The role of lysosomal rupture in neuronal death

Progress in Neurobiology 89 (2009) 343–358

Contents lists available at ScienceDirect

Progress in Neurobiology journal homepage: www.elsevier.com/locate/pneurobio

The role of lysosomal rupture in neuronal death Tetsumori Yamashima a,*, Shinji Oikawa b a b

Department of Restorative Neurosurgery, Kanazawa University Graduate School of Medical Science, Kanazawa 920-8641, Japan Department of Environmental and Molecular Medicine, Mie University Graduate School of Medicine, Tsu 514-8507, Japan

A R T I C L E I N F O

A B S T R A C T

Article history: Received 3 February 2009 Received in revised form 11 September 2009 Accepted 15 September 2009

Apoptosis research in the past two decades has provided an enormous insight into its role in regulating cell death. However, apoptosis is only part of the story, and inhibition of neuronal necrosis may have greater impact than apoptosis, on the treatment of stroke, traumatic brain injury, and neurodegenerative diseases. Since the ‘‘calpain–cathepsin hypothesis’’ was first formulated, the calpain- and cathepsinmediated regulation of necrotic cascades observed in monkeys, has been demonstrated to be a common neuronal death mechanism occurring from simpler organisms to humans. However, the detailed mechanism inducing lysosomal destabilization still remains poorly understood. Heat-shock protein-70 (Hsp70) is known to stabilize lysosomal membrane and protect cells from oxidative stress and apoptotic stimuli in many cell death pathways. Recent proteomics approach comparing pre- and post-ischemic hippocampal CA1 neurons as well as normal and glaucoma-suffered retina of primates, suggested that the substrate protein upon which activated calpain acts at the lysosomal membrane of neurons might be Hsp70. Understanding the interaction between activated calpains and Hsp70 will help to unravel the mechanism that destabilizes the lysosomal membrane, and will provide new insights into clarifying the whole cascade of neuronal necrosis. Although available evidence is circumferential, it is hypothesized that activated calpain cleaves oxidative stress-induced carbonylated Hsp70.1 (a major human Hsp70) at the lysosomal membrane, which result in lysosomal rupture/permeabilization. This review aims at highlighting the possible mechanism of lysosomal rupture in neuronal death by a modified ‘‘calpain– cathepsin hypothesis’’. As the autophagy–lysosomal degradation pathway is a target of oxidative stress, the implication of autophagy is also discussed. ß 2009 Elsevier Ltd. All rights reserved.

Keywords: Neuronal death Oxidative stress Hsp70 Lysosome Hippocampus Calpain–cathepsin hypothesis Autophagy

Contents 1. 2. 3. 4.

5. 6.

Background: calpain activation (Fig. 1). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1. Background: oxidative stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neuronal apoptosis and necrosis (Fig. 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Autophagy–lysosomal system and neuronal death (Figs. 3 and 4). . Role of lysosomes in cell death (Fig. 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Implication of oxidative stress in lysosomal rupture (Fig. 6) . . . . . . Hsp70 and its major form Hsp70.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Upregulation and carbonylation of Hsp70.1 after ischemia (Fig. 7) . 4.2. Possible cleavage of carbonylated Hsp 70.1 by activated calpain. . . Modified ‘‘calpain–cathepsin hypothesis’’ (Fig. 8) . . . . . . . . . . . . . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

344 345 346 347 348 349 351 352 353 353 355 355 355

Abbreviations: CA1, cornu Ammonis 1; ER, endoplasmic reticulum; Hsp, heat-shock protein; Hsc, heat-shock constitutive protein; HNE, 4-hydroxy-2-nonenal; Lamp, lysosome-associated membrane protein; MALDI-TOF/TOF, matrix-assisted laser desorption ionization-time of flight/time of flight; MS/MS, mass spectrometry/mass spectrometry; ROS, reactive oxygen species; 2D DIGE, two-dimensional differential in-gel electrophoresis; DNPH, 2,4-dinitrophenylhydrazine; DNP, 2,4-dinitrophenylhydrazone; 2D Oxyblots, two-dimensional gel electrophoresis with immunoblot detection of carbonylated proteins; V-ATPase, vacuolar-type proton ATPase. * Corresponding author at: Department of Restorative Neurosurgery, Kanazawa University Graduate School of Medical Science, Takara-machi 13-1, Kanazawa 920-8641, Japan. Tel.: +81 76 265 2381; fax: +81 76 234 4264. E-mail address: [email protected] (T. Yamashima). 0301-0082/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.pneurobio.2009.09.003

344

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

1. Background: calpain activation (Fig. 1) During embryonic development, neurons are pruned by apoptosis with excess ones being removed to ensure proper and precise synaptic connections. In contrast, during adult life, neurons prematurely die mostly by necrosis when subject to acute or chronic neurotoxic insults. Typically, neuronal necrosis is prominent in ischemic brain injury, and it underlies the pathology of also traumatic brain injury and neurodegenerative diseases. Nowadays, the incidence of brain ischemia is dramatically increasing in Western countries, becoming thus a major cause of chronic physical and/or cognitive disability (Flynn et al., 2008). Accordingly, detailed molecular analysis of ischemic neuronal death should have important clinical implications. Ischemic neuronal death develops selectively in cells most vulnerable to hypoxic damage, such as hippocampal CA1 neurons, medium-sized neurons in the striatum, and Purkinje cells in the cerebellum (Rami et al., 2008). From rodents to primates, transient brain ischemia is well known to cause delayed neuronal death in the hippocampal CA1 neurons. As ischemic CA1 neuronal death gradually develops within 5–7 days after the insult, it can provide an appropriate time window for studying the underlying mechanism of neuronal death. In the past two decades, substantial efforts focusing on CA1 have been made to elucidate the biochemical and molecular mechanisms of ischemic neuronal death. However, many basic questions remain unanswered. The sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA) pump as well as extrusion of calcium through the plasma-membrane Ca2+-ATPase, maintains free cytosolic Ca2+ concentrations at

approximately 100–200 nM, which is orders of magnitude less than extracellular levels. Excessive NMDA receptor activation induces Ca2+ influx and its release from the intracellular stores. Such disruption of Ca2+ homeostasis plays a key role in neuronal death, and the intracellular Ca2+ mobilization initiates neuronal death (Fig. 1). This, in turn, leads to autophagic death, apoptosis, or necrosis, depending upon the severity of the insult. Ca2+ overload triggers lethal downstream cascades, including calpain and caspase activation, and can also lead to mitochondrial dysfunction. The disappointing results of clinical trials using various Ca2+ blockers validate the importance of elucidating the downstream cascade of Ca2+-mobilization in order to develop a novel neuroprotective strategy. The molecular cascades of the cell death are diverse, and have been categorized into two main types, apoptosis or necrosis. In the embryonic brain, apoptosis plays an integral part for its anatomic and functional maturation. Intracellular ATP levels are a primary determinant of apoptosis or necrosis. In the adult brain, the apoptotic route will predominate when and where ATP is plentiful, whereas the necrotic route will predominate when and where ATP is depleted (Rami et al., 2008). Necrosis typically occurs following ischemia, hypoxia, stroke or trauma, but it has also been reported in neurodegenerative diseases such as Alzheimer’s, Huntington’s, and Parkinson’s disease. While a mild Ca2+ increase preferentially induces apoptosis, an abrupt and severe Ca2+ increase initiates necrosis. During stroke, for instance, the core area being immediately and drastically affected by the restricted blood flow usually suffers from necrosis. On the contrary, many neurons undergo apoptosis in the surrounding penumbra area which is less

Fig. 1. A flow chart of calpain–cathepsin and autophagy pathways from necrosis initiating insults to cell death. Upon induction of necrotic insults, intracellular Ca2+ is increased mainly by release from ER, which activates m-calpain. Parallel to activated calpain-induced lysosomal rupture, autophagy is upregulated directly and/or through calpain activation, but eventually synergizes with extra-lysosomal cathepsins to mediate cell death. A cup-formed phagophore (also called isolation membrane) surrounds cytosolic substrates thereby creating an autophagosome. As the cytoplasmic pH is reduced after the lysosomal rupture, this increases the potential for extra-lysosomal cathepsins to degrade cell constitutive proteins at an optimal pH. [Ca2+]i, cytoplasmic calcium concentration; ER, endoplasmic reticulum; InsP3R, inositol triposphate receptor; RyR, ryanodine receptor; SERCA, sarco-endoplasmic reticulum Ca2+-ATPase; V-ATPase, vacuolar H+-ATPase (cited from Kourtis and Tavernarakis, 2009).

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

affected by oxygen and nutrient deprivation (Broughton et al., 2009). As apoptosis has been extensively studied with over 170,000 scientific and clinical papers published on this topic in the past two decades, molecular mechanisms of apoptosis have been largely uncovered. Apoptosis proceeds via two parallel pathways; the intrinsic mitochondrial pathway and/or the extrinsic death receptor pathway. An elaborate series of apoptotic cascades has been illuminated, with particular focus on the caspase enzymes and the Bcl-2 family. However, accumulating evidence during the last decade suggests that programmed cell death is not confined to apoptosis, rather cells use conventional necrotic pathways for their active self-destruction. Since the ‘‘calpain–cathepsin hypothesis’’ was first formulated (Yamashima et al., 1996, 1998; Yamashima, 2000), the calpain-induced lysosomal rupture (Fig. 1) has been accepted as one of major cascades of necrosis. This occurs by dysregulation of normal cellular activity when cells are exposed to extreme stress or energy depletion. Necrosis was formerly considered to be chaotic, but its occurrence might instead be tightly regulated by controlled processes such as mitochondrial dysfunction, enhanced generation of oxidants, ATP depletion, proteolysis by calpains and cathepsins, and early plasma membrane rupture (Golstein and Kroemer, 2007). Lysosomes contain a large number of acidic hydrolases, and serve as the main site of degradation of constituent cell proteins and amino acid recycling. Lysosomes are intracytoplasmic organelles defined by an acidic milieu of pH around 4.5, and are separated from the surrounding cytosol which has a neutral pH, by a tough single membrane. This lysosomal membrane is essentially a physical barrier that prevents hydrolytic enzymes from digesting the cell proteins. Its destabilization not only influences their normal activities but also affects cell vitality. The role of lysosomes and lysosomal enzymes in initiation and execution of the apoptotic program has also become clear (Guicciardi et al., 2004). Partial rupture or permeabilization of the lysosomal membrane has been recently suggested to induce apoptosis by mitochondrial transmembrane potential loss or caspase activation. In contrast, intense lysosomal rupture can induce necrotic cell death due to leakage of cathepsin enzymes (Tardy et al., 2006). Lysosomal rupture is a crucial event for living cells, but very little is known about the molecular mechanisms that mediate lysosomal membrane destabilization. Studies on the monkey hippocampal CA1 neurons after transient global ischemia indicated that rupture of their lysosomal membrane should be inflicted enzymatically by Ca2+-mediated activation of calpains (EC 3.4.22.17), although there has been no demonstration that blocking calpain prevents lysosomal rupture. Activated m-calpain translocates to lysosomal membranes after ischemia, including the subsequent spillage of cathepsins into the cytoplasm and neuropil that eventually dismantle the whole CA1 neurons (Yamashima et al., 1996). This observation in the monkey experimental paradigm led to the formulation of the ‘‘calpain– cathepsin hypothesis’’, according to which calpains compromise the integrity of lysosomal membranes and cause leakage of lysosomal acidic hydrolases into the cytoplasm (Fig. 1) (Yamashima et al., 1998, 2003; Yamashima, 2000, 2004). Whereas cytoplasmic pH is regulated by transporters such as the Na+/H+ exchanger, the pH within many intracellular compartments is regulated by ATP-dependent proton pumps known as the vacuolar-type proton (H+)-ATPase (V-ATPase) (Nishi and Forgac, 2002). By pumping protons (H+) from the cytosol into the lysosomal lumen, V-ATPase physiologically maintains an acidic microenvironment within lysosomes, and regulates cellular pH at the expense of ATP. After lysosomal rupture, however, the VATPase pump contributes to the subsequent cytoplasmic acidification that is responsible for the increased activation of low-pH

