An update on the unfolded protein response in brain ischemia: Experimental evidence and therapeutic opportunities

Abstract

After ischemic stroke or cardiac arrest, brain ischemia occurs. Currently, no pharmacologic intervention that targets cellular processes has proven effective in improving neurologic outcome in patients after brain ischemia. Recent experimental research has identified the crucial role of proteostasis in survival and recovery of cells after ischemia. In particular, the unfolded protein response (UPR), a key signaling pathway that safeguards cellular proteostasis, is emerging as a promising therapeutic target for brain ischemia. For some time, the UPR has been known to play a critical role in the pathophysiology of brain ischemia; however, only in the recent years has the field grown substantially, largely due to the extensive use of UPR-specific mouse genetic models and the rapidly expanding availability of pharmacologic tools that target the UPR. In this review, we provide a timely update on the progress in our understanding of the UPR in experimental brain ischemia, and discuss the therapeutic implications of targeting the UPR in ischemic stroke and cardiac arrest.

1. Introduction

Almost all cellular processes rely on a functional proteome. Thus, protein homeostasis (proteostasis) is essential for cell survival and organismal health. Cellular proteostasis is rigorously and delicately maintained by a protein quality control network (Wang and Yang, 2019). One major component of the cellular proteome resides in the endoplasmic reticulum (ER), the key organelle for the folding and maturation of membrane and secretory proteins. Disruption of ER proteostasis leads to ER stress, which subsequently activates multiple adaptive response pathways, collectively known as the unfolded protein response (UPR) (Yang and Paschen, 2016). The UPR has 3 major branches (Fig. 1), named after 3 ER-membrane stress sensor proteins: ATF6 (activating transcription factor 6), IRE1 (inositol-requiring enzyme 1), and PERK (protein kinase RNA-like ER kinase). The primary function of the UPR is to safeguard ER function and cellular proteostasis, and thus promote cell survival.

Figure 1. Targeting the UPR in brain ischemia.

The UPR branches are mediated by 3 ER stress membrane sensors: ATF6, IRE1, and PERK. Small molecules that have been used to determine the effects of pharmacologically boosting specific UPR signaling components in brain ischemia are indicated. P, phosphorylation.

The UPR is implicated in many aging- and/or ischemia-related diseases including neurodegenerative diseases, heart ischemia, and ischemic stroke (Hetz and Saxena, 2017; Wang et al., 2018; Yang and Paschen, 2016). Importantly, data have shown that manipulating the UPR branches can enhance memory in aged mice (Martinez et al., 2016), mitigate neurodegenerative disease progression (Hetz and Saxena, 2017), and improve outcome after ischemic injury (Blackwood et al., 2019; Jiang et al., 2017; Li et al., 2021; Shen et al., 2021; Wang et al., 2020; Wang et al., 2021; Wang et al., 2014; Yu et al., 2017). Notably, given the great therapeutic potential of such proteostasis-based strategies in many diseases of high clinical significance, there has been sustained enthusiasm in industry and academia to search for small molecules that target the UPR. Many UPR-specific inhibitors and activators have been identified (Grandjean and Wiseman, 2020; Hetz et al., 2019), constituting a great resource for pre-clinical research in diseases associated with ER stress, such as brain ischemia.

In the ischemic brain, ATP is rapidly depleted, Ca²⁺ homeostasis is disrupted, and upon reperfusion, reactive oxygen species (ROS) are massively produced. All these changes detrimentally impact cellular proteostasis, as under such unfavorable conditions, protein folding, maturation, and degradation processes are undermined, leading to impaired ER function and accumulation of damaged and misfolded/unfolded proteins in brain cells. Indeed, a large body of evidence has demonstrated that brain ischemia/reperfusion causes ER stress and activates the UPR (Wang and Yang, 2019; Yang and Paschen, 2016). Although dissecting the role of the UPR in brain ischemia has been actively pursued for many years, only recently has such knowledge been substantially obtained, largely due to availability of UPR-specific mouse genetic and pharmacologic tools. In this review, we aim to update the progress in our understanding of the UPR in the pathophysiology after experimental brain ischemia, and discuss the therapeutic implications of targeting the UPR in ischemic stroke and cardiac arrest.

