Protein Aggregation in the Pathogenesis of Ischemic Stroke

Abstract

Despite the distinction between ischemic stroke and neurodegenerative disorders, they share numerous pathophysiologies particularly those mediated by inflammation and oxidative stress. Although protein aggregation is considered to be a hallmark of neurodegenerative diseases, the formation of protein aggregates can be also induced within a short time after cerebral ischemia, aggravating cerebral ischemic injury. Protein aggregation uncovers a previously unappreciated molecular overlap between neurodegenerative diseases and ischemic stroke. Unfortunately, compared with neurodegenerative disease, mechanism of protein aggregation after cerebral ischemia and how this can be averted remain unclear. This review highlights current understanding on protein aggregation and its intrinsic role in ischemic stroke.

Introduction

Stroke is a common neurological disorder and a major cause of permanent disability (Hankey 2017). It refers to the loss of brain functions because of a disturbance in the blood supply to the brain. The shortage of blood flow during a stroke is due to either ischemia or hemorrhage (Liang et al. 2016). Therefore, strokes can be divided into two categories: ischemic and hemorrhagic stroke, and over 80% are ischemic stroke (Zhou et al. 2018).

Stroke has presented a long-unmet clinical challenge as current therapies are ineffective at reducing the morbidity and outcomes associated with the disease. Currently, the only US Food and Drug Administration (FDA) approved therapy for acute ischemic stroke is intravenous administration of recombinant tissue plasminogen activator (rt-PA) (Alberts 2017), which has a narrow therapeutic time window and is associated with various side effects (Zerna et al. 2018). Furthermore, most neuroprotective agents have shown efficacy in experimental stroke models but failed to provide significant benefits in clinical trials (Rother 2008). Consequently, further studies on the mechanisms that causes brain damage in stroke and novel treatment approaches are urgently needed.

Proteins must have the right structure to perform their functions. As such, aggregated proteins typically have little or no biological activity and negative impacts on the health and safety of an individual. Aggregation of proteins has been shown to be associated with neurodegenerative disorders (Droppelmann et al. 2019; Yedlapudi et al. 2019), aging (David et al. 2010; Magalhaes et al. 2020) and enhanced cytotoxicity (Sun et al. 2011; Huang et al. 2019; Zhang et al. 2020). Emerging evidence shows that aggregation of proteins occurs in acute neuronal injury after focal brain ischemia (Hayashi et al. 1992; Hu et al. 2001; Zhang et al. 2006). Previous studies have shown that misfolding and aggregation of proteins are major pathological events in post-ischemic neurons (Luo et al. 2013; Zhang et al. 2015). In a rat ischemic model, two hours of focal brain ischemia induced protein aggregation in ischemic neocortical neurons at 1 h of reperfusion, and protein aggregation persisted until neuronal death at 24 h of reperfusion (Hu et al. 2001). Additionally, abnormal aggregation of ribosomal proteins lasted until the onset of delayed neuronal death, 24–48 h after ischemia reperfusion (Zhang et al. 2006).

Deposition and accumulation of undesirable protein aggregates may contribute to the neuronal death after focal ischemia (Hu et al. 2001). However, detailed mechanism of protein aggregation and subsequent proteotoxicity after brain ischemia are currently not well understood. Here, we review current knowledge on the underlying molecular mechanisms of protein aggregation in ischemic stroke, providing a foundation for future research.

Protein Aggregation in Ischemic Stroke

Initially, electron microscopy revealed dense, fluffy and dark materials in the pyramidal neurons of hippocampus CA1 region destined for delayed neuronal death after transient cerebral ischemia (Kirino et al. 1984; Deshpande et al. 1992). And the post-ischemic materials were thought to be aggregates of proteins or fragments of internalized plasma membrane, which may disrupt vital cellular structure and functions, leading to cell death. (Deshpande et al. 1992). Hu and colleagues latter demonstrated that these dark deposits were likely to be composed of aberrant proteins (Hu et al. 2000), because they conjugated with ethanolic phosphotungstic acid (EPTA), which preferentially reacts with proteins. Elsewhere, using nanoLC-MS/MS, Kahl and colleagues identified 196 proteins that aggregated after cerebral ischemia/reperfusion injury (Kahl et al. 2018). They provided the first comprehensive proteomic analysis on protein aggregation after ischemic stroke. This proved that ischemia affects the normal function of various proteins, interfering with intracellular biochemical process. Increasing evidence shows that ischemia/reperfusion induces aggregation of ubiquitin (Hochrainer et al. 2012), small ubiquitin-related modifier (SUMO) (Hochrainer et al. 2015) and neurodegeneration-related disease proteins (Kahl et al. 2018).

