microRNA-29b prevents liver fibrosis by attenuating hepatic stellate cell activation and inducing apoptosis through targeting PI3K/AKT pathway

Abstract

microRNA-29b (miR-29b) is known to be associated with TGF-β-mediated fibrosis, but the mechanistic action of miR-29b in liver fibrosis remains unclear and is warranted for investigation. We found that miR-29b was significantly downregulated in human and mice fibrotic liver tissues and in primary activated HSCs. miR-29b downregulation was directly mediated by Smad3 through binding to the promoter of miR-29b in hepatic stellate cell (HSC) line LX1, whilst miR-29b could in turn suppress Smad3 expression. miR-29b transduction in the liver of mice prevented CCl4 induced-fibrogenesis, concomitant with decreased expression of α-SMA, collagen I and TIMP-1. Ectopic expression of miR-29b in activated HSCs (LX-1, HSC-T6) inhibited cell viability and colony formation, and caused cell cycle arrest in G1 phase by downregulating cyclin D1 and p21cip1. Further, miR-29b induced apoptosis in HSCs mediated by caspase-9 and PARP. miR-29b inhibited its downstream effectors of PIK3R1 and AKT3 through direct targeting their 3'UTR regions. Moreover, knockdown of PIK3R1 or AKT3 suppressed α-SMA and collagen I and induced apoptosis in both HSCs and in mice. In conclusion, miR-29b prevents liver fibrogenesis by inhibiting HSC activation and inducing HSC apoptosis through inhibiting PI3K/AKT pathway. These results provide novel mechanistic insights for the anti-fibrotic effect of miR-29b.

Keywords: AKT3; hepatic stellate cell; liver fibrosis; miR-29b.

Figure 1. miR-29b is downregulated in liver fibrosis and in activated hepatic stellate cells (HSCs) and down-regulation of miR-29b is mediated by Smad3

(A) Level of miR-29b was significantly lower in human fibrotic liver tissues (n = 20) than in normal liver tissues (n = 13). (B) Representative morphological images of primary cultured quiescent HSCs and activated HSCs of rat origin. (C) mRNA expression of key genes involved in the activation of HSCs including α-SMA, DDR2, FN1, ITGB1 and PDGFRb were analyzed by real-time PCR. (D) miR-29b expression is reduced in activated HSCs compred to the quiescent HSCs. (E) Human HSC cell line LX1 was treated with TGF-β1 (2 ng/ml) for 0, 3h, 6h, 12h and 24h, respectively. Protein expression of phosphorylated-Smad3 (p-Smad3) and α-SMA was determined by Westerb blot, GAPDH was used as loading control. (F) miR-29b expression was examined by real-time RT-PCR. (G) DNA sequence alignments indicate a highly conserved Smad3 binding site at the promoter region 22kb upstream of miR-29b in difference species (upper panel). Functional interaction between Smad3 and miR-29b promoter was evaluated by Chromatin Immunoprecipitation (ChIP)-PCR. LX1 cells were treated with TGF-β1 (2ng/ml) for 24 h, then cross-linked with formaldehyde and lysed. The soluble chromatin was immunoprecipitated with the anti-Smad3 Ab. Two pairs of primers were designed to detect the Smad3-containing promoter region of miR-29b by ChIP-PCR (lower panel) and a direct interaction be Smad3 and miR-29b was demonstrated.

Figure 2. Ultrasound-mediated gene transfer of miR-29b in CCl-induced liver fibrosis in C57BL6 mice

(A) C57/BL6 mice were administered CCl₄ twice per week by Intraperitoneal injection for 8 weeks to induce liver fibrosis. Separately, CCl₄ treated mice were introduced with miR-29b using pTRE2-miR-29b-Tet-on plasmid or control vector by tail vein injection, followed by 5 min ultrasound treatment (2W/cm2) transcutaneously on liver location as described in Materials and Methods. (B) The efficacy of miR-29b delivery to the liver of mice was determined by quantitative real-time PCR and consistent miR-29b expression was detected at week 1 and week 2. Thus, a tail vein injection was given for 3 times in 8 weeks duration in mice treated with CCl₄. (C) Over-expression of miR-29b was confirmed in miR-29b-transfected mice in the liver by real-time RT-PCR and (D) by in situ hybridization.

