Hsa_circ_0009096/miR-370-3p modulates hepatic stellate cell proliferation and fibrosis during biliary atresia pathogenesis

Cell lines and culture

LX-2 (#CC4023) and HEK-293T (#CC4003) cells were obtained from CellCook (Guangzhou, China). LX-2 and HEK-293T cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; #C11995500BT; Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (#26010074; Thermo Fisher Scientific, Waltham, MA, USA) and 1% penicillin/streptomycin (#15140122; Thermo Fisher Scientific, Waltham, MA, USA) at 37 °C in a humidified cell culture chamber with a supply of 5% CO₂. To induce fibrosis of LX-2 cells, these cells were cultured in DMEM containing 0, 0.5, or 2-ng/mL rhTGF-β1 (Recombinant Human TGF β1; #BPR147; Bersee, Beijing, China) under normal conditions, as previously described (Luo et al., 2019).

RNA interference

For suppression of hsa_circ_0009096 expression in LX-2 cells, siRNA-1 (5′-GAGAATGGTTTGATGCTATAA-3′), siRNA-2 (5′-AACATCTAGAGAATGG TTTGA-3′), siRNA-3 (5′-CTAGAGAATGGTTTGATGCTA-3′), and negative control (NC) (5′-TTCTCCGAACGTGTCACGTTT-3′) were prepared by Genepharma (Shanghai, China). Similarly, to repress the miR-370-3p expression in cultured LX-2 cells, the miR-370-3p (5′-ACCAGGUUCCACCCCAGCAGGC-3′) and negative control sequence inhibitors (NC: 5′-CAGUACUUUUGUGUAGUACA A-3′) were also prepared by Genepharma. The above mentioned siRNA or microRNA inhibitors were introduced into the LX-2 cells using the Invitrogen Lipofectamine 3000 Transfection Reagent (#L3000150; Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s instructions. The expression-suppressing efficacy was then evaluated using reverse transcription-qPCR (RT-qPCR) 24 h after transfection.

RNase R treatment and the actinomycin D assay

A total of 2 μg of total RNA was incubated for 8 min at 37 °C with or without 6 U RNase R (#R0300; Geneseed, Guangzhou, China), subsequently deactivated with RNase R at 70 °C for 10 min, and then analyzed using RT-qPCR. Actinomycin D (#50-76-0, APExBIO, Houston, TX, USA) at a concentration of 1 μM was applied to LX-2 cells at specified time intervals (0, 12, and 24 h). Following cell harvesting, total RNA was extracted for RT-qPCR. The PCR products were detected using agarose gel electrophoresis.

Reverse transcription-qPCR

The expression levels of circRNAs, microRNAs, and mRNAs of interest in the cultured LX-2 cells were determined using RT-qPCR. Briefly, total RNA was extracted from the cultured LX-2 cells using the Trizol kit (#B511311; Sangon Biotech, Shanghai, China) following the manufacturer’s instructions. To avoid contamination by genomic DNA, the samples were treated with a DNase kit (Ambion, Austin, TX, USA). Reverse transcription without reverse transcriptase was also performed to assess gDNA contamination. The RNA quantity and purity (A260/A280) were evaluated using Nano-100 (ALLSHENG, Hangzhou, China). Then, the cDNA samples were synthesized from approximately 1.5-μg RNA samples of each group using the EasyScript RT kit (#AE101-02; Transgen Biotech company, Beijing, China) following the manufacturer’s instructions and stored at −20 °C for further experiments. Subsequently, quantitative PCR assay was performed on the ABI Stepone Plus instrument (Applied Biosystems, Foster City, CA, United States) to measure the relative expression of circRNA, microRNA, or mRNAs using the TransStart qPCR kit (#AQ101-01; Transgen Biotech company, Beijing, China) following the manufacturer’s instructions. The expression levels were quantitated via the standard 2^(−ΔΔCt) calculation method using U6 or GAPDH as reference genes. The qPCR for each group comprised three replicates. The information on the primers used in expressional quantitation is listed in Table 1. All primers were synthesized by GenePharma and were PAGE-purified.