345

dependent cathepsins (Fig. 1) (Syntichaki et al., 2005; Kourtis and Tavernarakis, 2009). The calpain–cathepsin cascade actually supports de Duve’s original categorization of lysosomes as the cell’s ‘‘suicide bag’’ (de Duve, 1983). Since their discovery over 40 years ago, calpains have been linked to a variety of pathological conditions including ischemic neuronal death. However, despite multiple attempts at substrate sequence analysis, the specific in vivo calpain substrate remains unknown in all pathologic conditions. No definitive methodology has ever existed to predict whether a given compound is actually an in vivo calpain substrate (Bevers and Neumar, 2008). Given the large number of known in vitro calpain substrates, it remains unclarified how calpain-mediated cleavage of a certain protein in vivo can lead to execution of neuronal death. In the brain, potential substrates for m-calpain include a number of cytoskeletal proteins such as actin-binding proteins, membrane proteins such as growth factors, receptors and Ca2+ channels, adhesion molecules and ion transporters, enzymes such as kinases, phosphatases (calcineurin) and phospholipases, cytokines, as well as transcription factors (Rami et al., 2008). This diversity of calpain substrates suggests the existence of multiple mechanisms by which calpain causes neuronal death. However, the in vivo calpain substrate that directly contributes to ischemic neuronal necrosis remains unknown. Although very difficult, identification of intracellular calpain substrates in the living animal neurons is needed to precisely understand the mechanism of calpain-induced lysosomal disruption. 1.1. Background: oxidative stress While oxygen in its many states is a requirement for life, partial reduction of molecular oxygen leads to the formation of reactive oxygen species (ROS) which can lead to extensive damage of various cellular macromolecules including DNA, lipids, and proteins. Because the brain consumes a large percentage of inspired oxygen, and is rich in polyunsaturated fatty acids, this organ is especially vulnerable to oxidative stress. Protein oxidation in neurons is well known to occur in age-related cognitive decline, depression, stroke, and neurodegenerative disorders such as Alzheimer’s and Parkinson’s disease. One major action of ROS is the production of 4-hydroxy-2-nonenal (HNE; oxidative aldehyde products of linoleate- and arachidonate-containing lipids) and other reactive carbonyl species, which ultimately promote protein carbonylation. HNE becomes a key endogenous neurotoxin, and HNE-induced carbonylation of proteins leads to loss of their function, due to its irreversible and unrepairable nature. In spite of the considerable number of papers reporting increased levels of protein carbonyls in various human neurodegenerative diseases, the precise relationship between protein carbonylation and neuronal death remains unclear at present. Although oxidative stress underlies cell death in several pathological conditions, the molecular mechanisms involved are complex and variable. It is well known that oxidative stress causes DNA damage, mitochondrial damage, or lysosomal rupture. In neurons, mechanisms of DNA and mitochondrial damage have been mainly investigated, whereas only the ‘‘calpain–cathepsin hypothesis’’ has been reported concerning the mechanisms of the oxidative stress-induced lysosomal rupture (Yamashima et al., 1998; Yamashima, 2000). Ca2+-induced calpain activation and ROS production are thought to be perpetrators of lysosomal membrane rupture (permeabilization) that leads to ischemic neuronal damage. Although both events are thought to cause release of hydrolytic enzymes such as cathepsins from lysosomes, the exact mechanism of lysosomal membrane rupture remains unknown. Using the rat hippocampal slice preparation of oxygen–glucose deprivation, Windelborn and Lipton (2008) recently reported that lysosomal

346

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

rupture occurred with the synergy of activated calpains via the NMDA-mediated Ca2+ influx and superoxide via the cyclooxygenase-2 (COX-2)-mediated oxidation of arachidonic acid. Arachidonic acid, which is released from the membrane phospholipids by Ca2+dependent phospholipase A2, is increased under ischemic and reperfusion/reoxygenation conditions. Then, COX-2-mediated metabolism of arachidonic acid to prostaglandin is the major source of ROS production (Windelborn and Lipton, 2008). However, the biochemical cascade of how activated calpain and ROS work together to rupture lysosomal membranes remains unknown. Lysosomal rupture is a crucial event for living cells, but the mechanisms underlying this process are still unclear. Elucidation of the critical steps of lysosomal rupture should aid in understanding both the mechanism of neuronal necrosis and developing novel therapies of intervention in neuronal death. As a fundamental step in establishing a relationship between protein oxidation and ischemic neuronal necrosis, proteomics analysis was recently done (Nakajima et al., 2006; Oikawa et al., 2009) to identify specific targets of protein oxidation, by means of 2D-gel fingerprinting of oxidized proteins, immunochemical detection of protein carbonyls coupled with time-of-flight/time-of-flight mass spectrometry (TOF/TOF mass spectrometry), and database search (Castegna et al., 2002). Interestingly and surprisingly, the proteomics analysis using primate retina samples (Nakajima et al., 2006) has identified Hsp70 as a potential in vivo key substrate of calpain. Protein carbonyls are actually the most general indicator of protein oxidation. To determine protein oxidation levels of 2,4-dinitrophenylhydrazine (DNPH)-treated samples, Oikawa et al. (2009) studied protein carbonylation through two-dimensional gel electrophoresis with immunoblot detection of carbonylated proteins (2D Oxyblot) analysis using anti-2,4-dinitrophenylhydrazone (DNP) antibody. Significantly elevated levels (more than 10-fold compared to the non-ischemic controls) of cabonylated Hsp70.1 (a major protein of the human Hsp70 family) were detected in the day 3 monkey hippocampus undergoing transient global brain ischemia (Oikawa et al., 2009). A clear delineation of the causal connections between calpain activation and lysosomal rupture cannot be given at present, but growing evidence from two independent proteomic analyses using primate hippocampus and retina indicates that activated calpain and oxidants can induce distinct pathological effects at the lysosomal Hsp70.1. Heat-shock-induced protein was first discovered in 1962 by Ritossa as a highly conserved protein, being present in virtually all species from bacteria to humans with both chaperoning and cytoprotective functions under a variety of environmental stressors. In 2004, Nylandsted et al. demonstrated that Hsp70 is localized at the lysosomal membranes of tumor cells, and that Hsp70 inhibits lysosomal membrane permeabilization induced by diverse stimuli while depletion of Hsp70 triggers it. Although it has not been shown that Hsp70 becomes an in vivo substrate for calpain after the oxidative stress, the correlation between calpain-mediated Hsp70 breakdown and lysosomal rupture in the postischemic neurons is highly probable, given the ability of Hsp70 to stabilize the lysosomal membrane. The aim of this review is to give an overview of the current understanding of the mechanism of lysosomal rupture in neuronal death. Focusing on the mechanism of ischemic neuronal death in primates, the authors discuss on (1) the key substrate cleaved by the activated m-calpain at the lysosomal membrane, (2) the role of oxidative stress in the Arg469-carbonylation of Hsp70.1, and (3) the implication of carbonylated Hsp70.1 cleavage in lysosomal rupture. 2. Neuronal apoptosis and necrosis (Fig. 2) Apoptosis is usually a caspase-dependent cell death, while necrosis is essentially a cathepsin-dependent cell death. In both

Fig. 2. Systemic pathways from cerebral ischemia to neuronal apoptosis/necrosis. Note the molecular cross-talks between apoptosis (caspase: left) and necrosis (cathepsin: right) cascades. Ca2+-mobilization during the ischemic insult activates mcalpain at the lysosomal membrane of CA1 neurons to induce release of cathepsins B, L and D that leads to necrosis. Alternatively, cathepsin B triggers activation of caspase 11 and subsequently caspase 3, which leads to apoptosis. Further, the interaction of cytosolic cathepsins with Bcl-2 family members has the potential to induce mitochondrial apoptosis. Extents of Ca2+-mobilization and lysosomal destabilization may influence cell death pattern (cited from Rami et al., 2008).

clinical disorders and clinically relevant experimental models, however, there is a continuum of cell death patterns with multiple variants of the apoptosis–necrosis continuum coexisting (Yakovlev and Faden, 2004). In Greek, ‘apoptosis’ means ‘falling off’ while ‘necros’ means ‘dead’. Apoptosis has come to be used synonymously with ‘programmed cell death’, and it requires energy in the form of ATP. In contrast, necrosis has been traditionally thought to be a passive form of cell death, and that is the end result of a bioenergetic catastrophe resulting from ATP depletion to a level incompatible with cell survival. Apoptosis and necrosis have been distinguished by specific histological criteria (Kerr, 1972); apoptosis is characterized by preserved membrane integrity, plasma membrane blebbing, and nuclear condensation called apoptotic bodies, while necrosis is characterized by loss of membrane integrity, lysosomal disruption, and uncontrolled cell lysis. Apoptosis and necrosis are not necessarily two independent pathways, but rather share some common events of signal transduction pathways (Yakovlev and Faden, 2004). Apoptosis and necrosis may be induced by the same type of insult, but depending on the magnitude of the insult or the intracellular ATP levels, the cell death fate is determined by the choice of either apoptotic or necrotic cell death programs. It is likely that the degree of Ca2+ elevation and ensuing calpain activation can

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

determine whether cells die by apoptosis or necrosis (Fig. 2). Mild Ca2+ elevation favours apoptosis (Neumar et al., 2003), whereas intense Ca2+ elevation and calpain activation precipitate necrosis probably via catastrophic cleavage of regulatory and structural proteins (Syntichaki et al., 2002). A sequence of intracellular events occurring in the necrosis cascade includes Ca2+ mobilization, early signs of mitochondrial dysfunction with ATP depletion and production of ROS, activation of cysteine protease calpains and cathepsins, lysosomal rupture, and ultimately plasma membrane rupture (Golstein and Kroemer, 2007). Under pathological stress, necrosis might be programmed in terms of both its occurrence and course. The human brain comprises only 2% of body weight but requires as much as 20% of total oxygen consumption to generate sufficient ATP by oxidative phosphorylation for the maintenance of membrane potential and ionic gradients. During brain ischemia, ATP synthesis in mitochondria is inhibited and the available ATP is depleted within 2 min. This energy failure subsequently prevents the plasma membrane Ca2+ ATPase from maintaining the very low (0.1–0.2 mM) concentrations of Ca2+ that are normally present within each cell. As a result, intracellular Ca2+ levels rise to 50– 100 mM, causing only a limited amount of m-calpain activation, which efficiently cleaves and activates caspase-12 to mediate apoptosis (Nakagawa and Yuan, 2000; Doyle et al., 2008). In contrast, remarkable increase in intracellular Ca2+ levels, may induce sufficient amount of calpain activation to mediate necrotic cell death through lysosomal rupture. High levels of calpain activity may cleave and inactivate pro-apoptotic proteins, such as Apaf-1 (Reimertz et al., 2001) and caspase-7, -8, and -9 (Chua et al., 2000). These cross-inhibitory actions of alternative pathways may ensure the successful execution of programmed cell death either by apoptosis or necrosis. Since apoptosis is an ATP-dependent mechanism, loss of the cellular energy will prevent cells from entering this mode of death. Accumulating data in the past decade show that lysosomes may function as death signal integrators not only in necrosis (Yamashima, 2000) but also in apoptosis (Nylandsted et al., 2004). For example, upon incubation of Jurkat leukemia cells with low amounts of sphingosine, a partial lysosomal rupture followed by mitochondrial membrane permeabilization and caspase activation led to apoptosis, whereas higher concentrations of sphingosine induced total rupture of lysosomes and necrosis without caspase implication (Ka˚gedal et al., 2001). It is nowadays widely accepted that the magnitude of the lysosomal rupture and consequently, the amount of hydrolytic enzymes released into the cytosol may influence cell death fate; apoptosis or necrosis (Fig. 2) (Olejnicka et al., 1999; Brunk and Svensson, 1999; Kessel and Poretz, 2000; Li et al., 2000; Bursch, 2001; Boya and Kroemer, 2008; Pivtoraiko et al., 2009). Partial lysosomal rupture leads to apoptosis, while fulminant lysosomal rupture results in necrosis (Terman et al., 2006a,b). The neuronal response to injury is not uniform and neuronal subpopulations are differentially vulnerable to the same injury. Exposure of glutamergic neurons, such as hippocampal neurons, to potentially damaging stressors such as ischemia, induces calpain–cathepsin signals able to mediate neuronal necrosis. In contrast, the ischemia-resistant cortical GABAergic neurons can tolerate this imposed excitotoxicity and the same calpain-mediated extra-lysosomal release of cathepsins induces only decreased synthesis of glutamic acid decarboxylase (GAD) (Monnerie and Le Roux, 2008). It is likely that such paradoxical events in vulnerable or resistant neurons occur due to distinct degrees of lysosomal rupture or cathepsin release. 2.1. Autophagy–lysosomal system and neuronal death (Figs. 3 and 4) Two major pathways accomplish regulated protein catabolism: the ubiquitin–proteasome system degrades short-lived proteins

