Animal experiments, biochemical assessment and histology

C57BL/6J male mice aged 6 weeks were supplied by Beijing Vital River Laboratory. The mice were housed in a facility under a 12-hour light/dark cycle and constant environmental conditions with temperatures between 21-23 °C and humidity levels of 40-70%. Following a 7-day acclimatization period under standard laboratory conditions, the mice were randomly allocated to various groups. In the study of ensartinib-induced liver toxicity, mice received 60 mg/kg or 120 mg/kg of ensartinib dissolved in 0.5% CMC-Na, and treated by gavage every day for 4 weeks (with 7 mice in each group). To evaluate the potential of rutin as a treatment for liver damage caused by ensartinib, male mice were administered 120 mg/kg ensartinib and/or 10 mg/kg rutin dissolved in 0.5% CMC-Na every day for 4 weeks (with 6 mice per group). As a control, a separate group of mice were given 0.5% CMC-Na by gavage. After 4 weeks of drug treatment, the mice were fasted overnight and sacrificed through cervical dislocation under anesthesia.

Under isoflurane anesthesia, blood drawn from the retro-orbital plexus of the mice that had settled for 2 hours was then centrifuged at 3500 × g for 15 minutes. Next, the supernatant was collected for analysis. The ALT, AST and alkaline phosphatase (ALP) levels of the supernatant were measured by a chemical analyzer (XN-1000V, Sysmex). Liver tissues were sent to HaoKe Biotechnology Co., Ltd. for hematoxylin and eosin (H&E) staining, Masson’s trichrome staining, and Sirius red staining. Panoramic view of tissue staining sections were observed and captured under a pathological section scanner (HS6, SUNNY INSTRUMENT CO., LTD) and the magnification of representative images were 100× magnification or 200× magnification.

Primary hepatocyte isolation

Primary hepatocytes from 6- to 8-week-old male C57BL/6J mice were centrifuged by the collagenase perfusion and gradient method. Primary hepatocytes from mice were digested by infusion in IV collagenase solution (17104019, Sigma–Aldrich) dissolved in HBSS (14065056, Gibco). Next, the cells that had been isolated were sieved through a 70-mm cell strainer and subsequently cultured in dishes precoated with type I collagen (C8065, Solarbio).

Cell culture

HL-7702 cells were obtained from Guangzhou Jennio Biotech Co., Ltd. Primary hepatocytes were extracted from 6- to 8-week-old male C57BL/6J mice. The AML12 cell line was generously supplied by the Stem Cell Bank at the Chinese Academy of Sciences. Additionally, HEK293T cells used for cellular thermal shift assay (CETSA) were supplied by the Institute of Biochemistry and Cell Biology. NCI-H3122 cells were originally obtained from Wuhan Procell Life Science&Technology Co., Ltd. cultured in specialized cell culture medium for NCI-H3122 (PM150110, Procell). Primary hepatocytes and AML12 cells were cultured in DMEM (10569010, Gibco), while HL-7702 cells were maintained in RPMI-1640 (21870076, Gibco) supplemented with 10% fetal bovine serum (16140071, Gibco) and penicillin/streptomycin in a humidified atmosphere with 5% CO2 at 37 °C. Routine testing was consistently performed on all cell lines to confirm the absence of mycoplasma contamination.

Materials

Ensartinib (Ensar; T37585), brigatinib (Briga; T3621), crizotinib (Crizo; T1661), alectinib (Alec; T1936), ceritinib (Ceri; T1791), lorlatinib (Lorla; T3061), metformin (T8526), verapamil (Vera; T20656) and rutin (T0795) were purchased from Topscience. The natural products utilized in the drug screening were obtained from Nantong Jingwei Biotechnology Co., Ltd. Details regarding these products are presented in Table S1. Z-VAD-FMK (C1202) was obtained from Beyotime. Chloroquine (CQ; C6628) and MG132 (M8699) were obtained from Sigma–Aldrich. The antibodies used are shown in Table S2. Throughout the experiments (if not stated otherwise), HL-7702 cells were treated with 2 μM ensartinib for the indicated time periods or with 0, 1, 2 or 4 μM ensartinib for 36 h. In selected samples, 20 μM Z-VAD-FMK, 10 μM CQ, 10 μM MG132, 2.5 μM brigatinib, 2 μM crizotinib, 1.5 μM alectinib, 2 μM ceritinib, 1.5 μM lorlatinib, 0.5 mM metformin, 50 μM verapamil and 5 μM rutin were used.

RNA-seq transcriptome analysis

Using TRIzol reagent, RNA was extracted from HL-7702 cells treated with either DMSO or 2 μM ensartinib. RNA-seq analysis was conducted by LC-Bio, and the resulting data were stored in the NCBI Gene Expression Omnibus database (GSE249370).

