of 0.52 nm, which are assigned to (1 0 0) and (0 1 0) lattice planes of a cubic NaYF4 phase [31,32]. The SAED pattern (Fig. 2b, inset) shows diffraction spots characteristic of cubic NaYF4 crystals. The crystalline structure of the UCNPs was further confirmed by XRD (JCPDS 77-2042, Fig. 3S, SI). The NaYF4:Yb,Er UCNPs capped by oleic acid were then conjugated with CS with the help of N-hydroxysuccinimide. After encapsulation of DOX in the UCNPs@CS nanospheres, the protonated amino groups in the CS attract RBS anions to give UCNPs(DOX) @CS-RBS nanospheres, the morphology of which is shown in Fig. 2c. A spherical core-shell structure can be clearly seen, with the UCNPs encapsulated by the CS shell. Fig. 2d shows size distribution of the UCNPs(DOX) CS-RBS nanospheres based on DLS. The nanospheres are well dispersed in water, having an average size of 75.4 nm and a polydispersity index (PDI) of 0.159. TG analysis is useful for determining the UCNPs content in the nanospheres. As indicated in Fig. 2e, the mass of UCNPs(DOX)@CS-RBS nanospheres gradually decreased with increasing temperature

5

until beyond 700 °C. The remaining mass at 800 °C accounted for 9.47% of the total, which was the loading of UCNPs. FTIR spectroscopy was employed to evidence the conjugation of the oleic acid-capping NaYF4:Yb,Er UCNPs with the CS. As shown in Fig. 2f, the FTIR spectrum of the UCNPs@CS exhibited increasing absorption peaks at 1664 cm1 and 1572 cm1 compared with the CS, ascribed to C@O stretching vibration (amide I band) and NAH bending vibration (amide II band) of amide groups, respectively. Besides, the absorption band at 2868–2934 cm1 typical of A(CH2)A for the UCNPs@CS increased significantly compared with the CS. These results imply that the NaYF4:Yb,Er UCNPs were successfully conjugated with the CS. The EDX pattern of UCNPs(DOX) @CS-RBS nanospheres shows the existence of Fe, verifying the binding of RBS to the nanospheres (Fig. 4S, SI). The LC of DOX was 4.10 ± 0.29% and 4.07 ± 0.24% for UCNPs(DOX)@CS-RBS and UCNPs(DOX)@CS, respectively. The corresponding LE was 28.9 ± 1.9% and 29.2 ± 1.6%. The LC of RBS was 2.88 ± 0.22% and 2.65 ± 0.21% for UCNPs(DOX)@CS-RBS and UCNPs@CS-RBS, respectively. The corresponding LE was 85.3 ± 4.1% and 83.4 ± 4.4%. 3.2. Optical properties The upconversion luminescence (UCL) properties of UCNPs (DOX) CS-RBS nanospheres were examined. The UCL excitation spectra of the UCNPs(DOX)@CS-RBS and NaYF4:Yb,Er UCNPs show a maximum at 980 nm (Fig. 5S, SI). The UCL spectra obtained under 980 nm NIR excitation are shown in Fig. 3a. The UCNPs (DOX)@CS-RBS showed similar emission bands to the UCNPs, suggesting that the UCL properties of the UCNPs kept almost unchanged after they were conjugated with CS. The green emissions at 511–533 nm and 533–554 nm are assigned to the 2 H11/2 ? 4I15/2 and 4S3/2 ? 4I15/2 transitions, respectively [33]. The red emission at 660 nm corresponds to the transition from 4F9/2 to 4I15/2 [34]. The UCL lifetime measurements of the UCNPs(DOX) @CS-RBS (Fig. 2b) indicated a decay time of 112.6 ls for the emission at 554 nm. The long UCL lifetime favors biological applications of the UCNPs(DOX) CS-RBS as a NIR-triggered drug delivery system. The photostability of the UCNPs(DOX)@CS-RBS nanospheres was assessed by exposing them to continuous irradiation of a mercury lamp (Fig. 3c). The fluorescence intensity changes of Cy3 under the same irradiation conditions were also recorded for comparison, as the emission wavelength of Cy3 is close to that of the UCNPs(DOX) CS-RBS. The UCNPs(DOX)@CS-RBS maintained 77.2% of the original UCL intensity after being irradiated for 3 h, whereas the Cy3 showed a rapid fluorescence decrease and was nearly photobleached at 150 min. Incubated in different media at 37 °C, the UCNPs(DOX)@CS-RBS retained high colloidal stability, without noticeable reduction in UCL intensity over 7 days (Fig. 3d). All these results demonstrate high photostability and chemical stability of the UCNPs(DOX) CS-RBS nanospheres. 3.3. Drug release analysis Drug delivery properties are vital for a drug carrier and should be assessed. We first investigated the NIR-triggered NO release from UCNPs(DOX) CS-RBS nanospheres via quantitative detection of released NO under NIR laser irradiation. The NO concentration was detected using a sensitive NO electrode based on an amperometric technique. The NO release profiles of the UCNPs(DOX) CSRBS nanospheres in a PBS buffer under irradiation of varied power density are shown in Fig. 4a. Without irradiation, only a very small amount of NO was detected in the data range, as expected. By contrast, NO was released from the nanospheres under the irradiation of NIR laser, and the amount of release increased with increasing irradiation power density, indicating that the dose-controllable

