HCCC-9810, RBE and 293T cell lines were obtained from the Chinese Academy of Sciences, Shanghai Branch Cell Bank. HuCCT1 were purchased from Cyagen, and normal human intrahepatic biliary epithelial cells (HiBEpiCs) were obtained from Qingqi (Shanghai) Biotechnology development Co., Ltd. Cells were cultured in RPMI-1640 medium (cat. no. 11875119; Gibco, Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (cat. no. 10099141C; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin (cat. no. 15140122; Gibco; Thermo Fisher Scientific, Inc.) at 37°C in a humidified incubator (HERA cell VIOS; Themo Fisher Scientific, Inc.) with 5% CO₂. 293T cells were cultured in DMEM (cat. no. 11995065; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS and 1% penicillin/streptomycin. Dimethyl sulfoxide (DMSO) was obtained from Sigma-Aldrich; Merck KGaA, and DT (cat. no. HY-B1357) was purchased from MedChemExpress.

The cell suspension (100 µl/well) was seeded in 96-well culture plates (cat. no. 3599; Corning, Inc.) at a density range of 1–2×10⁵ cells/ml. The cells were treated and incubated with DT at serial concentrations for 24, 48 and 72 h. A total of 10 µl CCK-8 (cat. no. HY-K0301; MedChemExpress) solution was then added to each well of the plate. The plate was incubated for 3 h at 37°C and absorbance was measured at 450 nm using the Molecular Devices Spectra Max iD3 Microplate Detection System. The cell viability rate was calculated as experimental optical density (OD) value/control OD value.

A total of 2,538 compounds in the TCM Active Compound Library (cat. no. HY-L065; MedChemExpress) were used for the screening experiment. Compounds were initially screened at a concentration of 10 µM. A further screening was conducted with the criterion that ICC cell viability should be decreased by 70% compared with the control (DMSO). Subsequently, compounds that maintained a viability rate >80% in HiBEpiCs were selected as candidates for anti-ICC drugs.

The overexpression of ST6GAL1 plasmids and control plasmids (pHBLV-CMV-MCS-3FLAG-EF1-ZsGreen-T2A-PURO) was carried out by Hanbio Biotechnology Co., Ltd. For the construction of the ST6GAL1-overexpression cell line, transfer plasmid (10 µg; A260/A280=1.94), envelope plasmid (pMD2G, 5 µg; A260/A280=1.93) and packaging plasmids (pSPAX2, 10 µg; A260/A280=1.92) were co-transfected into 293T packaging cells using Lipofectamine 3000 (cat. np. L3000015; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions at 37°C. The culture medium was replenished 16 h later, and the culture supernatant was collected at 2 and 3 days. The virus supernatant was centrifuged in a 50-ml centrifuge tube at 4°C and 2,000 × g for 10 min to remove cell debris. Then, the clarified viral supernatant was collected and placed into an ultracentrifuge tube to be centrifuged at 4°C and 82,700 × g for 120 min. The viral pellet was resuspended in 400 µl complete medium and finally the resuspended solution was aliquoted into sterilized viral tubes. A total of 2×10⁵ HCCC-9810 or HuCCT1 cells were seeded in 12-well plates overnight and subsequently infected with 50 µl lentivirus expressing the ST6GAL1 gene or a control lentivirus. The transfected cells were screened using puromycin (1 µg/ml) 1 week later.

The overexpression of P65/NF-κB plasmids and control plasmids pcDNA3.1 was carried out by Hanbio Biotechnology Co., Ltd. The P65/NF-κB overexpression plasmids (4 µg; A260/A280=1.93) were transfected into 293T cells at 37°C using Lipofectamine 3000 (cat. np. L3000015; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The medium was refreshed 6 h after transfection, 50 nM DT was added 24 h post-transfection, and WB and RT-qPCR were performed at 48 h.

A total of 1,000 ICC cells/well treated with different concentrations of DT and the ST6GAL1-overexpressed ICC cells treated with DT were plated in 12-well plates (cat. no. 3513; Corning, Inc.) and cultured at 37°C for 7 days. Subsequently, they were fixed in 4% paraformaldehyde (w/v) for 10 min at RT. After washing three times with PBS, staining was performed using 0.1% crystal violet staining solution (cat. no. C0121; Beyotime Institute of Biotechnology) for 10 min at room temperature (RT). Images were captured using a BZ-X800 fluorescence microscope (Keyence Corporation), and the colonies with aggregates of ≥50 cells were counted.

