Autosis is a Na+,K+-ATPase-regulated form of cell death triggered by autophagy-inducing peptides, starvation, and hypoxia-ischemia.

A long-standing controversy is whether autophagy is a bona fide cause of mammalian cell death. We used a cell-penetrating autophagy-inducing peptide, Tat-Beclin 1, derived from the autophagy protein Beclin 1, to investigate whether high levels of autophagy result in cell death by autophagy. Here we show that Tat-Beclin 1 induces dose-dependent death that is blocked by pharmacological or genetic inhibition of autophagy, but not of apoptosis or necroptosis. This death, termed "autosis," has unique morphological features, including increased autophagosomes/autolysosomes and nuclear convolution at early stages, and focal swelling of the perinuclear space at late stages. We also observed autotic death in cells during stress conditions, including in a subpopulation of nutrient-starved cells in vitro and in hippocampal neurons of neonatal rats subjected to cerebral hypoxia-ischemia in vivo. A chemical screen of ~5,000 known bioactive compounds revealed that cardiac glycosides, antagonists of Na(+),K(+)-ATPase, inhibit autotic cell death in vitro and in vivo. Furthermore, genetic knockdown of the Na(+),K(+)-ATPase α1 subunit blocks peptide and starvation-induced autosis in vitro. Thus, we have identified a unique form of autophagy-dependent cell death, a Food and Drug Administration-approved class of compounds that inhibit such death, and a crucial role for Na(+),K(+)-ATPase in its regulation. These findings have implications for understanding how cells die during certain stress conditions and how such cell death might be prevented.

Dying neurons in thalamus of asphyxiated term newborns and rats are autophagic.

OBJECTIVE: Neonatal hypoxic-ischemic encephalopathy (HIE) still carries a high burden by its mortality and long-term neurological morbidity in survivors. Apart from hypothermia, there is no acknowledged therapy for HIE, reflecting the lack of mechanistic understanding of its pathophysiology. (Macro)autophagy, a physiological intracellular process of lysosomal degradation, has been proposed to be excessively activated in excitotoxic conditions such as HIE. The present study examines whether neuronal autophagy in the thalamus of asphyxiated human newborns or P7 rats is enhanced and related to neuronal death processes. METHODS: Neuronal autophagy and cell death were evaluated in the thalamus (frequently injured in severe HIE) of both human newborns who died after severe HIE (n = 5) and P7 hypoxic-ischemic rats (Rice-Vannuci model). Autophagic (LC3, p62), lysosomal (LAMP1, cathepsins), and cell death (TUNEL, caspase-3) markers were studied by immunohistochemistry in human and rat brain sections, and by additional methods in rats (immunoblotting, histochemistry, and electron microscopy). RESULTS: Following severe perinatal asphyxia in both humans and rats, thalamic neurons displayed up to 10-fold (p < 0.001) higher numbers of autophagosomes and lysosomes, implying an enhanced autophagic flux. The highly autophagic neurons presented strong features of apoptosis. These findings were confirmed and elucidated in more detail in rats. INTERPRETATION: These results show for the first time that autophagy is enhanced in severe HIE in dying thalamic neurons of human newborns, as in rats. Experimental neuroprotective strategies targeting autophagy could thus be a promising lead to follow for the development of future therapeutic approaches.

Neuroscience, quantum indeterminism and the Cartesian soul.

Quantum indeterminism is frequently invoked as a solution to the problem of how a disembodied soul might interact with the brain (as Descartes proposed), and is sometimes invoked in theories of libertarian free will even when they do not involve dualistic assumptions. Taking as example the Eccles-Beck model of interaction between self (or soul) and brain at the level of synaptic exocytosis, I here evaluate the plausibility of these approaches. I conclude that Heisenbergian uncertainty is too small to affect synaptic function, and that amplification by chaos or by other means does not provide a solution to this problem. Furthermore, even if Heisenbergian effects did modify brain functioning, the changes would be swamped by those due to thermal noise. Cells and neural circuits have powerful noise-resistance mechanisms, that are adequate protection against thermal noise and must therefore be more than sufficient to buffer against Heisenbergian effects. Other forms of quantum indeterminism must be considered, because these can be much greater than Heisenbergian uncertainty, but these have not so far been shown to play a role in the brain.

The limits of brain determinacy.