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

6

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

Fig. 2. (a) TEM image of NaYF4:Yb,Er UCNPs; (inset) corresponding size distribution histograms of the UCNPs. (b) High-resolution TEM image of NaYF4:Yb,Er UCNPs; (inset) SAED pattern of the UCNPs. (c) TEM image of UCNPs(DOX)@CS-RBS nanospheres; (inset) high-resolution TEM image of a UCNPs(DOX)@CS-RBS nanosphere. (d) Size distribution of UCNPs(DOX) CS-RBS nanospheres at pH 7.4. (e) TG curve of UCNPs(DOX)@CS-RBS nanospheres. (f) FTIR spectra of chitosan (I) and UCNPs@CS (II).

release of NO could be achieved by adjusting the NIR laser power output. The half-life of NO release was hardly influenced by the irradiation power density, as can be seen in Table 1. For biological studies and clinical applications, the conservative limit of 980 nm laser intensity set for human skin exposure is 0.726 W/cm2 [35]. Hence, a safe power density of 0.7 W/cm2 was chosen in the following cell assay and small animal experiments. The NIR-triggered NO release was investigated by means of instantaneous NO detection. As shown in Fig. 4(b), initial irradiation on UCNPs(DOX)@CS-RBS/PBS solution (50 lg/mL) for 30 s led to an immediate electrode response. When irradiation was

halted, NO was no longer generated, and the current signal decreased due to the oxidation of NO by the ambient molecular oxygen [28,36]. Continued irradiation resulted in new bursts of signal, the intensity of which increased with prolonged irradiation time. In addition, the NO release from the nanospheres showed little sign of fatigue subjected to the repeated ‘‘on-off” irradiation. Influences of pH on the diameter and zeta potential of the UCNPs(DOX)@CS-RBS nanospheres were examined to show the pH responsiveness of the nanospheres (Fig. 4c). At low pH values, the amine groups on the CS shell were strongly protonated (highly charged), as reflected by the high zeta potential. The dominant

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

7

Fig. 3. (a) UCL spectra of UCNPs and UCNPs(DOX)@CS-RBS excited by 980 nm laser. (b) UCL lifetime data of UCNPs(DOX)@CS-RBS at 555 nm, exhibiting a lifetime s = 113.7 ± 10.2 ls based on biexponential fitting. (c) Photostability of UCNPs(DOX)@CS-RBS and Cy3 under continuous irradiation of a mercury lamp (50 W/cm2) for 3 h. The data are presented as average ± standard deviation (n = 3). (d) UCL intensity changes of UCNPs(DOX)@CS-RBS stored in PBS buffer (pH = 7.4), 10% FBS solution and mouse whole blood at 37 °C during a period of 7 days. The data are presented as average ± standard deviation (n = 3).

Fig. 4. (a) NO release from UCNPs(DOX)@CS-RBS in a PBS buffer (pH 7.4, 37 °C) as a function of time under 980 nm laser irradiation at different power densities. (b) NO release from UCNPs(DOX) CS-RBS controlled by ‘‘on-off” irradiation of 980 nm laser at 0.7 W/cm2. The sequential signals are the results of excitation for 30, 30, 40, 50, 60 and 60 s. (c) Particle size and zeta potential of UCNPs(DOX) CS-RBS as a function of pH value. The data are presented as average ± standard deviation (n = 3). (d) DOX release from UCNPs(DOX) CS-RBS in PBS buffers of different pH values at 37 °C. The data are presented as average ± standard deviation (n = 3).

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

8

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

Table 1 Half-lives of NO release and DOX release from UCNPs(DOX) CS-RBS nanospheres under different conditions. Drug type

Irradiation power (W/cm2)

Half-life (min)

Drug type

pH

Half-life (h)