Dual luciferase reporter assay

The association of hsa_circ_0009096 with hsa-miR-370-3p in 293T cells was validated through the dual luciferase reporter assay using the Nano-Glo Dual-Luciferase^® Reporter kit (#N1630; Promega, Madison, WI, USA) following the manufacturer’s instructions. Briefly, the circ_0009096 wild-type or mutant sequences were, respectively, ligated with the pmirGLO vectors, which were induced into 293T cells using the Lipofectamine 3000 kit as introduced above, along with the hsa-miR-370-3p mimics (5′-GCCUGCUGGGGUGGAACCUGGU-3′) or its negative control sequences (5′-UUCUCCGAACGUGUCACGUTT-3′). Finally, the cultured 293T cells were lysed 48 h after transfection, and cell lysates were used to measure luciferase activities using a GloMax luminometer. The binding of circ_0009096 with hsa-miR-370-3p was assessed by alterations of luciferase activities between different groups.

Flow cytometry for cell cycle progression analysis

Cell cycle progression in the cultured LX-2 cells was determined using flow cytometry with the Cell Cycle Analysis Kit (#ab287852; Abcam, Cambridge, UK) following the manufacturer’s protocol. Briefly, LX-2 cells cultured in a six-well plate (5 × 10⁵ cells/well) were harvested via centrifugation at 400 g for 4 min, washed with pre-chilled cell cycle assay buffer, fixed by incubation with 70% ethanol on ice for 40 min, washed again with 2-ml cell cycle assay buffer, incubated with 450-ul staining solution in the dark for 40 min, and finally analyzed using a flow cytometer. The LX-2 cell cycle progression was detected through at least three biological replicates of flow cytometry assays.

Flow cytometry for cell apoptosis analysis

Apoptosis of the LX-2 cells was analyzed using flow cytometry following treatment with the Annexin V-PI Apoptosis Staining kit (#ab214484; Abcam, Cambridge, UK) according to the manufacturer’s instructions. Briefly, the cultured LX-2 cells (1 × 10⁶ cells) after designated transfections and treatments were harvested through centrifugation at 400 g for 5 min, resuspended in 1X Annexin-binding buffer (100 ul), and incubated with Annexin V (5 μL) and PI (5 μL) solutions in the dark at room temperature for 14 min. Following incubation, the cell mixtures were then mixed with 400 μL 1X Annexin-binding buffer and subjected to flow cytometry. Three biological replicates were performed to assess the percentages of apoptotic LX-2 cells.

Immunocytochemistry (ICC) staining

In situ collagen 1A1 (COL1A1) and α-SMA protein levels in the cultured LX-2 cells following specified transfections and treatments were detected using the ICC method. Briefly, LX-2 cell slides were first fixed via incubation with paraformaldehyde (4%) for 12 min at room temperature, washed thrice with PBS solution, permeabilized with PBS solution containing 0.1 Triton X-100 for 8 min, blocked with 1% BSA solution for 30 min at room temperature, incubated with diluted primary antibodies targeting COL1A1 (#8784; Santa Cruz, California, USA; 1:50) or α-SMA (#14395-1-Ap; ProteinTech, IL, US; 1:500) overnight at 4 °C, washed thrice times with PBS solution for 6 min, and incubated with diluted secondary antibodies for 1 h. After counter-staining with 0.2 μg/mL DAPI for 1 min, the LX-2 cell slides were then developed with HRP substrates, and then mounted on and observed under a microscope. ICC scores were calculated using the following formula: ICC score = intensity score × percentage score. The intensity score was measured according to the intensity of staining (0: negative, 1: light-yellow particle, 2: brown-yellow particle, and 3: brown particle). The percentage score was determined using the percentage of stained area (0: 0%, 1: <10%, 2: 10–49%, 3: 50–74%, and 4: 75–100%) (Guo et al., 2021).

Western blotting (WB)

The LX-2 cells were lysed using 1 × RIPA buffer (#89901; ThermoFisher Scientific, Waltham, MA, USA) for 20 min on ice. Following the manufacturer’s instructions, the supernatant was collected after centrifugation at 12,000 g for 10 min. The Pierce^™ BCA Protein Assay Kit (#23225; Thermo Scientific, Waltham, MA, USA) was then used to quantify the protein. Thirty micrograms of protein were separated using 8–10% SDS-PAGE and then transferred to a PVDF membrane. After the membrane was cleaned with 1 × TBST, it was allowed to soak for 1 h at room temperature in 5% non-fat milk. Subsequently, the membrane was incubated at 4 °C overnight with the corresponding primary antibodies targeting COL1A1 (#ab34710; Abcam; 1:1,000; Cambridge, UK), α-SMA (#23081-1-AP; Proteintech, IL, USA; 1:1,000), TGFBR2 (#BA0526-2; Boster, California, US; 1:1,000), or GAPDH (#60004-1-lg; Proteintech, Rosemont, IL, USA; 1:100,000). The membrane was then treated for 1 h at room temperature with a secondary antibody. The target protein bands were visualized using a chemiluminescent western blot detection kit (#32209, Thermo Fisher Scientific, Waltham, MA, USA). ImageJ software (NIH, Bethesda, MD, USA) was used to analyze the greyscale value of each band. The relative intensity was calculated by dividing the normalized target protein/GAPDH ratio of the test samples by that of the control samples.

