Excessive excitotoxicity has been reported to contribute to the neurodegenerative processes [32]. Glutamate and associated excitatory amino acids can impair neurons by inducing apoptosis in vitro and in vivo [33, 34]. As an analogue of glutamate-associated amino acids, KA has recently been reported to be associated with AD, especially with the production and deposition of Aβ [1] and hyperphosphorylation of tau [35, 2]. In this study, we extended our previous research to test the effects of a new antiepileptic drug, LEV, on memory impairment and neuropathology in APP23/MAPT mice. We found that LEV alleviated the behavioral deficits by reducing the burden of SPs and the phosphorylation of tau. In addition, LEV enhanced the degradation and clearance of Aβ through activating autophagy and enhancing Aβ transport across the blood-brain barrier (BBB) in APP23/MAPT mice. More interestingly, LEV suppressed neuroinflammation by deactivating inflammasomes and protected neurons from dystrophy and neuronal loss by suppressing KA-induced apoptosis.
As an antiepileptic drug, LEV was found to improve memory decline in APP23/MAPT mice. We demonstrated that chronic administration of low concentrations of LEV before training partially reversed memory impairment in aged APP23/MAPT mice. Our results are in agreement with those of recent studies showing that LEV treatment reversed abnormalities in the expression of neuronal activity-related proteins that reflect hippocampal remodeling and correlate well with cognitive deficits in hAPPJ20 mice [19, 36]. In the MWM test, hippocampal spatial memory declines in aged rats were rescued by chronic infusion or single injection of LEV and sodium valproate before training [37]. Using high-resolution fMRI techniques, LEV treatment has also been shown to normalize the activity of hippocampal CA3/DG, which improved memory performance in patients with amnestic mild cognitive impairment [22].
On the basis of our findings, we suggest that excess neural activity may be responsible for disrupting the processes required for encoding new information during the acquisition or early phase of memory consolidation [38, 39]. In addition, hippocampal hyperactivation has been hypothesized as an aberrant condition leading to AD [40]. This hypothesis has been confirmed in the brains of APP Tg mice that exhibit memory deficits [19]. In accordance with this, AD patients also have a higher incidence of seizures than those in the non-demented population [41], which may contribute to at least some of the cognitive impairments associated with this disease [42]. These observations also establish the relationship between epileptic seizures and AD, providing the evidential basis for the current study.
What mechanisms may underlie the ability of LEV to rescue memory impairment in APP23/MAPT mice? Excessive loading of Aβ has been suggested to cause aberrant excitatory neuronal activity, which results in profound remodeling of neuronal circuits through compensatory inhibitory mechanisms to counteract overexcitation in APP mice [19]. Consistent with these changes, equivalent proportions of abnormally hyperactive and hypoactive neurons are found in the cerebral cortex of APP Tg mice [43]. Based on previous studies, it is reasonable to suspect that chronic infusion of LEV rescues memory by inhibiting the production and deposition of Aβ in AD animals. As expected, our data revealed that high concentrations of LEV inhibited the production and deposition of Aβ, but did not affect memory deficits in APP23/MAPT mice (Figs. 1, 2). LEV together with TPM has also been reported to reduce Aβ generation by inhibiting the activity of γ-secretase [44], which is consistent with our findings. The underlying mechanism involves the activation of AMPK/Akt/GSK3β pathways in vitro and in vivo [44], leading to the activation of mTOR signaling, autophagy, and Aβ clearance [45]. Since the modulation of autophagy has been regarded as a new therapeutic target for the treatment of AD [46–48], our data extended the previous studies to find that LEV treatment enhanced autophagy, leading to the degradation of Aβ (Fig. 3A). Moreover, LEV promoted the clearance of Aβ in the brains of AD animals (Fig. 3B). In line with our findings, lack of clearance of Aβ is a major cause of sporadic AD [49]. For the reason, it might be caused by the impaired transport of Aβ from brain to the blood in AD animals [50]. With regard to the mechanism, LRP1 and RAGE are the most important receptors for transporting Aβ across the BBB [51]. Taken together, our results suggested that LEV plays a role in increasing the clearance of Aβ by upregulating LRP1 (Fig. 3B).
Tau hyperphosphorylation—another hallmark of AD pathology—changes the function of tau from assembly promotion to assembly disruption, leading to impaired synaptic plasticity and axonal transport [52]. Tau phosphorylation can also be induced by the production and deposition of Aβ [53, 54]. We found that LEV decreased the phosphorylation of tau in CDK5- and GSK3α/β-dependent mechanisms (Fig. 4). Indeed, tau phosphorylation has been reported to be mediated by CDK5 and GSK3α/β [55, 56]. Importantly, cognitive impairment was alleviated by low, but not high concentrations of LEV (Fig. 1). Based on these findings, we can infer that low concentrations of LEV improve memory deficits by inhibiting tau phosphorylation.
Excessive Aβ production and tau hyperphosphorylation give rise to neuroinflammation and neuronal loss in AD pathology [57]. Inflammasomes have been shown to contribute to various disorders in the central nervous system by causing neuroinflammation [31]. Extracellular accumulation of Aβ in SPs in AD brains is a principal event in AD pathogenesis [58]. Deposition of Aβ peptide initiates inflammasome activity in the microglia [59]. Moreover, the activation of inflammasomes causes impairment in AD pathogenesis—through Aβ deposition and loss of spatial memory—by mediating a harmful chronic inflammatory response [31]. It is important to note that NLRP3 activation in the brain is restricted to plaque-associated microglia [31], suggesting that microglial activation of the NLRP3 inflammasome is pivotal for AD pathogenesis. Furthermore, the correlation of AD pathogenesis with local neuroinflammation is already established. As an important component of inflammasome, IL-1β has the ability to induce the phosphorylation of tau [60], leading to the compromised learning and memory of AD animals [61, 62]. By blocking IL-1β, the pathogenesis of AD was improved in the animal models [63]. In agreement with these reports, the current study further shows that LEV treatment deactivates inflammasomes, thus suppressing the activity of glial cells.
Synaptic dystrophy and neuronal loss are known to be significantly increased in AD mice [64, 65]. Our data demonstrate that a low concentration of LEV can protect neurons from synaptic dystrophy and neuronal loss, whereas high concentrations show neurotoxicity, leading to the apoptosis in neurons (Fig. 6B). LEV administration after hypoxia also reduces neuronal apoptosis in a neonatal rat model of hypoxic ischemic brain injury [66]. In addition, a study showed that LEV conferred neuroprotective effects against focal cerebral ischemia-reperfusion injury in mice [67]. Unfortunately, the authors of that study did not measure the effects of different concentrations of LEV on neuronal apoptosis. In this study, we extended the previous investigations and found that high concentrations of LEV induced the apoptosis in neurons.
Our findings therefore provide the experimental basis for selecting appropriate concentrations of LEV for future clinical trials in treating AD.