Long Blood Residence and Large Tumor Uptake of Ruthenium Sulfide Nanoclusters for Highly Efficient Cancer Photothermal Therapy

Transition metal sulfide (TMS) holds great potential in cancer photothermal therapy (PTT) because of the high absorbance in the near-infrared (NIR) region. The short blood circulation time and limited tumor accumulation of TMS-based photothermal agents, however, limit their applications. Herein, we design a novel TMS-based PTT agent, ruthenium sulfide-based nanoclusters (NCs), to overcome the current limitations. We firstly develop a simple method to prepare oleic acid coated ruthenium sulfide nanodots (OA-RuS1.7 NDs) and assemble them into water-soluble NCs via sequentially coating with denatured bovine serum albumin (dBSA) and poly(ethylene glycol) (PEG). The obtained PEG-dBSA-RuS1.7 NCs possess excellent photothermal conversion ability. More significantly, they exhibit enhanced blood circulation time and tumor-targeting efficiency in vivo compared with other TMS-based PTT nanoagents, which may be attributed to their appropriate hydrodynamic diameter (~70 nm) and an ideal charge (~0 mV). These characteristics help the PEG-dBSA-RuS1.7 NCs to escape the removal by the reticuloendothelial system (RES) and kidney. All these advantages enable the PEG-dBSA-RuS1.7 NCs to selectively concentrate in tumor sites and effectively ablate the cancer cells upon NIR irradiation.

a facile solvothermal method by decomposing the diethyl dithiocarbamate ruthenium (Ru(DDTC) 3 ) dissolved in the mixture of oleic acid (OA) and ethyl alcohol (V:V = 2:1) in a Teflon-lined autoclave. X-ray fluorescence (XRF) spectroscopy, X-ray diffraction (XRD), and X-ray photoelectron spectroscopy (XPS) were conducted to confirm the chemical composition of the as-synthesized RuS 1.7 NDs. The XRF spectrum demonstrates the characteristic fluorescence peaks of Ru and S (Fig. S1). The molar ratio of Ru to S was calculated to be 1:1.7 (weight ratio of Ru to S is 65:35) based on the intensity of the fluorescence peaks of Ru and S. As can be seen in Fig. S2, different from that of polycrystalline ruthenium sulfide (Joint Committee on Powder Diffraction Standards, File No. , the XRD pattern only shows broad humps, which indicates that the as-obtained RuS 1.7 NDs are in an amorphous phase. To further investigate the composition and oxidation state of the RuS 1.7 NDs, we performed XPS characterization. In the Ru 3d spectrum (Fig. S3), two peaks centered at 279.7 and 283.8 eV represent the 3d 5/2 and 3d 3/2 states, with the range of the binding energy for Ru 2+ . The sulfur 2p doublet with peaks located near 162.0 eV falls into the range of the binding energy for (S 2 ) 30,31 . Our results are in good agreement with previous report of RuS 1.7 by Jeevanandam et al. 30 .
For PTT applications, good dispersibility in physiological solutions is of great importance. Because the synthesized RuS 1.7 NDs coated by OA are well dispersed in many organic solvents but not dissolved in aqueous solutions, we further functionalized them with dBSA and PEG. In this way, RuS 1.7 NDs were aggregated to NCs. The formed NCs are water-soluble, and fourier-transform infrared (FTIR) spectroscopy was used to determine the coating of RuS 1.7 NDs. The bands at 1625 cm −1 and 2988 cm −1 in the FTIR spectrum of OA-RuS 1.7 NDs can be assigned to the characteristic vibration of carbon-carbon double bond of OA and the vibration of unsaturated carbon-hydrogen bond from OA, respectively (Fig. S4) 24 . These characteristic bands of OA disappeared after modification with dBSA followed by the conjugation with PEG, while an obvious new band at 1650 cm −1 ascribed to the vibration of carbonyl group from amide of dBSA was recorded (Fig. S4) 23 . Also, typical transmission electron microscope (TEM) and hydrodynamic size were employed to evaluate our PTT agent since a suitable size is beneficial for tumor-specific accumulation. As illustrated in Fig. 1b, the as-synthesized RuS 1.7 NDs show a uniform size with ~1.5 nm in diameter. After sequentially modified by dBSA and PEG, the clusters were formed by the aggregation of the RuS 1.7 NDs with the diameter ~23 nm (Figs 1c and S5). The formed NCs exhibited well stability in water and PBS, and do not precipitate within at least 6 months even in high concentration (2 mg mL −1 ) (Fig. S6). As shown in Fig. 1d, the hydrodynamic size of the PEG-dBSA-RuS 1.7 NCs measured by dynamic light scattering (DLS) is 70 nm with a polydispersity index of 0.226. This size of nanoparticles is well suited for prolonging blood circulation time and improving tumor accumulation 28,29 . In addition, the zeta potential value of PEG-dBSA-RuS 1.7 NCs (close to 0 mV) can effectively make themselves "stealthy" thereby evading the recognition by RES (Fig. S7) 28 .
