Biomaterials 39 (2015) 67e74

Contents lists available at ScienceDirect

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

Magnetic nanoparticle clusters for photothermal therapy with near-infrared irradiation Shun Shen a, 1, Sheng Wang a, c, 1, Rui Zheng b, Xiaoyan Zhu d, Xinguo Jiang a, Deliang Fu c, Wuli Yang b, * a

School of Pharmacy & Key Laboratory of Smart Drug Delivery, Fudan University, Shanghai 201203, China State Key Laboratory of Molecular Engineering of Polymers & Department of Macromolecular Science, Fudan University, Shanghai 200433, China Pancreatic Disease Institute, Department of Pancreatic Surgery, Huashan Hospital, Shanghai Medical College, Fudan University, Shanghai 200040, China d Department of Integrative Oncology, Shanghai Cancer Center, Department of Oncology, Shanghai Medical University, Fudan University, Shanghai 200032, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 August 2014 Accepted 19 October 2014 Available online

In this study, the photothermal effect of magnetic nanoparticle clusters was firstly reported for the photothermal ablation of tumors both in vitro in cellular systems but also in vivo study. Compared with individual magnetic Fe3O4 nanoparticles (NPs), clustered Fe3O4 NPs can result in a significant increase in the near-infrared (NIR) absorption. Upon NIR irradiation at 808 nm, clustered Fe3O4 NPs inducing higher temperature were more cytotoxic against A549 cells than individual Fe3O4 NPs. We then performed in vivo photothermal therapy (PTT) studies and observed a promising tumor treatment. Compared with PBS and individual magnetic Fe3O4 NPs by NIR irradiation, the clustered Fe3O4 NPs treatment showed a higher therapeutic efficacy. The treatment effects of clustered Fe3O4 NPs with different time of NIR illumination were also evaluated. The result indicated that a sustained high temperature generated by NIR laser with long irradiation time was more effective in killing tumor cells. Furthermore, histological analysis of H&E staining and TUNEL immunohistological assay were further employed for antitumor efficacy assessment of PTT against A549 tumors. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Magnetic nanoparticle clusters Fe3O4 Tumor Photothermal therapy

1. Introduction Nanomaterials have been extensively used in biomedical research during the past twenty years [1e4]. Especially in recent years, more and more studies focused on photo-absorbing nanoagents, and photothermal therapy (PTT) employing photoabsorbers can convert optical energy into thermal energy to kill cancer cells without affecting healthy tissues [5,6]. Compared with radiotherapy, chemotherapy and surgical management, PTT is less invasive, controllable and highly efficient. Our and other research groups have developed a large number of nanomaterials as PTT agents, such as gold-based nanomaterials [5,7e9], carbon nanotubes and graphene [10,11], all of which show strong optical absorbance in the near-infrared (NIR) tissue optical transparency window. The previous results demonstrated that photoabsorberbased therapy might be a promising approach for cancer therapy,

* Corresponding author. Tel.: þ86 21 65642385. E-mail address: [email protected] (W. Yang). 1 Both authors contributed equally to this work. http://dx.doi.org/10.1016/j.biomaterials.2014.10.064 0142-9612/© 2014 Elsevier Ltd. All rights reserved.

which could effectively reduce tumor growth and enhance survival [12e15]. Even so, the potential toxicity induced by photothermal agents (especially for carbon nanotubes or graphene, and so on), is still an unresolved debate [16,17], which will inevitably limit future clinic applications of PTT. These non-degradable or slowly degradable nanoparticles easily accumulate in bodily organs, resulting in increased oxidative stress, inflammatory cytokine production and cell death [18]. Therefore, it is significant to explore a bio-safety and biodegradable photoabsorber for the photothermal ablation of cancer with NIR irradiation. Magnetic iron oxides nanoparticles (namely Fe3O4 NPs) have received tremendous attention for their excellent magnetic, biocompatible and potentially non-toxic properties [19,20]. Fe3O4 NPs, which can be manipulated and controlled by an external magnetic field, are additional important materials and have been employed in many areas, including biology, pharmaceuticals and diagnostics. Moreover, iron is a nutrient and readily metabolized by cellular regulation using the transferrin pathway. Thus, Fe3O4 NPs are easily degradable and passes in and out of cells across the plasma membrane [21]. For the acknowledged advantages of Fe3O4 NPs, they are extremely suitable for in vivo applications. In addition,

68

S. Shen et al. / Biomaterials 39 (2015) 67e74

Fe3O4 NPs are exceptional materials for hyperthermia treatment of tumors [22,23]. Under an alternating magnetic field (AMF), magnetic hyperthermia of Fe3O4 NPs is produced via dipole relaxation, which can be employed to destroy tumor cells because these cells are more sensitive to temperatures in excess of ca. 41  C than their normal counterparts [24]. Although magnetic hyperthermia has been clinically used [25e28], the technique requires high current and voltage due to large air volume within the applied field in which energy cannot be easily focused [29]. Very recently, NIR light induced photothermal effect for Fe3O4 NPs has been studied and it exhibits good photothermal converting efficiency. For example, Chu et al. applied individual Fe3O4 NPs with high concentration for the photothermal ablation of cancer by NIR laser irradiation [29]. Considering extra-high-dose magnetite might generate potential toxicity to the body, the number of usable elements should also be severely limited. Thus, Fe3O4 NPs used for PTT must be improved, mainly involving two routes. One is to modify NIR light-absorbing materials onto the surface of magnetic NPs, another is to concentrate the magnetic NPs. Newly, it has been shown that the clustering of magnetic NPs induces a significant increase in the magnetic moment, and consequently, the clustered magnetic NPs have a much higher relatively high saturation magnetization and specific absorption rate (SAR) than individual magnetic NPs [30e33]. Hayashi et al. have utilized the unique merit of magnetite clusters for the magnetic hyperthermia therapy of tumors by AMF [32,33]. More significantly, previous studies have also indicated that metallic NPs clusters can induce a red-shift in the light absorption spectra [34,35], which enhances the light absorbance in the NIR region, expanding new application fields for metallic NPs clusters to be utilized as photosensitizers in the NIR PTT. Therefore, numerous papers have been reported the photothermal ablation of tumors by utilizing aggregation-induced enhanced photothermo (AIEP) of metallic NPs clusters [36,37], such as gold NPs aggregation. Until now, to the best of our knowledge, the papers on the study of photothermal effect of magnetite clusters were rarely reported, let alone for the use of magnetite clusters with NIR irradiation to destroy tumor in vivo. Herein, we present the synthesis of magnetic colloidal nanocrystal clusters that effectively produce heat in response to harmless NIR irradiation. The biomedical parameters were not only assessed in vitro in cellular systems but also in vivo study. The results proved that the photothermal effects of clustered magnetic NPs for the treatment of A549 (Human lung adenocarcinoma epithelial cell line) tumor were better than those obtained by single crystal magnetic NPs.

