Targeting Epac2 and GluA3-containing AMPARs: a novel therapeutic strategy for Alzheimer’s disease

Alzheimer’s disease (AD) is a neurodegenerative disease that manifests progressive decline in memory and cognition. In the early stage of AD, memory retrieval is impaired preceding memory acquisition and consolidation (Roy et al., 2016). Prior to the onset of symptoms, pathological amyloid-β (Aβ) plagues and tau protein tangles accumulate in extracellular and intracellular spaces, respectively, leading to neurodegeneration. Among these hallmark pathologies, Aβ is proposed to be the primary etiology by triggering a cascade of pathogenic events, including neuroinflammation, oxidative stress, tau hyperphosphorylation, synaptic/neuronal dysfunction, and neuronal death (Zhang et al., 2023b). In the early stage of AD, the Aβ accumulation appears in the medial temporal lobe of the brain, such as the hippocampus and entorhinal cortex, which are crucial structures for memory and cognitive functions. Consequently, anti-amyloid therapy is a primary strategy against AD. A recent phase 3 trial reported that lecanemab, a newly developed monoclonal antibody against Aβ protofibrils, greatly decreased the makers of Aβ, tau, neuroinflammation, and neurodegeneration in cerebrospinal fluid, and moderately improved cognitive decline in patients with mild AD after 18-month treatment (van Dyck et al., 2023). The success of lecanemab supports the Aβ hypothesis and suggests Aβ pathology as a target for AD treatment. As the Aβ hypothesis suggests, Aβ oligomers induce subtle effects on synaptic transmission before other pathogenic events. Thus, in addition to the anti-amyloid therapy, targeting glutamatergic receptors in postsynaptic surfaces, particularly α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs) and N-methyl-D-aspartate receptors (NMDARs), is a promising strategy to prevent synaptic dysfunction and alleviate memory deficits in the early stage of AD. During induction of long-term potentiation (LTP), activation of NMDARs leads to a transient but strong Ca2+ influx into postsynaptic neurons. The Ca2+ elevation triggers several intracellular signaling pathways, such as calcium/calmodulin-dependent protein kinase II, protein kinase A (PKA), and protein kinase C. Subsequently, these pathways lead to phosphorylation of AMPARs, finally leading to insertion and/or activation of postsynaptic AMPARs (Nicoll and Schulman, 2023). Notably, these postsynaptic AMPARs play key roles in strengthening synaptic transmission and plasticity. However, in AD, Aβ oligomers induce a moderate Ca2+ influx via NMADRs or other Ca2+-permeable receptors, subsequently resulting in endocytosis of postsynaptic AMPARs (Reinders et al., 2016; Zhang et al., 2023b). This pathological process resembles long-term depression and thereby impairs memory. Therefore, activation and/or upregulation of synaptic AMPARs could be a therapeutic strategy to counteract the postsynaptic AMPAR loss and memory impairment in AD. In hippocampal CA1 neurons, two types of heteromeric AMPARs are predominant in equal amounts: GluA1-containing AMPARs and GluA3-containing AMPARs. During LTP induction, GluA1-containing AMPARs are phosphorylated and transit into high-conductance states. Subsequently, more GluA1-containing AMPARs are trafficked and anchored in the postsynaptic surface via the postsynaptic density (PSD) (Nicoll and Schulman, 2023). During LTP maintenance, GluA1-containing AMPARs are gradually replaced by GluA3-containing AMPARs (Figure 1), followed by constant cycling in and out of the postsynaptic surface, which is independent of neuronal activity and does not affect the magnitude of synaptic transmission. This trafficking event is suggested to stabilize LTP and be implicated in memory consolidation (Zhang et al., 2024).Figure 1: Epac2 signaling mediates synaptic plasticity of GluA3-containing AMPARs.In postsynaptic neurons, Epac2 and GluA3-containing AMPARs are anchored together via AKAP5 and PSD95. Activation of