Next, we studied NIR absorption and photothermal performance of PEG-dBSA-RuS 1.7 NCs. As shown in Fig. 1e, Vis-NIR spectrums indicate that RuS 1.7 NDs have a very broad absorption in the NIR region, which is not affected by the sequential modification with dBSA and PEG. The mass extinction coefficient of PEG-dBSA-RuS 1.7 NCs at 800 nm was determined to be 9.4 L g −1 cm −1 , much higher than that of graphene oxide used in cancer PTT (λ = 800 nm, 3.6 L g −1 cm −1 ) 32 . To survey the photothermal performance of our nanoclusters, the PEG-dBSA-RuS 1.7 NCs solutions of different concentrations were exposed to an 808 nm continuous-wave laser at a power density of 1.4 W cm −2 (Fig. 1f). The photothermal heating curves recorded by an IR thermal camera show a strong concentration-dependent photothermal effect with the highest temperature increment up to 47.2 °C. In contrast, only a slight temperature rise (~3 °C) was observed for water. According to the reported method 8,24,33 , the photothermal conversion efficiency of PEG-dBSA-RuS 1.7 NCs was calculated to be ~28.5% which was similar to that of Bi 2 S 3 NRs (~28.1%) 24 . Besides, the study on photostability of PEG-dBSA-RuS 1.7 NCs was carried out. The Vis-NIR spectrum of PEG-dBSA-RuS 1.7 solution after the continuous irradiation for 1 h, shows no noticeable change, while the Vis-NIR absorption of indocyanine green (ICG) is largely decreased ( Fig. 2a and b). The excellent photothermal effects and photothermal stability of PEG-dBSA-RuS 1.7 NCs make themselves highly potential to be developed as a photothermal therapeutic agent. In vitro and in vivo toxicity of PEG-dBSA-RuS 1.7 NCs. In order to use PEG-dBSA-RuS 1.7 NCs in PTT, firstly the in vitro and in vivo toxicity of them should be evaluated. In our study, a standard WST-1 cell proliferation assay 34,35 was conducted using the mouse mammary tumor cell line (4T1) and fibroblast cell line (L929). After the cells were incubated with various concentrations of PEG-dBSA-RuS 1.7 NCs for 24 or 48 h, we found that the relative cellular viabilities did not obviously decrease for PEG-dBSA-RuS 1.7 NC-treated cells even at the highest nanocluster concentration (200 μ g mL −1 ) (Fig. 3a). In addition, to prove the potential intravenous administration in vivo, we employed the hemolysis assay on red blood cells. No visually red color was found in the supernatant except the color of the suspended PEG-dBSA-RuS 1.7 NCs that cannot be removed by low-speed centrifugation (Fig. 3b). And the Vis-NIR spectrums result clearly shows there is no characteristic peak of hemoglobin for the samples with different concentrations of PEG-dBSA-RuS 1.7 NCs, suggesting that the NCs have excellent blood compatibility (Fig. 3b). Furthermore, the treatment with PEG-dBSA-RuS 1.7 NCs (intravenous (i.v.) injection, up to 14 mg kg −1 ) even for 28 days did not cause death between the PEG-dBSA-RuS 1.7 NC-treated mice and the control mice. The liver function markers including alanine aminotransferase (ALT), alkaline phosphatase (ALP) and aspartate aminotransferase (AST), and kidney function markers including uric acid (UA) and blood urea nitrogen (BUN), were measured at 1 day, 4 day and 7 day after injection. The presented error bars are based on four mice in each group. The results showed no obvious hepatic and kidney disorder of mice at the given dose (14 mg kg −1 ). Collectively, these results provide evidence of the low in vitro and in vivo totoxicity of the PEG-dBSA-RuS 1.7 NCs.