2. UPR-selective mouse genetic tools

Recent advances in the field of the UPR in brain ischemia have been driven primarily by the extensive use of UPR-selective mouse genetic tools. Thus here, we briefly summarize main genetically modified mouse lines available to study the UPR in vivo. Two IRE1 isoforms are expressed in mammals – IRE1α and IRE1β – but only IRE1α is an ER stress sensor (hereafter referred to as IRE1). Notably, global deletion of Ire1α results in embryonic death around E13 in mice (Zhang et al., 2005). IRE1 regulates X-box binding protein 1 (Xbp1). Interestingly, Xbp1^−/− mouse embryos also cannot survive past E14.5 (Reimold et al., 2000). These data indicate that the IRE1/XBP1 branch is essential for embryonic development. Therefore, Kaser et al generated a conditional Xbp1 knockout mouse line (Xbp1^f/f) in which Xbp1 exon 2 is floxed (Kaser et al., 2008). In 2013, a conditional and inducible Xbp1s transgenic mice (TRE-XBP1s) became available to study effects of activating the IRE1/XBP1 branch (Deng et al., 2013). Of note, the TRE-XBP1s mouse line is based on a Tet-inducible system, which can be controlled by doxycycline in drinking water.

It has been shown that Perk knockout mutants (Perk^−/−) progressively lose pancreatic cells after birth, and after postnatal week 4, Perk^−/− mice begin to succumb (Harding et al., 2001; Zhang et al., 2002). Thus, Perk floxed (Perk^f/f) mice were developed (Zhang et al., 2002), and have been widely used to study the role of the PERK branch in various cell types.

Lastly, 2 ATF6 isoforms are expressed in mammals – ATF6α and ATF6β. ATF6 α (hereafter referred to as ATF6) is the primary UPR stress sensor (Wu et al., 2007). Global deletion of Atf6α appears to be well tolerated in mice, as these mice develop normally and do not show any overt defects (Wu et al., 2007). In order to genetically activate the ATF6 branch, our group developed a conditional and tamoxifen-inducible sATF6 knock-in mouse line (sATF6-MER) in which sATF6 is fused to a mutated estrogen receptor (MER) (Yu et al., 2017). When exposed to tamoxifen, MER binds to tamoxifen, translocating sATF6-MER from the cytosol to the nucleus where ATF6-dependent genes are then activated. Table 1 summarizes major experimental brain ischemia studies that have used UPR-selective mouse genetic models as well as relevant pharmacologic tools.

Table 1.

Experimental brain ischemia studies on the UPR.

Animal species	Brain ischemia model	Pharmacologic treatment	Animals	Study duration (post brain ischemia)	Main outcomes	References
Mice	tMCAO		Neuron-specific Xbp1s transgenic mice	1 day	Improved neurologic scores and smaller infarct volumes	(Wang et al., 2021)
Mice	tMCAO pMCAO		Neuron-specific Xbp1 knockout mice	1–3 days	Worse neurologic function and larger infarct volumes	(Jiang et al., 2017)
Mice	tMCAO pMCAO	Thiamet-G	C57Bl/6	1–3 days	Improved neurologic function and smaller infarct volumes	(He et al., 2017; Jiang et al., 2017)
Aged mice	pMCAO	Thiamet-G	C57Bl/6	3 days	Improved neurologic scores and smaller infarct volumes	(Jiang et al., 2017)
Aged rats	tMCAO	Thiamet-G	F344	7–28 days	Improved long-term neurologic function	(Wang et al., 2021)
Young and aged mice	PT stroke tMCAO	Glucosamine	C57Bl/6	1–3 days or 21–24 days	Improved neurologic function and survival rate, and smaller infarct volumes	(Gu et al., 2017; Wang et al., 2021)
Rats	tMCAO	Glucosamine	Sprague-Dawley	2 days	Improved neurologic scores and smaller infarct volumes	(Hwang et al., 2010)
Mice	CA		Neuron-specific Xbp1s transgenic mice	1–3 days	Improved neurologic scores	(Li et al., 2021)
Mice	CA		Neuron-specific Xbp1 knockout mice	1–3 days	Worse neurologic function	(Li et al., 2021)
Young and aged mice	CA	Glucosamine	C57Bl/6	1–7 days	Improved neurologic function	(Li et al., 2021)
Mice	tMCAO		Neuron-specific Perk knockout mice	1–3 days or 21 days	Worse short- and long-term neurologic function and larger infarct volumes	(Wang et al., 2020)
Mice	tMCAO	Salubrinal	C57Bl/6	1–3 days	Improved neurologic scores and smaller infarct volumes	(Wang et al., 2020)
Rats	Forebrain ischemia	Salubrinal	Sprague–Dawley	7 days	Decreased hippocampal neuronal death	(Anuncibay-Soto et al., 2016)
Mice	pMCAO		Atf6α−/− mice	3–5 days	Reduced astroglia activation and scaring, and increased neuronal death	(Yoshikawa et al., 2015)
Mice	tMCAO		Neuron-specific sATF6 knock-in mice	1 day	Improved neurologic scores and smaller infarct volumes	(Yu et al., 2017)
Mice	tMCAO	147	C57Bl/6	1 day	Smaller infarct volumes	(Blackwood et al., 2019)
Mice	CA		Neuron-specific sATF6 knock-in mice	3 days	Improved neurologic function and less neuronal death in the hippocampus	(Shen et al., 2021)
Mice	CA	147	C57Bl/6	3 days	Improved neurologic function and survival rate	(Shen et al., 2021)

