Materials

HCl (36–38%), LiF, Ti3AlC2 (325-meshes), and TPAOH were supplied by Beijing Chemical Reagents Company (Beijing, China), Aladdin Bio-Chem Technology Corporation (Shanghai, China), Forsman Scientific Company (Beijing, China), and J&K Scientific Corporation (Beijing, China), respectively. Oleic acid (OA), oleylamine, thioacetamide, bismuth neodecanoate, plysorbate 20 (Tween 20), cyclohexane, mPEG-NH2 (Mw\(\approx\)2K), N-(3-dimethylaminopropyl)-N’ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), TPP, dimethyl sulfoxide (DMSO), 5,5-dimethyl-1-pyrroline N-oxide (DMPO), 2,2,6,6-tetramethylpiperidine (TEMP), CPZ, and colchicine were purchased from Sigma-Aldrich (MO, USA). MTT, DCFH-DA, mitochondrial membrane potential assay kit (JC-1), cell lysis buffer (RIPA), Annexin V-FITC/propidium iodide (PI) apoptosis and necrosis detection kit, H&E staining kit, and TUNEL apoptosis assay kit were obtained from Beyotime Institute of Biotechnology (Shanghai, China). Calcein Acetoxymethyl ester (Calcein-AM)/PI double stain kit was brought from Yeasen Biotechnology Corporation (Shanghai, China). PBS (pH = 7.4, 10 mM), fetal bovine serum (FBS), trypsin, penicillin-streptomycin (PS), reduced serum medium (Opti-MEM), DMEM, mitochondrion-selective probe (Mito-Tracker Red), and red hypoxia probe (Image-iT™) were provided by Gibco Life Technologies (CA, USA). All the polyclonal antibodies were obtained from Proteintech Group Incorporation (IL, USA). All chemicals were used as received without further purification.

Synthesis of Bi2S3/Ti3C2

Ti3C2 NSs were prepared in a typical way of etching the Al atomic layer from Ti3AlC2 with HCl/LiF. Briefly, LiF (5 g) was dissolved in a HCl aqueous solution (50 mL, 9 mmol) contained in a polytetrafluoroethylene (PTEF) beaker and magnetically stirred for 30 min at room temperature to acquire a hydrogen fluoride (HF) aqueous solution. HF and HCl are both highly corrosive liquids, and the experimenters must strictly follow the safety protocol during the experiment. Then, Ti3AlC2 powders (5 g) were slowly added in the HF aqueous solution and stirred at \(38^\circ C\) for 3 days. After being washed and centrifuged repeatedly with deionized water (DI) or ethanol (11,000 rpm \(\times\) 10 min) to make the supernatant pH close to 7, the collected precipitates were dispersed in 25 mL TPAOH with stirring at room temperature for 3 days. The resulting ultra-thin Ti3C2 NSs were collected by freeze-drying after washing and centrifugation with DI or ethanol several times (11,000 rpm \(\times\) 10 min). Particularly, Bi2S3/Ti3C2 NHs were fabricated by a solvothermal synthetic method in situ. Initially, Ti3C2 NSs (0.25 g) were evenly dispersed in ethanol (10 mL) with an OA (20 mL) addition, and bismuth neodecanoate (1.45 g) was added to the mixed solution under continuous agitation, which made the bismuth ion (Bi3+) attached to the Ti3C2 NSs by electrostatic absorption. Subsequently, an oleylamine solution (4 mL) with thioacetamide (0.15 g) dissolved in it was rapidly added to the above solution with violent stirring at room temperature for 1 h. Eventually, the prepared mixture was sealed in a Teflon-lined autoclave (50 mL) and heated at 150 \(^\circ C\) for 10 h, which was required to cool to room temperature naturally at the end of the reaction. After being washed and centrifuged with ethanol (10,000 rpm \(\times\) 5 min) three times, the oil-soluble OA-coated Bi2S3/Ti3C2 NHs were obtained by freeze-drying. Similarly, Bi2S3 NPs were fabricated without the addition of Ti3C2 NSs.