The ICC cells treated with different concentrations of DT and the ST6GAL1-overexpressed ICC cells treated with DT were seeded into a 24-well plate (cat. no. 3524; Corning, Inc.). Cell proliferation was assessed by incorporating EdU (cat. no. C0071S; Beyotime Institute of Biotechnology) following the manufacturer's instructions. The percentage of EdU⁺ cells was calculated by three random images captured with the BZ-X800 microscope.

The ICC cells treated with different concentrations of DT and the ST6GAL1-overexpressing ICC cells treated with DT were seeded in 12-well plates and allowed to incubate for 1 day. When the cells reached ~90% confluence, the bottom of each well was scratched with a 10-µl pipette tip. After washing with PBS, the culture medium was replenished with serum-free medium for culture. Images were captured under an inverted light microscope (cat. no. DMi8; Leica Microsystems) at 0 and 24 h, and the migration area was computed using ImageJ software (version 1.53a; National Institutes of Health).

The ICC cells treated with different concentrations of DT and the ST6GAL1-overexpressing ICC cells treated with DT were collected and diluted with serum-free DMEM to a concentration of 1×10⁵ cells/ml. The upper chamber (8 µm; cat. no. 353097; Falcon; Corning, Inc.) was filled with 200 µl cell suspension in order to measure the capacity for migration. The migratory cells were subsequently stained with 0.1% crystal violet staining solution for 10 min at RT and observed using a BZ-X800 microscope. The number of migratory cells was counted using three randomly selected images of selected fields.

RNA-seq and analysis were conducted by Shanghai Biotechnology Co., Ltd. RNA extraction was performed using the MJzol Animal RNA Isolation Kit (Majorivd; http://www.majorbio.com/shop/goods/71), followed by purification with the RNAClean XP Kit (Beckman Coulter, Inc.) and RNase-Free DNase Set (Qiagen, Inc.). RNA integrity and quality (260/280 nm ratio >1.8) were assessed using the Agilent 2100 Bioanalyzer/Agilent 4200 TapeStation and quantified with the Qubit 2.0^® Fluorometer (Thermo Fisher Scientific, Inc.) and NanoDrop ND-2000 (Thermo Fisher Scientific, Inc.). The mRNA was isolated from the purified RNA, fragmented and synthesized into cDNA, with subsequent library construction steps including end repair, adenylation, adapter ligation and enrichment. Purified libraries were quantified by Qubit^® 2.0 Fluorometer (Thermo Fisher Scientific, Inc.) and validated by Agilent 2100 bioanalyzer (Agilent Technologies, Inc.) to confirm the insert size and calculate the mole concentration. Cluster was generated by cBot with the library diluted to 10 pM and then were sequenced on the Illumina NovaSeq6000 platform (Illumina, Inc.) with the PE150 (Pair-end 150 bp) mode. Following data analysis at the transcript level and differential gene screening, differentially expressed genes were identified. Subsequent analyses were carried out to determine the Gene Ontology (GO) functional significance and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway significance.

The EZ-press RNA purification kit (cat. no. B0004D; EZBioscience) was used to extracted total RNA from cells. The concentration of the extracted RNA was measured by NanoDrop (Thermo Fisher Scientific, Inc.) at 260 nm. PrimeScript™ RT Reagent Kit (cat. no. RR037A; Takara Bio, Inc.) was applied to reverse transcribe RNA into cDNA according to the manufacturer's protocol. Each RT reaction contained 500 ng RNA. TB Green™ Premix Ex Taq™ (cat. no. RR420A; Takara Bio, Inc.) was used for qPCR, and data collection and analysis were carried out using an Applied Biosystems PCR System (7500 Real-Time PCR System; Thermo Fisher Scientific, Inc.). The PCR thermocycling conditions were as follows: i) Stage 1, 95°C for 30 sec; ii) Stage 2, 95°C for 5 sec, 60°C for 34 sec, 40 cycles; iii) Stage 3, 95°C for 30 sec, 60°C for 1 min, 95°C for 15 sec. The 2^−ΔΔCq method (15) was used to ascertain the relative expression levels of each gene, with GAPDH serving as the internal reference. Primer sequences were as follows: ST6GAL1 forward, 5′-CCCCAATCAGCCCTTTTACATCCTC-3′ and reverse, 5′-CCTGGTCACACAGCGTCATCATG-3′; and GAPDH forward, 5′-CCAGCAAGAGCACAAGAGGAAGAG-3′ and reverse, 5′-GGTCTACATGGCAACTGTGAGGAG-3′.