Ischemic Stroke Induces Aggregation of Ubiquitin Proteins

Ubiquitin (Ub) is a small, dynamic and highly conserved protein found in many organisms, ranging from yeast to mammals. It regulates virtually all biological processes including cell division, transcription, immunity, endocytosis, signaling pathways and quality control of proteins (Swatek and Komander 2016; Yau and Rape 2016; Kwon and Ciechanover 2017; Sato et al. 2019). The ubiquitin–proteasome system (UPS) identifies, labels and transports damaged or misfolded proteins to the proteasomes for degradation (Schwartz and Ciechanover 2009). Structurally, ubiquitin is an 8.5 kD protein with 76-amino acids. To effect its function, ubiquitin is first activated by ubiquitin activating enzyme (E1) then transferred to ubiquitin conjugating enzyme (E2) before being conjugated to Lys residues on target substrates by ubiquitin ligating enzyme (E3) (Graham and Liu 2017; Bernassola et al. 2019). Additional ubiquitin molecules are linked to the initial ubiquitin, forming polyubiquitin (poly-Ub) chains. The type of ubiquitination (mono-, multi- or polyubiquitination) as well as the length and topology of the polyubiquitin chains determines the fate of an ubiquitin-modified protein (Kwon and Ciechanover 2017). With the help of cytoskeletal proteins and other enzymes, 26S proteasome recognizes and hydrolyses K48-poly-Ub-linked proteins (Graham and Liu 2017; Kwon and Ciechanover 2017). Additionally, there is evidence that K11 and 29-linked chains serve as proteasomal recognition signals as well (Kwon and Ciechanover 2017). K63-linked polyubiquitin can initiate the production of unique downstream signals including DNA repair, NF-κB signaling and endocytic pathway between the interacting substrates (Buetow and Huang 2016). UPS plays an essential role in modulating protein functions and maintaining the homeostatic balance. Dysfunctional UPS leads to the development of multiple human diseases (Schwartz and Ciechanover 2009; Akutsu et al. 2016). Aggregation of ubiquitinated proteins is common in neurodegenerative disorders and has been associated with neuronal degeneration (Blokhuis et al. 2013), ubiquitinated protein aggregates have also been observed in cerebral ischemia (DeGracia and Hu 2007).

Ischemia–reperfusion injury is known to induce ubiquitination in the brain (Hochrainer 2018). Hayashi et al. first demonstrated that mild forebrain ischemia increases the molecular weight of ubiquitin conjugates in rat hippocampus (Hayashi et al. 1991). Using ethanolic phosphotungstic acid electron microscopy and immunoelectron microscopy, Hu et al. observed dense ubiquitin protein aggregates in CA1 neurons destined to die in 72 h, just after 15 min of cerebral ischemia (Hu et al. 2000). Meanwhile, using high-resolution confocal microscopy, they further demonstrated that protein aggregates containing ubiquitin persistently and progressively accumulated in all CA1 neurons that would latter die, but not in neuronal populations that survive ischemia (Hu et al. 2000). Ubiquitinated-protein aggregates were found in the neuronal soma, dendrites and axons, all linked to ischemia mediated neuronal death (Hu et al. 2001).

Reperfusion has been found to induce accumulation of ubiquitin aggregates in the ipsilateral neocortex and striatum, whereas permanent ischemia failed to increase levels of ubiquitin aggregates (Hochrainer et al. 2012). In the latter case, increase in ubiquitin aggregates depletes free ubiquitin. This therefore impairs ubiquitin mediated functions such as those described above.

Aggregation of Neurodegeneration-Linked Protein in Ischemic Stroke

Aggregation of protein is a widely accepted hallmark of neurodegenerative diseases. However, emerging evidence shows that these protein aggregates also participate in the pathophysiology of ischemic stroke. FUS and TDP-43 are RNA-binding protein linked to amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). Research shows that the aggregation of these proteins may be related to the altered RNA metabolism after ischemia (Kahl et al. 2018).

