Photothermal ablation of bone metastasis of breast cancer using PEGylated multi-walled carbon nanotubes

This study investigates therapeutic efficacy of photothermal therapy (PTT) in an orthotropic xenograft model of bone metastasis of breast cancer. The near-infrared (NIR) irradiation on Multi-Walled Carbon Nanotubes (MWNTs) resulted in a rapid heat generation which increased with the MWNTs concentration up to 100 μg/ml. MWNTs alone exhibited no toxicity, but inclusion of MWNTs dramatically decreased cell viability when combined with laser irradiation. Thermographic observation revealed that treatment with 10 μg MWNTs followed by NIR laser irradiation resulted in a rapid increase in temperature up to 73.4±11.98 °C in an intraosseous model of bone metastasis of breast cancer. In addition, MWNTs plus NIR laser irradiation caused a remarkably greater suppression of tumor growth compared with treatment with either MWNTs injection or NIR irradiation alone, significantly reducing the amount of tumor-induced bone destruction. All these demonstrate the efficacy of PTT with MWNTs for bone metastasis of breast cancer.


Results
Synthesis and characterization of MWNTs. As MWNTs have a highly hydrophobic surface, agglomeration will take place almost inevitably when they are dispersed in aqueous solutions. To stabilize the MWNTs in a solution, PEG coated MWNTs were prepared. The TEM images of MWNTs composites (Fig. 1A) demonstrated PEGylated MWNTs suspended individually in water in a single tube form while the raw MWNTs was winding up forming a large bundle. As shown in Fig. 1B, The average length of MWNTs was estimated to be 1126 ± 389 nm from 100 counts of MWNTs. In Fig. 1C showing the 1H NMR spectra (in D 2 O) of raw MWNTs and PEG functionalized MWNTs, PEG resonance peaks were clearly observed in the spectrum of PEG modified MWNTs, indicating successful synthesis of PEGylated MWNTs. MWNTs with a 1126 nm length resulted in a high absorbance at a region of 700-1,100 nm, as shown in Fig. 1D. In Fig. 1E showing the stability of raw MWNTs and PEGylated MWNTs, agglomeration and settlement occurred after raw MWNTs were dispersed in PBS for only 24 h, but PEGylated MWNTs were stably dissolved in PBS for 7 days or even up to weeks.

MWNTs mediated photothermal effect generates significantly temperature elevation in vitro.
To observe the photothermal effect of MWNTs in vitro, MWNTs solutions with different concentrations were irradiated with 808 nm laser at a power density of 5 W/cm 2 with PBS as a control. MWNTs plus NIR laser produced a significantly greater amount of heat than laser alone (Fig. 1F). A temperature elevation of approximately 18 °C resulted from 1 min NIR irradiation on 100 μ g/ml MWNTs while 1.5 °C from only 1 min NIR irradiation on PBS. Moreover, the temperature increase associated with MWNTs was positively related to exposure time. 4 min NIR irradiation elevated the temperature by 32.7 °C at a concentration of 100 μ g/ml. In addition, the temperature of MWNTs also rose with the concentration of MWNTs (0-100 μ g/ml) (Fig. 1F). These results indicated that MWNTs possessed a strong capacity of light-heat conversion under NIR laser irradiation.
MWNTs have low cytotoxicity. The cytotoxicity of the MWNTs was evaluated in MCF-7 and MDA-231 cells using the CCK8 assay. The viability of untreated cells was about 100% and the cellular viability of MCF-7 cells remained above 95% when they were cultured with MWNTs of different concentrations ( Fig. 2A). Similar situations were found in MDA-231 cells ( Fig. 2A). The results indicated that MWNTs have low cytotoxicity.

MWNTs mediated photothermal ablation efficiently kills cancer cells in vitro.
CCK8 assay demonstrated that MWNTs combined with NIR irradiation (808 nm, 5 W/cm 2 ) induced stronger cytotoxicity compared with either MWNTs or NIR light irradiation alone (Fig. 2B&C). 1 min 808 nm NIR irradiation at 5 W/cm 2 did not change the viability of the tumor cells, but the viability of the tumor cells decreased significantly to 74.3% when treated with NIR irradiation combined with 100 μ g/ml MWNTs. Moreover, the viability of the tumor cells decreased significantly as the NIR laser irradiation Scientific RepoRts | 5:11709 | DOi: 10.1038/srep11709 time increased. After 2 min NIR irradiation, the viability of the MCF-7 cells exposed to MWNTs at a concentration of 100 μ g/ml dramatically decreased to 35.2% (Fig. 2B). The same trend was observed in 50 μ g/ml MWNTs plus NIR irradiation (Fig. 2B). The situations in MDA-231 cells were similar (Fig. 2C). These demonstrated that MWNTs might be used as an effective photothermal agent for photothermal destruction of cancer cells.