Western blot analysis

Protein was isolated by employing a lysis buffer mixture consisting of of 150 mM NaCl, 50 mM Tris-HCl, 2 mM EGTA, 2 mM EDTA, 25 mM sodium glycerophosphate, 25 mM NaF, 0.3% Triton X-100, 0.3% NP-40, 0.3% leupeptin, 0.1% NaVO3, and 0.1% PMSF. Lysates were subjected to SDS–PAGE at concentrations of 8%, 10%, or 12%, transferred to PVDF membranes provided by Millipore Corporation and incubated with a 3% solution of bovine serum albumin. Subsequently, the PVDF membranes were washed three times in PBS containing 0.1% Tween-20 (T-PBS) and incubated with primary antibodies followed by three washes in T-PBS and incubation for 1 h with secondary antibodies. Following three washes in T-PBS, band detection was performed using the ECL-Plus Kit (P2300, NCM Biotech), and an Amersham Imager 600 (General Electric Company) was used for visualization. The primary antibodies utilized for western blot analysis are listed in Supplementary Table S2. HRP-labeled secondary antibodies (GAR007, GAM007, 1:1000) were obtained from LiankeBio. The primary antibodies used include: beta Actin (Diagbio, db7283, 1:2000), TXNIP (Huabio, ET1705-72, 1:1500), Cleaved PARP (Huabio, ET1608-10, 1:1500), Cleaved PARP (Abcam, ab32064, 1:1500), cleaved caspase-3 (Asp175) (Cell Signaling Technology, #9661, 1:2000), phospho-ASK1 (ABclonal, AP1215, 1:1000), γ-H2AX (Cell Signaling Technology, #9718, 1:2000), Ub (FL-76) (Santa Cruz Biotechnology, sc-9133, 1:1000), p-Akt (S473) (ABclonal, AP1208, 1:1000), LC3 A/B (Medical & Biological Laboratories, M186-3, 1:2500). We used Image J to perform a statistical analysis of the optical density of the western blot experiment. Specifically, the ratio was obtained by dividing the protein optical density by the ACTB optical density, and then quotient the ratio with the control ratio.

SRB (Sulforhodamine B) staining

To assess the cell survival rate, sulforhodamine B (SRB; S1402, Sigma–Aldrich) was utilized. In brief, in 96-well plates, the cells were fixed with 10% trichloroacetic acid (T104257, Aladdin) and treated with 4 mg/mL SRB for 30 minutes. Subsequently, the plates were washed with 1% acetic acid (1000218, Sinopharm). The residual SRB dye was dissolved in 10 mM unbuffered Tris base (1115KG001, Biofroxx). The absorbance at 515 nm was measured using a multiscan spectrophotometer (Tecan).

Quantitative polymerase chain reaction (qPCR)

Following treatment, the cells were collected, and RNA was isolated using the FastPure Cell/Tissue Total RNA Isolation Kit V2 (RC112, Vazyme) according to the manufacturer’s instructions. The RNA was then uniformly reverse transcribed into cDNA for use in subsequent qPCR experiments. The qPCR amplification was performed in two steps, beginning with an initial 3-minute phase at 95 °C, followed by 39 cycles of 95 °C for 3 seconds and 60 °C for 31 seconds each. Gene expression fold changes were determined using the comparative threshold cycle (Ct) method and the 2−(ΔΔCt) formula. The detailed primer sequences, provided by Youkang Biotechnology Co., Ltd., were shown as below:

TXNIP forward, 5'- GAGTACCTGCGCTATGAAGAC -3',

TXNIP reverse, 5'- TTTGAAGGATGTTCCCAGAGG -3',

IDH1 forward, 5'- CAAGTGACGGAACCCAAAAG -3',

IDH1 reverse, 5'- ACCCTTAGACAGAGCCATTTG -3',

DUSP6 forward, 5'- CGACTGGAACGAGAATACGG -3',

DUSP6 reverse, 5'- CAGCCAAGCAATGTACCAAG -3',

HNF4A forward, 5'- TGGACAAAGACAAGAGGAACC -3',

HNF4A reverse, 5'- ATAGCTTGACCTTCGAGTGC -3',

ACTB forward, 5'- CACCATTGGCAATGAGCGGTTC -3',

ACTB reverse, 5'- AGGTCTTTGCGGATGTCCACGT -3'.