Although physical damage after ischemic stroke tends to resolve, survivors have an increased risk of developing cognitive impairment. It has been reported that stroke is a risk factor for incident Alzheimer Disease (AD) (Pluta 2007; Pluta et al. 2009, 2010; Vijayan and Reddy 2016; Dong et al. 2018), Moreover, cerebral abnormalities in stroke, particularly ischemic stroke, may mediate biochemical dysfunctions in the brain, ultimately leading to vascular dementia (VaD) and AD (Vijayan and Reddy 2016). Emerging evidence suggests that ischemic stroke and AD share pathophysiological changes in brain tissue, including hypoperfusion, oxidative stress, and inflammation (Dong et al. 2018). The insufficient flow of blood to the brain after ischemic stroke accelerates deposition of amyloid-β (Aβ) within brain parenchyma and cerebral vessel walls. This deposition is the initial event leading to synaptic dysfunction, causing dementia and impaired cognitive function (Pluta 2007; Pluta et al. 2010; Goulay et al. 2019). Furthermore, vascular factors predisposing to stroke and cerebrovascular disease have been strongly related to vascular deposition of Aβ (Goulay et al. 2019). One of the most common predisposing factors is cerebral amyloid angiopathy (CAA). CAA is a cerebrovascular disease characterized by the accumulation of Aβ within the walls of cortical and leptomeningeal blood vessels, associated with cognitive decline and both ischemic and hemorrhagic stroke (Charidimou et al. 2012; Vales-Montero et al. 2019). Severe CAA is believed to play a fundamental role in the development of cerebral microbleeds (CMB) and cerebral microinfarcts (CMI) (Haglund et al. 2006; Freeze et al. 2019). Moreover, a case–control study on human brain biopsy showed that CAA is a risk factor for cerebral infarction (Cadavid et al. 2000). CAA-induced CMI may be due to the blockage of blood vessels by the aggregation of Aβ on the microvascular wall. Frequent microinfarcts caused by severe CAA may contribute to cognitive impairment in individuals with dementia (Soontornniyomkij et al. 2010). Accordingly, the formation of Aβ aggregates induced by stroke may aggravate brain ischemic injury as well as promoting the cognitive impairment of stroke survivors.

On the other hand, tau aggregates have also been associated with AD (Bittar et al. 2020). A recent study showed that high level of tau in plasma is a biomarker for ischemic stroke (Pase et al. 2019). In the disease pathology, tau disease is described as insoluble intracellular neurofibrillary tangles (NFTs) consisting of hyperphosphorylated and aggregated tau proteins (Gao et al. 2018; Bittar et al. 2020). Optimum phosphorylation is necessary to promote the biological activities of tau, but excessive phosphorylation induces aggregation (Alonso et al. 2001; Gao et al. 2018). Expression of tau gene increased to the maximum of 3.3-fold change on the second day after brain ischemia–reperfusion in rats (Pluta et al. 2018). In addition, brain ischemia induced hyperphosphorylation of tau protein, causing neurofibrillary tauopathy and neurofibrillary tangles (Wen et al. 2007; Bi et al. 2017; Kovalska et al. 2018). Excessive cerebral infarction in the territory of the middle cerebral artery is associated with development of NFTs in the ipsilateral basal nucleus of Meynert (Hatsuta et al. 2019). Aberrant assembly of tau into insoluble aggregates is accompanied by synaptic dysfunction and neural cell death under pathological conditions (Guo et al. 2017). Therefore, tau has been targeted by therapeutic interventions during ischemic stroke, and controlling tau phosphorylation may induce more protective effects under ischemic stimuli (Chen and Jiang 2019).

α-Synuclein is a small presynaptic protein involved in the maintenance and release of synaptic neurotransmitters, particularly dopamine (Bendor et al. 2013; Schaser et al. 2019). Under pathological conditions, aggregation of α-synuclein is implicated for neurotoxicity and neurodegeneration. This aggregation participates in the neurodegeneration associated with Parkinson disease (PD) together with related Lewy body disorders (e.g., Dementia with Lewy Bodies) (Stoica et al. 2012; Lashuel et al. 2013; Schaser et al. 2019). Although α-synuclein is thought to be a chronic neurodegeneration-related protein, it also mediates secondary brain damage after cerebral ischemia (Kim et al. 2016). In the rodent brain, transient focal cerebral ischemia induced expression of α-synuclein protein, nuclear translocation, oligomerization and aggregation (Kim et al. 2016). Moreover, researches in mouse models have found that preventing aggregation of neurotoxic α-synuclein decreases the infarction area and leads to better physiological and functional outcome after stroke (Kim et al. 2016, 2018; Wang et al. 2019). This suggests that abnormal aggregation of α-synuclein promotes neuronal death not only in chronic neurodegenerative conditions like PD but also in acute conditions like ischemic stroke.