In vitro photothermal ablation of tumor cells. Encouraged by low toxicity of PEG-dBSA-RuS 1.7 NCs, we further tested their PTT efficiency using 4T1 cells as representative cancer cells. Upon 10 min NIR irradiation at power density even up to 4.8 W cm −2 , the proliferation level determined by the WST1 assay was not decreased for the cells without adding PEG-dBSA-RuS 1.7 NCs (Fig. 4a). In contrast, for the cancer cells incubated with PEG-dBSA-RuS 1.7 NCs and subsequently irradiated by NIR laser, the cellular proliferation level gradually decreased with the increase in either laser power density or nanocluster concentration. Typically, when the cells were under a mild laser exposure condition (1.4 W cm −2 , 10 min) after being incubated with PEG-dBSA-RuS 1.7 NCs, a small amount of cells died immediately but the massive cell death was observed 24 hours later after irradiation (Fig. 4b). This should be probably caused by the heat-induced apoptosis 36 . In vivo biodistribution and tumor-targeting efficacy. Before conducting further in vivo PTT experiments, it is necessary to confirm the biodistribution of PEG-dBSA-RuS 1.7 NCs considering that the in vivo system is more complicated. Herein, inductively coupled plasma-optical emission spectroscopy (ICP-OES) was employed to determine the Ru amount in the organs and blood that were collected at different intervals from tumor-bearing mice with PEG-dBSA-RuS 1.7 NC injection. The Ru level in blood was 15.85% ID g −1 (percentage of injected dose per gram tissue) at 24 h post injection, which was approximately four times greater than Mo level in the blood of PEGylated MoS 2 nanosheet-treated mice (3.67% ID g −1 ) 32 . At 72 h post injection, the Ru level in blood was kept at 3.21% ID g −1 , and Ru was still detectable even at 96 h post injection (0.61% ID g −1 ) (Fig. 5a). These results indicate that the blood circulation time of PEG-dBSA-RuS 1.7 NCs is very long, which may be attributed to the suitable hydrodynamic size (70 nm) and almost neutral charge. These features effectively enable the PEG-dBSA-RuS 1.7 NCs to escape from the RES reorganization and avoid renal clearance 28,29 . Moreover, the half-life (t 1/2 ) of PEG-dBSA-RuS 1.7 was also calculated to be 2 h which is similar to that of the MoS 2 -GSH nanodots 21 . The long blood circulation time is especially advantageous for the nanoagent accumulation in solid tumor. Indeed, at 4 day post injection the Ru level in tumor was determined to be 18.1% ID g −1 (Fig. 5b). It is much higher than that of other TMS nanoagent-treated mice, such as 6.62% for PEGylated MoS 2 nanosheets 32 ; ~2% for Tween-functionalized Bi 2 S 3 nanorods 24 ; and ~12.7% for PEGylated WS 2 nanoflakes 25   mice were irradiated with an 808 nm NIR laser (1.4 W cm −2 ) for 10 min. The temperature of the tumor area was recorded by an IR thermal camera at 1 min intervals (Fig. 6a,b). Under NIR irradiation, the tumor site temperature of the group (4) rapidly increased from 37 to 61 °C. However, the temperature of the tumor site of the control group (2) only had a small change although the mice were also irradiated by the NIR laser. An obvious crusting was observed in the mice from group (4) at the next day after irradiation while negligible change was detected in those from group (2) (Fig. 6c). Tumors in the control groups (1, 2, and 3) rapidly grew, and no difference in tumor volume was found in these control groups on the 14th day post PTT. By contrast, tumors in the group (4) were no longer observed (Fig. 6d,e,f). These results reveal that PTT based on PEG-dBSA-RuS 1.7 NCs has exhibited an excellent curative effect and PEG-dBSA-RuS 1.7 NCs are capable of acting as an ideal photothermal agent for cancer therapy.

