neuronal autophagy in cerebral ischemia – a potential ... · companion of mtor, ros – reactive...

Review

Neuronal autophagy in cerebral ischemia– a potential target for neuroprotective strategies?

Bo¿ena Gabryel, Alicja Kost, Daniela Kasprowska

Department of Pharmacology, Silesian Medical University, Medyków 18, PL 40-752 Katowice, Poland

Correspondence: Bo¿ena Gabryel, e-mail: [email protected]

Abstract:

Although many attempts have been made, stroke treatment options are still extremely limited and brain ischemia remains the leadingcause of death and disability worldwide. Two major strategies for ischemic stroke, reperfusion and neuroprotection, are currently be-ing evaluated. Autophagy is a bulk protein degradation system that is involved in multiple cellular processes. Increasing data suggestthat activation of autophagy in ischemic brain may contribute to neuroprotection. However, it should also be noted that there are evi-dences that autophagy is a process involved in neurodegeneration. Targeting signaling pathways related to autophagy might bea promising option in the treatment of cerebral ischemia, but the exact role of autophagy activation due to ischemic episodes and itspotential applications in pharmacotherapy are still to be determined. In this paper we review recent evidences for cerebral ischemia-induced autophagy, briefly discuss mechanisms and signaling pathways that lead to this activation and we analyze its potential roles.

Key words:

autophagy, neuroprotection, apoptosis, necrosis, brain, ischemia

Abbreviations: 3-MA – 3-methyladenine, AA – amino acids,Akt/PKB – protein kinase B, AMP – adenosine monophos-phate, AMPK – AMP activated kinase, ATP – adenosine tri-phosphate, CaMKKb – Ca2+/calmodulin-dependent protein ki-nase b, CMA – chaperone-mediated autophagy, CNS – centralnervous system, CREB – cAMP response element binding pro-tein, eIF2a – eukaryotic initiation factor 2a, ER – endoplasmicreticulum, GTP – guanosine triphosphate, H/I – hypoxic/ischemic, HIF-1 – hypoxia inducible factor 1, IPC – ischemicpreconditioning, Ire1 – inositol-requiring protein-1, JNK – JunN-terminal protein kinase, LAMP2a – lysosome associated mem-brane protein type 2, LC3 – microtubule-associated protein 1light chain 3, LKB1 – serine threonine kinase 11 (or liver ki-nase B1), MAP1B – microtubule associated protein 1B,mLST8/GbL – G protein b-subunit like protein, mSin1 – mam-malian stress-activated protein kinase-interaction protein 1,mTOR – mammalian target of rapamycin, mTORC1 – mam-malian target of rapamycin complex 1, mTORC2 – mammaliantarget of rapamycin complex 2, OGD – oxygen-glucose depra-vation, PAS – preautophagosomal structure, PCD – pro-grammed cell death, PDK1 – phosphoinositide-dependent pro-

tein kinase-1, PE – phosphatidylethanolamine, PERK – proteinkinase RNA-like endoplasmic reticulum kinase, PI – propid-ium iodide, PI3K – phosphatidylinositol 3-kinase, PIP3 – phos-phatidylinositol triphosphate, PKCa – protein kinase C a,Raptor – regulatory associated protein of mTOR, Rheb – Rashomolog enriched in brain, Rictor – rapamycin-insensitivecompanion of mTOR, ROS – reactive oxygen species, rt-PA –recombinant tissue plasminogen activator, TRAF2 – Tumor ne-crosis factor receptor-associated factor, TSC1 – hamartin,TSC2 – tuberin, UPR – unfolded protein response

Introduction

Ischemic stroke is the third leading cause of death andmain cause of permanent adult disablement in indus-trialized countries [16, 54]. This type of brain injury

Pharmacological Reports, 2012, 64, 1�15 1

Pharmacological Reports2012, 64, 1�15ISSN 1734-1140

Copyright © 2012by Institute of PharmacologyPolish Academy of Sciences

often occurs during childbirth, during which, it cancause permanent brain damage, resulting in cognitiveand motor dysfunction [25]. The extent of hypoxic-ischemic (H/I) injury depends on the degree of matu-ration of the brain as well as on the severity and dura-tion of the insult [118]. According to actual knowl-edge on pathophysiology of stroke, understood as anischemic cascade that leads to irreversible cerebral in-farction, two major therapeutic strategies are currentlybeing evaluated. The aim of the first one is to restoreblood flow (reperfusion) to the compromised regionthrough the use of thrombolytic, antithrombotic andanti-aggregation drugs [90]. To date, the only onedrug approved for clinical use for the thrombolytictreatment is recombinant tissue plasminogen activator(rt-PA). However, its use is extremely limited due tovery short – three hours – therapeutic window andhigh risk of hemorrhagic complications [46]. The sec-ond therapeutic approach is neuroprotection, under-stood as an action that prevents neuronal cell death byincreasing endogenous cytoprotective response and/orreducing activation of metabolic pathways that lead tocell injury [72]. The concept of neuroprotection instroke is strongly supported by existence of penum-bra, which was confirmed with neuroimaging re-search, the area that surrounds the ischemic core andseparates the core from healthy tissue. In penumbra,gradual decrease of blood flow (but still the reductionis lower than in ischemic core), metabolic distur-bances and activation of many molecular pathwaysare observed. Hence, penumbra is considered as thearea that can be recovered. In the course of time, how-ever, the cells in the penumbra will also die and thecore will expand [33]. Novel and effective neuropro-tective strategy might therefore restrain core expan-sion both by extending penumbra survival for lateruse of reperfusion therapy and reducing ischemic in-flammation and reperfusion injury after successfulreperfusion [26].

Recent studies have revealed that disturbances inautophagy may play an important role in pathogenesisof H/I brain injury as well as neurodegenerative dis-eases, including those of Alzheimer’s, Parkinson’sand Huntington’s [3, 79]. Autophagy (cellular self-digestion) is defined as an evolutionary conservedprocess in which damaged, death or worn out cellularcomponents are degraded [68]. During this complexmechanism that involves endosomal-lysosomal path-way, excess, old and unneeded cytoplasmic macro-molecules (including long-lived proteins) and organ-

elles (mitochondria, peroxisomes, Golgi apparatus,endoplasmic reticulum) are digested by lysosomal en-zymes [98].

Autophagy is observed both under normal growthand cell differentiation, as well as under pathologicalconditions such as: starvation, bacterial infections,presence of misfolded proteins and damage organelles[49, 53, 87]. Furthermore, many studies have shownthat autophagy can protect cells from death by apop-tosis [94]. Besides autophagy’s cellular housekeepingand death-preventing role, it can lead to cell death aswell [18]. For that reason, autophagy is often de-scribed as Type II Programmed Cell Death (PCD) – todistinguish it from apoptosis – Type I PCD [97].

The role of autophagy in ischemic neurons remainscontroversial. It is clear, however, that determining thesignaling pathways that lead to autophagy activationduring ischemia, as well as mechanisms that regulateautophagy-apoptosis interaction, may contribute to thedevelopment of new therapeutic strategies.

Morphological and molecular aspects

of autophagy

The term “autophagy”, which literally means “self-eating”, was used for the first time in 1960’s [45].Nowadays, on the basis of the mechanism used fordelivery of intracellular cargo to lysosomes, three dif-ferent types of autophagy have been described: micro-autophagy, chaperone-mediated autophagy (CMA)and macroautophagy. Both chaperone-mediated auto-phagy and macroautophagy were identified in mam-mals as processes that play a role in central nervoussystem (CNS) damage and diseases. Chaperone-mediated autophagy is a highly selective pathway forthe degradation of misfolded, oxidized or damagedcytosolic proteins containing a targeting motif bio-chemically related to the pentapeptide KFERQ [41].The KFERQ motif is generated in about 30% of cyto-solic proteins by posttranslational modifications suchas deamination [20]. This motif is recognized by cyto-solic chaperone Hsc70 (heat-shock constitutive pro-tein 70), what allows unfolding of substrate protein.Subsequently, chaperone Hsc70-substrate proteincomplex binds to located on lysosomal membrane –receptor for CMA substrates – lysosome-associatedmembrane protein 2a (LAMP2a) [20, 41]. A second

2 Pharmacological Reports, 2012, 64, 1�15

chaperon Hsc70 (lyHsc70), located in the lumen oflysosome, is required for substrate translocationthrough pore made of LAMP2 oligomers. Afterreaching the lysosomal lumen, substrate is rapidly de-graded to amino acids by hydrolases [18].

Macroautophagy, unlike chaperone-mediated auto-phagy, is a bulk degradation system that involves for-mation, maturation and degradation of autophagyvacuoles (autophagosomes). Commonly, and also inthis review, the term “macroautophagy” is unequivo-cal with autophagy.

