BiVO4/Fe3O4@PDA SPs are constructed upon the self-assembly of BiVO4 and Fe3O4 NPs following by coating with the PDA shell. Typically, BiVO4 NPs are prepared through our previous two-phase method, and Fe3O4 NPs are prepared via the classical thermal decomposition method.32,33 Transmission electron microscopy (TEM) images in Fig. 1a and 1b show that both of BiVO4 and Fe3O4 NPs are monodispersed nanospheres with the average diameters around 7.12 and 5.49 nm, respectively. High-resolution TEM (HRTEM) images exhibit the lattice fringes with the interplanar spacings of 0.312 and 0.244 nm, corresponding to the (112) planes of monoclinic BiVO4 and the (311) planes of cubic Fe3O4. X-ray diffraction (XRD) patterns of BiVO4 and Fe3O4 NPs further identify the monoclinic crystal structure of BiVO4 NPs and the cubic crystal structure of Fe3O4 NPs (Fig. S1).
Then, oil-in-water microemulsion method is employed to construct BiVO4/Fe3O4 SPs using BiVO4 and Fe3O4 NPs as the building blocks while sodium dodecyl sulfate (SDS) as the surfactants.34 The as-prepared SDS-capped BiVO4/Fe3O4 SPs are nanospheres with the average diameter of 81.20 nm (Fig. 1c). The element distributions of BiVO4/Fe3O4 SPs are characterized by energy-dispersive X-ray spectroscopy (EDS) elemental mapping (Fig. S2). Bi, V and Fe are uniformly distributed throughout the entire SPs, further demonstrating the assembled configuration of the as-prepared BiVO4/Fe3O4 SPs. Benefiting from the flexibility of the self-assembly technique, the size and composition of BiVO4/Fe3O4 SPs is tunable deliberately. For example, by increasing the toluene-to-water ratio from 1:5 to 2:5, the size of BiVO4/Fe3O4 SPs can be increased from 81.20 to 164.50 nm (Fig. S3). In the meantime, upon adjusting the feeding ratio between BiVO4 and Fe3O4 NPs during self-assembly, the molar ratio of Bi/Fe in the as-prepared BiVO4/Fe3O4 SPs is varied from 3.5:1 to 1.2:1. The corresponding products are designated as BiVO4/Fe3O4-1, BiVO4/Fe3O4-2 and BiVO4/Fe3O4-3 SPs (Table S1).
At last, dopamine (DA) monomers are oxidized followed by spontaneous polymerization on the surface of BiVO4/Fe3O4 SPs under the alkaline condition.35 The thickness of the PDA shell is positively correlated with the amount of DA, which will be increased from 10.00 to 80.00 nm when the concentration of DA increase from 0.3 to 0.8 mg/mL (Fig. S4). Given that nanomaterials larger than 120.00 nm can hardly enter into cells upon cellular phagocytosis, BiVO4/Fe3O4 SPs with the diameter around 80.00 nm is selected as the core and the PDA shell thickness is designed to be around 10.00 nm.
Since the aggregation of Fe3O4 NPs and the presence of the PDA shell can remarkably increase the molar extinction coefficient of monodispersed Fe3O4 NPs (Fig. S5), leading to the enhancement in their photothermal conversion capability, the photothermal conversion capability of BiVO4/Fe3O4@PDA SPs suspension is evaluated under 808 nm irradiation. As shown in Fig. 2a, the photothermal conversion capability of BiVO4/Fe3O4@PDA SPs is enhanced by elevating the proportion of Fe3O4 in SPs. At the same time, the temperature of BiVO4/Fe3O4@PDA SPs aqueous solution rises rapidly by increasing the power density of the applied laser and the concentration of SPs (Fig. 2b and 2c). Under 1 W/cm2 irradiation for 10 min, the aqueous solution containing 200 µg/mL of BiVO4/Fe3O4@PDA SPs exhibits a noticeable temperature increment of 25 °C. Based on the reported model, the photothermal conversion efficiency of as-prepared BiVO4/Fe3O4@PDA SPs is estimated to be 33.42%, which is comparable to previous reports (Fig. S6).29 To balance the properties deriving from BiVO4 and Fe3O4, BiVO4/Fe3O4@PDA SPs with the Bi/Fe element ratio of 1.8/1 are selected for the following in vitro and in vivo experiments.