Transfection of siRNA oligonucleotides

Transfection of siRNA oligonucleotides was achieved using Oligofectamine™ transfection reagent (12252011, Invitrogen). The medium in a 6-well plate was removed and replaced with the prepared transfection solution containing Opti-MEM™ (31985070, Gibco). After 6 hours of transfection, the transfection medium was replaced with complete culture medium, and drug treatments were administered as per the experimental design. The following oligonucleotides were provided by GenePharma:

siRNA#1 targeting TXNIP, 5'- GGCAAUCUCCUGGGCCUUAdTdT-3',

siRNA#2 targeting TXNIP, 5'- AGAAUACAUGUUCCCGAAGdTdT-3',

siRNA targeting DUSP6, 5'- AAUGUCAUAGGCAUCGUUCdTdT-3',

siRNA targeting HNF4A, 5'- CCACAUGUACUCCUGCAGAdTdT-3',

siRNA targeting IDH1, 5'- CCUUUGUAUCUGAGCACCAdTdT-3',

siRNA negative control, 5'- UUCUCCGAACGUGUCACGUdTdT-3'.

Flow cytometry analysis of Annexin V-PI staining

The proportion of apoptotic cells was detected with an Apoptosis Assay Kit I (C1062L, Beyotime). Briefly, cells were processed for Annexin V-PI staining at the indicated times. A total of 1 × 104 cells per sample were measured by a flow cytometer (BD Biosciences). The gating strategies for classification were documented using BD CellQuest Pro software.

JC-1 staining

To assess mitochondrial membrane potential (MMP), a JC-1 staining kit (C2005, Beyotime) was used. After treatment, cells were collected and rinsed with PBS. The cells were then stained with 5 μM JC-1, as provided in the kit, for 30 minutes at 37 °C in the dark. As a positive control, 10 μM CCCP was applied 20 minutes prior to the assay. For flow cytometry, 1 × 104 cells were prepared and measured using a flow cytometer (BD Biosciences). For fluorescence microscopy, nuclei were stained with DAPI (D212, Dojindo), and images for every sample (400× magnification) were captured with a fluorescence microscope (IX81-FV1000, Olympus).

Reactive oxygen species detection

Reactive oxygen species (ROS) were measured using a Reactive Oxygen Species Assay Kit (S0033S, Beyotime). Before loading with the DCFH-DA probe, 50 μg/mL Rosup was added to the positive control wells for pretreatment for 30 min. The cells were incubated in serum-free medium containing the DCFH-DA probe and diluted at a 1:1000 ratio for 20 minutes according to the instructions. For flow cytometry analysis of ROS, 1 × 104 cells were analyzed by a BD FACSCalibur™ flow cytometer (BD Biosciences). Fluorescence images for every sample (200× magnification or 400× magnification) were captured after staining the nuclei with DAPI (D212, Dojindo) with a fluorescence microscope (IX81-FV1000, Olympus).

MitoTracker staining

To assess the amount of mitochondria, HL-7702 cells were subjected to selective staining using the deep red fluorescent marker MitoTracker (M22426, Invitrogen). These cells were grown on the top of a Petri dish that was filled with a suitable growth solution. Upon reaching the required confluency, the existing medium was replaced with a staining solution preheated to 37 °C containing a 100 nM concentration of the MitoTracker probe following the provided instructions. Under conditions conducive to hepatocyte growth, the cells were incubated for 25 minutes, and then images for every sample (400× magnification) were captured using a fluorescence microscope (IX81-FV1000, Olympus).

MitoSOX staining

Mitochondrial ROS were measured with a MitoSOX staining kit (M36008, Invitrogen). Before the experiment, cells were plated on a 25-mm cover slip for 24 hours and treated with ensartinib. According to the instructions, the cells were incubated with MitoSOX at a concentration of 50 μM for 45 minutes. Images for every sample (400× magnification) were captured under a fluorescence microscope (IX81-FV1000, Olympus).

TUNEL staining

Liver specimens were initially fixed thoroughly using formalin (F8775, Sigma–Aldrich), followed by paraffin embedding and sectioning. These sections were then deparaffinized and subjected to proteinase K treatment (ST532, Beyotime) for 15 minutes. The slides were subsequently rinsed three times in PBS, followed by the addition of 50 μL of TUNEL reaction mix (C1088, Beyotime) and incubation for 60 minutes. Then, the nuclei were stained with DAPI (D212, Dojindo). Finally, the stained sections were imaged for every sample (200× magnification) using a fluorescence microscope (IX81-FV1000, Olympus).