347

Fig. 3. A working model of the four steps in chaperone-mediated autophagy of longlived proteins. Hsc70 chaperone and its cochaperones play a role during substrate recognition, targeting, unfolding, and transport. Three players of Hsc70 work in the cytosol (red circle), at the lysosomal membrane (lymHsc70), and within the lumen of lysosome (lyHsc70). Accordingly, although Hsc70 is a very abundant cytosolic protein, its dysfunction may lead to impairment of damaged protein clearance, and the resultant aggregation of damaged proteins causes enhanced cellular stress. Lamp2a serves as a receptor for the cytosolic proteins that undergo degradation. The lysosomal membrane is protected from the acidic hydrolases by lysosomespecific expression of Lamp2a, which is heavily glycosylated and hence resists digestion (Eskelinen, 2006). Hsc, heat-shock constitutive protein; Lamp2a, lysosome-associated membrane protein 2a (cited from Agarraberes and Dice, 2001).

while autophagy degrades long-lived proteins, although both contribute to the degradation of defective proteins (Nedelsky et al., 2008). Upon induction of necrosis-initiating insults, autophagy is upregulated directly and/or through calpain activation, and synergizes with extra-lysosomal cathepsins to mediate cell death (Fig. 1) (Kourtis and Tavernarakis, 2009). The word ‘autophagy’ means ‘self-eating’ in Greek, and was first used in 1963 by Christian de Duve (Klionsky, 2008). Autophagy is the major housekeeping pathway involved in the degradation of proteins and organelles for recycling, and is regulated by well-preserved autophagy-related genes (Atg) and many different signaling cascades (Kurz et al., 2008). Without the aid of autophagy the turnover of cytosolic proteins for quality control is impaired, which increases their propensity to become damaged, misfolded and subsequently ubiquitinated and aggregated (Hara et al., 2006; Komatsu et al., 2006). Basal and optimally induced levels of autophagy are particularly essential for neurons, because most of them are unable to re-distribute damaged proteins and organelles through cell division. Increasing attention has been recently focused on the role of autophagy in neuronal cell death associated with acute neuronal injury and chronic neurodegenerative diseases (Jaeger and Wyss-Coray, 2009). Two main types of mammalian autophagy have been identified and implicated in CNS injury and disease: chaperone-mediated autophagy and macroautophagy. Chaperone-mediated autophagy has been described only in mammals, and is involved in the degradation of misfolded, damaged or oxidized proteins. Mechanisms of chaperone-mediated autophagy involved in such substrate proteins comprise of four steps (Fig. 3) (Agarraberes and Dice, 2001). Initially, the cytosolic Hsc70 (heat-shock constitutive protein 70), which is highly homologous to its induced type Hsp70, activates the lysosomal proteolytic pathway. The Hsc70 chaperone system binds to the substrate protein at its pentapeptide KFERQ motif in the cytosol (step ). Next, this cytosolic Hsc70 chaperone system–substrate protein complex docks at the lysosome-associated membrane protein 2a (Lamp2a) (step )

348

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Fig. 4. A flow-chart interfacing the core autophagic pathway with cell death and survival mechanisms. Cytoplasmic contents to be degraded are engulfed in doublemembrane vesicles of 300–900 nm in diameter, termed autophagosomes, which fuse with lysosomes to become autolysosomes in which the cargo contents are degraded. Autophagy operates to both promote cell survival and induce cell demise. For cell survival, autolysosomes must degrade damaged proteins or organelles to generate amino and fatty acids that are recycled for de novo biosynthesis. Then, its failure causes protein aggregation, leading to cell death, although the causal relationship between autophagy levels and cell death is unclear. The Ser/Thr protein kinase TOR (originally identified in yeast) is a nutrient sensor and negative regulator of autophagy, while rapamycin is an autophagy inducer. Rapamycin is a drug currently used to inhibit renal transplant rejection in patients. 3-MA, 3-methyladenine; TOR, target of rapamycin; TCA, tricarboxylic acid (cited from Kourtis and Tavernarakis, 2009).

(Cuervo and Dice, 1996). By forming oligomers (up to octamers), Lamp2a can form the protein translocation pore (Fig. 3). Then, the substrate protein itself can bind to a lysosomal receptor Lamp2a (Cuervo and Dice, 1996), and subsequently be transported across this pore (step ). This transport requires the assistance of a luminal Hsc70 chaperone (lyHsc70) (Agarraberes et al., 1997; Cuervo et al., 1997). Unfolding of the substrate protein is required for its transport into the lysosomal lumen (Salvador et al., 2000). Finally, the substrate protein is rapidly degraded to single amino acids within the lumen of lysosomes by low specificity hydrolases (step ). These recycled amino acids are used for synthesis of new proteins or as an energy source. Contrary to chaperone-mediated autophagy, macroautophagy is a bulk degradation pathway (Fig. 4), and quite distinct in its involvement in the physiological control of the intracellular organelles. At least 12 ATG and 4 other proteins are known to be involved in the initiation and execution of mammalian macroautophagy (Jaeger and Wyss-Coray, 2009). Basal and induced macroautophagy is important for the maintenance of tissue homeostasis, because it can degrade parts of the cytoplasm or whole organelles such as mitochondria, peroxisomes, and ER. In response to different forms of stress, including nutrient withdrawal, growth factor depletion, hypoxia, ER stress, and accumulation of protein aggregates, macroautophagy acts as a prosurvival process (Kourtis and Tavernarakis, 2009). Macroautophagy involves sequestration of the substrate organelles within autophagosomes which fuse with the lysosomal compartments to form autolysosomes. The pH of autophagosomes is initially the same as the surrounding cytosol, however, their fusion with lysosomes promotes the delivery of lysosomal hydrolases and provides the optimal acidic pH for effective degradation. In autolysosomes, hydrolases generate free fatty and amino acids for reutilization, and maintain ATP levels for cell survival. Macroautophagy initially plays a role in eliminating damaged organelles during stress (Xiong et al., 2007), but may eventually contribute to cell death if overwhelmed (Fig. 4) (Kourtis and Tavernarakis, 2009). One should, however, keep in mind that there is no evidence so far that autophagy plays any role in mammalian cell

death in vivo, despite many examples in cultured cells (Scarlatti et al., 2009). Autophagy is a self-degradation process that is involved in regulation of cell survival and death (Cuervo, 2004; Shintani and Klionsky, 2004; Baehrecke, 2005). Such dual role of autophagy is context-dependent: it can protect against cell demise in response to mild stressors, but eventually cause cell death in response to severe stress. Therefore, when discussing autophagy as a cell death mechanism, one should discern calpain–cathepsin mediated autophagy from calpain–cathepsin mediated necrosis. Impaired autophagy can be deleterious by failing to provide energy for essential cell functions, or by allowing accumulation of damaged cellular components and aggregation of dysfunctional proteins. Moreover, it may catabolize vital amounts of cell components, interfere with pro-survival mechanisms, and digest anti-apoptotic survival factors (Kourtis and Tavernarakis, 2009). Consequently, impaired autophagy may lead to autophagic cell death that is characterized by increased autophagosomes that are used for selfdegradation (Edinger and Thompson, 2004). However, the inhibition of specific ATG proteins involved in autophagy regulation can change the morphological appearance of cell death to necrosis (Golstein and Kroemer, 2007). The complex interplay among autophagy, cell survival, and cell death was discussed elsewhere in detail (Golstein and Kroemer, 2007; Pivtoraiko et al., 2009; Scarlatti et al., 2009; Kourtis and Tavernarakis, 2009). Taking into account the overall mechanism of cell maintenance, survival and death, the role of lysosomes is indispensable. 3. Role of lysosomes in cell death (Fig. 5) During the past decade, lysosomal rupture/permeabilization has quickly emerged as a prominent area of research to elucidate mechanisms of cell death. Lysosomes (indicating ‘Lytic bodies’) were first described in 1955 by de Duve and his collaborators in Belgium as a cellular organelle full of acidic hydrolases that are potentially harmful for the cell itself. Soon after the discovery, lysosomes were thought to be related to necrosis ensuing after cell damage, but have not been generally considered as its primary

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

349

Fig. 5. Hypothetical mechanisms of the lysosomal permeabilization for programmed cell death. The lysosomal permeabilization can be promoted by cathepsins, sphingosine, ROS or Bcl-2-like proteins that translocate to the lysosomal membrane (as shown in red dot lines). Of note, Hsp70 physiologically stabilizes lysosomal membrane. The interaction of released cathepsins with pro-apoptotic Bcl-2 family members, leads to the induction of mitochondrial apoptosis via release of cytochrome c and apoptosisinducing factor. SIMPs, soluble intermembrane mitochondrial proteins (cited from Tardy et al., 2006).

cause for long time. Whereas lysosomes have been considered for many years as a degradation compartment, it is now established that they can participate in many biological processes such as antigen presentation by MHC II, bone resorption, tumor progression, and programmed cell death (Dell’Angelica et al., 2000). Lysosomes contain more than 80 acidic hydrolases for degradation of proteins, nucleic acids, polysaccharides, lipids, and their conjugates. Lysosomal enzymes comprise a variety of proteases with the representative group being cathepsins, nucleases, glycosidases, sulfatases, and lipases. Cathepsins are subdivided into three subgroups according to the amino acid of the active site that confers catalytic activity: cysteine (cathepsins B, C, F, H, K, L, N, O, S, T, U, W and X), aspartyl (cathepsins D and E), and serine (cathepsins A and G) cathepsins. Among them, cathepsins B (EC 3.4.22.1), L (EC 3.4.22.15) and D (EC 3.4.23.5) are abundant in neurons. Although the putative action of lysosomal enzymes in cell death was first postulated by de Duve in 1955, it has long remained unexplored. Only in the last decade, an expanding interest in the implication of lysosomal constituents has been witnessed in both apoptosis and necrosis (Fig. 5). The cytoplasmic release of lysosomal cathepsins B and D precedes cell death following various insults such as tumor necrosis factor-a (TNF-a), staurosporine treatments, ultra-violet irradiation, and ¨ llinger, 2000; Foghsgaard oxidative stress (Olsson et al., 1989; O et al., 2001; Ka˚gedal et al., 2001; Werneburg et al., 2002; Johansson et al., 2003; Kurz et al., 2006; Nilsson et al., 2006; Bivik et al., 2006, 2007; Hwang et al., 2008). In each of these cases, a specific inhibitor of cathepsin B or D attenuated cell damage. Furthermore, activation of lysosomal pathways has been recognized as an early feature of Alzheimer’s disease (Adamec et al., 2000), because of a massive increase in cathepsin B and D levels as well as in the number of lysosomes in the vulnerable neurons (Nixon et al., 2000). Lysosomal dysfunction has been proposed to play a role in other neurodegenerative diseases such as Parkinson’s and Huntington’s disease, but evidence supporting

a role of lysosomal damage in neuronal cell death is still missing (Yuan et al., 2003). An important mechanism by which lysosomal hydrolytic enzymes can directly contribute to cell death is through lysosomal destabilization and enzyme leakage into the cytoplasm (Figs. 1 and 5). This phenomenon has been initially described in vitro during ¨ llinger and Brunk, 1995; oxidative stress in non-neuronal cells (O Brunk et al., 1997, 2001; Brunk and Svensson, 1999; Brunk et al., ¨ llinger, 1998), and has been first demonstrated 2001; Roberg and O in vivo during experimental brain ischemia in the hippocampal neurons of primates (Yamashima et al., 1996, 1998, 2003; Yamashima, 2000, 2004). Ischemic injury or stress agents can trigger cytosolic acidification and rupture/permeabilization of lysosomes. The lysosomal rupture/permeabilization can be blocked by Hsp70, whereas promoted by some sphingolipids, cathepsins, ROS or, possibly, by Bcl-2-like proteins that translocate to the lysosomal membrane. The lysosomal rupture/permeabilization would enable cathepsins to degrade vital proteins, and activate additional hydrolases including caspases and DNase II (Fig. 5) (Tardy et al., 2006). 3.1. Implication of oxidative stress in lysosomal rupture (Fig. 6) Previous studies have shown that lysosomal membrane becomes more susceptible to additional damage when inflicted by free radicals or ROS (Coyle and Puttfarcken, 1993; Beal, 1996; Owen et al., 1996; Simonian and Coyle, 1996; Jenner, 1998; Halliwell, 2001), because the incorporation of molecular oxygen into polyunsaturated fatty acids initiates a chain of reactions. ROS, especially hydroxyl radicals, can produce functional alterations in lipids and proteins. It is likely that oxidative lipid damage (lipid peroxidation) causes loss of membrane fluidity and increment of membrane permeability (Pignol et al., 2006). Lysosomal membrane permeabilization occurs without any apparent ultrastructural alterations, whereas lysosomal membrane rupture occurs

