Novel photodynamic therapy using two-dimensional NiPS 3 nanosheets that target hypoxic microenvironments for precise cancer treatment

: Photodynamic therapy (PDT) is a highly promising modality against cancer, but its efficacy is severely limited by the low oxygen content in solid tumors. In this study, a smart photosensitive NiPS 3 nanosheet was developed to solve the problem of low oxygen to allow PDT to be performed against tumors. The photosensitized ROS generation mechanism of NiPS 3 is the photon-generated electron-hole pathway, which can generate O 2 ·− and · OH at the conduction band and valance band, respectively. More crucial is that · OH generation doesn’t need O 2 , and


Introduction
Photodynamic therapy (PDT) is a novel anti-cancer strategy in which photosensitizing agents selectively kill tumor cells by generating ROS when exposed to a suitable wavelength of light [1][2][3]. However, the therapeutic effects of PDT are significantly limited in a hypoxic tumor microenvironment since currently-available photosensitizers require oxygen to generate cytotoxic ROS [4,5], including superoxide radicals (O 2 ·− ), singlet oxygen ( 1 O 2 ), hydrogen peroxide (H 2 O 2 ) and hydroxyl radicals (·OH) [6]. Studies indicated that, even under severe hypoxic environment (2% O 2 ), some photosensitive materials can generate considerable O 2 ·− through type I photoreactions, and partial O 2 ·− is transformed to high toxic ·OH through SOD-mediated cascade reactions. These radicals synergistically damage the intracellular organelles, which subsequently trigger cancer cell apoptosis, presenting a robust hypoxic PDT potency [7]. ·OH radical is one of the most damaging radical of all oxidants, having an oxidation potential (2.8 V) second only to that of fluorine, and relatively less dependent on molecular oxygen compared with other radicals [6,8,9]. Furthermore, it can trigger a rapid chain reaction with the majority of organic molecules in cells due to its high oxidation potential, oxidizing them to CO 2 and H 2 O without the formation of secondary polluting reactants [10,11]. Therefore, the efficacy of PDT can be substantially improved by generating large quantities of ·OH.
Various strategies have been utilized to alleviate treatment of disease [12][13][14]. Transition metal sulfides with narrow bandgap energy of approximately 0.2-2.0 eV, have been widely used in photocatalysis [15,16]. One research found that NiPS 3 was a more cost-effective platform for advancing photocatalytic [17]. Photocatalysis for water splitting is an effective strategy to realize O 2 generation with H 2 O as a source [18,19]. The flexible application of such a strategy for endogenous oxygen production precisely solves the problem of hypoxia in tumor microenvironments during the PDT of solid tumors [20]. Recent research has shown that a new nickel-related photosensitizer Ni 3 S 2 /Cu1.8S, which has a narrow bandgap, can stimulate the creation of electron-holes when combined with near infrared (NIR) irradiation [21]. For photosensitizers, the position of the conduction band minimum (CBM) and valence band maximum (VBM) relative to the hydrogen and oxygen electrode potentials is relevant [14,22,23]. Zhang et al. and Chen et al. have shown that NiPS 3 nanomaterials are layered metal thiophosphites (MPS 3 ), and can operate as semiconductor photosensitizers. NiPS 3 can easily be prepared with high light stability compared with organic photosensitizers [24,25]. NiPS 3 has a puckered rather than planar structure, similar to that of phosphorene, two-dimensional (2D) NiPS 3 monolayers that function indirectly, with an energy bandgap of 1.63 eV [26]. NiPS 3 possesses [P 2 S 6 ] 4− clusters that 'ionically' interact with Ni 2+ within the material. Doping with NiPS 3 electrons (0.1 e− per atom) which results in an increased (8.01 eV) van der Waals gap while doping the holes causes the gap to be reduced (3.20 eV). The addition of an electron elongates the Ni-S bond length, while the opposite effect has been observed when doping the holes. Very little changes are observed in P-S and P-P bond lengths during electron and hole doping [27]. In addition, NiPS 3 nanomaterials have an adjustable bandgap that can generate O 2 ·− and ·OH when irradiated by laser light through the interaction of electron-hole pairs (e− -H+) and oxygen or water [22,28], making it possible to perform PDT both in normoxic and hypoxic conditions. NiPS 3 nanomaterials have gained considerable attention in PDT due to their biosafety, with a toxicity profile similar to that of black phosphorous, and significantly less toxic than CoPS 3 or FePS 3 , indicating good biocompatibility [29][30][31]. Owing to their extremely short diffusion range (∼20 nm) and half-life (<200 ns), ROS is more effective when produced in critical sites such as the mitochondria or nucleus rather than the cytoplasm or other organelles [32,33]. However, the majority of nanomaterials cannot pass through nuclear pores [34]. Mitochondria, on the other hand, are involved in multiple physiological and pathological processes [35], and therefore represent a viable therapeutic target. Furthermore, they are the cellular powerhouse that produces ATP to sustain cellular activities and survival, and also the primary site of ROS production. High levels of free radicals can trigger oxidative stress in the mitochondrial matrix, leading to the depolarization of mitochondrial membranes leading ultimately to apoptosis [36]. The mitochondrial electron transport chain is a major source of reactive oxygen species (ROS) and is also a target of ROS [37]. Meanwhile, the modification of positively charged triphenylphosphine (TPP) ligand on the surface of NiPS 3 provided the mitochondrial-targeting property for the NiPS3@TPP by taking advantage of negative mitochondrial membrane potential [38,39]. Consequently, the development of nano-systems that target organelles such as mitochondria during PDT is a promising therapeutic approach.
Thus, a safe and effective NiPS 3 -based PDT nanosystem was developed, able to produce a large quantity of ROS in both normoxic and hypoxic conditions. In the present study, surface modification with TPP drove NiPS 3 passage through the mitochondrial membrane via a delocalized positive charge, resulting in depolarization of the mitochondrial membrane and apoptosis of tumor cells [40]. More surprisingly, the cytotoxicity of NiPS 3 can be revealed only when stimulated by specific wavelengths of light. If the light stimulus was discontinued, the cytotoxicity of NiPS 3 also disappeared simultaneously. The strength of the NiPS 3 toxicity was related to the power of the external light source, making it possible to regulate the NiPS 3 @TPP nano-system in clinical applications.
while the mitochondrial membrane potential assay kit (JC-1), Annexin V-FITC/PI apoptosis detection kits, and primary and secondary antibodies were obtained from Thermo Fisher (USA). Enhanced chemiluminescent (ECL) detection system kits were obtained from Tanon (Shanghai, China). Female Balb/c mice (15-20 g) were purchased from the Institute of Model Zoology, Nanjing University and were bred in a specific pathogen-free (SPF) environment.