To further assess the photothermal effect of MWNTs combined with NIR irradiation, cells were stained with Live-Dead cell staining kit which can distinguish live cells (green fluorescence) from dead or dying ones (red fluorescence). The majority of cells treated with either MWNTs alone or NIR irradiation alone (808 nm, 5 W/cm 2 ) were living ones (green fluorescence), but a large number of cells treated with MWNTs plus NIR laser irradiation (808 nm, 5 W/cm 2 ) were dead ones (red or yellowish fluorescence) (Fig. 2D,E). Similar morphological damage to MCF-7 ( Fig. 2D i-l) and MDA-231 cells (Fig. 2E i-l) observed by confocal microscopy were shown by merging green and red fluorescence on bright field images. These results suggest that combination of MWNTs and NIR irradiation is necessary to achieve a lethal effect on tumor cells.

MWNTs mediated photothermal effect generates significantly temperature elevation in vivo.
We evaluated the photothermal effect of MWNTs in vivo by temperature monitoring using an infrared thermal imaging camera. As shown in Fig. 3A,B, laser irradiation (808 nm, 5 W/cm 2 ) combined with MWNTs led to significantly higher temperatures than irradiation alone. After 60 s of laser irradiation, the temperature on the tumor surface increased rapidly up to 73.4 ± 11.98 °C in the 10 μ g MWNTs plus laser group, to 47.3 ± 1.63 °C in the 1 μ g MWNTs plus laser group, but only to 42.8 ± 1.10 °C in the laser only group. Additionally, the temperature increase at the tumor site was associated with the concentration of MWNTs (Fig. 3B). 10 μ g MWNTs plus laser treatment caused a higher temperature increase at 60 s compared with the other two treatments (P < 0.001 vs laser group & P = 0.02 vs 1 μ g MWNTs plus laser group). Moreover, the temperature increase at the tumor site was associated with the duration of NIR irradiation. In the 10 μ g MWNTs plus laser group, the temperatures after irradiation for 15 s, 30 s, 45 s and 60 s at the tumor surface were constantly increased (Fig. 3B).
MWNTs mediated photothermal ablation reduces tumor volume and cancer-induced bone destruction. We next investigated the effectiveness of MWNTs-induced photothermal ablation in reducing tumor growth and cancer-induced bone destruction. As expected, MWNTs plus NIR irradiation (808 nm, 5 W/cm 2 ) significantly inhibited tumor growth compared with MWNTs alone or NIR and 100 μ g/ml MWNTs + Laser (d, h, l). The photothermal effect of MWNTs plus laser (d, h, l) led to a large number of dead cells (red or yellow fluorescence), but cells treated with MWNTs alone (c, g, k), laser alone (b, f, j) and nothing (a, e, i) were mostly living (green fluorescence). Fluorescence microscopy images (i-l) showed overlapping green and red fluorescence on bright field images. Magnification ×200 (a-d); magnification ×600 (e-l).
Scientific RepoRts | 5:11709 | DOi: 10.1038/srep11709 irradiation alone (Fig. 4A). There were no statistically significant differences regarding the mean tumor volume among the saline, NIR irradiation and MWNTs groups on day 10 after treatment (P > 0.05). However, the mean tumor volume in the 10 μ g MWNTs plus NIR irradiation group on day 10 after treatment (122.4 ± 120.9 mm 3 ) was significantly smaller than that in the saline group (1267.5 ± 327.5 mm 3 ; P < 0.001). In addition, when the dosage of MWNTs dropped to 1 μ g, their effect on suppressing tumor growth was greatly attenuated under the same NIR irradiation, indicating an association between the dosage and antitumor effect of MWNTs under NIR irradiation.
