Protective effects of farnesoid X receptor (FXR) on hepatic lipid accumulation are mediated by hepatic FXR and independent of intestinal FGF15 signal

Abstract

Background & aims: There is a growing evidence that bile acids are involved in the regulation of triglyceride-, cholesterol-homoeostasis and fat absorption. In this study organ-specific Fxr knockout mice were used to further investigate the influence of farnesoid X receptor FXR in lipogenesis.

Methods: Liver- and intestine-specific Fxr knockout mice were fed a 1% cholesterol diet for 28 days. Histological examination of frozen tissue sections included Sudan III/H&E, BODIPY staining and liver X receptor (LXR) immunohistochemistry. Liver triglycerides, serum cholesterol, serum bile acids and nuclear LXR protein were measured. mRNA expression of several genes involved in bile acid-, cholesterol-homoeostasis and lipogenesis was quantified by real-time PCR.

Results: Hepatic FXR deficiency contributes to lipid accumulation under 1% cholesterol administration which is not observed in intestinal Fxr knockout mice. Strong lipid accumulation, characterized by larger vacuoles could be observed in hepatic Fxr knockout sections, while intestinal Fxr knockout mice show no histological difference to controls. In addition, these mice have the ability to maintain normal serum cholesterol and bile acid levels. Hepatic Fxr knockouts were characterized by elevated triglycerides and bile acid levels. Expression level of LXR was significantly elevated under control and 1% cholesterol diet in hepatic Fxr knockout mice and was followed by concomitant lipogenic target gene induction such as Fas and Scd-1. This protective FXR effect against hepatic lipid accumulation was independent of intestinal Fgf15 induction.

Conclusion: These results show that the principal site of protective bile acid signalling against lipid accumulation is located in the liver since the absence of hepatic but not intestinal FXR contributes to lipid accumulation under cholesterol diet.

Keywords: FXR; NAFLD; bile acids; lipogenesis.

Figure 1

Liver histology. Representative hematoxylin and Sudan III staining of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice under control diet and 1% cholesterol diet (left panel). Representative BODIPY staining of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice under control diet and 1% cholesterol diet are shown in a lower magnification (right panel). Massive lipid accumulation can be seen in ΔL Fxr knockout mice under 1% cholesterol diet. Insert shows tissue in a higher magnification. Scale bars: 10 μm.

Figure 2

Lipid content and serum parameters in organ-specific knockout animals. A. liver triglycerides B. serum cholesterol in mg/dl C. total liver cholesterol in μg/g D. total serum bile acids in μM/l and E. serum glucose in % of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice under control diet and 1% cholesterol diet. Asterisk represents P<0.05 as determined by the Mann-Whitney U-test.

Figure 3

Relative quantification of the transcripts of nuclear receptors and their targets. RT-PCR was performed with RNA samples isolated from liver and ileum of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice. A. hepatic genes were analysed: Fxr, Shp, Cyp7a1 and Bsep. B. intestinal genes were analysed: Fxr, Ibap and Fgf15. Data represent means ± SEM (n = 4-5). Asterisk represents P<0.05 as determined by the Mann-Whitney U-test.

Figure 4

Liver × Receptor expression, localization and relative quantification of genes involved in cholesterol homeostasis. Respective liver tissue was used to isolate nuclear fractionations using the NE-PER Nuclear Extraction Reagents kit. A. hepatic LXR mRNA expression B. Nuclear LXR Protein content. Similar amounts of isolate nuclear fractionations from liver tissue, was analyzed. Nuclear LXR protein was determined using ELISA Kit for Liver X Receptor Alpha. LXR protein was normalized against total nuclear protein content. C. LXR immunohistochemistry. Increased nuclear localization in animals under 1% cholesterol diet. Increased localisation of LXR, is shown in (ΔL) Fxr knockout animals under standard diet. flox/flox littermates under the same treatment show lower nuclear LXR localisation. RT-PCR was performed with RNA samples isolated from liver of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice for D. HmgCoAr, E. Abcg5 and F. Abcg8. Data represent means ± SEM (n = 4-5). Asterisk represents P<0.05 as determined by the Mann-Whitney U-test.

Figure 5

Quantification of genes involved in lipogenesis. A. SREBP-1c protein expression was measured by western blot. Decreased concentration of high molecular weight (pre) SREBP1-c, results in increased lower molecular weight (nuclear) SREBP1-c, indicating SREBP1-c proteolytic activation. B. Densitometric quantification of Western blots was performed using Adobe Photoshop CS3, showing an increased SREBP1-c activation in ΔL Fxr animal under control diet. RTPCR was performed on RNA samples isolated from liver of flox/flox, ΔL Fxr and ΔIE Fxr knockout mice for C. Fas, D. Scd-1 and E. Pnpla3. Data represent means ± SEM (n = 4-5). Asterisk represents P<0.05 as determined by the Mann-Whitney U-test.

Figure 6

Organ-specific protective effect of FXR against lipid accumulation. A. flox/flox controls B. ΔIE Fxr knock out C. ΔL Fxr knock out mice under 1% cholesterol diet. The principal site of protective FXR signalling against lipid accumulation is located in the liver since the absence of hepatic but not intestinal FXR contributes to lipid accumulation under 1% cholesterol diet. In the absence of hepatic FXR elevated LXR expression can be observed together with the induction of lipogenic LXR-target genes such as SCD-1. It could be suggested that the observed effects are independent of intestinal Fgf15 induction, as exclusively hepatic and not intestinal FXR deletion contributes to lipid accumulation under 1% cholesterol diet.