temperature in nitrogen atmosphere, and NaOH solution (10 M, 50 mL) was then slowly dropped into the flask from the dropping funnel in 1 h. After that, the mixture was continuously stirred at room temperature for 1 h, which was then stirred at 90  C for another 2 h and cooled to room temperature under continuous stirring. The whole process was in the nitrogen atmosphere. The products were washed with deionized water three times, collected with the help of a magnet. After that, nitric acid solution (2 M, 100 mL) was added into the above products, which were then washed with deionized water several times and collected with the magnet, until the pH value of the supernatant is neutral. Then 100 mL (0.3 M) of trisodium citrate was added into the resulting products, the mixture was stirred for 30 min at 90  C in the nitrogen atmosphere. Then it was cooled to room temperature, transferred into a beaker and collected with the magnet, and 20 mL of deionized water and a lot of acetone were added into the beaker to wash the products twice. Finally, the obtained products were dispersed in 50 mL of deionized water, and stirred at 80  C for a long time to remove acetone.

2. Materials and methods

2.7. Cell viability assay

2.1. Materials

A549 cells were seeded in 96-well plates with a density of 104 cells/well, and allowed to adhere for 24 h prior to assay. The cells were exposed to the clustered magnetic NPs or individual magnetic NPs with the same concentration of 50 mg/mL, respectively. The cells were or were not irradiated by NIR laser light at a power density of 5 W/cm2 with different illumination time from 60 to 180 s. After the cells were incubated at 37  C for another 24 h, they were incubated with 0.5 mg/mL MTT in DMEM for 4 h in dark and then mixed with dimethyl sulfoxide after the supernatant was removed. The OD value at 570 nm was read using the microplate reader (Synergy TM2, BIO-TEK Instruments Inc. USA). Cell viability was determined by the percentage of OD value of the study group over the control group. Afterward, cells were co-stained by a mixture of Calcein AM and PI solution for 20 min. The samples were subjected to observe using a ZEISS LSM710 live cell confocal laser imaging System (Carl Zeiss, German).

Iron (III) chloride (FeCl3$6H2O), iron (II) chloride (FeCl2$4H2O), trisodium citrate dehydrate (C8H5Na3O7$2H2O), sodium acetate anhydrous (NaOAc), ethylene glycol (EG), acetone, sodium hydroxide (NaOH), and nitric acid (HNO3) were purchased from Shanghai Chemical Reagents Company (China). MTT (3-(4, 5-dimethylthiazol2-yl)-2,5-diphenyltetrazolium bromide) assay, and other biological reagents were purchased from Invitrogen Corp. Dulbecco's modified Eagle medium (DMEM), Fetal bovine serum (FBS), Penicillin-Streptomycin solution and Trypsin-EDTA solution were purchased from Gibco (Tulsa, OK, USA). All the other chemicals were of analytical grade, and purified water was produced by a Millipore water purification system.

2.3. Synthesis of clustered magnetic NPs The clustered magnetic NPs (magnetic particles) were prepared by a modified solvothermal reaction [39]. Briefly, FeCl3$6H2O (1.028 g), C8H5Na3O7$2H2O (0.24 g), and NaOAc (1.2 g) were first dissolved in 20 mL of EG under vigorous stirring for 0.5 h. The resulting solution was then transferred into a Teflon-lined stainless-steel autoclave with a capacity of 50 mL. The autoclave was sealed and heated at 200  C for 10 h. Then it was cooled to room temperature. The asprepared black products were washed with ethanol and deionized water several times, and collected with a magnet. The final products were dispersed in 10 mL of ethanol for the further use. 2.4. Characterization Ultravioletevisible (UVevis) spectra were performed using a PerkineElmer Lambda 750 spectrophotometer. Transmission electron microscopy (TEM) images were obtained on a Tecnai G2 20 TWIN transmission electron microscope. Magnetic characterization was carried out with a vibrating sample magnetometer (VSM) on a Model 6000 physical property measurement system (Quantum, USA) at 300 K. X-ray diffraction (XRD) measurements were recorded on a X'pert PRO diffractometer to determine the composition of Fe3O4 particles. All the diffraction peaks in the XRD patterns were indexed and assigned to the typical cubic structure of Fe3O4 (JCPDS 75-1609). 2.5. Cell lines and culture conditions A549 cells obtained from Chinese Academy of Sciences Cells Bank, Shanghai, China, were routinely cultured in RPMI-1640 cell medium supplemented with 10% FBS, 100 U/mL penicillin and 100 mg/mL streptomycin, at 37  C in 5% CO2 and 95% air atmosphere with >95% humidity. All experiments were performed on cells in the logarithmic phase of growth. 2.6. NIR-heating effect of magnetite NPs in solution The individual magnetic NPs or clustering of magnetic NPs stock solution at 1 mg/mL was diluted to the different concentrations (10e80 mg/mL) and 200 mL of aliquots were deposited into wells of a 48-well cell culture plate. Wells were illuminated by an 808-nm continuous-wave NIR laser (Changchun New Industries Optoelectronics Technology, Changchun, China; fluence: 5 W/cm2, spot size: 5 mm) with the different exposure time from 60 to 180 s. Pre- and postillumination temperatures were taken by thermocouple.