NO NO NO

1.0 0.7 0.5

8.4 8.3 8.3

DOX DOX DOX

5.0 6.0 7.4

4.4 5.9 7.1

charge repulsion between the protonated amine groups gave rise to polymer chain extension [23] and hence larger size of the nanospheres. With the increase of pH, the electrostatic repulsion between the CS polymeric chains decreased as deprotonation of the amine groups occurred (reduced zeta potential). On this occasion, the hydrophobic interaction between the oleoyl chains took hold, leading to decrease in the size of the nanospheres. It is noted that the size of nanospheres changed more dramatically in the pH range of 5.0–6.0 than in the range of pH > 6.0. The pH-triggered transition from a stretching structure to a contracting structure enables the nanospheres to act as an on-off switch in response to ambient pH changes. The cumulative amount of DOX released from the UCNPs(DOX) @CS-RBS nanospheres at varied pH values was measured by detecting the fluorescence signal characteristic of DOX. The release profiles over a 30 h period are shown in Fig. 4d. At pH 7.4, most of the DOX molecules were entrapped in the compact nanospheres, resulting in small doses of DOX release. At lower pH values, the nanospheres became swollen as the molecular chains of the shell were stretched, thereby releasing more DOX molecules. The release profile of a DOX solution with an equivalent DOX concentration under the same conditions (Fig. 6S, SI) demonstrate sustained release of DOX and the appropriate release media. It is known that the pH in endosomes and lysosomes of cancer cells usually lies in the range of 4.5–5.5 [37]. The acidic environment will trigger the DOX release from the nanospheres in the cancer cells. The half-life of DOX release decreased with decreasing pH (Table 1), indicating that the entrapped DOX was released at a higher rate under more acidic conditions. 3.4. Cellular uptake and cytotoxicity Before applying the UCNPs(DOX)@CS-RBS nanospheres to living cell imaging, we conducted MTT assay for CT26 cells incubated with UCNPs@CS-RBS and UCNPs@CS to evaluate the cytotoxicity of the carrier materials in the absence of drugs and/or drug release. The MTT results show low toxicity of the carrier materials to the CT26 cells (Fig. 7S, SI). Cellular uptake of the UCNPs(DOX) CSRBS nanospheres was investigated for CT26 cells to find out appropriate incubation time needed for related cell experiments throughout this work. The UCL characteristics of the nanospheres were utilized to observe internalization of the nanospheres. As shown in Fig. 5a, the UCL emission from the cells under 980 nm laser irradiation increased significantly within 4 h, indicating that a large amount of nanospheres entered the cells during the incubation. With the incubation time extended to 6 h, the emission intensity changed slightly compared with that at 4 h, suggesting that most of the nanospheres had been internalized within 4 h. We thereby chose 4 h as the time for cellular uptake in the following experiments. The cultured CT26 cells were incubated with UCNPs(DOX) CSRBS nanospheres and CuFL at 37 °C for 4 h to allow cellular uptake of the nanospheres. Then the cells were imaged by a fluorescence microscopy in different conditions. As shown in Fig. 5b, no fluorescence signal typical of CuFL reacted with NO was emitted from the cells under blue laser excitation, indicating that NO was not released without NIR irradiation. Very weak red fluorescence was

observed under green laser excitation, since almost all the NO was entrapped in the nanospheres at the beginning of incubation. After incubation for 5 h, the intracellular nanospheres entered endosomes/lysosomes and became swollen at reduced pH. DOX was released from the nanospheres, as evidenced by the appreciable increase of red fluorescence (Fig. 5c). Subsequent irradiation at 980 nm triggered release of NO, which turned on the green fluorescence of CuFL (Fig. 5d). In addition, we exposed the cells to NIR irradiation for 30 min immediately after internalization of the nanospheres, followed by 5 h incubation. Compared with that in Fig. 5d, the DOX signal was stronger, revealing a higher DOX concentration in the cells (Fig. 5e). The fluorescence change was confirmed by flow cytometry analysis, which showed a 21% increase. It seems that the presence of NO facilitated the accumulation of DOX in the cells. The proliferation of CT26 cells inhibited by UCNPs(DOX)@CSRBS nanospheres was evaluated by means of MTT assay to study the anticancer effects of the nanospheres (Fig. 6a). The cells were incubated for 4 h with free DOX, UCNPs(DOX)@CS, UCNPs@CSRBS or UCNPs(DOX) CS-RBS nanospheres at different equivalent concentrations of DOX or/and RBS for drug internalization. Then the cells were subjected to 980 nm-laser irradiation (0.7 W/cm2) for 30 min or 5 min. The untreated cells were used as a control. Free DOX did not have strong cancer-cell killing capability after 24 h incubation due to multidrug resistance (MDR) and thus insufficient DOX dose acting on the cell nuclei. At a low concentration, UCNPs@CS-RBS plus NIR irradiation had little effect on the cell viability compared with UCNPs(DOX)@CS and UCNPs(DOX)@CS-RBS plus NIR irradiation. With the increase of concentration, the viability of the cells containing UCNPs(DOX)@CS-RBS nanospheres decreased with a more significant downward trend than those containing UCNPs(DOX)@CS and UCNPs@CS-RBS plus NIR irradiation. It is noted that the UCNPs@CS-RBS killed more cells at higher concentrations. Nevertheless, the cell viability changed slightly once the concentration exceeded 70 lg/mL (Fig. 8S, SI), suggesting that the anticancer effect of NO without DOX is limited even at a much larger dose. The UCNPs(DOX) CS-RBS nanospheres released both NO and DOX in the cells under irradiation. P-gp acts as an efflux pump to expel chemotherapeutic agents out of cancer cells, hence reducing intracellular drug concentrations and pharmacological efficacy [38]. Overexpression of P-gp is often a main cause of MDR. The lower viability of the cells containing UCNPs(DOX) @CS-RBS nanospheres (either 5 min or 30 min irradiation) compared with the cells containing UCNPs(DOX) CS indicates that NO might help overcome MDR by inhibiting NO expression when combined with chemotherapy. The UCNPs@CS-RBS showed enhanced cancer-cell killing efficacy with increasing concentration. More importantly, the viability of the cells treated with UCNPs(DOX) CS-RBS nanospheres decreased dramatically as the concentration increased. Comparing the cells containing UCNPs (DOX)@CS-RBS nanospheres that were irradiated by the NIR laser for 5 min and for 30 min, we noticed that DOX in combination with a larger dose of NO gave rise to more cell death. Without RBS incorporated, the UCNPs(DOX)@CS nanospheres exhibited similar toxicity to CT26 cells under different NIR irradiation conditions (Fig. 9S, SI), indicating that NIR irradiation has no influence on viability of the cells containing the nanospheres. The toxicity of the UCNPs