MCAO: middle cerebral artery occlusion; tMCAO: transient MCAO; pMCAO: permanent MCAO; PT stroke: photothrombotic stroke; CA: cardiac arrest.

3. IRE1 UPR branch

The central function of the IRE1 branch is mediated via unconventional splicing of Xbp1 mRNA. Under ER stress, IRE1 turns into an endonuclease that acts on Xbp1 mRNA, removing its 26-nt fragment to generate spliced Xbp1 mRNA (Xbp1s). Xbp1s is then translated into a 54-kDa protein, XBP1s. XBP1s is a transcription factor that regulates expression of many genes related to protein homeostasis (Hetz et al., 2020). Notably, XBP1s can modulate the hexosamine biosynthetic pathway (HBP) by up-regulating expression of its major enzymes. The HBP produces UDP-GlcNAc, the substrate for the post-translational modification O-GlcNAcylation. Thus, the XBP1s/HBP/O-GlcNAc axis was proposed, and has been demonstrated to function in the brain and heart (Jiang et al., 2017; Wang et al., 2014). Of note, O-GlcNAcylation involves 2 enzymes that add and remove O-GlcNAc from proteins: O-GlcNAc transferase (OGT) and O-GlcNAcase (OGA), respectively.

It is well established that Xbp1s mRNA levels rapidly increase in the brain after ischemic stroke and cardiac arrest (Shen et al., 2018; Yu et al., 2017), indicating activation of the IRE1/XBP1s branch after brain ischemia. Early evidence suggesting the role of this branch in brain ischemia was largely obtained from in vitro studies. Data showed that Xbp1 splicing is induced in hippocampal neurons exposed to oxygen/glucose deprivation (OGD), an in vitro ischemia-like model, and overexpression of Xbp1s suppresses OGD-induced neuronal cell death (Ibuki et al., 2012). However, in vivo evidence to support this protective role was lacking for a long time. Recently, by using both loss-of-function and gain-of-function mouse models, considerable in vivo data on this branch have been generated in the context of stroke and cardiac arrest. Such studies have demonstrated that neuron-specific deletion of Xbp1 leads to worse outcome after transient or permanent ischemic stroke, while neuron-specific expression of Xbp1s improves short-term neurologic function and decreases infarct volumes (Jiang et al., 2017; Wang et al., 2021). Similar results were observed in a cardiac arrest mouse model (Li et al., 2021).

These XBP1-specifc mice were also used to study the XBP1s/HBP/O-GlcNAc axis in the brain. As expected, in Xbp1s transgenic mice, key HBP enzymes are up-regulated, and global O-GlcNAcylation is increased in the brain (Jiang et al., 2017). Interestingly, using a newly developed analytic method, our group further demonstrated that UDP-GlcNAc levels in the brain are closely associated with the XBP1 pathway activity, providing direct evidence of the coupling between XBP1s and the HBP (Wang et al., 2021). Moreover, by pharmacologically boosting O-GlcNAcylation with thiamet-G, a potent OGA inhibitor, worse stroke outcome is partially reversed in Xbp1 knockout mice (Wang et al., 2021). Together, the IRE1/XBP1 UPR is a pro-survival pathway in brain ischemia, and one crucial mechanism that underpins its neuroprotective effects involves the XBP1/HBP/O-GlcNAc axis.