2.2. Synthesis of individual magnetic NPs The individual magnetic NPs were prepared by chemical coprecipitation method [38]. In a typical recipe, FeCl3$6H2O (10.81 g, 0.04 mol) was dissolved in 50 mL deionized water. The resulting solution was transferred into a three-necked flask (250 mL), which was equipped with a mechanical stirrer, a dropping funnel, and a nitrogen inlet. Then 3.98 g of FeCl2$4H2O (0.02 mol) was also dissolved in another 50 mL deionized water, and the resulting solution was transferred into the abovementioned three-necked flask. The as-prepared mixture was then stirred at room

2.8. Flow cytometry analysis To investigate the photothermal ablation of A549 cells by clustered Fe3O4 NPs without or with the 808 nm laser irradiation using flow cytometry, clustered Fe3O4 NPs dispersion (100 mg/mL in PBS solution) was added to a 12-well cell culture plate containing A549 cells with a density of 5  105 cells/well in RPMI1640 medium supplemented with 10% FBS and 1% penicillinestreptomycin, then the A549 cells were incubated for 4 h at 37  C in 5% CO2 and 95% air atmosphere

S. Shen et al. / Biomaterials 39 (2015) 67e74 with >95% humidity. Then, a group of cell solution was exposed to an 808 nm laser at a power intensity of 5 W/cm2 for 180 s. After the laser irradiation treated, A549 cells were stained with Annexin-V-FITC and PI (Becton Dickinson, Mountain View, CA, USA), and analyzed by BD FACSAria I flow cytometry with excitation at 488 nm. Fluorescent emission of FITC was measured at 515e545 nm and that of DNA-PI complexes at 564e606 nm. Compensation was used wherever necessary. 2.9. In vivo photothermal treatment Male Balb/c mice, aging 4e5 weeks and weighing 18.8 ± 1.9 g, were supplied by the Department of Experimental Animals, Fudan University (Shanghai, China), and maintained under standard housing conditions. All animal experiments were carried out in accordance with guidelines evaluated and approved by the ethics committee of Fudan University. Through a subcutaneous injection of 2  106 cells suspended in 100 mL PBS into the flank region, the mice bearing A549 tumor were successfully established. The longest dimension and the shortest dimension of tumors were monitored by digital calipers. Once tumors were measured ~5.0 mm in longest dimension, mice were randomized into 4 treatment groups (n ¼ 5 per group): PBS with laser treated (Group I), individual Fe3O4 NPs with laser treated (Group II), clustered Fe3O4 NPs with laser treated (Group III and IV for different NIR illumination time). Each type of dispersion (2 mg/mL, injection volume of 25 mL) was intratumorally injected into mice. Two hours after injection, the 808-nm continuous-wave NIR laser was applied to irradiate tumors under a power intensity of 5 W/cm2 (Group I, II and IV for 180 s, Group III for 120 s). The temperature was recorded using an infrared camera thermographic system (InfraTec, VarioCAM®hr research, German). After treatment, all mice were returned to animal housing and tumor volumes were tracked every 2 days by digital caliper measurements. The tumor volume was calculated by the formula V ¼ 1/2(L$W2), where L is length (longest dimension) and W is width (shortest dimension). At 19th day, all the animals were euthanized, and the tumors were dissected and weighed. According to our previous reports [40e43], some histological examinations included hematoxylin and eosin (H&E) stains and TUNEL assays were carried out for the assessment of therapy efficacy.

69

2.10. The retention of the magnetic NPs in the tumor sites According to the previous literature [29], the retention of the Fe3O4 NPs in the tumor sites was also studied. Once tumors were measured ~5.0 mm in longest dimension, mice were randomized into 2 groups (n ¼ 9 per group): individual Fe3O4 NPs with laser treated (Group I), clustered Fe3O4 NPs with laser treated (Group II). Each type of dispersion (2 mg/mL, injection volume of 25 mL) was intratumorally injected into mice. Two hours after injection, the 808-nm continuous-wave NIR laser was applied to irradiate tumors under a power intensity of 5 W/cm2 for 180 s. They were sacrificed on day 1, 5 and 19 after injection, respectively. Three mice for one group were randomly selected and sacrificed at each time point and the tumors were extracted immediately after. The tumors were weighed and washed with distilled water. Afterward, the tissue was mixed with 0.5 mL of aqua regia and incubated at room temperature. Two days later, the samples were centrifuged and the upper clear solutions were collected and adjusted with distilled water. The concentrations of iron atoms in the solutions were performed by flame atomic absorption spectroscopy (Hitachi, Japan). An iron hollow cathode lamp was used as a light source. The excitation current, slit width and absorption wavelength of iron were set at 15 mA, 0.2 nm and 248.3 nm, respectively. 2.11. Statistical analysis Unpaired student's t test was used for between two-group comparison and oneway ANOVA with Fisher's LSD for multiple-group analysis. A probability (P) less than 0.05 was considered statistically significant. Results were expressed as mean ± standard deviation (SD) unless otherwise indicated.

