Cell culture and reagents (Sun et al. 2020; Zhou et al. 2017)

The HepG2 cells (American Type Culture Collection) and Bel-7402 cells were cultured in ATCC-formulated Eagle’s Minimum Essential Medium (EMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco), penicillin (100 units/mL, Gibco), and streptomycin (100 g/mL, Gibco). The cell lines were incubated with 5% CO2 at 37 °C. ICA with a purity of 99% was obtained from Selleck and dissolved in DMSO (concentration 1 mg/mL) at various concentrations (2.5, 5, 10, 20, and 40 mM) and stored at − 20 °C. The medium was changed to phenol red-free medium with 10% FBS before ICA was added. All cells were confirmed to be negative for mycoplasma by Mycoplasma Detection Kit (Solarbio, Beijing, China). Autophagy inhibitors 3-methyaldenine (3-MA) and agonists rapamycin were obtained from Sigma-Aldrich (St. Louis, MO).

In our initial experiments, we employed both Bel-7402 and HepG2 cells to establish a broader understanding of ICA’s effects across different HCC models. Bel-7402 cells were specifically used in early experiments to validate the reproducibility of HepG2 findings and to ensure that the observed effects were not cell line–specific. After preliminary experiments, we focused on HepG2 cells for detailed investigation due to their consistent response profiles and extensive characterization in literature, which supports their relevance for elucidating the molecular mechanisms underpinning HCC responses to treatment.

Cell transfection assay

The siRNAs targeting β-catenin and LOXL1-AS1 were designed and synthesized by RiboBio (Guangzhou, China). The sequences for LOXL1-AS1-targeting siRNA (si LOXL1-AS1) were 5′-GCUCAGUCUUACUAAUAAAGG-3′ (sense) and 5′-UUUAUUAGUAAGACUGAGCAC-3′ (antisense). For β-catenin-targeting siRNA (si β-catenin), the sequences were 5′-GCUCAGUCUUACUAAUAAAGG-3′ (sense) and 5′-UUUAUUAGUAAGACUGAGCAC-3′ (antisense). The plasmid and siRNAs were transfected into HepG2 cells using Lipofectamine 3000 (L3000150, Thermo Fisher Scientific, USA) according to the manufacturer’s instructions [41]. The HepG2 cells were divided into 7 groups: PBS (phosphate buffer saline) (control), ICA (20 μmol/L), ICA (20 μmol/L) + 3-methyladenine (3-MA, 1.0 mM), 3-methyladenine (3-MA, 1.0 mM), si-β-catenin (100 μmol/L), si-β-catenin (100 μmol/L) + ICA (20 μmol/L), si-β-catenin (100 μmol/L) + rapamycin (100 nmol/L) + ICA (20 μmol/L), si-LOXL1-AS1 (100 μmol/L), si-β-catenin (100 μmol/L). The Bel-7402 cells were divided into 2 groups: PBS (phosphate buffer saline) (control) and ICA (20 μmol/L). Cells in all groups were harvested after 48 h to measure the expression of corresponding cellular markers.

Cell growth assay (Sun et al. 2020)

Cell proliferation was evaluated using the Cell Counting Kit-8 (CCK-8) assay. Cells were isolated and seeded in a complete growth medium at a density of 1.5 × 103 cells. HepG2 and Bel-7402 cells were pretreated with PBS, 20 μmol/L ICA, 1.0 mM 3-MA, and ICA (50 μmol/L) + 3-MA (1.0 mM) for 48 h. Subsequently, the 96-well plates were placed in an incubator at 37 °C with 5% CO2. CCK-8 assay was performed according to the manufacturer’s instructions. Absorbance at 490 nm was obtained by a microplate reader (BioTek Instruments, Inc).