Statistical analysis

Each experiment in this study was repeated three times independently. All quantitative data are presented as mean ± standard deviation. Student’s t-test or one- and two-way analysis of variance (ANOVA) followed by Dunnett’s multiple-comparison posttest were performed to assess the statistical significance using Graphpad Prism (v6.0) software. NS, not significant (p > 0.05), *p < 0.05, **p < 0.01.

Hsa_circ_0009096 was resistant to RNase R and actinomycin D treatment

Hsa_circ_0009096 was characterized using PCR with divergent primers followed by agarose gel electrophoresis and Sanger sequencing in our previous study (Zhang et al., 2023), which confirmed the specialty and back-spliced junction of hsa_circ_0009096. In this study, we performed additional RNase R and actinomycin D assays on the LX-2 cells. After RNase R and actinomycin D treatment, the linear RNAs reduced significantly, but the circular RNAs were more stable and resisted RNase R (Fig. 1A) and actinomycin D (Fig. 1B), further verifying the circular nature of hsa_circ_0009096.

Elevated hsa_circ_0009096 expression in the induced hepatic fibrosis model

To investigate the functions of circRNAs in BA development, we previously conducted RNA sequencing to identify differentially expressed circRNAs in liver tissues collected from patients with BA, using those from patients with congenital choledocho cysts as the control group. The expression of hsa_circ_0009096 was significantly elevated in the hepatic tissues of patients with BA (Zhang et al., 2023). To investigate the effect of hsa_circ_0009096 on hepatic fibrosis in BA, we established a hepatic fibrosis model by treating LX-2 cells with different TGF-β1 concentrations. Using an RT-qPCR assay, we demonstrated that α-SMA and COL1A1 mRNA expression in the LX-2 cells concentration-dependently increased after TGF-β1 treatment (Figs. 2A and 2B). Furthermore, the WB results were consistent with the RT-qPCR data (Fig. 2C). These results suggested that fibrosis was effectively induced in the cultured LX-2 cells. Interestingly, we also confirmed via the RT-qPCR method that TGF-β1 treatment significantly upregulated the hsa_circ_0009096 expression in LX-2 cells (Fig. 2D), suggesting the potential functions of hsa_circ_0009096 in liver fibrosis during BA development.

Hsa_circ_0009096 sponges miR-370-3p to modulate TGFBR2 expression in LX-2 cells

To explore of the potential roles of hsa_circ_0009096 in liver fibrosis in BA development and its downstream targets, we silenced hsa_circ_0009096 expression in the LX-2 cells by transfecting them with three different siRNAs targeting hsa_circ_0009096. The following RT-qPCR assay validated the significant decrease in hsa_circ_0009096 expression in the LX-2 cells by the three siRNAs compared with the NC group (Fig. 3A). The LX-2 cells that were transfected with siRNA-2 showed the greatest decrease in hsa_circ_0009096 expression, which was used for the following assay.

