Single Atom Iron‐Doped Graphic‐Phase C3N4 Semiconductor Nanosheets for Augmented Sonodynamic Melanoma Therapy Synergy with Endowed Chemodynamic Effect

Abstract Sonodynamic therapy (SDT) is a non‐invasive therapeutic modality with high tissue‐penetration depth to induce reactive oxygen species (ROS) generation for tumor treatment. However, the clinical translation of SDT is restricted seriously by the lack of high‐performance sonosensitizers. Herein, the distinct single atom iron (Fe)‐doped graphitic‐phase carbon nitride (C3N4) semiconductor nanosheets (Fe‐C3N4 NSs) are designed and engineered as chemoreactive sonosensitizers to effectively separate the electrons (e−) and holes (h+) pairs, achieving high yields of ROS generation against melanoma upon ultrasound (US) activation. Especially, the single atom Fe doping not only substantially elevates the separation efficiency of the e−‐h+ pairs involved in SDT, but also can serve as high‐performance peroxidase mimetic enzyme to catalyze the Fenton reaction for generating abundant hydroxyl radicals, therefore synergistically augmenting the curative effect mediated by SDT. As verified by density functional theory simulation, the doping of Fe atom significantly promotes the charge redistribution in the C3N4‐based NSs, which improves their synergistic SDT/chemodynamic activities. Both the in vitro and in vivo assays demonstrate that Fe‐C3N4 NSs feature an outstanding antitumor effect by aggrandizing the sono‐chemodynamic effect. This work illustrates a unique single‐atom doping strategy for ameliorating the sonosensitizers, and also effectively expands the innovative anticancer‐therapeutic applications of semiconductor‐based inorganic sonosensitizers.

entific Inc., US). Transmission electron microscopy (TEM) images were taken on a JEOL JEM-F200 transmission electron microscope (JEOL Ltd., Japan). The samples for TEM analysis were prepared by dipping the carbon-coated copper grids into ethanol solutions the samples and drying them under ambient conditions. Biological electron microscope images were taken on a Hitachi HT7800 electron microscopy (Hitachi Ltd., Japan). The thickness of the samples was measured by atomic force microscopy (AFM) on Bruker Dimension Icon (Bruker, Scientific Technology Ltd., US). The dissolved iron concentration was detected by an inductively coupled plasma-atomic emission spectrometer (ICP-AES) Agilent 7800 (Agilent Technologies Ltd., US). The structures of the materials were confirmed by X-ray powder diffraction (XRD) analysis on a Rigaku Ultima IV diffractometer (Cu Kα, λ = 1.5418 nm) ( Rigaku Co., Japan) with a scanning angle ranging from 5 ° to 90 ° of 2θ. The surface chemical composition and the binding information of required elements were characterized by the X-ray photoelectron spectroscopy (XPS) technique on a Thermo Scientific K-Alpha system (Thermo Fisher Scientific Inc., US). The electron spin resonance (ESR) spectrum characterization was performed on a Bruker A5000 EMX electron paramagnetic resonance spectrometer (Bruker, Scientific Technology Ltd., US). Ultraviolet-visible diffuse reflection spectra (UV-vis DRS) were performed on a UV-3600i Plus spectrophotometer (Shimadzu Co., Japan) with the wavelength ranging from 200 nm to 800 nm by using BaSO4 powder as the reference.
Fluorescence measurements were performed on a Edinburgh FLS1000 spectrometer (Edinburgh Instruments, England). Photoelectrochemical tests were conducted on a CHI760E electrochemistry workstation (Chenhua, Shanghai, China) with Ag/AgCl as the reference electrode and the counter electrode. Fluorescence microscope images were recorded by the Nikon Eclipse Ts2-FL (Nikon Co., Japan). Cell phagocytosis, apoptosis, and ROS were obtained by FACS Calibur flow cytometry (Becton, Dickinson, and Co., USA). Ultrasound (US) irradiation for sonodynamic therapy was conducted by an Intelect Transport Ultrasound (Well.d Medical Electronics Ltd., China).

