Targeting hepatic miR-221/222 for therapeutic intervention of nonalcoholic steatohepatitis in mice

Abstract

Background: Effective targeting therapies for common chronic liver disease nonalcoholic steatohepatitis (NASH) are in urgent need. MicroRNA-targeted therapeutics would be potentially an effective treatment strategy of hepatic diseases. Here we investigated the functional role of miR-221/222 and the therapeutic effects of antimiRs-221/222 in NASH mouse models.

Methods: We generated the miR-221/222^flox/flox mice on a C57BL/6 J background and the hepatic miR-221/222 knockout (miR-221/222-LKO) mice. The mice were challenged with the methionine and choline deficient diet (MCDD) or chronic carbon tetrachloride (CCl₄) treatment to generate experimental steatohepatitis models. Adenovirus-mediated re-expression of miR-221/222 was performed on the MCDD-fed miR-221/222-LKO mice. The MCDD and control diet-fed mice were treated with locked nucleic acid (LNA)-based antimiRs of miR-221/222 to evaluate the therapeutic effects. Histological analysis, RNA-seq, quantitative PCR and Western blot of liver tissues were carried out to study the hepatic lipid accumulation, inflammation and collagen deposition in mouse models.

Findings: Hepatic deletion of miR-221/222 resulted in significant reduction of liver fibrosis, lipid deposition and inflammatory infiltration in the MCDD-fed and CCl4-treated mouse models. The hepatic steatosis and fibrosis were dramatically aggravated by miR-221/222 re-expression in MCDD-fed miR-221/222-LKO mice. AntimiRs of miR-221/222 could effectively reduce the MCDD-mediated hepatic steatosis and fibrosis. Systematically mechanistic study revealed that hepatic miR-221/222 controlled the expression of target gene Timp3 and promoted the progression of NASH.

Interpretation: Our findings demonstrate that miR-221/222 are crucial for the regulation of lipid metabolism, inflammation and fibrosis in the liver. LNA-antimiRs targeted miR-221/222 could reduce steatohepatitis with prominent antifibrotic effect in NASH mice. FUND: This work is supported by the Natural Science Foundation of China (81530020, 81390352 to Dr. Ning and 81522032 to Dr. Cao and 81670793 to Dr. Jiang); National Key Research and Development Program (No. 2016YFC0905001 and 2017YFC0909703 to Dr. Cao); the Shanghai Rising-Star Program (15QA1402900 to Dr. Cao); Shanghai Municipal Education Commission-Gaofeng Clinical Medicine Grant (20171905 to Dr. Jiang).

Keywords: Liver fibrosis; NASH; miR-221/222; miRNA-targeted therapeutics.

Fig. 1

MiR-221/222-LKO mice under the challenge of MCDD or CCl₄. (a) The expression of miR-221 and 222 in the livers of control mice treated with MCDD or control diet for 6 weeks (n = 4–6). (b) The expression of miR-221 and 222 in the livers of control mice treated with CCl₄ or vehicle for 6 weeks (n = 3–7). (c) The expression of miR-221 and 222 in the livers of 8-week-old miR-221/222-LKO and control mice (n = 4 for each group). The data represent the mean ± SD, Student's t-test. (d) Serum ALT levels of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks (n = 3–9), and CCl₄ or vehicle for 6 weeks (n = 3–5). The data represent the mean ± SD, one-way ANOVA. (e) H&E staining of the liver sections from miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks, and CCl₄ or vehicle for 6 weeks. Diffuse inflammatory foci containing mononucleated cells (arrowheads) apparent in the livers of control mice under MCDD or CCl₄ treatment. *P < .05, **P < .01, ***P < .001. Scale bars, 100 μm.

