Identification of two functional nuclear localization signals mediating nuclear import of liver receptor homologue-1

Abstract

Liver receptor homologue-1 (LRH-1) is a member of the nuclear receptor superfamily. We characterized two functional nuclear localization signals (NLSs) in LRH-1. NLS1 (residues 117-168) overlaps the second zinc finger in the DNA binding domain. Mutagenesis showed that the zinc finger structure and two basic clusters on either side of the zinc finger loop are critical for nuclear import of NLS1. NLS2 (residues 169-204) is located in the Ftz-F1 box that contains a bipartite signal. In full-length LRH-1, mutation of either NLS1 or NLS2 had no effect on nuclear localization, but disruption of both NLS1 and NLS2 resulted in the cytoplasmic accumulation of LRH-1. Either NLS1 or NLS2 alone was sufficient to target LRH-1 to the nucleus. Both NLS1 and NLS2 mediate nuclear transport by a mechanism involving importin α/β. Finally, we showed that three crucial basic clusters in the NLSs are involved in the DNA binding and transcriptional activities of LRH-1.

Fig. 1

Identification of two nuclear localization domains in LRH-1. a Schematic representation of LRH-1 deletion fragments fused to C-terminal EGFP tag. The full-length LRH-1 contains the A/B domain, DNA binding domain (DBD), hinge region and ligand binding domain (LBD). Two putative NLSs are indicated above. The subcellular localization of LRH-1 mutants is summarized on the right (N nuclear, C cytoplasmic). b GFP-LRH-1 wild-type or deletion mutants were transfected into COS-7 cells. After 24 h, cells were fixed and nuclei counterstained with DAPI. Fusion proteins were visualized by confocal microscopy. Scale bar 25 μm

Fig. 2

Detection of key basic residues in NLS1 for nuclear import. a Sequence of the putative NLS1 is shown. The full-fragment (residues 117–168) and three segments within this region (residues 126–142, 137–160 and 151–169) were fused to EGFP. The basic amino acids lysine and arginine (in boldface) in the 117–168 region were substituted by alanines as indicated by arrows. The subcellular localization of LRH-1 mutants is summarized on the right (N nuclear, C cytoplasmic). b Wild-type and mutant forms were transfected into COS-7 cells. After 24 h, cells were fixed and the nuclei counterstained with DAPI. Fusion proteins were visualized by confocal microscopy. Scale bar 25 μm

Fig. 3

Detection of key basic residues in NLS2 for nuclear import. a Sequence of the putative NLS2 is shown. Two segments of either residues 169–193 or residues 169–204 were fused to EGFP. The basic amino acids lysine and arginine (in boldface) in residues 169–204 were substituted by alanines as indicated by arrows. The subcellular localization of LRH-1 mutants is summarized on the right (N nuclear, C cytoplasmic). b Wild-type and mutant forms were transfected into COS-7 cells. After 24 h, cells were fixed and the nuclei counterstained with DAPI. Fusion proteins were visualized by confocal microscopy. Scale bar 25 μm

Fig. 4

Simultaneous disruption of NLS1 and NLS2 abolishes the nuclear localization of LRH-1. a Basic amino acids within either or both NLSs were mutated to alanines in the full-length LRH-1 tagged with Myc. These constructs were transfected into COS-7 cells. After 24 h, cells were fixed and immunostained with anti-Myc antibody. The nuclei were counterstained with DAPI. Fusion proteins were visualized by confocal microscopy. Wild-type (a) and single cluster mutants in NLS1 (b, c) or NLS2 (d, e) are exclusively localized to the nucleus. The subcellular localization of double cluster mutants in representative cells is shown in f–i. Scale bar 25 μm. b Statistics of subcellular distribution of wild-type (a in a) and double cluster mutants (f–i in a) are summarized. A minimum of 100 cells were counted for each experiment and the means ± SEM three independent experiments are presented

Fig. 5

Nuclear import mechanisms of LRH-1 NLS. a Residues 117–168 (NLS1) or 169–204 (NLS2) of LRH-1 were fused to GFP tag. The classical NLS of SV40 fused to GFP was used as a positive control. The N-terminal residues 1–76 fused to GFP and GFP alone were used as negative controls. [³⁵S]-labelled GFP or fusion proteins were synthesized in vitro and subjected to GST pull-down assays using GST, GST-importin α or GST-importin β. The molecular mass in kilodaltons is indicated on the right-hand side. b The GFP-NLS fusion constructs were transfected into COS-7 cells. After 24 h, cells were incubated in energy-depletion medium for 30 or 60 min, fixed and then counterstained with the nuclear dye DAPI. Fusion proteins were visualized by confocal microscopy. Scale bar 25 μm. c Statistics of subcellular distribution of GFP-NLS fusion proteins in b are summarized. A minimum of 100 cells were counted for each experiment and the means ± SEM of three independent experiments are presented

Fig. 6

DNA binding and transcriptional activity of wild-type and mutant LRH-1. a Myc-tagged LRH-1 was synthesized in vitro. EMSA was performed using a [³²P]-labelled probe containing a LRH-1 binding site in the absence (lane 2) or presence of a 20-fold molar excess of unlabelled wild-type (lane 3) or mutant (lane 4) oligonucleotide or anti-Myc antibody (lane 5). Migration of DNA–protein complexes is indicated by the arrow. b Myc-tagged wild-type (WT) and mutated LRH-1 were synthesized in vitro and subjected to EMSA using a [³²P]-labelled probe containing a LRH-1 binding site. Densitometric quantification of EMSA data are presented as the values relative to the wild-type in c (open boxes). c The CYP11A1 promoter–luciferase construct was cotransfected in COS-7 cells with an empty expression vector (−) or expression vectors for wild-type (WT) or mutated LRH-1 as indicated. Cells were lysed and assayed for luciferase activity after transfection for 24 h. The relative luciferase activity was compared to that with the wild-type construct. The data presented are the means ± SEM (closed boxes) of six independent experiments; *p < 0.05 compared to wild-type. d Expression of Myc-LRH-1 proteins in COS-7 cells was confirmed by immunoblotting with an anti-Myc antibody

Fig. 7

Comparison with other nuclear receptors and the structure of LRH-1 NLS domain. a Sequence alignment of the zinc-finger domain and the adjacent region of several nuclear receptors. The cysteine residues that coordinate zinc binding are highlighted by grey boxes. The nuclear localization domains NLS1 and NLS2 of LRH-1 are indicated. The conserved basic amino acids that were mutated in this study are highlighted by dark boxes. The key basic amino acids identified for nuclear import of NLS1 and NLS2 are indicated by asterisks. b Diagram of the zinc binding loops of LRH-1. The unique Ftz-F1 box for NR5A subfamily members is indicated