Cell cultureHepG2 cells were purchased from RIKEN cell bank (Wako, Saitima, Japan). HepG2 cells were maintained in low-glucose Dulbecco’s Modified Eagle’s Medium (DMEM) (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 10% USDA-tested fetal bovine serum (FBS; Cytiva, Marlborough, MA, USA), 100 U/mL penicillin, and 100 µg/mL streptomycin (FUJIFILM Wako Pure Chemical Corporation) in a humidified atmosphere of 5% CO2 at 37 °C34. For subculture, the cells were trypsinized with 0.05% EDTA-trypsin (FUJIFILM Wako Pure Chemical Corporation).Isolation and identification of baicalin target proteinThe baicalin target protein was extracted by the method reported previously by our group35.Baicalin-treated HepG2 cellsHepG2 cells were inoculated into 35-mm dishes (1 × 106 cells/2 mL) for 48 h, and treated with baicalin at a concentration of 0, 10, 25, 50, or 75 µM in 2 mL of DMEM containing 2% FBS at 37 °C for a further 48 h34. Comparisons of mRNA expression levels were performed by real-time quantitative PCR (RT-qPCR), and apoptosis activity was examined by caspase-3 activity assay.Real-time quantitative PCRRNA was extracted using ISOSPIN Cell and Tissue RNA (314-08211; Nippon Gene Co., Ltd., Tokyo, Japan) according to the manufacturer’s instructions. Reverse transcription of total RNA (500 ng) into cDNA was performed with TKR TB Green® Premix Ex Taq™ II (Tli RNaseH Plus) (RR820A; TaKaRa, Kyoto, Japan) on a PTC-200 DNA Engine Cycler (Bio-Rad, Hercules, CA, USA). Real-time PCR was performed using LightCycler 480 SYBR Green I Master Mix (TaKaRa) on an ABI 7500 Real Time PCR System (Applied Biosystems, Foster City, CA, USA). The primers used for analysis of ANXA2, cIAP-1, cIAP-2, STAT-3, and β-actin36,37,38,39 gene expression are listed in Table 1. The mRNA expression levels were measured and quantified using ABI Prism 7000 sequence Detection System (Applied Biosystems). RT-qPCR was performed in quadruplicate, and the mean values relative to β-actin were calculated. The 2−ΔΔCT method was used to calculate the relative mRNA expression.Table 1 Primer sequences for each gene.Caspase-3 activity assayApoptosis was analyzed by caspase-3 activity assay with an APOPCYTO Caspase-3 Colorimetric Assay Kit (Medical and Biological Laboratories, Tokyo, Japan) according to the manufacturer’s protocol40. A Pierce Coomassie (Bradford) Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) was used to measure the total protein concentration. Protein samples were prepared at a concentration of 100 µg/50 µL and the absorbance was measured at 405 nm using a microplate reader (SpectarMax® i3; Molecular Devices, Sunnyvale, CA, USA). All assays were performed in triplicate.siRNA transfectionHepG2 cells were transfected with 10 μM Annexin II small interfering RNA (siRNA) (h2) (sc-270151; Santa Cruz Biotechnology, Santa Cruz, CA, USA) by electroporation. HepG2 cells were inoculated onto 35-mm dishes (1 × 106 cells/2 mL), cultured for 24 h, washed, and the medium was replaced with Opti-MEM (Gibco, by Life Technologies, Carlsbad, CA, USA). The HepG2 cells in 100 μL of Opti-MEM and 2 μL of siRNA (2.6 μg) were mixed and transferred to a 2-mm gap electroporation cuvette (EC-002; Nepa Gene, Chiba, Japan). The HepG2 cells were transfected in an electroporator (NEPA21; Nepa Gene), cultured in DMEM supplemented with 10% FBS, seeded into 35-mm culture dishes, and cultured for 48 h. The siRNA transfection efficiency was confirmed by RT-qPCR and Western blotting. Cells transfected with control siRNA were used as a negative control (Stealth RNAi siRNA Negative Control, sc-37007; Santa Cruz Biotechnology).Western blottingTotal cellular protein was extracted with radioimmunoprecipitation assay (RIPA) lysis buffer (Cell Signaling Technology, Danvers, MA, USA). Total protein concentration was measured using a Pierce Coomassie (Bradford) Protein Assay Kit (Thermo Fisher Scientific). Proteins (total 10 μg) were separated by SDS-PAGE and then transferred onto polyvinylidene difluoride (PVDF) membranes (Zhuzhou Hongda Polymer Materials Co., Ltd., Zhuzhou, China).The