Changes in cerebrovascular disease-related deaths and their location during the COVID-19 pandemic in Japan.

OBJECTIVE: The COVID-19 pandemic placed an enormous strain on healthcare systems and raised concerns for delays in the management of patients with acute cerebrovascular events. In this study, we investigated cerebrovascular excess deaths in Japan. STUDY DESIGN: Vital mortality statistics from January 2012 to May 2022 were obtained from the Japanese Ministry of Health, Labour and Welfare. METHODS: Using quasi-Poisson regression models, we estimated the expected weekly number of cerebrovascular deaths in Japan from January 2020 through May 2022 by place of death. Estimates were calculated for deaths in all locations, as well as for deaths in hospitals, in geriatric health service facilities, and at home. The age subgroups of ≥75 and <75 years were also considered. Weeks with a statistically significant excess of cerebrovascular deaths were determined when the weekly number of observed deaths exceeded the upper bound of 97.5% prediction interval. RESULTS: Excess deaths were noted in June 2021 and became more pronounced from February 2022 onward. The trend was notable among those aged ≥75 years and for those who died in hospitals. With respect to the location of deaths, the excess was significant in geriatric health services facilities from April 2020 to June 2021, whereas no evidence of excess hospital deaths was observed during the same period. CONCLUSIONS: Beginning in the late 2021, excess cerebrovascular deaths coincided with the spread of the Omicron variant and may be associated with increased healthcare burden. In 2020, COVID-19 altered the geography of cerebrovascular deaths, with fewer people dying in hospitals and more dying in geriatric health service facilities and at home.

B-type natriuretic peptide levels are decreased by reducing dietary salt intake in patients with compensated heart failure with preserved ejection fraction.

BACKGROUND: Dietary salt restriction is believed to be a mainstay in the management of patients with heart failure. However, the effect of salt intake on heart failure has not been well evaluated in outpatient medical practice. AIMS: The aim of the present study was to assess the hypothesis that B-type natriuretic peptide (BNP) level, as an objective marker of heart failure, is associated with salt intake in patients with heart failure. METHODS: One hundred and thirteen consecutive patients with mild compensated heart failure (77 ± 10 years old, 51 female) were included. We estimated dietary salt intake by the concentration of sodium and creatinine in spot urine. We measured BNP at the time of urine sampling and assessed the relationship between the % changes in BNP levels (%ΔBNP) and the changes in the estimated daily salt excretion (ΔNaCl) during the follow-up period. RESULTS: The baseline median BNP level was 150 (interquartile range: 83-263) pg/mL and the estimated daily salt excretion was 162 ± 45 mmol/day. There was a positive correlation between %ΔBNP and ΔNaCl (r = 0.61, P < 0.01). Multiple regression analysis revealed that %ΔBNP was associated with ΔNaCl (P < 0.01), but not with changes in systolic blood pressure and bodyweight. CONCLUSIONS: Changes in BNP levels were associated with changes in the estimated daily salt excretion in outpatients with compensated heart failure. Salt restriction may be beneficial for the management of patients with heart failure.

Activation of the brainstem precedes and outlasts the K-complex in humans.

A K-complex (KC) in the electroencephalographs (EEGs) indicates a moderate depth of slow-wave sleep (SWS) in humans and animals. The cortical activities are upregulated during the periods between the KCs ("up state"), and it is proposed that temporarily stored memories during wakeful periods will be consolidated during these periods. Although the underlying mechanism for KCs is proposed to be in the cortex itself, the involvement of the brainstem has not been explored. Here we investigate the excitability changes of the brainstem preceding, during, and after KCs in humans. We simultaneously recorded brainstem auditory evoked potentials (BAEPs) with EEGs, and sequentially analyzed BAEPs around the KCs. The results showed a transient activation of the ventral brainstem preceding the KC and a sustained activation of the dorsal brainstem outlasting the KC. Thus, it is suggested that KCs are triggered by the activation of the brainstem and that the "up state" is maintained by the sustained activation of the brainstem.

Risk stratification of coronary revascularization patients by using clinical and angiographic data.