Fabrication of Bi2S3/Ti3C2-TPP

In order to connect TPP to Bi2S3/Ti3C2 NHs, oil-soluble OA-coated Bi2S3/Ti3C2 NHs must be converted to water-soluble NHs and encapsulated with mPEG-NH2. Firstly, OA-coated Bi2S3/Ti3C2 NHs (100 mg) and Tween 20 (150 \(\mu\)L) were uniformly dispersed in cyclohexane (20 mL) with stirring at room temperature for 1.5 h. After that, the mixture was added dropwise into DI (30 mL) in a 70 \(^\circ C\) water bath and continuously stirred for at least 3 h to evaporate the cyclohexane. Later, the water-soluble Tween 20-functionalized Bi2S3/Ti3C2 NHs were collected by freeze-drying after washing and centrifugation with ethanol (10,000 rpm \(\times\) 5 min) three times. Then, mPEG-NH2 (20 mg) and Tween 20-functionalized Bi2S3/Ti3C2 NHs (20 mg) were sufficiently dispersed in DI (10 mL) and stirred overnight at room temperature, while EDC (50 mg) and NHS (25 mg) were added into an aqueous solution (10 mL) containing TPP (50 mg) and stirred in the dark at room temperature for 6 h to activate TPP. Finally, the two prepared solutions were mixed with stirring overnight in the dark, and Bi2S3/Ti3C2-TPP was obtained by freeze-drying after washing and centrifugation with DI (10,000 rpm \(\times\) 5 min) three times.

Characterization

SEM (1000000X, TESCAN, CZ), TEM (JEM-2100, JEOL, JPN), and AFM (Dimension Fastscan, BRUKER, GER) were used to describe the morphology and structure of the samples. XRD (X’PERT, Panalytical, NL), XPS (ESCALAB 250Xi, ThermoFisher, USA), and FT-IR (Nicolet is50, ThermoFisher, USA) were conducted to determine the composition of the samples. ICP-MS, PL, UV-vis, and ESR spectroscopy were collected on an iCAP7400 spectrometer (ThermoFisher, USA), a FLS1000 spectrophotometer (Edinburgh, UK), a LAMBDA950 spectrophotometer (PE, USA), and an A200S-95/12 spectrometer (BRUKER, GER), respectively. The average size and zeta potential of the samples were measured by Nano-MS2000/ZS90 (Malvern, UK). The NIR source was purchased from LeiShi (808 nm, China). Oxygen concentration and temperature were severally monitored by a dissolved oxygen meter (JPBJ-607 A, INESA, China) and a portable thermal imager (E5-XT, FLIR, USA). All fluorescent imaging was visualized by an inverted fluorescence microscope (AF6000, Leica, GER). Cell viability was evaluated by a microplate reader (ELX800, Bio Tek, USA), while a flow cytometer (NovoCyte, Agilent, USA) was used to conduct quantitative flow cytometry. Western blot exposure was performed on a chemiluminescence imager (ChemiDoc Touch, Bio-Rad, USA).

Photodynamic performance and O2 evolution evaluation

ROS assay kits were used to detect the ROS generation capacity of Bi2S3/Ti3C2, Ti3C2, and Bi2S3 under 808 nm laser irradiation. Initially, the DCFH-DA was diluted with ethanol to acquire a solution with a molar concentration of 1 mM, and the mixture (0.5 mL) was added to the NaOH solution (2 mL, 10 mM) with stirring at room temperature for 30 min in the dark. After that, the obtained mixed solution was uniformly dispersed in PBS (10 mL, pH = 7.4, 10 mM) and stored at 4 \(^\circ C\) in a refrigerator. Then, the prepared DCFH-DA solution (100 \(\mu\)L) was mixed with the sample solution (100 \(\mu\)L, 200 \(\mu\)g mL-1) and irradiated with an 808 nm laser (1 W cm-2, 10 min). Meanwhile, a PL spectrophotometer was used to measure the fluorescence intensity of DCF per minute in order to determine how ROS generation changed over time.