350

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Fig. 6. Synthesis of hydroxyl radicals from hydrogen peroxide (a): the cell’s antioxidant defense system, diffusion of hydrogen peroxide (H2O2) into lysosomes, and Fentontype reactions (Fe2+ + H2O2 ! Fe3+ + HO + OH) producing the most powerful oxidant hydroxyl radicals (HO). Hydrogen peroxide is generated continuously in cells, mainly by loss of electrons from the mitochondrial electron transport chain, and some may diffuse into lysosomes by escaping anti-oxidative defense. As lysosomes degrade ironcontaining macromolecules during autophagy, they contain abundant iron which unfortunately contributes to generation of hydroxyl radicals (cited from Kurz et al., 2008), Oxidation of arginine (b): structure of non-oxidized arginine (R) and carbonylated arginine (R*) by oxidation, showing a decrease in molecular weight. Because of this decrease (from 157.20 to 113.12), arginine carbonylation can be precisely detected as shown in Fig. 7 (cited from Oikawa et al., 2009).

with apparent ultrastructural alterations (Yamashima et al., 1996). Then, such a question emerges ‘how is the lysosomal membrane rupture induced?’. It is far from elucidated at present how ROS may react with various cellular constituents. ROS, including hydrogen peroxide (H2O2) and oxygen-derived free radical superoxide (O2), and, most importantly, hydroxyl radical (HO), are known to play a major role in neuronal death, but the exact source of ROS has been unknown until now. It is quite reasonable to think that different pathways might dominate in different insults and different species. In neurodegenerative diseases, mitochondria are the main generator of ROS which induce mitochondrial dysfunction, and this, in turn, leads to higher ROS production, establishing a destructive malignant cycle (Trushina and McMurray, 2007). Another source of ROS, especially of hydroxyl radical (HO), is lysosomes. Lysosome-mediated autophagy allows turnover of proteins and other macromolecules, and results in intralysosomal release of low mass redox-active iron when iron-rich materials, such as damaged mitochondria, iron-saturated ferritin, and ironcontaining metalloproteins are degraded. Consequently, lysosomes become a compartment with a high concentration of low mass iron (Fe) (Terman et al., 2006b), being potentially redoxactive. Since the lysosomal compartment is acidic, and rich in reducing equivalents such as cysteine, iron would partly exist in the ferrous (Fe2+) form, capable of preparing Fenton-type reactions (Terman et al., 2006a). The resultant accumulation of Fe3+ enhances the local production of HO through the Fenton reaction and catalyzes conversion of H2O2 (being produced within mitochondria and diffusing into lysosomes) into HO (Fig. 6a). This eventually leads to lysosomal membrane damage (Figs. 1 and 5) (Terman et al., 2006a). Redox-active lysosomal iron seems to be crucial for lysosomal stability during oxidative stress. Intriguingly, HNE promotes zinc accumulation in lysosomes before their rupture. Because zinc itself elicits increased HNE levels in lysosomes, this leads to a detrimental cycle. It is difficult to determine which of the two is the main cause of lysosomal membrane permeabilization, because chelation of zinc not only inhibits permeabilization, but also suppresses HNE levels (Hwang

et al., 2008). Still, it is more important to identify the substrate of ROS at the lysosomal membrane of ishemia-vulnerable neurons. HNE is generated through peroxidation of n-6 polyunsaturated fatty acids, such as linoleic acid and arachidonic acid in the membrane phospholipids (Benedetti et al., 1980; Esterbauer et al., 1991). Abundant HNE is produced in response to ROS and has been proposed to mediate many of the toxic effects of ROS in vivo (Esterbauer et al., 1991). HNE reacts with various proteins, promotes protein carbonylation (Trevisani et al., 2007), and ultimately causes their dysfunction (Yoritaka et al., 1996; Sayre et al., 2006). In lysosomes, accumulation of HNE-modified proteins leads to lysosomal stress and rupture (Marques et al., 2004; Hwang et al., 2008). As HNE is a key endogenous neurotoxin produced under oxidative stress (Trevisani et al., 2007), exposure of neurons to HNE triggers cell death via lysosomal rupture (Castino et al., 2007; Hwang et al., 2008). Then, what is the substrate protein being carbonylated by HNE at the lysosomal membrane? Exactly how ROS and HNE together damage lysosomal membrane constituents has been far from elucidated. Support for the possibility that oxidative processes may provide the link between activated calpain and Hsp70.1, was first provided by recent proteomic analyses indicating that activated m-calpain can proteolyze Hsp70.1 in the primate retina suffering from glaucoma (Nakajima et al., 2006). Furthermore, in the pre-heated human epidermal melanocytes exposed to ultra-violet radiation, Hsp70 colocalized with lysosomes and mitochondria in the surviving cells, whereas in the apoptotic cells both organells showed membrane permeabilization, as shown by the release of cathepsin D and cytochrome c (Bivik et al., 2007). Because heat pre-treatment prevented release of cathepsin D and cytochrome c, and Hsp70 short interfering RNA (siRNA) also eliminated apoptosis prevention, Hsp70 was thought to stabilize membranes of lysosomes and mitochondria in favor of cell survival (Bivik et al., 2007). Since the proposal of the ‘‘calpain–cathepsin hypothesis’’ (Yamashima et al., 1996, 1998), the role of calpain in executing lysosomal rupture has been unclarified. However, following recent proteomic analyses, a new possibility of ‘calpain cleavage of Hsp70 at the lysosomal

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

membrane’ has emerged as a cause of its rupture (Nakajima et al., 2006; Oikawa et al., 2009). 4. Hsp70 and its major form Hsp70.1 The discovery of heat-inducible chromosome puffs in Drosophila larvae in 1962 initiated a rapidly expanding research field on heat-shock response proteins (Ritossa). The latter consist of constitutive (Hsc) and inducible (Hsp) subtypes that function differently at various physiological conditions. Under unstressed conditions, Hsc serves as molecular chaperones assisting the process of protein recycling (Fig. 3), whereas under stressed conditions Hsp functions as stress-resistant protein (Becker and Craig, 1994; Yang et al., 2008). Since 1962, several independent groups have reported that non-lethal heat-shock-induced Hsp70 (Li and Werb, 1982; Landry et al., 1982; Riabolow et al., 1988; Johnston and Kucey, 1988) can enhance recovery of stressed cells from subsequent severe heat shock as well as other lethal stimuli (Gerner et al., 1976; Sapareto et al., 1978; Henle et al., 1978; Petersen and Mitchell, 1981; Ja¨a¨ttela¨, 1999). Hsp70 is structurally and functionally conserved in evolution, and found in all organisms from bacteria to humans (Hunt and Morimoto, 1985; Lindquist and Craig, 1988; Gupta and Singh, 1994). All eukaryotes have more than one gene encoding up to 10 putative family member proteins. The relationships between the hsp70 genes in rats, humans and

351

mice can be summarized as follows: rhsp70.1 = hhsp70.2 = mhsp70.1, rhsp70.2 = hhsp70.1 = mhsp70.3, and rhsp70.3 = hhsp70.Hom = mhsc70t (r indicates rat; h, human; and m, mouse) (Lee et al., 2001). Whereas the expression of Hsp70 is mainly stress-inducible in normal cells, all cells constitutively express Hsc70 (Doulias et al., 2007). In normal cells under non-stressed conditions, Hsp70 is poorly expressed but induced in response to damaging stimuli. Hsp70 is comprised of Hsp70.1 and Hsp70.2, and is rapidly induced in response to environmental, chemical or physical stimuli. As Hsp70.1 (Fig. 7a) is a major protein of the human Hsp70 family, a large part of the data published on the human Hsp70 family deals with Hsp70.1 (Daugaard et al., 2007). Eight highly homologous members of Hsp70 differ from each other in the intracellular localization and expression pattern (Tavaria et al., 1996; Ja¨a¨ttela¨, 1999). Six of them (Hsp70.1A, Hsp70.1B, Hsp70.1L, Hsp70.2, Hsp70.6, and Hsc70) reside mainly in the cytosol, while mt-Hsp70 localizes to the mitochondria (Daugaard et al., 2007). Hsp70 and Hsc70 show an extensive amino acid sequence similarity, but disclose distinct bands on twodimensional Western blotting (Manzerra et al., 1998). Different tissues show considerable variation in the basal levels of Hsp70 and Hsc70 proteins. Hsp70 and Hsc70 are predominantly expressed in neurons (Manzerra et al., 1998). Neural tissues, such as retina and cerebellum, show high basal levels of Hsp70,

Fig. 7. Proteomics of Hsp70.1 for analyzing protein expression and carbonylation. 3D structure of human Hsp70.1 (a) (cited from www.thesgc.com/.../StructureDescription/ 3GDQ.php) Upregulation of Hsp70.1 protein after ischemia (b): a representative two-dimensional differential in-gel electrophoretic analysis (2D DIGE) image from both the control (red) and day 5 (green) CA1 proteins after the ischemia/reperfusion. Two green spots within the red circle indicate up-regulation, and were identified as Hsp70.1. MW: molecular weight marker, pI: isoelectric point. Dynamic change of Hsp70.1 carbonylation (c): two-dimensional gel electrophoresis with immunoblot detection of carbonylated protein analysis (2D Oxyblot) of the control and days 3, 5, 7 CA1 samples. The Hsp70.1 spots were identified by peptide mass fingerprinting. The pI difference between C3 and C2 spots (the same as green sptos within the red circle of b) was 0.05, suggesting phosphorylation of Hsp70.1. Arg469 carbonylation on Hsp70.1 (d): matrixassisted laser desorption ionization-time of flight/time of flight (MALDI-TOF/TOF) analysis with the Mascot search. Both the peptide sequence of the carbonylated peptide ion (459-FELSGIPPAPR*G-470) and the presence of y2 fragment ion at m/z 113.12, indicated that a critical carbonylation occurred at Arg469 in Hsp70.1. R*: Carbonylated arginine (cited from Oikawa et al., 2009).

352

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

although its induction by hyperthermia is less remarkable than in non-neural tissues. In contrast, neural tissues exhibit 3-fold higher amounts of Hsc70, compared to non-neural tissues. Levels of Hsc70 are similar between the control and hyperthermic tissues. The high constitutive level of Hsc70 in neural tissues may play a role in modulating the level of Hsp70 induction in the nervous system (Manzerra et al., 1998). In 1984, Pelham suggested that the ability of Hsp70 to enhance recovery of stressed cells was mediated by its chaperoning function of the house-keeping proteins against stress. Hsp70 family members function as ATP-dependent molecular chaperones that assist folding of newly synthesized polypeptides, assembly of multiprotein complexes, and targeting of proteins for lysosomal degradation. Their ability to disaggregate, refold, and renature misfolded proteins as well as to stabilize both lysosomal and mitochondrial membranes offsets the otherwise fatal consequences of damaging stimuli (Bivik et al., 2006). Accumulating data in the last decade clearly demonstrate that Hsp70 has essential anti-apoptotic properties (Lanneau et al., 2008). Hsp70 was reported to be a decisive negative regulator of the mitochondrial pathway of apoptosis that can block apoptosis by inhibiting stress-inducing signaling, preventing mitochondrial membrane permeabilization through the blockage of Bax translocation, and interacting with apoptosis-inducing factor and Apaf-1 (Schmitt et al., 2007). Hsp70.1 mainly functions as a chaperone enabling the cell to cope with harmful aggregations of denatured proteins during and following the insults, such as heat, ischemia, and oxidative stress (Hartl, 1996; Ja¨a¨ttela¨, 1999). The chaperone function of Hsp70.1 is required for its cytoprotection against heat stress by inhibiting accumulation of protein aggregates and thereby removing the stimulus that triggers cell death (Nollen et al., 1999; Mosser et al., 2000). Accordingly, its expression at the luminal side of the lysosomal membrane confers cytoprotection against stress (Nylandsted et al., 2000, 2004; Gyrd-Hansen et al., 2004; Mambula and Calderwood, 2006; Bivik et al., 2007; Doulias et al., 2007). Of note, by immunoelectron microscopy, Nylandsted et al. (2004) found that Hsp70 is localized at the lysosomal membranes in tumor cells in order to prevent lysosomal rupture, presumably by direct binding of its amino terminus to an anionic lipid, lysobisphosphatidic acid (Ja¨a¨ttela¨, 1999). Hsp70 effectively inhibits permeabilization of the lysosomal membrane after treatment with TNF-a, oxidative stress, irradiation, and etoposide. By stabilizing the lysosomal membrane, Hsp70.1 inhibits release of lysosomal hydrolases into the cytosol (Leist and Ja¨a¨ttela¨, 2001; Nylandsted et al., 2004; Kroemer and Ja¨a¨ttela¨, 2005; Figueiredo et al., 2008). Although the role of Hsp70 in apoptotic cell death has been extensively investigated (Ravagnan et al., 2001; Nylandsted et al., 2000, 2004; Stankiewicz et al., 2005; Beere, 2005; Kroemer and Ja¨a¨ttela¨, 2005), the molecular mechanism underlying its antiapoptotic action still remains elusive (Doulias et al., 2007). In the delay period between the ischemic insult and overt cell death, CA1 neurons contain elevated levels of anti-apoptotic proteins such as Bcl-2, phospho-Akt, X-linked inhibitor of apoptosis (XIAP), brainderived neurotrophic factor (BDNF), calpastatin, BDNF/tyrosine kinase B (TrkB), glial cell line-derived neurotrophic factor (GDNF), neuronal apoptosis inhibitor protein (NAIP) or CuZn superoxide dismutase (Mu¨ller et al., 2007). The most important among them is Hsp70 (Vass et al., 1988; Aoki et al., 1993; Mu¨ller et al., 2007), and the role of Hsp70, for example, in ischemic insults is now beginning to be better understood. Cerebral infarction after 6 h of ischemia was significantly smaller in the transgenic mice overexpressing rat Hsp70 than in the wild-type mice (Rajdev et al., 2000). In contrast, the hsp70.1-knockout mice showed larger and denser infarcted areas than their wild-type littermates (Lee et al., 2001). Furthermore, in the pancreas of hsp70.1-knockout mice, increased amount