Factors Contributing to Protein Aggregation After Brain Ischemia

Acidification of Ischemia Micro-environment Activates Aggregation of Protein

The pH of the protein environment greatly affects the conformation and stability of proteins, impacting on subsequent aggregation (Skamris et al. 2016; Wang and Roberts 2018; Santos et al. 2020). A brief exposure of monoclonal IgG to low pH (pH 1) induced immediate formation of transient molten globules, which mediated the formation of aggregates (from dimers to 10-μm particles) (Filipe et al. 2012). Altering the pH significantly influenced the size and solubility of protein aggregates. Aggregates at pH 3.5 grew primarily by monomer addition and remained small and soluble, however, higher pH of between 4.5 and 5.5 led to formation of insoluble aggregates. At pH 4.5, protein aggregates first grew by chain polymerization, followed by condensation polymerization, which ultimately led to formation of large insoluble particles. At pH 5.5, loss of monomer resulted to formation of primarily insoluble aggregates (Brummitt et al. 2011). During cerebral ischemia, hypoxia leads to the production of lactate which usually causes tissue acidosis. Severe acidosis with a pH below 6.4 exacerbates ischemic injury through mechanisms involving protein denaturation, acid-sensitive calcium channels and release of ferrous compounds (Lam et al. 2013). Conversely, a small degree of acidosis in the pH range of 6.5–7.0 can reduce ischemic injury (Lam et al. 2013). This difference was thought to be influenced by the pH of the around the aggregating proteins. To investigate this, a pH titration experiments were performed on mAb 1. Adjusting the pH from 3.5 to pH 5 was found to increase in aggregation of proteins, but there was no difference in aggregation levels at pH levels ranging from 5 to 7 (Kent et al. 2018).

Protein Aggregation and UPS Dysfunction After Cerebral Ischemia

Improperly folded proteins (misfolded proteins) may clump together, promoting aggregation in cells (Galves et al. 2019). Ubiquitin proteasome pathway is the major route involved in the degradation of abnormal proteins. Thus, it is reasonable to hypothesize that impaired UPS may be responsible for aggregation of proteins in post-ischemic stroke. Notably, ubiquitin/proteasome-dependent degradation of proteins is impaired after transient forebrain ischemia (Suh et al. 2010; Graham and Liu 2017). Before degradation, abnormal protein is first conjugated to polyubiquitin chains. It is then recognized and hydrolyzed by proteasomes. Elevated ubiquitination after ischemia could be caused by upregulated expression of ubiquitin gene, leading to greater availability of ubiquitin for conjugation (Hochrainer 2018). However as already mentioned, as ubiquitin aggregates increase, free ubiquitin is depleted (Hochrainer et al. 2012), which intern may increase accumulation of abnormal protein.

Alternatively, aggregation of proteins associated with proteasomal dysfunction has also been observed in cerebral ischemia (Hu et al. 2001; Ge et al. 2007). Elevated protein aggregation could be caused by a blockade in proteasome function. In fact, ischemia and reperfusion have been shown to decreases the 26S proteasomal activity, but which recovers gradually (Tai et al. 2010; Hochrainer et al. 2012; Caldeira et al. 2013). Furthermore, proteasome subunits particularly the 19S components were deposited into the protein fraction of the aggregate after an episode of transient cerebral ischemia (Ge et al. 2007). Of note is that 26S proteasomes are transiently disassembled during the early period of reperfusion and this may contribute to the accumulation of protein aggregation during the early post-ischemic phase (Ge et al. 2007). Particularly, Li and colleagues demonstrated that proteasome dysfunction contributed to protein aggregation caused by ischemic insults and trehalose prevented protein aggregation via preservation of proteasome activity (Li et al. 2017). Together, these findings imply that ischemia-induced proteasome dysfunction contributes, at least in part, to aggregation of proteins after brain ischemia.