membranes were subjected to Western blotting analysis using specific rabbit and mouse antibodies against ANXA2 (dilution 1:1000, Q25; Cell Signaling Technology) and β-actin (dilution 1:10000, A5316; Sigma, St. Louis. MO, USA). We detected immunoreactive ANXA2 and β-actin using an enhanced chemiluminescence (ECL) Prime Western Blotting System (RPN2232; Cytiva) after the isolated cellular proteins were complexed with horseradish peroxidase-conjugated anti-rabbit and mouse immunoglobulin G (IgG).Amersham Imager 680 (Cytiva) was used for development and visualization, and the gray values of the target bands were analyzed with Image J38.Baicalin-treated ANXA2 knockdown HepG2 cellsThe transfected ANXA2 knockdown cells were inoculated into 35-mm dishes (1 × 106 cells/2 mL), cultured for 48 h, and then treated with baicalin at a concentration of 0, 10, or 75 µM in 2 mL of DMEM supplemented with 2% FBS at 37 °C for a further 48 h34. Comparisons of mRNA expression levels were carried out by RT-qPCR, and the apoptosis activity was examined by caspase-3 activity assay.Statistical analysisStatistical analyses were performed using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA). All data are presented as the mean ± standard deviation (SD). For panels, n.s., not significant, *P < 0.05 (one-way analysis of variance (ANOVA)).Molecular docking simulationsTo examine the binding mode of ANXA2–baicalin complex, we conducted docking experiments using AutoDock Vina (Scripps Research, La Jolla, CA, USA). As 3D structures are necessary for molecular docking experiments, the 3D structures of the delta N-terminal domain of human ANXA2 (ΔN_ANXA2) were obtained from PDB (5LQ0, 2HYW, 4HRE, 1XJL). To confirm the binding mode of full-length ANXA2 including the N-terminal domain to baicalin, the 3D structure of full-length human ANXA2 was constructed by AlphaFold 2. Hydrogen atoms and AM1-BCC atomic charge were added to baicalin using the Hgene program in the myPresto portal41. Then, energy optimization to generate 3D conformations of baicalin was conducted using the cofgene program in the myPresto portal. For all docking simulations, box centers were set to the default settings. To include all ANXA2 structures, a grid box size of 60 × 60 × 60 Å against the monomeric structure of ANXA2 and the exhaustiveness of 100 were used as docking parameters42,43. The spacing between grid points was adjusted to 1 Å and the maximum number of docking modes was set to 20. The docking poses were visualized using UCSF Chimera and pymol. The ligand (baicalin) was then sorted based on the AutoDock score (kcal/mol).Molecular dynamics simulationsThe protein–ligand complexes with the highest binding affinities (kcal/mol) based on the docking simulations performed in AutoDock Vina (Scripps) were used as the starting structures for MD simulations. First, each ANXA2–baicalin complex was immersed in a transferable intermolecular potential 3P (TIP3P) water box, with water molecules extending 15.0 Å from any solute atom in each direction using Bondi radii. Then, Na+ or Cl− ions were added to neutralize each complex. The complexes were minimized in two steps: first, all of the heavy backbone atoms of ANXA2 were restrained with a weight of 2 kcal/mol Å2. Next, the system was minimized without any restraints. These systems were optimized by 5000 cycles of steepest descent and 5000 cycles of conjugate gradient minimization. All minimizations were performed using the sander Amber package.Following minimization, the AmberTools 23 simulation package was used to carry out MD simulations43,44. The total simulation time for baicalin and ANXA2 protein complexes was 100 ns, in which the complexes were heated from 0 to 300 K at 25 ps, in constant volume mode with a restraint weight of 2.0 kcal/mol Å2, then density balance was carried out with a restraint weight of 2.0 kcal/mol Å2 (constant pressure of 10 ps), equilibrium (constant pressure of 100 ps), and MD production. During 100 ns MD calculations, the docking poses with the lowest energy were defined as the binding models.