Conclusions
In summary, we have successfully prepared a ruthenium sulfide-based nanomaterial, PEG-dBSA-RuS 1.7 NCs, which can be applied as a novel photothermal agent with excellent physiological stabilities. The PEG-dBSA-RuS 1.7 NCs demonstrate good photothermal effects upon NIR laser irradiation, excellent biocompatibility, and no obvious toxicity. More importantly, in vivo cancer treatment study further reveals that the as-prepared NCs exhibit quite long blood circulation time and markedly high tumor accumulation compared with many other explored TMS-based photothermal agents, which may benefit from the right hydrodynamic size and almost neutral charge of PEG-dBSA-RuS 1.7 NCs. As a result, our novel NCs possess the outstanding ability to eradicate the tumor in a mouse model. Our work suggests that as a new class of photothermal agents, ruthenium sulfide-based nanostructures with improved blood residence and tumor uptake may facilitate the in vivo applications of TMS-based PTT. Material Preparation. Calcein-AM and propidium iodide (PI) were provided by Invitrogen, mPEG-NH 2 (MW = 5 K) was purchased from Seebio Co. Ltd., Shanghai, China, and other chemicals were purchased from Sigma-Aldrich. The dBSA were synthesized according to our previous report 37 . All the chemicals were used as received without further purification. For a typical synthesis, Ru(DDTC) 3 was firstly prepared according to the previous literature 38 by mixing 100 mmol of RuCl 3 and 300 mmol of diethyl dithiocarbamate (Na(DDTC)) in 500 mL of distilled water. After constant stirring for 1 h, the solution was kept stationary under ambient condition for another 3 h. The resulting precipitate was filtered, washed with distilled water and dried in air at 50 °C for 24 h to obtain Ru(DDTC) 3 . Then, 272 mg of Ru(DDTC) 3 was dissolved with 10 mL of ethyl alcohol. After stirring for 30 min, 20 mL of OA was added and kept stirring for another 15 min. The Ru(DDTC) 3 colloidal solution was then transferred to a 50 mL Teflon-lined autoclave, sealed, and heated at 200 °C for 12 h. The mixture was then cooled to room temperature. The products were collected by centrifugation at 20000 g and washed with ethyl alcohol several times to remove the remnant. The final products were dissolved with 40 mL of chloroform.

Sequential dBSA and PEG Modifications of RuS 1.7 NDs.
To transfer RuS 1.7 NDs from an organic phase to an aqueous phase, sequential dBSA and PEG modifications were conducted according to our previous report 37 . In a typical procedure, 10 mg of OA-RuS 1.7 NDs was dispersed in 9 mL of chloroform. Then 100 mg of dBSA dissolved with 1 mL of dimethyl sulfoxide (DMSO) was added to the mixture. After stirring overnight at room temperature, dBSA coated RuS 1.7 NCs were collected by centrifuging the mixture at 50000 g and washing with distilled water. The purified dBSA-RuS 1.7 NCs were dispersed with 5 mL of DMSO. For PEG conjugation, 50 mg of mPEG-NH 2 (MW = 5 K) and 20 mg of 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDAC) were added to dBSA-RuS 1.7 NCs under vigorous stirring overnight at room temperature. The as-synthesized PEG-dBSA-RuS 1.7 NCs were intercepted using a 100 kDa ultra-filter and washed with distilled water. The final purified PEG-dBSA-RuS 1.7 NCs were redispersed with distilled water and filtered by a membrane with 0.22 μ m pore size.