Figure 3. Gene transfer of miR-29b prevents CCl4-induced liver fibrosis in mice

(A) Collagen deposition by Picrosirius red staining of liver. (B) Semi-quantitative analysis of collagen area by Picrosirius red staining. (C) The quantification of collagen in liver tissues of mice was displayed by content of hydroxyproline (ug/mg liver) in different treatment groups. (D) Effects of CCl₄ and transduced with miR-29b on hepatic protein expression of p-Smad3, collagen I, α-SMA and TIMP-1 by Western blot. GAPDH was used as loading control. Values are mean ± SD, *P < 0.001 compared with control mice (treated with oil); #P < 0.01, ##P < 0.001 compared with mice treated with CCl₄. (E) Proposed scheme of the mechanism by which miR-29b prevents liver fibrogenesis in mice.

Figure 4. miR-29b inhibits HSC proliferation and arrests cell cycle in G1 phase

(A) Ectopic expression of miR-29b in HSC cell lines LX1 and HSC-T6 was confirmed by RT-PCR. (B) mRNA expression of α-SMA, DDR2, FN1, ITGB1 and PDGFRb in miR-29b expressed HSCs and control HSCs by real-time PCR. (C) Eectopic expression of miR-29b suppressed protein expression of p-Smad3, α-SMA, collagen I and TIMP-1 in LX1 and HSC-T6. (D). miR-29b inhibited HSC cell growth as determined by cell viability assay, and (E) by colony formation assay. (F) miR-29b increased G1 phase cell population, but decreased S phase cell population as examined by flow cytometry analysis. (G) Protein expression of key G1 cell cycle regulators cyclinD1 and p21^cip1 by Western blot. GAPDH was used as loading control.

Figure 5. miR-29b induces HSC apoptosis

(A) HSCs apoptosis was determined by Annexin V/7-AAD staining and analyzed by flow cytometry. Annexin V positive apoptotic cells were significantly increased in LX1 cells transfected with miR-29b compared with those transfected with control-miR. Data are mean ± SD from four independent experiments in duplicate. (B) Cell apoptosis in HSCs was examined by TUNEL staining. Apoptosis index was quantified by counting the proportion of TUNEL-positive cells. (C) Effects of miR-29b on protein expression of apoptosis-related genes by Western blot. GAPDH was used as loading control.

Figure 6. miR-29b inhibits liver fibrosis and suppresses the activation of HSCs through direct targeting PIK3R1 and AKT3

(A) miR-29b potential binding sites on the 3′-UTR of four candidate genes, PIK3R1, AKT3, Col1A2 and Col3A1. (B) LX1 cells were transfected with firefly luciferase transcript containing either wild-type or mutant form of 3′-UTR of the 4 candidate genes, in the presence of either control or miR-29b precursor, and then assessed for luciferase reporter activity at 48 hours post-transfection. The luciferase reporter activity of PIK3R1ang AKT3 was suppressed by wildtype miR-29b. (C) Protein expression of PIK3R1, AKT3 and p-AKT3 was reduced by miR-29b in LX1 and in HSC-T6 cells. (D) PIK3R1, AKT3 and p-AKT expressions which were induced in CCl₄-treated mice were down-regulated in mice with gene transfer of miR-29b. Values are mean ± SD, *P < 0.01 compared with control mice (treated with oil); #P < 0.05, ##P < 0.01 compared with mice treated with CCl₄.

Figure 7. Knockdown of PIK3R1 or AKT3 reduces HSC activation and induces HSC apoptosis

(A) Knockdown efficiency of PIK3R1 or AKT3 in LX1 cell line was confirmed by RT-PCR and (B) by Western blot. (C) Knockdown of PIK3R1 or AKT3 inhibits the protein levels of α-SMA and collagen I in LX1 cells by western blot, and (D) by dual-immunofluorescence staining. (E) Knockdown of PIK3R1 or AKT3 induced Annexin V positive apoptotic LX1 cells by flow cytometry. Values are mean ± SD from four independent experiments in duplicate. (F) Induced apoptosis in LX1 cells by Knockdown of PIK3R1 or AKT3 was confirmed by TUNEL staining. Apoptosis index was quantified by counting the proportion of TUNEL-positive cells.