Colony formation assay (Sun et al. 2020)

HepG2 and Bel-7402 cells were pretreated with PBS and 20 μmol/L ICA. Cells were cultured in a drug-free medium for approximately 14 days. The cells were fixed with cold methanol-glacial acetic acid and stained with crystal violet for 4 h. Finally, the staining results were observed under an inverted microscope at × 100 magnification (Tokyo, Japan).

Cell invasion assay and cell scratch assay (Sun et al. 2020)

A Transwell assay was performed to detect cell invasion in HepG2 and Bel-7402 cells. HepG2 and Bel-7402 cells were seeded (5 × 104 cells/200 µL) into the upper Transwell chamber with serum-free DMEM. The lower chamber was filled with 500 µL EMEM containing 10% FBS. Simultaneously, PBS (control) or 20 μmol/L ICA was added into the upper chamber. After a 24-h incubation, cells in lower chambers were fixed in 95% ethanol and stained with hematoxylin. The invasive number of cells was calculated and imaged under an inverted microscope at × 400 magnification (Tokyo, Japan). For cell scratch assay, 5 × 105 cells were scraped in a 6-well plate using a sterile pipette for a gap of about 1 mm. HepG2 and Bel-7402 cells were pretreated with PBS, 20 μmol/L ICA at 37 °C for 24 h. Cell images in the scratch area were captured under an inverted microscope at × 100 magnification (Tokyo, Japan).

Cell apoptosis assay (Li et al. 2021; Ni et al. 2020; Sun et al. 2020)

TUNEL staining was performed to detect cell apoptosis. HepG2 cells treated with ICA for 24 hours were processed using a TUNEL assay kit (Roche Applied Science, Indianapolis, in the USA) at 37 ℃ for 60 min and at 18–25 ℃ for 15 min with DAPI-Fluoromount-G. Fluorescence was evaluated using an inverted microscope (Eclipse Ti-U, Nikon, Japan) with an excitation range of 450–500 nm and detection at 515–565 nm (green) and 358–461 nm (blue). The number of TUNEL-positive cells and the DAPI-stained nuclei was determined and repeated at least 3 times. To avoid bias and ensure that all results were reliable, counting was anonymously conducted by two independent individuals (Ferretti et al. 2019).

Dihydroethidium (DHE) staining

DHE staining was employed to assess intracellular levels of superoxide anions, serving as an indicator of ROS. Cells were seeded at a density of 7500 cells per well in a black 96-well plate with a clear bottom and incubated at 37 °C for 24 h to allow attachment. After 24 h incubation, the medium was replaced with one containing previously specified concentrations of various substances. Subsequently, the medium was with a 10 µM fluorescent probe (DHE 100 µL/well). The probe stock solution was prepared in DMSO, and the working solution was prepared in PBS. Fluorescence was measured at a wavelength of 475/579 nm (DHE) at 1-min intervals for 2 h (Pieńkowska et al. 2021).

JC-1 staining

JC-1 staining was used to measure the mitochondrial membrane potential. The cells were stained with 2.5 mmol/L JC-1 staining solution for 30 min at 37 °C in the dark. The images were obtained using fluorescence microscopy (Tokyo, Japan) at × 400 magnification.

Cellautophagy assay (Sun et al. 2020)

The position of LC3 was detected by immunofluorescence staining. MRFP-GFP-LC3 adenovirus (Hanbio, China) was transfected into liver cancer cells for 24 h, followed by treatment with 20 µmol/L ICA for 48 h. Cells were then fixed in 4% paraformaldehyde and examined under a confocal microscope (Leica, Germany) to observe LC3 staining. The GFP-positive/mRFP-positive (yellow) and GFP-negative /mRFP-positive (red) spots in cells were counted by confocal to evaluate autophagy flux. The experiment was repeated thrice, and each group counted 50–100 cells (Zhou et al. 2019).