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

9

Fig. 5. (a) UCL intensity changes of CT26 cells incubated with UCNPs(DOX)@CS-RBS nanospheres against incubation time, excited by 980 nm laser. Bright-filed and fluorescence images of CT26 cells containing UCNPs(DOX)@CS-RBS nanospheres and the NO probe CuFL: (b) immediately after internalization of the nanospheres; (c) at 5 h after internalization of the nanospheres; (d) as (b) followed by 980 nm laser irradiation with a power density of 0.7 W/cm2 for 30 min; (e) irradiated by 980 nm laser immediately after internalization of the nanospheres and then incubated for 5 h before imaging. The scale bars represent 50 lm.

(DOX)@CS-RBS nanospheres to MCF-7/ADR cells (drug-resistant) was also examined (Fig. 10S, SI). Against the MCF-7/ADR cells, the inhibition of cell proliferation by the nanospheres under irradiation for 5 min or 30 min was much more obvious than that by the UCNPs(DOX) CS. In the case of UCNPs@CS-RBS under NIR irradiation for 30 min, the viability of MCF-7/ADR cells decreased remarkably at higher concentrations. The MDR reversal caused by released NO results in intracellular accumulation of DOX at a concentration above the cell-killing threshold. Large-dose NO is capable of killing cancer cells directly despite the cell types. The above results demonstrate that a higher killing efficacy is achieved as the released NO and DOX exert combined effect upon cancer cells. The synergistic effect of the released NO and DOX was further studied by treating the CT26 cells with free DOX and UCNPs (DOX) CS-RBS nanospheres under different conditions and determining the cell viability after incubation for 48 h. The equivalent DOX concentration at which 50% of the cells survive (IC50) was determined. Specifically, the IC50 of UCNPs(DOX) CS-RBS nanospheres irradiated by 980 nm laser for 5 min and 30 min was 0.417 ± 0.011 lg/mL and 0.274 ± 0.008 lg/mL, respectively, smal-

ler than that of the UCNPs(DOX)@CS-RBS nanospheres without irradiation (0.828 ± 0.015 lg/mL). NIR irradiation for 5 min triggered release of small-dose NO that inhibited P-gp expression and thus increased the intracellular DOX concentration. A longer NIR irradiation time of 30 min led to a larger dose of NO, which could kill the cancer cells along with the DOX in addition to the MDR reversal. To examine closely the influences of NO dose on P-gp inhibition, immunofluorescence stains were employed to quantify the changes of P-gp expression with NIR irradiation time (Fig. 6b). The relative fluorescence intensity decreased significantly after the CT26 cells containing UCNPs(DOX) CS-RBS nanospheres were irradiated for 5 min. With prolonged irradiation time, the fluorescence signal hardly changed, indicating that inhibition of P-gp expression is almost independent of NO dose when it increases beyond a certain value. This phenomenon remains to be further studied. On the other hand, caspase-3 protein is regarded as a key effector of apoptosis that will be activated in response to cytotoxic drugs [39,40]. To identify the activation of caspase-3 by the

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

10

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

Fig. 6. (a) Cytotoxicity of DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS plus 30 min irradiation, UCNPs(DOX)@CS-RBS plus 5 min irradiation and UCNPs(DOX)@CS-RBS plus 30 min irradiation at varied concentrations to CT26 cells based on MTT assay. The irradiation was provided by a 980 nm laser with a power density of 0.7 W/cm2. The data are presented as average ± standard deviation (n = 5). (b) Quantitative results of flow cytometric analysis presenting the reversal of P-gp-mediated MDR in CT-26 cells containing UCNPs(DOX) CS-RBS nanospheres under 980 nm-laser irradiation for varied times. Statistical significance: *P < 0.05. (c) The expression levels of caspase-3 (NIR irradiation for 30 min) and P-gp (NIR irradiation for 5 min) in CT26 cells induced by DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS and UCNPs(DOX)@CS-RBS at the same concentration of 50 lg/ mL after incubation for 30 min, analyzed by Western blot. Untreated cells were used as a control, and b-actin was the loading control.

UCNPs(DOX)@CS-RBS nanospheres and confirm the released NO could serve as a P-gp inhibitor, Western blot technique was employed to analyze the expressions of caspase-3 and P-gp. CT26 cells were incubated with DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS

or UCNPs(DOX)@CS-RBS nanospheres (at an equivalent DOX concentration of 1 lg/mL and an equivalent RBS concentration of 1.4 lg/mL) for 1 h. Then the cells were divided into two groups for caspase-3 expression and P-gp expression, respectively. Prior

Fig. 7. Fluorescence images of CT-26 cells with live-dead staining after treatments with different formulations: (a) PBS; (b) DOX; (c) UCNPs(DOX)@CS; (d) UCNPs@CS-RBS plus 30 min; (e) UCNPs(DOX)@CS-RBS plus 5 min; (f) UCNPs(DOX) CS-RBS plus 30 min. The living cells were labeled by calcein AM (green emission), and the dead cells were labeled by propidium iodide (red emission). The scale bar represents 100 lm. (g) Flow cytometry analysis indicating apoptosis and necrosis of CT26 cells induced by DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS plus 30 min irradiation and UCNPs(DOX) CS-RBS plus 30 min irradiation at the same concentration of 50 lg/mL. The numbers inserted denote the cell percentage in each area. The irradiation was provided by a 980 nm laser with a power density of 0.7 W/cm2.