According to our previous prediction of the hsa_circ_0009096-miRNA-TGFβ signaling pathway based on the ceRNA mechanism (Zhang et al., 2023), we selected four target genes TGF beta receptor 2 (TGFBR2), Mothers Against Decapentaplegic Homolog 4 (SMAD4), Transforming Growth Factor Beta 2 (TGFB2), and Thrombospondin 1 (THBS1) for further study based on the following criteria: (1) the target genes are located in the middle and upper reaches of the TGFβ signaling pathway and activate the pathway upon upregulation; (2) multiple miRNAs are likely to regulate the target genes. Using RT-qPCR, we confirmed that the TGFBR2 expression was remarkably suppressed in LX-2 cells by the transfection with hsa_circ_0009096 siRNA in contrast to the negative control group (Fig. 3B). Next, we screened miR-370-3p and miR-21-5p, which target TGFBR2, for validation (Li et al., 2022; Wang et al., 2023). The results showed that the expression of miR-370-3p, but not the miR-21-5p, was then greatly elevated in LX-2 cells transfected with hsa_circ_0009096 siRNA (Fig. 3C). Finally, we performed a dual luciferase reporter assay, which showed that the luciferase activity in the cultured 293T cells expressing the wild-type hsa_circ_0009096 sequence #1 was significantly repressed by miR-370-3p mimics, but not in cells expressing the mutant hsa_circ_0009096 sequence (Fig. 3D). No alteration of luciferase activity was observed in the 293T cells expressing the wild-type hsa_circ_0009096 sequence #2 (Fig. 3D). These results proved that hsa_circ_0009096 binds to miR-370-3p at predicted site #1 to regulate TGFBR2 expression in LX-2 cells.

Hsa_circ_0009096 silencing suppressed TGF-β1-induced HSC proliferation and fibrosis

We analyzed the influence of hsa_circ_0009096 silencing on HSC proliferation and fibrosis using the cellular model established with TGF-β1 treatment. We observed that hsa_circ_0009096 siRNA significantly reduced the COL1A1 and a-SMA mRNA expression in the TGF-β1-treated LX-2 cells compared with the siNC + TGF-β1 group (Figs. 4A and 4B). Moreover, the TGFBR2 mRNA expression levels in the LX-2 cells were greatly promoted following the TGF-β1 treatment, which was also effectively decreased by hsa_circ_0009096 siRNA (Fig. 4C). In contrast, the miR-370-3p expression in LX-2 cells was drastically suppressed by TGF-β1 treatment, which was then significantly recovered by hsa_circ_0009096 siRNA (Fig. 4D). Furthermore, flow cytometry showed that TGF-β1 promoted cell cycle progression from the S phase to the G2/M phases and that hsa_circ_0009096 siRNA significantly induced cell cycle arrest at the G1 stage in the LX-2 cells (Fig. 4E). Hsa_circ_0009096 siRNA partially counteracted the inhibition of apoptosis in the LX-2 cells after TGF-β1 treatment (Fig. 4F). Using ICC, we confirmed that the a-SMA and COL1A1 protein levels in the LX-2 cells increased significantly after TGF-β1 treatment, which were then significantly mitigated by hsa_circ_0009096 siRNA (Fig. 4G). Consistent with the RT-qPCR results, the WB results showed that interference with hsa_circ_0009096 reversed the TGF-β1-induced upregulation of COL1A1, a-SMA, and TGFBR2 protein expression (Fig. 4H). These results revealed that hsa_circ_0009096 could promote TGF-β1-induced HSC proliferation and liver fibrosis by inhibiting miR-370-3p expression and upregulating TGFBR2 expression.

MiR-370-3p mediated hsa_circ_0009096-regulated proliferation and fibrosis of TGF-β1-treated HSCs

To study the potential mediating effects of miR-370-3p in HSC proliferation and fibrosis regulated by hsa_circ_0009096, we suppressed miR-370-3p expression in the LX-2 cells transfected with hsa_circ_0009096 siRNA by treatment with miR-370-3p inhibitors. The miR-370-3p inhibitor significantly inhibited the hsa_circ_0009096 siRNA-induced elevation of the miR-370-3p expression in the LX-2 cells (Fig. 5A). In contrast, hsa_circ_0009096 siRNA repressed the expression of TGFBR2, a-SMA, and COL1A1 in the LX-2 cells, which were all significantly recovered by miR-370-3p inhibitors (Figs. 5B–5D). Furthermore, the LX-2 cell cycle arrest at the G1 stage induced by hsa_circ_0009096 siRNA transfection was significantly alleviated by miR-370-3p inhibitor treatment (Fig. 5E). Similarly, the increase in apoptotic LX-2 cells caused by hsa_circ_0009096 siRNA transfection was also remarkably mitigated by treatment with miR-370-3p inhibitors (Fig. 5F). Finally, our ICC assay showed that the miR-370-3p inhibitors significantly alleviated the hsa_circ_0009096 siRNA-induced downregulation of the a-SMA and COL1A1 protein levels in the LX-2 cells (Fig. 5G). Together, these assays proved that the regulation of HSC proliferation and fibrosis by hsa_circ_0009096 was mediated by its binding with miR-370-3p to modulate TGFBR2 expression.