3. Results and discussion 3.1. Characterization of magnetic NPs A TEM image shows the dimension of our obtained individual Fe3O4 NPs is about 15 nm (Fig. 1a), while that of the clustered Fe3O4

Fig. 1. TEM images of (a) individual magnetic Fe3O4 NPs and (b) clustered magnetic Fe3O4 NPs. (c) The XRD pattern of individual and clustered magnetic Fe3O4 NPs. (d) Magnetization loops of individual and clustered magnetic Fe3O4 NPs.

70

S. Shen et al. / Biomaterials 39 (2015) 67e74

NPs has a nearly uniform size of ca. 225 nm with spherical shape (Fig. 1b). A TEM image of the clustered Fe3O4 NPs at higher magnification (Fig. S1) illustrates that the magnetite conglobulations are loose clusters, and the particles are composed of nanocrystals with the size of about 5e10 nm. According to the previously reported literature [44], these nanocrystals are connected with each other by amorphous matrix as a bridge. The samples of individual and clustered Fe3O4 NPs both show similar typical XRD patterns of magnetite (JCPDS no. 75-1609; Fig. 1c), which indicates that the nanocrystalline structure of the clustered Fe3O4 NPs is the same as individual Fe3O4 NPs. The magnetic property characterization (Fig. 1d) shows that both individual and clustered Fe3O4 NPs have superparamagnetic property and high magnetization with saturation value of 58.69 emu/g for the individual Fe3O4 NPs, and 63.70 emu/g for the clustered Fe3O4 NPs, respectively. 3.2. Photothermal effect of magnetic NPs Fig. 2a illustrates the UVevis spectra of the individual and clustered Fe3O4 NPs dispersed in deionized water. The individual Fe3O4 NPs have a very weak absorption in the NIR spectroscopy from 600 to 1000 nm. Surprisingly, the clustered Fe3O4 NPs show an obvious absorption band containing a maximum at ca. 420 nm. Moreover, compared with individual Fe3O4 NPs, the clustered Fe3O4 NPs also resulted in a significant ~3.6 fold increase in the NIR absorption at 808 nm for the aggregation of nanoparticles. Thus, the clustered Fe3O4 NPs that have a stronger absorption in the NIR spectroscopy may be conducive to destroying tumors in vivo as an effective thermal generator due to its in-depth tissue penetration. The photothermal effects of the PBS solutions of the individual and clustered Fe3O4 NPs were firstly examined and compared in vitro by NIR laser irradiation (l ¼ 808 nm, 5 W/cm2). As seen from Fig. 2b, the clustered Fe3O4 NPs were more efficient than the individual Fe3O4 NPs in inducing a temperature increase with the same concentration and exposure time. For example, the temperature raised by the clustered Fe3O4 NPs solution could reach 51.4 ± 1.2  C with the NPs concentration of 50 mg/mL and NIR exposure time of 180 s, while that of the individual Fe3O4 NPs only reached 42.0 ± 1.5  C. Compared with the previous report using individual Fe3O4 NPs with high concentration [29], the higher temperature was easily acquirable by clustered Fe3O4 NPs with lower concentration and

shorter NIR irradiation time in our study, which was beneficial for the safe application in vivo. In addition, the NIR-heating effects of the individual and clustered Fe3O4 NPs were both time and concentration-dependent. The NPs with higher concentration performed markedly better than that of low concentration. Likewise, the longer the laser irradiation time, the higher the solution temperature is. The results indicate that optoelectronic excitation of clustered Fe3O4 NPs by NIR irradiation can occur rapidly, and the extra energies are efficiently transferred into molecular vibration modes to generate significant amounts of thermal energy, which could be useful as photothermal mediators. In order to validate the feasibility of the Fe3O4 NPs in biomedicine, the nanomaterials of individual or clustered Fe3O4 NPs were administered to A549 cells in 96-well plates, respectively, and each group was subdivided into groups of with and without NIR laser exposure. The cell counting Kit-8 (CCK-8) assay was employed for the quantitative evaluation of cell viability [45]. The viability of untreated cells was assumed to be 100%. The cell viability remained about 92.63% when they were incubated with individual or clustered Fe3O4 NPs at higher concentration of 500 mg/mL for 24 h (Fig. S2, Supporting information), which indicated our magnetic nanomaterials had no inherent toxicity. Subsequently, the magnetic nanomaterials were applied for the photothermal ablation of cancer cells. The cytotoxic effects by NIR irradiation were shown in Fig. 3a. About 8.9%, 33.5% and 72.8% of cells were killed by clustered Fe3O4 NPs at the concentration of 50 mg/mL with a power density of 5 W/cm2 for different illumination time of 60, 120 and 180 s, respectively. And only 0.8%, 3.5% and 14.5% of cells were killed by individual Fe3O4 NPs. In addition, direct irradiation of the cells remained close to 100%, exhibiting no effect on cell viability. This is mainly because the low light absorption by natural endogenous cytochromes of these cells caused minimal temperature elevation accounts for their high survival [5]. The CCK-8 experimental result indicated that clustered Fe3O4 NPs inducing higher temperature were more cytotoxic against A549 cells than individual Fe3O4 NPs by NIR irradiation. These results also proved that cancer cells could be effectively killed by clustered Fe3O4 NPs with lower concentration and shorter NIR irradiation time compared with previous report [29]. Fluorescence images of calcein AM (green, live cells) and PI (red, dead cells) co-stained cells after PTT were further confirmed the effectiveness of PTT using clustered Fe3O4 NPs (Fig. 3b). A large number of killed A549 cells were observed by

Fig. 2. (a) UVevis spectra of aqueous dispersion of individual and clustered magnetic Fe3O4 NPs. (b) The photothermal effects of the PBS solutions of individual and clustered magnetic Fe3O4 NPs with the concentration of 0e80 mg/mL were examined ex vitro by irradiating NIR (l ¼ 808 nm, 5 W/cm2) for 60e180 s. Temperature was measured by thermocouple.