Preparation of NiPS 3 nanosheets
NiPS 3 nanosheets were prepared by the synthesis of bulk NiPS 3 crystals followed by electrochemical exfoliation, as described previously [22].

Characterization of NiPS 3 nanosheets
X-ray diffraction (XRD) was performed using a Philips X'Pert Pro Super diffractometer with Cu Kα radiation. X-ray photoelectron spectroscopy (XPS) was conducted using an ESCALAB MK II with Mg Kα as the excitation source. Atomic force microscopy (AFM) was performed using a Veeco DI Nanoscope Multi Mode V system. UV-Vis DRS was performed using a Hitachi U-3010 UV-Vis spectrometer, with absorption spectra recorded on a PerkinElmer Lambda 950 UV−vis−NIR spectrophotometer [22].

Mechanisms for ·OH production ESR experiment
The ESR experiment was primarily focused on the analysis of ·OH. In the ·OH experiment, 5 mg of NiPS 3 were dispersed in 5 mL water via ultrasound for 20 min. Argon was then bubbled through the solution for 20 min to remove oxygen. DMPO (100 mM) solution (also bubbled with argon to remove oxygen) was used for hybrid acquisition within 660 nm light. Simultaneously, the same experiment was performed in normoxic conditions as a control. All tests were conducted in weak acidic conditions (pH = 6.8). Finally, whether H+ was able to generate ·OH was determined [41].

Preparation of the NiPS 3 @TPP system
A 1:1 mass ratio of NiPS 3 to TPP was dissolved in anhydrous ethanol and fully dissolved while exposed to ultrasound for 30 min to form the NiPS 3 @TPP targeting nano-system. The zeta potentials of NiPS 3 and NiPS 3 @TPP were measured using a Malvern Mastersizer 2000 (Zetasizer Nano ZS90, Malvern Instruments Ltd, UK).

Cell culture
Huh-7 human liver cancer cells and LO 2 were obtained from the Shanghai Cell Bank, Chinese Academy of Sciences, and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% antibiotics (penicillinstreptomycin, 10,000 U/mL) at 37 • C within an atmosphere containing 5% CO 2 . Normoxic and hypoxic conditions were simulated by culturing the cells within 21 and 1% O 2 , respectively. Hypoxia preconditioning performed using tri-gas incubator (Smartor118, Ningbo Huayi Ningchuang Intelligent Technology Co., LTD).

In vitro toxicity of NiPS 3 nanosheets
Huh-7 and LO 2 cells were seeded at a density of 8000 cells per well in 96-well plates and incubated overnight. The cells were then incubated with NiPS 3 and NiPS 3 @TPP at different concentrations (0, 12.5, 25, 50, 100, and 200 ppm) for 12 and 24 h (n = 5). The culture medium was replaced with CCK-8 reagent, and the cells were incubated for an additional 1 h at a predetermined time. The absorbance at 450 nm was measured using a microplate reader (ELx808; BioTek). Untreated cells were used as control. Normoxic and hypoxic conditions were simulated by culturing the cells within 21 and 1% O 2 , respectively.