Fig. 2

Hepatic deletion of miR-221/222 reduces hepatic inflammation and lipid deposition in NASH mice. (a and b) Triglyceride (TG) levels in the serum (a) and liver (b) of control and miR-221/222-LKO mice treated with MCDD or control diet for 6 weeks (n = 3–5). (c) Oil Red O staining of liver sections from miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks. Scale bars, 100 μm. (d) The expression of Argre1 in the livers of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks (n = 4 for each group). (e) Western blot analyses of F4/80 in the livers of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks (n = 2 for each group). The data shown represent three independent experiments. (f) The expression of Argre1 in the livers of miR-221/222-LKO and control mice treated with CCl₄ or vehicle for 6 weeks (n = 3–5). (g and h) The expression of Il1b, Tnf and Il6 in the livers of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks (n = 3 for each group) (g), and (h) CCl₄ or vehicle for 6 weeks (n = 3 for each group). The data represent the mean ± SD, *P < .05, **P < .01, ***P < .001, one-way ANOVA.

Fig. 3

Hepatic deletion of miR-221/222 reduces liver fibrosis in NASH mice. (a and b) Representative Sirius red staining (a) and Masson staining (b) of the liver sections from miR-221/222-LKO and control mice treated with MCDD or control diet (upper part) and CCl₄ or vehicle (bottom part). Scale bars, 100 μm. (c and d) Analysis of the Sirius red-positive area (c) and Masson-positive area (d) of the liver sections from miR-221/222-LKO and control mice treated with MCDD or control diet (upper part) and CCl₄ or vehicle (bottom part) (n = 4–5). (e) Hepatic hydroxyproline concentration of miR-221/222-LKO and control mice treated with MCDD or control diet (upper part) and CCl₄ or vehicle (bottom part) (n = 3–5). (f) TEM micrographs (arrowheads, collagen) of liver samples from miR-221/222-LKO and control mice treated with MCDD or control diet. Scale bars, 5 μm. Data represent mean ± SD. *P < .05, **P < .01, ***P < .001, one-way ANOVA.

Fig. 4

Hepatic deletion of miR-221/222 reduces the expression of procollagen genes in NASH mice. (a and b) The expression of sma, Col1a1, Col1a2 and Col3a1 in the livers of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks (a) (n = 3–9), and (b) CCl₄ or vehicle diet for 6 weeks (n = 3–5). (c) Western blot analyses of α-SMA in the livers of miR-221/222-LKO and control mice treated with MCDD or control diet for 6 weeks. The relative intensity of the bands was analyzed for three independent experiments. Data represent mean ± SD. *P < .05, **P < .01, ***P < .001, one-way ANOVA.

Fig. 5

Re-expression of hepatic miR-221/222 deteriorates liver steatosis and fibrosis in miR-221/222-LKO mice. (a) The expression of miR-221 and miR-222 in the livers of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 4 for each group). (b) The hepatic TG of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 4 for each group). (c) H&E staining of the liver sections from miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP. (d and e) Representative (d) Sirius red staining and (e) Masson staining of the liver sections from miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP. Scale bars, 100 μm. (f and g) Analysis of the (f) Sirius red-positive area and (g) Masson-positive area of the liver sections of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 4–5). (h and i) The expression of (h) Argre1, Il1b, Tnf, Il6 and (i) sma, Col1a1, Col1a2, Col3a1 in the livers of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 4 for each group). (j) Western blot analyses of F4/80 and α-SMA in the livers of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 3 for each group). The relative intensity of the bands was analyzed. Data represent mean ± SD. *P < .05, **P < .01, ***P < .001, Student's t-test.