Malfunction of Molecular Chaperones After Cerebral Ischemia

The protein quality control (PQC) system is essential in preventing accumulation of misfolded or damaged proteins and their aggregates (Silva et al. 2013). A fundamental component of PQC is molecular chaperones which promotes correct folding of proteins, prevents the formation of nonfunctional and potentially deleterious protein aggregates (Ciechanover and Kwon 2017; D'Andrea et al. 2018). Heat-shock proteins (HSPs) including Hsp70, Hsp90, Hsp60, Hsp40 (DnaJ), and small HSPs constitute the majority of molecular chaperones (Ciechanover and Kwon 2017). HSPs are important participants in protein homeostasis system. Their ability to bind misfolded proteins may play a crucial role in preventing protein aggregation in cells (Roman et al. 2017). Over-expression of Hsp70 in fruit flies suppresses neurodegeneration caused by aggregation of mutated α-synuclein (Auluck et al. 2002). However, in post-ischemic neurons, irreversible malfunctioning of molecular chaperones could result in production of large amounts of unfolded proteins (Liu et al. 2005b, a; Zhang et al. 2006). After ischemia, co-translational chaperone heat shock cognate protein 70 and co-chaperone HSP40-Hdj1 as well as other heat-shock proteins were irreversibly clumped into large abnormal protein aggregates (Liu et al. 2005a, b; Kahl et al. 2018). Therefore, ischemia damages co-translational chaperone, resulting in irreparable protein aggregation after cerebral ischemic injury.

Protein disulfide isomerase (PDI) is another endoplasmic reticulum (ER) molecular chaperone, critical for proper folding of proteins in the ER (Lyles and Gilbert 1991). PDI is also a component of signal peptide peptidase (SPP)-mediated degradation of misfolded proteins (Lee et al. 2010). Cerebral ischemia/reperfusion injury has been found to enhance expression of nitrogen oxide (NO), which mediates S-nitrosylation of PDI (Murphy 2000; Chen et al. 2012). S-nitrosylation of PDI probably inactivates the normal properties of PDI and this post-translational modification may weaken its protective effect on SOD1 in ischemia/reperfusion injury. PDI plays a key role in forming disulfide bonds during the biosynthesis of secreted proteins and other disulfide-bonded proteins within the ER (Freedman et al. 2017). As one of the molecular targets for PDI, SOD1 can bind to PDI through disulfide bond (Atkin et al. 2006; Toldo et al. 2011). However, if PDI were S-nitrosylated, it could not bind to SOD1 as efficiently, leading the disulfide shuffling between SOD1 molecules; and the disulfide-reduced SOD1 would be prone to forming aggregates (Tokuda and Furukawa 2017; Cohen et al. 2020). As a consequence, NO-mediated S-nitrosylation of PDI may take part in the formation of SOD1 aggregates in cerebral ischemia/reperfusion injury (Chen et al. 2012).

Cerebral Ischemia may Affect Phase Separation Leading to Protein Aggregation

Phase separation has recently been thought to be the operational principle governing the formation of membraneless organelles to regulate biological functions and activities (Boeynaems et al. 2018; Dolgin 2018). These membraneless organelles are composed of proteins, nucleic acids and other molecular components, and they are present in the nucleus (nucleolus, nuclear speckles, Cajal bodies) as well as in the cytoplasm (stress granules (SGs), processing bodies, centriole and Germ granules) (Mitrea and Kriwacki 2016; Dolgin 2018). In addition, various cellular functions including signaling, cytoskeletal organization and transcriptional regulation have been shown to be accompanied with phase separation (Banani et al. 2017). Although phase separation is critical for various features of cellular architecture and physiology, it can have deleterious outcomes. Abnormal phase separation is linked to aggregation of amyloid in proteins, and amyloid aggregation is on the other hand implicated in many neurodegenerative diseases and cancer. (de Oliveira et al. 2019). Soluble tau species can undergo phase separation under cellular conditions and that phase-separated tau droplets can induce tau aggregate formation (Wegmann et al. 2018). In vitro, phosphorylation of tau also promotes phase separation and aggregation (Ambadipudi et al. 2017). Interestingly, phosphorylation and aggregation of tau also occurred in ischemia stroke. Similarly, aggregation of RNA-binding proteins FUS (Ding et al. 2020), TDP-43 (Shorter 2019; Wolozin 2019), hnRNPA1 (Lin et al. 2015) and hnRNPA2 (Ryan et al. 2018) would correspond to irreversible phase separation, and these protein aggregates are also found in ischemic stroke as mentioned before. Accordingly, it is reasonable to hypothesize that stroke promotes formation of protein aggregates by influencing phase separation.

Of note is that currently, there is no solid evidence that ischemia/reperfusion injury affects phase separation in addition to mediating aggregation of pathological proteins. Interestingly, the inhibition of mitochondria function increased levels of reactive oxygen species and enhances protein aggregation of Caenorhabditis elegans (Civelek et al. 2019). Additionally, accumulative evidence suggests that mitochondrial function could be affected by protein aggregates in neurodegenerative disorders (Lin and Beal 2006; Cabezas-Opazo et al. 2015; Cenini and Voos 2016). Together, these findings show that mitochondrial dysfunction may be closely associated with the formation of protein aggregates. Indeed it was found that mitochondrial abnormalities initiate pathophysiological events under stroke (Catanese et al. 2017). Meanwhile, maintaining mitochondrial function is crucial in promoting neuron survival and neurological improvement (Liu et al. 2018).

