The farnesoid X receptor regulates transcription of 3beta-hydroxysteroid dehydrogenase type 2 in human adrenal cells

Abstract

Recent studies have shown that the adrenal cortex expresses high levels of farnesoid X receptor (FXR), but its function remains unknown. Herein, using microarray technology, we tried to identify candidate FXR targeting genes in the adrenal glands, and showed that FXR regulated 3beta-hydroxysteroid dehydrogenase type 2 (HSD3B2) expression in human adrenocortical cells. We further demonstrated that FXR stimulated HSD3B2 promoter activity and have defined the cis-element responsible for FXR regulation of HSD3B2 transcription. Transfection of H295R adrenocortical cells with FXR expression vector effectively increased FXR expression levels and additional treatment with chenodeoxycholic acid (CDCA) caused a 25-fold increase in the mRNA for organic solute transporter alpha (OSTalpha), a known FXR target gene. HSD3B2 mRNA levels also increased following CDCA treatment in a concentration-dependent manner. Cells transfected with a HSD3B2 promoter construct and FXR expression vector responded to CDCA with a 20-fold increase in reporter activity compared to control. Analysis of constructs containing sequential deletions of the HSD3B2 promoter suggested a putative regulatory element between -166 and -101. Mutation of an inverted repeat between -137 and -124 completely blocked CDCA/FXR induced reporter activity. Chromatin immunoprecipitation assays further confirmed the presence of a FXR response element in the HSD3B2 promoter. In view of the emerging role of FXR agonists as therapeutic treatment of diabetes and certain liver diseases, the effects of such agonists on other FXR expressing tissues should be considered. Our findings suggest that in human adrenal cells, FXR increases transcription and expression of HSD3B2. Alterations in this enzyme would influence the capacity of the adrenal gland to produce corticosteroids.

Fig.1

FXR transcript levels in human steroidogenic tissues. Quantitative PCR was used to quantify FXR mRNA levels in adult and fetal adrenal, testes, ovary, placenta, and liver as described in Materials and Methods. Data represent the mean ±SEM of at least 5 separate DNase-treated RNA samples and are expressed in attomoles of mRNA per μg of 18S ribosomal RNA. (ND: not detectable)

Fig.2

Immunohistochemical staining for FXR and HSD3B2 in representative adult human adrenal sections. Immunohistochemical localization of FXR (brown) was detected in the nuclei of zona fasciculata and cytoplasm of zona reticularis cells (panel A), and HSD3B2 was expressed in the cytoplasm (red) in human adrenals zona fasciculata (panel B). Negative control, in which primary antibody was not included, was also performed, and no specific immunoreactivity was detected in the tissue section with DAB staining (panel C) or FastRed staining (panel D). Hemotoxin nuclear staining was performed with FastRed staining (panel B and D).

Fig.3

FXR response genes in H295R cells. Panel A. Microarray of the whole transcriptome of FXR transfected H295R cells. H295R adrenocortical cells were transfected with FXR expression vector and treated with/without GW4064 (1 μM) for 24 h, followed by RNA isolation and microarray analysis. Data is shown as a scatter plot that compares basal and GW4060 treatment. Panel B. Microarray data comparing only 42 steroidogenic enzyme genes between basal and GW4064 treated samples. Panel C. FXR effects on HSD3B2 mRNA as determined by qPCR. Results represent the mean ± SD of data from at least three independent experiments, each performed in triplicate. Statistics were calculated by 1-way ANOVA followed by Dunnett’s method, ** P<0.01.

Fig.4

FXR effects on endogenous HSD3B2 and OSTα levels. Panel A. Concentration-dependent effects of FXR on HSD3B2 and OSTα expression levels. H295R adrenocortical cells were transfected with empty vectors or indicated doses of FXR expression vector, and treated with CDCA (10 μM) for 24 h. Quantitative PCR data were normalized to 18S ribosomal RNA and are shown as the fold induction over basal (no ligand treatment). Results represent the mean ± SD of data from at least three independent experiments each performed in triplicate. Statistics were calculated by 1-way ANOVA followed by Dunnett’s method, * P<0.05, compared to empty vector HSD3B2; †† P<0.001, compared to OSTα with empty vector transfection. Panel B. Time course of CDCA effects on FXR-mediated induction of endogenous HSD3B2 and OSTα expression. H295R adrenocortical cells were transfected with FXR and treated with CDCA for the indicated time. Results represent the mean ± SD of data from at least three independent experiments. Statistics were calculated by 1-way ANOVA followed by Dunnett’s method, ** P<0.01, compared to 12 h HSD3B2 level; † P<0.05, compared to 12 h OSTα level.