Autophagy can be divided into four phases: (i) in-duction that consists of enclosure of a portion of cyto-plasm by the isolation membrane also known as thephagophore (nucleation); (ii) conversion of pha-gophore to enclosed, double-membrane autophago-some that engulfs the cytoplasmic materials and or-ganelles, (iii) maturation of autophagosome by fusionof its outer membrane with lysosomes to build auto-phagolysosome (termed also autolysosome) and (iv)degradation of the inner membrane and the luminalcontent of autophagolysosomes by lysosomal hydro-lases [24, 74] (Fig. 1).

The process of autophagy is controlled by familyof evolutionary conserved Atg (AuTophaGy-related)genes. So far, 32 Atg genes have been described in

yeast [31], and 14 homologues of yeasts Atg geneswere identified in mammals [44]. The vast majority ofthem participate in initiation of autophagosome for-mation (Atg1 to 10, Atg12 to 14, Atg16 to18, Atg29,Atg31). The process depends on the concerted actionsof autophagy machinery that involves Atg proteinscascade based on conjugation mechanisms [106].

Phagophore membrane is probably initiated pri-marily from the endoplasmic reticulum (ER) in dy-namic equilibrium with trans-Golgi and late endo-somes [4, 111]. In the induction area accumulation ofAtg proteins is observed [89]. It seems that the trans-membrane protein Atg9 is essential for autophago-some formation. It was shown that Atg9 cycles be-tween the trans-Golgi network and late endosomes,probably carrying sources for phagophore membraneelongation [77]. Furthermore, there are data that Atg9is absent from the completed vesicles, indicating thatit is retrieved prior to the fusion of those vesicles withthe vacuole [42]. The retrieval of Atg9 from preauto-phagosomal structure (PAS) depends on the ser-ine/threonine protein kinase Atg1 (ULK1 in mam-mals). Activity of Atg1 is presumably regulatedthrough protein-protein interactions, primarily withthe phosphoprotein Atg13 [78, 89]. Hyperphosphory-lated Atg13 (e.g., during normal, growth conditions)


Autophagy in cerebral ischemiaBo¿ena Gabryel et al.

Fig. 1. Steps involved in the auto-phagy process. Induction and nuclea-tion of isolation membrane requirespresence of Beclin1/Vps34/p150 pro-tein complexes. The process of pha-gophore elongation involves two ubi-quitin-like conjugation systems. Cova-lent conjugation of Atg12 to Atg5 ismediated by Atg7 (E1-like enzyme)and Atg10 (E2-like enzyme). TheAtg12-Atg5 conjugate binds to Atg16,which in turn undergoes tetrameriza-tion creating a multi-conjugate com-plex. Second pathway starts with theaction of cysteine protease Atg4,which proteolytically removes the C-terminal arginine (R) of LC3 I. Atg7(E1-like enzyme) activates exposedglycine residue (G) and transfers acti-vated LC3 to Atg3 (E2-like enzyme),which mediates the covalent conjuga-tion of phosphatidylethanolamine (PE)to LC3. LC3 II-PE is recruited to form-ing vesicle depend on Atg12-Atg5-Atg16 complexes. LC3 II on the vesi-cle outer membrane is cleaved off byAtg4 protease. Autophagosomes un-dergo maturation by fusion with lyso-somes to form autolysosomes, inwhich autophagic cargo as well as in-ner membrane are degraded by lyso-somal enzymes

has a low affinity for Atg1. However, upon conditionssuch as nutrient starvation, hypoxia, growth factors orenergy depletion, Atg13 is rapidly dephosphorylated,resulting in its higher affinity for Atg1 and inductionof autophagy due to triggering Atg proteins cascade.Following induction, a phagophore or isolation mem-brane is formed (nucleation). This process is regulatedby Atg1, Atg9 and the protein complex, that includesa class III phosphatidylinositol 3-kinase (PI3K)(Vps34), Beclin 1 (the mammalian homolog of yeastAtg6) and p150 (mammalian Vps15) [40]. Vps34 ki-nase is involved in regulation of autophagy only whencomplexed with Beclin 1 and other regulatory pro-teins. Vps34-Beclin 1 interaction promotes kinasecatalytic activity and increases production of phos-phatidylinositol triphosphate (PIP3), which is essen-tial for phagophore elongation and allows the recruit-ment of other Atg proteins to phagophore [31].

For elongation phase two ubiquitin-like conjuga-tion systems are essential. First one requires the for-mation of an isopeptide bond between a C-terminalglycine of Atg12 and an internal lysine of Atg5 [99].This step is mediated by Atg7, a homologue of ubiq-uitin carrier protein E1. Atg7 induces the activation ofAtg12 through hydrolysis of ATP. During the nextstep, activated Atg12 is transferred to the Atg10(which acts like E2) and eventually conjugates toAtg5 [31]. Subsequently, Atg16 protein dimer(Atg16L) binds noncovalently to Atg5-Atg12 conju-gate to form multimeric Atg5-Atg12-Atg16L complex[7]. It is suggested that Atg5-Atg12-Atg16L com-plexes are required for induction of the curvature ofgrowing phagophore through asymmetric recruitmentof processed product of second conjugation systemthat involves conjugation of phosphatidylethanola-mine (PE) to the LC3 protein, mediated alternately byAtg4, Atg7 (E1-like enzyme) and Atg3 (E2-like en-zyme) [62]. This conjugation leads to conversion ofsoluble, cytoplasmic LC3-I form to the lipidated andlocalized in autophagosomal membranes LC3-IIform. Level of LC3-II is tightly correlated withnumber of autophagosomes and for that reason it isconsidered as the most reliable marker of active auto-phagosomes and autophagolysosomes [112]. LC3 hasalso been proposed to act as a receptor for p62, alsoknown as sequestosome-1 or SQSTM1, a protein thattargets poly-ubiquitinated protein aggregates for deg-radation at the autophagolysosomes [70]. p62 is re-cruited to the autophagosome via evolutionarily con-

served cargo receptor-binding domain of LC3 andtherefore it is constantly degraded by autophagy sys-tem thus level of its degradation is often use as anautophagic marker [70].

As a result of complete induction, nucleation andelongation phases a mature, double-membrane auto-phagosome is formed. Its size varies among mammal-ian cells usually from 0.5 to 1.5 µm [68]. LC3-II re-mains coupled to the autophagosome membrane untilfusion with the lysosome and autophagolysosome for-mation. The fusion is mediated by SNARE and Rabproteins, particularly Rab7, which is required forautophagolysosome maturation [24].

Basal autophagy in neurons

Mizushima [67] proposed a sub-classification of auto-phagy into “basal” and “induced”, depending on itsrole. Autophagy occurs at a low, basal level in mostcells and from physiological point of view is impor-tant for constitutive turnover of proteins or cytoplas-mic components and selective removal of damagedorganelles (e.g., mitochondria or peroxisomes) [98].Elimination of unnecessary cells during embryo de-velopment is also an example of basal autophagy.Moreover, it was proposed that basal autophagy maybe a determinant of life span [110].

Proper course of basal autophagy in neurons deter-minates maintenance of axonal homeostasis throughretrograde transport of autophagosomes with theircargo to the soma, where their fusion with lysosomestakes place [115] (Fig. 2). On the contrary, pathologi-cal conditions (e.g., stress or injury) might induceautophagy, what is observed as a local intensificationof autophagosomes biosynthesis and their accumula-tion in axons. Degradation, that takes place in axonsduring induced autophagy requires access to lyso-somes, what forces their anterograde transport andmight result in untimely autophagosomes degrada-tion. Furthermore, axons deficient in autophagy andimpaired in autophagic cargo transport might amassproteins, organelles and aberrant membrane structuresin axons, which could result in axonal dystrophy ordegeneration [115] (Fig. 2).

It is generally accepted that in normal conditionsneurons display low levels of autophagosomes andperhaps a low rate of autophagosomes biosynthesis


[115]. The low level of basal autophagy in neurons isprobably associated with their ability to use not onlyglucose, but also ketones as an energy source [9]. Fur-thermore, under conditions of extreme energy defi-ciency (such as ischemia) neighboring astrocytes,which metabolize glucose into lactate through glyco-lysis and contain glycogen stores, can provide a short-term energy supply for neurons. Moreover, astrocytessecrete growth factors and neuropeptides what alsocontributes to their neuroprotective action [29]. Lowrate of basal autophagy in neurons might also resultfrom some neuron-specific proteins and/or proteinmodifications. One of the examples is microtubule as-sociated protein 1B (MAP1B), which binds to LC3with high affinity and affects the formation of auto-phagosomes [115]. However, alternative thesis alsoexists, it assumes that basal level of autophagy in neu-rons is as high as in other cell types, but newly formedautophagosomes are efficiently eliminated throughrapid fusion with lysosomes, thereby building-up ofautophagic intermediates is avoided and detection ofautophagosomes is almost impossible [8].

