Glutamate Release from Astrocytes in Physiological Conditions and in Neurodegenerative Disorders Characterized by Neuroinflammation

GLUTAMATE RELEASE FROM ASTROCYTES IN PHYSIOLOGICAL CONDITIONS AND IN NEURODEGENERATIVE DISORDERS CHARACTERIZED BY NEUROINFLAMMATION

Sabino Vesce,* Daniela Rossi,y Liliana Brambilla,y and Andrea Volterra*,y *Department of Cell Biology and Morphology, University of Lausanne Rue du Bugnon 9, 1005 Lausanne, Switzerland y Department of Pharmacological Sciences Center of Excellence on Neurodegenerative Diseases University of Milan, Via Balzaretti 9, 20133 Milan, Italy

I. Introduction II. Ca2+-Dependent Glutamate Release from Astrocytes III. Excitotoxicity Involving Ca2+-Dependent Glutamate Release from Astrocytes in Pathological Conditions: The Case of ADC IV. Astrocytic Alterations and Ca2+-Dependent Glutamate Release Dysfunction in AD V. Conclusions References

Although glial cells have been traditionally viewed as supportive partners of neurons, studies of the last 20 years demonstrate that astrocytes possess functional receptors for neurotransmitters and other signaling molecules and respond to their stimulation via release of chemical transmitters (called gliotransmitters) such as glutamate, ATP, and D-serine. Notably, astrocytes react to synaptically released neurotransmitters with intracellular calcium ([Ca2þ]i) elevations, which result in the release of glutamate via regulated exocytosis and possibly other mechanisms. These findings have led to a new concept of neuron–glia intercommunication where astrocytes play an unsuspected dynamic role by integrating neuronal inputs and modulating synaptic activity. The additional discovery that glutamate release from astrocytes is controlled by molecules linked to inflammatory reactions, such as the cytokine tumor necrosis factor- (TNF-) and prostaglandins, suggests that glia-to-neuron signaling may be sensitive to changes in production of these mediators in pathological conditions. Indeed, a local, parenchymal brain inflammatory reaction (neuroinflammation) characterized by astrocytic and microglial activation has been reported in several neurodegenerative disorders, including Alzheimer’s disease and AIDS dementia complex. This transition to a reactive state may be accompanied by a disruption of the cross talk normally occurring between astrocytes and neurons and so contribute to disease development. INTERNATIONAL REVIEW OF NEUROBIOLOGY, VOL. 82 DOI: 10.1016/S0074-7742(07)82003-4

57

Copyright 2007, Elsevier Inc. All rights reserved. 0074-7742/07 $35.00

58

VESCE et al.

The findings reported in this chapter suggest that a better comprehension of the glutamatergic interplay between neurons and glia may provide information about normal brain function and also highlight possible molecular targets for therapeutic interventions in pathology.

I. Introduction

Glia are a group of cells largely represented in the central nervous system (CNS). Based on their diverse morphology and function they can be distinguished in various subclasses, including oligodendrocytes, responsible for axon myelination; microglia, immunocompetent, specialized brain macrophages; and astrocytes, which represent the major CNS population of glial cells and perform multiple tasks owing to their strategic anatomical location between neurons and blood vessels (Kettenmann and Ransom, 1995). On the one hand, astrocytes are intimately associated with neurons: in the hippocampus, for example, their processes tightly enwrap 50% of the synapses (Pfrieger and Barres, 1996; Ventura and Harris, 1999). This close interaction allows astrocytes to provide nerve cells with structural, metabolic, and trophic support. On the other hand, glucose, the main source of energy for nerve cells, enters the brain parenchyma from the cerebral circulation via uptake by the astrocytes. In the astrocytes, glucose is either stored in the form of glycogen or enters glycolysis to produce lactate, a major metabolic substrate for neurons (Tsacopoulos and Magistretti, 1996). By means of permeable channels and active pumps located on the plasma membrane, astrocytes play also a critical role in maintaining ionic homeostasis. For instance, they maintain the extracellular potassium concentration within the physiological range (Karwoski et al., 1989), which is critical for assuring normal neuronal excitability. Excess extracellular potassium may indeed result in neuronal depolarization and eventually cause action potential blockage. Moreover, astrocytes control the extracellular concentration of synaptically released neurotransmitters by means of specific plasma membrane transporter proteins (Coco et al., 1997; Dehnes et al., 1998). At glutamatergic synapses, glutamate uptake from astrocytes is the main mechanism for removing the neurotransmitter from the extracellular compartment. Because of their established functions of energy suppliers for neurons and regulators of the extracellular brain fluid composition, for decades astrocytes have been regarded solely as passive supportive elements maintaining the optimal neuronal microenvironment but devoid of any direct role in brain communication. However, this vision has been challenged by studies of the last 20 years showing that astrocytic cells modulate synaptic activity and actively participate to

