Experimental proceduresCell lines and culture conditions

We purchased Hep-55.1C and Hep-53.4, mouse hepatoma cell lines, from Cell Line Service GmbH (Oppenheim, Germany). The human hepatoma cell line, HuH7 (JCRB0403) was purchased from the Japanese Collection of Research Bioresource (JCRB) Cell Bank (Tokyo, Japan). Hepatoma cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Invitrogen Cell Culture Co., Auckland, New Zealand) containing 10% fetal bovine serum (Biowest, Nuaille, France) and 100 units/mL penicillin–streptomycin (Nacalai Tesque, Inc., Kyoto, Japan).

Reverse transcription–quantitative real-time PCR

To evaluate the effect of MTAs on hepatoma cells (Hep-55.1C and HuH7), reverse transcription–quantitative (RT-q) PCR was performed. The cells were cultured under serum-starved conditions at 37 °C overnight and then treated with the indicated concentration of each MTAs or dissolved in dimethyl sulfoxide (DMSO) for 24 h. The extracted RNA from cultured cells using an Isogen kit (Nippon Gene Co., Ltd, Tokyo, Japan) was reversely transcribed into cDNA, and RT-qPCR was performed to detect the mRNA levels of CD274 (PD-L1; programmed cell death ligand 1). Single-strand cDNA was synthesized using a high-capacity RNA-to-cDNA kit (Applied Biosystems; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The thermocycling parameters were as follows: 2 min at 50 °C and 10 min at 95 °C, followed by 45 cycles of 15 s at 95 °C, and 1 min at 60 °C. Relative quantification of gene expression was performed according to the 2−ΔΔCT method using StepOne Software 2.0 (Applied Biosystems). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a reference for normalization of the target gene expression data. The primers used were purchased from TaqMan™ (Applied Biosystems, Foster City, CA, USA) and were as follows: GAPDH, Hs02758991_g1; Gapdh, Mm99999915_g1; Cd274, Mm00452054_m1; CD274, Hs00204257_m1.

Establishment of the immune syngeneic orthotopic HCC mouse models and animal experiments

Five-week-old female C57BL/6 J mice were purchased from Kyudo KK (Fukuoka, Japan) and housed in cages. Each cage contained six or fewer mice per group at the animal facility of Kurume University School of Medicine (Kurume, Fukuoka, Japan). In total, 2 × 106 Hep-55.1C or Hep-53.4 cells were suspended in 1:1 phosphate-buffered saline (PBS) plus Matrigel matrix (Corning, NY, USA) and orthotopically inoculated into the left lobe of the liver to establish an immune syngeneic orthotopic Hep-55.1C or Hep-53.4 mouse model.

Two weeks after hepatoma cell transplantation, the tumor-bearing mice were allocated to different treatment groups (n = 5–6). The treatment groups were as follows: (I) Sorafenib; sorafenib tosylate (CAS No.: 475207-59-1; purchased from Santa Cruz Biotechnology, Inc., TX, USA), administered 30 mg/kg orally/once daily; (II) Lenvatinib; lenvatinib mesylate (CAS No.: 417716-92-8; kindly provided by Eisai co. ltd., Tokyo, Japan), administered 10 mg/kg/mouse orally/once daily; (III) VEGFR2 (VEGF receptor 2) antibody; DC101, an anti-VEGFR2 antibody, purchased from Bio X Cell (Lebanon, NH, USA), administered intraperitoneally thrice/week at a dose of 40 mg/kg; (IV) Regorafenib; regorafenib (CAS No.: 755037-03-7; purchased from Tokyo Chemical Industry, Tokyo, Japan), administered 3 mg/kg orally/once daily; (V) Cabozantinib; cabozantinib malate (CAS No.: 849217-68-1; purchased from ChemScene, Monmouth Junction, NJ, USA), administered 30 mg/kg orally/once daily; (VI) FGFR (FGF receptor) inhibitor; AZD4547, FGFR-1/2/3/4 inhibitor (CAS No.: 1035270-39-3; purchased from ChemScene), administered 12.5 mg/kg intraperitoneally/once daily; (VII) VT (vehicle treatment); PBS as vehicle was administered orally or intraperitoneally. The mice were euthanized by cervical dislocation under anesthesia with isoflurane and pentobarbital two weeks after treatment initiation, and the tumors were resected and evaluated. We calculated the estimated tumor volume as 0.52 × length × width2.