Prior to assessing the imaging performance of BiVO4/Fe3O4@PDA SPs, their cytotoxicity is evaluated via standard Cell Counting Kit 8 (CCK-8) assay. After incubation with BiVO4/Fe3O4@PDA SPs at different concentrations for 24 h, the cell viability of oral epithelial carcinoma (KB) cells is higher than 80% even at a high concentration of 300 µg/mL, which strongly manifest the negligible cytotoxicity of BiVO4/Fe3O4@PDA SPs (Fig. S7). The colloidal stability of BiVO4/Fe3O4@PDA SPs is tested as well. After storage in water, saline, cell culture or serum-containing cell culture for 24 h, BiVO4/Fe3O4@PDA SPs are well dispersed without any visible coagulation (Fig. S8). The low toxicity plus the high colloidal stability provide a powerful guarantee for the utilization of BiVO4/Fe3O4@PDA SPs in tumor theranostic.
Subsequently, the in vitro imaging performances of BiVO4/Fe3O4@PDA SPs are exhibited in Fig. 3a-c. Owning to the high X-ray attenuation coefficient of Bi, the CT signal intensities of BiVO4/Fe3O4@PDA SPs increase linearly and sharply with their concentrations. The Hounsfield units (HU) value of BiVO4/Fe3O4@PDA SPs is calculated to be 28.2136 HU·mL·mg− 1, which is comparable to the clinically used CT contrast agent iobitridol (25.6570 HU·mL·mg− 1) (Fig. 3a). Further increasing the proportion of BiVO4 in SPs can improve the CT imaging performance of BiVO4/Fe3O4@PDA SPs undoubtedly, but may lose their PA and MR imaging performances as the price. In addition, BiVO4/Fe3O4@PDA SPs are also anticipated to be the MR imaging contrast agents owing to the superparamagnetic property of Fe3O4 NPs. Their MR imaging contrast is enhanced in a concentration-dependent manner, and the r2 value is estimated to be 186 mM− 1·s− 1, which is higher than current commercial MR contrasts, such as Resovist (143 mM− 1·s− 1) and Feridex (93 mM− 1·s− 1) (Fig. 3b). Furthermore, benefiting from the excellent photothermal conversion capability, there is a good linear relationship between the concentration of BiVO4/Fe3O4@PDA SPs and their PA signal under NIR irradiation, suggesting their great potentials as the PA contrast agents (Fig. 3c). Then, in vivo multimode CT/MR/PA imaging properties of BiVO4/Fe3O4@PDA SPs are explored on the subcutaneous tumor model. As shown in Fig. 3d, the tumor tissue will possess the enhanced CT imaging signal after the intratumoral injection of BiVO4/Fe3O4@PDA SPs. In contrast, only normal bone structures can be observed without the injection of BiVO4/Fe3O4@PDA SPs. Meanwhile, the mouse treated by BiVO4/Fe3O4@PDA SPs displays a clear T2-weighted MR imaging in the tumor region comparing to the region without the SP injection (Fig. 3e). As for PA imaging, the PA signal of tumor is notably enhanced after intratumoral injection of BiVO4/Fe3O4@PDA SPs. As a comparison, only extremely weak PA signal arising from the tumor blood can be detected in the tumor site without the injection of BiVO4/Fe3O4@PDA SPs (Fig. 3f). The results above suggest the great potentials of BiVO4/Fe3O4@PDA SPs in multimodal imaging, which could combine advantages of each technique to provide complementary information for accurate diagnosis.
Thereafter, the in vitro synergistic therapeutic effect of BiVO4/Fe3O4@PDA SPs are evaluated via clonogenic assay (Fig. 4a and 4b). KB cells are treated by X-rays with different radiation doses (2 to 8 Gy) or NIR laser (0.33 W cm− 2) in the absence or presence of SPs (100 µg/mL). The result manifests that NIR alone and X-ray alone treatments can decrease the colony formation of KB cells to 88.2% and 61.3%, whereas SPs + NIR and SPs + X-ray can inhibit cell survival to 10.1% and 12.0%. Surprisingly, only 2.1% cells survive after the treatment of SPs + X-ray + NIR. Moreover, compared to the colony forming efficiency under X-ray treatment alone, the same therapeutic effect can be achieved under lower X-ray dose in the X-ray + NIR group, which strongly certify the considerable synergistic therapeutic efficacy between RT and PTT (Fig. 4c).