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

11

Fig. 8. (a) Growth of mice-bearing CT26 tumors treated with intravenous injection of PBS, DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS plus 30 min irradiation, UCNPs(DOX)@CSRBS plus 5 min irradiation and UCNPs(DOX)@CS-RBS plus 30 min irradiation via tail vein every other day. Statistical significance: *P < 0.05; **P < 0.01. The data are presented as average ± standard deviation (n = 5). (b) Body weight changes of the tumor-born mice after treatment with different formulations. The data are presented as average ± standard deviation (n = 5). (c) Histology analysis of sectioned tumor tissues treated with H&E staining. The images were taken at 500 magnification. (d) ICP-MS analysis of tumor and five major organs of the mice sacrificed at different time points after injection of UCNPs(DOX) CS-RBS. The data are presented as average ± standard deviation (n = 5).

to Western blot analysis, the cells were irradiated by 980 nm laser for 30 min (for caspase-3 expression) and 5 min (for P-gp expression). The untreated CT26 cells were used as a negative control.

The Western blot data suggest that the expression of caspase-3 protein was substantially up-regulated by UCNPs(DOX)@CS and UCNPs(DOX)@CS-RBS nanospheres compared with the control

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

12

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

(Fig. 6c), indicating that the caspase-3 was readily activated by DOX. The caspase-3 protein expression was only slightly upregulated by DOX and UCNPs@CS-RBS. This indicates that a large dose of NO can induce expression of caspase-3 protein in cancer cells. The expression of P-gp was repressed by both UCNPs@CSRBS and UCNPs(DOX)@CS-RBS nanospheres but not by DOX and UCNPs(DOX)@CS after exposure to the NIR irradiation for 5 min. Reduced expression of P-gp by a small dose of NO will prevent the efflux of anticancer drugs, leading to a higher intracellular DOX concentration and hence enhanced anticancer effects. This is in good accordance with the cell imaging results. Fluorescence microscopy observation is a powerful tool to judge whether the cells are alive or not. The CT26 cells were subjected to different treatments and then stained with calcein AM (denoting living cells) and propidium iodide (denoting dead cells) for observation. For the control group, all the cells in the field of view emitted green fluorescence characteristic of calcein, showing no sign of cell death (Fig. 7a). Only a tiny part of the cells treated with free DOX were dead, as indicated by a small quantity of red emission (Fig. 7b). Much more cells were killed when treated with UCNPs(DOX)@CS (Fig. 7c), since nanoscaled drugs have preferable internalization mechanisms by cancer cells through endocytosed endolysosomal trafficking [41]. A large dose of NO caused considerable cell death (Fig. 7d), confirming the caner-cell killing ability of NO. The UCNPs(DOX)@CS-RBS nanospheres under NIR irradiation exhibited even stronger cell inhibition (Fig. 7e,f). A larger amount of cells was killed when the irradiation lasted for 30 min, in good accordance with the findings mentioned above. Furthermore, the FITC-Annexin V/PI technique was adopted to investigate whether the death of cancer cells caused by the UCNPs(DOX)@CS-RBS nanospheres was induced by apoptosis (Fig. 7g). CT26 cells containing UCNPs(DOX)@CS, UCNPs@CS-RBS or UCNPs(DOX) CS-RBS nanospheres were incubated for 30 min under 980 nm NIR irradiation. Then the cells were stained with FITC-Annexin V/PI and measured on a flow cytometer. The untreated cells were used as controls. The ratio of apoptotic cells (both early and late apoptosis) containing UCNPs(DOX)@CS and UCNPs@CS-RBS nanospheres was 39.20% and 25.69%, respectively. The ratio of necrotic cells was 12.58% and 9.45% in the two cases. For the cells containing UCNPs(DOX)@CS-RBS nanospheres, the apoptotic and necrotic cells accounted for 57.65% and 11.83% of the total, respectively. It is evident that more apoptosis and necrosis were induced by the UCNPs(DOX)@CS-RBS nanospheres at the same drug concentration, demonstrating the synergistic treating effect of NO and DOX. 3.5. In vivo antitumor efficacy In order to evaluate whether the UCNPs(DOX)@CS-RBS nanospheres can be used for tumor therapy, free DOX, UCNPs(DOX) @CS, UCNPs@CS-RBS, UCNPs(DOX)@CS-RBS or PBS as a control was intravenously injected into BALB/c mice bearing a CT26 tumor via tail vein every other day. The tumor volume and body weight of the treated mice were monitored every other day for a total of 12 days (Fig. 8a,b). For the mice treated with UCNPs(DOX)@CSRBS nanospheres, the tumor sites were irradiated by the 980 nm laser for 5 min and 30 min (0.7 W/cm2) respectively at 24 h postinjection. For the mice treated with UCNPs@CS-RBS, the irradiation was applied for 30 min. The average tumor volume of the mice treated with DOX and UCNPs@CS-RBS was smaller than that of the mice treated with PBS at 12 days postinjection. In contrast, the tumor growth was suppressed by UCNPs(DOX)@CS to a larger extent over such a long period. The UCNPs(DOX)@CS-RBS nanospheres exhibited relatively high tumor growth inhibitory efficacy,