AO/PI staining
Huh-7 cells were seeded into 96-well plates and cultured overnight, then incubated with 100 ppm NiPS 3 or NiPS 3 @TPP for 6 h. After irradiation with a 660 nm laser at 0.3 W/cm for 10 min, the cells were stained with acridine orange (AO, green, live cells) and propidium iodide (PI, red, dead cells) purchased from Logos Biosystems (South Korea), as per standard protocols. The cells were rinsed three times with PBS and observed using a confocal microscope to assess cell viability.

In vitro photodynamic effects
Huh-7 cells were seeded into 96-well plates, cultured overnight, and then cultivated with NiPS 3 for 6 h. The oxygen content within the culture environment was adjusted in the incubator, normoxic and hypoxic conditions represented by 21% O 2 and 1% O 2 , respectively. 2,7-Dichlorodihydro-fluorescein·diacetate (DCFH-DA) probe dissolved in DMEM without FBS at a concentration of 1:1500 was added to the plates, then incubated for 30 min prior to irradiation with a 660 nm laser at 0.3 W/cm 2 for 10 min. The cells were rinsed three times with PBS then observed using a confocal microscope.

Scanning electron microscopy
Huh-7 cells were seeded in 6-well plates and cultured until 80% confluent. The culture medium was replaced with 2 mL fresh DMEM medium supplemented with 100 ppm NiPS 3 . Three groups: (A) NiPS 3 , (B) NiPS 3 @TPP, and (C) NiPS 3 @TPP + 660 nm, were tested, respectively. After incubation for 6 h, the experimental groups were irradiated with a 660 nm laser at 0.3 W/cm 2 for 10 min and cultured for a further 12 h. The cells were then harvested, fixed, and sectioned prior to observation using a scanning electron microscope.

Mitochondrial membrane potential (△ m) measurements
Huh-7 cells were seeded in 24-well plates at a density of 5 × 10 5 cells/well and cultured until 80% confluent. After incubation with the NiPS 3 @TPP nano-system for 6 h, then laser irradiated at 0.3 W/cm 2 for 10 min, the cells were washed with PBS then stained using a mitochondrial membrane potential assay kit (JC-1) (Thermo Fisher, M34152) in accordance with the manufacturer's instructions. The stained cells were analyzed by flow cytometry (BD FACSAria).

Annexin staining
Huh-7 cells were seeded in 24-well plates at a density of 5 × 10 5 cells/well and cultured until 80% confluent. After incubation with the NiPS 3 @TPP nano-system for 6 h, then laser irradiated at 0.3 W/cm 2 for 10 min, the cells were washed with PBS prior to staining using an Annexin V-FITC kit (Thermo Fisher,V13242), in accordance with the manufacturer's instructions. Early and late apoptotic cells were analyzed by flow cytometry (BD FACSAria).

-H 2 AX foci assay
Huh-7 cells were cultured on coverslips in six-well plates to 80% confluency, and then incubated with NiPS 3 or NiPS 3 @TPP for 4 h prior to irradiation with 660 nm laser light for 10 min at a light density of 0.3 W/cm 2 . Four hours after irradiation, the cells were fixed with 4% paraformaldehyde for 10 min. The cells were then permeabilized with 0.2% Triton X-100 for 15 min, incubated in blocking buffer (3% bovine serum albumin in PBS) for 45 min, then washed with PBS and incubated with a primary monoclonal antibody against γ-H 2 AX (Abcam, ab81299) at 4 C overnight. The cells were then washed with PBS and incubated with a FITC-labeled goat anti-rabbit IgG secondary antibody (Abcam, ab97050) for 1 h. The nuclei were additionally stained with DAPI for 10 min. γ-H 2 AX foci were imaged via confocal laser-scanning microscopy (CLSM).

In vitro bio-distribution
To detect the in vitro Bio-distribution, NiPS 3 and NiPS 3 @TPP were labeled with PEG-Cy7 and incubated with Huh-7 cells in a 27 mm confocal petri dish. After that, Hoechst 33,342 (beyotime.com, C1027) and Mito-Tracker Green (beyotime.com, C1048) were added for staining cell nucleus and mitochondria, respectively. The confocal microscopy (CLSM, Zeiss 710 NLO) was used to take fluorescent photographs of the stained cells.

