Alzheimer's disease (AD) is the most common neurodegenerative disease causing dementia, pathologically characterized by extracellular deposits of amyloid-beta (Aβ) protein, intracellular neurofibrillary tangle (NFT) composed of hyperphosphorylated forms of the microtubule-associated Tau protein, neuronal loss and neurotransmitter dysfunction (Selkoe 2004). Numerous attempts at developing novel therapeutics targeting AD based on these pathophysiological findings have failed, but acetylcholinesterase inhibitors (AChEIs; donepezil, galamtamine, and rivastigmine) and N-methyl-d-aspartate (NMDA) antagonist, memantine, are currently the only approved agents for the treatment of AD (Farrimond et al. 2012; Gauthier and Molinuevo 2013). In general, AChEIs have been known to possess no ability to halt the progression of the disease, providing only symptomatic efficacy and marginal therapeutic benefits. However, mounting evidence from pre-clinical and clinical studies points toward neuroprotective roles of AChEIs, especially, donepezil (Leyhe et al. 2009; Shen et al. 2010; Min et al. 2012). Moreover, even in AD patients, neuroprotective roles of AChEIs have been demonstrated through retardation of the progression of brain atrophy, indicating plausible disease modifying effect of donepezil by attenuating neuronal death (Hashimoto et al. 2005). A thorough study of the neuroprotective potential of donepezil, therefore, is appropriate at this juncture.
In our recent study, neuroprotective effects of donepezil against Aβ-induced neuronal toxicity were mediated through inhibition of GSK-3β activity via the activation of Akt in a cholinergic-dependent manner (Noh et al. 2009). One shortcoming, however, was that the effects of donepezil on the protein phosphatase 2A (PP2A) activity were not evaluated.
Glycogen synthase kinase-3β (GSK-3β) is one of the major kinases responsible for Tau hyperphosphorylation (Kosik 1992). Evidence suggesting the presence of mutual regulatory systems between kinases including GSK-3β or Akt and phosphatases such as PP2A (Wang et al. 2007; Chen et al. 2010; Hashiguchi and Hashiguchi 2013) prompted us to evaluate the effects of Aβ on cell viability and regulation of PP2A, GSK-3β activity, and the neuroprotective effects of donepezil on them and the expression level of tau phosphorylation against Aβ-induced neuronal injury. Numerous studies have shown that tau is a downstream target of Aβ, which increases GSK-3β activity resulting in hyperphosphorylation of tau (Koh et al. 2007; Noh et al. 2009; Lahmy et al. 2013). On the other hand, it is reported that Aβ induces tau phosphorylation by decreasing PP2A activity (Park et al. 2012).
On the basis of observations suggesting that Aβ induces tau hyperphosphorylation by decreasing PP2A activity or increasing GSK-3β activity, we first tried to evaluate the effects of Aβ42 on cell viability, and the time course of PP2A and GSK-3β activity. And then, to explore whether donepezil in Aβ-induced toxic condition has neuroprotective role by activating PP2A, direct measurement of PP2A activity and quantitative analysis of methyl/demethylated form of PP2A and expression levels of tau phosphorylation were analyzed with/without depleting PP2A activity by a selective antagonist of PP2A, okadaic acid (OA), or PP2A siRNA transfection.
According to recent evidence, the neuroprotective roles of AChEIs are mediated through the stimulation of α7-nAChRs and PI3K/Akt pathways against Aβ toxicity (Takada et al. 2003; Noh et al. 2009) or glutamate-induced excitotoxicity (Akaike et al. 2010; Shen et al. 2010). Therefore, lastly, we evaluated whether donepezil-mediated nAChRs activation also has a role in regulating PP2A activity in the same model.
In this study, we demonstrated that neuroprotective effects of donepezil against Aβ42-induced model of neurotoxicity were mediated through the activation of PP2A, although additional mechanisms including regulation of GSK-3β and nAChRs activities may contribute to its neuroprotective roles.
- Top of page
- Materials and methods
Donepezil (R,S-1-benzyl-4-[(5,6-dimethoxy-1-indanon)-2-yl] methylpiperidine hydrochloride) (E2020), a potent AChE inhibitor (Sugimoto et al. 1995), has been suggested as an effective symptomatic therapeutic agent for AD (Takada et al. 2003; Arias et al. 2005; Takada-Takatori et al. 2006, 2008). There is evidence that donepezil exerts its neuroprotective effect by activating PI3K-Akt in various cell types related to cognitive function (Arias et al. 2005; Takada-Takatori et al. 2006) by inhibiting GSK-3 (Koh et al. 2007). However, the exact effects of donepezil on PP2A activity in β42-induced neuronal toxicity have yet to be evaluated.
The main purpose of this study was to demonstrate whether the neuroprotective effects of donepezil against Aβ42-induced neurotoxicity could be mediated by enhancing PP2A activity and to evaluate its effect on the level of tau phosphorylation.
In this study, 20 μM Aβ42 treatment sufficiently suppressed PP2A activity in primary cultured cortical neurons, and 10 μM donepezil pre-treatment suitably activated PP2A enzymes for our subsequent studies. And both okadaic acid and PP2A siRNA transfection abolished donepezil's PP2A activating capacity and tau de-phosphorylating effect. And PP2A activity to be enhanced by donepezil was also partially blocked by nAChRs antagonist. These results suggest that the neuroprotective effects of donepezil against Aβ42-induced neurotoxicity are highly dependent on its stimulatory effect on PP2A activity as well as nAChRs.