This raises the need to understand the relationship between mitochondrial dysfunction, phase separation and protein aggregation in ischemic stroke. The fundamental role of mitochondria is to generate ATP energy for the cells. Elsewhere, research shows that the level of ATP influences the dynamics of phase separation (Wright et al. 2019). Cellular ATP can also act as a chemical hydrotrope, directly preventing phase separation and protein aggregation (Patel et al. 2017). Hence, the overall reduction in the level of cellular ATP caused by mitochondrial dysfunction in ischemic stroke may contribute to the impairment of phase separation. This intern promotes aggregation of protein.

Altered Non-coding RNA may Promote Protein Aggregation in Ischemic Stroke

A class of non-coding RNAs including microRNAs (miRNAs), long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) are increasingly being found to be involved in the pathogenesis of ischemic stroke (Mehta et al. 2017; Chung et al. 2018; Li et al. 2018a, b). One research showed that silencing long non-coding RNA GAS5 suppressed neuron cell apoptosis and nerve injury in ischemic stroke (Deng et al. 2020). Elsewhere, over-expression of circDLGAP4 enhanced the integrity of blood–brain barrier through miR-143 mediated regulation of endothelial-mesenchymal transition in stroke (Bai et al. 2018). However, the precise relationship between non-coding RNAs and protein aggregation in brain ischemia remains unclear. Recently in animals with PD, decrease in the abundance of miR-7 was implicated for toxicity and increased aggregation of α-synuclein (Junn et al. 2009; McMillan et al. 2017). Interestingly, pre- or post-ischemic treatment with miR-7 analogs ameliorated ischemic brain damage by repressing post-ischemic induction of α-synuclein (Kim et al. 2018). In addition, one other study revealed that SINEB1 element of the long non-coding RNA Malat1 is essential for proteostasis of TDP-43 (Nguyen et al. 2019). The loss of the SINEB1 in lncRNA Malat1 causes mis-localization and aggregation of TDP-43 in cytoplasm (Nguyen et al. 2019). Similarly, circRNAs can potentially regulate the formation of abnormal protein aggregates. circRNAs have been found to regulate RBPs function by serving as sponges for RBPs (Zang et al. 2020). This interaction between circRNAs and protein possibly prevents the aggregation of certain proteins. The major function of RNA debranching enzyme (Dbr1) is to prevent formation of lariat introns during pre-mRNA splicing (Khalid et al. 2005). Deletion of Dbr1 suppresses aggregation and toxicity of TDP-43, both in yeast and mammalian cells (Armakola et al. 2012). This may be due to the accumulation of intronic lariat in the cytoplasm of dbr1Δ cells, acting as decoys to sequester toxic TDP-43, and it suggests that circRNAs have the ability to inhibit the formation of RNA-binding protein aggregates. Taken together, abnormal expression of non-coding RNAs in ischemic stroke may cause aberrant aggregation of proteins.

The Toxicity of Aggregated Protein in Ischemic Stroke

Protein Aggregation and Oxidative Stress

Aggregated proteins are commonly present in neurodegenerative disorders and have been considered to cause neuronal degeneration. After focal ischemia, neurons in penumbra region ultimately undergo delayed neuronal death (Liu et al. 2005a, b). Several lines of evidence suggest that abnormal protein aggregates may contribute to ischemia mediated delayed neuronal death (Hu et al. 2000, 2001). Cytotoxicity and neuron death caused by accumulation of protein aggregates may be largely associated with oxidative stress (Levy et al. 2019). Hypoxia and aging can induce the formation of protein aggregates in mitochondria, which leads to mitochondrial dysfunction and the production of excess reactive oxygen species (ROS) (Banerjee et al. 2010; Kaufman et al. 2017). Weids and colleagues showed that the formation of aggregates induced production of ROS (Weids and Grant 2014). When yeast mutants lacking TSA1 were exposed to proline analogue azetidine-2-carboxylic acid (AZC), it significantly induced aggregation of proteins and production of ROS. Interestingly, treatment with cycloheximide prevented AZC-induced protein aggregation and abrogated generation of ROS by inhibiting synthesis of nascent proteins. This strengthens the view that in this case, production of ROS is mediated by the formation of protein aggregates (Weids and Grant 2014). Oxidative stress in ischemic brain results in cell death in neurons, glial cells, and vascular cells, and impairments in neurological recovery after stroke (Li et al. 2018a, b). Therefore, although aggregation of pathological proteins may be caused by different factors, redox perturbations appear to be a consequence of protein aggregation.

