Neuritic Beading Induced by Activated Microglia Is an Early Feature of Neuronal Dysfunction Toward Neuronal Death by Inhibition of Mitochondrial Respiration and Axonal Transport*

Recent studies suggest that excitotoxicity may contribute to neuronal damage in neurodegenerative diseases including Alzheimer disease, Parkinson disease, amyotrophic lateral sclerosis, and multiple sclerosis. Activated microglia have been observed around degenerative neurons in these diseases, and they are thought to act as effector cells in the degeneration of neural cells in the central nervous system. Neuritic beading, focal bead-like swellings in the dendrites and axons, is a neuropathological sign in epilepsy, trauma, ischemia, aging, and neurodegenerative diseases. Previous reports showed that neuritic beading is induced by various stimuli including glutamate or nitric oxide and is a neuronal response to harmful stimuli. However, the precise physiologic significance of neuritic beading is unclear. We provide evidence that neuritic beading induced by activated microglia is a feature of neuronal cell dysfunction toward neuronal death, and the neurotoxicity of activated microglia is mediated through N-methyl-d-aspartate (NMDA) receptor signaling. Neuritic beading occurred concordant with a rapid drop in intracellular ATP levels and preceded neuronal death. The actual neurite beads consisted of collapsed cytoskeletal proteins and motor proteins arising from impaired neuronal transport secondary to cellular energy loss. The drop in intracellular ATP levels was because of the inhibition of mitochondrial respiratory chain complex IV activity downstream of NMDA receptor signaling. Blockage of NMDA receptors nearly completely abrogated mitochondrial dysfunction and neurotoxicity. Thus, neuritic beading induced by activated microglia occurs through NMDA receptor signaling and represents neuronal cell dysfunction preceding neuronal death. Blockage of NMDA receptors may be an effective therapeutic approach for neurodegenerative diseases.

Recent studies suggest that excitotoxicity may contribute to neuronal damage in neurodegenerative diseases including Alzheimer disease, Parkinson disease, amyotrophic lateral sclerosis, and multiple sclerosis. Activated microglia have been observed around degenerative neurons in these diseases, and they are thought to act as effector cells in the degeneration of neural cells in the central nervous system. Neuritic beading, focal bead-like swellings in the dendrites and axons, is a neuropathological sign in epilepsy, trauma, ischemia, aging, and neurodegenerative diseases. Previous reports showed that neuritic beading is induced by various stimuli including glutamate or nitric oxide and is a neuronal response to harmful stimuli. However, the precise physiologic significance of neuritic beading is unclear. We provide evidence that neuritic beading induced by activated microglia is a feature of neuronal cell dysfunction toward neuronal death, and the neurotoxicity of activated microglia is mediated through N-methyl-D-aspartate (NMDA) receptor signaling. Neuritic beading occurred concordant with a rapid drop in intracellular ATP levels and preceded neuronal death. The actual neurite beads consisted of collapsed cytoskeletal proteins and motor proteins arising from impaired neuronal transport secondary to cellular energy loss. The drop in intracellular ATP levels was because of the inhibition of mitochondrial respiratory chain complex IV activity downstream of NMDA receptor signaling. Blockage of NMDA receptors nearly completely abrogated mitochondrial dysfunction and neurotoxicity. Thus, neuritic beading induced by activated microglia occurs through NMDA receptor signaling and represents neuronal cell dysfunction preceding neuronal death. Blockage of NMDA receptors may be an effective therapeutic approach for neurodegenerative diseases.
Central nervous system inflammation including microglial activation likely contributes to the neurotoxicity observed in neurodegenerative diseases such as Alzheimer disease, Parkin-son disease, amyotrophic lateral sclerosis, and multiple sclerosis (1)(2)(3)(4)(5)(6)(7). Additionally, excitotoxicity might lead to neuronal damage in these neurodegenerative diseases (8,9). Microglia can act as not only antigen-presenting cells but also effector cells to damage central nervous system cells directly in vitro and in vivo (10 -18). Conversely, microglia may have neuroprotective effects mediated by neurotrophin release, glutamate uptake, and ingesting neurotoxic substances (19 -22). Therefore, the role of microglia in either the pathogenesis of or protection from neurodegenerative diseases is still entirely unresolved.
