Research Article: Endoplasmic reticulum proteostasis impairment in aging

Date Published: April 23, 2017

Publisher: John Wiley and Sons Inc.

Author(s): Gabriela Martínez, Claudia Duran‐Aniotz, Felipe Cabral‐Miranda, Juan P. Vivar, Claudio Hetz.


Perturbed neuronal proteostasis is a salient feature shared by both aging and protein misfolding disorders. The proteostasis network controls the health of the proteome by integrating pathways involved in protein synthesis, folding, trafficking, secretion, and their degradation. A reduction in the buffering capacity of the proteostasis network during aging may increase the risk to undergo neurodegeneration by enhancing the accumulation of misfolded proteins. As almost one‐third of the proteome is synthetized at the endoplasmic reticulum (ER), maintenance of its proper function is fundamental to sustain neuronal function. In fact, ER stress is a common feature of most neurodegenerative diseases. The unfolded protein response (UPR) operates as central player to maintain ER homeostasis or the induction of cell death of chronically damaged cells. Here, we discuss recent evidence placing ER stress as a driver of brain aging, and the emerging impact of neuronal UPR in controlling global proteostasis at the whole organismal level. Finally, we discuss possible therapeutic interventions to improve proteostasis and prevent pathological brain aging.

Partial Text

As the world population gets older, dementia emerges as a major public health issue worldwide, particularly in middle‐ and middle‐to‐high‐income countries. The prevalence of dementia increases exponentially with age, affecting 5–10% of people over 65, and about 50% of people over 85. In 2011, dementia was estimated to affect 35.6 million people around the world, and it is expected to reach about 135 million by 2050 (Brayne, 2007; World Health Organization and Alzheimer’s Disease International 2012). Reduced cognitive function is a common trait present in elderly individuals, which correlates with substantial alterations to functional synapses and normal neuronal physiology at the cellular and molecular level (Leal & Yassa, 2015). Accordingly, a significant percentage of aged individuals will manifest some sort of dementia in the form of a collection of neurodegenerative diseases, transposing the line between normal aging (healthspan) to pathological brain aging (Brayne, 2007). Recently, several interconnected processes have been defined as the hallmarks of aging, where substantial alterations to cellular proteostasis is proposed as one of the major pillars of aging (Lopez‐Otin et al., 2013; Kennedy et al., 2014).

The ER is the main site for the synthesis and folding of around one‐third of the total proteome of a cell (Braakman & Bulleid, 2011). Considered a key component of the proteostasis network, ER‐located proteins regulate folding and quality control through the activity of multiple chaperones, foldases, and co‐factors that assist the folding of nascent proteins as well as degradation pathways, thus preventing abnormal protein aggregation and resultant proteotoxicity (Ellgaard & Helenius, 2003; Kourtis & Tavernarakis, 2011; Hetz et al., 2015). Stressful stimuli such as hypoxia (Badiola et al., 2011), nutrient deprivation (Szegezdi et al., 2006), increased protein oxidation (Santos et al., 2009), and disturbance of the secretory pathway (Badr et al., 2007) may lead to an excessive accumulation of misfolded proteins at the ER, a process termed ER stress (Walter & Ron, 2011; Hetz, 2012). To cope with ER stress, a highly conserved signaling pathway is engaged known as the UPR (Wang & Kaufman, 2016). The UPR is initiated by the activation of at least three types of stress sensors including inositol‐requiring enzyme‐1 (IRE1), PKR‐like ER kinase (PERK), and activating transcription factor 6 (ATF6). IRE1 catalyzes the unconventional splicing of the mRNA encoding X‐box binding protein‐1 (XBP1) (Yoshida et al., 2001; Calfon et al., 2002; Lee et al., 2002), resulting in the expression of an active transcription factor called XBP1s that controls the expression of a cluster of genes related to folding and quality control mechanisms (Hetz et al., 2011). Additionally, IRE1 also degrades several mRNAs, ribosomal RNAs, and microRNAs through a process known as regulated IRE1‐dependent decay (RIDD), having an impact on different processes including inflammation and apoptosis (Maurel et al., 2014). IRE1 also engages distinct stress pathways, including JNK and NF‐κB, through the binding of adapter proteins (Hetz et al., 2015). Activation of PERK leads to the phosphorylation of the eukaryotic translation initiation factor 2 alpha (eIF2α), which in turn inhibits translation decreasing protein influx into the ER (Ron & Walter, 2007). Paradoxically, some mRNAs, including activating transcription factor 4 (ATF4), are differentially translated, leading to the upregulation of genes related to redox homeostasis, amino acid metabolism, autophagy, and apoptosis control (Harding et al., 2003; Ye & Koumenis, 2009; B’Chir et al., 2014). Moreover, PERK activation has been shown to modulate the activity of nuclear factor erythroid 2‐related factor 2 (NRF2) and forkhead box O (FOXO), linking this pathway to the antioxidant response, insulin responsiveness, and autophagy (Chevet et al., 2015). Under ER stress conditions, ATF6 translocates to the Golgi apparatus where it is cleaved releasing ATF6f, a cytosolic active form. ATF6f exerts its action at the nuclear level as a transcription factor regulating genes associated with ERAD, in addition to enhancing XBP1 mRNA transcription (Yamamoto et al., 2007). Importantly, the control of gene expression by the UPR depends on the cellular context and the stimuli considering that UPR transcription factors can interact with other proteins to drive specific responses, in addition to be regulated by several post‐translational modifications (Hetz et al., 2015). Under chronic ER stress, the UPR triggers apoptosis through different mechanisms that involve the upregulation of CHOP, the induction of oxidative stress, exacerbated RIDD, upregulation of pro‐apoptotic components of the BCL‐2 family, among other mechanisms (Tabas & Ron, 2011; Urra et al., 2013). Thus, under conditions of ER stress, the UPR reprograms the cell toward adaptation, sustaining cell function or the engagement of cell death programs to eliminate irreversibly damaged cells.

