Nuclear respiratory factor 2 regulates the transcription of AMPA receptor subunit GluA2 (Gria2)

Abstract

Neuronal activity is highly dependent on energy metabolism. Nuclear respiratory factor 2 (NRF-2) tightly couples neuronal activity and energy metabolism by transcriptionally co-regulating all 13 subunits of an important energy-generating enzyme, cytochrome c oxidase (COX), as well as critical subunits of excitatory NMDA receptors. AMPA receptors are another major class of excitatory glutamatergic receptors that mediate most of the fast excitatory synaptic transmission in the brain. They are heterotetrameric proteins composed of various combinations of GluA1-4 subunits, with GluA2 being the most common one. We have previously shown that GluA2 (Gria2) is transcriptionally regulated by nuclear respiratory factor 1 (NRF-1) and specificity protein 4 (Sp4), which also regulate all subunits of COX. However, it was not known if NRF-2 also couples neuronal activity and energy metabolism by regulating subunits of the AMPA receptors. By means of multiple approaches, including electrophoretic mobility shift and supershift assays, chromatin immunoprecipitation, promoter mutations, real-time quantitative PCR, and western blot analysis, NRF-2 was found to functionally regulate the expression of Gria2, but not of Gria1, Gria3, or Gria4 genes in neurons. By regulating the GluA2 subunit of the AMPA receptor, NRF-2 couples energy metabolism and neuronal activity at the transcriptional level through a concurrent and parallel mechanism with NRF-1 and Sp4.

Keywords: AMPA receptor; GABP; Gene regulation; GluA2; Nuclear respiratory factor 2; Transcription factor.

Figure 1

In vitro binding of NRF-2 to putative binding sites on the AMPA receptor subunit gene promoters as determined with EMSA and supershift assays. ³²P-labeled oligonucleotides, excess unlabeled oligos as competitors, excess unlabeled mutant NRF-2 oligos as competitors, N2a nuclear extract, and NRF-2α antibodies are indicated by a + or a − sign. Arrowheads indicate specific NRF-2 shift, supershift, and non-specific complexes. The positive control, COX6b, shows a shift and supershift band (lanes 1 and 3, respectively). The addition of excess unlabeled probe competed out the shift band (lane 2). The addition of N2a nuclear extract yielded specific shift bands for both Gria1 and Gria2 (lanes 4 and 9, respectively) that were competed out by an excess of unlabeled oligos (lanes 5 and 10, respectively). The addition of NRF-2 antibody yielded a supershift band for both Gria1 and Gria2 (lanes 6 and 11, respectively). The addition of excess unlabeled probes with mutated NRF-2 binding sites did not compete out the shift reaction (lanes 7 and 12, respectively). The addition of NRF-2 antibody to labeled Gria1 and Gria2 probes in the absence of N2a extract did not reveal any antibody-to-probe reaction (lanes 8 and 13, respectively). Labeled Gria1 and Gria2 probes with mutated NRF-2 sites did not yield a specific NRF-2 shift band (lanes 14 and 16, respectively), nor a supershift band with the addition of NRF-2 antibody (lanes 15 and 17, respectively).

Figure 2

In vivo interaction of NRF-2 with the AMPA receptor subunit gene promoters using the ChIP assay in N2a cells (A) and murine visual cortical tissue (B). Nuclear extract was immunoprecipitated with anti NRF-2α antibodies (NRF-2 IP lane), anti-nerve growth factor receptor p75 antibody (negative control, NGFR IP lane), or no antibody (negative control, No Ab lane). Control reactions for PCR were performed with 0.5% (Input 0.5% IP lane) and 0.1% (Input 0.1% IP lane) of input chromatin. COX6B promoter was used as a positive control, and Exon 8 of NRF-1 was used as a negative control. Results indicate NRF-2 interactions with the tested region on the Gria2 promoter, but not the Gria1, Gria3, or Gria4 promoters, in both N2a cells and murine visual cortical tissue.

Figure 3

Site-directed mutational analysis of promoters using luciferase reporter gene constructs. Wild type promoters (wt) and those with mutated NRF-2 binding site (mut) for COX6b and Gria2 are indicated. COX6b served as a positive control. Mutating the NRF-2 site resulted in a significant decrease in the luciferase activity as compared to the wild type. Similarly, mutating the NRF-2 binding sites on the Gria2 promoter resulted in significant decreases in luciferase activity. KCl depolarization significantly increased promoter activity in all wild type, but not in the COX6b and Gria2 promoters with mutated NRF-2 sites. N = 6 for each construct. ***= P < 0.001; X = NS. All mutants and wild type + KCl are compared to the wild type. All mutant + KCl are compared to mutants.