In vivo biodistribution
For the in vivo fluorescence imaging experiments, the BALB/c (nu/nu) mice with Huh-7 tumors were divided into two groups (n = 3 in each group). NiPS 3 and NiPS 3 @TPP were labeled with PEG-Cy7. The mice were then injected via their tails vein with NiPS 3 /Cy7 and NiPS 3 @TPP/Cy7 (dose: 100 μL, 5 mg/kg bodyweight), respectively. An IVIS Spectrum (PerkinElmer) was used to obtain in vivo images of the mice at 12 and 24 h postinjection. A 740 nm wavelength light was used as the excitation source, and 760 nm emitted light was detected.
Afterward, mice were sacrificed, and the tumor, heart, liver, spleen, lung, and kidney were collected. The fluorescence in all organs was acquired using the IVIS Spectrum (PerkinElmer) imaging system.

In vivo toxicity evaluation
Female Balb/c mice (15-20 g) were purchased from the Institute of Model Zoology, Nanjing University. All experiments were performed in accordance with the guidelines of the National Regulations of China for the Care and Use of Laboratory Animals. The mice were injected with 3 mg/kg NiPS 3 or isotonic saline (100 μL), NiPS 3 was dissolved in saline and injected via tail vein. The duration of the toxicity test was 1, 7, or 14 days, respectively. On the last day, 0.8 mL of blood was sampled from the eyes of the mice. A blood routine test (BRT) was conducted at the Shenzhen Sun Yat-sen Cardiovascular Hospital. The heart, liver, spleen, lungs, and kidneys of the mice were resected and stained with hematoxylin and eosin (H & E) on the fourteenth day.

Establishing a xenograft liver tumor model and in vivo PDT
Huh-7 cells (1.0 × 10 6 cells, 100 μL) were injected into the right armpit of each mouse (as described for the in vivo toxicity evaluation experiment), and after the tumors had grown to 4-6 mm in diameter, the mice were randomly divided into one of 6 groups: (1) Control (saline); (2) 660 nm laser illumination; (3) NiPS 3 ; (4) NiPS 3 + 660 nm laser; (5) NiPS 3 @TPP; (6) NiPS 3 @TPP + 660 nm laser (n = 5 per group). The dosage of NiPS 3 @TPP particles was 3 mg/kg and sample was dissolved in saline and injected via tail vein. Ten hours after injection, the tumors were irradiated for 10 min with a 660 nm laser at 0.3 W/cm 2 . PDT was performed 3 times with gaps of 2 days between treatments. The volumes of tumors and body weights of the mice were recorded every 2 days for 14 days. The perpendicular diameters of the tumors were measured using calipers, and the volume (mm 3 ) was calculated as 0.5 × length × width 2 .

Statistical analysis
Data were presented as means ± standard deviation of three independent experiments. All statistical analyses were conducted using OriginPro-8 software. P-values < 0.05 were considered statistically significant.

Morphological and physical property characterization
In the present study, NiPS 3 @TPP produced ROS in mitochondria, which induced apoptosis, as displayed in the schematic diagram (Scheme 1). As shown in Figure S1, the NiPS 3 particles crystallized into space group C2/m (No. 12) with triclinic unit cell dimensions of a = 5.812 Å, Scheme 1: Graphic illustration describing the structure of NiPS 3 @TPP and its roles in hypoxic photodynamic therapy. b = 10.070 Å, c = 6.632 Å, and V = 371.2 Å 3 . This is an ideal structure for splitting water due to fully exposed S and P atoms that form an octahedron [22,28,43]. These parameters were shown by Li [22] and summarized in Supplementary Table 1. In this model, the Ni ions were immobilized on a [P 2 S 6 ] 4 -framework, with a metal layer sandwiched by distorted octahedral S layers forming a 2D structure stacked together via van der Waals forces [22]. These structural characteristics have important implications for its optical applications, which have been described in the journal, Nature etc. [27,44]. As shown in Figure 1A, the X-ray diffraction (XRD) peaks of both bulk NiPS 3 and ultrathin NiPS 3 nanosheet correspond to the standard XRD spectrum (JCPDS# 01-78-0499), with no additional peaks observed. This indicated that pure phase NiPS 3 was successfully synthesized and no impurities were introduced during electrochemical exfoliation. The ultrastructure of the NiPS 3 nanosheets was analyzed using scanning transmission electron microscopybright field (STEM-BF) and scanning transmission electron microscopy-high-angle annular dark field (STEM-HAADF). The entire crystal domain had a homogeneous and almost flawless structure ( Figure 1B and C). Furthermore, mapping by electron energy-loss spectroscopy in a transmission electron microscope (TEM-EELS) revealed a uniform distribution of Ni, P, and S elements over the entire exfoliated flake ( Figure 1D). The XPS spectra of the bulk crystals and nanosheets revealed peaks of Ni2p (∼885-850 eV), P2p (∼130.9 eV), and S2p (∼161.3 eV). Gaussian fitting indicated that the ultrathin NiPS 3 nanosheets displayed almost the same spectra as bulk NiPS 3 . Furthermore, peaks corresponding to oxidized P and S were not observed in the NiPS 3 nanosheets, suggesting that negligible oxidation had occurred ( Figure 1E and F). The optical absorption properties of a photosensitizer represent the key determinants of its photodynamic capabilities [45]. The photoelectric properties of NiPS 3 make it possible to utilize in optically controlled applications. The photo-response of NiPS 3 using 405, 516, 638, and 800.5 nm laser light are displayed in Figure 2A-D. It should be noted that the photo-response at 638 nm was beyond the individual response range, which suggested that this may be a key characteristic of NiPS 3 regarding light control. The NiPS 3 nanosheets absorbed light in the visible and near-infrared range (NIR), taking to account penetration during light therapy and exhibits absorption at 660 nm, NiPS 3 nanosheets is suitable for photocatalytic activity in PDT ( Figure 2E). To confirm the exact form of ROS, an ESR test was conducted. ESR spectroscopy further revealed the presence of DMPO-·OH under both normoxic or hypoxic conditions ( Figure 2F), indicating that irrespective of whether it was utilized in normoxic or hypoxic conditions, NiPS 3 in PDT was able to generate ·OH radicals. During hypoxia, with H+ generated by the NiPS 3 nano-particles oxidizing OH− in water to produce ·OH, resulting in effective PDT even at low oxygen levels.
The ESR spectrum of NiPS 3 displayed distinct electron-hole pairs, indicative of the presence of ·OH. As shown in Figure 2G, the signal intensity of light was evidently strong in comparison with the peak for the same sample in darkness, indicating the presence of H+. Intracellular ROS generation was further evaluated using Huh7 cells as typical cancer cells ( Figure 2I and J), Statistical analysis of fluorescence intensity data is shown in Figure 2H. Compared with the normoxic group, the hypoxic group also revealed DCFH-DA green fluorescence, demonstrating intracellular ROS was also producted in hypoxic condition. Compared with other 2D materials, NiPS 3 is more stable, less toxic, and has greater catalytic activity (Supplementary Table 2).

