The farnesoid X receptor regulates adipocyte differentiation and function by promoting peroxisome proliferator-activated receptor-gamma and interfering with the Wnt/beta-catenin pathways

Abstract

The bile acid receptor farnesoid X receptor (FXR) is expressed in adipose tissue, but its function remains poorly defined. Peroxisome proliferator-activated receptor-γ (PPARγ) is a master regulator of adipocyte differentiation and function. The aim of this study was to analyze the role of FXR in adipocyte function and to assess whether it modulates PPARγ action. Therefore, we tested the responsiveness of FXR-deficient mice (FXR(-/-)) and cells to the PPARγ activator rosiglitazone. Our results show that genetically obese FXR(-/-)/ob/ob mice displayed a resistance to rosiglitazone treatment. In vitro, rosiglitazone treatment did not induce normal adipocyte differentiation and lipid droplet formation in FXR(-/-) mouse embryonic fibroblasts (MEFs) and preadipocytes. Moreover, FXR(-/-) MEFs displayed both an increased lipolysis and a decreased de novo lipogenesis, resulting in reduced intracellular triglyceride content, even upon PPARγ activation. Retroviral-mediated FXR re-expression in FXR(-/-) MEFs restored the induction of adipogenic marker genes during rosiglitazone-forced adipocyte differentiation. The expression of Wnt/β-catenin pathway and target genes was increased in FXR(-/-) adipose tissue and MEFs. Moreover, the expression of several endogenous inhibitors of this pathway was decreased early during the adipocyte differentiation of FXR(-/-) MEFs. These findings demonstrate that FXR regulates adipocyte differentiation and function by regulating two counteracting pathways of adipocyte differentiation, the PPARγ and Wnt/β-catenin pathways.

FIGURE 1.

Obese FXR^−/−/ob/ob mice are resistant to the effects of PPARγ activation on adipocyte recruitment. A, increase of brown adipose tissue (BAT) mass in FXR^−/− and FXR^+/+/ob/ob mice after rosiglitazone (RSG) treatment. B, adipocyte size distribution in adipose tissue of FXR^−/− and FXR^+/+/ob/ob mice treated or not with RSG. 200 adipocytes were studied per section. Adipocyte size was measured using ImageJ. C, morphology of white adipocytes of FXR^−/− and FXR^+/+/ob/ob mice (n = 3/group) treated with RSG (10 mg/kg). Small adipocytes recruited after RSG treatment are indicated by the arrow. **, p < 0.01; ***, p < 0.001.

FIGURE 2.

FXR deficiency alters the expression profile of white adipose tissue genes following rosiglitazone treatment. A–C, mRNA expression of PPARγ target genes (A), adipogenic transcription factor genes (B), and lipid droplet genes (C) in white adipose tissue of FXR^−/− and FXR^+/+/ob/ob mice after rosiglitazone treatment. mRNA levels were measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to control FXR^+/+ mice. LPL, lipoprotein lipase. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 3.

FXR^−/− MEFs are resistant to PPARγ activation. mRNA levels of adipogenic genes of 4- and 8-day-differentiated FXR^−/− and FXR^+/+ MEFs in the presence or absence of 1 μm RSG were measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to those at day 0, which are arbitrarily set to 1. These results are representative of three experiments. *, p < 0.05; **, p < 0.01.

FIGURE 4.

FXR deficiency alters triglyceride storage, lipolysis, de novo lipogenesis, and the expression of lipid droplet genes in MEFs during differentiation to adipocytes. A, TG content in FXR^−/− and FXR^+/+ MEFs at days 0, 4, and 8 of differentiation treated with 1 μm rosiglitazone; lipolysis was measured in FXR^−/− and FXR^+/+ MEFs at days 0, 4, and 8 as glycerol release under basal and stimulated (isoproterenol: ISO) conditions. De novo lipogenesis in FXR^−/− and FXR^+/+ MEFs was measured at days 0, 4, and 8 of differentiation. The results are representative of three experiments and are presented as means ± S.D. B, quantification of the lipid droplet size of 8-day-differentiated FXR^−/− and FXR ^+/+ MEFs. C, mRNA expression of genes coding for lipid droplet proteins in FXR^−/− and FXR^+/+ MEFs during differentiation measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to those at day 0, which are arbitrarily set to 1. The results are representative of three experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 5.