Fig. 6

The effects of LNA-antimiR-221/222 in the MCDD-fed mice. (a) The expression of miR-221 and miR-222 in the livers from control mice treated with LNA-antimiRs (LNA-i-miR-221, LNA-i-miR-222, LNA-i-miR-221 + 222) or control (LNA-i-Ctrl) (n = 5–6). The mice were injected with the LNA-antimiRs on days 1, 4, 8,15 and 22 at the dose of 25 mg/kg. All the mice were fed the MCDD from day 1 to 29. (b) The hepatic TG of control mice treated with LNA-antimiRs or control (n = 6 for each group). (c) The expression of Argre1, Il1b, Tnf and Il6 in the livers of control mice treated with LNA-antimiRs or control (n = 5–6). (d) H&E staining of the liver sections from control mice treated with LNA-antimiRs or control. (e and f) Representative (e) Sirius red staining and (f) Masson staining of the liver sections from control mice treated with LNA-antimiRs or control. Scale bars, 100 μm. (g and h) Analysis of the (g) Sirius red-positive area and (h) Masson-positive area of the liver sections (n = 6 for each group). (i) The expression of sma, Col1a1, Col1a2 and Col3a1 in the livers of control mice treated with LNA-antimiRs or control (n = 5–6). Data represent mean ± SD. *P < .05, **P < .01, ***P < .001, one-way ANOVA.

Fig. 7

Hepatic miR-221/222 directly targeted Timp3 in NASH mice. (a) RNA-seq analysis and target gene prediction in the liver of MCDD-fed miR-221/222-LKO mice. Significantly upregulated (P < .05 and fold change >1.5) and predicted target genes of miR-221/222 are listed. (b) The expression of Bmf, Ppargc1b, Timp3, Ddit4 and G6pc in the liver of MCDD-fed miR-221/222-LKO mice by RNA-seq analysis. (c and d) The mRNA and protein levels of Timp3 in the liver of miR-221/222-LKO and control mice treated with MCDD for 6 weeks (n = 3–4). The relative intensity of the bands was analyzed. The protein levels of Timp3 were normalized to that of α-tubulin. (e and f) The mRNA (e) and protein levels (f) of Timp3 in the livers of miR-221/222-LKO mice injected with Ad-miR-221/222 or Ad-GFP (n = 3–4). The relative intensity of the bands was analyzed for Western blots. The data represent the mean ± SD, Student's t-test. (g) The expression of Timp3 in the hepa1–6 cells transfected with LNA-antimiRs or control at the concentration of 50 nM and 100 nM (n = 3 for each group). (h) Western blot analyses of Timp3 in the hepa1–6 cells transfected with LNA-antimiRs or control at the concentration of 100 nM. The data shown represent three independent experiments (n = 3 for each group). (i) The expression of Timp3 in the livers of control mice treated with LNA-antimiRs or control (n = 5–6). The data represent the mean ± SD, *P < .05, **P < .01, one-way ANOVA.

Fig. 8

Inhibition of hepatic Timp3 deteriorates liver steatosis and fibrosis in miR-221/222-LKO mice. (a) The expression of Timp3 in the livers of MCDD-fed miR-221/222-LKO mice injected with adeno-associated virus (AAV8) AAV-Timp3-shRNA or AAV-Ctrl (n = 4 for each group). (b) The hepatic TG and TC of MCDD-fed miR-221/222-LKO mice injected with AAV-Timp3-shRNA or AAV-Ctrl (n = 4 for each group). (c) H&E staining of the liver sections from MCDD-fed miR-221/222-LKO mice injected with AAV-Timp3-shRNA or AAV-Ctrl. (d and e) The expression of (d) Argre1, Il1b, Tnf, Il6 and (e) sma, Col1a1, Col1a2, Col3a1 in the livers of MCDD-fed miR-221/222-LKO mice injected with AAV-Timp3-shRNA or AAV-Ctrl (n = 4 for each group). (f and g) Representative (f) Sirius red staining and (g) Masson staining of the liver sections from MCDD-fed miR-221/222-LKO mice injected with AAV-Timp3-shRNA or AAV-Ctrl. Scale bars, 100 μm. (h and i) Analysis of the (h) Sirius red-positive area and (i) Masson-positive area of the liver sections of MCDD-fed miR-221/222-LKO mice injected with AAV-Timp3-shRNA or AAV-Ctrl (n = 4). Data represent mean ± SD. *P < .05, Student's t-test.