GLUTAMATE RELEASE FROM ASTROCYTES

59

CNS function and formation. One initial puzzling discovery was that astrocytes express a large repertoire of receptors for neurotransmitters, often mirroring the ones seen in the neighboring synapses. Subsequently, it was found that such receptors can be activated by the spillover of neurotransmitters during synaptic activity and induce [Ca2þ]i elevations in the astrocytes. This event can in turn drive the release of gliotransmitters such as glutamate from the astrocytes and trigger intercellular communication, including astrocyte-to-neuron signaling (Araque et al., 1999). Astrocyte-released gliotransmitters are able to activate neuronal receptors and thereby modify the neuronal electrical excitability and synaptic transmission (Schipke and Kettenmann, 2004; Volterra and Meldolesi, 2005). The amino acid glutamate can be released from astrocytes through multiple pathways that are activated under diVerent conditions, at diVerent loci and/or with diVerent modalities. Both Ca2þ-dependent and Ca2þ-independent processes have been described. Ca2þ-independent ones include reversed operation of reuptake carriers, notably under ischemic conditions (Rossi et al., 2000), exchange with cystine, the essential substrate for astrocytic production of glutathione, mediated by specific cystine-glutamate antiporters (Warr et al., 1999), and molecular permeation of large pore channels, including P2X7 receptors (Duan et al., 2003), gap-junction hemichannels (Ye et al., 2002), and volume-sensitive organic anion channels (Kimelberg et al., 1990), although for the latter a Ca2þ-dependent mechanism has also been described (Takano et al., 2005). In this chapter, we will focus on the mechanism(s) underlying astrocytic Ca2þ-dependent glutamate release in physiological circumstances and discuss the implications that alterations of this mechanism may have in neuroinflammatory and degenerative processes, notably in AIDS dementia complex (ADC) and Alzheimer’s disease (AD).

II. Ca2+-Dependent Glutamate Release from Astrocytes

The initial demonstration that astrocytes signal to neurons in response to physiological stimuli by Ca2þ-dependent glutamate release came in 1994 by the group of Philip Haydon. The authors described that stimulation of mixed neuron–glia cultures with the peptide bradykinin triggered [Ca2þ]i rises and glutamate release from astrocytes. In addition, they observed [Ca2þ]i elevations in neurons following those in astrocytes and showed that neuronal [Ca2þ]i elevations were mediated by astrocyte-released glutamate (Parpura et al., 1994). Support to these first in vitro studies and to the hypothesis that a bidirectional neuron–glia signaling exists in intact brain tissue was subsequently provided by Pasti et al. (1997). These authors showed that stimulation of neuronal aVerents in acute cortical and hippocampal slices induced [Ca2þ]i oscillations in surrounding astrocytes. Synaptically released glutamate seemed the most likely initiator of such

60

VESCE et al.

events via stimulation of metabotropic receptors (mGluRs) located in the astrocytic plasma membrane. [Ca2þ]i oscillations in astrocytes were followed by [Ca2þ]i rises in the surrounding neurons, possibly as a result of glutamate release from astrocytes. Indeed, a parallel study by our group (Bezzi et al., 1998) provided the direct demonstration that activation of astrocyte mGluRs (and even more eVectively the combined activation of AMPA/kainate and mGluRs) triggers glutamate release from astrocytes via a Ca2þ-dependent mechanism involving prostaglandins. In addition, such glutamate- and prostaglandin-dependent glutamate release from astrocytes in the hippocampal stratum radiatum induced GluR-dependent [Ca2þ]i responses in neighboring pyramidal neurons, definitively proving the existence of a glutamatergic neuron-astrocyte cross talk in situ. Interestingly, glutamate-induced glutamate release from astrocytes was occlusive with the release (Ca2þ-dependent) triggered by bradykinin, but additive to the release (Ca2þ-independent) mediated by glutamate transporters, suggesting that distinct astrocyte neuroligand receptors are coupled to a common Ca2þ-dependent transduction pathway ultimately leading to glutamate release; and that Ca2þ-dependent and Ca2þ-independent release mechanisms probably use diVerent glial glutamate pools. A wide range of more recent studies have confirmed these initial observations, reinforcing the view that the Ca2þ-dependent pathway is triggered by activation of G-protein–coupled receptors (GPCR) and mediated by inositol 1,4,5-triphosphate-induced Ca2þ release from stores of the endoplasmic reticulum (Bezzi et al., 2001; Jeremic et al., 2001; Kang et al., 2005; Sanzgiri et al., 1999; Takano et al., 2005). However, the ultimate process by which glutamate is released, whether regulated exocytosis or permeation via volume-regulated anion channels, has been more debated (Nedergaard et al., 2003). The evidence for the existence of a regulated exocytosis pathway in astrocytes is now overwhelming. Notably, our group has provided the first ultrastructural demonstration that astrocytes in adult brain tissue contain a vesicular compartment whose vesicles strictly resemble microvesicles of glutamatergic nerve terminals. These synaptic-like microvesicles (SLMVs), found in perisynaptic astrocytic processes, express vesicular glutamate transporters (VGLUT1/2) and vesicular SNARE proteins such as cellubrevin, indicating that they are equipped for taking up, storing, and releasing glutamate (Bezzi et al., 2004). By complementing the ultrastructural studies in situ with dynamic total internal reflection fluorescence (TIRF) real-time imaging studies in cultured astrocytes, we could directly document individual vesicle fusion events accompanied by glutamate release in response to mGluR stimulation (Bezzi et al., 2004). Such exocytic fusions occurred within tens of milliseconds from GPCR stimulation and were abolished by a classical blocker of exocytosis, tetanus neurotoxin. The presence of regulated exocytosis in astrocytes has been now documented with a wide range of experimental approaches, including optical detection techniques diVerent from TIRF