Neuronal autophagy is regulated by a few parallel,intracellular signaling pathways that are involved intransition from basal level (specific for neurons) tohighly conserved induced level (activation). Recentlypublished results of studies performed on primarycortical neurons cultures indicate that active insulinsignaling pathway is implicated in that role [113].

Also DJ-1/PARK4 and p53 proteins were identified aspotential regulators of neuronal autophagy. Krebiehlet al. [51] used an in vitro model of Parkinson’s dis-ease to show that E64D mutation in the DJ-1/PARK7gene leads to reduced basal autophagy and accumula-tion of dysfunctional mitochondria [51]. In turn, p53protein seems to have dichotomous functions: in nor-mal, metabolic conditions it acts as a negative regula-tor, whereas in response to acute stress activated p53triggers autophagy. As it was shown on human andmice cell lines (including neuroblastoma SH-SY5Y),basal role of p53 in metabolic conditions is to suppressautophagy [93], whereas, when cells are exposed to nu-trient depravation or oncogenic or else genotoxicstress, p53 acts as an autophagy inducer [55].

Defective basal autophagy has been implicated inthe pathogenesis of several neurodegenerative dis-eases [15]. This has been proven by the latest resultsof studies on transgenic mice with neuronal-cell-specific deletion of Atg5 (Atg5flox/flox; nestin-Cre) [37].Mice deficient in Atg5 showed growth retardation anddeveloped progressive motor and behavioral deficits.Moreover, accumulation of large, ubiquitin-positive(marker of misfolded proteins) inclusion bodies in cy-toplasm of large neurons in the thalamus, pons, me-dulla, dorsal root ganglion and midbrain was ob-served. Similar abnormalities and marked atrophy ofthe cerebral cortical region, implicated into neurode-generation, were observed in Atg7 flox/flox; nestin-Cre



Fig. 2. Significance of effective auto-phagy in neurons. Under physiologicalconditions autophagosomes are trans-ported retrogradely to the soma wherelysosomes perform degradation. Underpathological conditions lysosomes aretransported to axons due to locally in-duced and increased autophagy. Neu-rons that are impaired in autophagyaccumulate autophagic substrates,what disrupts axonal homeostasis andresults in axonal dystrophy and subse-quent neurodegeneration

mice [48]. These data strongly suggest that effectiveclearance of diffuse cytosolic proteins through basalautophagy prevents the accumulation of abnormalproteins, which can disrupt neuronal function and ul-timately lead to neurodegeneration.

Induced autophagy in neurons

Autophagy in CNS is activated not only by cerebralH/I but also by nutrient deprivation, neurotoxins, ex-citotoxic stimuli, closed head injuries and neuro-pathogenic pathways [55, 66, 115]. The primary roleof this activation is to provide proper cellular re-sponse for nutrients limitation via activation ofautophagosome-lysosome degradation mechanism.During the process of autophagy induction, neuronsundergo changes leading to deregulation of autophagymechanism that allows the transition from basal levelto the induced (activated) level, when intensificationof autophagosomes biosynthesis occurs [115].

Data gathered so far, unquestionably showed thatautophagy is controlled by a few regulators, which in-

duction depends on nutrients availability such as: in-sulin, amino acids and AMP activated kinase (AMPK)acting via serine-threonine protein kinase mTOR(mammalian target of rapamycin) [116] (Fig. 3).

The idea that mTOR is involved in negative controlof autophagy is now generally accepted. mTOR playsmultiple role in CNS and is involved in regulation ofcell viability, differentiation, transcription, translation,protein degradation, actin cytoskeletal organizationand autophagy [38]. In mammalian cells mTOR as-sembles into two functionally distinct protein com-plexes: mTORC1 and mTORC2. The first one acts asa regulator of autophagy and is highly sensitive to ra-pamycin (specific mTOR inhibitor). The latter one isinsensitive to rapamycin administration and its role isto regulate actin cytoskeletal organization and activityof PKCa and Akt kinases. Both complexes consist ofmTOR kinase and mLST8/GbL protein (G proteinb-subunit like protein). Additionally mTORC1 con-tains Raptor (regulatory associated protein of mTOR)and mTORC2 Rictor (rapamycin-insensitive compan-ion of mTOR) and mSin1 [105].

Mechanism by which mTOR acts as a negativeregulator of autophagy is not fully understood. How-ever, it seems that stimulated by insulin, phosphatidyl-


Fig. 3. Regulation of induced auto-phagy in neurons. Transition from ba-sal into induced autophagy could bemediated by multiple pathways. ERstress triggers autophagy via PERK/eIF2a pathway that up-regulatesAtg12 and via pathway that involvesactivation of IRE1, recruitment ofTRAF2 and activation of JNK. Further-more, increased [Ca2+]i and/orAMP/ATP ratio activate CaMKK andLKB1 kinases, respectively, what inturn phosphorylates and activatesAMPK kinase. AMPK mediates the in-duction of autophagy through directactivation of ULK1 or by inhibition ofmTOR (mTORC1) by phosphorylationof Raptor or TSC2. Activity of mTOR isalso regulated by insulin, which stimu-lates generation of PIP3 from PIP2 byPI3K. Increase in PIP3 results in theactivation of Akt by PDK1 and mTOR(mTORC2). Activated Akt kinase inhib-its the activity of TSC1/TSC2 complex,what reduces the GTPase activity ofRheb and leads to activation of mTORand subsequent inhibition of auto-phagy. Amino acids also lead to de-creased Rheb GTPase activity andthus mTOR activation and autophagyinhibition. Bcl-2 negatively regulatesautophagy through association withBeclin1 and thereby inhibition of Be-clin1/Vps34 complexes formation. ®activation, z inhibition

inositol 3-kinase (PI3K) pathway is the key pathwaythat contributes to mTOR activation. PI3K stimulatesproduction of PIP3 and PIP3 in turn recruits phos-phoinositide-dependent kinase-1 (PDK1) and Akt ki-nase to the membrane [105]. The Akt kinase is acti-vated through phosphorylation at two sites: specificthreonine residue (Thr308) by PDK1 and specific ser-ine residue (S473) by mTORC2 complex [80]. Acti-vated Akt kinase phosphorylates and thus inhibits theactivity of hamartin (TSC1) and tuberin (TSC2) pro-tein complex (tuberous sclerosis complex, TSC1/TSC2), the main negative regulator of mTOR. Inacti-vation of TSC1/TSC2 complex reduces the GTPaseactivity of small GTPase Rheb (Ras homolog en-riched in brain). Numerous studies have shown thatRheb-GTP strongly stimulates activity of mTOR bothin vivo and in vitro [73]. Besides insulin signalingpathway, mTOR can also be activated by amino acids.This activation also occurs via decreased RhebGTPase activity [17]. Therefore, it is generally ac-cepted that both activation of PI3K pathway by insu-lin signaling and direct phosphorylation induced byamino acids are required for developing effective en-zymatic activity of mTOR and subsequent autophagysuppression.

Contrary to insulin and amino acids, active AMPKkinase activates autophagy via suppression of mTOR

signalization pathway (Figs. 3 and 4). AMPK acts asa sensor of cellular energy and regulates metabolicpathways by switching on catabolic pathways to gen-erate ATP and switching off ATP-consuming anabolicpathways [96], what enables optimization of intracellu-lar energetic processes and cell survival during stressperiods. In response to increased cytosolic AMP lev-els and/or decline in ATP (in lesser extent) produc-tion, AMPK is activated through phosphorylation ata specific threonine residue (Thr172) by its upstreamkinases: LKB1 (serine threonine kinase 11) and Ca2+/calmodulin-dependent protein kinase b (CaMKKb)[96]. Active AMPK intensifies autophagy through in-hibition of mTORC1 complex via direct phosphoryla-tion of at least two proteins: TSC2 at specific and dif-ferent from other up-stream kinases serine residuesand mTOR binding partner – Raptor [36, 56]. Further-more, recent results suggest that AMPK might acti-vate autophagy also by direct interaction with Atgproteins machinery. Kim et al. [43] showed that underglucose starvation, AMPK promotes autophagy by di-rect activation of Ulk1 through phosphorylation ofSer 317 and Ser 777 [43].