During aging, organisms gradually accumulate intracellular aggregates composed by misfolded proteins, an event that is associated with a prominent decline in the buffering capacity of the proteostasis network and a consequent decrease in tissue and cellular function (Fig. 1) (Taylor & Dillin, 2011; Triplett et al., 2015). Several studies in model organisms have uncovered the significance of UPR signaling to the aging process, associated with protection against proteotoxicity (Ben‐Zvi et al., 2009; Labunskyy et al., 2014). For example, caloric restriction has been used as a major strategy to prevent the adverse effects of aging on healthspan (Riera & Dillin, 2015). In yeast, this intervention correlates with increased expression of HAC1, the functional homologue of XBP1s (Choi et al., 2013). Remarkably, genetic ablation of HAC1 abrogates the lifespan extension conferred by caloric restriction (Choi et al., 2013). Other studies indicated that the deletion of distinct UPR‐target genes impact replicative lifespan in yeast, a process dependent on the Ire1p/HAC1 axis (Labunskyy et al., 2014). Furthermore, genetic modifications to improve the activity of the UPR enhance replicative lifespan in Saccharomyces cerevisiae (Cui et al., 2015).

A novel concept is emerging based on research using fly and worm models of aging, indicating that the ER proteostasis network promotes health and lifespan through cell‐nonautonomous mechanisms, impacting whole organismal proteostasis (Mardones et al., 2015). Studies in Caenorhabditis elegans revealed that besides its importance in individual cells, the UPR acts as a key player in modulating global organism adaptability to stress during aging by integrating information at the level of the nervous system (Martinez et al., 2016a). Accordingly, UPR can be activated on a cell‐nonautonomous manner (Taylor & Dillin, 2013). The ectopic expression of XBP1s in neurons is able to engage a distal UPR activation in the intestine, thus increasing stress resistance and longevity in Caenorhabditis elegans (Taylor & Dillin, 2013). These results suggest that the nervous system may act as a central integrator and adjustor of global proteostasis, with possible major distal effects in the intestine. Importantly, other studies previously demonstrated that neuronal UPR regulates the innate immunity in the gut on a cell‐nonautonomous manner (Martinez & Hetz, 2012; Sun et al., 2012; Aballay, 2013). Chromatin remodeling factors in neurons can also engage ER stress responses through a cell‐nonautonomous mechanism (Kozlowski et al., 2014). Thus, accumulating evidence supports the idea that when an organism is exposed to environmental or pathogenic challenges, the ability of the nervous system to integrate these signals through the activation of the UPR favors the maintenance of homeostasis in various peripheral organs (Mardones et al., 2015). A similar model has been proposed for the heat‐shock response by Morimoto’s group, where HSF‐1 in neurons regulates global responses to aging in the gut (Morley & Morimoto, 2004; van Oosten‐Hawle & Morimoto, 2014a; Douglas et al., 2015). Importantly, cell‐nonautonomous control of aging‐related pathways has been extensively described in different model organisms mediated by distinct signaling molecular mediators (Taylor et al., 2014; Leiser et al., 2015; Schinzel & Dillin, 2015). A recent study indicated that PERK is activated in intestinal stem cells by JAK/Stat signaling in response to ER stress in neighboring cells, regulating intestinal homeostasis and lifespan in flies (Wang et al., 2015). A cell‐nonautonomous mechanism has been also described in mammals, where overexpression of XBP1s in the hypothalamus modulates global energy balance through the propagation of signals to the liver and adipose tissue to adjust energy metabolism (Williams et al., 2014). Furthermore, the concept of ‘transcellular chaperone signaling’ was proposed in Caenorhabditis elegans where cells suffering stress from the accumulation of protein aggregates propagate signals to the neighbor tissue to induce adaptive responses and resist further damage (van Oosten‐Hawle & Morimoto, 2014b).