Immunohistochemical staining

Five-μm-thick of paraffin-embedded tumor tissue sections were boiled for 30 min in a high-pH target retrieval solution for antigen retrieval, followed by incubation with the primary antibodies followed by secondary antibodies. The antibodies used are listed in Supplementary Table S1. To visualize the immunoreactivity, we used an EnVision + system (DAKO North America, Carpinteria, CA, USA) and diaminobenzidine commercial kit (Liquid DAB + Substrate Chromogen System; Dako North America). To quantify CD3-ε-, CD8-α-, Foxp3-, NK1.1-, Granzyme B-, F4/80-, Arginase-1-, CD11c-, programmed cell death-1 (PD-1)-, PD-L1-positive cells, and CD31-positive tumor microvessels were counted at a magnification of × 200 (n = 5–10 random fields per group). We defined these positive cells as follows; CD3, pan T cells; CD8, cytotoxic T cells; Foxp3, Tregs; NK1.1, natural killer (NK) cells; Granzyme B, activated cytotoxic T cells or NK cells; F4/80, pan macrophages; Arginase-1, M2 macrophages; CD11c, dendritic cells (DCs). We counted PD-1-positive non-tumor cells and PD-L1-positive tumor cells. All slides were photographed using a confocal microscope BZ-X700 (Keyence Corporation, Osaka, Japan) and quantified using Adobe Photoshop CC 2021 version 22.2.0 (Adobe Systems, Inc., San Jose, CA, USA).

Total RNA isolation and RNA sequencing data analysis

From the resected tumor tissues, total ribonucleic acid (RNA) was isolated using TRIzol Reagent (Thermo Fisher Scientific, Inc., MA, USA) and then purified using the SV Total RNA Isolation System (Promega, WI, USA), according to the manufacturer’s instructions. We quantified RNA samples using ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) and confirmed the quality using TapeStation (Agilent Technologies, Inc., Santa Clara, CA, USA). Using the MGIEasy rRNA Depletion Kit and MGIEasy RNA Directional Library Prep Set (MGI Tech Co., Ltd., Shenzhen, China), sequencing libraries were prepared from 200 ng of RNA. The libraries were sequenced on a DNBSEQ-G400 FAST Sequencer (MGI Tech Co., Ltd.) with a paired-end 150 nt strategy. Quality trimming and adapter clipping of the read data were performed using Trimmomatic version 0.38. Trimmed reads were mapped to the transcript in the reference human hg38 using the Bowtie2 aligner within RNA-Seq by Expectation–Maximization (RSEM). Estimation of gene and isoform abundance using RSEM-generated basic count data (expected count). We used edgeR to detect differentially expressed genes (DEGs). Normalized counts per million (CPM), log-fold change (logFC), and p values were obtained from normalized CPM values. Subsequently, the criteria for DEGs were established as follows: p value ≤ 0.05, logFC ≥ 1. A heat map of differentially expressed genes was generated using MeV software. Hierarchical clustering was used to sort the genes. The color indicates the distance from the median of each row. The distance metric was “Pearson correlation,” and the linkage method was “average linkage clustering.” Gene functional enrichment was analyzed using the database for annotation, visualization, and integrated discovery [10].

Microarray data information

The human RNA sequencing data used in this study were obtained from the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/gds/), which is a public repository containing high-throughput gene expression data. We selected GSE214324 from the GEO database, and the array data comprised MHCC-97H cells, HCC cell line, treated with various doses of lenvatinib (40, 60, 80 and 100 μM) or DMSO for 24 h.

Statistical analyses

All data are expressed as mean ± standard error of the mean (SEM). Differences between two groups were examined for statistical significance using unpaired Student’s t test, and differences among multiple groups were examined using one-way analysis of variance (ANOVA), followed by Fisher’s least significant difference test. p was set than 0.05. Data analysis was performed using JMP Pro 16.0 software (JMP, Tokyo, Japan).