Participation of AMPK in autophagy regulation wasconfirmed during an in vitro experiment in which itwas shown that glucose deprivation-induced autophagyis inhibited in the presence of dominant negative



Fig. 4. Cerebral ischemia-inducedautophagy. Autophagy activation un-der hypoxic/ischemic conditions is as-sociated with presence of differentstress factors like: enhanced ROSgeneration, increased cytosolic Ca2+and AMP concentrations, decrease inATP, insulin and amino acids supply.ER stress is induced when unfoldedproteins accumulate in the ER or cyto-sol. Impaired capacity of the ER to foldproteins results from disturbance inCa2+, oxygen and glucose homeosta-sis. ER stress activates autophagythrough signaling pathways involvedin UPR response. Additionally, in-crease in Ca2+ intracellular concentra-tion associated with ER-Stress or exci-totoxity can activate autophagy via

AMPK-dependent pathway. A de-crease in insulin and amino acids lev-els leads to reduction of activity of themajor autophagy inhibitor – mTORC1

AMPK [63]. In addition, it was proved that compoundc (6-[4-(2-piperidin-1-yl-etoxy)-phenyl)]-3-pyridin-4-yl-pyrazolo[1,5-a]pyrimidine), specific AMPK in-hibitor, effectively restrained Ca2+ induced autophagyin vitro [39].

In the most recently published data, there appearsmore and more evidence that ER stress and mitochon-dria dysfunction might also act as strong autophagyinducers (Figs. 3 and 4). Disruptions in proper ERfunctioning, involving reduction of posttranslationalmodifications ability and protein folding control, arecommonly termed as unfolded protein response(UPR). Moreover, all mentioned above autophagy in-ducers (including brain hypoxia/ischemia) are alsoknown for their ability to induce ER stress [40]. It isnow accepted that two UPR signaling pathways con-tribute to ER stress-induced autophagy: PERK/eIF2aand Ire1/TRAF2/JNK [40, 61]. It seems that eIF2a(eukaryotic initiation factor 2a) triggers autophagy bythe induction of Atg12 expression and finally LC3-Ito LC3-II conversion [50]. Increased [Ca2+] has alsobeen implicated in ER stress-induced autophagy pos-sibly via CaMKK/AMPK/mTORC1 pathway [40].

ER stress-induced autophagy might be inhibited byanti-apoptotic protein Bcl-2 localized in ER membrane[40]. At least two mechanisms have been proposed forthis suppression. First one supposes that Bcl-2 inhibitsautophagy possibly by lowering of the steady-state levelof Ca2+ in the ER ([Ca2+]ER) and thereby reducesstimulus-induced Ca2+ flux from the ER. Such reductionprevents CaMKK activation and thus CaMKKK/AMPK/mTORC1 pathway activation. Second scenarioassumes that Bcl-2 directly interacts with Beclin 1 andthereby avoids autophagosomes formation due to in-complete formation of PI3K complex [40].

Many studies have suggested that there is a relationbetween mitochondrial dysfunction and autophagy.Disorders of mitochondrial function associated withthe excess ROS generation are leading activators ofautophagy as a consequence of stroke, traumatic in-jury and neurodegenerative diseases (Fig. 4).

Excess of ROS leads to DNA damage, lipid peroxi-dation and lysosomal membrane rupture/permeabili-zation [108]. The lysosomal rupture/permeabilizationenables interaction between cathepsins released fromlysosomal lumen and pro-apoptotic Bcl-2 familymembers (Bax, Bak, tBid) and thus leads to the induc-tion of mitochondrial apoptosis pathway via release ofcytochrome c and activation of caspases and DNase II[92]. ROS might also affect the molecular mechanism

of autophagy. It was shown that H2O2 generated inoxidative conditions causes inactivation of cysteineprotease Atg4 at the site of autophagosome formation,thereby promoting LC3 conversion – an essential stepin the process of autophagy [81].

Autophagy, apoptosis and necrosis in

cerebral ischemia

Symptoms of brain injury caused by ischemia due toclot or thrombosis are the consequence of a massivecell death in the infract area. Within seconds after theloss of blood flow, so called ischemic cascade is initi-ated, which comprises a series of subsequent bio-chemical events that lead to degradation of cell mem-branes and structures and eventually to neuronal death[54, 103]. Patients experiencing acute ischemic strokewill lose 120 million neurons each hour [54].Ischemic cell death is mediated mainly by necroticand apoptotic pathways. Markers of both these pro-cesses were frequently observed in experimentalstroke models both in vivo and in vitro [58]. Thesetwo types of cell death are significantly different fromeach other at the morphological and molecular level,and intracellular ATP levels are a primary determinantof cell fate. Necrosis represents a passive form of celldeath, that occurs when cells are exposed to acutestress or extreme energy depletion. Necrosis is char-acterized by rapid cell and its organelles swelling andloss of membrane integrity, what in turn leads to re-lease of various inflammation inducing factors andcauses damage in neighboring tissues [28, 108]. Untilrecently, necrosis was considered as chaotic and non-regulated process. The results of the latest experi-ments suggest, however, that its occurrence andcourse might be tightly regulated by such signals as:mitochondrial dysfunction, enhanced generation ofROS, ATP depletion, proteolysis by calpains andcathepsins and early plasma membrane rupture [32].Necrotic cell death due to stroke is observed primarily inneurons located in the core, that are directly and drasti-cally subjected to limited blood flow and deleterious ac-tions of agents released from adjacent cells [69].

Morphological changes characteristic for cells under-going apoptosis are: nuclei shrinkage, DNA fragmenta-tion and subsequent collapse of the cell into small mem-brane enclosed fragments called apoptotic bodies.


Apoptotic cell death is possible only in condition ofsufficient ATP availability, what ensures the mainte-nance of proper ion channels function and preservesintact cells integrity [75]. Apoptosis, therefore, is notaccompanied by inflammatory response or surround-ing tissue damage [91]. After a stroke, apoptosis isobserved mainly in neurons located in the ischemicpenumbra, a zone of tissue less severely affected byreduced oxygen and nutrients supply in comparison tothe core [10].

In regard to the progress in understanding the pa-thophysiology of brain ischemia, Sharp et al. [85] pro-posed a differentiation of penumbra to several zones,on the basis of molecular events that dominate within:(i) zone of selective neuronal death, adjacent to in-fract, (ii) zone of protein denaturation, where induc-tion of heat shock proteins (HSP) occurs, (iii) zonethat is chronically defective in perfusion and reperfu-sion, where hypoxia-inducible factor 1 (HIF-1) is ex-pressed and the most external (iv) zone induced byspreading depression or repeated ischemic depolariza-tion of a large number of early genes, mainly c-fos [85].

It seems, however, that mechanisms that underlayischemic cell death as well as their kinetics are muchmore complex, than we initially thought. Especially,since it was shown that next to apoptotic and necroticfeatures, some cells in the area challenged with ische-mia demonstrate the presence of protein markerscharacteristic for autophagy, such as Beclin 1 andLC3-II [118]. Accelerated number of autophago-somes and autophagic proteins expression were ob-served in experimental models of focal [1, 102] andtransient [59] cerebral ischemia. Autophagy inducedby focal ischemia was observed in cerebral cortex,hippocampus and striatum of animals that sufferedfrom brain ischemia [1, 76]. Similar results were ob-tained during in vitro experiments performed on pri-mary cortical neuronal cultures challenged with ische-mia [65] and on hippocampal cell-lines (HT22) ex-posed to serum deprivation [88]. Evidences forischemia-induced autophagy were further provided byin vivo imaging studies on mice subjected to transientmiddle cerebral artery occlusion (tMCAO) [95].

Moreover, the differential response of neurons tothe ischemic challenge may be proved by the fact thatsome cells simultaneously activate both apoptotic andautophagic pathways. In a rat model of temporary fo-cal cerebral ischemia a penumbral subpopulation ofthe cortical and striatal neurons presented simultane-ously both up-regulation of Beclin 1 and activation of

caspase-3 [76]. Recent study by Carloni et al. [12]performed on neonatal rats subjected to ligation ofa right common carotid artery, followed by hypoxia,also revealed that some Beclin 1-positive cells wereTUNEL-positive as well. Cells that have co-localizedapoptosis and autophagy simultaneously were foundmainly in the superficial layers of the cerebral cortex.In the deeper layers of the cortex and in the CA1 re-gion of the hippocampus levels of cells expressingboth markers were much lower. Importantly, only sev-eral both Beclin 1 and propidium iodide (PI – necrosismarker) positive cells were found in the CA1 andCA2 areas of the hippocampus – what strongly sug-gests that increased protein expression occurs mainlyin cells that induce apoptosis pathway [12]. Neuronsthat presented both enhanced level of autophagy andapoptotic features were also observed in a rat modelof severe perinatal asphyxia [30]. The study also con-firmed that autophagy-apoptosis relation varies ac-cording to the cerebral area – it seems that in the cor-tex those processes are closely linked to each other,while in the hippocampus they rather act independ-ently. On the basis of published data and their ownstudies Rami and Kögel [75] introduced term“apophagy” – a kind of apoptotic cell death with con-comitant upregulation of Beclin 1 and activation ofcaspase-3.