S. Shen et al. / Biomaterials 39 (2015) 67e74

71

Fig. 3. (a) Relative viabilities of PBS, individual and clustered magnetic Fe3O4 NPs with the concentration of 50 mg/mL treated A549 cells without or with NIR laser irradiation (l ¼ 808 nm, 5 W/cm2) for 60e180 s. (b) Confocal fluorescence images of calcein AM (green, live cells) and propidium iodide (red, dead cells) co-stained A549 cells treated by clustered magnetic Fe3O4 NPs at the concentration of 50 mg/mL with NIR laser irradiation (l ¼ 808 nm, 5 W/cm2) for (1) 0 s, (2) 60 s, (3) 120 s, (4) 180 s. (c) Flow cytometry graphs of A549 cells treated by clustered magnetic Fe3O4 NPs at the concentration of 50 mg/mL without or with NIR laser irradiation (l ¼ 808 nm, 5 W/cm2) for 180 s. The treated A549 cells were double stained by Annexin-V-FITC/PI and analyzed by flow cytometry. Quadrant I doesn't representatively correspond to live or damaged cells; Quadrant II representatively corresponds to the population of dying cells. The membranes of these cells had been compromised so PI could penetrate and then hybridize with their DNAs; Quadrant III represents the population of live cells. Since the live cells have intact cell membranes, PI could not stain their DNAs; The population of dead cells, or stained DNAs, exhibits a high PI and small forward scattering signal as shown in quadrant IV. These are free DNAs, no longer enclosed within a membrane. The analysis data reveal that cells targeted by the clustered magnetic Fe3O4 NPs combined with NIR lighting display irreversible damage, as shown by the larger population in quadrant IV, where the cellular membrane is broken to such an extent that the cell can no longer function nor recover from the damage [46].

72

S. Shen et al. / Biomaterials 39 (2015) 67e74

clustered Fe3O4 NPs exposed to laser illumination with a power density of 5 W/cm2 for 180 s. Once the irradiation time was reduced to 120 s, most of cells remained survival. In addition, cells incubated with clustered Fe3O4 NPs for irradiation time of 60 s or without NIR laser treatment both showed limited damage, which could be attributable to insufficiently lethal temperature rise. The fluorescence imaging result was consistent with the CCK-8 result. To make clear the cell death mode after photothermal treatment, an Annexin-V-FITC/PI method was conducted by flow cytometry. Fig. 3c showed the flow cytometry graphs of the cells by clustered Fe3O4 NPs with and without laser irradiation. Annexin-VFITC emission signal was plotted on the x-axis, while PI emission signal was plotted on the y-axis. The graphs quadrants were divided into four quadrants. The quantities of living cells, apoptotic cells and necrotic cells were determined by the percentage of Annexin V/PI, Annexin Vþ/PI and Annexin Vþ/PIþ. Almost no apoptosis or necrosis cells were observed in the group of no NIR treatment. After laser irradiation, the apoptosis rate of cells reached 67.7%, while the necrosis rate was only 6.4%. The flow cytometry data revealed that cells treated by clustered Fe3O4 NPs displayed irreversible damage upon photothermal treatment, and their cellular membranes were broken to such an extent that the cells could no longer function nor recover from the damage. Therefore, the mechanism of in vitro PTT is mainly triggered by cell apoptosis, not necrosis. 3.3. Photothermal ablation of tumor in vivo Encouraged by the effective photothermal outcome in vitro, we next performed a pilot in vivo photothermal treatment study. In this work, A549 tumor-bearing mice were established by subcutaneous injection of 2  106 A549 cells suspended in RMPI-1640 medium into the flank region. Once the tumors were measured ca. 5.0 mm in longest dimension, mice were randomized into four treatment groups (n ¼ 5 per group) mentioned in the experimental section. There was no statistical difference among group mean tumor volumes at the onset of treatment (P > 0.05). Twenty-four hours before irradiation, mice were then injected intratumorally with 25 mL of treatment solution containing 50 mg of individual or clustered Fe3O4 NPs in Group II, III and IV, leaving light gray scars on the original tumor sites. The tumors were illuminated with an 808nm NIR laser (5 W/cm2; spot size, 5 mm) for 120 s in Group III and for 180 s in Group II, IV. No mice died during the course of therapy.

