doi:10.1016/S0378-1119(00)00518-7 
Copyright © 2001 Elsevier Science B.V. All rights reserved.
Regulation of cholesterol 7α-hydroxylase gene (CYP7A1) transcription by the liver orphan receptor (LXRα)
John Y. L. Chiang
, 
, Rhonda Kimmel and Diane Stroup1, 1
Department of Biochemistry and Molecular Pathology, Northeastern Ohio Universities College of Medicine, P.O. Box 95, Rootstown, OH 44272, USA
Received 31 July 2000;
revised 5 October 2000;
accepted 24 October 2000
Received by J.A. Engler
Available online 9 February 2001.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
The cholesterol 7α-hydroxylase gene (CYP7A1) plays an important role in regulation of bile acid biosynthesis and cholesterol homeostasis. Oxysterol receptor, LXR, stimulates, whereas the bile acid receptor, FXR, inhibits CYP7A1 transcription. The goal of this study was to investigate the role of LXRα on the regulation of rat, human and hamster CYP7A1 transcription in its native promoter and cellular context. Cotransfection with LXRα and RXRα expression plasmids strongly stimulated rat CYP7A1/luciferase reporter activity in HepG2 cells and oxysterol was not required. However, LXRα had much less effect on hamster and no significant effect on human CYP7A1 promoter activity in HepG2 cells. In Chinese hamster ovary cells, cotransfection with LXRα stimulated reporter activity by less than 2-fold and addition of 22(R)-hydroxycholesterol caused a small but significant stimulation of rat, human and hamster CYP7A1 promoter activity. At least two direct repeats of AGGTCA-like sequences with 4-base spacing (DR4) and five-base spacing (DR5), in previously identified bile acid response elements of the rat CYP7A1 were able to bind LXRα/RXRα and confer LXRα stimulation. However, LXRα did not bind to the corresponding sequences of the human gene and bound weakly to hamster and mouse DR4 sequences. Therefore, rats and mice have the unusual capacity to convert cholesterol to bile acids by LXRα-mediated stimulation of CYP7A1 transcription, whereas other species do not respond to cholesterol and develop hypercholesterolemia on a diet high in cholesterol.
Author Keywords: Bile acid synthesis; Reverse cholesterol transport; Cytochrome P450; Nuclear receptors; Gene regulation
Abbreviations: LXR, liver orphan receptor; LXRE, LXR response element; RXR, retinoid X receptor; RAR, retinoic acid receptor; HNF4, hepatocyte nuclear factor 4; PPARα, peroxisome proliferator activated receptorα; CPF, cholesterol 7α-hydroxylase promoter factor; COUP-TFII, chicken ovalbumin upstream promoter transcription factor II; FXR, farnesoid X receptor; DR, direct repeat; CYP7A1, cholesterol 7α-hydroxylase gene; BARE, bile acid response element; EMSA, electrophoretic mobility shift assay; TK, thymidine kinase; Luc, luciferase; nt, nucleotide
Fig. 1. Effect of cotransfection of RXRα and LXRα expression plasmids on CYP7A1/Luc reporter gene activity. (A) Rat reporter plasmid P-416/Luc (2.5 μg) which has 416 bp 5′-upstream sequences of the rat CYP7A1 gene linked to a pGL2 basic plasmid was cotransfected with 0.5 μg RXRα or LXRα or 0.5 μg each of expression plasmids in HepG2 cells. (B) The human (1887bp) and hamster (1607 bp) CYP7A1/Luc reporter genes were linked to pGL3 basic plasmid and were transfected with or without RXRα/LXRα expression plasmids in HepG2 cells. Results presented in this figure are the representation of three independent experiments and each experiment was performed with triplicate assays. Error bar represents the standard error of the mean. The Student's t-test was used to analyze statistic significance between each pair of experiment. *Indicates statistically significant different from the control.
Fig. 2. Effects of oxysterol on rat, human and hamster CYP7A1/Luc reporter gene transcription in HepG2 cells transfected with RXRα/LXRα. The human, hamster and rat CYP7A1/Luc reporter genes were cotransfected with RXRα/LXRα expression plasmids in HepG2 cells. Cells were overlayed with medium containing 22(R)-hydroxycholesterol (10 μM). (A) Rat CYP7A1 reporter plasmid, p-416/Luc, was used in transfection assays. (B) Human CYP7A1 (p-1877/Luc) and hamster (p-1607/Luc) reporter plasmids were used in transfection assays. P-values for 22(R)-hydroxycholesterol (22(R) OHCH) vs. control are 0.197, 0.09, and for human, hamster and rat, respectively.