Qualitative assessment of bone architecture showed the bone structure was protected completely in the group treated with 10 μ g MWNTs plus NIR irradiation (808 nm, 5 W/cm 2 ). The group treated with 1 μ g MWNTs plus NIR irradiation (808 nm, 5 W/cm 2 ) showed partial destruction of cortical bone and new bone extending from the cortex. However, apparent osteolytic destruction and massive new bone formation was observed in the saline, NIR irradiation (808 nm, 5 W/cm 2 ) and MWNTs groups (Fig. 4B). The bone volume of tumor-bearing tibia in the 10 μ g MWNTs plus NIR irradiation group was significantly smaller than that in the saline, NIR irradiation and MWNTs groups (P < 0.05) (Fig. 4C). However, the 1 μ g MWNTs plus NIR irradiation only slightly suppressed bone destruction, with no significant difference in bone volume from the saline, NIR irradiation or MWNTs group. (P > 0.05) (Fig. 4C).
MWNTs mediated photothermal ablation does not affect tactile allodynia and body weight in mice. Previous study showed that thermal ablation resulted in neurodestruction and behavioral hypersensitivity to allodynia 27 . We therefore tested the effects of photothermal therapy of MWNTs on the tactile allodynia in mice using nociceptive tests. The maximum tactile allodynia in five animal groups was similar before treatment, and there were no significant between-group differences in the maximum tactile allodynia after photothermal therapy (Fig. 5A) (P > 0.05). We further monitored the body weights following treatments in mice. As expected, the animals had no significant body weight loss after treatment with MWNTs plus laser at the final follow-up (Fig. 5B). These results clearly indicated that MWNTs combined with NIR irradiation did not affect the withdrawal threshold or body weight, thus implying the PTT might be a safe therapy used in tumor-bearing mice.

Discussion
Our current work demonstrated that PTT via MWNTs plus NIR irradiation effectively generated great heat which led to significant damage to MCF-7 and MDA-231 cells cultured in vitro. In addition, MWNTs plus NIR irradiation significantly increased the local temperature at the bone metastatic foci, reduced the tumor size in the mice, and protected the bone from cancer-induced destruction in an intraosseous Tumor growth curves for mice treated with saline, laser alone, 10 μ g MWNTs, 1 μ g MWNTs + laser, and 10 μ g MWNTs + laser. There were no statistically significant differences (P > 0.05) in the mean tumor volume among the groups treated with saline, laser alone and 10 μ g MWNTs, but tumor volume in the group treated with 10 μ g MWNTs + laser was significantly smaller (P < 0.01). The effect of suppression tumor growth was attenuated with decreased dosage of MWNTs of 1 μ g. 1 μ g MWNTs + laser produced modest suppression in tumor volume, which was significantly larger than that in the group of 10 μ g MWNTs + laser (P < 0.05). The bars represent means ± s.d. (n = 5). (B) Micro-CT images of tumor-bearing tibiae showed changes of bone structure subjected to treatment with saline (rank a), laser alone (rank b), (c) 10 μ g MWNTs (rank c), 1 μ g MWNTs + laser (rank d) and 10 μ g MWNTs + laser (rank e) from total bone, transverse-section (X-section) and cross-section (Y-section) views respectively. In the groups treated with saline (a), saline + laser (b) and 10 μ g MWNTs (c), the bone structure was all severely destroyed and massive osteosclerotic growth was observed. However, in the groups treated with 1 μ g MWNTs + laser (d), and 10 μ g MWNTs + laser (e), there was only mild bone destruction. Especially in the groups treated with 10 μ g MWNTs + laser, the bone structure was significantly protected. (C) Tumorbearing tibiae treated with 10 μ g MWNTs + laser showed a statistically significant decrease in bone volume compared with those treated with saline, laser alone and 10 μ g MWNTs (P < 0.05), but 1 μ g MWNTs + laser led to only a slight reduction in bone volume compared with the other three groups (P > 0.05). The bars represent means ± s.d. (n = 5). model of bone metastasis of breast cancer. We further found that duration of NIR irradiation and dosage of MWNTs might be critical to maximum tumor destruction. Prolonged duration of NIR irradiation and increased dosage of MWNTs enhanced photothermal effect of PTT. Significantly, our therapy for the bone metastasis of breast cancer, MWNTs plus NIR irradiation, achieved high antitumor efficacy without significant side effects, as proved by nociceptive tests and body weight measurements.