In vitro cytotoxicity
The biocompatibility and photodynamic effects of NiPS 3 were then evaluated under hypoxic conditions (1% O 2 ) to determine its feasibility as a photosensitizer for anti-tumor PDT. As shown in Figure 3A and B, in the absence of 660 nm laser irradiation, the viability of Huh-7 and LO 2 cells was not significantly affected by 0-200 ppm NiPS 3 , even when incubated with NiPS 3 for 24 h, indicating minimal cytotoxicity of the NiPS 3 nanosheet, a result consistent with previous research [29,46]. As shown in Figure 3C, its   optimal power was achieved by irradiating at 0.3 W/cm 2 for 10 min, a setting used for subsequent experiments. As shown in Figure 3D, NiPS 3 @TPP displayed a potent photodynamic effect against Huh-7 cells when irradiated with a 660 nm laser in a time and concentration-dependent manner. To directly observe cell death, acridine orange (AO, green) and propidium iodide (PI, red) were used to label living and dead cells. As displayed in Figure 3E and F, with increasing concentrations of NiPS 3 @TPP, green fluorescence was attenuated and red fluorescence increased in intensity, indicating cell death due to the generation of ROS. Image J was used to analyze the fluorescence (Red) intensity of dead cells, and the data analysis was shown in Figure 3G. To exclude the possibility of any inherent cytotoxicity of NiPS 3 @TPP, a proportion of the culture dishes were irradiated with 660 nm laser light only for 10 min, and the proportion of live and dead cells measured after staining with AO/PI. Only a few dead cells were observed in the non-photodynamic region, whereas almost all cells had died in the PDT region, as shown in Figure 3H. Similar results were obtained using electron microscopy, as shown in Figure S2. Under an inverted microscope, the effects of different concentrations of NiPS 3 @TPP on cell growth were observed. The results indicated that NiPS 3 nanosheets exhibited low toxicity toward Huh7 cells. Taken together, the results suggested that NiPS 3 nanosheets were biocompatible and effective at killing cells using PDT.

