Evidence of the neuron-restrictive silencer factor (NRSF) interaction with Sp3 and its synergic repression to the mu opioid receptor (MOR) gene

Abstract

Previously, we reported that the neuron-restrictive silencer element (NRSE) of mu opioid receptor (MOR) functions as a critical regulator to repress the MOR transcription in specific neuronal cells, depending on neuron-restriction silence factor (NRSF) expression levels [C.S.Kim, C.K.Hwang, H.S.Choi, K.Y.Song, P.Y.Law, L.N. Wei and H.H.Loh (2004) J. Biol. Chem., 279, 46464-46473]. Herein, we identify a conserved GC sequence next to NRSE region in the mouse MOR gene. The inhibition of Sp family factors binding to this GC box by mithramycin A led to a significant increase in the endogenous MOR transcription. In the co-immunoprecipitation experiment, NRSF interacted with the full-length Sp3 factor, but not with Sp1 or two short Sp3 isoforms. The sequence specific and functional binding by Sp3 at this GC box was confirmed by in vitro gel-shift assays using either in vitro translated proteins or nuclear extract, and by in vivo chromatin immunoprecipitation assays. Transient transfection assays showed that Sp3-binding site of the MOR gene is a functionally synergic repressor element with NRSE in NS20Y cells, but not in the NRSF negative PC12 cells. The results suggest that the synergic interaction between NRSF and Sp3 is required to negatively regulate MOR gene transcription and that transcription of MOR gene would be governed by the context of available transcription factors rather than by a master regulator.

Figure 1

Mithramycin A activates expression of the endogenous MOR gene. (A) The putative Sp family binding site and NRSE sequence among three species, mouse, human and rat is located from −9 to +20 in the MOR gene and have highly homologous (^*) sequence to the consensus Sp family binding site and NRSE DNA element. (B) NS20Y cells (NRSF positive cells) were pre-incubated with mithramycin A (5, 10, 20, 50 and 100 nM) for 24 h. Total RNAs were prepared and analyzed by RT–PCR for the MOR gene expression using specific primer pairs as described in Materials and Methods. Expression of the β-actin gene that is insensitive to mithramycin A was used as a negative control. (C) PC12 cell (NRSF negative cells) was pre-incubated with mithramycin A (10, 50, 100 nM) for 24 h. (D) Quantification of transcripts was performed with real-time quantitative PCR (RT-qPCR). NS20Y cells were treated either DMSO or 50 nM of mithramycin A for 24 h. Total RNA was prepared and treated with DNase I, and primer pairs specific for the coding sequence of each gene were used for RT-qPCR with the dye SYBR Green. The mRNA levels from DMSO or mithramycin A treated cells were normalized to β-actin levels. Error bars indicate the range of SE and threshold cycle (C_t) values were obtained from triplicate data points and changes in transcript levels for mithramycin A treated samples were compared with DMSO-treated samples, which were assigned a value of 1.

Figure 2

The NRSE- and Sp-binding site additively repress the MOR promoter activity in NS20Y but not in PC12 cells. (A) Schematic diagram representing the constructs of the mouse MOR promoter/luciferase from nucleotide −4.7 kb to +20 bp, containing the translation start site (ATG), which is designated +1 and showing NRSE and putative Sp-binding sequences. (B) NS20Y and PC12 cells individually were transfected with the pGL4.7NRmSPm (mutated the both of NRSE and Sp-binding sites), pGL4.7NRSP (wild-type promoter), pGL4.7NRmSP (mutated the NRSE site) and pGL4.7NRSPm (mutated the Sp-binding site) reporter constructs. The promoter activity of each construct was expressed as relative luciferase activity, and transfection efficiencies were normalized to β-galactosidase activity by co-transfection of the internal control plasmid pCH110. Error bars indicate the range of SE, and the activities of the luciferase reporter were expressed as n-fold relative to the activity of each corresponding luciferase reporter with vector-alone transfection, which was assigned an activity value of 1.0. (C) Analysis of NRSF expression in NS20Y and PC12 cells by western blotting.

Figure 3

Mithramycin A enhances the MOR promoter activity. (A) Schematic representation of the promoter constructs pGLNRSP4.7 (wild type) and pGLNRSPm (mutated the Sp-binding site). (B) NS20Y cells were transfected with the pGL4.7NRSP and pGL4.7NRSPm luciferase reporter gene constructs and treated with DMSO (as a vechicle) and mithramycin A (100 and 400 nM) for 24 h. The relative luciferase activity of each construct was expressed as the n-fold increase over the value for the DMSO-treated NS20Y cells. Data were normalized by protein concentration and expressed as the n-fold activation of the luciferase activity of DMSO-treated activity, which is arbitrarily defined as 1.0. Error bars indicate the range of SE. In addition, the n-fold increase in promoter activity by mithramycin A compared with the corresponding DMSO-treated control is shown on the top of the error bars.