Protein Aggregation and Inflammation

Neuroinflammation is one of the main causes of mortality in cerebral ischemia (Stonesifer et al. 2017). Interestingly, there is evidence showing that aggregation of proteins was closely associated with inflammatory cascade in neurodegenerative disorders (Currais et al. 2017). In SOD1 mice models with ALS, aggregation of proteins was found to impair ubiquitin–proteasome system and also generates peptides for the major histocompatibility complex class I molecules. This activates immune responses, enhancing the pathological process of ALS (Bendotti et al. 2012). Presence of α-synuclein amyloid deposits has been found to trigger inflammatory response in Parkinson's disease (Gustot et al. 2015). In addition, astrocytes expressing ALS-linked mutant FUS adopt a reactive phenotype and become toxic to motor neurons via the release of pro-inflammatory cytokines (Kia et al. 2018). Therefore, a wide variety of protein aggregates may activate an inflammatory cascade that eventually culminates with cerebral ischemic injury.

Resident microglia are the predominant immune cells in the brain, detecting signals from injured neurons and other cells in the CNS (An et al. 2014). Activated microglia secretes pro-inflammatory cytokine that causes secondary brain damage in cerebral ischemia. Notably, aggregated α-synuclein may directly activate microglia, predominantly increasing the production of pro-inflammatory cytokines such as tumor necrosis factor alpha (TNFα) and interleukin-1beta (IL-1β), which in turn modulates subsequent inflammatory responses (Hoffmann et al. 2016). During the acute phase of cerebral ischemia, the activation of microglia may be mainly due to the infiltration of peripheral immune cells caused by the destruction of the blood–brain barrier. In later stages, activation of microglia mediated by aggregation of proteins may explain why degenerative diseases such as dementia (Corraini et al. 2017) and Alzheimer Disease (Zhou et al. 2015) are prevalent in stroke survivors.

Aggregation of RNA-Binding Proteins may Prevent These Proteins from Executing Their Normal Function

As mentioned before, RNA-binding proteins such as FUS, TDP-43 and hnRNPA1 constitute the largest group of aggregating proteins in ischemia. These proteins could be involved in all aspects of RNA metabolism; from transcription, processing and transport/stability to the formation of cytoplasmic and nuclear stress granules (Ratti and Buratti 2016; Deshaies et al. 2018). The regulatory activity on RNA metabolism of FUS and TDP-43 depends on their RNA-binding properties. FUS and TDP-43 could autoregulate their expression by binding to their respective pre-mRNA (Ayala et al. 2011; Zhou et al. 2013). Target sequences for RNA-binding proteins are mainly at the intronic or exonic sites, but also include non-coding RNAs and 3′UTRs. In line with this, they can participate in the metabolism of nuclear RNA, including splicing, transcriptional repression, no-coding RNA synthesis, mRNA stability and RNA transport (Blokhuis et al. 2013; Ratti and Buratti 2016). Consequently, aberrant aggregation of these RNA-binding proteins during disease can impair multiple RNA metabolic pathways and ultimately lead to cell death or inactivation.

On the other hand, aggregation of proteins could exert cytotoxicity via sequestration of their binding partners or interactomes essential for cell function. FUS and TDP-43 could interact and bind with multiple proteins including survival motor neuron (SMN) protein, Drosha protein and small nuclear ribonucleoprotein particles (snRNPs) (Yamazaki et al. 2012; Tsuiji et al. 2013). TDP-43 and FUS have been described as an accessory component of Drosha complex, which is crucially involved in basal and tissue-specific RNA processing events (Gregory et al. 2004). Furthermore, TDP-43 is an essential factor that controls the stability of Drosha protein during neuronal differentiation (Di Carlo et al. 2013). SMN plays important roles in multiple fundamental cellular homeostatic pathways, including the assembly of spliceosome and biogenesis of ribonucleoproteins (Chaytow et al. 2018). SMN complex is mainly located in the cytoplasm and nuclear Gems (Yamazaki et al. 2012). Interaction of FUS with SMN proteins is a requisite for the formation of Gem, whereas knockdown of FUS or TDP-43 results in the loss of Gems (Yamazaki et al. 2012). Therefore, FUS or TDP-43 cytoplasmic aggregates may affect the level and function of Drosha protein and SMN.