FXR deficiency impairs PPARγ-induced differentiation and lipid droplet formation of primary preadipocytes. A and C, expression of adipogenic marker (A) and lipid droplet protein (C) genes in FXR^−/− when compared with FXR^+/+ preadipocytes treated with 1 μm rosiglitazone. Preadipocytes were isolated from white adipose tissue of obese FXR^−/−/ob/ob and FXR^+/+/ob/ob mice. mRNA levels were measured by quantitative PCR. Values (± S.D.) are normalized to the expression of cyclophilin and are expressed relative to those at day 0, which are arbitrarily set to 1. B, smaller size of lipid droplets in FXR^−/− when compared with FXR^+/+ preadipocytes. Representative Oil Red O staining of FXR^−/− and FXR^+/+ preadipocytes at days 0 and 8 of differentiation is shown (×20 magnification). *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 6.

Re-expression of FXR reverses the impaired adipocyte differentiation of rosiglitazone-treated FXR^−/− MEFs. FXR^−/− MEFs were infected with a retrovirus coding for FXRα3 or the empty vector, subjected to adipogenic differentiation, and treated with 1 μm rosiglitazone. A, increased number of differentiated cell clusters after FXR retroviral infection of rosiglitazone-treated FXR^−/− MEFs. Representative Oil Red O staining of FXR^−/− and FXR^+/+ MEFs at day 8 of differentiation is shown (×20 magnification). B and C, mRNA levels of adipogenic markers (B) and lipid droplet protein (C) genes in empty retrovirus- and FXR retrovirus-infected rosiglitazone-treated FXR^−/− MEFs measured by quantitative PCR. Empty retrovirus-transfected FXR^+/+ MEFs were used as reference. Values are normalized to cyclophilin mRNA and are expressed relative to those at day 0, which are arbitrarily set to 1. The results are representative of three experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 7.

Up-regulation of the Wnt/β-catenin signaling pathway in adipose tissue of FXR^−/−/ob/ob mice and FXR^−/− MEFs. A, mRNA expression of β-catenin, LRP5, c-Myc, Axin2, and cyclin D1 in inguinal adipose tissue of FXR^−/− and FXR^+/+/ob/ob mice. mRNA levels were measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to FXR^+/+ mice, which are arbitrarily set to 1. The results are presented as means ± S.D. B, mRNA expression of sFRP1 and sFRP5 during differentiation of FXR^−/− and FXR^+/+ MEFs treated with 1 μm rosiglitazone. C, β-catenin protein levels during FXR^−/− and FXR^+/+ MEF differentiation. D, mRNA levels of Wnt/β-catenin target genes during differentiation of FXR^−/− and FXR^+/+ MEFs. E, top, β-catenin protein levels. Bottom, mRNA levels of c-Myc, Axin2, and LRP5 in FXR^+/+ and FXR^−/− MEFs transfected for 2 days with empty of FXR retrovirus as indicated. mRNA levels were measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to those at day 0, which are arbitrarily set to 1. The results are representative of two experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001. C.I. acid red, Ponceau S.

FIGURE 8.

sFRP1 reduces β-catenin protein (A) and increases PPARγ and aP2 gene expression in FXR−/− MEFs (B). FXR^−/− MEFs were differentiated in the presence or absence of recombinant sFRP1 (75 nmol/liter). A, β-catenin protein levels. B, mRNA levels of adipogenic genes measured by quantitative PCR. Values are normalized to the expression of cyclophilin and are expressed relative to those in FXR^+/+ MEFs, which are arbitrarily set to 1. *, p < 0.05; **, p < 0.01. C.I. acid red, Ponceau S.