The genes do not control everything that happens in a cell or an organism, because thermally induced molecular movements and conformation changes are beyond genetic control. The importance of uncontrolled events has been argued from the differences between isogenic organisms reared in virtually identical environments, but these might alternatively be attributed to subtle, undetected differences in the environment. The present review focuses on the uncontrolled events themselves in the context of the developing brain. These are considered at cellular and circuit levels because even if cellular physiology was perfectly controlled by the genes (which it is not), the interactions between different cells might still be uncoordinated. A further complication is that the brain contains mechanisms that buffer noise and others that amplify it. The final resultant of the battle between these contrary mechanisms is that developmental stochasticity is sufficiently low to make neurobehavioural defects uncommon, but a chance component of neural development remains. Thus, our brains and behaviour are not entirely determined by a combination of genes-plus-environment.

Excitotoxicity-induced endocytosis mediates neuroprotection by TAT-peptide-linked JNK inhibitor.

Excitotoxicity and cerebral ischemia induce strong endocytosis in neurons, and we here investigate its functional role in neuroprotection by a functional transactivator of transcription (TAT)-peptide, the c-Jun N-terminal kinase (JNK) inhibitor D-JNKI1, against NMDA-excitotoxicity in vitro and neonatal ischemic stroke in P12 Sprague-Dawley rats. In both situations, the neuroprotective efficacy of D-JNKI1 was confirmed, but excessively high doses were counterproductive. Importantly, the induced endocytosis was necessary for neuroprotection, which required that the TAT-peptide be administered at a time when induced endocytosis was occurring. Uptake by other routes failed to protect, and even promoted cell death at high doses. Blocking the induced endocytosis of D-JNKI1 with heparin or with an excess of D-TAT-peptide eliminated the neuroprotection. We conclude that excitotoxicity-induced endocytosis is a basic property of stressed neurons that can target neuroprotective TAT-peptides into the neurons that need protection. Furthermore, it is the main mediator of neuroprotection by D-JNKI1. This may explain promising reports of strong neuroprotection by TAT-peptides without apparent side effects, and warns that the timing of peptide administration is crucial.

A peptide inhibitor of c-Jun N-terminal kinase protects against excitotoxicity and cerebral ischemia.

Neuronal death in cerebral ischemia is largely due to excitotoxic mechanisms, which are known to activate the c-Jun N-terminal kinase (JNK) pathway. We have evaluated the neuroprotective power of a cell-penetrating, protease-resistant peptide that blocks the access of JNK to many of its targets. We obtained strong protection in two models of middle cerebral artery occlusion (MCAO): transient occlusion in adult mice and permanent occlusion in 14-d-old rat pups. In the first model, intraventricular administration as late as 6 h after occlusion reduced the lesion volume by more than 90% for at least 14 d and prevented behavioral consequences. In the second model, systemic delivery reduced the lesion by 78% and 49% at 6 and 12 h after ischemia, respectively. Protection correlated with prevention of an increase in c-Jun activation and c-Fos transcription. In view of its potency and long therapeutic window, this protease-resistant peptide is a promising neuroprotective agent for stroke.

Role of the c-Jun N-terminal kinase pathway in retinal excitotoxicity, and neuroprotection by its inhibition.

Retinal excitotoxicity is associated with retinal ischemia, and with glaucomatous and traumatic optic neuropathy. The present study investigates the role of c-Jun N-terminal kinase (JNK) activation in NMDA-mediated retinal excitotoxicity and determines whether neuroprotection can be obtained with the JNK pathway inhibitor, D-form of JNK-inhibitor 1 (D-JNKI-1). Young adult rats received intravitreal injections of 20 nmol NMDA, which caused extensive neuronal death in the inner nuclear and ganglion cell layers. This excitotoxicity was associated with strong activation of calpain, as revealed by fodrin cleavage, and of JNK. The cell-permeable peptide D-JNKI-1 was used to inhibit JNK. Within 40 min of its intravitreal injection, FITC-labeled D-JNKI-1 spread through the retinal ganglion cell layer into the inner nuclear layer and interfered with the NMDA-induced phosphorylation of JNK. Injections of unlabeled D-JNKI-1 gave unprecedentedly strong neuroprotection against cell death in both layers, lasting for at least 10 days. The NMDA-induced calpain-specific fodrin cleavage was likewise strongly inhibited by D-JNKI-1. Moreover the electroretinogram was partially preserved by D-JNKI-1. Thus, the JNK pathway is involved in NMDA-mediated retinal excitotoxicity and JNK inhibition by D-JNKI-1 provides strong neuroprotection as shown morphologically, biochemically and physiologically.