Tumor-targeting efficacy
Due to the extremely short diffusion range and half-life of ROS, greater damage is caused if it is produced within key organelles such as mitochondria rather than in the cell membrane or cytoplasm. Mitochondria are the optimum site for PDT since the decrease in mitochondrial membrane potential (MMP) following excessive ROS production is irreversible, triggering apoptosis that can enhance the effect of PDT. Furthermore, as tumor cells require high quantities of energy for proliferation and drug resistance, targeting mitochondria can significantly impair their bioenergetic status [47]. Finally, mitochondrial targeting can also induce endogenous apoptosis by promotion of the release of cytochrome C from the depolarized membrane, therefore sensitizing the cells to PDT. Selective mitochondrial uptake of a drug can be achieved in vivo by linking a lipophilic cation to the molecule [48]. Therefore, TPP was attached to the surface of NiPS 3 to target the mitochondria of tumor cells. This enhanced the efficiency of ROS in PDT. The stability of NiPS 3 and NiPS 3 @TPP is shown in Figure 4A, their zeta potentials in Figure 4B, and the absorption electron cloud diagrams of TPP and NiPS 3 NiPS 3 in Figure 4C.
In the present study, NiPS 3 @TPP produced ROS and induced apoptosis. Therefore, we explored the intracellular localization and photocatalytic activity of the NiPS 3 @TPP targeting system under hypoxic conditions. As shown in Figure 4D, Huh7 cells were incubated with NiPS 3 , NiPS 3 @TPP, and NiPS 3 @TPP + 660 nm. In NiPS 3 group, the nanomaterials were mainly swallowed by lysosomes and randomly dispersed in cytoplasm. In the NiPS 3 @TPP group and the NiPS 3 @TPP + 660 nm group, nanomaterials showed obvious enrichment in mitochondria, and the nanomaterials were distributed in or around the inner mitochondrial membrane. However, when Huh7 cells were incubated with NiPS3@TPP + 660 nm, their mitochondrial inner membrane collapsed and shrank, with large vacuoles formed in the inner cavity and the outer membrane ruptured.
Intracellular trafficking of NiPS 3 @TPP in Huh-7 cells was investigated by CLSM. Cy7-labeled NiPS 3 @TPP was efficiently internalized and localized in mitochondria compartments after incubation for 12 h, as demonstrated by colocalization of Cy7 with mitochondria label fluorescence, shown as yellow fluorescence in the merge image ( Figure 5A). Compared with NiPS 3 , the mitochondria targeting of NiPS 3 @TPP were significantly increased, Fluorescence intensity quantified for Cy7 in vitro targeting were shown as ( Figure 5D).
A Huh-7 xenograft tumor model was established for the investigation of in vivo therapeutic effects of NiPS 3 @TPP to explore its potential for further pre-clinical application. The in vivo biodistribution and tumor accumulation of NiPS 3 @TPP/Cy7 after intravenous injection were determined via in vivo fluorescence imaging. As shown in Figure 5B, in NiPS 3 /Cy7 group, fluorescent signals reduced significantly after 24 h, this was probably caused for the reason that the tumor was rich in blood supply and had relatively high blood drug concentration effect for a short time, but due to the lack of targeting, it was soon carried to other parts by the blood flow. The NiPS 3 @TPP/Cy7 group exhibited relatively high and stable accumulation levels in tumors at both measured time-points, which could generally be attributed to the enhanced permeability and retention effect (EPR). After reaching the blood vessels around the tumor, due to the mitochondrial targeting effect of TPP, the material could quickly pass through the blood vessels and the tumor cells, and then riveted in the tumor cells. In addition to fluorescent signals detected at tumor sites, they were also detected in the major mouse organs including the liver, kidney, and lung, shown as

Mitochondrial targeting mechanism research
Previous research have found that loss of mitochondrial membrane potential (MMP), mitochondrial depolarization, and cytochrome c (Cyt c) release induced cell apoptosis [49,50]. Therefore, we investigated the PDT effect of NiPS 3 @TPP by measuring MMP and the rate of apoptosis. In normoxic conditions (21% O 2 ), experimental group therapy at 660 nm, 0.3 W/cm 2 for 10 min, the MMP of the experimental group ( Figure 6B) decreased 69.32% compared with the control group ( Figure 6A). In hypoxic conditions (1% O 2 ), the MMP of the experimental group ( Figure 6D) decreased 56.23% compared with the control group ( Figure 6C). Although the decrease in MMP within hypoxic conditions was less than that observed in normoxia, the decrease was nevertheless significant. Similar results were also observed for the experiments of apoptosis. In normoxia, the proportion of dead and early/late-stage apoptotic cells in the experimental group ( Figure 6F) was 52.09% higher than the control group ( Figure 6E). In hypoxia (1% O 2 ), the proportion of dead and early/late-stage apoptotic cells in the experimental group ( Figure 6H) was 32.85% higher than that of the control group ( Figure 6G). A significant increase in the expression levels of cytochrome C and pro-apoptotic cleaved-caspase 3, with a concomitant decrease in pro-survival Bcl-2 ( Figure 6I) was observed.
Furthermore, γ-H 2 AX was detected in the PDT group, indicative of DNA double-strand breaks caused by ROS ( Figure 6J). Taken together, the results indicated that the NiPS 3 @TPP nano-system was capable of catalyzing the generation of ROS within both normoxic and hypoxic conditions.