With the neuroprotective effects of donepezil focused solely on the aspect of PP2A activation, we may overlook important issues raised in this study. First, there is a mutual regulatory system between kinases and phosphatases activity (Zhou et al. 2009). Second, besides its PP2A activating capacity, donepezil has other important functions including the inhibitory effect on GSK-3β activity via the activation of PI3K/Akt pathway, and the modulating effect of nAChRs activity (Noh et al. 2009). Moreover, in given concentrations Aβ42 tended to accelerate neuronal cell death by increasing GSK-3β activity and decreasing PP2A activity as shown in Fig. 1c. However, in conditions such as Aβ42 toxic insult in neuronal cells, normal compensatory or counteracting mutual regulations of these kinases/phosphatases were found to be disturbed (Fig. 1c). If it is a physiological condition, or if knockdown is induced in one part of the kinase/phosphatase, activity in the counter part of kinase/phosphatase would be appropriately regulated toward the direction of cell survival or maintenance of this balancing system (Zhou et al. 2009).
The most interesting finding was that even in the 20 μM Aβ42-induced toxic state, aligned with the PP2A-depleted condition via OA or PP2A siRNA transfection, donepezil still exhibited significant effect not only on cellular viability, but also on the level of phosphorylated tau at the S-396 site. Thus, effects of donepezil in this condition were presented as additional increment of neuronal viability and tau de-phosphorylation at S-396 site (Figs 4, 5). We therefore assumed that other neuroprotective mechanisms of donepezil, including its inhibitory role in GSK-3β activity or stimulatory effect on nAChRs, might be involved in these processes.
PP2A is one of the major serine/threonine phosphatases that plays important roles in many biological processes (Janssens and Goris 2001; Lechward et al. 2001). It can dephosphorylate tau at multiple sites (Gong et al. 2000; Liu et al. 2005). PP2A activity is regulated by post-translational modification via phosphorylation (Tyr307) and methylation (Leu309) of its catalytic C subunit (Xing et al. 2008; Eichhorn et al. 2009). Methylation of the C subunit is performed by cytoplasmic leucine carboxyl methyltransferase-1 (LCMT-1), and demethylation is made by nuclear phosphatase methyltransferase-1 (PME-1) (Longin et al. 2004). Decreased methylation of PP2A has been proposed as a link between the elevated plasma homocysteine (Hcy) levels and tau hyperphosphorylation observed in AD (Zhang et al. 2008). Importantly, PP2A activity is down-regulated in brain tissue from AD patients (Gong et al. 1994). The specific inhibitor of serine/threonine PP1 and PP2A, OA, can induce tau hyperphosphorylation and neuronal cell death (Wang et al. 2001; Vale and Botana 2008). In particular, a previous study demonstrated that PP2A regulates an amyloid-beta-induced apoptosis (Yin et al. 2006). Another study showed that the caspase cleavage of the amyloid precursor protein induces tau phosphorylation by decreasing PP2A activity (Park et al. 2012). PP2A has also been implicated in the phosphorylation of tau and aggregation involved in aging and AD (Liu et al. 2005; Wang et al. 2007). Previous studies have shown that PP2A site-specifically dephosphorylates tau in vitro (Qian et al. 2010) and that Aβ induced tau phosphorylation (Koh et al. 2007; Noh et al. 2009; Lahmy et al. 2013) by decreasing PP2A activity or increasing GSK-3 activity (Park et al. 2012). The data presented here demonstrate that exposure to Aβ42 decreases PP2A activity. These results are consistent with the observation that Aβ peptides decrease PP2A activity in endothelial cells (Hsu et al. 2007) and cleavage of the amyloid precursor protein (APP) at the Asp664 residue-induced tau phosphorylation by decreasing PP2A activity (Park et al. 2012).
Recently, the role of PP2A in the pathogenic mechanisms of AD has been emphasized (Sontag et al. 2007). PP2A links Hcy metabolism with regulation of tau and amyloid precursor protein (Liu and Wang 2009). We observed that PP2A inhibitor (OA) treatment increased the level of tau phosphorylation at the S-369 site, which is consistent with previous findings (Lim et al. 2010; Akasofu et al. 2003; Plattner et al. 2006; Zhou et al. 2008; Liu and Wang 2009). Our results also confirm that the effects of donepezil through PP2A activation are inhibited by PP2A inhibitor (OA) (Fig. 4b). As shown in our previous report, donepezil decreases phosphorylated tau (Ser396, Thr231, Ser199) in models of Aβ42-induced neurotoxicity (Noh et al. 2009). Also, effects of donepezil were associated with an increase in PP2A activity, correlated with the level of tau phosphorylation. Consequently, we tried to inhibit PP2Ac expression by means of RNA interference. The effects of donepezil through PP2A activation were also inhibited by PP2A-siRNA (Fig. 5c).
On the other hand, nAChRs exerted neuroprotective effects against a variety of toxicants (Picciotto and Zoli 2008; Del Barrio et al. 2011). Nicotinic agonists have been reported to be effective against Aβ-induced toxicity (Arias et al. 2005). Our previous results suggested that activation of nAChRs plays an important role in the protective effects of donepezil against Aβ-induced toxicity with the PI3K/Akt signaling pathway (Noh et al. 2009). Examining whether PP2A activity was regulated by nAChRs stimulation, we found that PP2A activity was regulated via nAChRs activity. However, even in the depleted activity of nAChRs induced by mecamylamine, a non-specific antagonist of nAChRs, donepezil pre-treatment demonstrated to have an additional residual PP2A activity. This finding indicates that the effects of donepezil on PP2A activity might be partially related to nAChRs stimulation, with the main part being mediated via a direct stimulation of PP2A or other mediating pathway including a mutual regulatory system of kinase/phosphatase.