SREBP-2 positively regulates transcription of the cholesterol efflux gene, ABCA1, by generating oxysterol ligands for LXR

Abstract

Cholesterol accumulation and removal are regulated by two different transcription factors. SREBP-2 (sterol-regulatory-element-binding protein-2) is best known to up-regulate genes involved in cholesterol biosynthesis and uptake, whereas LXR (liver X receptor) is best known for up-regulating cholesterol efflux genes. An important cholesterol efflux gene that is regulated by LXR is the ATP-binding cassette transporter, ABCA1 (ATP-binding cassette transporter-A1). We have previously shown that statin treatment down-regulated ABCA1 expression in human macrophages, probably by inhibiting synthesis of the LXR ligand 24(S),25-epoxycholesterol. However, it was subsequently reported that ABCA1 expression is down-regulated by SREBP-2 through binding of SREBP-2 to an E-box element in ABCA1's proximal promoter. As statin treatment induces SREBP-2 activation, this may provide an alternative explanation for the statin-mediated down-regulation of ABCA1. In the present study, we employed a set of CHO (Chinese-hamster ovary) mutant cell lines to investigate the role of SREBP-2 in the regulation of ABCA1. We observed increased ABCA1 mRNA levels in SREBP-2-overexpressing cells and decreased levels in cells lacking a functional SREBP-2 pathway, which were restored when the SREBP-2 pathway was reinstated. Moreover, ABCA1 gene expression was positively associated with synthesis of 24(S),25-epoxycholesterol in these cell lines. In studies using a human ABCA1 promoter reporter assay, mutation of the E-box motif had a similar response as the wild-type construct to either statin treatment or addition of 24(S),25-epoxycholesterol. By contrast, these responses were completely ablated when the DR4 element to which LXR binds was mutated. These results support the idea that 24(S),25-epoxycholesterol and statin treatment influence ABCA1 transcription via supply of an LXR ligand and not through an SREBP-2/E-box-related mechanism. In addition, our results indicate a critical role of SREBP-2 as a positive regulator of ABCA1 gene expression by enabling the generation of oxysterol ligands for LXR.

Figure 1. A functional SREBP pathway is required for ABCA1 gene expression

Various CHO cell lines were incubated in the absence or presence of compactin (CPN; 5 μM) for 24 h. Cell lines were CHO-7 (wild-type), SRD-1 (overexpressing the nuclear form of SREBP-2), M19 (lacking Site-2 Protease), SRD-13A (lacking SCAP) and 13A/pSCAP (an SRD-13A cell line stably expressing pTK3-SCAP). mRNA levels for (A) LDL-R and (B) ABCA1 were measured using QRT–PCR. Data are presented relative to vehicle-treated controls and are means+S.E.M. (three replicate cultures representative of two separate experiments).

Figure 2. SREBP-2 is required for the synthesis of 24(S),25-epoxycholesterol (24,25EC)

Various CHO cell lines and PMA-differentiated THP-1 macrophages were metabolically labelled by incubation with [1-¹⁴C]acetate. Neutral lipid extracts were separated by TLC and bands corresponding to authentic cholesterol and 24(S),25-epoxycholesterol were visualized by a phosphoimager. (A) This phosphoimage is representative of three separate experiments. (B) Values are means from the three separate experiments represented by Figures 1(B), 3(F) and 2A. Key: 1, CHO-7; 2, SRD-1; 3, M19; 4, SRD-13A; 5, 13A/pSCAP. Equation of the line for (B): y=1.32x−0.36; R²=0.88; P<0.02.

Figure 3. Addition of oxysterols and a synthetic LXR ligand restores ABCA1 gene expression in cells lacking SREBP-2

CHO-7 cells (wild-type) or SRD-13A cells (lacking SCAP) were incubated with increasing concentrations of (A, B) 24(S),25-epoxycholesterol or (C, D) cholesterol complexed with methyl-β-cyclodextrin. mRNA levels for (A, C) LDL-R and (B, D) ABCA1 were measured using QRT–PCR. Data are presented relative to vehicle-treated controls and are means+S.E.M. (three replicate cultures representative of two separate experiments). Various CHO cell lines were incubated in the absence or presence of TO901317 (1 μM) for 24 h. Cell lines were CHO-7 (wild-type), SRD-1 (overexpressing the nuclear form of SREBP-2), M19 (lacking Site-2 Protease), SRD-13A (lacking SCAP), and 13A/pSCAP (an SRD-13A cell line stably expressing pTK3-SCAP). mRNA levels for (E) LDL-R and (F) ABCA1 were measured using QRT–PCR. Data are presented relative to vehicle-treated controls and are means+S.E.M. (three replicate cultures representative of two separate experiments).

Figure 4. SREBP-2 is required for ABCA1 promoter activation and the effects are mediated by the LXR response element

CHO-7 cells and SRD-13A cells were transiently transfected for 24 h with phRL-TK Renilla internal control plasmid together with either (A, B) pGL3-hABCA1-wild-type or (C) SRE-luc (CHO-7 cells only). (D) Schematic illustration of the E-box and DR4 elements within the proximal −200 bp promoter region of a human ABCA1 promoter-driven luciferase (luc) reporter. (E) CHO-7 cells were transiently transfected with phRL-TK Renilla internal control plasmid together with pGL3-hABCA1: wild-type, E-box mutant, DR4 mutant, E-box/DR4 mutant. Following transfection, cells were incubated for 24 h in the absence or presence of (A, C, E) compactin (5 μM) or (B) 24(S),25-epoxycholesterol (10 μM) or TO901317 (1 μM). Data are presented relative to the wild-type construct, vehicle-treated control condition, and for (A) are means+S.E.M. (n=5 separate experiments; *P≤0.05 compared with CHO-7 untreated cells); for (C), data are means+S.E.M. (n=3 separate experiments; †P=0.005); for (B, E), data are means+S.E.M. (three replicate cultures representative of two or three separate experiments).

Figure 5. Addition of 24(S),25-epoxycholesterol increases ABCA1 promoter activation via the LXR response element independent of the E-box

CHO-7 cells were transiently transfected for 24 h with phRL-TK Renilla internal control plasmid together with either (A) SRE-luc or (B–D) pGL3-hABCA1: (B) wild-type, (C) E-box mutant, or (D) DR4 mutant. Following transfection, cells were incubated for 24 h in the absence or presence of indicated concentrations of 24(S),25-epoxycholesterol. Data are presented relative to the wild-type construct, vehicle-treated control condition, and for (A, B) are means+S.E.M. (n=3 separate experiments); for (C, D) data are means+half range (n=2 separate experiments).