ESR measurements were performed to identify the ROS species generated by Bi2S3/Ti3C2 under NIR irradiation. Briefly, TEMP as a spin-trapping agent was added to the methanol solution of Bi2S3/Ti3C2 (500 \(\mu\)L, 100 \(\mu\)g mL-1) to reach a molar concentration of 200 \(\mu\)mol L-1, and the mixture was irradiated with an 808 nm laser (1 W cm-2, 10 min). Whereafter, an ESR measurement was immediately conducted to determine whether 1O2 was formed. The formation of ·OH and ·O2- was tested by the same experimental procedure in DI and methanol, respectively, and DMPO was used as the spin-trapping agent.

Nitrogen (N2) was blown into the Bi2S3/Ti3C2 solution (10 mL, 100 \(\mu\)g mL-1) for 15 min to remove the inherent O2 in it, and the solution was sealed with liquid paraffin. After that, the amount of O2 produced by the samples under 808 nm laser irradiation (1 W cm-2) was measured per minute within 20 min, utilizing a dissolved oxygen meter.

Photothermal performance evaluation

To evaluate the photothermal effect of the samples, Bi2S3/Ti3C2, Ti3C2, or Bi2S3 aqueous solutions (4.5 mL, 100 \(\mu\)g mL-1) were added to a tailor-made quartz cuvette and irradiated with an 808 nm laser (1 W cm-2, 10 min). A thermal imager was used to record temperature changes every 30 s, until it reached room temperature approximately. Simultaneously, the above experimental procedure was continuously repeated five times to further investigate the photothermal stability of Bi2S3/Ti3C2. Subsequently, the temperature changes of Bi2S3/Ti3C2 aqueous solutions with different concentrations (0, 25, 50, 100, and 200 \(\mu\)g mL-1) were recorded within 10 min under the same conditions, aiming to determine the influence of sample concentration on the photothermal performance.

Enrichment analysis of hypoxic-related genes in glioma

Firstly, gene expression and clinical data from glioma samples were gathered through the TCGA and CGGA databases. Then, the hypoxic-related gene set was chosen from The Molecular Signatures Database (MSigDB, http://www.gsea-msigdb.org/gsea/msigdb/human/search.jsp). The samples were classified into low-grade glioma and glioblastoma based on tumor grade, and the enrichment degree of hypoxic-related genes in the two samples was determined by GSEA. Whereafter, the R package of gene set variation analysis (GSVA) and ssGSEA were used to acquire hypoxic-related functional scores for each glioma patient sample in the TCGA and CGGA databases. Finally, the scores were analyzed in conjunction with tumor grade and survival time to explore their relationship.

Cellular experiments

U251 human glioma cells were selected for the following cellular investigations and cultured according to the standard protocol. For cell culture in a hypoxic microenvironment, the conditions were set as a humidified atmosphere containing 1% O2 and 5% CO2 at 37 \(^\circ C\), which was formed by flowing N2.

For cellular uptake, 8 \(\times\) 104 U251 cells in 1 mL of DMEM were seeded in each well of a 12-well plate for 24 h of incubation. Then, cells were treated with Opti-MEM containing Bi2S3/Ti3C2-TPP or Bi2S3/Ti3C2 (12.5, 25, 50, 100, or 200 \(\mu\)g mL-1) for 4 h. After washing with PBS three times, cells were digested with trypsin and collected by centrifugation (1000 rpm \(\times\) 5 min, 4 \(^\circ C\)). Finally, cells were counted, and the Bi or Ti content was measured by ICP-MS tests.