There are many evidences that autophagy andapoptosis may share common molecular inducers andregulatory mechanism. Furthermore, autophagic path-ways overlap with the apoptosis through the involve-ment of both autophagic and apoptotic proteins [62].It was shown that Atg5, in addition to its essential rolein autophagy, also enhances susceptibility towardsapoptotic stimuli. Moreover, calpain-mediated cleav-age of Atg5 promotes cytochrome C release and cas-pases activation and thus induces autophagy to apop-tosis switch [114]. Another example is Beclin 1,which possesses a BH3 domain interacting with anti-apoptotic protein Bcl-2. Interestingly, binding ofBcl-2 to Beclin 1 inhibits autophagy [62] (Fig. 3). Re-cently, it was shown that caspase-mediated cleavageof Beclin-1 not only inhibits autophagy but also en-hances apoptosis by promoting the release of proa-poptotic factors from mitochondria [104]. Addition-ally, latest results obtained in study by Grishchuk etal. [34] clearly reveled that also Beclin 1-independentautophagy contributes to both caspase-dependent and-independent components of neuronal apoptosis.



Role of autophagy in cerebral ischemia

At present stage of knowledge it is not possible to une-quivocally answer the question whether ischemia-inducedautophagy contributes to neuron death, or reflects to anendogenous neuroprotection mechanism, or maybe is justan epiphenomenon of apoptosis and necrosis? On the ba-sis of data published so far, however, it is possible to indi-cate three main thesis concerning the potential impact ofautophagy process on ischemic neurons.

Firstly, it seems that autophagy in brain ischemiamight contribute to neurite degeneration. These thesiswas confirmed by the recent research of Adhami et al.[1], where rapid and massive degeneration of corticalneuronal axons after 24 h of ischemia/hypoxia wasobserved. It was also showed that AMPK mediatedautophagy could contribute to enlarged ischemic areaand worse outcome [22, 56]. In turn, in mice withneuronal deletion of Atg7 (Atg7 flox/flox, nestin-Cre)a week after the occlusion of the left common carotidartery, followed by hypoxia, reduced loss of the py-ramidal neurons in the hippocampus and a great re-duction of ischemia damaged area were observed[47]. Pharmacological inhibition of autophagy like-wise resulted in attenuated focal cerebral ischemia-associated neural damage in rats [52]. Moreover, Wenet al. [102] showed that single, intracerebral ven-tricule injection of 3-methyladenine (3-MA) – a spe-cific inhibitor of Vps34 kinase – directly after perma-nent occlusion of the rat’s middle cerebral artery(pMCAO) significantly reduced infract volume [102].Similar conclusions can be made on the basis of thestudy on the in vitro ischemia model, where inhibitionof autophagy with 3-MA significantly increased corti-cal neurons viability [65]. Recently, it was also shownthat inhibition of autophagy with 3-MA in rats sub-jected to severe global ischemia prevented the onsetof programmed necrosis in hippocampal CA1 neuronsand thus contributed to neuroprotection, suggestingthat detrimental role of autophagy in cerebral ische-mia might be related with autophagy-necrosis cross-talk [100]. Moreover, neuroprotectants such as lith-ium and glial cell line-derived neurotrophic factor(GDNF) were also shown to have strong both anti-autophagic and antiapoptotic effect [57, 84].

Secondly, however, it was repeatedly demonstratedthat autophagy plays an important role in protectingneurons from ischemia-induced death. Inhibition ofautophagy with 3-MA and bafilomycin A1 (an inhibi-

tor of autophagosomes-lysosomes fusion) dramati-cally enhanced the protein level of cleaved caspase-3and increased cell mortality in hippocampal cell-linesHT22 exposed to serum deprivation [88]. In turn, inrat model of neonatal H/I induced brain injury, treat-ment with rapamycin – a pharmacological autophagyinducer – caused augmented Beclin 1 expression incerebral cortex and hippocampus, reduced necroticcell death and decreased cerebral injury [12]. Benefi-cial effects of rapamycin were also observed in micemodel of traumatic brain injury [23]. Neuroprotectiveeffect of rapamycin is probably associated with acti-vation of PI3K/Akt/mTOR axis and phosphorylationof transcriptional factor CREB (cAMP response ele-ment binding protein) [13]. Increased Beclin 1 ex-pression and a long-lasting neuroprotective effect wasalso observed as a result of simvastatin administrationto neonatal rats before the onset of the ischemic pro-cedure [5, 6]. Autophagy is also a possible mecha-nism that underlies the neuroprotective effect of mela-tonin [35]. Moreover, recently it was shown that ad-ministration of SB216763, an inhibitor of a serine/threonine kinase GSK-3b – the main player in theneurodegeneration, results in suppression of neuroin-flammation due to increased autophagy activation inthe brain cortex of rats subjected to pMCAO [117].

Currently it seems that elimination of damaged mi-tochondria (mitophagy) and subsequent apoptosis in-terruption could represent the most important mecha-nism responsible for neuroprotective effects of auto-phagy in cerebral ischemia. Selective elimination ofmitochondria has been reported after blockade of cas-pases, thus sequestration of organelles might indicatemechanism that prevents apoptosis [107]. Neverthe-less, elimination of mitochondria was partially sup-pressed by the autolysosomal inhibitor bafilomycinA1, what strongly supports the suggestion that auto-phagy may be involved in that process [107]. Further-more, elimination of damaged mitochondria in auto-phagolysosomes might represent an adaptive meta-bolic response to hypoxia, mediated by transcriptionalfactor HIF-1, which induction is observed in the zone(iii) of penumbra – see above [83]. Therefore, mito-phagy in cells exposed to hypoxia is currently consid-ered as an adaptive mechanism that is necessary tomaintain redox homeostasis and cell survival [83].However, not only mitochondria, but also ROS-damaged ER fragments are sequestrated in autopha-golysosomes, what prevents the release of calciumstorages into the cytoplasm and subsequent activation


of inflammatory caspase 11-dependent apoptosis [21].Active caspase 11 is implicated in release of inter-leukin 1b, thus it activates an inflammatory responseand might contribute to ischemic injury extension [27,101]. Then, the proper sequestration of damaged, dueto RFT activity, ER fragments could prevent neuroin-flammation and/or neurons apoptosis [59]. It is of ut-most importance, when considering autophagy as anendogenous neuroprotective mechanism that onsetsafter cerebral ischemia, to mention that also damagedproteins and/or aggregated proteins are beingremoved in that process. Liu et al. [59] using the2-vessel occlusion model in rats revealed that failureof the autophagy pathway results in the accumulationof protein aggregate-associated organelles (ER, Golgiapparatus, mitochondria), what in turn increases cel-lular stress and can lead to delayed neuronal death[59].

Probably, autophagy activation is also associatedwith the neuroprotective effect of the ischemic pre-conditioning (IPC), a phenomenon that makes neu-rons more tolerant to ischemia, due to former brief(and less intensive) exposure to ischemia episodes[11, 14, 60]. Park et al. [71] reported enhanced PC12cells viability accompanied by increased synthesisand degradation of autophagosomes after IPC in vitro

[71]. Moreover, the addition of 3-MA to the cultureincreased the proportion of necrotic and apoptoticcells after preconditioned lethal OGD. Similar sup-pression of neuroprotective effect was also observedas a result of 3-MA and bafilomycin A1 treatment inrat model of IPC [86]. Elevated autophagic activitycontributing to increased tolerance against transientfocal cerebral ischemia was also observed as a resultof hyperbaric oxygen preconditioning [109].

Thirdly, autophagy is a catabolic process and there-fore it might delay the onset of necrosis through main-taining ionic homeostasis [2]. Importantly, mTOR ki-nase – a key autophagy regulator – acts as a sensor ofthe changes in the intracellular ATP concentration[19]. Probably, changes in ATP are transmitted tomTOR by active AMPK, because even a small de-cline in cytosolic ATP, results in a relatively large in-crease in AMPK concentration through the adenylatekinase activation. Activation of AMPK kinase inhib-its mTOR-dependent signaling pathways, inducesautophagy and turns off ATP-dependent metabolicpathways [17]. Moreover, it is worth to mention theperfect mitochondrial localization of two enzymes:mTOR associated with the outer membrane and

adenylate kinase in the intermembrane space, whatenables rapid and integrated reaction for changes incellular AMP/ATP ratio [64, 82]. However, if ener-getic deficits during reperfusion are not corrected,a high level of “autophagic stress” will lead to mas-sive lysosomal activation and finally necrotic neurondeath [2].