Fig. 4 showed the infrared thermal images of tumor surface in Group I, II and IV mice. In Group IV, the maximum temperature of the tumor surface rapidly increased 21.6  C to reach 55.9  C attributing to the clustered Fe3O4 NPs with a good NIR-heating property. In addition, as can be seen from the figure, the increase in NIR irradiation time from 120 s to 180 s could not significantly increase the temperature of tumor surface, showing the temperature reached a balance after two minutes of NIR irradiation. At that time, the NIR laser didn't stop working, and kept irradiating for one minute to maintain a high temperature in the tumor site to effectively kill A549 cells. In Group II, the maximum temperature of the tumor surface reached 50.3  C, which was inferior to Group IV due to the weak photothermal effects. In contrast, less temperature changed in Group I by PBS with NIR treated, and increased only 4.2  C. Furthermore, in order to investigate the effect of PTT using clustered Fe3O4 NPs with different NIR irradiation time, the 808nm continuous-wave NIR laser was employed to irradiate tumors under a power intensity of 5 W/cm2 for 120 s in Group III. Compared with Group IV, the duration time of high temperature in Group III reduced one minute. As we know, high temperature can lead to cell apoptosis, even instantaneous coagulative necrosis and irreversible cell death [47]. Tumor sizes after diverse treatments were measured every two days (Fig. 5a). The tumors of Group I, II and III grew rapidly. There was no statistically significant difference in final tumor size (P > 0.05), indicating that tumor growth was not affected by PBS or individual Fe3O4 NPs with NIR irradiation for 180 s, and clustered Fe3O4 NPs with NIR irradiation time of 120 s. On the contrary, a statistically significant of tumor growth was observed in mice treated with clustered Fe3O4 NPs plus NIR laser for illumination time of 180 s (P < 0.05). Effects were most pronounced in the Group IV, where mean tumor volume at 19 days was reduced from 955.3 mm3 in controls to 222.8 mm3 (P ¼ 0.0139). At 19th days, mice were euthanized, tumors were excised and weighted (Fig. 5). The weights of tumors for Group I, II, III, IV and V were 0.5311 ± 0.0599 g, 0.4369 ± 0.0452 g, 0.4072 ± 0.0479 g and 0.2297 ± 0.0333 g, respectively (Fig. 5d). Obviously, the antitumor efficiency of Group IV was particularly prominent and was superior to all the other groups (P < 0.01), showing an inhibition rate of 56.75%, 47.43% and 43.59% compared with Group I, II and III. However, there was no statistically significant difference in Group III compared with Group I, II, indicating that a lack of sustained high temperature would hardly effectively kill the tumor cells.

Fig. 4. Infrared thermal images of PBS, individual and clustered magnetic Fe3O4 NPs with the concentration of 100 mg/mL injected A549 tumor sample under NIR laser irradiation (l ¼ 808 nm, 5 W/cm2) for 0e180 s.

S. Shen et al. / Biomaterials 39 (2015) 67e74

73

Fig. 5. (a) Tumor growth curves of different groups after treatment. Tumor sizes were measured every 2 days. Means and standard errors are shown. The groups of PBS, individual magnetic Fe3O4 NPs with NIR irradiation for 180 s and clustered magnetic Fe3O4 NPs with NIR irradiation for 120 s were statistically identical. The antitumor efficiency of clustered magnetic Fe3O4 NPs with NIR irradiation for 180 s was particularly prominent and was superior to all the other groups (P < 0.05). (b) Body weight changes were recorded after therapy every 2 days. Means and standard errors are shown. (c) Photograph of tumors after excision from PBS, individual and clustered magnetic Fe3O4 NPs under NIR irradiation. (d) Tumor weights of each group. n.s. no significant difference, *P < 0.05, **P < 0.01.

Accordingly, histological analysis of H&E staining illustrated that the tumors in test groups exhibited a scarlike structure containing numerous collagen bundles (Fig. S3), especially for Group IV. However, a lot of tumor cells were observed in Group I, II and III. The antitumor efficacy of PTT against A549 tumor was also evaluated by the TUNEL immunohistological assay (Fig. S4), which exclusively detected apoptosis in tumor tissues. TUNEL signals (brown fluorescence) were observed in Group II, III and IV. Tumor tissues treated with clustered Fe3O4 NPs plus laser irradiation for 180 s showed more TUNEL signals than did tumors treated with PBS and individual Fe3O4 NPs plus laser irradiation for 180 s, and clustered Fe3O4 NPs plus laser irradiation for 120 s, indicating that photothermal ablation of clustered Fe3O4 NPs with long time NIR treatment was more efficient for tumor therapy. Last but not the least, to investigate the metabolism of the Fe3O4 NPs after therapy, the iron atoms captured in the tumors were determined by a flame atomic absorption spectroscopy. The analytical results exhibited that 98.7%, 77.8%, 53.5% of individual Fe3O4 NPs or 99.1%, 72.3%, 46.7% of clustered Fe3O4 NPs in the tumors for 1, 5 and 19 days postinjection with NIR irradiation with 180 s (Fig. S5). The analytical result showed that our Fe3O4 NPs could be slowly cleared from the tumors after therapy. Besides, the high Fe3O4 NPs uptake in the first few days is particularly important for photothermal cancer treatment if repeated NIR irradiation therapy was needed. 4. Conclusions The individual and clustered Fe3O4 NPs with the same nanocrystalline structure were successfully synthesized. The clustered Fe3O4 NPs were more efficient than the individual Fe3O4 NPs in inducing a temperature increase. Significantly greater cell killing was detected when A549 cells incubated with clustered Fe3O4 NPs

were irradiated with NIR illumination. The flow cytometry analysis demonstrated that the mechanism of in vitro PTT was mainly triggered by cell apoptosis, not necrosis. Using an A549 tumor model, a higher therapeutic efficacy was obtained by the NIRinduced hyperthermia of the clustered Fe3O4 NPs. We believe that our study has the potential in the future clinical application of cancer therapies for the clustered Fe3O4 NPs with good photothermal effect and excellent biological safety. Acknowledgment This work was supported by the National Natural Science Foundation of China (grant, No. 51273047, 81301974 and 81102847). W. Yang thanks Prof. Hao Zhang (Jilin University) for the discussion about the properties of Fe3O4. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2014.10.064. References [1] Karakoti AS, Das S, Thevuthasan S, Seal S. PEGylated inorganic nanoparticles. Angew Chem Int Ed 2011;50:1980e94. [2] Yang K, Hu L, Ma X, Ye S, Cheng L, Shi X, et al. Multimodal imaging guided photothermal therapy using functionalized graphene nanosheets anchored with magnetic nanoparticles. Adv Mater 2012;24:1868e72. [3] Zhao F, Zhao Y, Liu Y, Chang XL, Chen CY, Zhao YL. Cellular uptake, intracellular trafficking, and cytotoxicity of nanomaterials. Small 2011;7:1322e37. [4] Liu HY, Chen D, Li LL, Liu TL, Tan LF, Wu XL, et al. Multifunctional gold nanoshells on silica nanorattles: a platform for the combination of photothermal therapy and chemotherapy with low systemic toxicity. Angew Chem Int Ed 2011;50:891e5.