In this study, we sought to elucidate the mechanisms regulating the induction of neuritic beading by activated microglia. We show that neuritic beading induced by activated microglia is a feature of neuronal cell dysfunction preceding neuronal death, and this neurotoxicity is mediated by N-methyl-D-aspartate (NMDA) receptor signaling following glutamate binding. Blockade of the NMDA receptor may protect against the development of neurodegenerative diseases.

EXPERIMENTAL PROCEDURES
Cell Culture-All reagents except those specifically mentioned were obtained from Sigma. The protocols for animal experiments were approved by the Animal Experiment Committee of Nagoya University. * This work was supported by grants from the Ministry of Health, Labor and Welfare of Japan, and a Center of Excellence grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Microglia were isolated from primary mixed glial cell cultures from newborn C57/BL6 mice on the 14th day with the "shaking off " method as described previously (40). The purity of the cultures was 97 to 100% as determined by Fc receptor-specific immunostaining as described previously (40). Cultures were maintained in Dulbecco's modified Eagle's minimum essential medium supplemented with 10% fetal calf serum (JRH Biosciences, Lenexa, KS), 5 g/ml bovine insulin, and 0.2% glucose. Neuron cultures were prepared from C57BL/6 mice at embryonic day 17 using the Nerve-Cell Culture System (Sumitomo Bakelite, Akita, Japan) as described previously (41)(42)(43). Briefly, cortices were dissected and freed of meninges. Cortical fragments were dissociated into single cells using dissociation solution, and they were resuspended in Nerve-Cell Culture Medium (serum-free conditioned medium from 48-h rat astrocyte confluent cultures based on Dulbecco's modified Eagle's minimum essential medium/F-12 with N2 supplement, Sumitomo Bakelite). Primary neuronal cells were plated on 12-mm polyethyleneimine-coated coverslips (Asahi Techno Glass Corp., Chiba, Japan) in 24-well multidishes at a density of 5 ϫ 10 4 cells/well. The purity of the cultures was more than 95% as determined by NeuN-specific immunostaining as described previously (42,43).
Assessment of Neuritic Beading-To assess neuritic beading, neurons were observed under a phase-contrast microscope at each time point (0, 1, 3, 6, 12, and 24 h) by a modification of the method previously reported by Park et al. (32). More than 200 neurons in duplicate wells were assessed blindly in three independent trials. The ratio of bead-bearing neurons was calculated as a percentage of total cells.
Assessment of Cell Death-Cell death was assessed by the dye-exclusion method with propidium iodide (PI; Molecular Probes, Eugene, OR) as described previously (44). At each time point (0, 1, 3, 6, 12, and 24 h) after stimulation, cells were incubated with 2 g/ml PI-containing medium for 15 min at 37°C. More than 200 neurons in duplicate wells were assessed blindly in three independent trials under a conventional fluorescent microscope. The ratio of dead cells was calculated as a percentage of PI-positive cells among total cells.
To detect apoptosis, we used the terminal deoxynucleotidyltransferase-mediated UTP end labeling (TUNEL) assay with the in situ cell death detection kit (Roche Diagnostics) as described previously (44). TUNEL assay was carried out at each time point (0, 1, 3, 6, 12, and 24 h) according to the manufacturer's protocol. As a positive control, neurons were incubated with 10 nM staurosporin for 24 h. More than 200 neurons in duplicate wells were assessed blindly in three independent trials under a conventional fluorescent microscope. The ratio of apoptotic cells was calculated as a percentage of TUNEL-positive cells among total cells.
Assessment of Intracellular ATP Levels-To measure intracellular ATP levels, we used a luminometric assay with ApoSENSOR Cell Viability Assay Kit (BioVision, Mountain View, CA) according to the manufacturer's protocol. Assays were carried out at each time point (0, 1, 3, 6, 12, and 24 h) in six independent trials. ATP concentration at each time point was calculated as a percentage of control.