as reflected by a 73.5% reduction (5 min NIR irradiation) and a 80.2% reduction (30 min NIR irradiation) in tumor volume after 12 days. The NO released from the UCNPs(DOX)@CS-RBS nanospheres under NIR irradiation could reduce MDR effect and thereby enhance the tumor inhibition ability of DOX. Meanwhile, increased NO dose could kill the tumor cells. The synergistic treating effects agree well with that found in cell experiments. The mice treated with DOX lost 8.6% of the body weight at the end of the 12-day period, implying side effects of DOX on the animals. The mice treated with other formulations showed less body weight loss. The weight loss of the mice treated with UCNPs(DOX)@CS-RBS nanospheres (either 5 min irradiation or 30 min irradiation) was below 4%, indicating that the UCNPs(DOX)@CS-RBS nanospheres were welltolerated without severe side effects. Histology analysis was also employed to evaluate antitumor efficacy of free DOX, UCNPs(DOX)@CS, UCNPs@CS-RBS and UCNPs(DOX)@CS-RBS, with the case of PBS as a control. Hematoxylin and eosin (H&E) stained tumor tissue sections with different treatments had distinct morphologies (Fig. 8c). The tumor cells treated with PBS showed large nucleus and spindle shape, signifying a rapid tumor growth. A similar phenomenon was observed from the treatment of DOX, implying the limited treating effect of free DOX. Pyknotic or fragmented nuclei can be observed in the UCNPs(DOX)@CS, UCNPs@CS-RBS and UCNPs(DOX)@CS-RBS treated groups, especially for the UCNPs(DOX)@CS-RBS treated tumor cells. Moreover, more cell necrosis was found in the mice treated with UCNPs(DOX)@CS-RBS nanospheres plus 5 min and 30 min irradiation as well as UCNPs(DOX)@CS in comparison to those treated with DOX and UCNPs@CS-RBS, as revealed by greater occurrence of missing nuclei (smaller or absent purple staining). The UCNPs(DOX)@CS-RBS nanospheres resulted in severer cell death than the UCNPs(DOX)@CS. The histology analysis suggests that more apoptosis and necrosis was induced by combined dose of DOX and NO. Distribution of the element Y in the tumor-bearing nude mice at different time points after injection of the UCNPs(DOX)@CS-RBS nanospheres was evaluated using ICP-MS (Fig. 8d). The Y concentration in the tumor was much higher than those in five major organs, indicating more accumulation of the nanospheres in the tumor resulted from the enhanced permeability and retention (EPR) effect. The Y concentration in liver and spleen was higher compared with that in stomach, heart or kidney, reflecting a comparatively high affinity of the nanospheres for liver and spleen. Furthermore, the Y concentration inside the tumor and organs began to decrease after 6 h because of metabolism. Compared with those in the organs, the downward trend of Y content in the tumor was less obvious, indicative of longer retention of the UCNPs(DOX) @CS-RBS nanospheres in the tumor. 4. Conclusion To sum up, we have synthesized UCNPs(DOX)@CS-RBS nanospheres with well-defined structure for synergistic tumor therapy. Exposed to deeply-penetrating NIR irradiation, the UCNPs(DOX) @CS-RBS nanospheres can provide on-demand and dosecontrollable NO release. On the other hand, the anticancer drug DOX can be released from the UCNPs(DOX) CS-RBS nanospheres triggered by lowered pH typical of endosomes and lysosomes in cancer cells. The experimental results demonstrate that simultaneous release of NO and DOX exhibits synergistic therapeutic effect on tumors based on two mechanisms: (i) a small dose of NO acts as a P-gp-mediated MDR reversal agent, thus increases intracellular DOX concentration; (ii) a larger dose of NO kills cancer cells in coordination with DOX. Combination of DOX and a larger dose of NO has more potent cell-killing ability. This work provides