Fig. 3. Effects of oxysterol on rat, human and hamster CYP7A1/Luc reporter gene transcription in CHO cells. Transfection assays were done in CHO cells using rat (p-416/Luc), human (p-1887/Luc) and hamster (p-1607/Luc) reporter genes without (−) or with cotransfection with RXRα/LXRα expression plasmids. *Indicates statistically significant difference in activity between 22(R)-hydroxycholesterol vs. non-treated control in cells transfected with LXRα (P=0.035, 0.029, and 0.004, respectively, for human, hamster and rat gene). **Indicates statistically significant difference in activities between cells transfected with LXRα and non-transfected control (open bars) (P=0.01, 0.006, and 0.031, respectively, for human, hamster and rat reporter).
Fig. 4. Effect of inhibitors and metabolites of the mevalonate pathway on the reporter activity of the rat CYP7A1/luciferase reporter gene in HepG2 cells. Rat CYP7A1/Luc construct (p-416/Luc) containing 416 bp of the upstream sequence was transfected into HepG2 cells with or without cotransfection with RXRα/LXRα expression plasmids. 22(R)-hydroxycholesterol (10 μM), geranylgeranyl-PP (50 μM), lovastatin (10 μg/ml), or mevalonate (5 mM), was added in HepG2 cell culture medium. Luciferase activities were measured as described under experimental procedures. *Indicates significant difference (P<0.05) from the control.
Fig. 5. Effect of RXR-specific ligand LG100268 on rat p-416/Luc reporter gene activity. P-416/Luc (2.5 μg) was cotransfected with 0.5 μg RXRα or LXRα or 0.5 μg each of expression plasmids in HepG2 cells. LG100268 (0.1 μM) was added in cell culture and harvested 40 h later for assay of luciferase activity. Normalized luciferase activities are expressed as percentage of control, which was transfected with empty plasmid. Results presented in this figure are the representation of three independent experiments and each experiment was performed with triplicate assays. Error bar represents the standard error of the mean. *Indicates significant difference (P<0.05) from the control.
Fig. 6. Mapping of the LXRE in the rat CYP7A1 gene by mutation analysis. Transfection assays were done by cotransfection of RXRα/LXRα with a wild type plasmid p-416/Luc, or a plasmid (p-416ΔBARE-I) containing −416 bp upstream sequence with BARE-I (−74/−54) deleted, or a plasmid (p-416ΔBARE-II/Luc) with BARE-II (−149/−128) deleted, or a plasmid (p-416ΔBARE-I+BARE-II/Luc) with both BARE-I and BARE-II deleted. The P-values of LXRα co-transfection vs. control are 0.006, 0.02, 0.027, and 0.003, for p-416/Luc, p-416/ΔBARE-I/Luc, p-416/ΔBARE-II/Luc, and p-416/ΔBARE-I+II/Luc, respectively. *Indicates statistically significant difference in activities between cells transfected with LXRα and non-transfected control (P<0.05).
Fig. 7. Identification of the LXRE in rat CYP7A1 gene. Cotransfection assays using a control reporter plasmid p-GL2 promoter (SV40/Luc), p2 (BARE-I)/SV40/Luc containing two copies of BARE-I sequence (−74 to −54) of the rat CYP7A1gene linked to pGL2 promoter plasmid (SV40/Luc), or plasmids containing the −344/−240, or −240/−128, or −149/−118 fragments of the rat CYP7A1 upstream sequences linked to pGL2 Promoter reporter plasmid. RXR/LXRα were cotransfected as described under Fig. 1. The P-values are 0.65, 0.0034, 0.09, 0.44, and 0.0058, for pGL2-promoter, p-(2×BARE-I)/SV40Luc, p-344/-240/SV40Luc, p-240/-128/SV40Luc, and p-149/-118/SV40Luc, respectively. *Indicates statistically significant difference in activities between cells transfected with LXRα and non-transfected control (P<0.05).
Fig. 8. EMSA of in vitro synthesized RXRα and LXRα using rat BARE-I (−74/−53) and BARE-II (−149/−118) probes. (A) In vitro synthesized LXRα and/or RXRα were incubated with rat BARE-I (−74/−53) and BARE-II (−149/−118) probes as indicated. Electrophoresis was performed as described in experimental procedures. In vitro synthesized HNF4 was used as a positive control. (B) Probes corresponding to mouse, hamster, and human sequences in BARE-I region were used for EMSA were incubated with in vitro synthesized LXRα/RXRα (L/R) as indicated. Nucleotide sequences of the probes are aligned on the bottom of the figure. Arrows indicate the AGGTCA-like half-site sequences. A consensus DR4 probe was used as a positive control for RXRα/LXRα binding.
Abstract
Purchase PDF (492 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (135)
Purchase PDF (191 K)