Neuroprotection for optic nerve disorders.

PURPOSE OF REVIEW: The concept that optic nerve fiber loss might be reduced by neuroprotection arose in the mid 1990s. The subsequent research effort, focused mainly on rodent models, has not yet transformed into a successful clinical trial, but provides mechanistic understanding of retinal ganglion cell death and points to potential therapeutic strategies. This review highlights advances made over the last year. RECENT FINDINGS: In excitotoxicity and axotomy models retinal ganglion cell death has been shown to result from a complex interaction between retinal neurons and Müller glia, which release toxic molecules including tumor necrosis factor alpha. This counteracts neuroprotection by neurotrophins such as nerve growth factor, which bind to p75NTR receptors on Müller glia stimulating the toxic release. Another negative effect against neurotrophin-mediated protection involves the action of LINGO-1 at trkB brain-derived neurotrophic factor (BDNF) receptors, and BDNF neuroprotection is enhanced by an antagonist to LINGO-1. As an alternative to pharmacotherapy, retinal defences can be stimulated by exposure to infrared radiation. SUMMARY: The mechanisms involved in glaucoma and other optic nerve disorders are being clarified in rodent models, focusing on retrograde degeneration following axonal damage, excitotoxicity and inflammatory/autoimmune mechanisms. Neuroprotective strategies are being refined in the light of the mechanistic understanding.

Enhancement of autophagic flux after neonatal cerebral hypoxia-ischemia and its region-specific relationship to apoptotic mechanisms.

The multiplicity of cell death mechanisms induced by neonatal hypoxia-ischemia makes neuroprotective treatment against neonatal asphyxia more difficult to achieve. Whereas the roles of apoptosis and necrosis in such conditions have been studied intensively, the implication of autophagic cell death has only recently been considered. Here, we used the most clinically relevant rodent model of perinatal asphyxia to investigate the involvement of autophagy in hypoxic-ischemic brain injury. Seven-day-old rats underwent permanent ligation of the right common carotid artery, followed by 2 hours of hypoxia. This condition not only increased autophagosomal abundance (increase in microtubule-associated protein 1 light chain 3-11 level and punctuate labeling) but also lysosomal activities (cathepsin D, acid phosphatase, and beta-N-acetylhexosaminidase) in cortical and hippocampal CA3-damaged neurons at 6 and 24 hours, demonstrating an increase in the autophagic flux. In the cortex, this enhanced autophagy may be related to apoptosis since some neurons presenting a high level of autophagy also expressed apoptotic features, including cleaved caspase-3. On the other hand, enhanced autophagy in CA3 was associated with a more purely autophagic cell death phenotype. In striking contrast to CA3 neurons, those in CA1 presented only a minimal increase in autophagy but strong apoptotic characteristics. These results suggest a role of enhanced autophagy in delayed neuronal death after severe hypoxia-ischemia that is differentially linked to apoptosis according to the cerebral region.

Excitotoxicity-induced endocytosis confers drug targeting in cerebral ischemia.

OBJECTIVE: Targeting neuroprotectants specifically to the cells that need them is a major goal in biomedical research. Many peptidic protectants contain an active sequence linked to a carrier such as the transactivator of transcription (TAT) transduction sequence, and here we test the hypothesis that TAT-linked peptides are selectively endocytosed into neurons stressed by excitotoxicity and focal cerebral ischemia. METHODS: In vivo experiments involved intracerebroventricular injection of TAT peptides or conventional tracers (peroxidase, fluorescein isothiocyanate-dextran) in young rats exposed to occlusion of the middle cerebral artery at postnatal day 12. Cellular mechanisms of uptake were analyzed in dissociated cortical neuronal cultures. RESULTS: In both models, all tracers were taken up selectively into stressed neurons by endocytosis. In the in vivo model, this was neuron specific and limited to the ischemic area, where the neurons displayed enhanced immunolabeling for early endosomal antigen-1 and clathrin. The highly efficient uptake of TAT peptides occurred by the same selective mechanism as for conventional tracers. All tracers were targeted to the nucleus and cytoplasm of neurons that appeared viable, although ultimately destined to die. In dissociated cortical neuronal cultures, an excitotoxic dose of N-methyl-D-aspartate induced a similar endocytosis. It was 100 times more efficient with TAT peptides than with dextran, because the former bound to heparan sulfate proteoglycans at the cell surface, but it depended on dynamin and clathrin in both cases. INTERPRETATION: Excitotoxicity-induced endocytosis is the main entry route for protective TAT peptides and targets selectively the neurons that need to be protected.