Currently, targeting the IRE1 branch in brain ischemia has focused on the XBP1/HBP/O-GlcNAc axis. To boost this axis, approaches can be designed to intervene the O-GlcNAcylation cycle by targeting 2 enzymes (ie, OGA and OGT), or to increase the substrate UDP-GlcNAc by targeting the HBP. Both approaches have been used in experimental brain ischemia. We and others have demonstrated that thiamet-G treatment to increase O-GlcNAcylation is beneficial in both stroke and cardiac arrest (Gu et al., 2017; He et al., 2017; Jiang et al., 2017; Shen et al., 2018; Wang et al., 2021). For example, mice treated with thiamet-G exhibit smaller infarct volumes and better short-term functional outcome after transient and permanent stroke. Recently, we provided evidence that thiamet-G treatment also improves long-term functional outcome in young mice and aged rats after ischemic stroke (Wang et al., 2021). To increase UDP-GlcNAc, glucosamine has been evaluated. Glucosamine can readily be converted to glucosamine-6-phosphate, and then enter the HBP flux. Indeed, dosing glucosamine increases O-GlcNAcylation in the brain, and exerts beneficial effects in ischemic stroke (both transient and permanent strokes) and cardiac arrest (Gu et al., 2017; Hwang et al., 2010; Li et al., 2021; Wang et al., 2021). For example, glucosamine treatment significantly reduces infarct volumes in rats after transient stroke, and improves short-term functional recovery in both young and aged mice after permanent stroke. Moreover, this treatment can also benefit long-term recovery of neurologic function after transient middle cerebral artery occlusion (MCAO; a widely used stroke model) in young mice (Gu et al., 2017). Collectively, there is compelling evidence to support the notion that increasing O-GlcNAcylation is cytoprotective in brain ischemia.

Due to the lack of tools, pharmacologically targeting the IRE1/XBP1 pathway (not just the downstream O-GlcNAcylation) has not been performed in brain ischemia. However, this approach could be highly interesting, based on the literature, as many studies have shown remarkably positive effects of the IRE1/XBP1 pathway in animal models of various diseases (Casas-Tinto et al., 2011; Cisse et al., 2017). Consequently, the search for small molecules that specifically activate this pathway has attracted much attention. It is important to note that, in addition to splicing Xbp1 mRNA, activated IRE1 can degrade mRNAs via a mechanism called regulated IRE1-dependent decay (RIDD), a process that contributes to apoptosis. Recently, the Wiseman group identified a few IRE1/XBP1 activators that selectively activate IRE1-dependent pro-survival XBP1s signaling without promoting the detrimental RIDD process (Grandjean et al., 2020). One of these activators in particular, IXA4, has been shown to ameliorate amyloid precursor proteins (APP)-mediated toxicity (Grandjean et al., 2020). Thus, it would be intriguing to assess the therapeutic potential of using IXA4 to target the IRE1/XBP1 pathway in brain ischemia.

Current brain ischemia studies for this branch have focused predominantly on neurons. However, when pharmacologically targeting the branch for therapeutic purposes, we need to consider other cell types. For example, the IRE1/XBP1 branch also plays a critical role in glial cells. A recent study found that activation of the IRE1/XBP1s pathway in astrocytes promotes their pathogenic activities in experimental autoimmune encephalomyelitis (EAE), by driving pro-inflammatory responses, eg, upregulation of inflammatory genes and monocyte recruitment (Wheeler et al., 2019). This critical aspect must be clarified in the context of brain ischemia, because neuroinflammation is a critical contributor to ischemia-induced brain damage.

4. PERK UPR branch

When activated, PERK functions as a kinase, and phosphorylates the α subunit of eukaryotic translation initiation factor 2α (eIF2α), which then inhibits the initiation step of translation, resulting in global protein synthesis inhibition (PSI). PERK-induced PSI is essential to the survival of cells under acute ER stress, as it reduces the ER workload. However, if ER stress is prolonged, PERK-induced phosphorylated eIF2α (p-eIF2α) is associated with induction of apoptosis, because p-eIF2α selectively facilitates translation of activating transcription factor 4 (Atf4) mRNA. ATF4 is a transcriptional factor that activates expression of Chop, a pro-apoptotic player in ER stress-induced cell death (Hetz et al., 2020). Taken together, in ER-stressed cells, the PERK branch appears to play a dual role – early pro-survival by reducing the ER workload, and late pro-apoptosis if ER homeostasis is not re-established. Indeed, studies have shown that ER stress activates all 3 UPR branches to promote cell survival, but when stress persists, only PERK signaling is maintained, which promotes apoptosis (Lin et al., 2007; Lin et al., 2009).