AIM: Patients undergoing coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI) are still at a substantial risk of death after their procedures. A core group of preoperative and preprocedural risk factors and conditions contributes the majority of inherent mid- and long-term mortality risk in these patients. Therefore, we sought to develop a classification tree model as a practical and user-friendly method of predicting mid-term survival after coronary revascularization procedures. METHODS: We retrospectively analyzed data from a single, large-volume institution. Specifically, we examined all-cause three-year mortality in 3387 consecutive patients with multivessel or single proximal left anterior descending coronary artery disease who underwent either PCI with stenting or CABG. RESULTS: Recursive partitioning indicated that the best single predictor of death within three years was a history of heart failure (HF), followed by a proximal left circumflex artery (pLCX) lesion and age greater than 65 years for patients with and without a history of HF, respectively. With these variables, patients were readily stratified into low-, intermediate-, and high-risk groups whose risks of death over three years ranged from 2.3% to 36.2%. Among patients with a history of HF, pLCX stenosis was an independent predictor of mid-term mortality after adjustment for other known risk factors (hazard ratio, 1.46; 95% CI, 1.04-2.03). CONCLUSION: The constructed risk stratification scheme stratified patients into groups at low, intermediate, and high risk of death within three years. Stenosis of the pLCX seems to be an important prognostic factor for patients with a history of HF.

Inhibition of N-methyl-D-aspartate receptor activity resulted in aberrant neuronal migration caused by delayed morphological development in the mouse neocortex.

Embryonic and neonatal neocortical neurons already express functional N-methyl-D-aspartate (NMDA) receptors before they form synapses. To elucidate the role of NMDA receptors in neuronal migration in the developing neocortex, we visualized radially migrating neurons by transferring the enhanced green fluorescent protein (EGFP) gene into the ventricular zone (VZ) of the mouse neocortex using in utero electroporation at E15.5. Two days later, we prepared neocortical slices and examined the EGFP-positive cells using time-lapse imaging in the presence of the NMDA receptor antagonist Cerestat. The EGFP-positive cells generated in the VZ in the control slices exhibited a multipolar morphology, but within several hours they became bipolar (with a leading process and an axon-like process) and migrated toward the pial surface. By contrast, many of the multipolar cells in the Cerestat-treated slices failed to extend either process and become bipolar, and frequently changed direction, although they ultimately reached their destination even after Cerestat-treatment. To identify the molecules responding for mediating NMDA signaling during neuronal migration and the changes in morphology observed above, we here focused on Src family kinases (SFKs), which mediate a variety of neuronal functions including migration and neurite extension. We discovered that the activity of Src and Fyn was reduced by Cerestat. These findings suggest that NMDA receptors are involved in neuronal migration and morphological changes into a bipolar shape, and in the activation of Src and Fyn in the developing neocortex.

Pigment epithelium-derived factor up-regulation induced by memantine, an N-methyl-D-aspartate receptor antagonist, is involved in increased proliferation of hippocampal progenitor cells.

Memantine is classified as an NMDA receptor antagonist. We recently reported that memantine promoted the proliferation of neural progenitor cells and the production of mature granule neurons in the adult hippocampus. However, the molecular mechanism responsible for the memantine-induced promotion of cellular proliferation remains unknown. In this study we searched for a factor that mediates memantine-induced cellular proliferation, and found that pigment epithelium-derived factor (PEDF), a broad-acting neurotrophic factor, is up-regulated in the dentate gyrus of adult mice after the injection of memantine. PEDF mRNA expression increased significantly by 3.5-fold at 1 day after the injection of memantine. In addition, the expression level of PEDF protein also increased by 1.8-fold at 2 days after the injection of memantine. Immunohistochemical study using anti-PEDF antibody showed that the majority of the PEDF-expressing cells were protoplasmic and perivascular astrocytes. Using a neurosphere assay, we confirmed that PEDF enhanced cellular proliferation under the presence of fibroblast growth factor-2 (FGF-2) and epidermal growth factor (EGF) but was not involved in the multilineage potency of hippocampal progenitor cells. Over expression of PEDF by adeno-associated virus, however, did not stimulate cellular proliferation, suggesting PEDF per se does not promote cellular proliferation in vivo. These findings suggest that the memantine induced PEDF up-regulation is involved in increased proliferation of hippocampal progenitor cells.

The origin and cell lineage of microglia: new concepts.