Postischemic treatment of neonatal cerebral ischemia should target autophagy.

OBJECTIVE: To evaluate the contributions of autophagic, necrotic, and apoptotic cell death mechanisms after neonatal cerebral ischemia and hence define the most appropriate neuroprotective approach for postischemic therapy. METHODS: Rats were exposed to transient focal cerebral ischemia on postnatal day 12. Some rats were treated by postischemic administration of pan-caspase or autophagy inhibitors. The ischemic brain tissue was studied histologically, biochemically, and ultrastructurally for autophagic, apoptotic, and necrotic markers. RESULTS: Lysosomal and autophagic activities were increased in neurons in the ischemic area from 6 to 24 hours postinjury, as shown by immunohistochemistry against lysosomal-associated membrane protein 1 and cathepsin D, by acid phosphatase histochemistry, by increased expression of autophagosome-specific LC3-II and by punctate LC3 staining. Electron microscopy confirmed the presence of large autolysosomes and putative autophagosomes in neurons. The increases in lysosomal activity and autophagosome formation together demonstrate increased autophagy, which occurred mainly in the border of the lesion, suggesting its involvement in delayed cell death. We also provide evidence for necrosis near the center of the lesion and apoptotic-like cell death in its border, but in nonautophagic cells. Postischemic intracerebroventricular injections of autophagy inhibitor 3-methyladenine strongly reduced the lesion volume (by 46%) even when given >4 hours after the beginning of the ischemia, whereas pan-caspase inhibitors, carbobenzoxy-valyl-alanyl-aspartyl(OMe)-fluoromethylketone and quinoline-val-asp(OMe)-Ch2-O-phenoxy, provided no protection. INTERPRETATION: The prominence of autophagic neuronal death in the ischemic penumbra and the neuroprotective efficacy of postischemic autophagy inhibition indicate that autophagy should be a primary target in the treatment of neonatal cerebral ischemia.

Limited role of the c-Jun N-terminal kinase pathway in a neonatal rat model of cerebral hypoxia-ischemia.

D-JNKI1, a cell-permeable peptide inhibitor of the c-Jun N-terminal kinase (JNK) pathway, has been shown to be a powerful neuroprotective agent after focal cerebral ischemia in adult mice and young rats. We have investigated the potential neuroprotective effect of D-JNKI1 and the involvement of the JNK pathway in a neonatal rat model of cerebral hypoxia-ischemia (HI). Seven-day-old rats underwent a permanent ligation of the right common carotid artery followed by 2 h of hypoxia (8% oxygen). Treatment with D-JNKI1 (0.3 mg/kg intraperitoneally) significantly reduced early calpain activation, late caspase 3 activation and, in the thalamus, autophagosome formation, indicating an involvement of JNK in different types of cell death: necrotic, apoptotic, and autophagic. However, the size of the lesion was unchanged. Further analysis showed that neonatal HI induced an immediate decrease in JNK phosphorylation (reflecting mainly JNK1 phosphorylation) followed by a slow progressive increase (including JNK3 phosphorylation 54 kDa), whereas c-jun and c-fos expression were both strongly activated immediately after HI. In conclusion, unlike in adult ischemic models, JNK is only moderately activated after severe cerebral HI in neonatal rats and the observed positive effects of D-JNKI1 are insufficient to give neuroprotection. Thus, for perinatal asphyxia, D-JNKI1 can only be considered in association with other therapies.

[The two faces of autophagy in the nervous system]. / Les deux visages de l'autophagie dans le système nerveux.

Autophagy is a cellular mechanism for degrading proteins and organelles. It was first described as a physiological process essential for maintaining homeostasis and cell survival, but understanding its role in conditions of stress has been complicated by the recognition of a new type of cell death ("type 2") characterized by deleterious autophagic activity. This paradox is important in the central nervous system where the activation of autophagy seems to be protective in certain neurodegenerative diseases but deleterious in cerebral ischemia. The development of new therapeutic strategies based on the manipulation of autophagy will need to take into account these opposing roles of autophagy.