β-adrenergic receptors leads to an elevated intracellular level of cAMP, which subsequently activates Epac2. Epac2 initiates its downstream Ras-p42/p44 pathway, leading to activation of GluA3-containing AMPARs. This activation increases their opening probability of GluA3-containing AMPARs, thereby potentiating synaptic plasticity. Created with BioRender.com. AKAP5: A-kinase anchoring protein 5; AMPARs: α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors; Epac2: exchange protein directly activated by cAMP 2; MAPK: mitogen-activated protein kinase; PSD95: postsynaptic density 95.Recently, GluA3-containing AMPARs are found to be involved in a novel form of synaptic plasticity in hippocampal CA1 neurons and cerebellar Purkinje cells (Gutierrez-Castellanos et al., 2017; Renner et al., 2017). This synaptic plasticity has several distinct characteristics compared to GluA1-mediated plasticity. Firstly, this synaptic plasticity is potentiated via rapid activation of the GluA3-containing AMPARs without involving receptor trafficking. Secondly, this synaptic plasticity is characterized by a significant increase in the frequency of miniature excitatory postsynaptic currents and an increasing trend in miniature excitatory postsynaptic current amplitude. Thirdly, this GluA3-mediated plasticity does not require NMDARs but is dependent on a rise of intracellular cyclic AMP (cAMP). The elevated cAMP results in an increased opening probability of GluA3-containing AMPARs, thereby facilitating basal transmission and LTP. Furthermore, other key components in cAMP signaling, including β-adrenergic receptors, PKA, exchange protein directly activated by cAMP (Epac) and Ras (Figure 1), are implicated in this novel GluA3 plasticity (Gutierrez-Castellanos et al., 2017; Renner et al., 2017). In AD, both the expression and function of GluA3-containing AMPARs are compromised. We have reported that the protein level of hippocampal GluA3 subunits is significantly downregulated in post-mortem materials of AD patients and an Aβ-overexpressing mouse line (J20) (Zhang et al., 2023a). In addition, postsynaptic density 95 (PSD95), the anchoring protein for synaptic AMPARs, tends to decrease in the hippocampus of AD patients and J20 mice (Zhang et al., 2023a). Functionally, knockout of GluA3 subunit leads to resistance to a series of Aβ-mediated events including synaptic depression, spine loss and LTP blockage in hippocampal CA1 neurons, and memory impairment in APP/PS1 mice (Reinders et al., 2016). These studies suggest that the Aβ-induced synaptic and memory impairment in AD are specifically triggered via removing postsynaptic GluA3-containing AMPARs. Epac, a main cAMP effector implicated in the regulation of GluA3 plasticity, has two isoforms: Epac1 and Epac2. Compared to the low expression of Epac1, Epac2 is abundantly expressed in the brain (Ostroveanu et al., 2010). Epac initiates its downstream signaling via exchanging guanine nucleotides (GDP to GTP) on small GTPases, such as Ras and Rap. Epac plays different roles in synaptic plasticity via enhancing presynaptic and/or postsynaptic plasticity in different brain regions (Zhang et al., 2024). It is worth noting that Epac is not involved in LTP induction, but LTP maintenance which converts early LTP into long-lasting LTP, a cellular event related to memory retrieval. In presynaptic areas, Epac regulates vesicular docking and release of glutamate, while in postsynaptic areas, it mediates the activation of GluA3-containing AMPARs (Zhang et al., 2024). Due to the lack of Epac2 modulators, previous studies used the Epac activator 8-pCPT, which could not distinguish the effect of Epac2 from Epac1 on synaptic plasticity (Zhang et al., 2024). However, by using genetic deletion of Epac2, a study revealed that it was Epac2, rather than Epac1, potentiating the cAMP-driven synaptic plasticity in hippocampal neurons (Fernandes et al., 2015). Furthermore, our lab previously revealed that intrahippocampal injection of 8-pCPT specifically