NIR light at a region of 700-1,100 nm in wavelength is often used in PTT because it penetrates deeply into the tissue and is hardly absorbed by normal tissue 18 . In our study, we demonstrated that MWNTs show a high degree of absorption in this NIR region in the optical absorption measurement, which is consistent with previous studies 22 . Study show that MWNTs combined with 808 nm NIR light induce highly effective photothermal effect 21 . So, we chose 808 nm NIR light to stimulate MWNTs.
Neither MWNTs nor NIR irradiation alone resulted in cell destruction. MWNTs alone led to relatively low toxicity, with less than 5% cell damage under a concentration of 100 μ g/ml. NIR irradiation alone at 5 W/cm 2 for 2 min only caused minimal temperature elevation and no change in cell viability. However, MWNTs combined with NIR irradiation at 5 W/cm 2 for 2 min dramatically enhanced the temperature and killed a large number of cells. Previous studies indicated that if NIR irradiation was used alone, the laser needed to be as high as 50.9 W/cm 2 and prolonged to 10 min for destruction of the majority of cells 22 . Therefore, our study has demonstrated that MWNTs used together with NIR irradiation can significantly enhance the efficiency of heat generation, implying their role as a robust photothermal agent in tumor PTT.
Although inclusion of MWNTs is important for generation of high heat in PTT, irradiation duration is also an essential factor. In combination with 100 μ g/ml MWNTs, NIR irradiation at 5 W/cm 2 for 30 s increased the temperature to 66.7 °C, but 60 s NIR irradiation enhanced the temperature to 73.4 °C. Prolonged irradiation may allow MWNTs to absorb more NIR laser to generate more heat, leading to a maximum damage to cancer cells.
Moreover, antitumor capacity of PTT was associated with the dosage of MWNTs delivered to the tumor. In combination with NIR irradiation, 10 μ g MWNTs yielded dramatical temperature elevation and significant suppress of tumor growth, showing a statistical significance compared with the saline group. However, 1 μ g MWNTs only caused mild temperature increase and modest suppress of tumor growth, displaying no statistical significance compared with the saline group. This is consistent with a previous study, which presented a dose-dependent effect of MWNTs on tumor regression 26 .
Our photothermal ablation experiment revealed that 10 μ g MWNTs plus NIR irradiation resulted in dramatically temperature increase of 73.4 ± 11.98 °C in 60 s. Although the temperature was far above the damage threshold of 43 °C supposed to cause irreversible damage to cancer cells 28 , the tumors were still not completely eradicated. We think this might be associated with two possible factors. First, the dispersibility of MWNTs in the tumor might not be good enough. The poor dispersibility of the photothermal agent was also observed by Hashida et al. who indicated that the poor dispersibility of nanotubes in tumor after intra-tumor injection might undermine the effect of PTT 29 . Secondly, since the bone metastasis induced by breast cancer had highly irregular margins it was hard to eliminate the remaining tumor cells thoroughly. As we know, if even a trace of tumor cells survive PTT treatment they can regrow to form a tumor tissue in a short period. Therefore, in order to completely eradicate cancer cells, it is necessary to optimize the parameters of PTT, such as NIR irradiation time, dosage of photothermal agent and dispersibility of photothermal agent. PTT combined with other therapeutic algorithms is another approach to enhancing the therapeutic effect. Our study has other limitations. First, our animal models of bone metastasis induced by breast cancer were established by direct inoculation of tumor cells into the tibia of the mice. This, of course, did not simulate the steps of metastatic lesion spreading from the primary tumor to bone. However, our models allowed for relatively consistent tumor burden 30 and a lesion location suitable for PTT. Secondly, we showed that the temperature rose with the period of NIR irradiation (0-4 min) and also with the concentration of MWNTs (0-100 μ g/ml), but we did not determine the optimal irradiation duration and the optimal concentration of MWNTs which might lead to the maximal therapeutic efficacy. Optimization of all parameters for PTT using MWNTs plus NIR irradiation is a huge challenge for further studies. Thirdly, although we studied the effects of MWNTs-associated photothermal therapy on tumor growth and cancer-induced bone destruction, we did not address the effects on other tumor-related factors, like macrophages. According to Yang et al., macrophages influenced by MWNTs plays an important role in the tumor progression and metastasis 31 . Thus, we should take macrophages into account in the further research.