For the hemolysis assay, 0.5 mL of blood was obtained from mice, and erythrocytes were extracted by centrifugating and washing with PBS five times (3000 rpm \(\times\) 5 min, 4 \(^\circ C\)) and diluted at 10 times their initial volume in PBS. Then, diluted suspensions (200 \(\mu\)L) were mixed with 800 \(\mu\)L of PBS, Triton-100 (0.025% in PBS), and Bi2S3/Ti3C2-TPP (25, 50, 100, 200, and 300 \(\mu\)g mL-1 in PBS) and incubated with cells at 37 \(^\circ C\) for 2 h. All samples were centrifuged (3000 rpm \(\times\) 5 min, 4 \(^\circ C\)), and the supernatant absorbance at 541 nm was detected using a microplate reader. The hemolysis rate was calculated by dividing the supernatant absorbance value (I) of the samples by the value (I0) of the positive control group. The calculation formula is as follows:

Hemolysis rate (%) = (I/I0) \(\times\) 100%

For mitochondrial targeting performance, 1.6 \(\times\) 105 U251 cells in 2 mL DMEM were seeded in dishes (d = 35 mm) for 24 h of incubation. After treatment with Opti-MEM containing Bi2S3/Ti3C2-TPP (100 \(\mu\)g mL-1) for 4 h, Mito-Tracker Red (1 \(\mu\)L, 400 nmol L-1) was added to the medium, and cells were incubated for another 15 min. Finally, cells were washed with PBS three times and imaged with an inverted fluorescence microscope.

For mitochondrial membrane potential assessment, 8 \(\times\) 104 U251 cells in 1 mL of DMEM were seeded in each well of a 12-well plate for 24 h of incubation. After treatment with Opti-MEM containing different NMs (100 \(\mu\)g mL-1) for 4 h, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min). After another 2 h of incubation, cells were stained with JC-1 (5 \(\mu\)mol L-1, 20 min) in Opti-MEM for mitochondrial membrane potential. Finally, cells were washed with PBS three times and directly observed by an inverted fluorescence microscope or quantitatively analyzed by flow cytometry.

For intracellular O2 detection, 1 \(\times\) 104 U251 cells in 100 \(\mu\)L of DMEM were seeded in each well of a 96-well plate for 24 h of incubation. After treatment with Opti-MEM containing Bi2S3/Ti3C2-TPP (100 \(\mu\)g mL-1) for 4 h, Image-iT™ was added to the medium to a concentration of 10 \(\mu\)mol L-1, and cells were incubated for another 30 min. Then, the medium was renewed, and cells were incubated in a hypoxic culture environment for 4 h. After the incubation, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min) and imaged with an inverted fluorescence microscope.

For intracellular ROS detection, 8 \(\times\) 104 U251 cells in 1 mL of DMEM were seeded in each well of a 12-well plate for 24 h of incubation. After treatment with Opti-MEM containing Bi2S3/Ti3C2-TPP (100 \(\mu\)g mL-1) for 4 h, cells were stained with DCFH-DA (10 \(\mu\)mol L-1) in Opti-MEM for 15 min and irradiated with or without an 808 nm laser (1 W cm-2, 10 min). Finally, the cellular DCF fluorescence was directly observed by an inverted fluorescence microscope or quantitatively analyzed by flow cytometry after washing with PBS three times.

For in vitro phototherapy, 1 \(\times\) 104 U251 cells in 100 \(\mu\)L DMEM were seeded in each well of a 96-well plate for 24 h of incubation. After treatment with Opti-MEM containing different NMs (100 \(\mu\)g mL-1) for 4 h, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min) and incubated for another 20 h. Then, cell viability was evaluated by MTT and live/dead staining assays. For PDT fractions, the temperature during NIR irradiation should be 4 \(^\circ C\) to avoid the influence of PTT.

For the cell clone formation assay, 1 \(\times\) 103 U251 cells in 2 mL DMEM were seeded in each well of a 6-well plate for 2 d of incubation. After treatment with Opti-MEM containing different NMs (100 \(\mu\)g mL-1) for 4 h, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min) and incubated for another 10 d with medium changed every other day. When significant clone clusters were visible under the microscope, the incubation was terminated. Then, cells were washed with PBS and fixed with formaldehyde for 15 min. Finally, 0.1% crystal violet was added to the samples to stain for 10 min, which was photographed for recording.

For the cell apoptosis assay, 1.6 \(\times\) 105 U251 cells in 2 mL DMEM were seeded in each well of a 6-well plate for 24 h of incubation. After treatment with Opti-MEM containing different NMs (100 \(\mu\)g mL-1) for 4 h, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min) and incubated for another 20 h. Then, cells were washed with cold PBS, collected, and stained with the Annexin V-FITC/PI assay kit. Finally, they were detected by flow cytometry.