Singlet oxygen ( 1 O 2 ) generation:
First of all, TEMP was used as a radical spin trap for 1 O 2 detected by ESR spectra. The five groups included C 3 N 4 , Fe-C 3 N 4 , US, C 3 N 4 + US, and Fe-C 3 N 4 + US groups were set to explore and compare the production of 1 O 2 , and the concentrations of C 3 N 4 and Fe-C 3 N 4 NSs were 80 μg mL -1 . 5 μL of TEMP (100 mM) was added into 100 μL of the Fe-C 3 N 4 /C 3 N 4 solution and then irradiated by US (1.5 W cm −2 , 1.0 MHz, 50% duty cycle) for 2 min. In the end, 2 µL of the reaction mixture was injected into the quartz capillary and measured by the ESR spectrum immediately.
Moreover, DPBF was typically used as a molecular probe to detect 1 O 2 generation. Fe-C 3 N 4 NSs (80 μg mL -1 ) were mixed with 40 µL DPBF (1 mg mL -1 ) and then irradiated by US (1.5 W cm −2 , 1.0 MHz, 50% duty cycle) for different durations in the dark (0, 1, 2, 3, 4 and 5 min). In the course of time, the absorption intensity change of DPBF at 423 nm was recorded by a Tecan Spark microplate reader (Tecan, M. nnedorf, Switzerland). For comparison, the ROS generation of pure DI-Water and C 3 N 4 triggered by US respectively acted as US group and C 3 N 4 + US group were also detected in the same way.
Hydroxy radical (•OH) generation: DMPO was used as a radical spin trap for •OH observed by ESR spectra. Typically, Fe-C 3 N 4 NSs (80 μg mL -1 ) and H 2 O 2 (100 µM) were mixed with 10 µL DMPO (100 mM) solution in the mildly acidic environment (pH 5.4) with or without US irradiation and then detected by ESR spectra at once. Moreover, H 2 O 2 with or without US irradiation and C 3 N 4 NSs mixed with H 2 O 2 were also performed with a similar approach for comparison.
In addition, TMB was used as a typical molecular probe to detect the •OH production.
Firstly, Fe-C 3 N 4 NSs (80 μg mL -1 ) were mixed with varying concentrations of H 2 O 2 (100 µM, min. Subsequently, Fe-C 3 N 4 NSs (80 μg mL -1 ) were mixed with H 2 O 2 (100 µM) and 5 µL TMB (0.1 mg mL -1 ) at pH 5.4, and then the mixture was further treated with or without US (1.5 W cm −2 , 1.0 MHz, 50% duty cycle) in the dark for 0, 1, 2, 3, 4 and 5 min. After the reaction of above mixture, the absorption change of TMB at 650 nm was recorded by the microplate reader. Besides, H 2 O 2 under US activation was performed in the same way for comparison.
Density functional theory calculations: All the geometry optimization and energy calculations were performed by using density functional theory (DFT) implemented in the Vienna ab initio simulation package (VASP). [4] The electron and core interactions were described using the frozen-core projected augmented wave (PAW) approach. Generalized gradient approximation (GGA) formulated by Perdew-Burke-Ernzerhof (PBE) was chosen for the exchangecorrelation between electrons. [5] A kinetic energy cutoff of 400 eV was used for the plane wave. The Brilliouin zone was sampled using 5×5×2 k-point Gamma mesh for the orthogonal supercell of g-C 3 N 4 containing 6 C and 8 N atoms to ensure adequate convergence. The van der Walls interaction was accounted for with the Grimme DFT-D3 correction. [6] The related structures were optimized until the energy differences converged within 10 −4 eV and the forces of all atoms were less than 0.01 eV/Å. Cellular uptake in vitro: B16F10 mouse melanoma cells were seeded in a 6-well plate at 2×10 5 cells in the logarithmic growth phase for 12 h. To discard the old medium after adhered plate of cells, and then added 2 ml Fe-C 3 N 4 /C 3 N 4 NSs (40 μg mL -1 ) which were dissolved in 1640 medium to the plate for different incubation times (1, 2, 4, 6 h). In the course of time, the cellular uptake of the NSs was observed by fluorescence microscope and biological electron microscope.