In vivo anticancer experiments
The therapeutic potential of the NiPS 3 @TPP nano-system was finally tested in a murine tumor-bearing model. As shown by H & E staining ( Figure 7A), NiPS 3 @TPP injection did not cause apparent organ damage or inflammation. In addition, almost all blood routine examination (BRT) parameters ( Figure S3) and body weight ( Figure 7B) of NiPS 3 -treated nude mice were similar to those of the untreated controls. Minimal toxicity of NiPS 3 @TPP at the tested dose indicated that it was a promising biocompatible photosensitizer for in vivo biomedical applications. Therefore, we evaluated an enhancement of PDT of the NiPS 3 @TPP targeted system against solid tumors. Perpendicular diameters of the tumors were measured using calipers. Relative tumor size in the different groups and time points were displayed in Figure 7C. When the tumor volumes reached approximately 100 mm 3 , this experiment was conducted within six experimental groups: (1) normal saline, (2) 660 nm laser illumination, (3) NiPS 3 , (4) NiPS 3 + 660 nm laser, (5) NiPS 3 @TPP, (6) NiPS 3 @TPP + 660 nm laser. After injection of NiPS 3 @TPP, PDT with 660 nm laser light (0.3 W/cm 2 , 10 min) via tail vein was performed 3 times with gaps of 2 days between treatments, as shown in Figure 7D and E, the tumors of the mice treated with the NiPS 3 @TPP nano-system and 660 nm laser irradiation were significantly smaller compared with those in untreated mice. By contrast, NiPS 3 , NiPS 3 @TPP, or 660 nm laser light alone had no significant effect on the tumor volume relative to the control. Therefore, consistent with in vitro results, NiPS 3 @TPP can effectively kill tumor cells by PDT using 660 nm laser light stimulation.