Next, 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA) fluorescent probe is employed to detect intracellular oxidative stress level of KB cells after different treatments.36 As shown in Fig. 4d, there is no detectable fluorescence in cells treated by PBS, SPs, NIR or SPs + NIR. In contrast, cells under X-ray, SPs + X-ray, SPs + NIR + X-ray treatments exhibit the bright fluorescence, and their fluorescence intensities are gradually enhanced. Since high radiation energy deposition and enhanced oxidative stress may facilitate the damage of DNA, γ-H2AX staining is performed to analyze the damage of DNA double-strand in cell nuclei quantitatively (Fig. 4e).37 As similar as the ROS assay, no visible fluorescence is found in cells without the X-ray treatment. The apparent fluorescence can be seen in cells under the treatments of X-ray, SPs + X-ray and SPs + NIR + X-ray, and the SPs + NIR + X-ray treatment produce the highest fluorescence intensity. Both DCFH-DA and γ-H2AX assays demonstrate the synergistic therapeutic efficacy of RT and PTT.
Motivated by the effective in vitro therapeutic outcome, the subcutaneous tumor model is employed to investigate the antitumor efficacy of BiVO4/Fe3O4@PDA SPs in vivo. The KB tumor-bearing BALB/c nude mice are randomly divided into 7 groups according to various treatments: (1) PBS, (2) SPs, (3) NIR, (4) X-rays, (5) SPs + NIR, (6) SPs + X-rays, (7) SPs + NIR + X-rays. Mice in group (3), (5) and (7) are irradiated by an 808 nm laser (0.33 W/cm2) for 10 min after the injection of BiVO4/Fe3O4@PDA SPs. Fig. S9 exhibits the IR thermographs of mice at different time intervals. The temperature of tumor tissue treated by SPs exhibits a rapid increase of 19 °C within 10 min, which is sufficient for tumor ablation. In contrast, there is no significant temperature elevation in the tumor without SPs injection. This dramatic difference lead to the remarkable localized overheat at the tumor site under NIR irradiation, causing severe tumor damage without influencing the adjacent normal tissues. Tumor volumes within 16 days in each group are recorded (Fig. 5a). The tumor volumes in group (1), (2) and (3) increase rapidly, suggesting the negligible effect of SPs alone and NIR alone treatments on the inhibition of tumor growth. Tumors in group (4) grow slowly comparing with those in group (1), revealing the irradiation of X-ray can only inhibit the tumor growth mildly. Despite SPs + X-ray and SPs + NIR treatments have the significantly inhibition on the growth of tumor at the initial stage of treatment, there are recurrences can be found after the treatment for about 10 days. Surprisingly, nearly complete tumor inhibition is realized in group (7) in the absence of recurrence. The weights and photographs of tumors exhibited in Fig. 5b and 5c further verifies that the synergistic therapeutic efficacy of RT and PTT are better than any single treatment. Because the weights of mice in each group are steady without distinct fluctuation, the side-effects of all the treatments during the therapeutic process can be excluded (Fig. 5e). According to the hematoxylin and eosin (H&E) staining images of tumor tissues (Fig. 5g), the death and nucleus rupture and ablation of cancer cells can be observed in tumors under SPs + X-ray and SPs + NIR treatments, whereas the tumors in group (7) have the most serious cell damage. This result further demonstrates the combination of RT with PTT can greatly improve the therapeutic effect compared to any single treatment. Besides the overlap of RT and PTT, the excellent synergistic therapeutic efficacy of BiVO4/Fe3O4@PDA SPs on tumor inhibition may come from the alleviation of hypoxia status in tumor tissues by boosting intratumoral blood circulation under NIR irradiation. Fig. S10 shows the in vivo PA images of tumors under various treatments: (1) PBS, (2) NIR, (3) SPs, (4) SPs + NIR, which imply that the photothermal conversion capability of BiVO4/Fe3O4@PDA SPs can remarkably increase tumor oxygenation, making tumor cells more sensitive to RT.
At last, the biosafety profile of BiVO4/Fe3O4@PDA SPs is evaluated by using BALB/c nude mice. H&E staining assays of major organs show that different treatments have no significant effects in the organ tissues of heart, liver, spleen, lung and kidney (Fig. S11). Serum biochemistry analysis exhibits the negligible side-effects on blood glucose and lipid, liver and renal function tests after the injection of SPs and combined treatments (Fig. S12). All these results mentioned above testify the excellent biocompatibility and powerful lethality of BiVO4/Fe3O4@PDA SPs.