Cells treated with different concentrations of DT were lysed using cell lysis buffer (cat. no. 9803s; Cell Signaling Technology, Inc.) containing ProtLytic Protease and Phosphatase Inhibitor Cocktail (cat. no. P002; NCM Biotech). Nuclear proteins were extracted using the Nuclear and Cytoplasmic Protein Extraction Kit (cat. no. P0028; Beyotime Institute of Biotechnology). The protein concentrations were quantified by the BCA protein assay kit (cat. no. P0011; Beyotime Institute of Biotechnology). A total of 30 µg proteins were separated by 7.5 or 10% SDS-PAGE, and the resulting proteins were then moved from the gel onto PVDF membranes (cat. no. ISEQ00010; MilliporeSigma).

After 1 h of incubation with 5% skimmed milk at RT, membranes were incubated with primary antibodies against Bcl-2 (1:1,000; cat. no. 12789-1-AP; Proteintech Group, Inc.), ST6GAL1 (1 µg/ml; cat. no. AF5924; R&D Systems, Inc.), BAX (1:1,000; cat. no. 50599-2-Ig; Proteintech Group, Inc.), phosphorylated (p)-NF-κB p65 (1:1,000; cat. no. 3033; Cell Signaling Technology, Inc.), NF-κB p65 (1:1,000; cat. no. 8242; Cell Signaling Technology, Inc.), p-p38 MAPK (1:1,000; cat. no. 4511; Cell Signaling Technology, Inc.), p38 MAPK (1:1,000; cat. no. 8690; Cell Signaling Technology, Inc.), GAPDH (1:5,000; cat. no. 10494-1-AP; Proteintech Group, Inc.) and biotinylated Sambucus nigra (SNA) lectin recognizing α2,6-linked sialic acid (1:2,000; cat. no. B-1305; Vector Laboratories) at 4°C overnight. The membranes were then incubated with secondary antibodies at RT, goat anti-rabbit (1:5,000; cat. no. 7074; Cell Signaling Technology, Inc.), goat anti-mouse (1:5,000; cat. no. 7076; Cell Signaling Technology, Inc.), or streptavidin-HRP (1:5,000; cat. no. 3999; Cell Signaling Technology, Inc.). Protein visualization was carried out using a chemiluminescence ECL Detection kit (NcmECL Ultra; cat. no. P10300; NCM). The software used for densitometric analysis was ImageJ software (version 1.53a; National Institutes of Health).

The ICC cells treated with varying concentrations of DT and the DT-treated ST6GAL1-overexpressing ICC cells were fixed for 10 min with 4% paraformaldehyde at RT. Following fixation, cells were blocked for 1 h using blocking buffer (cat. no. P0260; Beyotime Institute of Biotechnology) for immunological staining at RT. Next, the cells were incubated with Cy5 labeled SNA lectin (cat. no. CL-1305-1; Vector Laboratories) overnight at 4°C. Each step involved washing three times with PBS for 5 min. Finally, the slides were sealed using an antifade mounting medium containing 1 µg/ml DAPI (cat. no. A4084; UElandy; http://www.uelandy.com/productDe_235.html); then images were captured by a BZ-X800 microscope.

After harvesting ICC cells with ST6GAL1 overexpression treated with DT, the cells were blocked for 20 min on ice using a carbo-free blocking solution (cat. no. SP-5040; Vector Laboratories) at a volume of 0.2 ml for 1×10⁶ cells. Subsequently, these cells were incubated with Cy5-labeled SNA lectin (1 µl per 1×10⁶ cells). Cells were analyzed with a BD FACSCanto II Flow Cytometer (BD Biosciences). FlowJo software (version 10.8.1; Becton Dickinson and Company) was used for analysis.