enhanced retrieval of contextual fear memory, whereas Epac2 knockdown by its siRNA abolished this enhancement, indicating the specific role of Epac2 in promoting memory retrieval (Ostroveanu et al., 2010). Intriguingly, this enhancement was observed for a time-limited duration, further suggesting that the memory retrieval mediated by Epac2 is hippocampus-dependent. In AD, Epac signaling is also altered. In postmortem AD patients, Epac1 and Epac2 have distinct expression patterns in different brain regions: in the frontal cortex, the mRNA expression of Epac2 is decreased whereas Epac1 mRNA is increased (Zhang et al., 2024). In contrast, in the hippocampus, the protein expression of Epac2 is decreased while the Epac1 protein is not altered (Zhang et al., 2023a). The specific decrease of Epac2, but not Epac1, suggests that the loss of Epac2 may contribute to the memory deficits in AD. Intriguingly, the protein expression of A-kinase anchoring protein 5 (AKAP5), the anchoring protein of both Epac and PKA, is increased in AD patients but not in J20 mice, probably resulting from PKA elevation in tau hyperphosphorylation. In addition, the downstream signaling of Epac2, Akt/protein kinase B pathway, but not p42/p44 mitogen-activated protein kinase (MAPK) pathway, is reduced in AD patients and J20 mice. The downregulation of Epac2 signaling might suggest impaired synaptic plasticity and memory retrieval in AD. Due to their comparable roles in memory and synaptic plasticity, and evidence that Epac regulates GluA3 activation, we propose a novel crosstalk model of synaptic plasticity between GluA3-containing AMPARs and Epac2 (Fernandes et al., 2015; Gutierrez-Castellanos et al., 2017; Renner et al., 2017; Zhang et al., 2024; Figure 1). During basal neuronal activities, GluA3-containig AMPARs are trafficked in or out of postsynaptic areas via transient activation of Ras-p42/p44 MAPK or Rap-p38 MAPK pathways, respectively. Subsequently, the GluA3-containig AMPARs are anchored together with Epac2 via PSD95 and AKAP5 in postsynaptic spines and stay in low-conductance states. Internal cues (e.g., emotions) or external cues (e.g., contexts) trigger the release of neurotransmitters such as norepinephrine, leading to activation of β-adrenergic receptors and rapid intracellular cAMP elevation. Subsequently, Epac2 is activated and initiates its downstream Ras-p42/p44 MAPK, leading to the activation of the GluA3-containing AMPARs. This Epac2 mediated event promotes neurotransmission, LTP, and potentially memory function. In AD, as reviewed, the expression and/or function of GluA3-containing AMPARs and Epac2 are decreased, thereby impairing synaptic plasticity and memory. Hence, we propose that pharmacological activation of Epac2 might counteract this synaptic and memory impairment. Furthermore, an additional advantage of pharmacological targeting of Epac2 to the brain is the minimal side effects on other tissues. Until recently, Schwede and colleagues reported on a novel Epac2 activator S220 based on the structure of the cAMP binding pocket of Epac2 (reviewed in Zhang et al., 2024). Compared to the Epac activator 8-pCPT in a previous study (Zhang et al., 2024), S220 exhibits 10 times the efficacy, 35 times the potency, and much higher specificity exhibiting almost no effect on Epac1 or PKA. Furthermore, S223-AM, a membrane preamble analog of S220, bears the potential to cross blood–brain barrier and represents another promising candidate as an Epac2 activator. A recent study reported that approximately 20% elderly population exhibits pathologies in the preclinical stage of AD (Parnetti et al., 2019). Therefore, treatment by using these specific Epac2 activators could potentially alleviate the early deficits in memory retrieval associated with preclinical AD. We thank the Department of Molecular Pharmacology and Molecular Neurobiology for the support. This work was supported by Alzheimer Nederland grant [WE.03-2019-05] (to MS). C-Editors: Zhao M, Liu WJ, Qiu Y; T-Editor: Jia Y