GLUTAMATE RELEASE FROM ASTROCYTES

61

(Crippa et al., 2006; Kang et al., 2005), membrane capacitance measurements (Kreft et al., 2004; Zhang et al., 2004a), electrochemical amperometry (Chen et al., 2005), as well as by selective interference with proteins of the exocytotic machinery (Montana et al., 2004; Zhang et al., 2004b). Although astrocytes express the machinery for regulated exocytosis, several relevant diVerences have emerged in terms of protein composition, organization, and kinetics with respect to synaptic exocytosis. Molecularly, four vesicular proteins (SNAP23, complexin 2, Munc18a, and synaptotagmin IV) are certainly expressed in astrocytes, while other exocytotic proteins typically expressed in synapses have not been detected (for a review, see Volterra and Meldolesi, 2005). Although similar in size to their neuronal counterparts, astrocytic vesicles are less densely packed and not as orderly arranged. However, they are frequently observed at sites adjacent to neuronal structures that bear glutamate receptors. The exocytotic event itself appears to be relatively slow in astrocytes, about 100 times slower than at synapses (Kreft et al., 2004). Such property, however, is justified by the slower stimulus-secretion coupling mechanism via GPCR signaling and Ca2þ release from the internal stores, and may well fit the modulatory role that astrocytes are supposed to play in synaptic function. In addition to mGluRs, other GPCRs trigger Ca2þ-dependent glutamate exocytosis from astrocytes. In a study, Domercq et al. (2006) documented an exocytic response to stimulation of purinergic P2Y1 receptors (P2Y1R, activated by ATP in physiological conditions). In complement to TIRF imaging evidence in cultured astrocytes, the authors showed that P2Y1R stimulation induces Ca2þdependent glutamate release in hippocampal slices (but not in hippocampal synaptosomes), and that the release in situ shares the same pharmacological properties with the P2Y1R-evoked release in cultured astrocytes. Notably, the process in situ is sensitive to the exocytosis blocker, bafilomycin A1 (Baf A1). The rate by which Baf A1 inhibits P2Y1R-dependent glutamate release is slower than that by which the drug acts on high Kþ-evoked neuronal exocytosis, suggesting that the P2Y1R-evoked process is distinct from the one in nerve terminals. Also, Takano et al. (2005) found that stimulation of metabotropic P2Y receptors evokes Ca2þ-dependent glutamate release in cultured astrocytes. These authors, however, reported the release to be insensitive to blockers of exocytosis and sensitive to inhibitors of volume-regulated anion channels. According to the purinergic receptor pharmacology for this release, ATP activates P2Y2 or P2Y4 receptors, not P2Y1R. Why the same endogenous ligand activates distinct P2Y receptor subtypes in the two studies remains to be established. One possibility is that astrocytes express diVerent P2YR subtypes in diVerent culture conditions. In this context, it remains to be confirmed that the mechanism described by Takano et al. takes place in situ. If so, one can hypothesize that ATP leads to glutamate release from astrocytes via activation of multiple P2YR subtypes coupled to distinct Ca2þ-dependent intracellular pathways.

62

VESCE et al.

Shedding of the cytokine tumor necrosis factor- (TNF-) as well as formation of prostaglandin E2 (PGE2) (see next paragraph) are apparently primordial signaling events accompanying ATP-induced glutamate exocytosis. Thus, this release is strongly diminished in the presence of pharmacological inhibitors of TNF- and prostaglandins, and the same occurs in cultures of astrocytes from TNF- knockout mice (Domercq et al., 2006). The relatively fast kinetics of glutamate secretion from astrocytes indicates that the production of TNF- and PGE2 should occur in the timescale of milliseconds following P2Y1R activation. This is perhaps possible according to observations (Goddard et al., 2001; Zonta et al., 2003), but remains to be directly demonstrated. Alternatively, a tonic basal release of TNF- and PGE2 from astrocytes might appropriately tune the secretory apparatus for eYcient glutamate liberation. Moreover, both TNF- and PGE2 are connected to [Ca2þ]i elevations in astrocytes (Bezzi et al., 1998; Domercq et al., 2006) and could boost the intracellular Ca2þ response initiated by P2Y1R activation.

III. Excitotoxicity Involving Ca2+-Dependent Glutamate Release from Astrocytes in Pathological Conditions: The Case of ADC

Several lines of evidence indicate the presence of a local, parenchymal inflammatory response in a number of chronic neurodegenerative disorders. Such condition is characterized by specific morphological and functional changes of astrocytes and microglia, broadly defined as ‘‘reactive gliosis.’’ The signals exchanged between the two glial cell types during these events are largely unknown, yet their transition from the resting to the activated state appears to be associated with a marked upregulation of several genes and the secretion of factors like cytokines, eicosanoids, reactive oxygen species, nitric oxide, and excitatory amino acids (Perry et al., 1995). Although the inflammatory glial reaction was originally thought to be mostly beneficial for tissue-repairing processes, in several cases it may actually contribute to the exacerbation of neurodamaging processes. In view of the control exerted by prostaglandins and TNF- on Ca2þ-dependent glutamate release from astrocytes, overproduction of these mediators during neuroinflammation might favor an increased and deleterious glutamatergic input from astrocytes to neurons. This hypothesis is substantiated by direct experimental evidence. Thus, our group discovered that, in addition to classical transmitters, such as glutamate or ATP, the chemotactic cytokine (chemokine) stromal-derived factor1 (SDF-1) also induces Ca2þ-dependent glutamate release from astrocytes (Bezzi et al., 2001). SDF-1 acts via stimulation of its specific GPCR, CXCR4. The process is accompanied by shedding of TNF- which, once released from

63

GLUTAMATE RELEASE FROM ASTROCYTES

astrocytes, acts autocrinally or paracrinally on its p55, TNF receptor type-1 (TNFR1). In turn, TNFR1 activation controls downstream signaling events, including PGE2 production, [Ca2þ]i elevation, and glutamate release. The latter is blocked by the exocytosis blockers Baf A1 and tetanus neurotoxin (Fig. 1A). Importantly, we found that CXCR4 is expressed also in microglia and that