On the basis of the aforementioned data, we mayspeculate that autophagy, apoptosis and necrosis co-exist in the ischemia challenged area and that thisthree processes lead to cell death with mixed bio-chemical and morphological features. Though manydata suggest that process of autophagy could be a keydecision-maker that contributes to neurons fate.

Conclusions

Mounting evidence has indicated involvement ofautophagy in the pathomechanism of brain ischemia.It remains controversial whether activation of auto-phagy is a manifestation of endogenous neuroprotec-tive mechanism or quite the opposite, it contributes tocell death. The dominating thesis is, however, thatautophagy acts as a protective mechanism activatedearly as a response for apoptogenic factors. What ismore, it is believed that signaling pathways associatedwith autophagy might represent a potential target ofnew neuroprotective strategies.

Acknowledgment:

The work was supported by the grant No. N N401 072139 (B.G.)

from the Ministry of Sciences and Higher Education, Warszawa,

Poland.

References:

1. Adhami F, Liao G, Morozov YM, Schloemer A,Schmithorst VJ, Lorenz JN, Dunn RS et al.: Cerebralischemia-hypoxia induces intravascular coagulation andautophagy. Am J Pathol, 2006, 169, 566–583.

2. Adhami F, Schloemer A, Kuan CY: The roles of auto-phagy in cerebral ischemia. Autophagy, 2007, 3, 42–44.

3. Alirezaei M, Kemball CC, Lindsay Whitton J: Auto-phagy, inflammation and neurodegenerative disease. EurJ Neurosci, 2011, 33, 197–204.

4. Axe EL, Walker SA, Manifava M, Chandra P, RoderickHL, Habermann A, Griffiths G, Ktistakis NT: Autopha-gosome formation from membrane compartments en-



riched in phosphatidylinositol 3-phosphate and dynami-cally connected to the endoplasmic reticulum. J CellBiol, 2008, 182, 685–701.

5. Balduini W, Carloni S, Buonocore G: Autophagy in hy-poxia–ischemia induced brain injury: evidence andspeculations. Autophagy, 2009, 5, 221–223.

6. Balduini W, De Angelis V, Mazzoni E, Cimino M: Sim-vastatin protects against long-lasting behavioral and mor-phological consequences of neonatal hypoxic/ischemicbrain injury. Stroke, 2001, 32, 2185–2191.

7. Barth S, Glick D, Macleod KF: Autophagy: assays andartifacts. J Pathol, 2010, 221, 117–124.

8. Boland B, Kumar A, Lee S, Platt FM, Wegiel J, Yu WH,Nixon RA: Autophagy induction and autophagosomeclearance in neurons: relationship to autophagic pathol-ogy in Alzheimer’s disease. J Neurosci, 2008, 28,6926–6937.

9. Boland B, Nixon RA: Neuronal macroautophagy: fromdevelopment to degeneration. Mol Aspects Med, 2006,27, 503–519.

10. Broughton BR, Reutens DC, Sobey CG: Apoptoticmechanisms after cerebral ischemia. Stroke, 2009, 40,331–339.

11. Brown JE, Zeiger SL, Hettinger JC, Brooks JD, Holt B,Morrow JD, Musiek ES et al.: Essential role of theredox-sensitive kinase p66shc in determining energeticand oxidative status and cell fate in neuronal precondi-tioning. J Neurosci, 2010, 30, 5242–5252.

12. Carloni S, Buonocore G, Balduini W: Protective role ofautophagy in neonatal hypoxia–ischemia induced braininjury. Neurobiol Dis, 2008, 32, 329–339.

13. Carloni S, Girelli S, Scopa C, Buonocore G, Longini M,Balduini W: Activation of autophagy and Akt/CREB sig-naling play an equivalent role in the neuroprotective ef-fect of rapamycin in neonatal hypoxia-ischemia. Auto-phagy, 2010, 6, 366–377.

14. Chang S, Jiang X, Zhao C, Lee C, Ferriero DM: Exoge-nous low dose hydrogen peroxide increases hypoxiain-ducible factor-1alpha protein expression and induces pre-conditioning protection against ischemia in primary cor-tical neurons. Neurosci Lett, 2008, 441, 134–138.

15. Cherra III SJ, Dagda RK, Chu CT: Autophagy and neu-rodegeneration: survival at a cost? Neuropathol ApplNeurobiol, 2010, 36, 125–132.

16. Chu CT: Eaten alive: autophagy and neuronal cell deathafter hypoxia-ischemia. Am J Pathol, 2008, 172, 284–287.

17. Codogno P, Meijer AJ: Autophagy and signaling: theirrole in cell survival and cell death. Cell Death Differ,2005, 12, 1509–1518.

18. Cuervo AM: Autophagy: in sickness and in health.Trends Cell Biol, 2004, 14, 70–77.

19. Dennis PB, Jaeschke A, Saitoh M, Fowler B, Kozma SC,Thomas G: Mammalian TOR: a homeostatic ATP sensor.Science, 2001, 294, 1102–1105.

20. Dice JF: Chaperone-mediated autophagy. Autophagy,2007, 4, 295–299.

21. Ding WX, Ni HM, Gao W, Yoshimori T, Stolz DB, RonD, Yin XM: Linking autophagy to ubiquitin-proteasomesystem is important for the regulation of endoplasmic re-ticulum stress and cell viability. Am J Pathol, 2007, 171,513–524.

22. Du L, Hickey RW, Bayir H, Watkins SC, Tyurin VA,Guo F, Kochanek PM et al.: Starving neurons showsex difference in autophagy. J Biol Chem, 2009, 284,2383–2396.

23. Erlich S, Alexandrovich A, Shohami E, Pinkas-KramarskiR: Rapamycin is a neuroprotective treatment for traumaticbrain injury. Neurobiol Dis, 2007, 26, 86–93.

24. Eskelinen EL: Maturation of autophagic vacuoles inmammalian cells. Autophagy, 2005, 1, 1–10.

25. Ferrari DC, Nesic O, Perez-Polo JR: Perspectives onneonatal hypoxia/ischemia-induced edema formation.Neurochem Res, 2010, 35, 1957–1965.

26. Fisher M: New approaches to neuroprotective drug de-velopment. Stroke, 2011, 42, 24–27.

27. Fradejas N, Pastor MD, Burgos M, Beyaert R, TranqueP, Calvo S: Caspase-11 mediates ischemia-induced astro-cyte death: involvement of endoplasmic reticulum stressand C/EBP homologous protein. J Neurosci Res, 2010,88, 1094–1105.

28. Gabryel B, Bielecka A, Bernacki J, £abuzek K, HermanZS: Immunosuppressant cytoprotection correlates withHMGB1 suppression in primary astrocyte cultures ex-posed to combined oxygen-glucose deprivation. Pharma-col Rep, 2011, 63, 392–402.

29. Gabryel B, Trzeciak HI: Role of astrocytes in pathogenesisof ischemic brain injury. Neurotox Res, 2001, 3, 205–221.

30. Ginet V, Puyal J, Clarke PG, Truttmann AC: Enhance-ment of autophagic flux after neonatal cerebral hypoxia-ischemia and its region-specific relationship to apoptoticmechanisms. Am J Pathol, 2009, 175, 1962–1967.

31. Glick D, Barth S, Macleod KF: Autophagy: cellular andmolecular mechanisms. J Pathol, 2010, 221, 3–12.

32. Golstein P, Kroemer G: Cell death by necrosis: towardsa molecular definition. Trends Biochem Sci, 2007, 32,37–43.

33. Green AR, Shuaib A: Therapeutic strategies for the treat-ment of stroke. Drug Discov Today, 2006, 11, 681–693.

34. Grishchuk Y, Ginet V, Truttmann AC, Clarke PG, PuyalJ: Beclin 1-independent autophagy contributes to apopto-sis in cortical neurons. Autophagy, 2011, 7, 1115–1131.

35. Guo Y, Wang J, Wang Z, Yang Y, Wang X, Duan Q:Melatonin protects N2a against ischemia/reperfusion in-jury through autophagy enhancement. J Huazhong UnivSci Technolog Med Sci, 2010, 30, 1–7.

36. Gwinn DM, Shackelford DB, Egan DF, Mihaylova MM,Mery A, Vasquez DS, Turk BE, Shaw RJ: AMPK phos-phorylation of raptor mediates a metabolic checkpoint.Mol Cell, 2008, 30, 214–226.