Immunofluorescence

Both cells and liver tissue sections were fixed with 4% paraformaldehyde for 15 minutes (P6148, Sigma–Aldrich), followed by permeabilization with 0.3% Triton X-100 (1139ML100, Biofroxx) for 5 minutes. After three PBS washes, the cells were blocked in 4% bovine serum albumin (BSA; B2064, Sigma–Aldrich), and the liver tissues were blocked with 5% goat serum (S9070, Solarbio) for 30 minutes. The samples were then incubated overnight with the target antibody at 4 °C. After another three PBS washes, the cells were treated with Alexa Fluor 488- or Alexa Fluor 568-conjugated secondary antibodies (A21202, A10042, A31573, Thermo Fisher Scientific) for 1 hour, followed by DAPI staining (D212, Dojindo). The primary antibodies used included anti-TXNIP (Huabio, ET1705-72, 1:200) and anti-Tomm20 (Santa Cruz Biotechnology, sc-17764, 1:200). Fluorescence images for every sample (200× magnification or 400× magnification) were captured using a fluorescence microscope (IX81-FV1000, Olympus). The antibodies used for immunofluorescence are listed in Table S2.

Immunohistochemical staining

The immunohistochemistry process began with deparaffinization of the tissue sections in xylene, followed by rehydration using graded alcohol. Next, the sections were treated with 3% H2O2 (PV-6001, ZSGB-BIO) for 5 minutes and subsequently blocked for 30 minutes using 5% goat serum (16210064, Gibco). After this blocking step, the liver tissues were incubated overnight with specific antibodies against TXNIP (Huabio, ET1705-72, 1:200) and c-caspase-3 (Asp175) (Cell Signaling Technology, 9661, 1:200). Following three PBS washes, the sections were further incubated with an enzyme-conjugated secondary antibody (PV-6001, ZSGB-BIO) for 60 minutes. Staining was completed using DAB chromogen (ZLI-9017, ZSGB-BIO), and the nuclei were counterstained with hematoxylin (C0107, Beyotime) for a brief period of 3 seconds. The resulting images for every sample (200× magnification) were scanned with a pathological section scanner (HS6, SUNNY INSTRUMENT CO., LTD). The specific antibodies utilized for staining are detailed in Table S2.

ATP measurement

ATP levels were measured using an ATP detection kit (S0027, Beyotime). Following treatment with ensartinib, the cells were lysed using the ATP lysis buffer provided in the kit. After centrifugation, the supernatant was collected from the cell lysates. The ATP assay solution was then prepared according to the kit instructions and added to the samples for 3-5 minutes at room temperature. Samples or standards, 20 μL each, were dispensed into a black bottom 96-well plate (165305, Thermo Fisher Scientific). Relative light units (RLU) were quantified using a multifunctional microplate detector (Tecan).

Cellular thermal shift assay (CETSA)

CETSA procedures were performed based on previous studies (Jafari et al. 2014). Briefly, 100 μL of 293T cell lysate was mixed with rutin, silibinin, or salidroside at a final concentration of 5 μM. In the control group, the cell lysate was coincubated with DMSO, the solvent used for rutin. The mixture was incubated at various temperatures (42.0 °C, 42.5 °C, 43.7 °C, 45.6 °C, 48.2 °C, 53.2 °C, 55.8 °C, 58.4 °C, 60.3 °C, 61.5 °C, and 62.0 °C) for 6 minutes. To obtain the supernatant, all the mixtures were centrifuged at 12,000 rpm for 15 minutes. SDS–PAGE was used to analyze the expression of TXNIP in the supernatant.

Assessment of DNA damage using the comet assay

For the single-cell gel electrophoresis (SCGE) assay, as described by Velma and Tchounwou (Velma and Tchounwou 2013), HL-7702 cells were suspended in ice-cold PBS at a concentration of 1 × 105 cells/mL, for a total volume of 1 mL. Images for every sample (400× magnification) were captured using a fluorescence microscope (IX81-FV1000, Olympus). For the analysis of comet experimental results, we use Image J to analyze the tail length.

Molecular docking

The crystallographic structure of TXNIP (PDB ID 4II1) was sourced from the Protein Data Bank, while the molecular structure of rutin (PubChem CID 5280805) was retrieved from the PubChem database. Preparation of the TXNIP and rutin structures for docking was performed using AutoDockTools 1.5.6. The docking parameters were established as outlined in a previous study (Ferreira et al. 2015). The optimal docking configurations were visualized using PyMOL software.

Statistical analysis

The data are presented as the means ± standard deviations (SDs). Differences were deemed significant when the P value was less than 0.05. To assess significant differences between two groups, Student’s t test was employed, while one-way ANOVA was used for comparisons involving three or more groups. The data were analyzed using Microsoft Excel and GraphPad Prism software. Comprehensive statistical details, including the precise number of mice used, are provided in the legends of the figures for each experiment. No data were excluded from any of the experiments.