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019

L. Tan et al. / Acta Biomaterialia xxx (2017) xxx–xxx

promising drug nanocarriers with high antitumor efficacy for synergistic cancer therapy. Acknowledgements This work was financially supported by the National Natural Science Foundation (Grant No. 51403125, 81671802, 11505110 and 53503083) and the Specialized Research Fund for the Doctoral Program of Higher Education (Grant No. 20130073120087). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.actbio.2017.05. 019. References [1] X. Zhang, Real time and in vivo monitoring of nitric oxide by electrochemical sensors – from dream to reality, Front. Biosci. 9 (2004) 3434–3446. [2] D.A. Wink, J.B. Mitchell, Chemical biology of nitric oxide: insights into regulatory, cytotoxic and cytoprotective mechanisms of nitric oxide, Free Radical Biol. Med. 25 (1998). 434-434. [3] L. Ying, L.J. Hofseth, An emerging role for endothelial nitric oxide synthase in chronic inflammation and cancer, Cancer Res. 67 (2007) 1407–1410. [4] K. Harada, Supriatno, S. Kawaguchi, O. Tomitaro, H. Yoshida, M. Sato, Overexpression of iNOS gene suppresses the tumorigenicity and metastasis of oral cancer cells, In Vivo 18 (2004) 449–455. [5] A.W. Carpenter, M.H. Schoenfisch, Nitric oxide release: Part II. Therapeutic applications, Chem. Soc. Rev. 41 (2012) 3742–3752. [6] D. Fukumura, S. Kashiwagi, R.K. Jain, The role of nitric oxide in tumour progression, Nat. Rev. Cancer 6 (2006) 521–534. [7] S.K. Choudhari, M. Chaudhary, S. Bagde, A.R. Gadbail, V. Joshi, Nitric oxide and cancer: a review, World J. Surg. Oncol. 11 (2013). 118-118. [8] D.A. Wink, Y. Vodovotz, J. Laval, F. Laval, M.W. Dewhirst, J.B. Mitchell, The multifaceted roles of nitric oxide in cancer, Carcinogenesis 19 (1998) 711–721. [9] B.M. Choi, H.O. Pae, S.I. Jang, Y.M. Kim, H.T. Chung, Nitric oxide as a proapoptotic as well as anti-apoptotic modulator, J. Biochem. Mol. Biol. 35 (2002) 116–126. [10] C. Riganti, E. Miraglia, D. Viarisio, C. Costamagna, G. Pescarmona, D. Ghigo, A. Bosia, Nitric oxide reverts the resistance to doxorubicin in human colon cancer cells by inhibiting the drug efflux, Cancer Res. 65 (2005) 516–525. [11] M.-F. Chung, H.-Y. Liu, K.-J. Lin, W.-T. Chia, H.-W. Sung, A pH-responsive carrier system that generates NO bubbles to trigger drug release and reverse Pglycoprotein-mediated multidrug resistance, Angew. Chem. Int. Ed. 54 (2015) 9890–9893. [12] S. Sortino, Photoactivated nanomaterials for biomedical release applications, J. Mater. Chem. 22 (2012) 301–308. [13] H. Nakagawa, Photocontrollable nitric oxide (NO) and nitroxyl (HNO) donors and their release mechanisms, Nitric Oxide 25 (2011) 195–200. [14] L. Tan, S. Liu, X. Li, I.S. Chronakisc, Y. Shen, A new strategy for synthesizing AgInS2 quantum dots emitting brightly in near-infrared window for in vivo imaging, Colloids Surf., B 125 (2015) 222–229. [15] J.V. Garcia, J. Yang, D. Shen, C. Yao, X. Li, R. Wang, G.D. Stucky, D. Zhao, P.C. Ford, F. Zhang, NIR-triggered release of caged nitric oxide using upconversion nanostructured materials, Small 8 (2012) 3800–3805. [16] L. Tan, A. Wan, X. Zhu, H. Li, Nitric oxide release triggered by two-photon excited photoluminescence from engineered nanomaterials, Chem. Commun. 50 (2014) 5725–5728. [17] X. Zhang, G. Tian, W. Yin, L. Wang, X. Zheng, L. Yan, J. Li, H. Su, C. Chen, Z. Gu, Y. Zhao, Controllable generation of nitric oxide by near-infrared-sensitized upconversion nanoparticles for tumor therapy, Adv. Funct. Mater. 25 (2015) 3049–3056. [18] F.W. Flitney, I.L. Megson, J.L.M. Thomson, G.D. Kennovin, A.R. Butler, Vasodilator responses of rat isolated tail artery enhanced by oxygendependent, photochemical release of nitric oxide from iron-sulphurnitrosyls, Br. J. Pharmacol. 117 (1996) 1549–1557.