37. Hara T, Nakamura K, Matsui M, Yamamoto A, NakaharaY, Suzuki-Migishima R, Yokoyama M et al.: Suppres-sion of basal autophagy in neural cells causes neurode-generative disease in mice. Nature, 2006, 441, 885–889.

38. Harris TE, Lawrence JC Jr: TOR signaling. Sci STKE,2003, 212, re15.

39. Høyer-Hansen M, Jäättelä M: AMP-activated protein ki-nase, a universal regulator of autophagy? Autophagy,2007, 3, 381–383.

40. Høyer-Hansen M, Jäättelä M: Connecting endoplasmicreticulum stress to autophagy by unfolded protein responseand calcium. Cell Death Differ, 2007, 14, 1576–1582.


41. Kiffin R, Christian C, Knecht E, Cuervo AM: Activationof chaperone-mediated autophagy during oxidativestress. Mol Biol Cell, 2004, 15, 4829–4840.

42. Kim J, Huang WP, Stromhaug PE, Klionsky DJ: Conver-gence of multiple autophagy and cytoplasm to vacuoletargeting components to a perivacuolar membrane com-partment prior to de novo vesicle formation. J BiolChem, 2002, 277, 763–773.

43. Kim J, Kundu M, Viollet B, Guan KL: AMPK andmTOR regulate autophagy through direct phosphoryla-tion of Ulk1. Nat Cell Biol, 2011, 13, 132–141.

44. Klionsky DJ: Autophagy: from phenomenology to mo-lecular understanding in less than a decade. Nat Rev MolCell Biol, 2007, 8, 931–937.

45. Klionsky DJ: Autophagy revisited: a conversation withChristian de Duve. Autophagy, 2008, 4, 740–743.

46. Kobayashi A, Cz³onkowska A: Trombolysis in acuteischemic stroke (Polish). Farmakoter Psychiatr Neurol,2005, 1, 5–18.

47. Koike M, Shibata M, Tadakoshi M, Gotoh K, KomatsuM, Waguri S, Kawahara N et al.: Inhibition of autophagyprevents hippocampal pyramidal neuron death afterhypoxic-ischemic injury. Am J Pathol, 2008, 172,454–469.

48. Komatsu M, Waguri S, Chiba T, Murata S, Iwata J,Tanida I, Ueno T et al.: Loss of autophagy in the centralnervous system causes neurodegeneration in mice. Na-ture, 2006, 441, 880–884.

49. Komatsu M, Waguri S, Ueno T, Iwata J, Murata S,Tanida I, Ezaki J et al.: Impairment of starvation-inducedand constitutive autophagy in Atg7-deficient mice. J CellBiol, 2005, 169, 425–434.

50. Kouroku Y, Fujita E, Tanida I, Ueno T, Isoai A, KumagaiH, Ogawa S et al.: ER stress (PERK/eIF2a phosphoryla-tion) mediates the polyglutamine-induced LC3 conver-sion, an essential step for autophagy formation. CellDeath Differ, 2007, 14, 230–239.

51. Krebiehl G, Ruckerbauer S, Burbulla LF, Kieper N,Maurer B, Waak J, Wolburg H et al.: Reduced basalautophagy and impaired mitochondrial dynamics due toloss of Parkinson’s Disease-Associated Protein DJ-1.PLoS One, 2010, 5, e9367.

52. Kubota C, Torii S, Hou N, Saito N, Yoshimoto Y, ImaiH, Takeuchi T: Constitutive reactive oxygen species gen-eration from autophagosome/lysosome in neuronal oxi-dative toxicity. J Biol Chem, 2010, 285, 667–674.

53. Kuma A, Hatano M, Matsui M, Yamamoto A, Nakaya H,Yoshimori T, Ohsumi Y et al.: The role of autophagyduring the early neonatal starvation period. Nature, 2004,432, 1032–1036.

54. Lakhan SE, Kirchgessner A, Hofer M: Inflammatorymechanisms in ischemic stroke: therapeutic approaches.J Transl Med, 2009, 7, 97.

55. Levine B, Abrams J: p53: The Janus of autophagy?Nature Cell Biol, 2008, 10, 637–639.

56. Li J, McCullough LD: Effects of AMP-activated proteinkinase in cerebral ischemia. J Cereb Blood Flow Metab,2010, 30, 480–492.

57. Li Q, Li H, Roughton K, Wang X, Kroemer G, BlomgrenK, Zhu C: Lithium reduces apoptosis and autophagy af-

ter neonatal hypoxia-ischemia. Cell Death Dis, 2010, 15,1:e56.

58. Lipton P: Ischemic cell death in brain neurons. PhysiolRev, 1999, 79, 1431–1568.

59. Liu C, Gao Y, Barrett J, Hu B: Autophagy and proteinaggregation after brain ischemia. J Neurochem, 2010,115, 68–78.

60. Liu J, Narasimhan P, Yu F, Chan PH: Neuroprotection byhypoxic preconditioning involves oxidative stress-mediated expression of hypoxia-inducible factor anderythropoietin. Stroke, 2005, 36, 1264–1269.

61. Liu L, Cash TP, Jones RG, Keith B, Thompson CB,Simon MC: Hypoxia-induced energy stress regulatesmRNA translation and cell growth. Mol Cell, 2006, 21,521–531.

62. Maiuri MC, Zalckvar E, Kimchi A, Kroemer G: Self-eating and self-killing: crosstalk between autophagy andapoptosis. Nature Rev Mol Cell Biol, 2007, 8, 741–752.

63. Matsui Y, Takagi H, Qu X, Abdellatif M, Sakoda H,Asano T, Levine B, Sadoshima J: Distinct roles of auto-phagy in the heart during ischemia and reperfusion: rolesof AMP-activated protein kinase and Beclin 1 in mediat-ing autophagy. Circ Res, 2007, 100, 914–922.

64. Meijer AJ, Dubbelhuis PF: Amino acid signalling andthe integration of metabolism. Biochem Biophys ResCommun, 2004, 313, 397–403.

65. Meloni BP, Meade AJ, Kitikomolsuk D, Knuckey NW:Characterisation of neuronal cell death in acute and de-layed in vitro ischemia (oxygen-glucose deprivation)models. J Neurosci Methods, 2011, 195, 67–74.

66. Mizushima N: Autophagy: process and function. GenesDev, 2007, 21, 2861–2873.

67. Mizushima N: The pleiotropic role of autophagy: fromprotein metabolism to bactericide. Cell Death Differ,2005, 12, 1535–1541.

68. Mizushima N, Ohsumi Y, Yoshimori T: Autophagosomeformation in mammalian cells. Cell Struct Funct, 2002,27, 421–429.

69. Obrenovitch TP: The ischemic penumbra: twenty yearson. Cerebrovasc Brain Metab Rev, 1995, 7, 297–332.

70. Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA,Outzen H, Øvervatn A et al.: p62/SQSTM1 binds di-rectly to Atg8/LC3 to facilitate degradation of ubiquiti-nated protein aggregates by autophagy. J Biol Chem,2007, 282, 24131–24145.

71. Park HK, Chu K, Jung KH, Lee ST, Bahn JJ, Kim M,Lee SK, Roh JK: Autophagy is involved in the ischemicpreconditioning. Neurosci Lett, 2009, 451, 16–19.

72. Patyar S, Prakash A, Modi M, Medhi B: Role of vinpo-cetine in cerebrovascular diseases. Pharmacol Rep, 2011,63, 618–628.

73. Perycz M, Œwiech £, Malik A, Jaworski J: mTOR inphysiology and pathology of the nervous system(Polish). Post Biol Kom, 2007, 34, 511–525.

74. Rami A: Review: autophagy in neurodegeneration: fire-fighter and/or incendiarist? Neuropathol Appl Neurobiol,2009, 35, 449–461.

75. Rami A, Kögel D: Apoptosis meets autophagy-like celldeath in the ischemic penumbra: Two sides of the samecoin? Autophagy, 2008, 4, 422–426.



76. Rami A, Langhagen A, Steiger S: Focal cerebral ische-mia induces upregulation of Beclin 1 and autophagy-likecell death. Neurobiol Dis, 2008, 29, 132–141.

77. Reggiori F, Shintani T, Nair U, Klionsky DJ: Atg9 cyclesbetween mitochondria and the pre-autophagosomalstructure in yeasts. Autophagy, 2005, 1, 101–109.

78. Reggiori F, Tucker KA, Stromhaug PE, Klionsky DJ:The Atg1-Atg13 complex regulates Atg9 and Atg23 re-trieval transport from the pre-autophagosomal structure.Dev Cell, 2004, 6, 79–90.