of lysosomal cathepsin activity was found outside the lysosomes (Hwang et al., 2008). These data altogether indicate that Hsp70 can stabilize the lysosomal membranes to prevent release of lysosomal proteases to the cytosol (Fig. 5). Lysosomes are rich in low mass redox-active iron, and the interaction of H2O2 with ferrous iron in lysosomes causes H2O2induced lysosomal destabilization, DNA damage and apoptosis (Brunk et al., 2001; Terman et al., 2006a,b; Doulias et al., 2007). The release of redox-active iron into the cytosol with subsequent movement to the nucleus may determine DNA sensitivity toward H2O2-induced oxidative modifications. Compared to control HeLa cells that express normal Hsp-70 levels, the Hsp70-overexpressing ones were significantly resistant to H2O2-induced DNA damage, while Hsp70-depleted cells showed an enhanced vulnerability. In addition, pretreatment with 1 mM iron-chelator desferrioxamine protected both Hsp70-overexpressing and Hsp70-depleted HeLa cells equally well from the ensuing H2O2-induced oxidative stress (Doulias et al., 2007). Accordingly, Doulias et al. (2007) suggested that Hsp70 may modulate H2O2-induced nuclear DNA singlestrand breaks, and this effect is related to the regulation of cellular iron homeostasis and stabilization of lysosomal membranes. Furthermore, dermal fibroblats of aged people (61–77 years) showed both decreased response of Hsp70 after heat-preconditioning and decreased resistance to ischemic and oxidative stress, compared to those of young people (15–28 years) (Tandara et al., 2006). The main mechanisms by which Hsp70 confers cell protection appear to be not only the chaperoning function of denaturing proteins following stress but also the inhibition of rupture and/or permeabilization of lysosomal membranes (Figueiredo et al., 2008). Therefore, it is proposed that the dysfunction of Hsp70 at the lysosomal membrane is evoked by oxidative stress and induced with the aid of redox-active iron. Why and how does the dysfunction of Hsp70 occur at the lysosomal membrane to cause its destabilization and the resultant leakage of cathepsins? 4.1. Upregulation and carbonylation of Hsp70.1 after ischemia (Fig. 7) ROS play an important role in the progression of neuronal death associated with various diseases. For example, loss of cholinergic neurons frequently occurs in the forebrain of Alzheimer’s disease patients, dopaminergic neurons in the substantia nigra are selectively injured in Parkinson’s disease, and motor neurons in the spinal cord are selectively lost in amyotrophic lateral screlosis (Ischiropoulos and Beckman, 2003). Oxidative protein damage has been implicated as one of the leading causes of these neurodegenerative diseases. In order to elucidate rational mechanisms for the relationship between protein oxidation and neuronal death, the identification of specifically modified substrate proteins is indispensable. Oxidative modification of a protein, especially carbonylation, affects the function and/or metabolic stability of the modified proteins (Levine, 2002). Carbonylation makes the substrate protein more susceptible to proteolysis due to unfolding of its targeted domains. Unfolding exposes amino-acid residues that were normally hidden in the protein structure but became more prone for degradation in response to the proteasome and the Lon protease (Grune et al., 2004; Nystro¨m, 2005). We speculate that the same may occur between activated calpain and carbonylated Hsp70.1 at the lysosomal membrane of postischemic CA1 neurons. Such interaction was reinforced by both protein expression and functional proteomics analysis. With the aid of two-dimensional differential in-gel electrophoresis (2D DIGE), the authors’ group recently reported postischemic upregulation of Hsp70.1 in postischemic monkey CA1 as compared to normal tissue (Oikawa et al., 2009). Hsp70.1 protein (Fig. 7a) expression was remarkably upregulated on days 3 and 5

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

(Fig. 7b) after ischemia, compared to the non-ischemic control. Protein carbonylation is the major and most common oxidative modification of proteins (Stadtman and Berlett, 1997), being increased by ischemia–reperfusion injury in the brain of gerbils (Oliver et al., 1990), and beagles (Liu et al., 1993). Carbonyl derivatives are formed by ROS-mediated oxidation of side-chains of such amino acid residues as threonine, lysine, arginine (Fig. 6b), and proline (Aksenov et al., 2001; Bizzozero et al., 2007). Carbonyl groups, being introduced into amino-acid side chain following oxidative modification of proteins (Butterfield and Stadtman, 1997), were detected by reaction with 2,4-dinitrophenylhydrazine (DNPH). The 2,4-dinitrophenylhydrazone (DNP) derivative produced by this reaction can be immunodetected by the specific antiDNP protein adduct antibody. Using two-dimensional gel electrophoresis with immunoblot detection of carbonylated proteins (2D Oxyblots), oxidized proteins were analyzed in the CA1 tissues of the non-ischemic control, and 3, 5, and 7 days after the ischemia– reperfusion insult. Carbonyl levels of two Hsp 70.1 spots were negligible before the ischemic insult, but increased significantly on days 3 and 5 after ischemia (Fig. 7c). Furthermore, using matrixassisted laser desorption ionization-time of flight/time of flight (MALDI-TOF/TOF) analysis of peptides obtained from the carbonylated Hsp70.1 spots, the authors’ group first found that the critical carbonylation in the Hsp70.1 protein occurs at the key amino acid, Arg469 (R*) (Fig. 7d) (Oikawa et al., 2009). 4.2. Possible cleavage of carbonylated Hsp 70.1 by activated calpain During cerebral ischemia, lack of oxygen and glucose leads to a shortage of energy for sustaining the neuronal membrane potential. As a consequence, plasma membrane depolarization occurs, which results in massive release of glutamate at the synaptic clefts. Overstimulation of NMDA and AMPA receptors by glutamate leads to remarkable Ca2+ mobilization in the postsynaptic neurons. Additionally, the secondary activation of voltage-gated Ca2+ channels also contributes to Ca2+ overload (Artal-Sanz and Tavernarakis, 2005). Massive Ca2+ increase occurs only during ischemia, whereas calpain activation persists long after Ca2+ has returned to normal levels (Yamashima, 2004). Calpain activation has been demonstrated in ischemic neuronal death, but the exact role of calpains in cell death still remains unknown. Possible cross-talks (Fig. 2) among the three cysteine protease systems of calpain, cathepsin and caspase, comprise of calpain-mediated cathepsin release (Yamashima et al., 1996, 1998), cathepsin-mediated caspase-activation (Ishisaka et al., 1998; Vancompernolle et al., 1998), and caspase-mediated calpastatin degradation (Wang et al., 1998; Rami et al., 2000). It is likely that these cross-talks are also related to neuronal death associated with various neurodegenerative diseases. The calpain family of proteases was first established with the discovery of m-calpain in 1964 (Guroff, 1964). Among 15 identified calpain family members within the human genome (Bevers and Neumar, 2008), the most abundant brain calpains are two major isoforms; m-calpain (calpain 1) and m-calpain (calpain 2), which differ in their sensitivity to Ca2+. Calpains are heterodimers, consisting of two different polypeptide subunits. The larger 80-kDa subunit has a catalytic activity, whereas the smaller 30-kDa subunit has a regulatory function. The catalytic site cysteine is present at domain II that is composed of two subdomains (IIa and IIb) forming the substrate binding pocket between them (Suzuki et al., 2004). Domains IV and VI contain five sets of EF-hand Ca2+binding motifs similar to those found in calmodulin. When Ca2+ binds to the EF-hands, the N-terminus of calpain undergoes autolytic cleavage accompanied by a conformational change, and the catalytic domain becomes activated. For this autolytic activation to occur, m-calpain requires 30–50 mM Ca2+ for half-

353

maximal activity, whereas m-calpain requires 400–800 mM Ca2+ (Goll et al., 2003). Calpain exists in the cytosol as an inactive enzyme and translocates to membranes in response to an increase in cellular Ca2+ levels. Binding of calpain to membrane phospholipids induces its conformational changes, which brings IIa and IIb closer to form a functional catalytic site. Activated calpain cleaves substrate proteins into two large fragments with retaining intact domains. Calpain is regarded as a bio-modulator, because properties of the substrate proteins are often modulated upon hydrolysis by calpain (Suzuki et al., 2004). Calpain activity is regulated by the ubiquitous specific inhibitor, calpastatin. In hippocampus, m-calpain is localized in both pyramidal and granular neurons while m-calpain is localized in the interneurons (Zatz and Starling, 2005). An increase in the concentrations of cytoplasmic Ca2+ may play a major role in the initiation of downstream cell death events. Hippocampal interneurons are generally more resistant than pyramidal cells to excitotoxic insults (Avignone et al., 2005). This is possibly because Ca2+ mobilization during insults is sufficient to activate m-calpain but insufficient to activate m-calpain. Increases in intracellular Ca2+ concentration during transient ischemia were sufficient to trigger activation of m-calpain, which may in turn damage lysosomal membrane with the aid of oxidants, and induce cathepsin release (Fig. 1). Currently, there is insufficient evidence to place oxidative stress in the center of the pathogenic mechanism that leads to neuronal death. Nevertheless, the pathway of calpain-mediated cleavage of carbonylated Hsp70 accompanied by the resultant lysosomal membrane rupture, represents an emerging paradigm for the coordinated, multistep regulation of signaling events to facilitate neuronal degeneration and death after exposure to the ischemic insult. We speculate that the extent of calpain activation and Hsp70 carbonylation in each injury may determine the extent of lysosomal permeabilization and eventually the cell death pattern. Even in the presence of Ca2+-induced calpain activation, antioxidants such as N-tert-butyl-a-phenyl nitrone (a spintrapping agent) or ()-6-hydroxy-2,5,7,8-tetramethylchromane-2carboxylic acid (a water-soluble analog ofa-tocopherol) can inhibit extra-lysosomal release of cathepsins in the hippocampal slices following 5 min oxygen and glucose deprivation followed by 60– 120 min reperfusion (Windelborn and Lipton, 2008). Accordingly, it is likely that oxidation of lysosomal membrane proteins such as Hsp70 may be an upstream event of calpain-mediated rupture/permeabilization of the lysosomal membrane. Without oxidative stress of the lysosomal membrane, mild calpain activation alone cannot induce drastic release of cathepsins. Calpain activation may be necessary but not sufficient, if the extent of activation is less intense. The substrate oxidization might be indispensable for slightly activated calpain to achieve cleavage of Hsp70. On the contrary, 10 min oxygen and glucose deprivation followed by 15 min reperfusion can immediately induce lysosomal rupture in the early phase of reperfusion (Windelborn and Lipton, 2008). This is presumably because calpain activation, if occurring drastically, can alone cleave lysosomal membrane without the help of ROS. This is consistent with the fact that damage in the ischemic core of stroke patients occurs more rapidly without a detectable change in ROS and is not attenuated by free radical scavengers, whereas neuronal damage in the ischemic penumbra occurs gradually within hours to days following the initial insult and is attenuated by free radical scavengers (Solenski et al., 1997, 2000). 5. Modified ‘‘calpain–cathepsin hypothesis’’ (Fig. 8) The role of lysosomes and more particularly of the lysosomal enzymes cathepsins in cell death has been unmasked in the past decade, and the importance of lysosomal membrane rupture/