74

S. Shen et al. / Biomaterials 39 (2015) 67e74

[5] Shen S, Tang HY, Zhang XT, Ren JF, Pang ZQ, Wang DG, et al. Targeting mesoporous silica-encapsulated gold nanorods for chemo-photothermal therapy with near-infrared radiation. Biomaterials 2013;34:3150e8. [6] Shen S, Ren JF, Zhu XY, Pang ZQ, Lu XH, Deng CH, et al. Monodisperse magnetites anchored onto carbon nanotubes: a platform for cell imaging, magnetic manipulation and enhanced photothermal treatment of tumors. J Mater Chem B 2013;1:1939e46. [7] Okuno T, Kato S, Hatakeyama Y, Okajima J, Maruyama S, Sakamoto M, et al. Photothermal therapy of tumors in lymph nodes using gold nanorods and near-infrared laser light. J Control Release 2013;172:879e84. [8] Tsai MF, Chang SH, Cheng FY, Shanmugam V, Cheng YS, Su CH, et al. Au nanorod design as light-absorber in the first and second biological nearinfrared windows for in vivo photothermal therapy. ACS Nano 2013;7: 5330e42. [9] Bagley AF, Hill S, Rogers GS, Bhatia SN. Plasmonic photothermal heating of intraperitoneal tumors through the use of an implanted near-infrared source. ACS Nano 2013;7:8089e97. [10] Robinson JT, Welsher K, Tabakman SM, Sherlock SP, Wang H, Luong R, et al. High performance in vivo near-IR (>1 mum) imaging and photothermal cancer therapy with carbon nanotubes. Nano Res 2010;3:779e93. [11] Markovic ZM, Harhaji-Trajkovic LM, Todorovic-Markovic BM, Kepic DP, Arsikin KM, Jovanovic SP, et al. In vitro comparison of the photothermal anticancer activity of graphene nanoparticles and carbon nanotubes. Biomaterials 2011;32:1121e9. [12] Zhou ZG, Sun YA, Shen JC, Wei J, Yu C, Kong B, et al. Iron/iron oxide core/shell nanoparticles for magnetic targeting MRI and near-infraredphotothermal therapy. Biomaterials 2014;35:7470e8. [13] Li XJ, Takashima M, Yuba E, Harada A, Kono K. PEGylated PAMAM dendrimerdoxorubicin conjugate-hybridized gold nanorod for combinedphotothermalchemotherapy. Biomaterials 2014;35:6576e84. [14] Tao Y, Ju EG, Liu Z, Dong K, Ren JS, Qu XG. Engineered, self-assembled nearinfrared photothermal agents for combined tumor immunotherapy and chemo-photothermal therapy. Biomaterials 2014;35:6646e56. [15] Bai J, Liu YW, Jiang XE. Multifunctional PEG-GO/CuS nanocomposites for nearinfrared chemo-photothermal therapy. Biomaterials 2014;35:5805e13. [16] Zhu MT, Nie GJ, Meng H, Xia T, Nel A, Zhao YL. Physicochemical properties determine nanomaterial cellular uptake, transport, and fate. Acc Chem Res 2013;46:622e31. [17] Nystrom AM, Fadeel B. Safety assessment of nanomaterials: implications for nanomedicine. J Control Release 2012;161:403e8. [18] Nel A, Xia T, Meng H, Wang X, Lin SJ, Ji ZX, et al. Nanomaterial toxicity testing in the 21st century: use of a predictive toxicological approach and highthroughput screening. Acc Chem Res 2013;46:607e21. [19] Li WP, Liao PY, Su CH, Yeh CS. Formation of oligonucleotide-gated silica shellcoated Fe3O4-Au coreeshell nanotrisoctahedra for magnetically targeted and near-infrared light-responsive theranostic platform. J Am Chem Soc 2014;136:10062e75. [20] Torres-Lugo M, Rinaldi C. Thermal potentiation of chemotherapy by magnetic nanoparticles. Nanomedicine 2013;8:1689e707. [21] Dormer KJ, Awasthi V, Galbraith W, Kopke RD, Chen K, Wassel R. Magnetically-targeted, technetium 99m-labeled nanoparticles to the inner ear. J Biomed Nanotechnol 2008;4:174e84. [22] Gogoi M, Sarma HD, Bahadur D, Banerjee R. Biphasic magnetic nanoparticlesnanovesicle hybrids for chemotherapy and self-controlled hyperthermia. Nanomedicine 2014;9:955e70. [23] Shi DL, Cho HS, Chen Y, Xu H, Gu HC, Lian J. Adv Mater 2009;21:2170e3. [24] Storm FK, Harrison WH, Elliott RS, Morton DL. Cancer Res 1979;39:2245e51. [25] Maier-Hauff K, Ulrich F, Nestler D, Niehoff H, Wust P, Thiesen B, et al. Efficacy and safety of intratumoral thermotherapy using magnetic iron-oxide nanoparticles combined with external beam radiotherapy on patients with recurrent glioblastoma multiforme. J Neurooncol 2011;103:317e24. [26] Maier-Hauff K, Rothe R, Scholz R, Gneveckow U, Wust P, Thiesen B, et al. Intracranial thermotherapy using magnetic nanoparticles combined with external beam radiotherapy: results of a feasibility study on patients with glioblastoma multiforme. J Neurooncol 2007;81:53e60. €fner N, [27] Johannsen M, Gneveckow U, Thiesen B, Taymoorian K, Cho CH, Waldo et al. Thermotherapy of prostate cancer using magnetic nanoparticles: feasibility, imaging, and three-dimensional temperature distribution. Eur Urol 2007;52:1653e61.

