Human constitutive androstane receptor mediated methotrexate induction of human dehydroepiandrosterone sulfotransferase (hSULT2A1)

Abstract

Sulfotransferases (SULTs) catalyzed sulfation is important in the regulation of biological activities of hormones and neurotransmitters, the metabolism of drugs, and the detoxification of xenobiotic toxicants. Sulfation also leads to the bioactivation of procarcinogens. Human dehydroepiandrosterone sulfotransferase (hSULT2A1) is a major SULT catalyzing the sulfation of hydroxysteroids and xenobiotic alcohols. Our previous studies had shown that the anti-folate drug methotrexate (MTX) can up-regulate several major isoforms of human SULTs. To determine the mechanisms controlling the regulation of hSULT2A1, the 5'-flanking region of hSULT2A1 was constructed into the pGL3-Basic luciferase reporter vector. The transcriptional regulation mechanism of hSULT2A1 promoter was studied using Caco-2 cell line based on the reporter gene assay. Nuclear receptor co-transfection results indicated that human constitutive androstane receptor (hCAR) and human retinoid X receptor alpha (hRXRalpha) were involved in the transcriptional regulation of hSULT2A1. RNA interference experiments further proved the role of hCAR in hSULT2A1 regulation. Progressive promoter deletion, DNA sequence alignment, and site directed promoter mutation results suggested that an imperfect inverted repeat DNA motif, IR2 (-186AGCTCAGATGACCC-173), within the hSULT2A1 promoter region mediated the hSULT2A1 induction by MTX. Furthermore, electrophoretic mobility shift assay and super shift assay were employed to characterize the interactions of hCAR and hRXRalpha with the IR2 element. In summary, we identified an IR2 DNA cis-element located at -186/-173 of hSULT2A1 promoter region; the IR2 element mediates the MTX induction of hSULT2A1 through interacting with hCAR and hRXRalpha.

Figure 1. Effect of hCAR and hRXRα on CITCO and MTX induction of hSULT2A1 promoter activity in Caco-2 cells

Luciferase constructs containing 5′-flanking sequences of hSULT2A1 were co-transfected with expression vector of hCAR, hRXRα and treated with MTX, CITCO, retinoic acid or alcohol (control) as described in the methods section. pRL-TK was used as internal control for transfection assay. Fold inductions were calculated relative to the promoter activity in vehicle-treated cells. Concentrations used: hCAR (25 ng), hRXRα (25 ng), MTX (100 nM), CITCO (100 nM) and 9-cis-RA (1 μM). Dual-luciferase activities were measured according to the manufacturer’s recommendations. The histograms with standard deviations are average values from three independent transfections; each independent transfection was performed in duplicate. One way ANOVA/Dunnett’s test was used in the data analysis, * was used when all treatment was compared with control and # was used when treatments were compared with MTX + CAR group. *^/#, p < 0.05; and **^/##, p < 0.01.

Figure 2. Effect of RNA interference of hCAR on hSULT2A1 transcriptional activity

Figure 2A, the luciferase reporter vector regulated by hSULT2A1 promoter was transfected into Caco-2 cells and dual luciferase activity was measured according to manufacture’s recommendation. Negative control siRNA (NC siRNA) or hCAR specific siRNA (CAR siRNA) was transfected into Caco-2 cells with (column 4) or without (column 2) the co-transfection of hCAR expression vector. The control cells for reporter gene assay received only the reporter vectors. For figure 2B and figure 2C, Caco-2 cells were transfected with hCAR expression vector, and then negative control siRNA or hCAR specific siRNA were transfected after 6 hours. Cells were harvested after 48 hours and analyzed for real-time PCR with gene specific primers for human β-actin, hCAR and hSULT2A1 as described under Methods section. Relative concentrations were calculated using standard curves method. Each treatment group was analyzed in triplicate and the data shown were average of three independent experiments. One way ANOVA/Dunnett’s test was used in the data analysis, * was used when all treatment was compared with control and # was used when CAR+CAR siRNA group was compared with CAR group. *^/#, p < 0.05; and **^/##, p < 0.01.

Figure 3. Deletion analysis of hSULT2A1 promoter

Figure 3A, schematic diagram of deleted hSULT2A1 promoter constructs. Figure 3B–3I, transactivation of deleted promoter constructs by hCAR and MTX. The luciferase reporter vector was regulated by hSULT2A1 5′-flanking region from the number indicated in the figure to +48. 25 ng hCAR was used for transfection. 100 nM MTX was used for treatment. Promoter activity is expressed as normalized luciferase activity. The histograms with standard deviations are average values from three independent transfections; each independent transfection was performed in duplicate. ANOVA/Dunnett’s test was used in the data analysis, *, p < 0.05; and **, p < 0.01; compared with control.

Figure 4. Sequence alignments of rat, mouse, and human SULT2A1 promoter regions

MultAlin program was used for the alignment of rat (GenBank accession no. M29301), mouse (Echchgadda et al., 2004a), and human (GenBank accession no. U54701) SULT2A1 promoter regions.

Figure 5. Point mutation analysis of the IR2 DNA cis-element

Mutated 5′-flanking region of hSULT2A1 (−414 to +48) as shown in the figure was inserted into pGL3-Basic vector. Luciferase expression regulated by this mutated promoter was determined in Caco-2 cells. The histograms are average values from three independent experiments (each performed in duplicate). ANOVA/Dunnett’s test was used in the data analysis, *, p < 0.05 compared with control.

Figure 6. EMSA of Caco-2 nuclear extract and super shift assay of hCAR and hRXRα with the IR2-containing oligonucleotide (31 bp)

Figure 6A, gel shift assay of Caco-2 nuclear extract with the IR2-containing oligonucleotide. The labeled IR2 wild type oligonucleotide (lane 1 – 4) or labeled mutated IR2 oligonucleotide (lane 5) was incubated with Caco-2 nuclear extract (lane 2–5, lane 1 for control without nuclear extract) for 20 minutes at room temperature to form possible DNA protein complexes. 125-fold excess of wild type cold competitors (lane 3) and mutated cold competitors (lane 4) was added to test the specificity of the IR2. Figure 6B, super shift assay of hCAR and hRXR. Antibody specific to hCAR (lane 6) and hRXRα (lane 7) was added to the nuclear extract and incubated for 20 minutes at room temperature, then followed with the shift reaction.