Fig.5

Effects of FXR on steroidogenic enzyme promoters in H295R adrenocortical cells. Cells were transfected with indicated luciferase promoter constructs (1 μg/well) with or without FXR (0.1 μg/well), and treated with or without FXR ligand CDCA (10 μM). Data were normalized to co-transfected β-gal vector and results are expressed as fold induction over basal promoter. Data represent the mean ± SEM of data from at least three independent experiments performed in triplicate. Statistics were calculated by 1-way ANOVA followed by Dunnett’s method, ** P<0.01. (StAR: Steroidogenic Acute Regulatory protein; CYP11A1: cholesterol side-chain cleavage; CYP17: 17α-hydroxylase)

Fig.6

FXR regulation of HSD3B2 reporter gene activity. Panel A. Concentration-dependent effects of FXR on HSD3B2 reporter gene activity. H295R cells were co-transfected with HSD3B2 promoter constructs (1 μg/well) and indicated amount of the FXR expression plasmid or the empty expression plasmid. Luciferase activity was tested after treatment with/ without CDCA (10 μM) for 24 h. Panel B. Concentration-dependent effects of CDCA on FXR-mediated induction of HSD3B2 reporter gene activity. H295R adrenocortical cells were co-transfected with HSD3B2 reporter construct and FXR expression vector in the presence of increasing amounts of CDCA for 24 h. Data were normalized to co-transfected β-gal vector and data are shown as the fold induction over the basal reporter. Results represent the mean ± SEM of data from at least three independent experiments each performed in triplicate. Statistical significance was analyzed by a paired t-test, * P<0.05, ** P<0.01, *** P<0.001.

Fig.7

A FXRE was necessary for FXR activation of HSD3B2 promoter activity. Panel A. Deletion analysis was used to localize the FXRE within the human HSD3B2 promoter. H295R adrenocortical cells were co-transfected with HSD3B2 reporter constructs (1 μg/well) and FXR expression plasmid (0.1 μg /well) in the presence of CDCA (10 μM) for 24 h. Cells were then lysed and assayed for luciferase activity. Data were normalized to co-transfected β-gal vector and data are shown as fold-induction over the basal pGL3 reporter activity. Results represent the mean ± SD of data from at least three independent experiments, each performed in triplicate. Statistics were calculated by paired t-test, * P<0.05, ** P<0.01. Panel B. Illustration of potential FXRE in HSD3B2 promoter region, and mutated sites in the FXRE and NBRE. Panel C. Effects of CDCA or NGFI-B on mutated HSD3B2 promoter in H295R cells. Cells were transfected with HSD3B2 promoter constructs and FXR expression plasmid with or without CDCA or NGFI-B (0.1 μg/ well). Data were normalized to co-transfected β-gal vector and results are expressed as fold induction over basal promoter. Results were presented as mean ± SD of data from at least three independent experiments performed in triplicate. Statistics were calculated by paired t-test, * P<0.05, ** P<0.01.

Fig.8

Chromatin immunoprecipitation (ChIP) assay demonstration of FXR binding to the HSD3B2 promoter. ChIP was performed as described in the Materials and methods, where indicated cells were treated with CDCA (10 μM) for 24 h. Panel A: Gel electrophoresis of the PCR products amplified from HSD3B2 and OSTα promoter regions. Each is representative of three independent experiments. Panel B: a quantification of fold-change by comparing the signal strength of the samples compared to that of the FXR basal treatment. Results were presented as mean ± SD of data from three independent experiments.

Fig.9

EMSA for FXR binding to HSD3B2 to the FXRE. EMSA was performed using the IRDye 700 tagged oligonucleotide probe corresponding to the putative HSD3B2 FXRE (−137/-124). Samples containing labeled probe alone (FP), probe incubated with FXR over-expressing H295R nuclear extract (NE; 1 μg) or in vitro prepared FXR/RXR (2:1) mix (FXR/RXR; 3 μl) are shown. Non-labeled self-competitor DNA (WT) or non-specific Ad1 oligonucleotide (Ad1) was added to reaction mixtures to define specific protein/ DNA interactions.