Despite intense study, the precise origin and cell lineage of microglia, the resident mononuclear phagocytes of the nervous system, are still a matter for debate. Unlike macroglia (astrocytes and oligodendrocytes) and neurons, which are derived from neuroectoderm, microglial progenitors arise from peripheral mesodermal (myeloid) tissue. The view still commonly held is that tissue-resident mononuclear phagocytes (including microglia) are derived from circulating blood monocytes and these take up residence late in gestation and postnatally. However, microglial progenitors colonise the nervous system primarily during embryonic and fetal periods of development. Recent evidence indicates differences between the lineage of mononuclear phagocytes during the embryonic and fetal period from that in the neonate and adult-mononuclear phagocytes that take up residence within tissues are derived from a lineage of myeloid cells that is independent of the monocyte lineage. Our own findings on the development and differentiation of microglial progenitors, taken together with findings by other investigators, and in the context of the heterogeneity between myeloid differentiation in the fetus and in the adult, support the view that microglia are derived prenatally from mesodermal progenitors that are distinct from monocytes. Furthermore, microglial progenitors colonise the nervous system via extravascular routes initially. These findings challenge the concept that resident microglia in the nervous system are derived from circulating blood monocytes. Work is still underway to establish the tissue of origin and lineage of microglial progenitors in vivo. This information is critical not only from a developmental perspective, but significantly from a therapeutic viewpoint, as (i) the unique property of microglial progenitors to colonise the nervous system from the periphery allows these cells to be exploited as a biological and non-invasive means for cell therapy by delivering genes to the nervous system (microglial engraftment), and (ii) there are indications that microglial progenitors are specifically able to home to the nervous system. Use of microglial progenitors for therapeutic purposes becomes feasible only if the origin and cell lineage of these microglial progenitors are known and these cells can be isolated and manipulated in vitro (i.e., to express specific trophic factors) prior to therapeutic transfer (e.g., intravenously) in vivo. In this paper, we shall briefly consider the existing concepts on the origin and lineage of microglial progenitors and discuss new hypotheses in the light of emerging data that suggest clear differences between fetal and adult ontogeny of myeloid cells.

The mutation in the ATP-binding region of JAK1, identified in human uterine leiomyosarcomas, results in defective interferon-gamma inducibility of TAP1 and LMP2.

The presentation of human leukocyte antigens (HLA) class I requires the coordinated expression of numerous components involved in antigen presentation. Tumor cells may alter the antigen presentation by HLA class I, allowing them to evade antitumor immunity. In many cases, the lack of antigen presentation can be attributed to the downregulation of genes needed for antigen processing, such as the transporters associated with antigen processing (TAP) 1, and the proteasomal component, low molecular weight proteins (LMP) 2. The TAP1 and LMP2 genes are transcribed from a shared bidirectional promoter containing an interferon (IFN)-gamma-response factor element; thus, the IFN-gamma-signal strongly induces both TAP1 and LMP2 expression. Low molecular weight proteins2-deficient mice exhibited the development of uterine leiomyosarcomas. Here, the differential responsiveness to IFN-gamma of the SKN human uterine leiomyosarcomas cell line was investigated. We now identify the G871E mutation in the ATP-binding region of Janus kinases 1, suggesting that the loss of TAP1 and LMP2 induction is a defect in the earliest steps of the IFN-gamma-signal pathway, resulting in the inability of SKN cells to upregulate the antigen-processing pathway. Understanding the mechanisms by which these tumors circumvent cytokine signalling, thereby evading antitumor-specific immunity, would greatly aid the efficacy of immunotherapy for treating uterine leiomyosarcomas.

Visualization of microglia in living tissues using Iba1-EGFP transgenic mice.

Microglia are thought to play important roles not only in repairing injured tissue but in regulating neuronal activity, and visualizing the cells is very useful as a means of further investigating the function of microglia in vivo. We previously cloned the ionized calcium-binding adaptor molecule 1 (Iba1) gene, which is expressed selectively in microglia/microphages. To generate new transgenic mice to visualize microglia with enhanced green fluorescent protein (EGFP), we here constructed a plasmid carrying EGFP cDNA under control of the Iba1 promoter. This construct was injected into C57B/6 mouse zygotes, and the Iba1-EGFP transgenic line was developed. Fluorescent in-situ hybridization analysis revealed that the Iba1-EGFP transgene was located on chromosome 11D. No obvious defects were observed during development or in adulthood, and the EGFP fluorescence remained invariant over the course of at least four generations. Judging from the immunoreactivity with anti-Iba1 antibody, all EGFP-positive cells in the adult brain were ramified microglia. In the developing transgenic embryos, EGFP signals were detected as early as embryonic Day 10.5. The most prominent EGFP signals were found in forebrain, spinal cord, eye, foreleg, yolk sac, liver, and vessel walls. At postnatal Day 6, clear EGFP signals were observed in the supraventricular corpus callosum, known as "fountain of microglia", where ameboid microglia migrate into the brain parenchyma and mature into ramified microglia. Iba1-EGFP transgenic mice thus permit observation of living microglia under a fluorescence microscope and provide a useful tool for studying the function of microglia in vivo.