Assessment of Mitochondrial Respiration Inhibition-To analyze mitochondrial respiration inhibition, we carried out a mitochondrial respiration recovery assay by modifying the method previously reported by Rego et al. (45). If complex I is inhibited, mitochondrial respiration recovers following the addition of succinate as a substrate for complex II/III. If complex II/III are inhibited, mitochondrial respiration recovers after the addition of ascorbic acid and TMPD as substrates for complex IV. If complex IV is inhibited, mitochondrial respiration does not recover following the addition of these drugs. Neurons were preincubated for 1 h with each drug, and neurons were then incubated with activated microglia-conditioned medium containing drug. The final concentration of each drug was as follows: 5 mM succinate, 5 mM ascorbic acid, and 0.25 mM TMPD. 10 M Rotenone was added to inhibit complex I activity. 0.25 g/ml Antimycin A was added to inhibit complex II/III activity. 0.5 mM NaN 3 was added to inhibit complex IV activity. Recovery of mitochondrial respiration was detected by measuring the intracellular ATP levels. Assays were carried out after a 3-h incubation in six independent trials. Each ATP concentration was calculated as a percentage of control.
Assessment of Mitochondrial Impairment-To assess mitochondrial membrane potential, we used MitoTracker Red CMXRos (Molecular Probes), a dye whose staining intensity is directly proportional to mitochondrial membrane potential at concentrations lower than 50 nM. At each time point (0, 1, 3, 6, 12, and 24 h), cells were incubated with 20 nM MitoTracker-containing medium for 30 min at 37°C. Cells were fixed and permeabilized as described below under "Immunocytochemistry." Fluorescent signal intensity was quantified with a confocal laser scanning microscopic system (LSM510; Carl Zeiss, Oberkochen, Germany). More than 100 cells in duplicate wells were assessed blindly in three independent trials. Relative signal intensity at each time point was expressed as a percentage of control.
Assessment of NO Generation-To measure NO production, we performed the Griess reaction on the media at each time point (0, 1, 3, 6, 12, and 24 h) as described previously (16). Absorbance at 540 nm was measured in a multiple plate reader. Assays were carried out in six independent trials.
Assessment of Glutamate Release-To measure extracellular glutamate concentrations, we used the Glutamate Assay Kit colorimetric assay (Yamasa Corp., Tokyo, Japan) according to the manufacturer's protocol at each time point (0, 1, 3, 6, 12, and 24 h). Absorbance at 600 nm was measured in a multiple plate reader. Assays were carried out in six independent trials.
Analysis of Axonal Transport-Axonal transport was analyzed under a time-lapse phase-contrast microscopic system (Axiovert 200M/ Cell Observer System, Carl Zeiss). The number of particles (diameter Ն50 nm) moving antegrade and retrograde in axons were counted for each hour (0ϳ6 h) as described previously (46). Three independent fields were assessed blindly. Data were expressed as a percentage of control.
Statistical Analysis-All results were analyzed by one-way analysis of variance with a Tukey-Kramer post-hoc test using Statview software version 5 (SAS Institute Inc., Cary, NC).

Activated Microglia Induce Neuritic Beading and Subsequent Neuronal Cell
Death-Activated microglia conditioned medium rapidly induced numerous beads in most neurites (Figs. 1B and 2A). Beaded neurites were narrow and pale compared with control cells, reflecting neurite damage (Fig.  1B). Activated microglia-conditioned medium also induced neuronal cell death at a later phase (Fig. 2B), but very little apoptotic cell death was detected by TUNEL assay (Fig. 2B). Thus, activated microglia appear to induce necrotic cell death in neurons; this is consistent with previous reports (47,48).

A Rapid Increase in Neuritic Beading and a Rapid Drop in Intracellular ATP Levels Are Early Features of Neuronal Cell
Dysfunction Induced by Microglia-Activated microglia-conditioned medium induced a rapid increase in bead-containing neurons ( Fig. 2A) and a rapid drop in intracellular ATP levels (Fig. 2D), whereas cell viability (Fig. 2B), mitochondrial reduction activity (Fig. 2C), and mitochondrial membrane potential (Fig. 2E) were relatively unchanged during the early stages after treatment (Fig. 2, shaded areas).
During this early period, neurons were likely running an energy deficit to maintain homeostasis. Neuritic beading could be an early feature of neuronal cell dysfunction resulting from energy loss induced by activated microglia.