Neuronal death in the lateral geniculate nucleus of young ferrets following a cortical lesion: time-course, age dependence and involvement of caspases.

In humans and many other mammalian species, the behavioural consequences of a cortical lesion tend to be milder when it occurs early in life, and there is evidence that an important factor contributing to the behavioural sparing in the young is the formation of new thalamo-cortical connections by thalamic neurons initially connected with the lesioned area. However, this plasticity may be hindered by the secondary death of many of these neurons owing to the elimination by the primary lesion of their trophic support from the cortex. With the long-term aim of preventing this neuronal death, we have here characterised its timing in the lateral geniculate nucleus of ferrets following lesions of the visual cortex on postnatal days 5, 10, 20 or 35. After the earliest lesions (P5 or P10), this cell death began rapidly and occurred synchronously, being maximal at 48 h and declining to zero over the next few days. Following later lesions the cell death began more slowly and continued for longer. The dying neurons contained activated caspase-3 and fragmented DNA and their number 2 days after a P5 lesion was reduced by the broad-band caspase inhibitor z-VAD.fmk. These experiments open the way for a concerted effort to enhance adaptive plasticity by neuroprotection in the hours or days following a cortical lesion.

Soman-induced convulsions: the neuropathology revisited.

The organophosphorus compound soman, an irreversible inhibitor of cholinesterases, produces seizure activity and related brain damage. Studies using various biochemical markers of programmed cell death (PCD) suggested that soman-induced cell damage in the brain was apoptotic rather than necrotic. However, it has recently become clear that not all PCD is apoptotic, and the unequivocal demonstration of apoptosis requires ultrastructural examination. Therefore, the present study was undertaken to reinvestigate the damage produced in the brains of mice sacrificed at various times within the first 24 h or at 7 days after a convulsive dose of soman. Classical histology and ultrastructural examination were performed. The immunohistochemical expression of proteins (p53, Bax) involved in PCD, DNA fragmentation (TUNEL method at light and electron microscopy levels) and the glial reaction were also explored. Our study confirms that the severity of lesions depended on the duration of convulsions and shows that cerebral changes were still occurring as late as 7 days after the onset of long-lasting convulsions. Our observations also establish that there was a large variety of ultrastructurally distinct types of cell damage, including hybrid forms between apoptosis and necrosis, but that pure apoptosis was very rare. A prominent expression of p53 and Bax proteins was detected indicating that PCD mechanisms were certainly involved in the morphologically diverse forms of cell death. Since purely apoptotic cells were very rare, these protein expressions were presumably involved either in nonapoptotic cell death mechanisms or in apoptotic mechanisms occurring in parallel with nonapoptotic ones. Moreover, evidence for DNA fragmentation by the TUNEL method was found in apoptotic but also in numerous other morphotypes of cell damage. Therefore, TUNEL-positivity and the expression of PCD-related proteins, in the absence of ultrastructural confirmation, were here shown not to provide proof of apoptosis. In soman poisoning as well as in other cerebral pathologies, premature conclusions on this question can potentially be misleading and might even lead to detrimental therapies.

Involvement of cyclin-dependent kinases in axotomy-induced retinal ganglion cell death.

We have tested the role of cyclin-dependent kinases (CDKs) in the type 3B death of axotomized retinal ganglion cells, by injecting intraocularly olomoucine, roscovitine, or butyrolactone I. Each of these inhibits CDK1, CDK2, and CDK5; CDK1 and CDK2 are involved in cell proliferation, whereas CDK5 is involved in neuronal differentiation. The inhibitors partially protected ganglion cells against the effects of axotomy. These agents may affect the ganglion cells directly, because CDK1, its regulatory subunit cyclin B1, and CDK5 were identified immunohistochemically in the perikarya of ganglion cells, and this was confirmed for CDK1 and CDK5 in Western blots of the ganglion cell layer. These blots showed an axotomy-induced phosphorylation of CDK5 occurring remarkably quickly (within 6 hours of axotomy) but little if any change in the phosphorylation state of CDK1. In addition, we studied the expression of proliferation markers, including proliferating cell nuclear antigen (PCNA) and the synthesis of DNA, by immunohistochemical and autoradiographic methods. Normal or axotomized ganglion cells did not express PCNA and did not synthesize DNA. Although we cannot exclude the possibility that axotomized ganglion cells may leave their quiescent state, our data show that they did not progress beyond the G1 phase of the cell cycle. Finally, in contrast to inhibitors of CDKs, cell cycle blockers with different targets than CDKs did not protect ganglion cells. Globally, our results suggest that axotomy-induced death of ganglion cells involves the activation of CDK1, CDK2, or CDK5 (most probably CDK5) but not the full cell cycle machinery.