Distribution of the galectin-1 mRNA in the rat nervous system: its transient upregulation in rat facial motor neurons after facial nerve axotomy.

Galectin-1 is a member of the animal lectin family that displays conserved consensus sequences and similar carbohydrate binding specificities. Recent analyses revealed that galectin-1 plays an important role in the process of nerve regeneration. We analyzed the topological expression of galectin-1 mRNA in adult rat nervous system. Galectin-1 mRNA was predominantly observed in the cell bodies of neurons such as oculomotor nucleus (III), trochlear nucleus (IV), trigeminal motor nucleus (V), abducens nucleus (VI), facial nucleus (VII), hypoglossal nucleus (XII), red nucleus, and locus ceruleus. Neurons in pineal gland and dorsal root ganglia expressed galectin-1 mRNA. We next tested whether the axotomy of facial nerve altered the expression of galectin-1 mRNA in motor neurons. In the adult rats, the axotomy of facial nerve induced transient upregulation of galectin-1 mRNA around 6 h after axotomy. These results indicate that galectin-1 may play roles in the early event of the nerve injury and regeneration through the transient change of its expression level.

Widespread activation of the brainstem preceding the recruiting rhythm in human epilepsies.

The excitability change of the brainstem was investigated before and during the conspicuous epileptic discharge in six patients with generalized convulsive seizures. The discharge consisted of a short duration of recruiting rhythm, which was considered equivalent to the seizure discharge on electroencephalogram. The excitability of the brainstem was measured with the parameters (amplitude and area) of component waves (wave-III and -V) of brainstem auditory evoked potentials. The theoretical background of the analysis is that brainstem auditory evoked potentials are 'far-field' potentials, by which they convey the information on the activity change of the brainstem even during the paroxysmal discharge within the cortex. The excitability of both the ventral (parameters of wave-III) and the dorsal brainstem (parameters of wave-V) exhibited a synchronized change (activation-inactivation). They were enhanced from -2.4+/-0.4 s, reaching the maxima before the onset of the seizure discharge, and decayed corresponding to the emergence of the recruiting rhythm. The results suggest the possibility that the widespread (ventral and dorsal) and synchronized activation of the brainstem triggers the seizure discharge in human generalized epilepsy. During the widespread activation of the brainstem, both the thalamus and the cortex probably undergo a suppressed inhibitory state through the cholinergic activation, precipitating the seizure discharge.

Mechanisms underlying extracellular ATP-evoked interleukin-6 release in mouse microglial cell line, MG-5.

Microglia play various important roles in the CNS via the synthesis of cytokines. The ATP-evoked production of interleukin-6 (IL-6) and its intracellular signals were examined using a mouse microglial cell line, MG-5. ATP, but not its metabolites, produced IL-6 in a concentration-dependent manner. Although ATP activated two mitogen-activated protein kinases, i.e. p38 and extracellular signal-regulated protein kinase, only p38 was involved in the IL-6 induction. However, the activation of p38 was not sufficient for the IL-6 induction because 2'- and 3'-O-(4-benzoylbenzoyl) ATP, an agonist to P2X7 receptors, failed to produce IL-6 despite the fact that it activated p38. Unlike in other cytokines in microglial cells, P2Y rather than P2X7 receptors seem to have a major role in the IL-6 production by the cells. The ATP-evoked IL-6 production was attenuated by Gö6976, an inhibitor of Ca(2+)-dependent protein kinase C (PKC). The P2Y receptor responsible for these responses was insensitive to pertussis toxin (PTX) and was linked to phospholipase C. Taken together, ATP acting on PTX-insensitive P2Y receptors activates p38 and Ca(2+)-dependent PKC, thereby resulting in the mRNA expression and release of IL-6 in MG-5. This is a novel pathway for the induction of cytokines in microglia.

