Blockade of EphA4 signaling ameliorates hippocampal synaptic dysfunctions in mouse models of Alzheimer's disease

Abstract

Alzheimer's disease (AD), characterized by cognitive decline, has emerged as a disease of synaptic failure. The present study reveals an unanticipated role of erythropoietin-producing hepatocellular A4 (EphA4) in mediating hippocampal synaptic dysfunctions in AD and demonstrates that blockade of the ligand-binding domain of EphA4 reverses synaptic impairment in AD mouse models. Enhanced EphA4 signaling was observed in the hippocampus of amyloid precursor protein (APP)/presenilin 1 (PS1) transgenic mouse model of AD, whereas soluble amyloid-β oligomers (Aβ), which contribute to synaptic loss in AD, induced EphA4 activation in rat hippocampal slices. EphA4 depletion in the CA1 region or interference with EphA4 function reversed the suppression of hippocampal long-term potentiation in APP/PS1 transgenic mice, suggesting that the postsynaptic EphA4 is responsible for mediating synaptic plasticity impairment in AD. Importantly, we identified a small-molecule rhynchophylline as a novel EphA4 inhibitor based on molecular docking studies. Rhynchophylline effectively blocked the EphA4-dependent signaling in hippocampal neurons, and oral administration of rhynchophylline reduced the EphA4 activity effectively in the hippocampus of APP/PS1 transgenic mice. More importantly, rhynchophylline administration restored the impaired long-term potentiation in transgenic mouse models of AD. These findings reveal a previously unidentified role of EphA4 in mediating AD-associated synaptic dysfunctions, suggesting that it is a new therapeutic target for this disease.

Keywords: Abeta; drug discovery; ephrin; receptor tyrosine kinase.

Fig. 1.

Aβ stimulates the activation of EphA4 signaling. (A and B) Western blot of Tyr602 EphA4 phosphorylation (P-Tyr EphA4) in hippocampal synaptosomal fractions of WT and APP/PS1 mice. Quantification analysis (*P < 0.05; one-way ANOVA, Student–Newman–Keuls test; n ≥ 3 hippocampi). (C and D) Aβ increased EphA4 tyrosine phosphorylation. Acute rat hippocampal slices were treated with Aβ at different concentrations for 2 h or 500 nM for various periods. EphA4 was immunoprecipitated, and subjected to Western blotting for P-Tyr EphA4. (*P < 0.05, **P < 0.01; one-way ANOVA, Student–Newman–Keuls test; n ≥ 3 slice samples.) (E and F) Aβ increased the number of EphA4 clusters. (E) Representative images. (Scale bar, 10 µm.) (F) Quantification analysis (***P < 0.001, Student t test, n = 15 neurons per group). (G and H) Blockade of EphA4–ligand interaction abolished the Aβ-stimulated activation of EphA4. Acute rat hippocampal slices were pretreated with KYL peptide followed by Aβ. (G) EphA4 was immunoprecipitated and subjected to Western blot analysis for P-Tyr EphA4. (H) Fold change of P-Tyr EphA4/EphA4. [**P < 0.01; one-way ANOVA followed by the Student–Newman–Keuls test (n ≥ 3 slice samples)].

Fig. 2.

Inhibition of EphA4 activity prevents Aβ-mediated neurotransmission impairment. (A and B) KYL peptide attenuated the Aβ-triggered reduction of dendritic spines in hippocampal neurons. (A) Representative images. (Scale bar, 10 µm.) (B) Quantification analysis [***P < 0.001; two-way ANOVA followed by the Bonferroni posttest (n ≥ 22 neurons for each condition)]. (C and D) Unclustered EphA4-Fc, which inhibits the EphA4 forward signaling, rescued the Aβ-triggered increase in the interevent interval of mEPSCs. (C) Representative mEPSC traces. (D) mEPSC interevent interval (mean ± SEM, n ≥ 39 neurons for each condition). (E and F) Small-molecule inhibitor for EphA4 (Cpd1) and Cdk5 (Ros) prevented the Aβ-induced increase in interevent interval. (E) Representative mEPSCs. (F) mEPSC interevent interval (mean ± SEM, n ≥ 22 neurons for each condition). (D and F) ***P < 0.001, ^###P < 0.001; one-way ANOVA with Kruskal–Wallis test.

Fig. 3.

