Circular RNA cVIM promotes hepatic stellate cell activation in liver fibrosis via miR-122-5p/miR-9-5p-mediated TGF-β signaling cascade

Abstract

Hepatic stellate cell (HSC) activation is considered as a central driver of liver fibrosis and effective suppression of HSC activation contributes to the treatment of liver fibrosis. Circular RNAs (circRNAs) have been reported to be important in tumor progression. However, the contributions of circRNAs in liver fibrosis remain largely unclear. The liver fibrosis-specific circRNA was explored by a circRNA microarray and cVIM (a circRNA derived from exons 4 to 8 of the vimentin gene mmu_circ_32994) was selected as the research object. Further studies revealed that cVIM, mainly expressed in the cytoplasm, may act as a sponge for miR-122-5p and miR-9-5p to enhance expression of type I TGF-β receptor (TGFBR1) and TGFBR2 and promotes activation of the TGF-β/Smad pathway, thereby accelerating the progression of liver fibrosis. Our results demonstrate a vital role for cVIM in promoting liver fibrosis progression and provide a fresh perspective on circRNAs in liver fibrosis.

Fig. 1. Deregulated circRNAs and up-regulation of cVIM in liver fibrosis.

Masson staining was performed in CCl₄ mice (a) and BDL mice (b), n = 6 mice per group. Scale bar, 100 μm. c Heat map for differentially expressed circRNAs analyzed by circRNA arraystar ChIP between the fibrotic tissues and the control tissues (n = 3 per group). Expression of cVIM in the fibrotic tissues in CCl₄ (d) and BDL mice (e), n = 6 mice per group. f Expression of cVIM in primary HSCs isolated from healthy mice during culture days (n = 6 per group). Each value is the mean ± SD of six independent experiments. **P < 0.01 compared to the control.

Fig. 2. The characteristics of cVIM.

Primary 1-day-old HSCs were isolated from CCl₄-treated mice. a Scheme illustrating the production of cVIM. b Random hexamer or oligo (dT)₁₈ primers were applied to perform the reverse transcription experiments and qRT-PCR was used to detect cVIM expression in primary 1-day-old HSCs (n = 3 per group). The random hexamer primers group was used as the control. c Relative cVIM expression was detected in primary 1-day-old HSCs after Rnase R treatment (n = 3 per group). d Relative cVIM expression was detected in primary 1-day-old HSCs by qRT-PCR after Actinomycin D at the indicated time points (n = 3 per group). e cVIM was mainly expressed in the cytoplasm in primary 1-day-old HSCs (n = 3 per group). GAPDH and U6 were applied as positive controls in the cytoplasm and nucleus, respectively. f RNA FISH for cVIM in primary 1-day-old HSCs (n = 3 per group). DAPI stained nuclei blue. Scale bar, 50 μm. Each value is the mean ± SD of three independent experiments. **P < 0.01 and ***P < 0.001 compared to the control.

Fig. 3. Loss of cVIM inhibits the progression of liver fibrosis in vitro and in vivo.

Primary 1-day-old HSCs isolated from CCl₄-treated mice were transfected with cVIM siRNA using Lipofectamine RNAiMAX for 24 h, 48 h and 72 h. a CCK8 assay showed the inhibitory role of loss of cVIM in HSCs (n = 3 per group). b mRNA expressions of α-SMA and Col1A1 (n = 3 per group). c Immunofluorescence staining for α-SMA (green) and type I collagen (red) were evaluated by confocal laser microscopy (n = 3 per group). DAPI stained the nuclei blue. The scale bar represents 20 μm. d Masson staining and α-SMA immunohistochemistry in CCl₄ mice after cVIM knockdown (n = 6 per group). The scale bar represents 100 μm. e Analysis of Masson staining and α-SMA immunohistochemistry (n = 6 per group). f ALT value (n = 6 per group). g Type I collagen expression in vivo (n = 3 per group). Each value is the mean ± SD of three independent experiments. *P < 0.05 and **P < 0.01 compared to the control.

Fig. 4. cVIM may act as a sponge for miR-122-5p and miR-9-5p.

a RIP experiments (n = 3 per group). b circRIP was performed in cVIM-overexpressing HSCs using a cVIM-specific probe and control probe, respectively (n = 3 per group). The enrichments of cVIM and miRNAs were examined by qRT-PCR and normalized to the control probe. c The luciferase activity of luc-cVIM in HEK-293T cells with miR-122-5p or miR-9-5p mimics (n = 3 per group). d Scheme illustrating the putative binding sites of miR-122-5p and miR-9-5p with respect to cVIM. e The luciferase activity of luc-cVIM or luc-cVIM-mutant in HEK-293T cells after co-transfection with miR-122-5p or miR-9-5p (n = 3 per group). f Pull-down assay to validate the direct interaction between cVIM and miR-122-5p/miR-9-5p (n = 3 per group). Bio-miR-NC is not complementary to cVIM. Each value is the mean ± SD of three independent experiments. **P < 0.01 and ***P < 0.001.

Fig. 5. cVIM promotes HSC activation via miR-122-5p/miR-9-5p-midaited TGF-β pathway.

Primary 1-day-old HSCs isolated from CCl₄-treated mice were transduced with Ad-cVIM for 48 h and then transfected with miR-122-5p/miR-9-5p mimics for additional 24 h. In addition, cells were transduced with Ad-shcVIM for 48 h and then transfected with miR-122-5p/miR-9-5p inhibitor for additional 24 h. qRT-PCR (a) and Western blotting (b) analysis showed the mRNA and protein expressions of TGFBR1 and TGFBR2 after overexpressing or silencing cVIM (n = 3 per group). c p-Smad2 level in cVIM overexpressing-HSCs after co-transfection with miR-122-5p or miR-9-5p mimics (n = 3 per group). d Cell proliferation in cVIM overexpressing-HSCs after co-transfection with miR-122-5p or miR-9-5p mimics and cells with loss of cVIM after co-transfection with miR-122-5p or miR-9-5p inhibitor (n = 3 per group). e Col1A1 and α-SMA mRNA in cVIM overexpressing-HSCs after co-transfection with miR-122-5p or miR-9-5p mimics and cells with loss of cVIM after co-transfection with miR-122-5p or miR-9-5p inhibitor (n = 3 per group). f Type I collagen in cVIM overexpressing-HSCs after co-transfection with miR-122-5p or miR-9-5p mimics (n = 3 per group). Each value is the mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 and ***P < 0.001.

Fig. 6. Sp1 activates cVIM expression in liver fibrosis.

Primary 1-day-old HSCs isolated from CCl₄-treated mice were transfected with Sp1 siRNA using Lipofectamine RNAiMAX for 48 h. a Sp1 binding site prediction in the cVIM promoter region using JASPAR. b qRT-PCR analysis of cVIM and Sp1 expression in HSCs after Sp1 siRNA or negative control treatment (n = 3 per group). c ChIP-qPCR analysis of Sp1 occupancy in the cVIM promoter in HSCs (n = 3 per group). DHFR was used as positive control and IgG was used as a negative control. d Construction of the luciferase reporter vector cVIM-P1/P2 (containg all Sp1 binding sites), cVIM-P1 (containg −672 and −663 binding sites) and cVIM-P2 (containg −129 and −119 binding sites). e Luciferase assays of the cells indicated that were transfected with cVIM-P1/P2, cVIM-P1, cVIM-P2 vectors, the Sp1 vector, or an empty vector (n = 3 per group). f Summary of the regulation and mechanism of cVIM in liver fibrosis. Each value is the mean ± SD of three independent experiments. **P < 0.01 compared to the control.