It has long been known that brain ischemia triggers PSI, and an increase in p-eIF2α has been repeatedly demonstrated in the brain after ischemic stroke and cardiac arrest (Shen et al., 2018; Yu et al., 2017). However, since at least 3 other kinases (GCN2, HRI, and PKR) can also phosphorylate eIF2α, the extent to which this PSI involves PERK/p-eIF2α signaling and affects pathophysiology after brain ischemia remained undefined for a long time. In 2005, a study on mice with global Perk deletion, except in pancreatic β-cells, provided evidence that PERK is responsible for post-ischemic eIF2α phosphorylation (Owen et al., 2005). However, in this study, protein synthesis was analyzed in the test tube, but not in the intact brain. Recently, we conducted a study in tamoxifen-induced and conditional Perk knockout mice, and found that deletion of Perk in neurons reduces ischemia-induced phosphorylation of eIF2α, and significantly worsens stroke outcome after transient MCAO (Wang et al., 2020). Using a new method based on the surface sensing of translation (SUnSET) technology, this study also showed that the reduced levels of post-ischemic p-eIF2α in Perk knockout mouse brain is associated with higher protein synthesis, thus supporting the mechanistic link between PERK activation, p-eIF2α induction, and suppression of protein synthesis in the ischemic brain. Not surprisingly then, salubrinal, a p-eIF2α de-phosphorylation inhibitor that can increase levels of p-eIF2α in the brain, largely reverses the marked effects of Perk deletion on post-ischemic p-eIF2α and protein synthesis (Wang et al., 2020). These data together clarify the longstanding question about the role of PERK in post-ischemic protein synthesis in the brain, and also indicate that post-ischemic activation of the PERK branch in neurons during the acute phase is neuroprotective in stroke.

A large number of small molecules can be used to specifically modify the PERK/eIF2α UPR branch (Grandjean and Wiseman, 2020; Hetz et al., 2019). Our lab and others have used salubrinal to enhance PERK/p-eIF2α signaling, and have found that acute treatment with salubrinal is beneficial after stroke and global brain ischemia (Anuncibay-Soto et al., 2016; Wang et al., 2020). However, the role of PERK/p-eIF2α in the recovery phase after brain ischemia remains largely unknown. Although Perk knockout mice exhibit worse long-term stroke outcome (Wang et al., 2020), this is likely due to large infarct volumes formed during the acute phase, rather than chronic effects of Perk deletion. Indeed, many studies have demonstrated that under physiologic and pathologic conditions, chronic PERK activation is detrimental to brain function, probably due to the sustained reduction in global rates of translation, while chronic PERK suppression improves neurologic functions (Hughes and Mallucci, 2019; Ma et al., 2013; Sharma et al., 2018). For example, long-term PERK suppression, via genetic modulation or pharmacologic inhibitors, prevents neurodegeneration in animal models of neurodegenerative diseases (Hughes and Mallucci, 2019). Interestingly, suppression of PERK signaling in middle-aged mice enhances learning and memory, and restores them to the normal performance levels found in young mice (Sharma et al., 2018). These exciting findings strongly suggest that targeting the PERK branch is likely to offer long-term benefits in restoring ischemia-induced neurologic deficits including cognitive function, even in aged brains. This appealing possibility warrants investigation in future studies.

In the context of brain ischemia, as in research on the IRE1/XBP1 branch, the PERK branch has been mostly studied in neurons. However, a recent study indicates that activation of PERK/p-eIF2α signaling in astrocytes can lead to a distinct “UPR” reactive state that is pathogenic and trends toward an A1 (neurotoxic) phenotype (Smith et al., 2020). This important finding aligns with the current opinion that astrocytes are highly heterogeneous, and play divergent roles in different states (Sofroniew, 2020). More strikingly, this study showed that suppressing the PERK branch exclusively in astrocytes alone is sufficient to prevent neuronal loss in a mouse model of prion disease, even without any intervention to alter on-going neuronal Perk activation (Smith et al., 2020). Clearly, the PERK branch does more than suppress protein synthesis, and importantly, PERK-modulated astrocytes can be a pathogenic driver of disease progression. Thus, further work is required to better understand PERK in astrocytes after brain ischemia, in order to pave the way for PERK branch-based pharmacologic interventions in brain ischemia.