A Normal brain

B AIDS - related neuropathology Activated microglia

Resting microglia 1 gp120

SDF-1a

1

1

TNF-a

TNF-a 3

2

3

2

Glu Astrocyte-neuron signaling

Astrocyte

Glu

Astrocyte

Excitotoxicity

FIG. 1. Microglia-dependent transformation of CXCR4-evoked glutamate release from astrocytes into an excitotoxic pathway. (A) In physiological conditions, stimulation of the CXCR4 receptor by the endogenous ligand SDF-1 is followed by a sequence of intracellular events that triggers the shedding of TNF- from its membrane-bound precursor (1). Once released, the cytokine acts as an autocrine/ paracrine factor and stimulates TNFR1 in the same and/or in surrounding astrocytes (2). The interaction of TNF- with its cell surface receptor initiates additional intracellular signaling events, including prostaglandin production, which act in parallel or sequence with the initial events triggered by CXCR4 stimulation, to eventually lead to Ca2þ-dependent glutamate (Glu) release (3). This, in turn, may have modulatory eVects on synaptic activity. Microglia are in the resting state and do not participate in astrocyte-neuron signaling. (B) In AIDS-related neuropathology, microglia infected by HIV become reactive and shed viral particles, including the envelope glycoprotein gp120. This can act as an agonist on CXCR4 receptors that are present in astrocytes and also in the reactive microglia itself, thus inducing a considerably higher TNF- release compared to physiological conditions (1). The enhanced activation of TNFR1 receptors in astrocytes (2), results in a massive and ultimately excitotoxic release of glutamate (3).

64

VESCE et al.

stimulation of CXCR4 in these cells results in potent TNF- release, several-fold higher than in astrocytes, but occurring only when microglia are in a reactive state. Stimulation of microglial CXCR4 does not induce glutamate release. In the pathological brain, astrocytes and microglia often form local foci of reactive cells around the sites of lesion or infection. We mimicked this condition by adding LPSactivated microglia to astrocyte-pure cultures in about the same ratio (1:10) existing in the brain. Simultaneous stimulation of CXCR4 in contiguous reactive microglia and astrocytes resulted in a dramatic amplification of TNF- release from both cells and, as a consequence, in strong potentiation of Ca2þ-dependent glutamate release from the astrocytes. In separate experiments in hippocampal cultures containing neurons, astrocytes, and microglia, we could demonstrate that the microglia- and TNF--dependent potentiation of astrocyte glutamate release had excitotoxic consequences, inducing slow apoptotic death of neuronal subpopulations (Fig. 1B). Not only identification of the above CXCR4-dependent signaling cascade described a novel role for TNF- in rapid glial communication, but also provided a novel mechanistic hypothesis (Bezzi et al., 2001) for the neurodegenerative processes underlying HIV-associated neuropathology, notably the ADC (Kaul et al., 2001). Previous evidence suggested that an increasing number of neurons die by apoptosis (Bagetta et al., 1996; Shi et al., 1996) during progression of the disease, apparently via NMDA-mediated excitotoxicity. However, microglial cells, not neurons, are infected by penetration of the HIV-1 virus into the brain and so neuronal death is thought to occur via interactions with these cells, as well as with the astrocytes (Kaul et al., 2001; Meucci and Miller, 1996; Toggas et al., 1996). In this context, viral particles, including the HIV-1 coat protein gp120, shedded from infected cells, may play a key role. Indeed, the sole expression of gp120 in transgenic mice was found to reproduce several features of the ADC neuropathology (Toggas et al., 1994). We discovered that a gp120 isoform, gp120IIIB, derived from the T-tropic HIV-1 IIIB strain, acts as CXCR4 agonist in both astrocytes and microglia and, similar to the endogenous ligand SDF-1, triggers potent TNF--dependent glutamate release from astrocytes. Since production and shedding of viral proteins in HIV-1-infected brains is an uncontrolled process, it is not unlikely that a consequence of it is the overstimulation of glial CXCR4 with a consequent excessive glutamate release from astrocytes eventually triggering excitotoxic neuronal cell death (Fig. 1B). Indeed, in our in vitro model, we obtain neuroprotection by blocking the astrocyte signaling cascade at any of the identified steps: CXCR4 receptor activation, TNF- release, prostaglandin formation, and, finally, by scavenging released glutamate or blocking NMDA receptors (Bezzi et al., 2001). Antiretroviral therapy does not fully control AIDS-associated neuropathology while significantly prolonging the life expectancy of AIDS-aVected subjects.

GLUTAMATE RELEASE FROM ASTROCYTES

65

Therefore, there is a need for drugs limiting the development of ADC and ensuring a better quality of life. Clinical trials have been conducted with memantine, an inhibitor of the NMDA receptor channels that might prove helpful in containing excitotoxicity. Our work suggests that molecules acting at other levels along the toxic pathway could be also beneficial.