For Nrf2, HO-1, and Hsp70 expressions, 1.6 \(\times\) 105 U251 cells in 2 mL DMEM were seeded in each well of a 6-well plate for 24 h of incubation. After treatment with Opti-MEM containing different NMs (100 \(\mu\)g mL-1) for 4 h, cells were irradiated with or without an 808 nm laser (1 W cm-2, 10 min) and incubated for another 20 h. Then, cells were washed with cold PBS and collected. The protein content was determined by the Bradford method, and the expressions of Nrf2, HO-1, Hsp70, and β-actin were evaluated through western blotting analysis.

Animal experiments

Four-week-old female Balb/c nude mice were purchased from Liaoning Changsheng Biotechnology Co., Ltd., and all animal experiments were approved by the Animal Care and Ethical Committee of the First Affiliated Hospital of Harbin Medical University, which were conducted in accordance with the guidelines from the Ministry of Science and Technology of China.

For in vivo phototherapy, the U251 tumor model was established by subcutaneous injection of PBS (100 \(\mu\)L) containing 1 \(\times\) 106 U251 cells into the right back of the hips of each mouse. When the tumor volume of mice reached approximately 80 mm3 (V = width2 \(\times\) length \(\times\) 1/2) under the same feeding conditions, total mice were randomly divided into 7 treatment groups (n = 5) and injected through the caudal vein with (I) PBS (pH = 7.4, 10 mM); (II) PBS (pH = 7.4, 10 mM) with an 808 nm laser irradiation (1 W cm-2, 10 min); (III) Bi2S3/Ti3C2-TPP; (IV) Bi2S3 with an 808 nm laser irradiation (1 W cm-2, 10 min); (V) Bi2S3 + Ti3C2 with an 808 nm laser irradiation (1 W cm-2, 10 min); (VI) Bi2S3/Ti3C2 with an 808 nm laser irradiation (1 W cm-2, 10 min); (VII) Bi2S3/Ti3C2-TPP with an 808 nm laser irradiation (1 W cm-2, 10 min), respectively. The injection concentration of the samples contained in PBS (200 \(\mu\)L) was 20 mg kg-1, with a Bi2S3 to Ti3C2 ratio of 3:2 in Group V. And the samples were irradiated at 24 h post-injection. Under NIR irradiation, the temperature of tumor sites in mice was recorded by a portable thermal imager. After treatments, the tumor volumes and body weights of mice were monitored every other day during 2 weeks. Furthermore, the mice were euthanized after 2-week tumor treatment, and the tumor tissues and major organs (heart, liver, spleen, lung, and kidney) of mice in each group were harvested, sliced, and stained for H&E and TUNEL staining to perform the histological analysis.

For in vivo biodistribution and optical imaging assessment, U251 tumor-bearing mice were intravenously injected with Bi2S3/Ti3C2-TPP (20 mg kg-1). At indicated time points (1, 2, 4, 8, 12, and 24 h post-injection), 50 \(\mu\)L blood was extracted from the caudal vein of each mouse and weighed every time (n = 5 for each time point). The mice were sacrificed after the last blood collection, and tumors and major organs were harvested. Then, the content of Bi or Ti in the blood and tissue samples was measured by ICP-MS tests. In vivo optical imaging was presented by CT imaging. The mouse was anesthetized at 0 and 8 h post-injection, and imaged by a CT scanner. Then, the CT value of the region of interest (ROI) at two time points was evaluated and compared with the corresponding analysis software.

Statistical analysis

SPSS 23.0 statistical software was conducted for the statistical analysis of the experimental data. A completely random design was performed to obtain measurement data. The normal distribution was described by mean \(\pm\) standard deviation, and the non-normal distribution was described by quartile and median. The analysis of variance (ANOVA) was used for comparison between multiple groups, and the t test was used for comparison between two groups. P \(<\) 0.05 was considered statistically significant.