Figure 4

NRSF interacts with Sp3 factor but not with Sp1 or Sp3 isoforms. (A) Protein was harvested and immunoprecipitated with NRSF and Sp1 or Sp3 antibodies. Immune complexes were subjected to western blot analysis with Sp1 or Sp3 and NRSF antibodies after washing with RIPA buffer (high stringency, HS) or PBS + 0.5% NP-40 (low stringency, LS). (B and C) NS20Y cells were transiently transfected with myc-tagged NRSF expression vector. Protein extract from NS20Y cell was precipitated with anti-c-myc, anti-Sp1 or anti-Sp3 antibodies, and the immune complexes were subjected to western blot with anti-Sp3, anti-Sp1 or anti-c-Myc antibodies.

Figure 5

Binding of NRSF and Sp3 factor to the negative regulatory region (NRSE/GC box) of MOR promoter. A supershift assay was performed using in vitro translated myc-tagged NRSF (A) or Sp3 (B) or both (C). EMSAs were performed as indicated in Materials and Methods. (A) Lane 1, probe alone; lane 2, 500-fold molar excess of unlabeled self-competitor probe; lane 3, probe and in vitro translated myc-tagged NRSF; lane 4, c-myc antibody; lane 5, pre-immune serum (PI); lane 6, non-specific antibody (IRF-4). (B) Lane 1, probe alone and reticulocyte (RBC); lane 2, in vitro translated Sp3; lane 3, PI; lane 4, 500-fold molar excess of unlabeled self-competitor probe; lane 5, Sp3 antibody; lane 6, non-specific antibody (IRF-4). All lanes contained 4 μl of in vitro translated NRSF and Sp3 products from a 50 μl reaction volume (Promega's in vitro translation kit). (C) Lane 1, probe alone; lane 2, probe and reticulocyte (RBC); lane 3, probe and in vitro translated myc-tagged NRSF; lane 4, c-myc antibody; lanes 5–9; probe and in vitro translated myc-tagged NRSF with increasing amount of Sp3 (0.1–2 μg).

Figure 6

NRSF and Sp3 from NS20Y cell extract bind NRSE/GC box element of MOR promoter. A supershift assay was performed with nuclear extracts from NS20Y cells. The double-stranded ³²P-labeled NRSE/GC box was used as a probe. Lane 1, probe alone; lanes 2–8, probe plus 10 μg of NS20Y cell nuclear extracts, lane 3, 100-fold molar excess of unlabeled self-competitor; lane 4, PI; lane 5, mouse monoclonal NRSF antibody; lane 6, Sp3 antibody; lane 7, 100-fold molar excess of NRmSP competitor; and lane 8, 100-fold molar excess of NRSPm competitor.

Figure 7

Functional binding of NRSF, Sp3 and HDAC2 at NRSE/GC box at MOR promoter in vivo. (A) Chromatin immunoprecipitation analysis of NRSF, Sp3 and HDAC2 binding at the MOR promoter. ChIPs with PI (sham), anti-IRF4 (sham), anti-NRSF, anti-HDAC1, anti-HDAC2, anti-Sp1 and anti-Sp3 performed on NS20Y chromatin were amplified with primers flanking the Sp-binding site of the MOR gene. Immunoprecipitated DNA was analyzed by PCR using primers that amplified a 188 bp region of the MOR promoter, which encompasses a Sp3-binding site. Lane 1, 100 bp DNA marker; lane 2, input; lane 3, without antibody; lane 4, PI; lane 5, IRF-4 antibody; lane 6, NRSF antibody; lane 7, HDAC1 antibody; lane 8, HDAC2 antibody; lane 9, Sp1 antibody; lane 10, Sp3 antibody. (B) Quantitative data for ChIPs were analyzed by using ImageQuant 5.2 software. Bars depict the sum of signal intensity in the same size area and SD between experiments. DNA fragments for input samples were compared with antibody-treated samples, which were assigned a value of 1. (C) Re-ChIP was performed using anti-NRSF antibody as described in Materials and Methods. Prior to reversal of formaldehyde cross-linking, precipitates were washed, resuspended and subjected to re-ChIP using anti-Sp1 and anti-Sp3 antibody as described. PCR was then performed as described ChIP assay. Lanes 1, 3 and 7: 100 bp DNA marker; lane 2: input; lane 4: PI; lane 5: IRF-4 antibody; lane 6: Sp1 antibody; lane 8: PI; lane 9: IRF-4 antibody; lane 10: Sp3 antibody.