IV. Astrocytic Alterations and Ca2+-Dependent Glutamate Release Dysfunction in AD

AD is a neurodegenerative disorder characterized by the progressive loss of cognitive function. One of its best-known biochemical hallmarks is the accumulation of the amyloid- (A ) protein into amyloid plaques that can be found in the extracellular space of forebrain regions (Glenner and Wong, 1984). Findings suggest that astrocytes may play an important role in the clearance of the A peptide and thus in preventing plaque formation (Wyss-Coray et al., 2003). By incubating astrocytes with brain sections from transgenic mice expressing amyloid precursor protein, the authors showed that adult astrocytes migrate in response to chemokines present in AD lesions and are able to bind, internalize, and degrade A . The mechanism of digestion is unclear, but could involve matrix metalloproteinase-9, a protease able to digest both soluble and fibrillary A and expressed by astrocytes found at the border of amyloid plaques (Yan et al., 2006). Interestingly, this protective action by astrocytes could be lost in AD, since endogenous astrocytes surround and contact plaques in AD brains, but appear unable to remove A (Wyss-Coray et al., 2003). Deficits in the astrocytic clearance of A may therefore contribute to amyloid plaque formation. One of the possible causes is a defect in apolipoprotein E (Koistinaho et al., 2004) that, in the mammalian brain, is mainly produced by astrocytes and represents a recognized genetic risk factor for AD. There are other indications that astrocytes are perturbed by the presence of A . For example, the administration of A to mixed cultures of hippocampal neurons and astrocytes causes abnormal [Ca2þ]i transients and mitochondrial depolarization in astrocytes long before any impairment is visible in neurons. Therefore, the subsequent neuronal death might be the result of oxidative stress generated by the astrocytic dysfunction (Abramov et al., 2004). Thus, the fate of neurons in AD may be closely dependent on the maintenance of normal astrocytic function. Chronic inflammatory glial cell reaction is well documented around A plaques (for a review, see Wyss-Coray, 2006). While it is not completely understood whether inflammation plays a causative role in AD, recent data from animal models suggest that some inflammatory processes might at least accelerate the development of the disease. For instance, -secretase, the A -generating enzyme, was found to be present in reactive astrocytes from aged Tg2576 mice

66

VESCE et al.

(Hartlage-Rubsamen et al., 2003), a transgenic AD mouse model expressing the double mutation K670N-M671L of the amyloid precursor protein APP. The proximity of these astrocytes to the plaques raises the possibility that they promote A accumulation. Thus, in the transition to their reactive state, astrocytes might undergo a deleterious double transformation: loss of the ability to digest A , followed by contribution to generating the toxic protein itself. High levels of proinflammatory cytokines such as interleukin-1 , -6, and TNF-, mostly produced by reactive microglia and astrocytes, are detected in the brain of AD subjects (Zhao et al., 2003). The consequences of this phenomenon are unclear, also because proinflammatory cytokines have varied eVects depending on the biological context (Wyss-Coray, 2006). However, recent work points to a toxic action of these cytokines in the neuroinflammatory reaction characterizing AD (Ralay Ranaivo et al., 2006). In this study, A directly injected in the mouse brain caused a reduction in the levels of the synaptic markers synaptophysin and postsynaptic density-95 (PSD-95), indicative of a synaptic defect. If the A treatment was followed, starting 3 weeks later, by the daily oral administration of a selective inhibitor of proinflammatory cytokine production, the levels of the synaptic markers were restored. This positive biochemical change was matched by an improved performance of the animals in a spatial learning task, suggesting that, indeed, targeting neuroinflammation might be an eVective therapeutic strategy in AD. Glial alterations, notably increased cytokine production, may also aVect the Ca2þ-dependent pathway of glutamate release from astrocytes. The observation that expression of both TNF- and TNFR1 receptors is enhanced in the brain of AD patients (Del Villar and Miller, 2004; Zhao et al., 2003) led our group to investigate possible alterations of the TNF--dependent pathway of glutamate release in PDAPP mice, a transgenic model of AD (Rossi et al., 2005). We utilized aged animals (12 months old), presenting abundant amyloid plaque deposition and reactive gliosis in the forebrain, and adult animals (4 months old) with little or no amyloid deposits. Ca2þ-dependent glutamate release was stimulated in brain slices from PDAPP animals by direct TNF- application. Our data indicate that the release does not originate from nerve terminals and that its pharmacological characteristics are identical to those of the process evoked by TNF- in cultured astrocytes (Bezzi et al., 2001). Contrary to our expectations and the findings in Bezzi et al. (2001), we detected a considerable reduction (not increase) of TNF-evoked glutamate release in hippocampal slices from aged PDAPP animals compared to adult animals and age-matched controls (Rossi et al., 2005). The defect was region-selective as the glutamate release response from cerebellar slices of aged PDAPP mice was identical to that of controls. Interestingly, the secretory process itself appeared intact, since stimulation with PGE2, which acts downstream of TNF-, evoked normal glutamate release. Therefore, the alteration must take place at the level of the stimulus-secretion coupling mechanism.

GLUTAMATE RELEASE FROM ASTROCYTES

67

An important intracellular mediator of TNF- is the protein DENN/MADD, which binds TNFR1 and triggers activation of multiple downstream signaling pathways, including cytosolic phospholipase A2, thus leading to arachidonic acid release and prostaglandin production. Importantly, a reduced expression of DENN/MADD was reported in the hippocampi from Alzheimer’s patients (Del Villar and Miller, 2004). We found that DENN/MADD is defective also in the hippocampus (not cerebellum) of aged PDAPP mice (Rossi et al., 2005), suggesting that defect of DENN/MADD expression may account for reduced glutamate release in response to TNF- stimulation. Downregulation of DENN/MADD may result from the chronic inflammation that characterizes the slow progression of AD and might be a consequence of long-term overstimulation of TNFR1 by excessive TNF-. This condition is very diVerent from the in vitro model of acute inflammation that we utilized in the studies concerning CXCR4-evoked glutamate release. In that case, we tested glutamate release within 24 h from adding reactive microglia to astrocyte cultures. It is therefore likely that acute and chronic TNF- overproduction cause opposite alterations of glutamate release from astrocytes. Our present evidence in aged PDAPP mice cannot identify whether the described impairment has a pathogenetic role in AD. Therefore, at present, we can only speculate on the functional significance of these findings. Astrocytes are known to exert a critical control on the formation, maintenance, and strength of synapses by actively releasing soluble factors (Beattie et al., 2002; Christopherson et al., 2005; Mauch et al., 2001; Ullian et al., 2001). Notably, astrocyte-released TNF- controls the strength of excitatory synapses in the hippocampus and thus is crucial for the stability and optimal performance of neuronal networks (Beattie et al., 2002; Stellwagen and Malenka, 2006). In addition, Ca2þ-dependent glutamate release from hippocampal astrocytes modulates neuronal excitability, synchronicity, and synaptic transmission (Angulo et al., 2004; Fellin et al., 2004; Fiacco and McCarthy, 2004; Kang et al., 1998; Liu et al., 2004; Parri et al., 2001; Perea and Araque, 2005). Synaptic failure in AD, probably dependent on the early formation of A oligomers in the extracellular milieu, becomes apparent well before aggregation into plaques occurs and neuronal degeneration is detectable (Selkoe, 2002). Thus, although more evidence is needed, it is possible that the disruption of TNF--mediated glutamate release from astrocytes participates in the progressive reduction of synaptic eYcacy underlying cognitive decline in AD.