Figure 4

Effect of RNA interference-mediated silencing of NRF-2α on the expression of COX and AMPA receptor subunit genes. (A) Western blots revealed a down-regulation of NRF-2α and GluA2 protein levels (lane 2) as compared to controls (lane 1), but not GluA1 protein levels, in NRF-2α shRNA-transfected cells. β-actin served as a loading control. N = 3 for each data point; ***= P < 0.001 and *= P < 0.05 when compared to pBS/U6 empty vector controls. (B) As determined by real-time PCR, NRF-2α shRNA transfection in N2a cells down-regulated mRNA levels of NRF-2α and Gria2, but not those of Gria1, Gria3, and Gria4. mRNA levels of the positive control, COX7c, were also reduced with NRF-2α silencing. N = 6 for each data point. ***= P < 0.001 and **= P < 0.01 when compared to pBS/U6 empty vector controls.

Figure 5

Effect of over-expressing NRF-2α and β on the transcript and protein levels of COX7c and AMPA receptor subunit genes. (A) NRF-2α/β over-expression increased protein levels of NRF-2α and NRF-2β, as well as GluA2 (lane 2) as compared to controls (lane 1). Protein levels of GluA1 did not increase significantly with NRF-2α/β over-expression. β-actin served as a loading control. N = 3 for each data point; ***= P < 0.001 when compared to pcDNA3.1 empty vector controls. (B) Real-time PCR revealed an up-regulation of NRF-2α and β mRNA with NRF-2α/β over-expression as compared to pcDNA3.1 empty vector controls. (C) In N2a cells, mRNA levels of Gria2, but not those of Gria1, Gria3, and Gria4, were increased with NRF-2α/β over-expression. mRNA levels of the positive control, COX7c, were also increased with NRF-2α/β over-expression. N = 6 for each data point. **= P < 0.01 and ***= P < 0.001 when compared to pcDNA3.1 empty vector controls.

Figure 6

Effect of KCl or TTX treatment in the presence of NRF-2 silencing or over-expression, respectively, on the transcript levels of the AMPA receptor subunit genes and COX7c in N2a cells (A) Cells treated for 5 h with 20 mM KCl revealed an up-regulation of all transcripts as compared to pBS/U6 empty vector controls. In the presence of shRNA against NRF-2α, 5 h treatment with 20 mM KCl failed to up-regulate the transcripts of Gria2 and COX7c, but it did up-regulate those of Gria1, Gria3, and Gria4. (B) N2a cells treated for 3 days with 0.4 μM TTX revealed a down-regulation of all tested transcripts as compared to pcDNA3.1 empty vector controls. Over-expression of NRF-2α and β rescued the down-regulation of the COX7c and Gria2 transcripts, but not those of Gria1, Gria3, and Gria4. N = 6 for each data point; ***= P < 0.001 when compared to controls; ### = P < 0.001 and X = non-significant when compared to KCl- or TTX-treated samples.

Figure 7

Effect of NRF-2 silencing and over-expression, with and without KCl or TTX treatment, respectively, on the transcript levels of the AMPA receptor subunit genes in visual cortical neurons. (A) NRF-2α shRNA transfection in primary neurons down-regulated mRNA levels of NRF-2α and Gria2, but not those of Gria1, Gria3, and Gria4. Primary neurons treated for 5 h with 20 mM KCl revealed an up-regulation of all transcripts as compared to pBS/U6 empty vector controls. In the presence of shRNA against NRF-2α, 5 h treatment with 20 mM KCl did not up-regulate transcripts of NRF-2α and Gria2, but it did up-regulate those of Gria1, Gria3, and Gria4. N = 3 for each data point. ***= P < 0.001, **= P < 0.01 and *= P < 0.05 when compared to pBS/U6 empty vector controls. ### = P < 0.001 and X = non-significant when compared to KCl- treated samples. (B) In primary neurons, NRF-2α/β over-expression led to an increase in the transcript levels of NRF2α and Gria2, but not those of Gria1, Gria3, and Gria4. Primary neurons treated for 3 days with 0.4 μM TTX revealed a down-regulation of all tested transcripts as compared to pcDNA3.1 empty vector controls. Over-expression of NRF-2α and β rescued the down-regulation of Gria2 transcripts, but not those of Gria1, Gria3, and Gria4. N = 3 for each data point; ***= P < 0.001 and **= P < 0.01 when compared to controls; ### = P < 0.001 and X = non-significant when compared to TTX-treated samples.

Figure 8

Aligned partial sequences of Gria2 promoter from mouse, rat, and human showed conservation of the NRF-2 binding site.