Regulation of the human hydroxysteroid sulfotransferase (SULT2A1) by RORα and RORγ and its potential relevance to human liver diseases

Abstract

The retinoid-related orphan receptors (RORs) were postulated to have functions in tissue development and circadian rhythm. In this study, we revealed a novel function of RORα (NR1F1) and RORγ (NR1F3) in regulating the human hydroxysteroid sulfotransferase (SULT2A1), a phase II conjugating enzyme known to sulfonate bile acids, hydroxysteroid dehydroepiandrosterone, and related androgens. A combination of promoter reporter gene assay and EMSA and chromatin immunoprecipitation (ChIP) assays showed that both RORα and RORγ transactivated the SULT2A1 gene promoter through their binding to a ROR response element found in the SULT2A1 gene promoter. Interestingly, this ROR response element overlaps with a previously reported constitutive androstane receptor response element on the same promoter. Down-regulation of RORα and/or RORγ by small interfering RNA inhibited the expression of endogenous SULT2A1. In primary human hepatocytes and human livers, we found a positive correlation between the expression of SULT2A1 and RORs, which further supported the regulation of SULT2A1 by RORs. We also found that the expression of RORα and RORγ was impaired in several liver disease conditions, such as steatosis/steatohepatitis, fibrosis, and hepatocellular carcinoma. The positive regulation of human SULT2A1 by RORs is opposite to the negative regulation of Sult2a1 by RORs in rodents. In summary, our results established SULT2A1 as a novel ROR target gene. The expression of RORs is a potential predictor for the expression of SULT2A1 as well as disease conditions.

Fig. 1.

RORα and RORγ activated the hSULT2A1 gene promoter and induced the expression and activity of endogenous SULT2A1 and down-regulation of RORα/γ inhibited the expression of endogenous SULT2A1. A, HepG2 cells were transiently transfected with pGL-SULT2A1 (500 bp) reporter gene, together with the empty vector or indicated receptors, and the transfection efficiency control pCMX-β-gal. Transfected cells were treated with vehicle or TCPOBOP (5 μm) for 24 h and then harvested and assayed for luciferase and β-gal activities. The transfection efficiency was normalized against the β-gal activity. B, The inductions of SULT2A1 in ROR-transfected HepG2 cells and further inductions upon cholesterol sulfate (C.S.) treatment (10 μm, overnight) were confirmed by enzymatic assay using DHEA as the substrate. C, The C.S.-responsive inductions of SULT2A1 mRNA in three cases of primary human hepatocytes was confirmed by real-time PCR. The three cases included a 61-yr-old male, a 68-yr-old male, and a 91-yr-old female, all Caucasians. D, HepG2 cells were transfected with control scrambled siRNA (siControl) or RORα and RORγ siRNA. Forty-eight hours after the transfection, cells were harvested and subjected to real-time PCR analysis to detect the expression of the endogenous SULT2A1. TC, TCPOBOP. *, P < 0.05 (n = 3 for each group).

Fig. 2.

Identification and characterization of a RORE in the SULT2A1 gene promoter that overlaps the CARE. A, Sequence of the shared binding sites with the CARE and RORE (boxed). Mutations intended to disrupt individual or combined CAR and ROR bindings are listed with mutated nucleotides (underlined). B, Binding of in vitro synthesized RORα, RORγ, and CAR to ³²P-labeled wild-type or mutant DNA was demonstrated by EMSA. The relative densities of the bands within the same probe are labeled. C, The binding of RORα and RORγ to the wild-type DNA was efficiently competed by unlabeled cold probes of self or ApoA-V/RORE but not by the RORE mutant. D, Competition of DNA binding by RORs and CAR. The protein ratios of receptors are indicated.

Fig. 3.

Recruitment of RORα and RORγ to the SULT2A1 gene promoter as shown by a ChIP assay. A and B, Formaldehyde cross-linked DNA was extracted from Huh-7 cells without (A) or with (B and C) indicated receptor transfections using the indicated antibodies. The final DNA extracts were amplified by PCR using the primer pairs encompassing SULT2A1/RORE. The recruitment of the receptor to a distal promoter in A and ApoA-V/RORE in B was included as negative and positive controls, respectively. C, The recruitment of transfected HA-CAR onto the SULT2A1 gene promoter as shown by a ChIP assay. The recruitment of CAR to CYP2B6/CARE was included as a positive control.

Fig. 4.

Functional characterization of RORE and its overlap with CARE using reporter gene assays. HepG2 cells were transiently transfected with the synthetic promoter reporter tk-WT (A) or the natural promoter reporter (pGL-SULT2A1) (B) or their respective mutant variants (A a-d and B a-c, respectively) together with the empty vector or indicated receptors. When applicable, transfected cells were treated with vehicle or TCPOBOP (5 μm) for 24 h before cell harvesting and assays for luciferase and β-gal activities. The transfection efficiency was normalized against the β-gal activity. All data in B are expressed relative to the basal WT promoter activity from Fig. 1A that was arbitrarily defined as 1. TC, TCPOBOP. *, P < 0.05 (n =3 for each group).

Fig. 5.

The expression of SULT2A1 was positively correlated with the expression of RORα and RORγ in primary human hepatocytes and human livers. Twenty-one cases of primary human hepatocyte samples (A) and 27 cases of human livers (B) were analyzed for their mRNA expression of SULT2A1, RORα, RORγ, and CAR by real-time PCR analysis. Diamonds represent individual patients. The correlations were analyzed by linear regression.

Fig. 6.

The expression of RORα, RORγ, CAR, and SULT2A1 was affected by liver diseases. The 27 cases of human livers were divided into normal and three disease groups, and the relative expression of RORα (A), RORγ (B), CAR (C), and SULT2A1 (D) in disease groups was compared with those of the normal group. *, P < 0.05. NS, Statistically not significant.