Western blot

Cells and tumor tissues were lysed using ice-cold RIPA buffer with protease inhibitors. Protein concentrations were determined colorimetrically using bicinchoninic acid protein dye reagent (Pierce, Thermo Fisher Scientific, Inc., Waltham, MA, USA). Proteins were separated using SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% BSA in TBST and incubated with primary antibodies sourced from Abcam: anti-PI3K, AKT and mTOR (Abcam), anti-GAPDH (Abcam), anti-Ki67 (Abcam), anti-caspase-3,8,9 (Abcam), anti-VEGF (Abcam), anti-MMP-9 (Abcam), anti-cleaved caspase-3,9, anti-β-catenin, anti-Bcl-2, anti-Bcl-XL, anti-Bax, anti-LC3-II/LC3-I, anti-Atg5, anti-Atg7, anti-Atg8, anti-Beclin-1, and anti-p62. After washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibody (Santa Cruz Biotechnology). The target proteins were visualized using the Amersham ECL Prime Western Blotting Detection Reagent (Amersham Pharmacia Biotech). Finally, protein images were obtained by a ChemiDoc XRS imaging system and analyzed by Quantity One analysis software (Bio-Rad Laboratories).

Quantificational real-Time PCR Analysis (Deng et al. 2018)

The expression of β-catenin and LOXL1-AS1 was quantified using quantitative real-time PCR (qRT-PCR) on the CFX96 Real-Time System (Bio-Rad, USA). Sangon Biotech (Shanghai, China) provided the primers, and GAPDH/U6 was used as the internal reference. The following primers were used: β-catenin forward: 5′-CTCCAAGAATGGAGGCTGTAGGAA-3′, reverse: 5′-CCTATGAGATGGAGCAGGCAAGA-3′; LOXL1-AS1 forward: 5′-TTCCCATTTACCTGCCCGAAG-3′, reverse: 5′-GTCAGCAAACACATGGCAAC-3′; and GAPDH forward: 5′-CTGGGCTACACTGAGCACC-3′, reverse: 5′-AAGTGGTCGTTGAGGGCAATG-3′. Data were analyzed using the comparative Ct (2−ΔΔCt) method.

The xenograft tumor model assay

Six-week-old female BALB/c thymus nude mice were used to establish an orthotopic xenograft mouse model, and HepG2 cells (2.0 × 106) were inoculated on the right side near the hind legs of each nude mouse. The model mice were divided into a control group (saline injection) and an ICA group (40 mg/kg ICA injection), with 6 mice in each group. Mice in the ICA group received intraperitoneal injections of 40 mg/kg ICA three times a week on the first day after the injection of tumor cells. All animals were killed on the 30th day, and the median survival time and tumor weight were measured (Huang et al. 2020a, b).

Immunohistochemistry assay

Mouse tumor tissues were fixed in a 4% neutral formaldehyde solution, dehydrated with gradient alcohol, transparent with xylene, embedded in paraffin, and sliced into 4-μm-thick sections. For antigen retrieval, sections were treated with EDTA (pH 8.0) in a composite-bottom aluminum pressure cooker. The sections were then blocked for endogenous peroxidase activity with 3% hydrogen peroxide at room temperature for 10 min, followed by three washes in PBS for 5 min each. Subsequently, sections were incubated with primary antibodies against Ki67 and VEGF (Abcam) at room temperature (25 ℃) for 2 h, followed by three PBS washes of 2 min each. Secondary antibodies were applied for 30 min at 25 ℃ and then washed off similarly. DAB was added to the tissue for 10 min. After hematoxylin counterstaining for 1 min and gradient ethanol dehydration, xylene was added, and the slides were sealed with neutral gum. Finally, the staining results were observed under an inverted microscope at × 100 magnification (Tokyo, Japan) (Nahari et al. 2007).

Statistics analysis (Sun et al. 2020)

SPSS software (version 19.0) was used for all the statistical analyses. The data were presented as mean ± SD. Statistical comparisons were analyzed by one-way analysis of variance (ANOVA) followed by the least significant difference (LSD) test. The difference was considered statistically significant when values of P < 0.05.