Selective coactivation of α7- and α4β2-nicotinic acetylcholine receptors reverses beta-amyloid-induced synaptic dysfunction

Abstract

Beta-amyloid (Aβ) has been recognized as an early trigger in the pathogenesis of Alzheimer's disease (AD) leading to synaptic and cognitive impairments. Aβ can alter neuronal signaling through interactions with nicotinic acetylcholine receptors (nAChRs), contributing to synaptic dysfunction in AD. The three major nAChR subtypes in the hippocampus are composed of α7-, α4β2-, and α3β4-nAChRs. Aβ selectively affects α7- and α4β2-nAChRs, but not α3β4-nAChRs in hippocampal neurons, resulting in neuronal hyperexcitation. However, how nAChR subtype selectivity for Aβ affects synaptic function in AD is not completely understood. Here, we showed that Aβ associated with α7- and α4β2-nAChRs but not α3β4-nAChRs. Computational modeling suggested that two amino acids in α7-nAChRs, arginine 208 and glutamate 211, were important for the interaction between Aβ and α7-containing nAChRs. These residues are conserved only in the α7 and α4 subunits. We therefore mutated these amino acids in α7-containing nAChRs to mimic the α3 subunit and found that mutant α7-containing receptors were unable to interact with Aβ. In addition, mutant α3-containing nAChRs mimicking the α7 subunit interact with Aβ. This provides direct molecular evidence for how Aβ selectively interacted with α7- and α4β2-nAChRs, but not α3β4-nAChRs. Selective coactivation of α7- and α4β2-nAChRs also sufficiently reversed Aβ-induced AMPA receptor dysfunction, including Aβ-induced reduction of AMPA receptor phosphorylation and surface expression in hippocampal neurons. Moreover, costimulation of α7- and α4β2-nAChRs reversed the Aβ-induced disruption of long-term potentiation. These findings support a novel mechanism for Aβ's impact on synaptic function in AD, namely, the differential regulation of nAChR subtypes.

Keywords: Alzheimer's disease; alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPA receptor, AMPAR); amyloid-beta (Aβ); nicotinic acetylcholine receptor (nAChR); synaptic plasticity.

Figure 1

Interaction between Aβ42 and nAChRs is important for Aβ42-induced Ca²⁺hyperexcitation. Representative traces of GCaMP6f fluorescence intensity in hippocampal neurons in each condition and a summary graph of the normalized average of total Ca²⁺ activity in neurons treated with either 250 nM sAβ42 (black) or 250 nM oAβ42 (red) in the absence or presence of 1 μM Aβcore or inactive 1 μM Aβcore (inAβcore) (n = number of neurons [sAβ42, n = 127; sAβ42+Aβcore, n = 23; sAβ42+inAβcore, n = 32; oAβ42, n = 71; oAβ42+Aβcore, n = 30; and oAβ42+inAβcore, n = 31], ∗∗p < 0.01 and ∗∗∗∗p < 0.0001 and, one-way ANOVA, Fisher's least significant difference test).

Figure 2

Aβ42 selectively interacts with α7- and α4β2-nAChRs, but not α3β4-nAChRs.A, Co-IP shows Aβ42 interacts with α7-nAChRs. B, Co-IP shows Aβ42 binds to α4β2-nAChRs. C, Co-IP shows Aβ42 is unable to interact with α3β4-nAChRs. D, sequence and numbering of human α7-nAChRs in the loop C region and its alignment with related human and mouse nAChR sequences. Y210 (bold) is the ligand-binding residue and conserved in all human and mouse α subunits. R208 (blue) and E211 (green) are predicted to be critical for interaction with the N terminus of Aβ, which are conserved only in both human and mouse α4 and α7 subunits except mouse α7 receptors that have positive-charged lysine (light blue), which is similar to positive-charged arginine. However, both mouse and human α3 receptors have uncharged residues in both positions (red). E, Co-IP shows Aβ42 is unable to interact with the α7 R208I mutant. F, Co-IP shows Aβ42 is unable to bind to the α7 E211N mutant. G, Co-IP shows Aβ42 is unable to interact with the α3 I284R mutant. H, Co-IP shows Aβ42 cannot interact with the α3 N287E mutant. I, double α3 I284R/N287E mutant is able to pull down Aβ42. Co-IP, coimmunoprecipitation.

Figure 3

Selective coactivation of α7- and α4β2-nAChRs reverses Aβ-induced reduction of AMPAR surface expression. Representative immunoblots of input (I) and surface (S) levels and quantitative analysis in (A) single activation of each nAChRs (n = 6 immunoblots from three independent cultures duplicated, ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, one-way ANOVA, Fisher's least significant difference [LSD] test). B, double activation of each nAChRs (n = 6 immunoblots from three independent cultures duplicated, ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001, one-way ANOVA, Fisher's LSD test). C, triple activation and cholinergic stimulation (n = 6 immunoblots from three independent cultures duplicated, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001, one-way ANOVA, Fisher's LSD test).

Figure 4

Coactivation of α7- and α4β2-nAChRs reverses Aβ-induced impaired AMPA receptor phosphorylation and synaptic plasticity.A, representative immunoblots of pGluA1 levels in each condition. B, quantitative analysis of pGluA1 levels under the basal condition in each condition (n = 9 immunoblots from four independent cultures, ∗p < 0.05 and ∗∗p < 0.01, one-way ANOVA, Fisher's least significant difference test). C, quantitative analysis of pGluA1 levels following cLTP induction in each condition (n = 11 immunoblots from five independent cultures, ∗p < 0.05 and ∗∗p < 0.01, one-way ANOVA, Fisher's least significant difference test).

Figure 5

Schematic model.A, impact of Aβ oligomers. In the hippocampus, α7- and α4β2-nAChRs are prominently expressed on inhibitory interneurons; thus, selective binding of soluble Aβ42 oligomers (oAβ42) to α7- and α4β2-nAChRs but not α3β4-nAChRs, reduces neuronal activity in inhibitory cells, leading to a decrease in the release of GABA onto hippocampal excitatory neurons. Consequently, excitatory cells have increased frequency of Ca²⁺ transients, resulting in elevated calcineurin (CaN) activity. Calcineurin then dephosphorylates the AMPA receptor (AMPAR) subunit, GluA1, promoting AMPAR endocytosis and resulting in an overall decrease of AMPAR surface expression. This ultimately contributes to disruptions of long-term potentiation. B, reversal of Aβ-induced synaptic and neuronal dysfunction by costimulation with α7- and α4β2-nAChRs agonists. As Aβ42 inhibits both α7- and α4β2-nAChRs but not α3β4-nAChRs, costimulation of α7- and α4β2-nAChRs by selective agonists, PNU-282987 (PNU) and RJR-2403 Oxalate (RJR), can restore normal activity of both hippocampal inhibitory and excitatory cells, reversing Aβ-induced synaptic dysfunction. This restoration of normal Ca²⁺ activity prompts a decrease in calcineurin activity, leading to a decrease in AMPAR dephosphorylation and AMPAR endocytosis, ultimately restoring normal long-term potentiation. However, an agonist for α3β4-nAChRs, NS-3861 (NS), does not appear to have neuroprotective effects. Moreover, nonspecific stimulation of nAChRs by using three agonists together or carbachol is unable to reverse the Aβ effects on neuronal activity and synaptic function, emphasizing the importance of selective costimulation of nAChRs as potential therapeutic approaches.