GraphPad Prism (version 9; Dotmatics) was used to conduct statistical analyses. Error bars in the experiments represent SD. The legends provide information about the number of events and independent experiments. Statistical comparisons between two experimental conditions were conducted using the Mann-Whitney test or unpaired Student's t-test, and the Wilcoxon rank-sum test was used for all paired samples. A one-way ANOVA was used to compare multiple experimental groups, followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

High-throughput drug screening was performed to evaluate the antitumor properties of TCM monomeric compounds in a TCM Active Compound Library containing 2,538 compounds. The screening aimed to assess the inhibitory effects of these compounds on the viability of ICC. In the initial screening (Fig. S1A), 83 monomers were identified that exhibited a growth inhibition rate >70% in HCCC9810 and RBE cells (Fig. S1B). Through a secondary screening, 64 monomers with consistent inhibitory activity >70% were selected (Fig. S1C). By assessing the viability of HiBEpiCs, 15 active monomers were identified that showed inhibitory effects on ICC without adverse effects on HiBEpiCs (Fig. S1D). DT emerged as a standout among the 15 active compounds tested, showcasing potent anti-ICC cell activity and promising therapeutic potential against ICC.

To assess the inhibitory effects of DT on ICC cells, a CCK-8 assay was conducted which allowed to evaluate the impact of varying concentrations and treatment durations of DT on cell viability. The findings revealed a significant dose-dependent reduction in the viability of both HuCCT1 and HCCC9810 cell lines (Fig. 1A). More specifically, the IC₅₀ values for HCCC9810 cells after 24, 48 and 72 h of DT treatment were 106.5±23.2, 92.4±12.1 and 77.9±10.9 nM, respectively. For HuCCT1 cells, the corresponding IC₅₀ values were 132.3±33.8, 105.1±26.6 and 98.5±4.2 nM, respectively, indicating a progressive decrease in IC₅₀ values, which suggests an increase in drug sensitivity over time. Subsequently, to further probe the effects of DT on cell proliferation, a series of concentration gradients were applied: 0, 1/4 IC₅₀, 1/2 IC₅₀, IC₅₀ of DT at 48 h (Table SI); and the proliferation capacity of ICC cells was assessed at 24 h. Through the analysis of the colony formation and EdU incorporation assays, it was observed that DT significantly decreased the colony formation and proliferation ability of HuCCT1 and HCCC9810 cells, demonstrating a dose-dependent inhibition (Fig. 1B and C). Taken together, these findings strongly suggest that DT possesses the ability to hinder both the viability and proliferation of ICC cells.

To explore the potential suppressive impact of DT on the migration capabilities of ICC cells, the cells were exposed to varying concentrations of DT, and wound healing and Transwell assays were carried out. The Transwell assay results showed that DT significantly decreased the migration of HuCCT1 and HCCC9810 cells in a dose-dependent manner (P<0.05; Fig. 2A). Additionally, an increase in DT concentration led to a decrease in the migration areas for both HCCC9810 (Fig. 2B) and HuCCT1 (Fig. 2C) cells, underscoring a progressive impediment to cell movement. Collectively, these findings demonstrated the ability of DT to inhibit the migration of ICC cells.