5. ATF6 UPR branch

ER stress unmasks the Golgi localization signals residing in ATF6. This leads to translocation of full-length ATF6 to the Golgi apparatus where it is cleaved by S1P and S2P endopeptidases to release the cytosolic DNA-binding portion fragment (p50) of ATF6, also called short-form ATF6 (sATF6). Short-form ATF6 is a transcription factor that enters the nucleus where it regulates many pro-survival genes involved in protein folding, maturation, and degradation (Shen et al., 2021). Notably, almost all in vitro and in vivo data suggest that the main function of the ATF6 UPR branch is to facilitate recovery of proteostasis and thus, promote cell survival under stress conditions (Glembotski et al., 2019; Wang and Yang, 2019).

Unlike the PERK and IRE1 UPR branches, the ATF6 branch appears not to be activated, or only modestly activated, in the brain after stroke or cardiac arrest. Since sATF6 is subjected to rapid degradation, studies have used GRP78, a key sATF6-regulated gene, as an indicator of ATF6 activation. A time-course study showed that after transient MCAO, no major change in GRP78 proteins was found in the stroke brain (Yu et al., 2017). Using a cardiac arrest model, we found a modest ~2-fold increase in GRP78 in the brain at both 3 and 24 hours after reperfusion (Shen et al., 2018). Consistently, a previous study indicated no evidence that sATF6 is increased in the brain at 4 hours after cardiac arrest (Kumar et al., 2003). The reason for the differential activation of 3 individual UPR branches after brain ischemia is unknown.

Early studies on 2 downstream components of the ATF6 branch, GRP78 and PDI, have indicated that GRP78 is involved in the neuroprotective effect of ischemic pre-conditioning and post-conditioning (Hayashi et al., 2003; Yuan et al., 2011), and overexpression of PDI significantly reduces the number of CA1 apoptotic neurons after global brain ischemia (Tanaka et al., 2000). To directly examine the role of this branch in brain ischemia, Atf6-specific mouse models have proven useful. In 2015, Yoshikawa et al subjected Atf6α^−/− mice to transient MCAO, and found that Atf6α^−/− mice have worse brain damage after stroke (Yoshikawa et al., 2015). Although the data from this study support a beneficial role for the ATF6 branch in brain ischemia, this conclusion relies on a global knockout model in which Atf6α is constitutively deleted in all cells. Thus, any potentially altered pathways during the development of Atf6α^−/− mice may complicate the results. Moreover, the relevant cell types cannot be precisely dissected. In this respect, conditional Atf6α knockout mice need to be developed, and should be useful.

Based on the Atf6α^−/− data and also on the reported protective effects of the ATF6 branch in cells or other organs under various stress conditions, one immediate question is: to what extent is boosting the ATF6 branch protective in brain ischemia? To answer this question, we generated neuron-specific and tamoxifen-inducible sATF6 knock-in mice (sATF6-KI^Neuron). After subjecting these mice to transient MCAO, we found that forced activation of the ATF6 branch in neurons significantly reduces infarct volumes and improves neurologic function (Yu et al., 2017). The sATF6-KI^Neuron mice also exhibit significantly better outcome than control mice after cardiac arrest (Shen et al., 2021). Mechanistically, enhanced autophagy activity in sATF6-KI^Neuron mouse brains likely contributes to protective effects in stroke (Yu et al., 2017). Recently, analysis of the ATF6-regulated transcriptome in the brain using sATF6-KI^Neuron mice and RNA-sequencing revealed that the pathway most activated by sATF6 in neurons is the ER-associated degradation (ERAD) pathway (Shen et al., 2021). ERAD is a major cellular degradation pathway that is specifically responsible for clearing misfolded proteins that accumulate in the ER under physiologic and pathologic conditions (Qi et al., 2017). Thus, increased ERAD capacity could be another mechanism responsible for ATF6-mediated neuroprotection in brain ischemia.