€fner N, Scholz R, [28] Johannsen M, Gneveckow U, Taymoorian K, Thiesen B, Waldo et al. Morbidity and quality of life during thermotherapy using magnetic nanoparticles in locally recurrent prostate cancer: results of a prospective phase I trial. Int J Hyperth 2007;23:315e23. [29] Chu MQ, Shao YX, Peng JL, Dai XY, Li HK, Wu QS, et al. Near-infrared laser light mediated cancer therapy by photothermal effect of Fe3O4 magnetic nanoparticles. Biomaterials 2013;34:4078e88. vy M, Bacri JC, et al. [30] Lartigue L, Hugounenq P, Alloyeau D, Clarke SP, Le Cooperative organization in iron oxide multi-core nanoparticles potentiates their efficiency as heating mediators and MRI contrast agents. ACS Nano 2012;6:10935e49. [31] Qiu P, Jensen C, Charity N, Towner R, Mao C. Oil phase evaporation-induced self-assembly of hydrophobic nanoparticles into spherical clusters with controlled surface chemistry in an oil-in-water dispersion and comparison of behaviours of individual and clustered iron oxide nanoparticles. J Am Chem Soc 2010;132:17724e32. [32] Hayashi K, Nakamura M, Sakamoto W, Yogo T, Miki H, Ozaki S, et al. Superparamagnetic nanoparticle clusters for cancer theranostics combining magnetic resonance imaging and hyperthermia treatment. Theranostics 2013;3: 366e76. [33] Hayashi K, Nakamura M, Miki H, Ozaki S, Abe M, Matsumoto T, et al. Magnetically responsive smart nanoparticles for cancer treatment with a combination of magnetic hyperthermia and remote-control drug release. Theranostics 2013;4:834e44. [34] Qin W, Lohrman J, Ren SQ. Magnetic and optoelectronic properties of gold nanocluster-thiophene assembly. Angew Chem Int Ed 2014;53:7316e9. [35] Ni WX, Qiu YM, Li M, Zheng J, Sun RWY, Zhan SZ, et al. Metallophilicity-driven dynamic aggregation of a phosphorescent gold(I)-ailver(I) cluster prepared by aolution-based and mechanochemical approaches. J Am Chem Soc 2014;136: 9532e5. [36] Zhang XD, Chen J, Luo ZT, Wu D, Shen X, Song SS, et al. Enhanced tumor accumulation of sub-2 nm gold nanoclusters for cancer radiation therapy. Adv Health Mater 2014;3:133e41. [37] Retnakumari A, Jayasimhan J, Chandran P, Menon D, Nair S, Mony U, et al. CD33 monoclonal antibody conjugated Au cluster nano-bioprobe for targeted flow-cytometric detection of acute myeloid leukaemia. Nanotechnology 2011;22:285102. [38] Deng YH, Yang WL, Wang CC, Fu SK. A novel approach for preparation of thermoresponsive polymer magnetic microspheres with coreeshell structure. Adv Mater 2003;15:1729e32. [39] Ma W, Xu S, Li JM, Guo J, Lin Y, Wang CC. Hydrophilic dual-responsive magnetite/PMAA core/shell microspheres with high magnetic susceptibility and pH sensitivity via distillation-precipitation polymerization. J Polym Sci Part A Polym Chem 2011;49:2725e33. [40] Ren JF, Shen S, Wang DG, Xi Z, Guo LR, Pang ZQ, et al. The targeted delivery of anticancer drugs to brain glioma by PEGylated oxidized multi-walled carbon nanotubes modified with angiopep-2. Biomaterials 2012;33:3324e33. [41] Zhang B, Sun X, Mei H, Wang Y, Liao Z, Chen J, et al. LDLR-mediated peptide22-conjugated nanoparticles for dual-targeting therapy of brain glioma. Biomaterials 2013;34:9171e82. [42] Zhang B, Shen S, Liao Z, Shi W, Wang Y, Zhao J, et al. Targeting fibronectins of glioma extracellular matrix by CLT1 peptide-conjugated nanoparticles. Biomaterials 2014;35:4088e98. [43] Zhang B, Wang H, Liao Z, Wang Y, Hu Y, Yang J, et al. EGFP-EGF1-conjugated nanoparticles for targeting both neovascular and glioma cells in therapy of brain glioma. Biomaterials 2014;35:4133e45. [44] Liu J, Sun Z, Deng Y, Zou Y, Li C, Guo X, et al. Highly water-dispersible biocompatible magnetite particles with low cytotoxicity stabilized by citrate groups. Angew Chem Int Ed 2009;48:5875e9. [45] Wang S, Zhang Q, Luo XF, Li J, He H, Yang F, et al. Magnetic graphene-based nanotheranostic agent for dual-modality mapping guided photothermal therapy in regional lymph nodal metastasis of pancreatic cancer. Biomaterials 2014;35:9473e83. [46] Au L, Zheng D, Zhou F, Li ZY, Li XD, Xia YN. A quantitative study on the photothermal effect of immuno gold nanocages targeted to breast cancer cells. ACS Nano 2008;2:1645e52. [47] Zhou YF. Noninvasive treatment of breast cancer using high-intensity focused ultrasound. J Med Imaging Health Informatics 2013;3:141e56.