Establishment of CHO cell lines expressing four N-methyl-D-aspartate receptor subtypes and characterization of a novel antagonist PPDC.

To develop an assay system that allows the N-methyl-D-aspartate (NMDA) receptor subtype-selective antagonistic potency of drugs, we have established Chinese hamster ovary cell lines expressing the four NMDA receptor subtypes (GluRepsilon1/zeta1-GluRepsilon4/zeta1) heat-indelibly. Using these clonal cells, we found that a novel antagonist, (1S,2R)-1-phenyl-2[(S)-1-aminopropyl]-N,N-diethylcyclopropanecarboxamide, was less selective for the GluRepsilon1/zeta1: the IC(50) values for the GluRepsilon1/zeta1-GluRepsilon4/zeta1 were 41.7, 13.3, 12.6 and 11.5 microM, respectively, while two well-known antagonists, DL-2-amino-5-phosphonovaleric acid and ifenprodil, showed the known potency and selectivity for each subtype. Thus, the established clonal cells are of use in characterizing the pharmacological properties of drugs that act on NMDA receptors.

Induction of CHOP and apoptosis by nitric oxide in p53-deficient microglial cells.

Excessive nitric oxide (NO) has been implicated in neurotoxicity after stresses such as ischemia. NO toxicity is generally thought to be mediated by the DNA damage-p53 pathway or mitochondrial dysfunction. We investigated the mechanism of NO toxicity by using murine microglial MG5 cells established from p53-deficient mice. When MG5 cells were exposed to bacterial lipopolysaccharide plus interferon-gamma, mRNA and protein for inducible NO synthase (iNOS) were markedly induced, and apoptosis occurred. Under these conditions, we found that mRNA and protein for CHOP/GADD153, a C/EBP family transcription factor which is involved in endoplasmic reticulum (ER) stress-induced apoptosis, are induced. iNOS mRNA was induced 2 h after treatment, whereas CHOP mRNA began to increase at 6 h with a time lag. CHOP mRNA was also induced by NO donors S-nitroso-N-acetyl-DL-penicillamine (SNAP) or NOC18, or a peroxynitrite generator 3-(4-morpholinyl)-sydnonimine hydrochloride (SIN-1). Bip/GRP78, an ER chaperone which is known to be induced by ER stress, was also induced by SNAP or SIN-1, indicating that NO causes ER stress. These results suggest that NO-induced apoptosis in MG5 cells occurs through the ER stress pathway involving CHOP, but is independent of p53.

Nitric oxide inhibits the proliferation of murine microglial MG5 cells by a mechanism involving p21 but independent of p53 and cyclic guanosine monophosphate.

We investigated the effect of nitric oxide (NO) on the proliferation of microglial MG5 cells established from p53-deficient mice. Cells were treated with bacterial lipopolysaccharide and interferon-gamma, and expression of inducible NO synthase (iNOS) and p21/waf1, a cyclin-dependent kinase inhibitor protein which is a critical downstream effector of p53, was investigated by RNA blot and immunoblot analyses. iNOS mRNA was induced 2 h after treatment and increased with time up to 24 h. p21 mRNA was expressed at a low level in untreated cells and increased with a kinetics similar to that for iNOS mRNA. iNOS and p21 proteins were also induced. An NO donor SNAP induced p21 mRNA and protein. SNAP inhibited incorporation of [(3)H]thymidine in MG5 cells in a dose-dependent manner. 8-Bromo-cGMP neither induced p21 mRNA nor inhibited [(3)H]thymidine incorporation. These results suggest that NO inhibits the proliferation of MG5 cells by induction of p21, which occurs independent of p53 and cGMP.

Impaired microglial activation in the brain of IL-18-gene-disrupted mice after neurovirulent influenza A virus infection.