V. Conclusions

The discovery that astrocytes are active partners of neurons in brain communication and possess regulated forms of transmitter release is dense of implications for understanding brain processes in health and disease. In particular, several

68

VESCE et al.

recent findings indicate that release of gliotransmitters from astrocytes, such as the release of glutamate that we briefly summarized in this chapter, is implicated in the physiological modulation of synaptic functions. Alteration of this process, leading either to excessive release and excitotoxicity or to reduced release and possibly failure to integrate synaptic activity, may have serious consequences on the surrounding neuronal population. This suggests that understanding the biological basis of the transformations that astrocytes and microglia undergo during reactive gliosis and neuroinflammatory processes could lead to a significant progress toward the cure of neurodegenerative disorders. For instance, [Ca2þ]i oscillations in hippocampal astrocytes stimulate glutamate release and synchronize the activity in neighboring neurons (Angulo et al., 2004; Fellin et al., 2004; Parri et al., 2001). However, in a model of traumatic injury, these oscillations are lost in reactive astrocytes surrounding the area of lesion (Aguado et al., 2002). Ca2þ signaling alterations and alterations in Ca2þ-dependent glutamate release in reactive astrocytes appear as functionally relevant defects that should be taken into consideration when studying neurodegenerative processes. The data here reviewed call for more studies looking at glial and neuronal pathological alterations as integrated phenomena, given that glial derangements cannot be simply considered as marginal events or late reactions to neuronal injury, but rather as intrinsic components of the neurodegenerative processes.

References

Abramov, A. Y., Canevari, L., and Duchen, M. R. (2004). -amyloid peptides induce mitochondrial dysfunction and oxidative stress in astrocytes and death of neurons through activation of NADPH oxidase. J. Neurosci. 24, 565–575. Aguado, F., Espinosa-Parrilla, J. F., Carmona, M. A., and Soriano, E. (2002). Neuronal activity regulates correlated network properties of spontaneous calcium transients in astrocytes in situ. J. Neurosci. 22, 9430–9444. Angulo, M. C., Kozlov, A. S., Charpak, S., and Audinat, E. (2004). Glutamate released from glial cells synchronizes neuronal activity in the hippocampus. J. Neurosci. 24, 6920–6927. Araque, A., Parpura, V., Sanzgiri, R. P., and Haydon, P. G. (1999). Tripartite synapses: Glia, the unacknowledged partner. Trends Neurosci. 22, 208–215. Bagetta, G., Corasaniti, M. T., Malorni, W., Rainaldi, G., Berliocchi, L., Finazzi-Agro, A., and Nistico, G. (1996). The HIV-1 gp120 causes ultrastructural changes typical of apoptosis in the rat cerebral cortex. Neuroreport 7, 1722–1724. Beattie, E. C., Stellwagen, D., Morishita, W., Bresnahan, J. C., Ha, B. K., Von Zastrow, M., Beattie, M. S., and Malenka, R. C. (2002). Control of synaptic strength by glial TNF-. Science 295, 2282–2285. Bezzi, P., Carmignoto, G., Pasti, L., Vesce, S., Rossi, D., Rizzini, B. L., Pozzan, T., and Volterra, A. (1998). Prostaglandins stimulate calcium-dependent glutamate release in astrocytes. Nature 391, 281–285.

