Synthesis and Characterization of TPT@B12H12
Detailed information on catalyst preparation can be found in the Experimental Section. Fourier transform infrared spectra (Fig. 1a) revealed that characteristic peaks corresponding to stretching vibration of the B-H bond near 2480 cm− 1 in the spectra of Cs2B12H12 and TPT@B12H12, while TPT@B12H12 exhibited a series of characteristic peaks of TPT at 1600 − 750 cm− 1, indicating that TPT@B12H12 was successfully obtained. Indeed, originating from the anion-cation Coulombic attraction, the bonds of hydrogen and B-H...π have enhanced the interaction between closo-[B12H12]2- and TPT, thus contributing to the robust properties of TPT@B12H12. The transformation of TPT@B12H12 before and after UV irradiation can be observed from powder X-ray diffraction (PRXD) (Fig. 1b), that shows slight alteration of several characteristic peaks between 2θ = 10–15° and 22–30° after UV irradiation. Electron paramagnetic resonance (EPR) spectrum (Fig. 1c) demonstrated that the EPR signal was rather weak at the illumination time of 0 min, and the peak of the radical signal corresponding to [B12H12]-· appeared near G = 3490. Notably, compared with the pure free radical signal of [B12H12]-· (Figure S1) in our previous literature32, the [B12H12]-· free radical signal of TPT@B12H12 had a blue shift. Similarly, the free radical signal of [B12H12]-· gradually enhanced and blue shifted when the irradiation time was prolonged. When the illumination time was 12h, a significant signal of the boron cluster free radical appears at G = 3480 while the corresponding intensity was not significantly strengthened. Solid UV-vis spectra of TPT@B12H12 under different radiation indicated that TPT@B12H12 had response to light containing ultraviolet (Fig. 1d). Upon extended exposure to visible light, two UV-vis signals were observed around 535 nm and 725 nm. Moreover, after a short UV irradiation, the intensities of the two peaks notably increased in comparison with TPT@B12H12 under visible light conditions. These results indicated that TPT@B12H12 had a significant response in the UV region. Meanwhile, the color of the TPT@B12H12 solid gradually changed from light yellow to dark purple with the increase of UV-containing illumination time (Figure S2). Therefore, the charge transfer to TPT in the presence of UV light and the accompanying boron cluster radical generation led to a photochromic effect. X-ray photoelectron spectra (XPS) of TPT@B12H12 before and after UV illumination revealed that the B1s spectrum changed from a single characteristic peak (186.70 eV) to two characteristic peaks corresponding to 186.89 and 187.35 eV, respectively (Fig. 1e and f). Despite the presence of a series of changes, the field emission scanning electron micrographs (FESEM) and high-resolution transmission electron micrographs (HRTEM) indicated no significant structural changes of TPT@B12H12 after illumination, thus keeping its rod-like structures (Fig. 1g-i). Element distribution and the energy-dispersive X-ray spectroscopy (EDS) spectrum of TPT@B12H12 demonstrated that the three elements B, C, N were distributed rather homogeneously (Fig. 1j and k).
To elucidate the nature of catalytic efficiency of the developed catalyst TPT@B12H12, we have performed a set of calculations. It is well-known from the literature that hydrogen peroxide decomposes when it acquires an electron, like in the Fenton reaction.40 In our case, an electron transfers form the anion to the cation upon light absorption forming a radical pair Cat•2+ An•–. Herein, Cat•2+ tends to release the electron back; thus, this species is expected to demonstrate the catalytic activity.
The calculations of spin density distribution in Cat•2+ reveal the preferable location on one pyridinium cycle (Fig. 2a). Sterically, such location of the unpaired spin is very convenient since the pyridinium cycles are directed towards the most morphologically important surfaces (Fig. 2b). All other stable surfaces of the studied crystalline catalyst are illustrated in Figure S1 in the ESI. Thus, we have compared activation barriers for the hydrogen peroxide decomposition with and without catalyst (Fig. 2c). Note all the discussed energies include the zero-point vibrational energy correction term. As one can see, our catalyst decreases the activation barrier more than 3.5 times being more effective than the Fe2+ ion (the Fenton reaction), which decreases the barrier about twice.
The corresponding optimized geometries of R', TS' and P' are presented in Figure S2 in the ESI. We should stress that the level of P' is much higher than that of P, which is due to the presence of a further stage of the Fenton process. Herein, we ignore this fact for the time being since the first stage already has a higher barrier compared to our catalyst (Fig. 2c). Meanwhile, the structural changes of the H2O2…HN moiety are illustrated in Fig. 2d. As one can see, a simultaneous process of the O–O bond elongation and the hydrogen atom abstraction is observed and a subsequent formation of a hydroxyl radical and a water molecule takes place. Obviously, further chemical transformations include the water molecule desorption and hydroxylation of a benzene molecule. We assume here a synchronous process of the HO• addition and H• detachment near the bare nitrogen, which should immediately acquire the latter regenerating the native structure of the catalyst. The proposed scheme of the whole catalytic cycle looks thus as follows:
Cat•2+ + H2O2 = [Cat-H] 2+ + OH• + H2O
An EPR test of TPT@B12H12/H2O2 system was carried out to verify the outcome of the calculations. The results observed here demonstrated that TPT@B12H12 was stimulated by UV light irradiation and consequently has the capacity to decompose H2O2 into OH• (Fig. 2e).