Activated Microglia Inhibit Mitochondrial Respiratory Chain Complex IV Activity in Neurons-We carried out mitochondrial respiration recovery assays to determine the mitochondrial respiratory chain complex in neurons that are inhibited by activated microglia (Fig. 2F). Rotenone, a complex I inhibitor, reduced intracellular ATP levels by ϳ80% that seen in controls, and this reduction was abrogated by the addition of succinate, a substrate of complex II/III. Antimycin A, a complex II/III inhibitor, reduced intracellular ATP levels by ϳ85% compared with controls, and ascorbic acid and TMPD, complex IV substrates, reversed these effects. NaN 3 , a specific complex IV inhibitor, led to a reduction in intracellular ATP levels of ϳ90% compared with controls, and no substrates were able to restore normal ATP levels. As seen with NaN 3 , no substrates were able to restore intracellular ATP levels or mitochondrial respiration of neurons treated with activated microglia (Fig. 2F, black  columns). Thus, activated microglia inhibit neuronal mitochondrial respiration primarily through inhibiting complex IV activity. We also confirmed that NMDA inhibited complex IV activity (Fig. 2F, dotted columns).
Neuritic Beads Colocalized with Collapsing Cytoskeletal Proteins-Time course immunocytochemical analysis revealed that neuritic bead formation was accompanied by the gradual collapse of the neuronal network (Fig. 3). Neuritic beads colocalized with tubulin (Fig. 3, A-F and Y), p-NF (Fig. 3, G-L and Y), and MAP2 (Fig. 3, M-R and Y). It was unclear whether neuritic beads colocalized with actin (data not shown). The fluorescent signal arising from these cytoskeletal proteins gradually weakened as the neuronal network was collapsing. Neurons were stained with rabbit polyclonal anti-MnSOD antibody (a mitochondrial marker, green, A-F), MitoTracker Red CMXRos (red, G-L), and mouse monoclonal anti-␤ III -tubulin antibody (blue, M-R). The mitochondria were first distributed throughout the neuronal network. Y and Z are enlargements of S and X, respectively. After addition of activated microglia-conditioned medium, the mitochondria gradually accumulated in the cell body and disappeared from distal neurites (A-F). Mitochondria morphology also changed from a granular pattern to a tubular and reticular pattern (A-F, Y, and Z). Note the sparing of MitoTracker fluorescence during an early stage following treatment (those quantitative data were indicated in Fig. 2E). Scale bar, 10 m.
Neuritic bead formation is associated with the collapse of cytoskeletal proteins such as tubulin, NF, and MAPs.
Activated Microglia Affect the Morphology and Distribution of Neuronal Mitochondria-Time course immunocytochemical analysis also revealed that activated microglia-conditioned medium changed the morphology and distribution of mitochondria in neurons (Fig. 4). Prior to treatment, mitochondria were distributed throughout the neurites and cell bodies (Fig. 4, A  and S). After the addition of activated microglia-conditioned medium, however, the mitochondria gradually collected in the cell body and disappeared from distal neurites (Fig. 4, B-F and T-X) as the neuronal network was collapsing (Fig. 4, N-R). The mitochondria partially colocalized with proximal neuritic beads because they disappeared from distal neurites. As the distribution of the mitochondria changed, their morphology shifted gradually from a granular pattern to a tubular and reticular pattern (Fig. 4, A-F, Y, and Z), reflecting a shift in the demand for mitochondrial energy (49).
As shown in Fig. 2E, the mitochondrial membrane potential was relatively unchanged during the early stages following treatment with activated microglia-conditioned medium (Fig.  4, G-I), but it gradually declined over time (Fig. 4, J-L) as assessed by the fluorescence intensity of the MitoTracker dye.
Neuritic Beads Colocalized with Collapsed Axonal Transport Motor Proteins-We next examined the effects of microglia on the motor proteins responsible for axonal transport because activated microglia disturbed the cytoskeletal network and mitochondrial distribution in neurons. Time course immunocytochemical analysis revealed that neuritic beads colocalized with kinesin, a motor protein involved in antegrade fast axonal transport (Fig. 5), and cytoplasmic dynein, a motor protein active in retrograde fast axonal transport (Fig. 6).