In conclusion, our experiments demonstrates that MWNTs plus NIR laser irradiation is a promising therapeutic alternative because it is safe and effective for bone metastatic foci induced by breast cancer. Of course, many problems related to this treatment warrant extensive research before it can be clinically applied. medium. After 24 h incubation, the cells were exposed to irradiation with an 808-nm NIR laser for 2 min at 5 W/cm 2 and stained with Live-Dead cell staining kit 16 hours later. Briefly, cells were stained with 0.5 ml staining solution containing 0.5 μ l Live-Dye (1 mM) and 0.5 μ l PI (2.5 mg/ml). After incubated for 15 min at 37 °C, signals were visualized by confocal microscopy (FV10i-W, Olympus, Tokyo, Japan).

Animals.
In vivo experiments were conducted using female Balb/c, weighing approximately 18 g (5-week-old), obtained from Southern Medical University Experimental Animal Center (Guangzhou, China). Animals had free access to chow and water in a 25 °C ± 1 °C environment with regular light and 55% ± 5% relative humidity. Experiments were conducted in accordance with Guiding Principles for the Care and Use of Experiment Animals in Southern Medical University. The animal study protocols were approved by the Institutional Animal Care and Use Committee at Southern Medical University (NFYY-2014-65).
Photothermal effect of MWNTs in vivo. The temperature changes in tumor sites were imaged by an infrared thermal imaging camera (FLIR T610, FLIR Systems Inc, Sweden). EMT6 cells were used to establish the tumor model because of their high rate of tumor formation in normal Balb/c mice. Single-cell suspensions of 1.0 × 10 6 murine breast cancer EMT6 cells were injected into the tibiae of female Balb/c mice of 5-week-old. After 8 days of tumor inoculation, the mice were randomized into three different groups subjected to treatments by (i) saline (100 μ l) + laser, (ii) MWNTs (100 μ l, 10 μ g/ ml) + laser, (iii) MWNTs (100 μ l, 100 μ g/ml) + laser. One day later, NIR laser irradiation was carried out for 60 s with an 808 nm NIR laser at 1.25 W (5 W/cm 2 ) under sodium pentobarbital anesthesia. During irradiation, the temperature changes in the tumor region were imaged using an infrared thermal imaging camera. Pictures were edited with FLIR QuickReports 1.2.
In vivo therapeutic effect of MWNTs with NIR irradiation. The therapeutic effect of MWNTs and NIR laser irradiation was evaluated by measuring growth inhibition of the tumor inoculated in mice. Single-cell suspensions of 1.0 × 10 6 murine breast cancer EMT6 cells were injected orthotopically into the tibiae of female Balb/c mice of 5-week-old. After 8 days of tumor inoculation, the mice were randomized into five different groups treated by (i) saline only (100 μ l), (ii) saline (100 μ l) + laser, (iii) MWNTs only (100 μ l, 100 μ g/ml), (iv) MWNTs (100 μ l, 10 μ g/ml) + laser, (v) MWNTs (100 μ l, 100 μ g/ml) + laser. At 24 h after the injection, the mice in the five groups were anesthetized for irradiation on the tumor sites with an NIR laser at 808 nm. The NIR laser treatment consisted of three rounds of 60 s of illumination at 1.25 W for 0.25 cm 2 (5 W/cm 2 ) with 30 s intervals. After NIR irradiation, the tumor sizes were measured by a caliper every other day and calculated as the volume = (tumor length) × (tumor width) 2 /2.

Micro computed tomography.
Excised formalin-fixed tibiae were scanned using a micro-CT system (μ CT80, SCANCO MEDICAL, Switzerland) at a resolution of 18 μ m, a tube voltage of 50 kV and a tube current of 0.1 mA. The volume of interest (VOI) was defined as 260 slices of approximately 4.7 mm in thickness starting from the growth plate of tibial interface and moving down the tibia. Bone volume was calculated to allow quantitative analysis of bone quality.
Nociceptive tests. After PTT, the paw withdrawal latency was elicited with a hand-held force transducer (Electronic Von Frey Anesthesiometer, IITC Life Science, CA, US) adapted with a 0.5 mm 2 polypropylene tip. Each animal was placed in an individual acrylic cage (12 × 10 × 17 cm) with a wire mesh floor and acclimatized to the cage for 15 minutes before the start of testing. An increasing force was applied perpendicular to the central area of the hindpaw with the polypropylene tip. The end-point was characterized by the removal of the paw followed by clear flinching movements. After paw withdrawal, the intensity of the pressure was automatically recorded.