Integrated Stress Response: Connecting ApoE4 to Memory Impairment in Alzheimer's Disease

There is much evidence to suggest that control of gene expression and protein translation contribute to memory formation and consolidation in the hippocampus (Costa-Mattioli et al., 2007; Cho et al., 2015). Therefore, the intricate coordination of mechanisms that couple extracellular signals to translational regulation is essential for memory stability. Disruption of such pathways might contribute to memory decline in cognitive disorders such as Alzheimer's disease (AD).

Most cases of AD are sporadic, meaning they are not tightly linked to gene mutations. Nonetheless, several genetic risk factors for late-onset, sporadic AD have been reported. The strongest genetic risk factor report is possession of the apolipoprotein E ε4 allele (ApoE4). Compared to the ApoE2 or ApoE3 alleles, ApoE4 appears to promote metabolic impairment, amyloid-β (Aβ) production, and tau aggregation, which are hallmarks of AD (Liu et al., 2013). The debate about the culprits of AD pathology is still open, but mounting evidence indicates that neuronal stress-related mechanisms (initiated by either Aβ, tau, or ApoE4) take place in AD brains to induce synapse dysfunction, memory loss, and neurodegeneration (Lourenco et al., 2015).

Cellular stress signals converge to increase phosphorylation of eukaryotic translation initiation factor 2α (eIF2α-P), resulting in transcriptional changes and protein synthesis attenuation, in a set of mechanisms collectively known as the integrated stress response (ISR). eIF2α can be phosphorylated by four different kinases, namely double-stranded RNA-dependent protein kinase (PKR), PKR-like endoplasmic reticulum kinase (PERK), general control nonderepressible 2 (GCN2), and heme-regulated kinase (HRI). eIF2α-P then acts by blocking its partner, eIF2B, in the translation initiation complex (Buffington et al., 2014). Recent evidence has demonstrated significant roles for ISR elements, including PKR and eIF2α-P, in suppressing memory, in either physiological or AD contexts (for review, see Buffington et al., 2014; Lourenco et al., 2015).

In a recent paper in The Journal of Neuroscience, Segev, Barrera et al. (2015) addressed whether PKR could mediate the deleterious impact of ApoE4 on memory. They used a mouse model harboring two genomic copies of either human ApoE3 or ApoE4 alleles in place of the murine homologs. They initially found that ApoE4 mice exhibited impaired contextual memory in the fear conditioning paradigm, in agreement with their previous observations (Segev et al., 2013). They further demonstrated that young (∼3-month-old) ApoE4 mice had higher hippocampal levels of mRNA encoding activating transcription factor 4 (ATF4) than age-matched ApoE3 or wild-type mice (Segev, Barrera et al., 2015).

This increase in ATF4 levels is interesting for several reasons. ATF4 mRNA translation is selectively induced when eIF2α-P increases, despite general translational attenuation, and this drives ISR-related transcriptional reprogramming (Buffington et al., 2014). Recently, ATF4 was described as essential for synapse development and morphogenesis (Liu et al., 2014), and hippocampal silencing of ATF4 impaired synapse plasticity and spatial memory formation in mice (Pasini et al., 2015). However, ATF4 overexpression has been shown to mediate oxidative stress-induced cell death (Lange et al., 2008) and alter expression of several ISR factors, including chaperones and the pro-apoptotic transcription factor C/EBP-homologous protein (CHOP) in neurons (Galehdar et al., 2010). Earlier reports have also indicated that ATF4 could negatively impact memory, as it was shown to oppose the actions of the cAMP response element binding (CREB) transcription factor in the hippocampus (Chen et al., 2003). Thus, fine tuning of ATF4 signaling appears essential for proper synapse and cognitive function. Furthermore, increased ATF4 levels have been observed in the brains of both AD patients (Baleriola et al., 2014) and mouse models (Ma et al., 2013). A recent study by Baleriola and colleagues (2014) identified ATF4 as a retrograde neurodegenerative signal that propagates from neurons exposed to toxic Aβ oligomers (AβOs), known for their central role in AD. Hence, ATF4 has the potential to explain, at least in part, how AD pathology spreads throughout brain regions.

Segev, Barrera et al. (2015) next demonstrated that PKR inhibition alleviates memory impairment and hippocampal ATF4 upregulation in ApoE4 mice. This is consistent with previous reports that blocking eIF2α kinases is effective in impeding AD-related memory impairment in different animal models (Lourenco et al., 2013; Ma et al., 2013) and further extends the notion that metabolic stress and ISR are integral components of AD pathogenesis.

Neuroinflammation has been linked to AD pathogenesis, and some evidence suggests that inflammatory mechanisms drive synapse and cognitive impairment through ISR and metabolic stress (Lourenco et al., 2015). Human ApoE4 carriers have elevated plasma levels of inflammatory markers (Ringman et al., 2012), and ApoE4 has been shown to exacerbate central inflammatory responses that coincide with decreased levels of synaptic markers in mice (Maezawa et al., 2006; Tai et al., 2015). A similar phenomenon is thought to underlie AβO-induced cognitive dysfunction in AD models (Lourenco et al., 2013). Therefore, current evidence suggests that ApoE4 and AβOs may contribute to a toxic process that includes brain inflammation and ISR and impairs synapse and memory (Fig. 1).

Figure 1.

ISR mediates synaptic and memory impairment triggered by ApoE4 and AβOs. ISR comprises the activation of PKR, GCN2, or PERK, resulting in abnormal eIF2α-P. ApoE4 and AβOs promote an inflammatory process, which likely enhances ISR through PKR-dependent eIF2α-P. PERK and GCN2 may further contribute to the increase eIF2α-P levels. This process favors upregulation of ATF4, whose nuclear actions antagonize CREB activity and facilitate stress-related gene transcription, including CHOP and molecular chaperones. Such transcriptional changes result in defective synaptic plasticity and cognition, and may further stimulate ISR. Aberrant ISR may thus comprise a common ground to explain memory loss in AD, possibly offering novel targets for therapeutic intervention.

The precise mechanisms by which ApoE4 acts to increase AD risk remain to be determined, and inflammation-dependent cellular stress pathways might offer a consistent explanation for ApoE4-linked AD cases. The description of higher ATF4 levels in brains of ApoE4 carriers than in noncarriers, and in AD subjects compared to cognitively healthy controls, offers additional support for this possibility (Baleriola et al., 2014; Segev, Barrera et al., 2015). Still, further investigation is required to establish a causal role for ATF4 in ApoE4-induced memory defects.

In addition to upregulation of ATF4, eIF2α-P-dependent disruption of translation appears to mediate memory impairment in AD (Ma et al., 2013), which is consistent with the notion that protein synthesis is essential for synaptic plasticity and memory consolidation (Buffington et al., 2014). One might thus envision that targeting aberrant translational regulation could offer therapeutic benefit for cognitive decline. In fact, ISRIB, a pharmacological agent that counteracts eIF2α-P signaling through eIF2B binding (Sekine et al., 2015), reduces ATF4-dependent gene expression while sustaining protein synthesis and enhancing memory in mice (Sidrauski et al., 2013). Nonetheless, further efforts in preclinical research and drug development are needed to support the promising possibility that ISR could become an effective target in neurodegenerative disorders.

In conclusion, Segev, Barrera et al. (2015) provide evidence that ISR mediates memory impairment caused by the ApoE4 allele, whose carriers are at higher risk of developing AD. This opens the possibility that modulating ISR could represent a potential preventive strategy against ApoE4-related cognitive impairment. Furthermore, such findings offer a novel perspective in which similar mechanisms may drive memory loss induced by different AD-associated agents.