354

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Fig. 8. A flow chart showing the modified ‘‘calpain–cathepsin hypothesis’’. Calpain plays a major role for causing lysosomal rupture while ROS play a supportive role (Yamashima, 2000). Oxidation of n-6 PUFA, such as linoleic and arachidonic acids, produces HNE that carbonylates Hsp70 (Hsc70) at the lysosomal membrane. Arg469 carbonylation of Hsp70 leads to not only loss of its chaperone function but also vulnerability to calpain cleavage, which altogether result in the rupture or permeabilization of lysosomal membranes. It is likely that severe calpain activation can alone rupture the lysosomal membrane if the insult is severe, but mild calpain activation can permeabilize it if Arg469 of the substrate protein Hsp70 is carbonylated by ROS beforehand. When Ca2+-induced calpain activation and ROS-induced Hsp70 carbonylation occur to the maximum and work in parallel, the lysosomal damages become most remarkable through increased calpain-proteolysis of carbonylated Hsp70. The representative event of the cascade in the upper column is illustrated in the lower from the representative data of monkey experimental paradigm. The major challenge for future studies is to demonstrate a direct interaction of activated m-calpain and Arg469-carbonylated Hsp70. ROS, reactive oxygen species; n-6 PUFA, n-6 polyunsaturated fatty acids such as linoleic and arachidonic acids; HNE, 4-hydroxy-2-nonenal; Hsp70, heat-shock protein-70.

permeabilization is becoming firmly established nowadays. Studies on postischemic monkey hippocampal CA1 neurons by Yamashima et al. (1996) using immunoelectron microscopy, revealed that activated m-calpain translocates to the lysosomal membrane and contributes to the disruption of the lysosomes. Using the same experimental paradigm, the ‘‘calpain–cathepsin hypothesis’’ was formulated in 1998, based on such observation that calpain-mediated lysosomal rupture caused necrosis of the CA1 neurons via the release of cathepsins, and cathepsin inhibition prevented CA1 neuronal death (Yamashima et al., 1998). In 1997, the role of lysosomal rupture in cell death was also proposed in vitro by Brunk and coworkers who showed that photo-oxidative disruption of lysosomal membrane causes apoptosis of cultured human fibroblasts. Since then, numerous in vitro and in vivo studies have indicated that lysosomal enzymes can mediate or modulate cell death programmes (Tardy et al., 2006). Necrotic neuronal death underlies the pathology of devastating neurological diseases such as stroke, trauma, or various neurodegenerative disorders. However, the molecular players that induce necrotic neuronal death have not been fully elucidated, compared to those of apoptotic neuronal death. Despite activation of apoptosis-related death programs such as caspase-3 and CAD/ICAD activation, the majority of CA1 neurons failed to execute apoptosis, instead, developed necrosis via the aid of calpain–cathepsin cascade in the postischemic hippocampus (Tsukada et al., 2001). Since the formulation of the ‘‘calpain– cathepsin hypothesis’’ in 1998 (Yamashima et al., 1998), it became widely accepted that calpain-mediated lysosomal rupture plays a major role in programmed cell necrosis. The calpain-induced cathepsin leakage was confirmed in various experimental para-

digms ranging from C. elegans (Syntichaki et al., 2002) to humans (Adamec et al., 2000; Taniguchi et al., 2001). Immunoblotting of the Alzheimer brain revealed upregulation of calpain (Taniguchi et al., 2001) and cathepsin (Adamec et al., 2000). Immunohistochemical analysis of the Alzheimer brain showed a massive increase and enlargement (actually indicating lysosomal permeabilization) of cathepsin D-positive lysosomes in pyramidal neurons of the CA1-subiculum (Adamec et al., 2000). In the numerous experimental paradigms also, lysosomal cathepsins were demonstrated to translocate from the lysosomal lumen to the cytosol in response to a wide variety of insults such as the ones occurring in b-amyloid protein (Boland and Campbell, 2004), TNF (Guicciardi et al., 2000; Foghsgaard et al., 2001), Fas (Brunk and Svensson, 1999), p53 activation (Yuan et al., 2002), microtubule stabilizing agents (Broker et al., 2004), oxidative stress (Brunk et al., 1995, 1997; Brunk and Svensson, 1999), quinolinic acid (Figueiredo et al., 2008), staurosporine (Bidere et al., 2003), growth factor deprivation (Brunk and Svensson, 1999) and lysosomotropic agents (Brunk et al., 1995, 1997; Boya et al., 2003). Furthermore, in tumor cells lysosomal cathepsins have also been found to translocate from the lysosomal lumen to the cytosol in response to a wide variety of death stimuli such as TNF (Guicciardi et al., 2000), Fas (Brunk and Svensson, 1999), p53 activation (Yuan et al., 2002), microtubule stabilizing agents (Broker et al., 2004), oxidative stress (Brunk and Svensson, 1999), staurosporine (Bidere et al., 2003), growth factor deprivation (Brunk and Svensson, 1999), and lysosomotropic agents (Boya et al., 2003). Accordingly, lysosomal disintegration appears to be a crucial common event for living cells, and the key process for lysosomal rupture and/or organelle destabilization during necrosis is now becoming uncovered.

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Unfortunately, definitive proof of the ‘‘calpain–cathepsin hypothesis’’ has been hampered by the lack of an appropriate experimental procedure demonstrating an in situ link between calpain activation and lysosomal rupture/permeabilization. Here, however, the authors outlined a central step on the calpain– cathepsin cascade, by speculating the interaction between activated calpain and Arg469-carbonylated Hsp70.1. Chang et al. (2001) reported that Arg469 of the rat 70-kDa Hsc70 is crucial for binding and releasing certain peptides or unfolded proteins. The Arg469Cyst mutation reduced the affinity of Hsc70 for polypeptide substrates and its capability of refolding denatured luciferase. Furthermore, compared to wild type Hsc70, the mutant protein was more accessible to proteolytic cleavage by endopeptidase LysC with the overall structure becoming relatively loose (Chang et al., 2001). In the postischemic CA1 neurons of monkeys, the authors’ group has found that oxidative carbonylation occurs at this Arg469 of Hsp70.1 (Fig. 7d) (Oikawa et al., 2009). There is no direct evidence showing that Arg469 carbonylation of Hsp70.1 makes the protein susceptible to calpain. Nevertheless, it is intriguing to speculate that carbonylation of Arg469 of the lysosomal Hsp70.1 as induced by ROS-mediated HNE, promotes not only dysfunction of Hsp70.1 but also its susceptibility to activated calpain (Fig. 8). Because calpain activation is usually a very early event occurring during and/or immediately after the ischemic insult, calpain-inhibition strategies are unlikely to be actually effective to treat patients hospitalized many hours after stroke. The therapeutic time window for m-calpain inhibition ranges from at least the first two hours after an insult as in the model of global brain ischemia while up to six hours after the insult in a reversible focal cerebral ischemia model (Rami et al., 2008). In contrast, the therapeutic time window for cathepsin inhibition lasts a couple of days, because upregulation of cathepsins and lysosomal rupture occur mainly 3 days after the induction of transient ischemia (Yamashima et al., 1998). Indeed, virtually all neuroprotective treatment trials in stroke have failed until now (Ovbiagele et al., 2003). However, Yamashima et al. (1998) have successfully inhibited ischemic neuronal death by blocking cathepsin enzymes that are upregulated predominantly 3 days after ischemia. Intervention at the levels of lysosomal rupture, cathepsin release or cathepsin inhibition in neuronal injury should be potentially desirable for two reasons. First, there may be an increased window of therapeutic opportunity for blocking lysosomal rupture and cathepsin release, as opposed to blocking calpains. Second, cathepsin inhibition might have less deleterious effects on normal cell function than intervention at the calpain level.

355

of Hsp70, Hsp70.1. It can promote cell survival by recycling longlived and damaged proteins, and by stabilizing lysosomal membrane. Hsp70.1 is localized to lysosomal membranes, and its cleavage after carbonylation (although speculated from the circumstantial evidence) may cause lysosomal rupture, release of lysosomal constituents into the cytosol, and caspase-independent programmed cell necrosis. A satisfactory hypothesis on the mechanisms behind lysosomal rupture after ischemia/reperfusion injury must provide firm and distinct links between ischemia-induced calpain activation and reperfusion-induced Hsp70 oxidation. Hydrogen peroxide, in combination with lysosomal redox-active iron, produces hydroxyl radicals by the Fenton reaction, that cause carbonylation of Hsp70.1 at the lysosomal membrane. Although additional studies are needed, the authors would like to conclude that Arg469-carbonylated Hsp70.1 may be susceptible to cleavage by activated calpain, and the resultant lysosomal rupture may be a relevant mechanism executing neuronal necrosis. A number of critical issues are still to be solved including calpain-mediated cleavage of lysosomal Hsp70.1 and its blockage, the nature of Hsp70-modulating cell death, the mode of cathepsin release, their target substrates, the regulatory pathways, and cross-talks with other death-signalling pathways such as autophagic cell death. In view of the increasing evidence supporting a role of lysosomal rupture in ischemic neuronal necrosis, blocking this mechanism and inhibiting cathepsins may facilitate the development of effective neuroprotective strategies not only in stroke but also in traumatic brain injury and various neurodegenerative diseases. Calpain is necessary for the lysosomal membrane rupture/ permeabilization, but there is still insufficient evidence showing that calpain causes the lysosomal membrane rupture/permeabilization through cleavage of Hsp70.1 at the lysosome. Further studies to confirm the molecular interaction between activated mcalpain and the lysosomal Hsp70.1, especially focusing on the vulnerability of Arg469-carbonylated Hsp70.1, are expected to elucidate the exact mechanism of lysosomal rupture. Acknowledgements This work was supported by a grant (Creative Scientific Research: 17GS0317, Kiban-Kennkyu (B): 1839039) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. The author is deeply indebted to Dr. N. Tavernarakis (Institute of Molecular Biology and Biotechnology, Heraklion, Greece) for the careful revision of this manuscript.

6. Concluding remarks References Neurons die by caspase-dependent apoptosis as a normal physiological process during development, or alternatively die by calpain- and cathepsin-dependent necrosis in diseases. Neurons are sensitive to lysosomal dysfunction, and the contribution of ROS to lysosomal dysfunction may lead to various neurodegenerative diseases. The role of lysosomes in what one calls programmed cell death today was initially suggested by de Duve, but the important role of lysosomes in programmed neuronal necrosis was not recognized until Yamashima’s consecutive works using macaque monkey experimental paradigms. It is well known that possible calpain substrates in neurons are present within synapses, plasma membrane, ER, lysosomes, mitochondria, and the nucleus. However, the in vivo substrate of calpain in the postischemic neurodegeneration still remains undecided. In this review, the authors highlighted the role of two neuronal necrosis executor proteins; calpain and Hsp70.1. The in vivo substrate of calpain responsible for lysosomal rupture and the subsequent cathepsin leakage was identified to be a major type

Adamec, E., et al., 2000. Up-regulation of the lysosomal system in experimental models of neuronal injury: implications for Alzheimer’s disease. Neuroscience 100, 663–675. Agarraberes, F.A., et al., 1997. An intralysosomal hsp70 is required for a selective pathway of lysosomal protein degradation. J. Cell. Biol. 137, 825–834. Agarraberes, F.A., Dice, J.F., 2001. A molecular chaperone complex at the lysosomal membrane is required for protein translocation. J. Cell Sci. 114, 2491– 2499. Avignone, E., et al., 2005. Differential responses to NMDA receptor activation in rat hippocampal interneurons and pyramidal cells may underlie enhanced pyramidal cell vulnerability. Eur. J. Neurosci. 222, 3077–3090. Aksenov, M.Y., et al., 2001. Protein oxidation in the brain in Alzheimer’s disease. Neuroscience 103, 373–383. Aoki, M., et al., 1993. Temporal profile of the induction of heat shock protein 70 and heat shock cognate protein 70 mRNAs after transient ischemia in gerbil brain. Brain Res. 601, 185–192. Artal-Sanz, M., Tavernarakis, N., 2005. Proteolytic mechanisms in necrotic cell death and neurodegeneration. FEBS Lett. 579, 3287–3296. Baehrecke, E.H., 2005. Autophagy: dual roles in life and death? Nat. Rev. Mol. Cell Biol. 6, 505–510. Beal, M.F., 1996. Mitochondria, free radicals, and neurodegeneration. Curr. Opin. Neurobiol. 6, 661–666.

