miR-21 ablation and obeticholic acid ameliorate nonalcoholic steatohepatitis in mice

Abstract

microRNAs were recently suggested to contribute to the pathogenesis of nonalcoholic fatty liver disease (NAFLD), a disease lacking specific pharmacological treatments. In that regard, nuclear receptors are arising as key molecular targets for the treatment of nonalcoholic steatohepatitis (NASH). Here we show that, in a typical model of NASH-associated liver damage, microRNA-21 (miR-21) ablation results in a progressive decrease in steatosis, inflammation and lipoapoptosis, with impairment of fibrosis. In a complementary fast food (FF) diet NASH model, mimicking features of the metabolic syndrome, miR-21 levels increase in both liver and muscle, concomitantly with decreased expression of peroxisome proliferator-activated receptor α (PPARα), a key miR-21 target. Strikingly, miR-21 knockout mice fed the FF diet supplemented with farnesoid X receptor (FXR) agonist obeticholic acid (OCA) display minimal steatosis, inflammation, oxidative stress and cholesterol accumulation. In addition, lipoprotein metabolism was restored, including decreased fatty acid uptake and polyunsaturation, and liver and muscle insulin sensitivity fully reinstated. Finally, the miR-21/PPARα axis was found amplified in liver and muscle biopsies, and in serum, of NAFLD patients, co-substantiating its role in the development of the metabolic syndrome. By unveiling that miR-21 abrogation, together with FXR activation by OCA, significantly improves whole body metabolic parameters in NASH, our results highlight the therapeutic potential of nuclear receptor multi-targeting therapies for NAFLD.

Figure 1

miR-21 ablation protects from MCD-induced steatohepatitis and fibrosis development. C57BL/6 WT and miR-21 KO mice were fed a control (n=5 for each time point) or MCD diet (n=8–10 for each time point and/or genetic background) for 2 and 8 weeks. (a) qRT-PCR analysis of miR-21 (left) and immunoblotting of PPARα (right). Representative blots are shown. Blots were normalized to endogenous β-actin. (b) H&E and Oil Red-O staining of representative liver sections. Scale bar, 100 μm. (c) qRT-PCR analysis of TNF-α and IL-1β. (d) Serum ALT levels. (e) TUNEL staining of liver tissue sections. Nuclei were counterstained with Hoechst 33258 (blue). Scale bar, 30 μm. Histograms showing the percentage of TUNEL-positive cells (right). (f) Immunoblotting of active caspase-2. Representative blots are shown. Blots were normalized to endogenous β-actin. (g) qRT-PCR analysis of TGF-β and collagen-1α1 in mouse liver (left) and representative images of Masson's Trichrome stained liver sections from MCD or control diet-fed mice for 8 weeks (right). Scale bar=100 μm. Results are expressed as mean±S.E.M. fold change. *P<0.05, **P<0.01 and ***P<0.001 from control; ^†P<0.05, ^‡P<0.01 and ^ϕP<0.001 compared with respective WT MCD-fed mice

Figure 1

miR-21 ablation protects from MCD-induced steatohepatitis and fibrosis development. C57BL/6 WT and miR-21 KO mice were fed a control (n=5 for each time point) or MCD diet (n=8–10 for each time point and/or genetic background) for 2 and 8 weeks. (a) qRT-PCR analysis of miR-21 (left) and immunoblotting of PPARα (right). Representative blots are shown. Blots were normalized to endogenous β-actin. (b) H&E and Oil Red-O staining of representative liver sections. Scale bar, 100 μm. (c) qRT-PCR analysis of TNF-α and IL-1β. (d) Serum ALT levels. (e) TUNEL staining of liver tissue sections. Nuclei were counterstained with Hoechst 33258 (blue). Scale bar, 30 μm. Histograms showing the percentage of TUNEL-positive cells (right). (f) Immunoblotting of active caspase-2. Representative blots are shown. Blots were normalized to endogenous β-actin. (g) qRT-PCR analysis of TGF-β and collagen-1α1 in mouse liver (left) and representative images of Masson's Trichrome stained liver sections from MCD or control diet-fed mice for 8 weeks (right). Scale bar=100 μm. Results are expressed as mean±S.E.M. fold change. *P<0.05, **P<0.01 and ***P<0.001 from control; ^†P<0.05, ^‡P<0.01 and ^ϕP<0.001 compared with respective WT MCD-fed mice