It is therefore logical to hypothesize that due to pathological aggregation, the decrease of soluble proteins in the cytoplasm may directly or indirectly affect interaction of molecules or proteins, which consequently alters downstream pathways that negatively impact on the cell.

Protein Aggregation may Contribute To The Excessive Autophagy After Brain Ischemia

Autophagy is a widely accepted self-protecting cellular catabolic pathway. It is essential in maintaining cellular homeostasis by enhancing the degradation of nonfunctional proteins and damaged organelles. There is increasing evidence that activated autophagy mediates many pathophysiological changes in ischemic stroke (Carloni et al. 2010; Zhang et al. 2014). Inhibiting autophagy either with 3-methyladenine or through Atg7 silencing was found to enhance ischemia induced release of cytochrome c, together with mediating downstream activation of apoptosis (Zhang et al. 2013). It has been shown that autophagy induced by rapamycin reduces brain infarct together with improving neurological outcome in tMCAO, pMCAO and embolic MCAO models (Chauhan et al. 2011; Buckley et al. 2014; Li et al. 2014; Wang et al. 2018). Despite the neuroprotection of autophagy in ischemia stroke, other research demonstrated that over activation of autophagy could lead to cellular death, popularly referred to as “autophagic cellular death” (Clarke and Puyal 2012). Consequently, autophagy in ischemia stroke has been considered to be a type of cell death, exemplified by its detrimental effect in ischemia stroke.

Autophagy is an essential process that facilitates the clearance of damaged organelles and proteins aggregates by delivering them to lysosomes for degradation. It is now known that disease-related proteins accumulate and aggregate when autophagic flux is compromised (Pavel et al. 2016). Consequently, boosting autophagy has been proposed as a novel therapeutic strategy for neurodegenerative diseases targeting aggregated proteins (Martin et al. 2015). During the early phase of ischemic stroke, autophagy plays an indispensable protective role against neural death by removing toxic protein aggregates and damaged organelles (Luo et al. 2013). However, sustained protein aggregation coupled with deterioration of other pathogenesis-related factors may cause excessive activation of autophagy. Propofol administration significantly decreased the infarct area, and attenuated the neurological deficits after ischemic stroke in a mouse model. Furthermore, the propofol-induced neuroprotective effect was associated with the decreased neurotoxic aggregation of α-synuclein and reduction of the excessive autophagy occurring after acute ischemic stroke (Wang et al. 2019).

Conclusions

It appears that protein aggregates modulate the pathogenesis of cerebral ischemia and are thus a new mechanism driving cell dysfunction and death following brain ischemia (Fig. 1). Protein aggregation triggers widespread disruptions of processes that determine cell fate. Aggregated proteins exert cytotoxicity leading to ischemic cell death on the basis of induced-inflammation. Although attenuating aberrant protein aggregation might be an effective strategy to rescue the neuronal death in mice model of transient cerebral ischemia (Zhang et al. 2010; Liang et al. 2012; Wang et al. 2019), few systematic studies have been done in relation to the therapeutic value of protein aggregation in ischemic stroke. Thus, further understanding of toxic effects of protein aggregates and the mechanisms underlying their formation will be vital if aggregation modulators are to be used effectively as neuroprotective therapies in ischemic stroke. Prophylactic or therapeutic strategies that target protein aggregation will bring new hope for combating ischemic neurodegeneration.

Fig. 1.

A diagram illustrating the role of protein aggregation in ischemic stroke. Brain ischemia disrupts the balance in the protein quality control system, leading to the accumulation of aberrant protein aggregates. These protein aggregates have lost their original cellular functions and thus exhibit cytotoxicity. In addition, protein aggregation may contribute to the excessive autophagy due to dysfunction of the proteasome in ischemic stroke. Cytotoxicity caused by protein aggregates can lead to glial cell activation, neural death, further aggravating cerebral ischemic-induced damage

Author Contributions

SW wrote the manuscript, Dr. LD provided constructive comments.

Funding

This study was supported by Grants from 2019 PhD Research Startup Fund of Affiliated Hospital of Yangzhou University (No. BS2019DLF).

Compliance with Ethical Standards

Conflict of interest

The authors have no competing financial interests to declare.

Research Involving Human and Animal Rights

This article does not contain any studies with human participants or animals performed by any of the authors.