Subsequently, the primary signaling pathways through which DT exerted its inhibitory effects on ICC cell functions were investigated. The aim was to delineate the in vitro molecular mechanisms underlying the suppression of ICC cell activities with DT. Firstly, the intersection of differentially expressed genes in HCCC9810 and HUCCT1 cells after treatment with DT was selected, including both upregulated and downregulated genes. Subsequently, GO and KEGG functional enrichment analyses were conducted. RNA-seq analysis unveiled significant alterations in pathways related to glycosylation, regulation of NF-κB imported into the nucleus and transcriptional mis-regulation in ICC cells following treatment with DT (Fig. 3A and B). The heatmap of differentially expressed genes suggested that DT could downregulate the transcriptional levels of ST6GAL1 (Fig. 3C). Furthermore, the RT-qPCR results confirmed that DT could effectively inhibit the mRNA expression of ST6GAL1 (Fig. 3D). Next, the effects of DT on the NF-κB, P38 and STAT3 signaling pathways, as well as ST6GAL1 expression were investigated. DT treatment resulted in inhibition of phosphorylation of p-NF-κB and p-STAT3, while it upregulated the phosphorylation of P38 and reduced the expression of the anti-apoptotic protein BCL-2 (Fig. 3E; statistical charts, Fig. S2). It was found that NF-κB interacted with the promoter region of the ST6GAL1 gene, initiating its transcription (16). To confirm the effect of NF-κB on ST6GAL1, a P65/NF-κB overexpression plasmid was constructed, transfected into 293T cells and followed by DT treatment (Fig. S3). The results indicated that overexpression of NF-κB promoted the expression of ST6GAL1, and also demonstrated the inhibitory effect of DT on ST6GAL1 in 293T cells. Finally, lectin blotting with SNA was used which selectively adheres to α2,6-sialic acid structures to evaluate the α2,6-sialylation levels in situ. A dose-dependent reduction in α2,6-sialylation, predominantly triggered by DT, was observed (Fig. 3F and G). These findings collectively revealed the involvement of the NF-κB/ST6GAL1 pathway in mediating the effects of DT in ICC cells, underscoring its significance in the modulation of ICC cell abilities.

The aforementioned results revealed that DT reduced the expression of ST6GAL1 in a dose-dependent manner. Consequently, the ability of overexpressing ST6GAL1 to reverse the inhibitory effects of DT on ICC cells was investigated. ST6GAL1-overexpressing-ICC cells were generated by a lentivirus vector system. The increased levels of ST6GAL1 in ST6GAL1-overexpressing HCCC9810 and HuCCT1 cells were confirmed by RT-qPCR and western blotting (Fig. S4A and B). The colony formation and EdU assays provided compelling evidence that overexpression of ST6GAL1 significantly promotes the proliferation of ICC cells. These assays highlighted that ST6GAL1 could not only enhance cell proliferation, but also counteract the suppression caused by DT (Fig. 4A and B). Further investigation into the impact of ST6GAL1 on cell migration was conducted using wound healing and Transwell migration assays. The outcomes from these assays confirmed that ST6GAL1 overexpression not only augmented the migratory capabilities of ICC cells, but also effectively neutralized the migration-inhibiting influence of DT (Fig. 4C and D). Collectively, these findings revealed that ST6GAL1 acted as a potent promoter of both proliferation and migration of ICC cells, effectively reversing the inhibitory effects imposed by DT treatment.

To ascertain the critical role of ST6GAL1 in the signaling pathway through which DT inhibited ICC cells, ST6GAL1-overexpressing HCCC9810 and HuCCT1 cells were treated with DT at IC₅₀ value. Western blotting results revealed that DT inhibited the expression of the anti-apoptotic protein Bcl-2, while overexpression of ST6GAL1 enhanced Bcl-2 expression and partially reversed the inhibitory effect of DT. Moreover, DT was shown to suppress ST6GAL1 expression in both control vector cells and ST6GAL-overexpressing cells (Fig. 5A and B). The aforementioned results indicated that DT could inhibit the phosphorylation of NF-κB, STAT3 and activate p-P38. Consequently, nuclear proteins were next extracted to examine the changes in nuclear p-NF-κB, p-STAT3 and p-P38 in ICC cells treated with DT at the IC₅₀ concentration. Western blotting results suggested that DT could activate p-P38, and inhibit p-NF-κB and p-STAT3 in whole cells (Fig. 5A and B; statistical charts, Fig. S5A and B), and further nuclear protein detection revealed an increased expression of nuclear p-P38, and a decrease in p-NF-κB and p-STAT3 (Fig. 5C and D; statistical charts, Fig. S5C and D). In addition, SNA lectin flow cytometry was employed to evaluate the α2,6 sialylation levels in cells, with the aim of exploring the enzymatic activity of ST6GAL1. The experimental findings revealed that DT could reduce the α2,6 sialylation on the cell membrane, and the overexpression of ST6GAL1 was able to counteract the reduction effect induced by DT (Fig. 5E and F). These findings further elucidated that DT can modulate the expression of ST6GAL1 by impeding NF-κB activation, diminishing the nuclear presence of p-NF-κB and ultimately suppressing ICC cell progression through the downregulation of the effector protein Bcl-2.