TPT@B12H12 induced ROS-dependent apoptosis of melanoma in vitro
As mentioned above, TPT@B12H12 can decompose H2O2 to produce hydroxyl radical (•OH) with ultraviolet radiation. Hence, we sought to determine whether TPT@B12H12 triggers ROS production in tumor cells. We firstly investigated the effect of ultraviolet radiation on the ROS generation in cells, and found that the ROS levels generated by the B16 cells (murine melanoma cells) were dependent on the pre-treatment duration by ultraviolet radiation (Fig. 3a and 3b). We then determine the effect of dose on the ROS generation by the cells. The results from the DCFH-DA staining showed that TPT@B12H12 increased the production of ROS in B16 cells in a dose-dependent manner (Fig. 3c-3f).
To determine the cytotoxicity of the ROS produced by the TPT@B12H12, we performed the cell viability assay and confirmed that TPT@B12H12 significantly suppress the cell viability of B16 in a dose-dependent manner (Fig. 4a and 4b). Longer exposure of TPT@B12H12 to the ultraviolet radiation led to higher cytotoxicity (Figure S6). Vitally, the ROS scavenger N-acetyl-L-cysteine (NAC) could significantly decrease the ROS levels (Figure S5) and rescue the cytotoxicity of TPT@B12H12 (Fig. 4b), suggesting that TPT@B12H12 mediated cell cytotoxicity through ROS generation. Since ROS could trigger the cell apoptosis via mitochondrial related apoptotic cascade,42 we evaluated whether TPT@B12H12 could induce B16 cell apoptosis and affect the functions of mitochondrial. With flow cytometry, we confirmed that TPT@B12H12 could lead to a dose-dependent apoptosis (Fig. 4c and 4d, FITC+). With a JC-1 probe, we verified TPT@B12H12 treatment dramatically increased the accumulation of JC-1 monomers in B16 cells, indicating the decreased of mitochondrial membrane potential (Fig. 4e and 4f). Collectively, these data indicated that TPT@B12H12 lead to apoptosis of B16 through triggering the ROS generation.
ROS modulation is a widely used anti-tumor strategy based on the fact that H2O2 plays an important role in the tumor proliferation, invasion and metastasis,43 making tumor cells more vulnerable to oxidative stress by H2O2 compared to normal cells.44 CDT exerts its anti-tumor effect depending on the high level of H2O2, suggesting that CDT is selective to tumor cells.10 These data indicates that TPT@B12H12 can produce a high levels of ROS. This property is sufficient to induce tumor cell apoptosis, making TPT@B12H12 a novel kind of effective CDT agent.
TPT@B12H12 significantly suppressed melanoma growth in vivo
The antitumor effects of TPT@B12H12 were evaluated in the animal model. C57BL/6 mice were subcutaneously injected with 0.5×106 B16 cells, and when tumor volumes reached 100 mm3, mice were randomly divided into 2 groups, one group mice were administrated with saline, and the other group with TPT@B12H12 (20 mg/kg, every three days, i.p. injection). The tumor volumes were monitored during treatment. All mice were sacrificed at the day22. The blood, serum, and organs including kidney, livers, heart, spleen, and lung samples were harvested (Fig. 5a). TPT@B12H12 could significantly inhibit melanoma tumor growth (Fig. 5b and 5c). The side effects assessment of TPT@B12H12 were evaluated with the blood, serum, and important organs. TPT@B12H12 administration rarely affected the mice body weight (Fig. 5d). There was no significant difference for the levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and creatinine (CREA) between the two groups (Fig. 5e). Cells counting including white blood cell (WBC), lymphocyte (LYM), neutrophil (NEUT), hemoglobin (HGB) and blood platelet (PLT) indicated that TPT@B12H12 has limited hematologic toxicity (Fig. 5f). Consistently, HE staining showed that TPT@B12H12 could promote melanoma cell death and showed no obvious toxicity in important organs, including the heart, liver, spleen, lung, and kidney (Fig. 5g). These data support that TPT@B12H12 inhibited B16 tumor growth with limited side effects.
As mentioned, metal-based Fenton materials may have systemic toxicity problems, such as nervous system abnormalities and organ metabolic function loss.13,14 Currently, most CDT agents are iron-based materials, and iron overload is an important problem in the process of drug administration. The bioavailability of iron is scarce and the excretion of excess iron is very slow.45–47 Long-term exposure to excess iron may cause organ toxicity, secondary malignancies, and even organ failure.15,16,21 At the same time, the use of iron-based agents in cancer patients need a careful consideration, since patients who are undergoing cancer treatment are at high risk for iron loading, especially under the condition of transfusion and chemotherapy which can suppress erythropoiesis and increased exposure to toxic non-transferrin-bound iron.19,20,48,49 As patients with tumors usually are receiving continuous treatment, this situation can make physicians neglect the complications induced by iron toxicity.21 Other metallic agents (Cu, Mn, etc.) also face the same risk in clinical use.50–52 Thus, TPT@B12H12 is a metal-free catalyst with a limited biological toxicity and its use does obviously not cause any of the above mentioned concerns about the accumulation of metal toxicity in the body and subsequent organ failure.