Notably, the ATF6 UPR branch has been considered a crucial protein quality control pathway in the heart, and many studies have demonstrated that boosting this pathway protects the heart under pathologic conditions, including ischemia (Glembotski et al., 2019). Thus, targeting the ATF6 branch is a promising strategy in ischemic diseases (Glembotski et al., 2019). In 2016, several novel small molecules were identified as specific ATF6 activators (Plate et al., 2016). Among these, compound 147 has been further studied in various ischemic models, including ischemic stroke. The data indicate that compound 147 induces ATF6 target genes in the heart and brain, protects them from ischemic injury, and is safe for long-term use (Blackwood et al., 2019). We recently found that mice treated with compound 147 exhibit better neurologic function and survival rates after cardiac arrest (Shen et al., 2021).

Collectively, current evidence supports the notion that the ATF6 UPR branch is a pro-survival pathway in stroke and cardiac arrest. However, more research is required to substantiate the therapeutic potential for this branch in brain ischemia. Indeed, many important questions need to be clarified. For example, we still do not know whether ATF6 activation is protective in permanent stroke. For both stroke and cardiac arrest, long-term functional recovery after chronic ATF6 modulation has not been evaluated. This is crucial because a better long-term functional recovery is the most meaningful outcome for patients. Moreover, it has been shown that after stroke, Atf6α^−/− mice exhibit less activation of astrocytes and accordingly, less astrocyte scarring, indicating the critical involvement of the ATF6 branch in reactive astrogliosis and glial scarring (Yoshikawa et al., 2015). Interestingly, astrocyte activation was also found to be markedly suppressed in Atf6α^−/− mice in a Parkinson’s model (Hashida et al., 2012). However, both studies used a whole-body Atf6 deletion mouse line and thus, the exact role of ATF6 in astrogliosis remains to be defined. Since reactive astrogliosis and glial scarring play a critical role in the pathophysiology after brain ischemia, it is critical to use an astrocyte-specific mouse tool to clarify the role of ATF6 in astrogliosis. Until we do, pharmacologic targeting of the ATF6 branch for brain ischemia need to remain cautious.

6. Conclusions and perspectives

Brain ischemia induced by ischemic stroke and cardiac arrest continues to be a leading cause of death and long-term disability worldwide and thus, poses a major burden on families and healthcare systems. It is, therefore, of high clinical significance to better understand protection and recovery of the brain from ischemic insult. Although decades of experimental research has markedly advanced our knowledge of the pathophysiology of brain ischemia, no effective pharmacologic intervention for acute cytoprotection or brain restoration in brain ischemia has yet been successfully translated into the clinic. Notably, cellular proteostasis has long been considered a defining factor in the survival and recovery of stressed cells, and the UPR is a key player in restoring proteostasis. Indeed, studies have indicated that the UPR is a promising therapeutic target for ischemia- and/or aging-related diseases. We and others have generated a wealth of information that supports the therapeutic potential of targeting the UPR in brain ischemia (Table 1; Fig. 1).

The IRE1/XBP1 branch has been the most studied UPR branch with respect to pharmacologic interventions in brain ischemia. Regarding the XBP1/HBP/O-GlcNAc axis, in particular, convincing data support the notion that increasing O-GlcNAcylation is beneficial after ischemic stroke or cardiac arrest, as has been manifested by improved short- and long-term outcomes in both young and aged animals. ATF6-based therapeutics also hold great promise for treating ischemia-related diseases. Moreover, the on-going robust research on the PERK branch in brain disorders is expected to guide PERK-based therapeutic strategies during both the acute and chronic recovery phases after brain ischemia. Additionally, UPR associated pathways could also serve as therapeutic targets in brain ischemia. For example, clearance of unfolded/misfolded proteins requires not only the ubiquitin-proteasome pathway, but also the lysosomal pathway (De Leonibus et al., 2019). Importantly, both degradation pathways are dysfunctional after stroke (Chen et al., 2020; Yuan et al., 2021).

Taken together, with growing efforts to identify new UPR-specific activators or inhibitors, and to clarify the role of the UPR in the post-ischemic brain, optimism remains for future UPR-based treatments in ischemic stroke and cardiac arrest.

Highlights:

• Brain ischemia induced by ischemic stroke and cardiac arrest is a leading cause of death and long-term disability worldwide.

• After brain ischemia, cellular proteostasis is disrupted and the unfolded protein response (UPR) is activated.

• Activation of the UPR is pro-survival at least during the acute phase after brain ischemia.