Blockade of EphA4 signaling rescues the impairment of hippocampal LTP in AD models. (A and B) Blockade of EphA4 activity rescued the Aβ-induced reduction of LTP. Acute hippocampal slices were treated with Aβ in the presence of KYL peptide. (C–F) Blockade of postsynaptic EphA4 signaling rescued the LTP impairment in APP/PS1 mutant mice. WT and APP/PS1 mutant mice were infused with KYL peptide (C and D); or CA1 regions of WT and APP/PS1 mutant mice were injected with EphA4-shRNA (shEphA4) or GFP virus (E and F). (A, C, and E) Points represent averaged slopes of fEPSP normalized to baseline values (mean ± SEM). Trace recordings 5 min before (1) and 50 min after (2) LTP induction (arrow) are shown. Inset traces are representative fEPSPs recorded before (gray) and after (black) HFS. (B, D, and F) Quantification of mean fEPSP slopes in the last 10 min of the recording after LTP induction; n = ≥ 9 slices from 5 brains for each condition. (B) **P < 0.01, two-way ANOVA followed by the Bonferroni posttest; ^###P < 0.001, one-way ANOVA followed by the Student–Newman–Keuls test. (D) *P < 0.05, ^#P < 0.05; one-way ANOVA followed by the Student–Newman–Keuls test. (F) ***P < 0.001, ^#P < 0.05; two-way ANOVA with Bonferroni posttest.

Fig. 4.

Rhy is a small-molecule EphA4 inhibitor. (A) Rhy binds the extracellular domain of EphA4 but not that of EphB2. In vitro pulldown assay of recombinant EphA4-Fc or EphB2-Fc proteins with biotinylated Rhy (n = 3 experiments). (B and C) Rhy inhibited the ephrin-A (A1)-stimulated EphA4 tyrosine phosphorylation in rat cortical neurons. Lysate was immunoprecipitated with EphA4 antibody and subjected to Western blot analysis for P-Tyr. (C) Quantification analysis (mean ± SEM, **P < 0.01, Student t test, n = 5 neuronal cultures per group). (D and E) Rhy inhibited EphA4-dependent signaling and cellular functions. (D) Rhy abolished the ephrin-A1–induced EphA4 clustering. ***P < 0.005, ^##P < 0.01, ^###P < 0.001; one-way ANOVA followed by the Student–Newman–Keuls test; n ≥ 9 neurons. (E) Rhy inhibited ephrin-A1–stimulated growth cone collapse (mean ± SEM, ≥150 neurons for each group). *P < 0.05, ***P < 0.001, two-way ANOVA followed by the Bonferroni posttest; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001, one-way ANOVA followed by the Student–Newman–Keuls test. (F and G) Rhy administration reduced EphA4 activation in the hippocampus of APP/PS1 mice. (F) Hippocampal synaptosomal fractions of APP/PS1 mice administered with Rhy were collected. (G) Fold change (mean ± SEM, n ≥ 3 mice, *P < 0.05, two-way ANOVA followed by the Bonferroni posttest).

Fig. 5.

Rhy rescues the Aβ-induced deficit in neurotransmission and LTP inhibition in AD mice. (A and B) Pretreatment with Rhy abolished the Aβ-induced reduction of neurotransmission in hippocampal neurons. (A) Representative mEPSC traces. (B) mEPSC interevent interval (mean ± SEM; n ≥ 19 neurons for each group; ***P < 0.001, ^###P < 0.001; one-way ANOVA followed by the Kruskal–Wallis test.) (C and D) Rhy prevented the Aβ-induced inhibition of LTP. Acute hippocampal slices were treated with Aβ in the presence of Rhy. (C) Averaged slopes of baseline-normalized fEPSP (mean ± SEM). (D) Quantification of mean fEPSP slopes during the last 10 min of the recording after LTP induction (n ≥ 10 slices from 6 brains; ***P < 0.001, two-way ANOVA followed by the Bonferroni posttest; ^###P < 0.001, one-way ANOVA followed by Student–Newman–Keuls test). (E and F) Rhy rescued the LTP impairment in APP/PS1 mutant mice. WT and APP/PS1 mutant mice were orally administered Rhy. (E) Averaged slopes of baseline-normalized fEPSP (mean ± SEM). (F) Quantification of mean fEPSP slopes during the last 10 min of the recording after LTP induction (n ≥ 8 slices from 6 brains; ***P < 0.001, ^###P < 0.001; two-way ANOVA followed by the Bonferroni posttest). (C and E) Trace recordings 5 min before (1) and 50 min after (2) LTP induction (arrow) are shown. Inset traces are examples of fEPSPs recorded before (gray) and after (black) HFS.