Although proteostasis impairment and ER stress are proposed as one of the hallmarks of aging, most of the studies addressing the contribution of the UPR to mammalian aging rely mostly on correlative data. For example, the expression of the ER chaperones BiP, calnexin, and PDI has been reported to be downregulated in the hippocampus of aged rats while pro‐apoptotic mediators such as CHOP and the ER‐located caspase‐12 are increased (Paz Gavilan et al., 2006; Gavilan et al., 2009). Additionally, another report demonstrated that levels of both ATF4 and BiP are decreased in this tissue (Hussain & Ramaiah, 2007). In aged brains, basal expression of PERK, GADD34, and total eIF2α is augmented contrasting with reduced levels of eIF2α phosphorylation (Hussain & Ramaiah, 2007). Moreover, young mice under sleep deprivation showed an increase in BiP and eIF2α phosphorylation, which was not observed in aged mice, but there was an upregulation of GADD34, CHOP, and caspase‐12 (Naidoo et al., 2008). Another study demonstrated similar observations in the pancreas of mice (Naidoo et al., 2014). Additionally, aged macrophages exhibit diminished IRE1 activation and increased susceptibility to ER stress‐dependent apoptosis (Song et al., 2013). These findings suggest that the ability to engage the UPR may be disrupted during aging; however, the functional significance of these observations is unknown (Fig. 2).

Abnormal aggregation of specific proteins is a hallmark of age‐related neurodegenerative diseases. Increasing evidence indicates that despite the fact that PMD‐related proteins distribute in different subcellular locations and have distinct binding partners, a common pathological consequence of their accumulation is the occurrence of ER stress. This mechanistic convergence is explained by the observation that disease‐related proteins actually disrupt the function of one or more components of the proteostasis network, highlighting the inhibition of ERAD, altered vesicle trafficking between the ER and Golgi, perturbed ER calcium homeostasis, autophagy dysregulation, and abnormal interactions with ER chaperones (Hetz & Mollereau, 2014; Vidal et al., 2014; Kaushik & Cuervo, 2015).

Imbalance of neuronal proteostasis is one of the pathological hallmarks of aging, and understanding its molecular defects will contribute to develop strategies to intervene age‐associated disorders. Because the nervous system is highly dynamic and plastic, the manifestation of clinical features in patients arises very late, after severe damage has already occurred. Likely, it is predicted that the development of strategies to improve the quality of the aging process will substantially reduce the probability to undergo PMDs. Despite the fact that proteostasis is composed of a complex network of individual interconnected signaling pathways, recent findings suggest that the maintenance of ER physiology is a prominent molecular target to prevent age‐related diseases affecting the nervous system. The involvement of ER stress in the biology of aging is complex as illustrated by most recent advances. The activity of the ER proteostasis network may not only operate as a mechanism to handle abnormal protein aggregation, but it is also proposed as an adjuster of brain function through fine‐tuning synaptic function. Specific neuronal populations are highly vulnerable to perturbations to ER function possibly because their metabolic state depends on the basal activity of the UPR. Furthermore, the UPR may orchestrate repair processes of the nervous system by controlling the expression of neurotrophins such as BDNF, and the regenerative capacity of axons and stem cells pools (Castillo et al., 2015; Martinez et al., 2016b; Onate et al., 2016). Regarding inflammatory reactions, the UPR is known to have important functions in macrophages and dendritic cells by modulating the secretion of pro‐inflammatory cytokines (Bettigole & Glimcher, 2015). In this context, future efforts should address the importance of the UPR to brain inflammation and the activity of astrocytes, microglia, and oligodendrocytes during aging. The fact that the UPR participates in the adjustment of energy and lipid metabolism, an additional layer of complexity, could be also explored to link the UPR with brain aging. Finally, the discovery of cell‐nonautonomous UPR responses and its relation to healthspan control adds a new concept as ER stress‐related signals in the brain may influence the capacity of the whole organism to adapt and cope with ER stress. All those aspects should be considered in future studies aiming to define the relative impact of ER stress on mammalian brain aging and its significance as a risk factor to develop neurodegenerative diseases.

This work was directly funded by FONDAP program 15150012, US Office of Naval Research‐Global (ONR‐G) N62909‐16‐1‐2003, Millennium Institute P09‐015‐F, FONDEF ID16I10223, FONDEF D11E1007, U.S. Air Force Office of Scientific Research FA9550‐16‐1‐0384, CONICYT‐Brazil 441921/2016‐7, FONDECYT no 11160760 (CDA), and FONDECYT no 3150637 (GM). We also thank the support from ALS Therapy Alliance 2014‐F‐059, Muscular Dystrophy Association 382453, Michael J Fox Foundation for Parkinson’s Research—Target Validation grant No 9277, FONDECYT no. 1140549, and ALSRP Therapeutic Idea Award AL150111 (CH).

Authors declare that they have no conflict of interest.