Characterization of RuS 1.7 NDs and PEG-dBSA-RuS 1.7 NCs. The morphology and size of the as-prepared nanoparticles were characterized by the transmission electron microscope (TEM, JEM-2100, JOEL). Vis-NIR spectrum was recorded by a UV spectrophotometer (UV2600, Shimadzu, Japan). X-ray diffraction (XRD) spectrum of the dry powder was measured using an ARL-X'TRA x-ray diffractometer. X-ray photoelectron spectra were obtained on a PHI 5000 VersaProbe XPS spectrometer. X-ray fluorescence spectra were recorded on an ARL-9800 x-ray fluorescence spectrometer.
Cytotoxicity and in vitro PTT Effect Study. 4T1 and L929 cell lines were cultured in 1640 cell medium contained 10% fetal bovine serum (FBS) at 37 °C and a 5% CO 2 atmosphere. Cells were seeded into 96-well plates at a density of 10 4 cells per well, and were incubated with different concentrations of PEG-dBSA-RuS 1.7 NCs for 24 or 48 h. Relative cellular viabilities were measured by the standard WST-1 assay. For in vitro PTT, 4T1 cells were incubated with various concentrations of PEG-dBSA-RuS 1.7 NCs for 12 h and then irradiated by an 808 nm laser at a series of power densities with different exposure times. At 24 h post laser exposure, the cells were stained with calcein AM and PI for 15 min, then imaged by a confocal fluorescence microscope (Olympus). For detecting cell proliferating viability, a standard WST1 assay was conducted at 24 h post laser exposure.
Hemolysis Assay of PEG-dBSA-RuS 1.7 NCs. Red blood cells were obtained by removing the serum from BALB/c female mice blood after being washed with 0.9% saline, and centrifuged five times. After that, blood cells were diluted to 1:10 with PBS solution. Then, 0.2 mL of diluted cells suspension was mixed with 0.8 mL of PBS (as a negative control), 0.8 mL of deionized water (as a positive control), and 0.8 mL of product suspensions. The samples were shaken and kept steady for 2 h. Finally, the mixtures were centrifuged at 1000 g for 5 minutes and the absorbance of the upper supernatants was measured by UV-vis spectroscopy.
Animals and preparation of tumor-bearing mice. BALB/c female mice with body weights of about 20 g were obtained from Tianqin Biotechnology Co., Ltd, Changsha, China, and housed under protocols approved by Hainan Medical College. Tumor-bearing mice were prepared by inoculating 10 6 4T1 cells in the backside of each mouse.
Inductively coupled plasma-optical emission spectroscopy (ICP-OES) for Ru element quantification. To quantify the tissue-distribution of PEG-dBSA-RuS 1.7 NCs, the NCs (14 mg kg −1 ) were i.v. injected to the tail vein of tumor-bearing BALB/c mice. The mice (n = 4) were euthanized at different time points (1, 4 and 7 days). Then tissue samples were stored at − 20 °C before analysis. For ICP-OES experiment, each sample was added with 0.4 mL of aqua regia and digested for 48 h at 60 °C. The final solutions were diluted to 1 mL and filtered. Ru contents of all samples were measured by ICP-OES.
Assessing in vivo PTT efficacy of PEG-dBSA-RuS 1.7 NCs. When the tumor size reached approximately 5 mm in width, the mice were divided into four groups: (a) blank; (b) laser only; (c) PEG-dBSA-RuS 1.7 NCs i.v. injection; and (d) PEG-dBSA-RuS 1.7 NCs (i.v. injection) + laser. The 4T1 tumor-bearing mice were i.v. injected with PEG-dBSA-RuS 1.7 NCs (dosage = 14 mg kg −1 ) and exposed to an 808 nm laser with a 1 W cm −2 of power density for 10 min. Temperature change under laser irradiation in the tumor site was recorded by a NIR camera. Photos were captured at every 1 minute interval. Tumorswere measured in the following days, and tumor sizes were calculated as follows 24 = V ab /2 2 where V (mm 3 ) is the tumor volume, and a (mm) and b (mm) are the tumor length and width, respectively.