GLUTAMATE RELEASE FROM ASTROCYTES

69

Bezzi, P., Domercq, M., Brambilla, L., Galli, R., Schols, D., De Clercq, E., Vescovi, A., Bagetta, G., Kollias, G., Meldolesi, J., and Volterra, A. (2001). CXCR4-activated astrocyte glutamate release via TNF: Amplification by microglia triggers neurotoxicity. Nat. Neurosci. 4, 702–710. Bezzi, P., Gundersen, V., Galbete, J. L., Seifert, G., Steinhauser, C., Pilati, E., and Volterra, A. (2004). Astrocytes contain a vesicular compartment that is competent for regulated exocytosis of glutamate. Nat. Neurosci. 7, 613–620. Chen, X., Wang, L., Zhou, Y., Zheng, L. H., and Zhou, Z. (2005). ‘‘Kiss-and-Run’’ Glutamate secretion in cultured and freshly isolated rat hippocampal astrocytes. J. Neurosci. 25, 9236–9243. Christopherson, K. S., Ullian, E. M., Stokes, C. C., Mullowney, C. E., Hell, J. W., Agah, A., Lawler, J., Mosher, D. F., Bornstein, P., and Barres, B. A. (2005). Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120, 421–433. Coco, S., Verderio, C., Trotti, D., Rothstein, J. D., Volterra, A., and Matteoli, M. (1997). Non-synaptic localization of the glutamate transporter EAAC1 in cultured hippocampal neurons. Eur. J. Neurosci. 9, 1902–1910. Crippa, D., Schenk, U., Francolini, M., Rosa, P., Verderio, C., Zonta, M., Pozzan, T., Matteoli, M., and Carmignoto, G. (2006). Synaptobrevin2-expressing vesicles in rat astrocytes: Insights into molecular characterization, dynamics and exocytosis. J. Physiol. 570, 567–582. Dehnes, Y., Chaudhry, F. A., Ullensvang, K., Lehre, K. P., Storm-Mathisen, J., and Danbolt, N. C. (1998). The glutamate transporter EAAT4 in rat cerebellar Purkinje cells: A glutamate-gated chloride channel concentrated near the synapse in parts of the dendritic membrane facing astroglia. J. Neurosci. 18, 3606–3619. Del Villar, K., and Miller, C. A. (2004). Down-regulation of DENN/MADD, a TNF receptor binding protein, correlates with neuronal cell death in Alzheimer’s disease brain and hippocampal neurons. Proc. Natl. Acad. Sci. USA 101, 4210–4215. Domercq, M., Brambilla, L., Pilati, E., Marchaland, J., Volterra, A., and Bezzi, P. (2006). P2Y1 receptor-evoked glutamate exocytosis from astrocytes: Control by tumor necrosis factor- and prostaglandins. J. Biol. Chem. 281, 30684–30696. Duan, S., Anderson, C. M., Keung, E. C., Chen, Y., Chen, Y., and Swanson, R. A. (2003). P2X7 receptor-mediated release of excitatory amino acids from astrocytes. J. Neurosci. 23, 1320–1328. Fellin, T., Pascual, O., Gobbo, S., Pozzan, T., Haydon, P. G., and Carmignoto, G. (2004). Neuronal synchrony mediated by astrocytic glutamate through activation of extrasynaptic NMDA receptors. Neuron 43, 729–743. Fiacco, T. A., and McCarthy, K. D. (2004). Intracellular astrocyte calcium waves in situ increase the frequency of spontaneous AMPA receptor currents in CA1 pyramidal neurons. J. Neurosci. 24, 722–732. Glenner, G. G., and Wong, C. W. (1984). Alzheimer’s disease: Initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem. Biophys. Res. Commun. 120, 885–890. Goddard, D. R., Bunning, R. A., and Woodroofe, M. N. (2001). Astrocyte and endothelial cell expression of ADAM 17 (TACE) in adult human CNS. Glia 34, 267–271. Hartlage-Rubsamen, M., Zeitschel, U., Apelt, J., Gartner, U., Franke, H., Stahl, T., Gunther, A., Schliebs, R., Penkowa, M., Bigl, V., and Rossner, S. (2003). Astrocytic expression of the Alzheimer’s disease -secretase (BACE1) is stimulus-dependent. Glia 41, 169–179. Jeremic, A., Jeftinija, K., Stevanovic, J., Glavaski, A., and Jeftinija, S. (2001). ATP stimulates calciumdependent glutamate release from cultured astrocytes. J. Neurochem. 77, 664–675. Kang, J., Jiang, L., Goldman, S. A., and Nedergaard, M. (1998). Astrocyte-mediated potentiation of inhibitory synaptic transmission. Nat. Neurosci. 1, 683–692.

70

VESCE et al.