79. Rubinsztein DC, DiFiglia M, Heintz N, Nixon RA, QinZH, Ravikumar B, Stefanis L, Tolkovsky A: Autophagyand its possible roles in nervous system diseases, damageand repair. Autophagy, 2005, 1, 11–22.

80. Sarbassov DD, Guertin DA, Ali SM, Sabatini DM: Phos-phorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science, 2005, 307, 1098–1101.

81. Scherz-Shouval R, Shvets E, Fass E, Shorer H, Gil L,Elazar Z: Reactive oxygen species are essential for auto-phagy and specifically regulate the activity of Atg4.EMBO J, 2007, 26, 1749–1760.

82. Schieke SM, Phillips D, McCoy JP Jr, Aponte AM, ShenRF, Balaban RS, Finkel T: The mammalian target of ra-pamycin (mTOR) pathway regulates mitochondrial oxy-gen consumption and oxidative capacity. J Biol Chem,2006, 281, 27643–27652.

83. Semenza GL: Mitochondrial autophagy: life and breathof the cell. Autophagy, 2008, 4, 534–536.

84. Shang J, Deguchi K, Yamashita T, Ohta Y, Zhang H,Morimoto N, Liu N et al.: Antiapoptotic and antiauto-phagic effects of glial cell line-derived neurotrophic fac-tor and hepatocyte growth factor after transient middlecerebral artery occlusion in rats. J Neurosci Res, 2010,88, 2197–2206.

85. Sharp FR, Lu A, Tang Y, Millhorn DE: Multiple molecu-lar penumbras after focal cerebral ischemia. J CerebBlood Flow Metab, 2000, 20, 1011–1032.

86. Sheng R, Zhang LS, Han R, Liu XQ, Gao B, Qin ZH:Autophagy activation is associated with neuroprotectionin a rat model of focal cerebral ischemic preconditioning.Autophagy, 2010, 6, 482–494.

87. Shintani T, Klionsky DJ: Autophagy in health and disease:a double-edged sword. Science, 2004, 306, 990–995.

88. Steiger-Barraissoul S, Rami A: Serum deprivation in-duced autophagy and predominantly an AIF-dependentapoptosis in hippocampal HT22 neurons. Apoptosis,2009, 14, 1274–1288.

89. Suzuki K, Kubota Y, Sekito T, Ohsumi Y: Hierarchy ofAtg proteins in pre-autophagosomal structure organiza-tion. Genes Cells, 2007, 12, 209–218.

90. Szczudlik A: Neuroprotection as a direction of brainischemia treatment (Polish). XX Winter School of theInstitute of Pharmacology of PAS, 2003, 99–105.

91. Taoufik E, Probert L: Ischemic neuronal damage. CurrPharm Des, 2008, 14, 3565–3573.

92. Tardy C, Codogno P, Autefage H, Levade T, Andrieu-Abadie N: Lysosomes and lysosomal proteins in cancercell death (new players of an old struggle). Biochim Bio-phys Acta, 2006, 1765, 101–125.

93. Tasdemir E, Maiuri MC, Galluzzi L, Vitale I, Djavaheri-Mergnt M, D’Amelio M, Criollo A et al.: Regulation of

autophagy by cytoplasmic p53. Nat Cell Biol, 2008, 10,676–687.

94. Thorburn A: Apoptosis and autophagy: regulatory con-nections between two supposedly different processes.Apoptosis, 2008, 13, 1–9.

95. Tian F, Deguchi K, Yamashita T, Ohta Y, Morimoto N,Shang J, Zhang X et al.: In vivo imaging of autophagy ina mouse stroke model. Autophagy, 2010, 6, 1107–1114.

96. Towler MC, Hardie DG: AMP-activated protein kinasein metabolic control and insulin signaling. Circ Res,2007, 100, 328–341.

97. Uchiyama Y: Autophagic cell death and its execution by ly-sosomal cathepsins. Arch Histol Cytol, 2001, 64, 233–246.

98. Uchiyama Y, Shibata M, Koike M, Yoshimura K, SasakiM: Autophagy-physiology and pathophysiology. Histo-chem Cell Biol, 2008, 129, 407–420.

99. Wang CW, Klionsky DJ: The molecular mechanism ofautophagy. Mol Med, 2003, 9, 65–76.

100. Wang JY, Xia Q, Chu KT, Pan J, Sun LN, Zeng B,Zhu YJ et al.: Severe global cerebral ischemia-inducedprogrammed necrosis of hippocampal CA1 neurons in ratis prevented by 3-methyladenine: a widely used inhibitorof autophagy. J Neuropathol Exp Neurol, 2011, 70,314–322.

101. Wang S, Miura M, Jung YK, Zhu H, Li E, Yuan J:Murine caspase-11, an ICE-interacting protease, is essen-tial for the activation of ICE. Cell, 1998, 92, 501–509.

102. Wen YD, Sheng R, Zhang LS, Han R, Zhang X, ZhangXD, Han F et al.: Neuronal injury in rat model of perma-nent focal cerebral ischemia is associated with activationof autophagic and lysosomal pathways. Autophagy,2008, 4, 762–769.

103. Werner C, Engelhard K: Pathophysiology of traumaticbrain injury. Br J Anaesth, 2007, 99, 4–9.

104. Wirawan E, Vande Walle L, Kersse K, Cornelis S,Claerhout S, Vanoverberghe I, Roelandt R et al.:Caspase-mediated cleavage of Beclin-1 inactivatesBeclin-1-induced autophagy and enhances apoptosis bypromoting the release of proapoptotic factors from mito-chondria. Cell Death Dis, 2010, 1, e18.

105. Wullschleger S, Loewith R, Hall MN: TOR signaling ingrowth and metabolism. Cell, 2006, 124, 471–484.

106. Xie Z, Klionsky DJ: Autophagosome formation: coremachinery and adaptations. Nat Cell Biol, 2007, 9,1102–1109.

107. Xue L, Fletcher GC, Tolkovsky AM: Mitochondria areselectively eliminated from eukaryotic cells after block-ade of caspases during apoptosis. Curr Biol, 2001, 11,361–365.

108. Yamashima T, Oikawa S: The role of lysosomal rupturein neuronal death. Prog Neurobiol, 2009, 89, 343–358.

109. Yan W, Zhang H, Bai X, Lu Y, Dong H, Xiong L: Auto-phagy activation is involved in neuroprotection inducedby hyperbaric oxygen preconditioning against focal cere-bral ischemia in rats. Brain Res, 2011, 1402, 109–121.

110. Yang Z, Klionsky DJ: An overview of the molecularmechanism of autophagy. Curr Top Microbiol Immunol,2009, 335, 1–32.

111. Ylä-Anttila P, Vihinen H, Jokitalo E, Eskelinen EL:3D tomography reveals connections between the phagophoreand endoplasmic reticulum. Autophagy, 2009, 5, 1180–1185.


112. Yorimitsu T, Klionsky DJ: Autophagy: molecular ma-chinery for self-eating. Cell Death Differ, 2005, 12,1542–1552.

113. Young JE, Martinez RA, La Spada AR: Nutrient depriva-tion induces neuronal autophagy and implicates reducedinsulin signaling in neuroprotective autophagy activa-tion. J Biol Chem, 2009, 284, 2363–2373.

114. Yousefi S, Perozzo R, Schmid I, Ziemiecki A, SchaffnerT, Scapozza L, Brunner T, Simon HU: Calpain-mediatedcleavage of Atg5 switches autophagy to apoptosis. NatCell Biol, 2006, 8, 1124–1132.

115. Yue Z, Friedman L, Komatsu M, Tanaka K: The cellularpathways of neuronal autophagy and their implication inneurodegenerative diseases. Biochim Biophys Acta,2009, 1793, 1496–1507.

116. Zeng X, Overmeyer JH, Maltese WA: Functional speci-ficity of the mammalian Beclin-Vps34 PI 3-kinase com-

plex in macroautophagy versus endocytosis and lysoso-mal enzyme trafficking. J Cell Sci, 2006, 119, 259–270.

117. Zhou X, Zhou J, Li X, Guo C, Fang T, Chen Z: GSK-3b

inhibitors suppressed neuroinflammation in rat cortex byactivating autophagy in ischemic brain injury. BiochemBiophys Res Commun, 2011, 411, 271–275.

118. Zhu C, Wang X, Xu F, Bahr BA, Shibata M, UchiyamaY, Hagberg H, Blomgren K: The influence of age on apop-totic and other mechanisms of cell death after cerebralhypoxia-ischemia. Cell Death Differ, 2005, 12, 162–176.

Received: July 26, 2011; in the revised form: October 21, 2011;

accepted: November 4, 2011.



neuronal autophagy in cerebral ischemia – a potential ... · companion of mtor, ros – reactive...

Documents