356

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Becker, J., Craig, E.A., 1994. Heat-shock proteins as molecular chaperones. Eur. J. Biochem. 219, 11–23. Beere, H.M., 2005. Death versus survival: functional interaction between the apoptotic and stress-inducible heat shock protein pathways. J. Clin. Invest. 115, 2633–2639. Benedetti, A., et al., 1980. Identification of 4-hydroxynonenal as a cytotoxic product originating from the peroxidation of liver microsomal lipids. Biochim. Biophys. Acta 620, 281–296. Bevers, M.B., Neumar, R.W., 2008. Mechanistic role of calpains in postischemic neurodegeneration. J. Cereb. Blood Flow Metab. 28, 655–673. Bidere, N., et al., 2003. Cathepsin D triggers Bax activation, resulting in selective AIF relocation in T lymphocytes entering the early commitment phase to apoptosis. J. Biol. Chem. 278, 31401–31411. Bivik, C.A., et al., 2006. UVA/B-induced apoptosis in human melanocytes involves translocation of cathepsins and Bcl-2 family members. J. Invest. Dermatol. 126, 1119–1127. Bivik, C., et al., 2007. Hsp70 protects against UVB induced apoptosis by preventing release of cathepsins and cytochrome c in human melanocytes. Carcinogenesis 28, 537–544. Bizzozero, O.A., et al., 2007. Lipid peroxidation scavengers prevent the carbonylation of cytoskeletal brain proteins induced by glutathione depletion. Neurochem. Res. 32, 2114–2122. Boland, B., Campbell, V., 2004. Ab-mediated activation of the apoptotic cascade in cultured cortical neurones: a role for cathepsin-L. Neurobiol. Ageing 25, 83–91. Boya, P., et al., 2003. Lysosomal membrane permeabilization induces cell death in a mitochondrion-dependent fashion. J. Exp. Med. 197, 1323–1334. Boya, P., Kroemer, G., 2008. Lysosomal membrane permeabilization in cell death. Oncogene 27, 6434–6451. Broker, L.E., et al., 2004. Cathepsin B mediates caspase-independent cell death induced by microtubule stabilizing agents in non-small cell lung cancer cells. Cancer Res. 64, 27–30. Broughton, B.R., et al., 2009. Apoptotic mechanisms after cerebral ischemia. Stroke 40, e331–e339. Brunk, U.T., et al., 1995. Exposure of cells to nonlethal concentrations of hydrogen peroxide induces degeneration–repair mechanisms involving lysosomal destabilization. Free Radic. Biol. Med. 19, 813–822. Brunk, U.T., et al., 1997. Photo-oxidative disruption of lysosomal membranes causes apoptosis of cultured human fibroblasts. Free Radic. Biol. Med. 23, 616–626. Brunk, U.T., Svensson, I., 1999. Oxidative stress, growth factor starvation and Fas activation may all cause apoptosis through lysosomal leak. Redox Rep. 4, 3–11. Brunk, U.T., et al., 2001. Lysosomal involvement in apoptosis. Redox Rep. 6, 91–97. Bursch, W., 2001. The autophagosomallysosomal compartment in programmed cell death. Cell Death Differ. 8, 569–581. Butterfield, D.A., Stadtman, E.R., 1997. Protein oxidation processes in aging brain. Adv. Cell Aging Gerontol. 2, 161–191. Castegna, A., et al., 2002. Proteomic identification of oxidatively modified proteins in Alzheimer’s disease brain. Part I: creatine kinase BB, glutamine synthase, and ubiquitin carboxy-terminal hydrolase L-1. Free Radic. Biol. Med. 33, 562–571. Castino, R., et al., 2007. Cathepsin D-Bax death pathway in oxidative stressed neuroblastoma cells. Free Radic. Biol. Med. 42, 1305–1316. Chang, T.C., et al., 2001. The effect of mutating arginine-469 on the substrate binding and refolding activities of 70-kDa heat shock cognate protein. Arch. Biochem. Biophys. 386, 30–36. Chua, B.T., et al., 2000. Direct cleavage by the calcium-activated protease calpain can lead to inactivation of caspases. J. Biol. Chem. 275, 5131–5135. Coyle, J.T., Puttfarcken, P., 1993. Oxidative stress, glutamate, and neurodegenerative disorders. Science 26, 689–695. Cuervo, A.M., Dice, J.F., 1996. A receptor for the selective uptake and degradation of proteins by lysosomes. Science 273, 501–503. Cuervo, A.M., et al., 1997. A population of rat liver lysosomes responsible for the selective uptake and degradation of cytosolic proteins. J Biol Chem. 272, 5606– 5615. Cuervo, A.M., 2004. Autophagy: many paths to the same end. Mol. Cell Biochem. 263, 55–72. Daugaard, M., et al., 2007. The heat shock protein 70 family: highly homologous proteins with overlapping and distinct functions. FEBS Lett. 19, 3702–3710. de Duve, C., et al., 1955. Tissue fractionation studies. 6. Intracellular distribution patterns of enzymes in rat-liver tissue. Biochem. J. 60, 604–617. de Duve, C., 1983. Lysosomes revisited. Eur. J. Biochem. 137, 391–397. Dell’Angelica, E.C., et al., 2000. Lysosome-related organelles. FASEB J. 14, 1265– 1278. Doulias, P.T., et al., 2007. Involvement of heat shock protein-70 in the mechanism of hydrogen peroxide-induced DNA damage: the role of lysosomes and iron. Free Radic. Biol. Med. 42, 567–577. Doyle, K.P., et al., 2008. Mechanisms of ischemic brain damage. Neuropharma 55, 310–318. Edinger, A.L., Thompson, C.B., 2004. Death by design: apoptosis, necrosis and autophagy. Curr. Opin. Cell Biol. 16, 663–669. Eskelinen, E.L., 2006. Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy. Mol. Aspects Med. 27, 495–502. Esterbauer, H., et al., 1991. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11, 81–128. Figueiredo, C., et al., 2008. Neuron-microglia crosstalk up-regulates neuronal FGF-2 expression which mediates neuroprotection against excitotoxicity via JNK1/2. J. Neurochem. 107, 73–85.

Flynn, R.W.V., et al., 2008. The cost of cerebral ischaemia. Neuropharmacology 55, 250–256. Foghsgaard, et al., 2001. Cathepsin B acts as a dominant execution protease in tumor cell apoptosis induced by tumor necrosis factor. J. Cell Biol. 153, 999– 1009. Gerner, E.W., et al., 1976. A transient thermotolerant survival response produced by single thermal doses in HeLa cells. Cancer Res. 36, 1035–1040. Goll, D.E., et al., 2003. The calpain system. Physiol. Rev. 83, 731–801. Golstein, P., Kroemer, G., 2007. Cell death by necrosis: towards a molecular definition. Trends Biochem. Sci. 32, 37–43. Grune, T., et al., 2004. Decreased proteolysis caused by protein aggregates, inclusion bodies, plaques, lipofuscin, ceroid, and ‘aggresomes’ during oxidative stress, aging, and disease. Int. J. Biochem. Cell Biol. 36, 2519–2530. Guicciardi, M.E., et al., 2000. Cathepsin B contributes to TNF-alpha-mediated hepatocyte apoptosis by promoting mitochondrial release of cytochrome c. J. Clin. Invest. 106, 1127–1137. Guicciardi, M.E., et al., 2004. Lysosomes in cell death. Oncogene 23, 2881–2890. Gupta, R.S., Singh, B., 1994. Phylogenetic analysis of 70 kDa heat shock protein sequences suggests a chimeric origin for the eukaryotic cell nucleus. Curr. Biol. 4, 1104–1114. Guroff, G., 1964. A neutral, calcium-activated proteinase from the soluble fraction of rat brain. J. Biol. Chem. 239, 149–155. Gyrd-Hansen, M., et al., 2004. Heat shock protein 70 promotes cancer cell viability by safeguarding lysosomal integrity. Cell Cycle 3, 1484–1485. Halliwell, B., 2001. Role of free radicals in the neurodegenerative diseases: therapeutic implications for antioxidant treatment. Drugs Aging 18, 685–716. Hara, T., et al., 2006. Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 441, 885–889. Hartl, F.U., 1996. Molecular chaperones in cellular protein folding. Nature 381, 571– 579. Henle, K.J., et al., 1978. Induction of thermotolerance in Chinese hamster ovary cells by high (45 degrees) or low (40 degrees) hyperthermia. Cancer Res. 38, 570– 574. Hunt, C., Morimoto, R.I., 1985. Conserved features of eucaryotic hsp70 genes releaved by comparision with the nucleotide sequence of human hsp70. Proc. Natl. Acad. Sci. U.S.A. 82, 6455–6459. Hwang, J.J., et al., 2008. Zinc and 4-hydroxy-2-nonenal mediate lysosomal membrane permeabilization induced by H2O2 in cultured hippocampal neurons. J. Neurosci. 28, 3114–3122. Ischiropoulos, H., Beckman, J.S., 2003. Oxidative stress and nitration in neurodegeneration: cause, effect, or association? J. Clin. Invest. 111, 163–169. Ishisaka, R., et al., 1998. Activation of caspase-3-like protease by digitonin-treated lysosomes. FEBS Lett. 435, 233–236. Ja¨a¨ttela¨, M., 1999. Heat shock proteins as cellular lifeguards. Ann. Med. 31, 261–271. Jaeger, P.A., Wyss-Coray, T., 2009. All-you-can-eat: autophagy in neurodegeneration and neuroprotection. Mol. Neurodegener. 4 (April (1)), 16 [Epub ahead of print]. Jenner, P., 1998. Oxidative mechanisms in nigral cell death in Parkinson’s disease. Mov. Disord. 13, 24–34. Johansson, A.C., et al., 2003. Cathepsin D mediates cytochrome c release and caspase activation in human fibroblast apoptosis induced by staurosporine. Cell Death Differ. 10, 1253–1259. Johnston, R.N., Kucey, B.L., 1988. Competitive inhibition of hsp70 gene expression causes thermosensitivity. Science 242, 1551–1554. Ka˚gedal, K., et al., 2001. The lysosomal protease cathepsin D mediates apoptosis induced by oxidative stress. FASEB J. 15, 1592–1594. Kerr, J.F., 1972. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26, 239–257. Kessel, D., Poretz, R.D., 2000. Sites of photodamage induced by photodynamic therapy with a chlorin e6 triacetoxymethyl ester (CAME). Photochem. Photobiol. 71, 9496. Klionsky, D.J., 2008. Autophagy revisited: a conversation with Christian de Duve. Autophagy 4, 740–743. Komatsu, M., et al., 2006. Loss of autophagy in the central nervous system causes neurodegeneration in mice. Nature 441, 880–884. Kourtis, N., Tavernarakis, N., 2009. Autophagy and cell death in model organisms. Cell Death Differ. 16, 21–30. Kroemer, G., Ja¨a¨ttela¨, M., 2005. Lysosomes and autophagy in cell death control. Nat. Rev. Cancer 5, 886–897. Kurz, T., et al., 2006. Intralysosomal iron chelation protects against oxidative stressinduced cellular damage. FEBS J. 273, 3106–3117. Kurz, T., et al., 2008. Lysosomes and oxidative stress in aging and apoptosis. Biochim. Biophys. Acta 1780, 1291–1303. Lanneau, D., et al., 2008. Heat shock proteins: essential proteins for apoptosis regulation. J. Cell Mol. Med. 12, 743–761. Landry, J., et al., 1982. Synthesis and degradation of heat shock proteins during development and decay of thermotolerance. Cancer Res. 42, 2457–2461. Lee, S.H., et al., 2001. Targeted hsp70.1 disruption increases infarction volume after focal cerebral ischemia in mice. Stroke 32, 2905–2912. Leist, M., Ja¨a¨ttela¨, M., 2001. Triggering of apoptosis by cathepsins. Cell Death Differ. 8, 324–326. Levine, R.L., 2002. Carbonyl modified proteins in cellular regulation, aging, and disease. Free Radic. Biol. Med. 32, 790–796. Li, G.C., Werb, Z., 1982. Correlation between synthesis of heat shock proteins and development of thermotolerance in Chinese hamster fibroblasts. Proc. Natl. Acad. Sci. U.S.A. 79, 3122–3218.