Discussion
PDT is severely limited by the hypoxic environment in a tumor. Current strategies focus primarily on the improvement of the intratumoral O 2 perfusion, while clinical trials suggest that O 2 enrichment may promote cancer cell proliferation [3]. ·OH are strong oxidizers that can oxidize and hydroxylate multiple biological macromolecules, including unsaturated fatty acids, sugars, enzymes, or related proteins, which eventually cause extensive cellular damage [51,52]. The stimulus-response of nanomaterials is promising for precision medicine [53]. As displayed in Figure 2, NiPS 3 nanosheets absorbed light in the visible and nearinfrared range, the absorption wavelength of which can be used as the key to control the biological effect of NiPS 3 nanosheets. It was non-toxic in absense of being activated by light, while ROS which is highly cytotoxic was generated after light activation. Previous reports suggest that NiPS 3 nanosheets with only a few layers represent promising catalysts [54]. NiPS 3 nanosheets generated high levels of ·OH through 660 nm photocatalysis under both normoxic and hypoxic conditions. As shown in Figure 2F, there were 4 distinct characteristic peaks with magnitudes of 1:2:2:1, suggesting the generation of ·OH. When the dissolved O 2 was removed by N 2 , the generation of ·OH decreased, but were maintained at a steady level. As shown in Figure 2G, the ESR spectrum of NiPS 3 also demonstrated a separation of electron-hole pairs. Within normoxic conditions, photoexcited electrons from NiPS 3 mainly captured O 2 to form O 2 ·− and ·OH, while in hypoxic conditions, H+ generated by NiPS 3 was more likely to oxidize OH− to produce ·OH.  [56,57]. We surmise that apart from ·OH, NiPS 3 catalyzes other reactions such as the decomposition of H 2 O 2 to produce ROS in a tumor microenvironment. Effective induced programmed cell death or apoptosis has been a mainstay and goal of clinical cancer therapy. The process of apoptosis can be divided into extrinsic or intrinsic pathways [58]. The extrinsic pathway, dependent on pro-death signals from outside a cell occurs mainly by cell surface receptors, whereas the intrinsic pathway is triggered by mitochondrial events [59,60]. Mitochondrial targeting in oncotherapy results in a two-pronged cellular damage strategy that destroys the energy supply, stalling proliferation and inducing apoptosis [61]. Nanoparticles with a high net positive charge, such as TPP cations, have the potential to promote endosomal escape and can participate in delivering chemotherapeutic drugs to mitochondria [59]. In the present study, TPP was attached to the surface of NiPS 3 to target mitochondria. Thus, ROS catalyzed by NiPS 3 induces mitochondrial outer membrane permeabilization, representing the decisive event that irrevocably commits Huh7 cells to die. In addition, cancer cells have a higher MMP compared with normal cells, conducive to selective targeting [62]. Therefore, we introduced TPP onto the surface of NiPS 3 to precisely target the mitochondria in Huh7 cells. TPP cations have a strong electric field that promotes their rapid uptake [35]. Since the mitochondrial inner membrane potential is 150-160 mV (negative inside) and that of the cancer cell plasma membrane is 30-60 mV (negative inside), mitochondria can absorb large quantities of cations, and so able to drive the uptake of NiPS 3 @TPP. The Nernst equation indicates that at 37 • C, for every 61.5 mV that the membrane potential increases, the number of singly charged cations increases 10-fold [35,63]. These properties can promote the absorption of NiPS 3 @TPP nanomaterials by tumor cells. From Figure 4A-C, we demonstrate that the NiPS 3 @TPP nano-system was successfully prepared. The results of scanning electron microscopy ( Figure 4D) demonstrated that NiPS 3 @TPP provided superior targeting efficiency than NiPS 3 . Huh7 cells were incubated with NiPS 3 , NiPS 3 @TPP, and NiPS 3 @TPP + 660 nm. In Huh7 cells incubated with NiPS 3 , the nanomaterials were mainly swallowed by lysosomes and randomly dispersed in cytoplasm; the content of nanomaterials in mitochondria was not high. Huh-7 cells had clear and intact nuclear membranes, visible nucleoli, and complete mitochondria without indentations on the inner ridge, all of which indicated healthy, metabolically active cells. In the NiPS 3 @TPP group and the NiPS 3 @TPP + 660 nm group, nanomaterials showed obvious enrichment in mitochondrial, and the nanomaterials were distributed in or around the inner mitochondrial membrane, as indicated by the red triangle in the figure. However, in NiPS 3 @TPP, the mitochondrial inner membrane was intact, while in NiPS 3 @TPP + 660 nm group, the mitochondrial inner membrane collapsed and shrank, with large vacuoles formed in the inner cavity and the outer membrane ruptured, as indicated by the white triangle in the figure. Combine with Figures 2  and 3, we demonstrated that NiPS 3 @TPP was shown to have good biosafety. The results were also verified in animal experiments, as shown in H & E staining ( Figure 7A) and routine examination (BRT) parameters ( Figure S3). Mitochondria were observed to show apparently morphological and functional destruction such as the loss of cristae, vacuolization, and even mitochondrial membrane rupture after photodynamic therapy following PDT ( Figure 4D). Following this, MMP declined and mitochondrial permeability transition pores (MPTP) increased.
This process preceded cytochrome-c release from the mitochondria into the cyto-sol, which activated caspase to trigger apoptosis [64]. As shown in Figure 6A-H, the MMP decreased and rate of apoptosis increased after PDT both in normoxic and hypoxic conditions. Bcl-2 is a membrane protein located principally on the outer membrane of mitochondria. Its overexpression prevents cells from undergoing apoptosis, while cytochrome c is required for the initiation of the apoptotic program [65,66]. Caspase-3 acts as an executioner factor and triggers the apoptosis process by cleaving its protein substrates. Cleavage of caspase-3 is considered a reliable marker of cell death by apoptosis [67,68]. As shown by the results of Western blotting ( Figure 6J), cytochrome c and cleaved caspase-3 were the highest in the NiPS 3 @TPP + 660 nm laser group, but Bcl-2 expression was the least among these three treatments. DNA damaging agents, such as ROS may activate both membrane death receptors and endogenous mitochondrial damage pathways leading to apoptosis [69]. As shown in Figure 6I, compared to the control group, the NiPS 3 @TPP + 660 nm group clearly exhibited double-strand breaks (DSBs), indicating severe photodynamic damage.
Following in vivo therapy for specific durations, as shown in Figure 7A, the heart, liver, spleen, lungs, and kidneys in the experimental group displayed no pathological changes compared with the control group. As shown in Figure 7B, there was a slight increase in body weight, and furthermore, no animals died in any in vivo experiment, suggesting that there were negligible toxic side effects for these treatments. This conclusion is consistent with previous in vitro experiments. As shown in Figure 7C-E, the tumor volumes of the photodynamic groups (Groups 4 and 6) were significantly smaller than those in other groups. The therapeutic effect of the targeted modification group (Group 6) was significantly superior to that of the non-targeted modification group (Group 4). In view of the observations described above, we conclude that the NiPS 3 @TPP has potential for applications in nano-medicine.

Conclusions
We developed a photosensitizer NiPS 3 @TPP that can produce ROS independent of oxygen and thus can kill cancer cells. The photosensitized ROS generation mechanism of NiPS 3 is the photon-generated electron-hole pathway, which can generate O 2 ·− and ·OH at the conduction band and valance band, respectively. More crucial is that ·OH generation doesn't need O 2 , and the O 2 ·− can also work in a low O 2 environment, and depleting oxygen in tumor cells.
In a hepatoma tumor model, NiPS 3 @TPP system-based PDT significantly inhibited tumor growth and displayed low biotoxicity. Thus, NiPS 3 is a promising next-generation photosensitizer representing a simple and effective nanoenhancer for PDT. All animal experiments were conducted in accordance with the standard guidelines for the care of animals, as approved by the Welfare Committee of the Center of Experimental Animals (Jinan University, Guangzhou, Guangdong, China). Data availability statement: All data generated or analyzed during this study are included in the published article (and its supplementary information files).