13

[19] Q. Yang, C. He, Y. Xu, B. Liu, Z. Shao, Z. Zhu, Y. Hou, B. Gong, Yu-M. Shen, Chitosan oligosaccharide copolymer micelles with double disulphide linkage in the backbone associated by H-bonding duplexes for targeted intracellular drug delivery, Polym. Chem. 6 (2015) 1454–1464. [20] L. Tan, A. Wan, H. Li, Q. Lu, Novel quantum dots-carboxymethyl chitosan nanocomposite nitric oxide donors capable of detecting release of nitric oxide in situ, Acta Biomater. 8 (2012) 3744–3753. [21] L. Tan, A. Wan, H. Li, Conjugating S-nitrosothiols with glutathiose stabilized silver sulfide quantum dots for controlled nitric oxide release and nearinfrared fluorescence imaging, ACS Appl. Mater. Interfaces 5 (2013) 11163– 11171. [22] M. Rinaudo, Chitin and chitosan: properties and applications, Prog. Polym. Sci. 31 (2006) 603–632. [23] K.-J. Chen, Y.-L. Chiu, Y.-M. Chen, Y.-C. Ho, H.-W. Sung, Intracellularly monitoring/imaging the release of doxorubicin from pH-responsive nanoparticles using Förster resonance energy transfer, Biomaterials 32 (2011) 2586–2592. [24] A. Montembault, C. Viton, A. Domard, Rheometric study of the gelation of chitosan in aqueous solution without cross-linking agent, Biomacromolecules 6 (2005) 653–662. [25] Y.L. Chiu, S.C. Chen, C.J. Su, C.W. Hsiao, Y.M. Chen, H.L. Chen, H.W. Sung, pHtriggered injectable hydrogels prepared from aqueous N-palmitoyl chitosan: in vitro characteristics and in vivo biocompatibility, Biomaterials 30 (2009) 4877–4888. [26] D. Seyferth, M.K. Gallagher, M. Cowie, The bridging sulfide anion reactivity of Roussin’s red salt, Organometallics 5 (1986) 539–548. [27] Q. Yang, L. Tan, C. He, B. Liu, Y. Xu, Z. Zhu, Z. Shao, B. Gong, Y.-M. Shen, Redoxresponsive micelles self-assembled from dynamic covalent block copolymers for intracellular drug delivery, Acta Biomater. 17 (2015) 193–200. [28] L. Tan, A. Wan, X. Zhu, H. Li, Visible light-triggered nitric oxide release from near-infrared fluorescent nanospheric vehicles, Analyst 139 (2014) 3398– 3406. [29] L. Tan, A. Wan, H. Li, Fluorescent chitosan complex nanosphere diazeniumdiolates as donors and sensitive real-time probes of nitric oxide, Analyst 138 (2013) 879–886. [30] K.M. Camacho, S. Kumar, S. Menegatti, D.R. Vogus, A.C. Anselmo, S. Mitragotri, Synergistic antitumor activity of camptothecin-doxorubicin combinations and their conjugates with hyaluronic acid, J. Control. Release 210 (2015) 198–207. [31] Z.J. Wang, F. Tao, L.Z. Yao, W.L. Cai, X.G. Li, Selected synthesis of cubic and hexagonal NaYF4 crystals via a complex-assisted hydrothermal route, J. Crystal Growth 290 (2006) 296–300. [32] G. Wang, W. Qin, J. Zhang, L. Wang, G. Wei, P. Zhu, R. Kim, Controlled synthesis and luminescence properties from cubic to hexagonal NaYF4:Ln3+ (Ln = Eu and Yb/Tm) microcrystals, J. Alloy. Compd. 475 (2009) 452–455. [33] Z.G. Chen, H.L. Chen, H. Hu, M.X. Yu, F.Y. Li, Q. Zhang, Z. Zhou, T. Yi, C. Huang, Versatile synthesis strategy for carboxylic acid-functionalized upconverting nanophosphors as biological labels, J. Am. Chem. Soc. 130 (2008) 3023–3029. [34] L.-Q. Xiong, Z.-G. Chen, M.-X. Yu, F.-Y. Li, C. Liu, C.-H. Huang, Synthesis, characterization, and in vivo targeted imaging of amine-functionalized rareearth up-converting nanophosphors, Biomaterials 30 (2009) 5592–5600. [35] American National Standard for Safe Use of Lasers, ANSI Z136. 1-2000, American National Standard Institute, Orlando, FI 2000. [36] S. Griveau, C. Dumezy, P. Goldner, F. Bedioui, Electrochemical analysis of the kinetics of nitric oxide release from two diazeniumdiolates in buffered aqueous solutions, Electrochem. Commun. 9 (2007) 2551–2556. [37] H. Xu, D. Yang, C. Cai, J. Gou, Y. Zhang, L. Wang, H. Zhong, X. Tang, Dualresponsive mPEG-PLGA-PGlu hybrid-core nanoparticles with a high drug loading to reverse the multidrug resistance of breast cancer: an in vitro and in vivo evaluation, Acta Biomater. 16 (2015) 156–168. [38] L. Wang, X. Lin, J. Wang, Z. Hu, Y. Ji, S. Hou, Y. Zhao, X. Wu, C. Chen, Novel insights into combating cancer chemotherapy resistance using a plasmonic nanocarrier: enhancing drug sensitiveness and accumulation simultaneously with localized mild photothermal stimulus of femtosecond pulsed laser, Adv. Funct. Mater. 24 (2014) 4229–4239. [39] G. Tian, X. Zhang, X. Zheng, W. Yin, L. Ruan, X. Liu, L. Zhou, L. Yan, S. Li, Z. Gu, Y. Zhao, Multifunctional RbxWO3 nanorods for simultaneous combined chemophotothermal therapy and photoacoustic/CT imaging, Small 10 (2014) 4160– 4170. [40] Y. Lazebnik, N.A. Thornberry, Caspases: enemies within, Science 281 (1998) 1312–1316. [41] Y. Zhao, D.Y. Alakhova, A.V. Kabanov, Can nanomedicines kill cancer stem cells?, Adv Drug Deliv. Rev. 65 (2013) 1763–1783.

Please cite this article in press as: L. Tan et al., Controllable release of nitric oxide and doxorubicin from engineered nanospheres for synergistic tumor therapy, Acta Biomater. (2017), http://dx.doi.org/10.1016/j.actbio.2017.05.019