Initially, kinesin was distributed throughout the neuronal network (Fig. 5, A and S). After addition of activated microgliaconditioned medium, however, kinesin strongly accumulated along the neurites in a pattern consistent with aggregation. These accumulations colocalized with sites of neuritic beading (Fig. 5, B-F and T-X and Y). The immunoreactivity of kinesin in the cell body also decreased with time.
Like kinesin, cytoplasmic dynein was initially distributed throughout the neuronal network (Fig. 6, A and S). After the addition of activated microglia-conditioned medium, cytoplasmic dynein immunoreactivity decreased gradually in the neurites. Cytoplasmic dynein accumulated in an aggregate pattern along the neurites to a lesser extent than kinesin with similar kinetics. These sites also colocalized with neuritic beads (Fig. 6, B-F, T-X,  and Y). The immunoreactivities of cytoplasmic dynein in the proximal neurite and cell body were relatively unaffected.
Taken together, these data suggest that activated microglia impair kinesin function to a greater extent than cytoplasmic dynein. This implies that microglia selectively inhibit fast axonal transport, and we further investigated this hypothesis.
Activated Microglia Inhibit Fast Axonal Transport-Timelapse phase-contrast imaging revealed that activated microgliaconditioned medium inhibited fast axonal transport (Fig. 7). Additionally, the decreased levels of intracellular ATP induced by activated microglia inhibited axonal transport. The inhibition of transport led to the focal accumulations of motor proteins and cytoskeletal proteins such as kinesin, cytoplasmic dynein, tubulin, NF, and MAPs, that become apparent as neuritic beads.
Interestingly, retrograde transport was affected to a lesser extent (1-3 h) than antegrade transport during the early phases of this response; this was consistent with the immunocytochemical observations of kinesin and cytoplasmic dynein described above. We further hypothesize that the asymmetric inhibition of axonal transport leads to the accumulation of mitochondria at the neuronal cell body that is observed after addition of activated microglia-conditioned medium.
Activated Microglial Neurotoxicity Is Primarily Mediated by Released Glutamate through NMDA Receptor Signaling-Activated microglia-conditioned medium had high concentrations of NO (Fig. 8A) and glutamate (Fig. 8B). In addition to NO and glutamate, activated microglia/macrophage release inflammatory cytokines such as interleukin-1␤, interleukin-6, interferon-␥, and TNF-␣ (42, 50 -54). We confirmed that recombinant cytokines did not induce significant cell death except for TNF-␣. 2 Moreover, recent studies reported that neuronal death in neurodegenerative diseases might be a non-apoptotic but FIG. 6. Collapsed cytoplasmic dynein accumulates in neuritic beads. Time course images were taken under a confocal laser-scanning microscope. Neurons were stained with mouse monoclonal anti-cytoplasmic dynein antibody (green, A-F), mouse monoclonal anti-phosphorylated NF (p-NF) antibody (SMI31, red, G-L), and mouse monoclonal anti-MAP2 antibody (blue, M-R). Y is an enlargement of V. Cytoplasmic dynein was first distributed throughout the neuronal network, then its immunoreactivity in the neurites decreased gradually. Simultaneously, a mild accumulation of cytoplasmic dynein in aggregates was observed along the neurites, and this colocalized with neuritic beads (T-X and arrows in Y). Its immunoreactivity in the cell body was relatively spared. Scale bar, 10 m. caspase-dependent form of programmed cell death (44,(55)(56)(57). Thus, we evaluated the effects of drugs that affect NO, glutamate, TNF-␣, and caspases.