Figure 2

miR-21 ablation and OCA prevent FF-induced hepatomegaly and strongly ameliorate steatosis, cholesterol accumulation and inflammation. C57BL/6 WT and miR-21 KO mice were fed a FF diet (n=12) or a control diet (SD; n=12), with or without supplementation with OCA (10 mg/kg/day) for 25 weeks. (a) Mice body weight (left) and liver to body weight ratio (right). (b) qRT-PCR analysis of miR-21 (left) and immunoblotting of PPARα (right). Representative blots are shown. Blots were normalized to endogenous β-actin. (c) Representative image of H&E-stained liver sections. Scale bar, 100 μm. (d) Steatosis score in blinded liver samples (top). Steatosis was graded on a scale 0–4. Liver cholesterol content (bottom). (e) qRT-PCR analysis of TNF-α, IL-1β, IL-6 and TLR4. Results are expressed as mean±S.E.M. fold change. *P<0.05, **P<0.01 and ***P<0.001 from WT control mice; ^†P<0.05, ^‡P<0.01 and ^ϕP<0.001 compared with WT FF-fed mice. SD; standard diet

Figure 3

Expression of metabolism-relevant transcriptional targets of PPARα is restored upon dual nuclear receptor activation. C57BL/6 WT and miR-21 KO mice were fed a FF diet (n=12) or a control diet (SD; n=12), with or without supplementation with OCA (10 mg/kg/day) for 25 weeks. (a) qRT-PCR analysis of Cyp4a14, FAT, ACOX-2 and CPT-1. (b) ROS levels (top) and qRT-PCR analysis of HO-1 (bottom). Results are expressed as mean±S.E.M. fold change. *P<0.05 and ***P<0.001 from WT control mice; ^†P<0.05, ^‡P<0.01 and ^ϕP<0.001 compared with WT FF-fed mice. SD, standard diet

Figure 4

Hepatic insulin sensitivity is restored upon PPARα and FXR activation. C57BL/6 WT and miR-21 KO mice were fed a FF diet (n=12) or a control diet (SD; n=12), with or without supplementation with OCA (10 mg/kg/day) for 25 weeks. (a) Immunoblotting of p-JNK. Representative blots are shown. Blots were normalized to total JNK. (b) Immunoblotting of p-IRS1 (left) and p-INSR (right). Representative blots are shown. Blots were normalized to total IRS1 and INSR respectively. (c) Immunblotting of p-AKT. Representative blots are shown. Blots were normalized to total AKT. Results are expressed as mean±S.E.M. fold change. *P<0.05 and **P<0.01 from WT control mice; ^†P<0.05 and ^ϕP<0.001 compared with WT FF-fed mice. SD, standard diet

Figure 5

The miR-21/PPARα axis is modulated in the muscle of mice with NASH, and muscle insulin resistance is prevented upon miR-21 ablation and FXR activation. C57BL/6 WT and miR-21 KO mice were fed a FF diet (n=12) or a control diet (SD; n=12), with or without supplementation with OCA (10 mg/kg/day) for 25 weeks. (a) qRT-PCR analysis of miR-21 (left) and immunoblotting of PPARα (right). Representative blots are shown. Blots were normalized to endogenous tubulin. (b) Immunoblotting of p-IRS1 (left) and p-INSR (right). Representative blots are shown. Blots were normalized to total IRS1 and INSR, respectively. (c) Immunblotting of p-AKT. Representative blots are shown. Blots were normalized to total AKT. Results are expressed as mean±S.E.M. fold change. *P<0.05 and **P<0.01 from WT control mice; ^†P<0.05 compared with WT FF-fed mice. SD, standard diet

Figure 6

miR-21 expression is increased in the liver, muscle and serum of NAFLD patients. (a) qRT-PCR analysis of liver miR-21. (b) Representative PPARα immunostaining (red) in liver tissue from patients. Nuclei were counterstained with Hoechst 33258 (blue). The corresponding histogram shows the quantification of PPARα mean fluorescence intensity, as described in Materials and Methods section. At least five images per liver sample were used. Scale bar=10 μm. (c) qRT-PCR analysis of muscle miR-21. (d) qRT-PCR analysis of serum miR-21. Results are expressed as mean±S.E.M. fold change. *P<0.05 and **P<0.01 from steatosis