The orphan nuclear receptor DAX-1 functions as a potent corepressor of the constitutive androstane receptor (NR1I3)

Abstract

Regulation of gene transcription is controlled in part by nuclear receptors that function coordinately with coregulator proteins. The human constitutive androstane receptor (CAR; NR1I3) is expressed primarily in liver and regulates the expression of genes involved in xenobiotic metabolism as well as hormone, energy, and lipid homeostasis. In this report, DAX-1, a nuclear receptor family member with corepressor properties, was identified as a potent CAR regulator. Results of transaction and mutational studies demonstrated that both DAX-1's downstream LXXLL and its PCFQVLP motifs were critical contributors to DAX-1's corepression activities, although two other LXXM/LL motifs located nearer the N terminus had no impact on the CAR functional interaction. Deletion of DAX-1's C-terminal transcription silencing domain restored CAR1 transactivation activity in reporter assays to approximately 90% of control, demonstrating its critical function in mediating the CAR repression activities. Furthermore, results obtained from mammalian two-hybrid experiments assessing various domain configurations of the respective receptors showed that full-length DAX-1 inhibited the CAR-SRC1 interaction by approximately 50%, whereas the same interaction was restored to 90% of control when the DAX-1 transcription silencing domain was deleted. Direct interaction between CAR and DAX-1 was demonstrated with both alpha-screen and coimmunoprecipitation experiments, and this interaction was enhanced in the presence of the CAR activator 6-(4-chlorophenyl)imidazo[2,1-b]thiazole-5-carbaldehyde O-(3,4-dichlorobenzyl)oxime (CITCO). Results obtained in primary human hepatocytes further demonstrated DAX-1 inhibition of CAR-mediated CITCO induction of the CYP2B6 target gene. The results of this investigation identify DAX-1 as a novel and potent CAR corepressor and suggest that DAX-1 functions as a coordinate hepatic regulator of CAR's biological function.

Fig. 1.

DAX-1 protein sequence showing the locations of the three LXXL/ML NR boxes (bold), the PCFQVL NR interacting sequence (bold and underlined), and the TSD (underlined).

Fig. 2.

Effect of DAX-1 or SHP on CAR1 activity. Results shown here represent single transfection experiments, with all of the treatments in quadruplicate. COS-1 cells were transfected with the CMV2-CAR1 and 3.1-RXRα expression vectors, the 2B6-XREM-PBREM reporter, the pRL-CMV vector for normalization of transfection efficiency, and varying amounts of PCMV6-DAX1 (A) or pCMV6-SHP (B). All of the treatments were for 24 h, and the data are represented as normalized luciferase values. Each data point represents the mean (±S.D.). Asterisks indicate significant difference from respective control (no DAX-1 or SHP) within a chemical treatment group (p < 0.01).

Fig. 3.

Effect of DAX-1 mutations on CAR variant activity. Results shown here represent single transfection experiments, with all of the treatments in quadruplicate. COS-1 cells were transfected with the 3.1-RXRα expression vector, the 2B6-XREM-PBREM reporter, the pRL-CMV vector for normalization of transfection efficiency, and varying amounts of CMV2-CAR1, -CAR2, or -CAR3 and PCMV6-empty, -DAX1 WT, -DAX1 M1, -DAX1 M2, -DAX1 M3, or -DAX1 PCF (A) or PCMV6-DAX1-TSD (B). All of the treatments were for 24 h, data are represented as normalized luciferase values, and each data point represents the mean (±S.D.). Asterisks indicate significant difference from respective control (no DAX1) within a chemical treatment group (p < 0.05).

Fig. 4.

A, mammalian two-hybrid assays to assess CAR1 interaction with DAX-1. COS- cells were cotransfected with pmGAL4-CAR1-LBD and VP16-empty (negative control), VP16-SRC1 (positive control), VP16-DAX1, or VP16-DAX1 TSD, 3.1 RXRα-LBD, pFR-luciferase reporter, and pRL-CMV vector for transfection efficiency normalization (left side). On the right half of the graph, CAR1 was in the VP16 vector, and the test constructs were in the pmGAL4 vector. B, mammalian two-hybrid assay to assess the effect of DAX-1 on CAR1-SRC1 interaction. COS-1 cells were cotransfected with pmGAL4-CAR1 and VP16-SRC1 along with either VP16-empty, -DAX1, or -DAX1-TSD. The 3.1 RXRα-LBD, pFR-luciferase reporter, and pRL-CMV vectors were included in all of the transfections. On the right half of the graph, VP16-CAR1-LBD and pmGAL4-SRC1 were tested in the presence of pmGAL4-empty, -DAX1, or -DAX1-TSD. Both panels are representative single transfection experiments, with all of the treatments in quadruplicate, and each data point represents the mean (±S.D.). Asterisks indicate that each treatment was significantly different from its respective empty vector control (p < 0.05).

Fig. 5.

Direct interaction between CAR1 and DAX-1. AlphaScreen assay for CAR1 interaction with DAX-1 (A). Purified CAR1-LBB and DAX-1-GST (∼40 nM each) were incubated in the presence of 5 μg/ml donor and acceptor beads. The DAX-1a construct contains amino acids 210 to 470 and DAX-1b contains amino acids 218 to 470. Bars represent the mean (±S.D.) of two separate experiments, each performed in duplicate. Representative coimmunoprecipitation of CAR1 and DAX1 (B). COS-1 cells were cotransfected with pCMV6-DAX1 and p3XFLAG CAR or 3XFLAFG empty. After ∼40 h, the cells were treated with DMSO, CITCO, ANDRO, or PB for 5 to 6 h and then harvested. Cell lysates were immunoprecipitated with anti-FLAG beads, and precipitated protein was subjected to Western blot analysis with anti-DAX1 antibody, followed by anti-FLAG antibody. C, densitometry analysis of the Western blot data.

Fig. 6.

Effect of DAX-1 on CAR-mediated induction of CYP2B6 by CITCO in primary human hepatocytes. Human primary hepatocytes were transfected with DAX-1, empty vector and then treated with CITCO or DMSO. Real-time polymerase chain reaction was performed on cDNA samples isolated from the cells to quantify DAX-1, CYP2B6, and GAPDH (as an internal reference). Values were expressed as fold induction relative to untransfected or empty vector-transfected, DMSO-treated cells and represent the average of two replicate samples. Percentages indicate the decrease in CYP2B6 induction relative to untransfected or empty vector-transfected, CITCO-treated cells.