NMDA induced neuritic beading and subsequent neuronal death markedly (Fig. 8, C and D, NMDA). Complete inhibition of microglial NO release did not prevent neuritic beading and neuronal death (Fig. 8, C and D, Mi L-NMMA). In contrast, inhibiting endogenous NO production in neurons partially reduced neuritic beading and cell death (Fig. 8, C and D, Neu L-NMMA). Blockage of NMDA receptors with MK801 dramatically ameliorated neuritic beading and cell death (Figs. 8, C and D, and 9C). No effect was seen when cells were incubated with a non-NMDA inhibitor (Fig. 8, C and D, CNQX). Ca 2ϩ influx through glutamate receptors is thought to lead to excitotoxic neuronal death (47,48,58,59). We evaluated the effect of the Ca 2ϩ chelator EDTA on neurite viability, and, surprisingly, it increased cell death (Fig. 8, C and D). Neither TNF-␣ neutralization (Fig. 8, C and D, anti-TNF) nor a pan-caspase inhibitor (Fig. 8, C and D, zVAD-fmk) affected neuritic beading or cell death. These data suggest that a programmed cell death pathway is not responsible for neuritic beading and neuronal death induced by activated microglia; this is consistent with the TUNEL assay data shown in Fig. 2B. We next assessed the effects of drugs that stabilize cytoskeletal protein polymerization on cell death, but both Taxol and phalloidin increased cell death (Fig. 8, C and D).
Blockage of NMDA Receptors Completely Rescues Neurons from Activated Microglial Neurotoxicity-Because MK801 nearly completely ameliorated neuritic beading and neuronal death induced by activated microglia, we assessed its effects on mitochondrial function and axonal transport. MK801 rescued neurons from mitochondrial impairment (Fig. 9D), intracellular ATP decreases (Fig. 9E), and axonal transport damage (Fig.  9F) induced by activated microglia. These data clearly link activated microglial neurotoxicity, including neuritic beading and cell death, to glutamate excitotoxicity downstream of NMDA receptor signaling. DISCUSSION Here we provide evidence that neuritic beading induced by activated microglia correlates with neuronal cell dysfunction and precedes neuronal death. Additionally, microglial neurotoxicity is primarily mediated by NMDA receptor signaling following ligation of released glutamate. Neuritic bead formation was accompanied by a drop in intracellular ATP levels, and it preceded neuronal death. The observed drop in intracellular ATP levels was because of inhibition of the mitochondrial respiratory chain complex IV activity, and the loss of intracellular energy pools negatively affected neuronal transport. Impaired transport caused cytoskeletal and motor protein accumulation at sites of neuritic beading. Finally, blockage of NMDA receptors abrogated all the observed signs of microglial neurotoxicity including neuritic bead formation, mitochondrial impairment, neuronal transport damage, and subsequent neuronal death. Taken together, the neuritic beads induced by activated microglia are thought to consist of the residual cargo of neuronal transport vesicles that accumulate following energy starvation downstream of NMDA receptor signaling. Furthermore, our data demonstrate that neurons do not undergo apoptosis or another form of caspase-dependent cell death.
Neuritic beading is observed in various pathological conditions (23)(24)(25)(26)(27)(28)(29)(30)(31). Our current results indicate that neuritic beads represent the accumulation of vesicular cargo and transport proteins following damage to the transport mechanisms by activated microglia. Additionally, the neuritic beading observed in diseased brain tissue likely arises during a period of relative cellular energy deficit. Activated microglia release large amounts of glutamate. Following glutamate ligation of the NMDA receptor, neuronal mitochondrial respiration was inhibited and this led to neuronal energy loss. Neuronal transport is an extremely energy intense process, and we observed gross defects in the activity of microtubule-associated ATPase motor proteins such as kinesin and cytoplasmic dynein. These motor proteins, as well as cytoskeletal proteins and organelles accumulated along the neurites because of impaired neuronal transport, and these accumulations formed neuritic beads. Time-lapse imaging showed that particles sequestrated in neuritic beads were rotating like a top inside the bead (data not shown). Interestingly, retrograde transport was affected to a lesser extent than antegrade transport during the early phase after treatment (Fig. 8). This was consistent with the immunocytochemical observations of kinesin and cytoplasmic dynein (Figs. 6 and 7). The mitochondria accumulated at the neuronal cell body and disappeared from distal neurites after the addition of activated microglia-conditioned medium (Fig. 5). During periods of energy stress, neurons retrieve mitochondria from the distal neurites to the cell body, and this process relies on retrograde axonal