The DREAM Protein Negatively Regulates the NMDA Receptor through Interaction with the NR1 Subunit

Identification of the binding sites between DREAM and NR1. A, Schematic representation of the plasmid construction of GST- and His₆-fusion protein containing different domains of the NR1-1a C terminus. B, GST-C0, C0C1, and C0C1C2 precipitated both His₆-DREAM (top) and endogenous DREAM present in hippocampal tissue (bottom), whereas GST, GST-C1, C2, and C1C2 did not. The bands detected by the GST antibody represent the loading amount. C, Schematic representation of the plasmid construction of GST- and His₆-fusion proteins containing different domains of DREAM. D, GST-DREAM-1-50, 101–256, Δ1-50, and full-length precipitated both His₆-NR1a Ct (top) and endogenous NR1 present in hippocampal tissue (bottom), whereas GST and GST-51–100 did not. E, Sensitivity of the binding between DREAM-1-50 or Δ1-50 and NR1a C terminus to Ca²⁺. F, EF hand mutation abolishes the Ca²⁺ sensitivity of the binding between Δ1-50 DREAM and NR1a C terminus, although it could not cancel their binding. NT, No treatment; EFmu DREAM, EF hand mutant of DREAM.

Inhibition of NMDAR-mediated current and NMDAR surface expression by coexpression of DREAM. A, Inhibition of the NMDAR-mediated current by coexpression of DREAM. Glutamate-evoked whole-cell currents were recorded from Xenopus oocytes expressing NR2B-containing NMDARs alone (gray traces) or together with DREAM (black traces). The bars indicate the duration of the application of agonists with different concentrations (10, 30, 100, and 300 μml-glutamate, respectively, plus 100 μm glycine). B, Quantitative analysis of the peak current responses in A. ***p < 0.001, two-way ANOVA followed by Bonferroni's posttests. Error bars indicate SEM. C, Coexpression of DREAM suppressed the surface expression of NR1 and NR2B subunits in oocytes. Protein surface expression was detected by surface biotinylation assay. D, Coexpression of DREAM inhibited the surface expression of NR1 and NR2B subunits in CHO cells transfected pEGFP-N1-DREAM along with pCDNA3-NR1-1a and PRK5-NR2B plasmids compared with control group.

Neuroprotective effect of DREAM overexpression in primary cultured hippocampal neurons. A, Coexpression of full-length DREAM and the Δ1-50 fragment inhibited NMDA-induced cell injury. HEK 293 cells were transfected with NMDARs together with GFP, full-length, Δ1-50 DREAM, or EF hand mutant DREAM (EFmu DREAM) and treated with 100 μm NMDA (plus 10 μm glycine) or 100 μm MK801 in Mg²⁺-free balanced salt solutions for 2 h. Then LDH release in the medium was assayed. *p < 0.05, ***p < 0.001; n = 3–4; one-way ANOVA followed by Newman–Keuls posttests. B, Adenovirus-mediated overexpression of DREAM in hippocampal neurons detected by Western blot analysis. C, Overexpression of DREAM decreased NMDA-induced LDH release in hippocampal neurons. *p < 0.05, **p < 0.01, ***p < 0.001; n = 4; unpaired t test. D, Overexpression of DREAM reduced OGD-induced LDH release in hippocampal neurons. *p < 0.05, ***p < 0.001; n = 4; unpaired t test. Error bars indicate SEM. E, Representative images of PI staining in hippocampal neurons after OGD or control treatment for 30 min and reperfusion for 24 h. Overexpression of DREAM reduced the number of PI-positive cells. F, Overexpression of DREAM specifically inhibited the surface expression of the NR1 subunit without effect on N-cadherin in hippocampal neurons.

Enhancement of NMDAR-mediated current and excitotoxic injury by DREAM knockdown. A, Knockdown of DREAM expression by DREAM siRNA in HEK 293 cells transfected with DREAM plasmids. B, Quantitative analysis of PI-positive cells in hippocampal neurons challenged with NMDA for 1 h, followed by reperfusion for 12 h after transfection with DREAM siRNA or control siRNA for 24–48 h. DREAM siRNA further exaggerated NMDA-induced neuronal injury. *p < 0.05; n = 25–35; two-way ANOVA followed by Bonferroni's posttests. C, Representative traces of NMDA-induced current recorded in cultured hippocampal neurons by whole-cell patch clamp. D, Quantitative analysis of the peak current amplitude. DREAM siRNA significantly potentiated NMDA-induced current in hippocampal neurons. *p < 0.05; n = 5; one-way ANOVA followed by Newman–Keuls posttests. Error bars indicate SEM. E, Knockdown of DREAM in hippocampal neurons significantly increased the surface expression of the NR1 subunit without affecting the total amount. F, Efficient knockdown of DREAM expression by DREAM siRNA in hippocampal neurons.

Construction of TAT-21-40 fusion peptide and its effect on NMDAR-mediated current. A, Schematic representation of GST fusion proteins containing different fragments of DREAM-1-50. B, DREAM 21-40 amino acid residues are the critical binding sites with NR1a C terminus. C, Sequence of TAT-fusion peptides, TAT-21-40 and scramble TAT. D, Effect of TAT-fusion peptides on the binding of GST-1-50 and His₆-NR1a Ct. TAT-21-40 significantly suppressed the binding of DREAM-1-50 and NR1a C terminus. E, Effect of TAT-fusion peptides on the surface expression of NR1 in hippocampal neurons. Treatment with 3 μm TAT-fusion peptides for 2 h significantly inhibited the surface expression of NR1. F, Representative traces of NMDA-induced current recorded in cultured hippocampal neurons using whole-cell patch clamp by TAT-fusion peptide pretreatment and posttreatment, respectively. Top, Perfusion of the neuron with scramble-TAT (3 μm) for 1 min did not affect NMDA-induced current. Bottom, Perfusion of the neuron with TAT-21-40 (3 μm) for 1 min greatly inhibited NMDA-induced current. G, Quantitative analysis of the peak current response in F. The values for the posttreatment response were normalized to the corresponding pretreatment values. ***p < 0.001; n = 6–7; two-way ANOVA followed by Bonferroni's posttests. Error bars indicate SEM.

Protection of TAT-21-40 against focal ischemia injury. A, Schematic representation of the TAT fusion peptide pretreatment program in vivo. TAT-fusion peptide was administered 20 min before MCAO through lateral ventricle injection. B–D, Pretreatment with TAT-21-40 but not scramble-TAT significantly reduced infarction volume (B) and edema (C), and improved neurological scores (D) 6 h after MCAO. *p < 0.05, ***p < 0.001, unpaired t test. E, Schematic representation of the TAT fusion peptide posttreatment program in vivo. TAT-fusion peptide was administered 20 min after the onset of ischemia through lateral ventricle injection. F–H, Posttreatment with TAT-21-40 but not scramble-TAT significantly reduced infarction volume (F) and edema (G), and improved neurological scores (H) 6 h after MCAO. **p < 0.01, ***p < 0.001, unpaired t test. Error bars indicate SEM. I, J, Representative photographs of TTC-stained coronal sections of mouse brains by TAT-fusion peptide pretreatment (I) and posttreatment (J). The lines illustrate the infarct border for each section.