Knockout of the interleukin-18 (IL-18) gene predisposed mice to impaired clearance of neurovirulent influenza A virus-infected neurons from the brain. In wild-type mice, IL-18 molecule-producing microglia/macrophages emerged in virally attacked regions as early as day 3 after infection. Microglial transformation into macrophages culminated at day 7 to 9, with upregulated expression of Iba1, a novel calcium-binding protein that controls phagocytic functions of microglia/macrophages. In IL-18-/- mice, microglial transformation was interrupted with reduced Iba1 expression. Interferon-gamma (IFN-gamma)-immunopositive neurons appeared in and around virally invaded regions in wild-type mice, peaking in number at day 7, whereas such cells were barely detected in IL-18-/- mice. Stereotaxic microinjection of recombinant IFN-gamma triggered microglial transformation in IL-18-/- mice and upregulated Iba1 expression, leading to effective eradication of virally infected neurons. Collectively, these results suggest that IL-18 plays a key role in activating microglial functions directed against the influenza virus infection by inducing neuronal IFN-gamma in the brain parenchyma.

Iba1 is an actin-cross-linking protein in macrophages/microglia.

Iba1 is a 17-kDa EF hand protein that is specifically expressed in macrophages/microglia and is upregulated during the activation of these cells. When exposed to macrophage colony-stimulating factor (M-CSF), microglia cell line MG5 immediately produces intense membrane ruffles in which Iba1 accumulates together with filamentous actin. In this study, we investigated the physical interaction between Iba1 and actin by centrifugation assay and electron microscopic examination and showed that Iba1 possesses actin-binding and -cross-linking activities. Inhibitory mutant Iba1 that suppresses M-CSF-induced membrane ruffling had lost the actin-cross-linking activity, and it inhibited the cross-linking activity of intact Iba1. These results indicate that Iba1 is a macrophage/microglia-specific actin-cross-linking protein essential for M-CSF-induced membrane ruffling.

Neurotrophin secretion from cultured microglia.

Because microglia have been suggested to produce neurotrophins, we tested this ability in vitro. Rat primary microglia were found to constitutively secrete a limited amount of brain-derived neurotrophic factor (BDNF), but nerve growth factor (NGF) and neurotrophin-3 (NT-3) were undetectable in the conditioned medium. Stimulation of the cells with lipopolysaccharide (LPS) increased BDNF secretion, and induced NGF secretion. As a first step to examine this regulation system, the association of protein kinase C (PKC) was pharmacologically analyzed. A PKC activator, phorbol-12-myristate-13-acetate, enhanced the secretion of BDNF. Pre-treatment of microglia with a PKC inhibitor, bisindolylmaleimide, suppressed LPS-stimulated BDNF secretion as well as the constitutive one. These results suggest that the PKC signaling cascade is closely associated with BDNF secretion. Among PKC isoforms, PKCalpha probably plays a role in BDNF secretion, based on the results of experiments using a specific PKC activator, 1-oleoyl-2-acetyl-sn-glycerol, and a specific PKC inhibitor, Gö 6976, and by immunoblotting. Taken together, these findings suggest that the secretion of BDNF from microglia is regulated through PKCalpha-associated signal transduction mechanism.

Microglia: activation and their significance in the central nervous system.

Microglia are resident monocyte-lineaged cells in the brain. Their characteristic feature is that they react to injury and diseases of the brain and become morphologically and functionally activated. Although some trigger molecules which activate microglia are predicted to be released from injured or affected cells, such molecules have not yet been identified. The main role of activated microglia is believed to be in brain defense, as scavengers of dead cells, and as immune or immunoeffector cells. Recent biochemical and neurobiological studies have further indicated that they significantly affect the pathological state and/or regulate the regenerative state and remodeling of the brain by producing a variety of biologically active molecules including cytotoxic and neurotrophic molecules.

Ability of rat microglia to uptake extracellular glutamate.

Since it has been suggested that microglia in vivo act as glutamate scavengers, this possibility was investigated in primary cultured microglia. The microglia showed specific abilities to uptake (14)C-glutamate depending on incubation time and numbers of cells used. The activity was suppressed by a specific inhibitor for a glial cell-type transporter, glutamate transporter-1 (GLT-1) (EAAT2). However, that of cultured astrocytes was not affected. These results suggest that microglia uptake glutamate by means of GLT-1. Supporting these results, immunoblotting revealed the presence of GLT-1 in the microglia, while only glutamate-aspartate transporter (GLAST) (EAAT1: another glial cell-type transporter) was detected in the astrocytes. All together, these results indicate that microglia can act as glutamate scavengers in vivo by expressing the glutamate transporter GLT-1.