transport (60,61). Thus, the preferential maintenance of retrograde transport following exposure to activated microglial-conditioned medium is consistent with this hypothesis. The morphology of mitochondria changed from a granular pattern to a tubular and reticular pattern as they FIG. 9. Blockage of NMDA receptors completely rescues neurons from activated microglial neurotoxicity. A-C, images were taken under a phase-contrast microscope. Scale bar, 10 m. A, control neurons. B, neurons incubated with activated microglia-conditioned medium for 6 h. C, neurons incubated with activated microglia-conditioned medium including MK801 for 6 h. D, MTS assay. •, neurons with activated microglia-conditioned medium; E, neurons incubated with activated microglia-conditioned medium including MK801. E, intracellular ATP levels. •, neurons with activated microglia-conditioned medium; E, neurons incubated with activated microglia-conditioned medium including MK801. F, assessment of fast axonal transport. •, antegrade transport in neurons with activated microglia-conditioned medium; f, retrograde transport in neurons incubated with activated microglia-conditioned medium; E, antegrade transport in neurons incubated with activated microglia-conditioned medium including MK801; Ⅺ, retrograde transport in neurons incubated with activated microgliaconditioned medium including MK801. *, p Ͻ 0.05 versus control. †, p Ͻ 0.05 versus antegrade transport in neurons with activated microglia-conditioned medium. Values are mean Ϯ S.D. accumulated at the neuronal cell body (Fig. 5). Alterations in mitochondrial shape and structure are known to correlate with increasing energy demands, and mitochondria with this morphology are frequently found in tumor cells (49). The changes in mitochondrial morphology induced by activated microglia are further evidence of the neuronal energy deficit produced by these cells.
Glutamate is known to cause neuronal damage through excitotoxicity (8,9). In particular, the NMDA receptor pathway has been extensively studied in this regard. According to previous studies, a Ca 2ϩ influx through NMDA receptor activates neuronal NOS and endogenous NO elevation leads to mitochondrial respiratory inhibition and direct neurotoxicity (58,59). We confirmed that a high dose of NMDA decreased respiratory chain complex IV activity and induced neuritic beading followed by neuronal death (Figs. 2F and 8, A-D). In our study, the NMDA receptor antagonist MK801 nearly completely blocked the neurotoxicity induced by activated microglia (Fig. 9). However, the NOS inhibitor L-NMMA only partially prevented microglial neurotoxicity despite its complete inhibition of NOS (Fig. 9). Direct inhibition of respiratory chain complex IV activity by NaN 3 also induced neuritic beading and subsequent neuronal death (data not shown), which was in accordance with a previous report (62). Taken together, in addition to NO, another unknown substance induced downstream of NMDA receptor signaling may participate in the inhibition of respiratory chain complex IV activity. Further investigations are needed to clarify this.
Controversy surrounds the issue of whether excitotoxic neuronal cell death through NMDA receptor is mediated by Ca 2ϩ (47,48,58,59) or Na ϩ (34,36). In our study, the Ca 2ϩ chelator EDTA significantly increased neuronal death (Fig. 9). Clearly, EDTA might disturb intracellular Ca 2ϩ homeostasis, and more thorough studies are needed to address this question. The present study did not provide any positive evidence that Ca 2ϩ influx mediates the neuronal death signal through the NMDA receptor.
Previous reports suggested that a caspase-dependent nonapoptotic form of cell death took place in neurodegenerative diseases (51,52). However, treatment with the broad caspase inhibitor z-VAD-fmk did not affect neuronal death in our study (Fig. 9). We propose that excitotoxic neuronal death through the NMDA receptor is much more like necrosis, not a programmed cell death.
In this study, we assessed the effect of drugs that stabilize cytoskeletal protein polymerization. However, both the tubulin polymerization stabilizer Taxol and the actin polymerization stabilizer phalloidin increased neuronal death (Fig. 9). These observations implied that collapse of the cytoskeleton per se was only a feature of neuronal cell dysfunction, not a cause of neuronal death. Our study suggested that either blockage of NMDA receptors or an increase in intracellular ATP levels was needed to avoid neuronal cell death induced by activate microglia.
In conclusion, we demonstrated that neuritic beading induced by activated microglia was a feature of neuronal cell dysfunction toward neuronal death downstream of NMDA receptor signaling. Blockage of NMDA receptor may be an effective strategy for the treatment of neurodegenerative diseases including Alzheimers disease, Parkinsons disease, amyotrophic lateral sclerosis, and multiple sclerosis.