The constitutive androstane receptor (CAR) is a ligand/activator-dependent transactivation factor that resides in the cytoplasm and forms part of an as yet unidentified protein complex. Upon stimulation, CAR translocates into the nucleus where it modulates the transactivation of target genes. However, CAR exogenously expressed in rat liver RL-34 cells is located in the nucleus even in the absence of activators. By transiently transfecting RL-34 cells with various mutated rat CAR segments, we identified two nuclear localization signals: a basic amino acid-rich sequence (RRARQARRR) between amino acids 100 and 108; and an assembly of noncontiguous residues widely spread over amino acid residues 111 to 320 within the ligand binding domain. A C-terminal leucine-rich segment corresponding to a previously reported murine xenochemical response signal was not found to exhibit nuclear import activity in cultured cells. Using rat primary hepatocytes transfected with various CAR segments, we identified the region required for the cytoplasmic retention of CAR. Based on these results, the intracellular localization of CAR would be determined by the combined effects of nuclear localization signals, the xenochemical response signal, and the cytoplasmic retention region.

No abstract

Fluorescence recovery after photobleaching (FRAP) in spontaneous multinuclear cells shows that both rat and human constitutive active/androstane receptors (CARs) are shuttling proteins with both nuclear localization signals (NLSs) and nuclear export signals (NESs). We previously identified two NLSs in rat CAR: NLS1 in the hinge region (residues 100-108) and NLS2 in the ligand-binding domain (residues 111-320). In the present study, we compared the intracellular localization signals between rat and human CARs. There was a marked difference in their intracellular localization in COS-7 cells because, unlike rat CAR, human CAR does not contain NLS1 due to an amino acid change at position 106. A CRM1-dependent leucine-rich NES, which is sensitive to an inhibitory effect of leptomycin B, was found in the cytoplasmic retention region previously identified within the ligand-binding domain of rat CAR (residues 220-258). We found that human CAR instead has a NES in the ligand-binding domain between residues 170 and 220. Also, we detected CRM1-independent C-terminal NESs between residues 317-358 of rat and human CARs. Removal of NLS1 by N-terminal truncation and mutation of xenochemical response signal caused rat CAR to localize in the cytoplasm of COS-7 cells, which we suspect is due to the masking of NLS2.

Vitellogenin (VTG) is a useful biomarker for detecting the estrogenic activity of chemicals in aquatic environments. However, little information is available on the regulatory mechanisms of the expression of each VTG subtype, particularly the relationship between expression patterns of VTG1/2 and estrogen receptor (ER) subtypes, such as ERα and ERβ. In this paper, we measured VTG1 and VTG2 mRNA induction in male medaka liver, which was treated with ERα-selective ligand, (17α, 20E)-3-hydroxy-17,20-[(1-methoxyethylidene)bis(oxy)]-19-norpregna-1,3,5(10),20-tetraene-21-carboxylic acid, methyl ester or ERβ-selective ligand, 2-(4-hydroxyphenyl)-5-hydroxy-1,3-benzoxazole and investigated the characteristics of ER subtype function in VTG1 and VTG2 inductions. Hepatic VTG1 mRNA was induced by ERα-selective ligands at even low concentration and maximum increases were the same as for E2. VTG2 mRNA was also increased, but its levels were very low. On the other hand, ERβ-selective ligands significantly increased VTG2 mRNA in the presence of ERα agonists. These results indicate that the expression of each VTG subtype is regulated by unique ER subtypes. VTG1 expression is only regulated by the action of ERα. In contrast, VTG2 expression is regulated by both ERα and ERβ, with ERα being essential for VTG2 gene expression and ERβ being essential for enhancement.

No abstract