Ultrastructure of retinal ganglion cell death after axotomy in chick embryos.

Axotomy often leads to neuronal death, which occurs after a particularly short delay in immature animals. Tectal lesions were made in embryonic day (E) 12 chick embryos, thereby axotomizing the retinal ganglion cells of the contralateral eye, which then died within 3 days. We here describe the ultrastructural changes in the axotomized ganglion cells. The main changes were nuclear invagination and type 3B (cytoplasmic type) cell death characterized by dilation of the perinuclear space, endoplasmic reticulum, and Golgi apparatus. However, nuclear invagination was never seen in type 3B dying cells. All the axotomy-induced retinal ganglion cell death appears to have been of type 3B; apoptosis was not induced by axotomy, as was confirmed by additional light microscopic experiments showing that it did not increase the frequency of apoptotic markers revealed by terminal deoxynucleotidyl transferase-mediated biotinylated UTP nick end labeling (the TUNEL method) labeling and immunoreactivity for activated caspase-3. However, the latter methods did show small numbers of apoptotic cells dying naturally even in control retinas. After the death of the axotomized ganglion cells, they were phagocytosed mainly in Müller processes. The present findings open up the chick tectal lesion model as a system for analyzing type 3B neuronal death in vivo.

Kainate-induced endocytosis in retinal amacrine cells.

Endocytosis is enhanced in some cases of neuronal death. We report for the first time that intraocular injections, in chick embryos, of excitotoxic doses of kainate induce strong endocytosis in retinal amacrine cells destined to die and that even subtoxic doses can induce some degree of endocytosis. That the uptake was due to endocytosis rather than passive diffusion through the plasma membrane was shown ultrastructurally. The endocytosis was demonstrated by using three unrelated tracers--horseradish peroxidase, microperoxidase, and 4.4-kDa fluorescein isothiocyanate (FITC)-labeled dextran--suggesting that it does not depend on the binding of the tracers to a particular receptor. However, it appears to be surprisingly sensitive to the size of the ligand, because a heavier (42-kDa) FITC-dextran was not endocytosed. The induction of endocytosis by kainate can occur even when protein synthesis is blocked. These results indicate that toxic or near-toxic doses of kainate induce endocytosis, raising the question of whether this is mechanistically implicated in causing or preventing excitotoxic neuronal death.

Involvement of autophagy in hypoxic-excitotoxic neuronal death.

Neuronal autophagy is increased in numerous excitotoxic conditions including neonatal cerebral hypoxia-ischemia (HI). However, the role of this HI-induced autophagy remains unclear. To clarify this role we established an in vitro model of excitotoxicity combining kainate treatment (Ka, 30 µM) with hypoxia (Hx, 6% oxygen) in primary neuron cultures. KaHx rapidly induced excitotoxic death that was completely prevented by MK801 or EGTA. KaHx also stimulated neuronal autophagic flux as shown by a rise in autophagosome number (increased levels of LC3-II and punctate LC3 labeling) accompanied by increases in lysosomal abundance and activity (increased SQSTM1/p62 degradation, and increased LC3-II levels in the presence of lysosomal inhibitors) and fusion (shown using an RFP-GFP-LC3 reporter). To determine the role of the enhanced autophagy we applied either pharmacological autophagy inhibitors (3-methyladenine or pepstatinA/E64) or lentiviral vectors delivering shRNAs targeting Becn1 or Atg7. Both strategies reduced KaHx-induced neuronal death. A prodeath role of autophagy was also confirmed by the enhanced toxicity of KaHx in cultures overexpressing BECN1 or ATG7. Finally, in vivo inhibition of autophagy by intrastriatal injection of a lentiviral vector expressing a Becn1-targeting shRNA increased the volume of intact striatum in a rat model of severe neonatal cerebral HI. These results clearly show a death-mediating role of autophagy in hypoxic-excitotoxic conditions and suggest that inhibition of autophagy should be considered as a neuroprotective strategy in HI brain injuries.