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358 Li, W., et al., 2000. Induction of cell death by the lysosomotropic detergent MSDH. FEBS Lett. 470, 3539. Lindquist, S., Craig, E.A., 1988. The heat shock proteins. Ann. Rev. Genet. 22, 631– 677. Liu, Y., et al., 1993. Inhibition of postcardiac arrest brain protein oxidation by acetyl-L-carnitine. Free Radic. Biol. Med. 15, 667–670. Mambula, S.S., Calderwood, S.K., 2006. Heat shock protein 70 is secreted from tumor cells by a nonclassical pathway involving lysosomal endosomes. J. Immunol. 177, 7849–7857. Manzerra, et al., 1998. Tissue-specific differences in heat shock protein hsc70 and hsp70 in the control and hyperthermic rabbit. J. Cell. Physiol. 170, 130–137. Marques, C., et al., 2004. Ubiquitin-dependent lysosomal degradation of the HNEmodified proteins in lens epithelial cells. FASEB J. 18, 1424–1426. Monnerie, H., Le Roux, P.D., 2008. Glutamate alteration of glutamic acid decarboxylase (GAD) in GABAergic neurons: the role of cysteine proteases. Exp. Neurol. 213, 145–153. Mosser, D.D., et al., 2000. The chaperone function of hsp70 is required for protection against stress-induced apoptosis. Mol. Cell. Biol. 20, 7146–7159. Mu¨ller, G.J., et al., 2007. Anti-apoptotic signaling and failure of apoptosis in the ischemic rat hippocampus. Neurobiol. Dis. 25, 582–593. Nakagawa, T., Yuan, J., 2000. Cross-talk between two cysteine protease families. Activation of caspase-12 by calpain in apoptosis. J. Cell Biol. 150, 887– 894. Nakajima, E., et al., 2006. Calpain-specific proteolysis in primate retina: contribution of calpains in cell death. Invest. Ophthalmol. Vis. Sci. 47, 5469–5475. Nedelsky, N.B., et al., 2008. Autophagy and the ubiquitin–proteasome system: collaborators in neuroprotection. Biochim. Biophys. Acta 1782, 691–699. Neumar, R.W., et al., 2003. Cross-talk between calpain and caspase proteolytic systems during neuronal apoptosis. J. Biol. Chem. 278, 14162–14167. Nilsson, C., et al., 2006. Cytosolic acidification and lysosomal alkalinization during TNF-alpha induced apoptosis in U937 cells. Apoptosis 11, 1149–1159. Nishi, T., Forgac, M., 2002. The vacuolar (H+)-ATPases—nature’s most versatile proton pumps. Nat. Rev. Mol. Cell Biol. 3, 94–103. Nixon, R.A., et al., 2000. The endosomal-lysosomal system of neurons in Alzheimer’s disease pathogenesis: a review. Neurochem. Res. 25, 1161–1172. Nollen, E.A., et al., 1999. In vivo chaperone activity of heat shock protein 70 and thermotolerance. Mol. Cell Biol. 19, 2069–2079. Nylandsted, J., et al., 2000. Selective depletion of heat shock protein 70 (Hsp70) activates a tumor-specific-death program that is indepenent of caspases and bypass Bcl-2. Proc. Natl. Acad. Sci. U.S.A. 97, 7871–7876. Nylandsted, J., et al., 2004. Heat shock protein 70 promotes cell survival by inhibiting lysosomal membrane permeabilization. J. Exp. Med. 200, 425–435. Nystro¨m, T., 2005. Role of oxidative carbonylation in protein quality control and senescence. EMBO J. 24, 1311–1317. Oikawa, S., et al., 2009. Proteomic identification of carbonylated proteins in the monkey CA1 after ischemia–reperfusion. Free Radic. Bio. Med. 46, 1472– 7477. Olejnicka, B.T., et al., 1999. Minute oxidative stress is sufficient to induce apoptotic death of NIT-1 insulinoma cells. APMIS 107, 747–761. ¨ llinger, K., Brunk, U.T., 1995. Cellular injury induced by oxidative stress is O mediated through lysosomal damage. Free Radic. Biol. Med. 19, 565–574. ¨ llinger, K., 2000. Inhibition of cathepsin D prevents free-radical-induced apoptosis O in rat cardiomyocytes. Arch. Biochem. Biophys. 373, 346–351. Oliver, C.N., et al., 1990. Oxidative damage to brain proteins, loss of glutamine synthetase activity, and production of free radicals during ischemia/reperfusion-induced injury to gerbil brain. Proc. Natl. Acad. Sci. U.S.A. 87, 5144– 5147. Olsson, G.M., et al., 1989. Lysosomal enzyme leakage during the hypoxanthine/ xanthine oxidase reaction. Virchows Arch. B Cell Pathol. Incl. Mol. Pathol. 56, 385–391. Ovbiagele, B., et al., 2003. Potential role of neuroprotective agents in the treatment of patients with acute ischemic stroke. Curr. Treat. Opt. Cardiovasc. Med. 5, 441–449. Owen, A.D., et al., 1996. Oxidative stress and Parkinson’s disease. Ann. N.Y. Acad. Sci. 786, 217–223. Pelham, H.R.B., 1984. Hsp70 accelerates the recovery of nucleolar morphology after heat shock. EMBO J. 3, 3095–3100. Petersen, N.S., Mitchell, H.K., 1981. Recovery of protein synthesis after heat shock: prior heat treatment affects the ability of cells to translate mRNA. Proc. Natl. Acad. Sci. U.S.A. 78, 1708–1711. Pignol, B., et al., 2006. Calpain inhibitors and antioxidants act synergistically to prevent cell necrosis: effects of the novel dual inhibitors (cysteine protease inhibitor and antioxidant) BN 82204 and its pro-drug BN 82270. J. Neurochem. 98, 1217–1228. Pivtoraiko, V.N., et al., 2009. Oxidative stress and autophagy in the regulation of lysosome-dependent neuron death. Antioxid. Redox Signal. 11, 481–496. Rajdev, S., et al., 2000. Mice overexpressing rat heat shock protein 70 are protected against cerebral infarction. Ann. Neurol. 47, 782–791. Rami, A., et al., 2000. m-Calpain activation, DNA-fragmentation, and synergistic effects of caspase- and calpain-inhibitors in protecting hippocampal neurons from ischemic damage. Brain Res. 866, 299–312. Rami, A., et al., 2008. Exploiting endogenous anti-apoptotic proteins for novel therapeutic strategies in cerebral ischemia. Prog. Neurobiol. 85, 273–296. Ravagnan, L., et al., 2001. Heat-shock protein 70 antagonizes apoptosis-inducing factor. Nat. Cell Biol. 3, 839–843.

357

Reimertz, C., et al., 2001. Ca2+-induced inhibition of apoptosis in human SH-SY5Y neuroblastoma cells: degradation of apoptotic protease activating factor-1 (APAF-1). J. Neurochem. 78, 1256–1266. Riabolow, K.T., et al., 1988. Heat shock is lethal to fibroblasts microinjected with antibodies against hsp70. Science 242, 433–436. Ritossa, F., 1962. A new puffing pattern induced by temperature shock and DNP in Drosophila. Experimentia 18, 571–573. ¨ llinger, K., 1998. Oxidative stress causes relocation of the lysosomal Roberg, K., O enzyme cathepsin D with ensuing apoptosis in neonatal rat cardiomyocytes. Am. J. Path. 152, 1151–1156. Salvador, N., et al., 2000. Import of a cytosolic protein into lysosomes by chaperonemediated autophagy depends on its folding state. J. Biol. Chem. 275, 27447– 27456. Sayre, L.M., et al., 2006. Protein adducts generated from products of lipid oxidation: focus on HNE and one. Drug Metab. Rev. 38, 651–675. Sapareto, S.A., et al., 1978. Effects of hyperthermia on survival and progression of Chinese hamster ovary cells. Cancer Res. 38, 393–400. Scarlatti, F., et al., 2009. Does autophagy have a license to kill mammalian cells? Cell Death Differ. 16, 12–20. Schmitt, E., et al., 2007. Intracellular and extracellular functions of heat shock proteins: repercussions in cancer therapy. J. Leukoc. 81, 15–27. Shintani, T., Klionsky, D.J., 2004. Autophagy in health and disease: a double-edged sword. Science 306, 990–995. Simonian, N.A., Coyle, J.T., 1996. Oxidative stress in neurodegenerative diseases. Annu. Rev. Pharmacol. Toxicol. 36, 83–106. Solenski, N.J., et al., 1997. Differential hydroxylation of salicylate in core and penumbra regions during focal reversible cerebral ischemia. Stroke 28, 2545– 2551. Solenski, N.J., et al., 2000. Cerebral ischemia–reperfusion injury: a novel therapeutic approach with TAK-218. Clin. Neuropharmacol. 23, 69–74. Stadtman, E.R., Berlett, B.S., 1997. Reactive oxygen-mediated protein oxidation in aging and disease. Chem. Res. Toxicol. 10, 485–494. Stankiewicz, A.R., et al., 2005. Hsp70 inhibits heat-induced apoptosis upstream of mitochondria by preventing Bax translocation. J. Biol. Chem. 280, 38729– 38739. Suzuki, K., et al., 2004. Structure, activation, and biology of calpain. Diabetes 53, 12– 18. Syntichaki, P., et al., 2002. Specific aspartyl and calpain proteases are required for neurodegeneration in C. elegans. Nature 419, 939–944. Syntichaki, P., et al., 2005. The vacuolar H+-ATPase mediates intracellular acidification required for neurodegeneration in C. elegans. Curr. Biol. 15, 1249–1254. Tandara, A.A., et al., 2006. Age effect on HSP70: decreased resistance to ischemic and oxidative stress in HDF. J. Surg. Res. 132, 32–39. Taniguchi, S., et al., 2001. Calpain-mediated degradation of p35 to p25 in postmortem human and rat brains. FEBS Lett. 489, 46–50. Tardy, C., et al., 2006. Lysosomes and lysosomal proteins in cancer cell death (new players of an old struggle). Biochim. Biophys. Acta 1765, 101–125. Tavaria, M., et al., 1996. A hitchhickers’ guide to human Hsp70 family. Cell Stress Chaperon. 1, 23–28. Terman, A., et al., 2006a. Lysosomal labilization. IUBMB Life 58, 531–539. Terman, A., et al., 2006b. The lysosomal–mitochondrial axis theory of postmitotic aging and cell death. Chem. Biol. Interact. 163, 29–37. Trevisani, M., et al., 2007. 4-Hydroxynonenal, an endogenous aldehyde, causes pain and neurogenic inflammation through activation of the irritant receptor TRPA1. Proc. Natl. Acad. Sci. U.S.A. 104, 13519–13524. Trushina, E., McMurray, C.T., 2007. Oxidative stress and mitochondrial dysfunction in neurodegenerative diseases. Neuroscience 145, 1233–1248. Tsukada, T., et al., 2001. Implications of CAD and DNase II in ischemic neuronal necrosis specific for the primate hippocampus. J. Neurochem. 79, 1196–1206. Vancompernolle, K., et al., 1998. Atractyloside-induced release of cathepsin B, a protease with caspase-processing activity. FEBS Lett. 438, 150–158. Vass, K., et al., 1988. Localization of 70-kDa stress protein induction in gerbil brain after ischemia. Acta Neuropathol. 77, 128–135. Wang, K.K., et al., 1998. Caspase-mediated fragmentation of calpain inhibitor protein calpastatin during apoptosis. Arch. Biochem. Biophys. 356, 187–196. Werneburg, N.W., et al., 2002. Tumor necrosis factor-alpha-associated lysosomal permeabilization is cathepsin B dependent. Am. J. Physiol. Gastrointest. Liver Physiol. 283, G947. Windelborn, J.A., Lipton, P., 2008. Lysosomal release of cathepsins causes ischemic damage in the rat hippocampal slice and depends on NMDA-mediated calcium influx, arachidonic acid metabolism, and free radical production. J. Neurochem. 106, 56–69. Xiong, Y., et al., 2007. Degradation of oxidized proteins by autophagy during oxidative stress in Arabidopsis. Plant Physiol. 143, 291–299. Yakovlev, A.G., Faden, A.I., 2004. Mechanisms of neural cell death: implications for development of neuroprotective treatment strategies. NeuroRX 1, 5–16. Yamashima, T., et al., 1996. Transient brain ischaemia provokes Ca2+, PIP2 and calpain responses prior to delayed neuronal death in monkeys. Eur. J. Neurosci. 8, 1932–1944. Yamashima, T., et al., 1998. Inhibition of ischaemic hippocampal neuronal death in primates with cathepsin B inhibitor CA-074: a novel strategy for neuroprotection based on ‘calpain–cathepsin hypothesis’. Eur. J. Neurosci. 10, 1723–1733. Yamashima, T., 2000. Implication of cysteine proteases calpain, cathepsin and caspase in ischemic neuronal death of primates. Prog. Neurobiol. 62, 273– 295.

358

T. Yamashima, S. Oikawa / Progress in Neurobiology 89 (2009) 343–358

Yamashima, T., et al., 2003. Sustained calpain activation associated with lysosomal rupture executes necrosis of the postischemic CA1 neurons in primates. Hippocampus 13, 791–800. Yamashima, T., 2004. Ca2+-dependent proteases in ischemic neuronal death: a conserved ‘calpain–cathepsin cascade’ from nematodes to primates. Cell Calcium 36, 285–293. Yang, T., et al., 2008. Heat shock protein 70 expression in epilepsy suggests stress rather than protection. Acta Neuropathol. 115, 219–230.

Yoritaka, Y., et al., 1996. Immunochemical detection of 4-hydroxynonenal protein adducts in Parkinson disease. Proc. Natl. Acad. Sci. U.S.A. 93, 2696– 2701. Yuan, X.M., et al., 2002. Lysosomal destabilization in p53-induced apoptosis. Proc. Natl. Acad. Sci. U.S.A. 99, 6286–6291. Yuan, J., et al., 2003. Diversity in the mechanisms of neuronal cell death. Neuron 40, 401–413. Zatz, M., Starling, A., 2005. Calpains and disease. New Engl. J. Med. 352, 2413–2423.