Kang, N., Xu, J., Xu, Q., Nedergaard, M., and Kang, J. (2005). Astrocytic glutamate release-induced transient depolarization and epileptiform discharges in hippocampal CA1 pyramidal neurons. J. Neurophysiol. 94, 4121–4130. Karwoski, C. J., Lu, H. K., and Newman, E. A. (1989). Spatial buVering of light-evoked potassium increases by retinal Muller (glial) cells. Science 244, 578–580. Kaul, M., Garden, G. A., and Lipton, S. A. (2001). Pathways to neuronal injury and apoptosis in HIVassociated dementia. Nature 410, 988–994. Kettenmann, H., and Ransom, B. R. (1995). ‘‘Neuroglia.’’ Oxford University Press, Oxford, UK. Kimelberg, H. K., Goderie, S. K., Higman, S., Pang, S., and Waniewski, R. A. (1990). Swelling-induced release of glutamate, aspartate, and taurine from astrocyte cultures. J. Neurosci. 10, 1583–1591. Koistinaho, M., Lin, S., Wu, X., Esterman, M., Koger, D., Hanson, J., Higgs, R., Liu, F., Malkani, S., Bales, K. R., and Paul, S. M. (2004). Apolipoprotein E promotes astrocyte colocalization and degradation of deposited amyloid- peptides. Nat. Med. 10, 719–726. Kreft, M., Stenovec, M., Rupnik, M., Grilc, S., Krzan, M., Potokar, M., Pangrsic, T., Haydon, P. G., and Zorec, R. (2004). Properties of Ca2þ-dependent exocytosis in cultured astrocytes. Glia 46, 437–445. Liu, Q. S., Xu, Q., Arcuino, G., Kang, J., and Nedergaard, M. (2004). Astrocyte-mediated activation of neuronal kainate receptors. Proc. Natl. Acad. Sci. USA 101, 3172–3177. Mauch, D. H., Nagler, K., Schumacher, S., Goritz, C., Muller, E. C., Otto, A., and Pfrieger, F. W. (2001). CNS synaptogenesis promoted by glia-derived cholesterol. Science 294, 1354–1357. Meucci, O., and Miller, R. J. (1996). gp120-induced neurotoxicity in hippocampal pyramidal neuron cultures: Protective action of TGF- 1. J. Neurosci. 16, 4080–4088. Montana, V., Ni, Y., Sunjara, V., Hua, X., and Parpura, V. (2004). Vesicular glutamate transporterdependent glutamate release from astrocytes. J. Neurosci. 24, 2633–2642. Nedergaard, M., Ransom, B., and Goldman, S. A. (2003). New roles for astrocytes: Redefining the functional architecture of the brain. Trends Neurosci. 26, 523–530. Parpura, V., Basarsky, T. A., Liu, F., Jeftinija, K., Jeftinija, S., and Haydon, P. G. (1994). Glutamatemediated astrocyte-neuron signalling. Nature 369, 744–747. Parri, H. R., Gould, T. M., and Crunelli, V. (2001). Spontaneous astrocytic Ca2þ oscillations in situ drive NMDAR-mediated neuronal excitation. Nat. Neurosci. 4, 803–812. Pasti, L., Volterra, A., Pozzan, T., and Carmignoto, G. (1997). Intracellular calcium oscillations in astrocytes: A highly plastic, bidirectional form of communication between neurons and astrocytes in situ. J. Neurosci. 17, 7817–7830. Perea, G., and Araque, A. (2005). Properties of synaptically evoked astrocyte calcium signal reveal synaptic information processing by astrocytes. J. Neurosci. 25, 2192–2203. Perry, V. H., Bell, M. D., Brown, H. C., and Matyszak, M. K. (1995). Inflammation in the nervous system. Curr. Opin. Neurobiol. 5, 636–641. Pfrieger, F. W., and Barres, B. A. (1996). New views on synapse-glia interactions. Curr. Opin. Neurobiol. 6, 615–621. Ralay Ranaivo, H., Craft, J. M., Hu, W., Guo, L., Wing, L. K., Van Eldik, L. J., and Watterson, D. M. (2006). Glia as a therapeutic target: Selective suppression of human amyloid- -induced upregulation of brain proinflammatory cytokine production attenuates neurodegeneration. J. Neurosci. 26, 662–670. Rossi, D. J., Oshima, T., and Attwell, D. (2000). Glutamate release in severe brain ischaemia is mainly by reversed uptake. Nature 403, 316–321. Rossi, D., Brambilla, L., Valori, C. F., Crugnola, A., Giaccone, G., Capobianco, R., Mangieri, M., Kingston, A. E., Bloc, A., Bezzi, P., and Volterra, A. (2005). Defective tumor necrosis factor-dependent control of astrocyte glutamate release in a transgenic mouse model of Alzheimer disease. J. Biol. Chem. 280, 42088–42096.

GLUTAMATE RELEASE FROM ASTROCYTES

71

Sanzgiri, R. P., Araque, A., and Haydon, P. G. (1999). Prostaglandin E2 stimulates glutamate receptordependent astrocyte neuromodulation in cultured hippocampal cells. J. Neurobiol. 41, 221–229. Schipke, C. G., and Kettenmann, H. (2004). Astrocyte responses to neuronal activity. Glia 47, 226–232. Selkoe, D. J. (2002). Alzheimer’s disease is a synaptic failure. Science 298, 789–791. Shi, B., De Girolami, U., He, J., Wang, S., Lorenzo, A., Busciglio, J., and Gabuzda, D. (1996). Apoptosis induced by HIV-1 infection of the central nervous system. J. Clin. Invest. 98, 1979–1990. Stellwagen, D., and Malenka, R. C. (2006). Synaptic scaling mediated by glial TNF-. Nature 440, 1054–1059. Takano, T., Kang, J., Jaiswal, J. K., Simon, S. M., Lin, J. H., Yu, Y., Li, Y., Yang, J., Dienel, G., Zielke, H. R., and Nedergaard, M. (2005). Receptor-mediated glutamate release from volume sensitive channels in astrocytes. Proc. Natl. Acad. Sci. USA 102, 16466–16471. Toggas, S. M., Masliah, E., Rockenstein, E. M., Rall, G. F., Abraham, C. R., and Mucke, L. (1994). Central nervous system damage produced by expression of the HIV-1 coat protein gp120 in transgenic mice. Nature 367, 188–193. Toggas, S. M., Masliah, E., and Mucke, L. (1996). Prevention of HIV-1 gp120-induced neuronal damage in the central nervous system of transgenic mice by the NMDA receptor antagonist memantine. Brain Res. 706, 303–307. Tsacopoulos, M., and Magistretti, P. J. (1996). Metabolic coupling between glia and neurons. J. Neurosci. 16, 877–885. Ullian, E. M., Sapperstein, S. K., Christopherson, K. S., and Barres, B. A. (2001). Control of synapse number by glia. Science 291, 657–661. Ventura, R., and Harris, K. M. (1999). Three-dimensional relationships between hippocampal synapses and astrocytes. J. Neurosci. 19, 6897–6906. Volterra, A., and Meldolesi, J. (2005). Astrocytes, from brain glue to communication elements: The revolution continues. Nat. Rev. Neurosci. 6, 626–640. Warr, O., Takahashi, M., and Attwell, D. (1999). Modulation of extracellular glutamate concentration in rat brain slices by cystine-glutamate exchange. J. Physiol. 514, 783–793. Wyss-Coray, T. (2006). Inflammation in Alzheimer disease: Driving force, bystander or beneficial response? Nat. Med. 12, 1005–1015. Wyss-Coray, T., Loike, J. D., Brionne, T. C., Lu, E., Anankov, R., Yan, F., Silverstein, S. C., and Husemann, J. (2003). Adult mouse astrocytes degrade amyloid- in vitro and in situ. Nat. Med. 9, 453–457. Yan, P., Hu, X., Song, H., Yin, K., Bateman, R. J., Cirrito, J. R., Xiao, Q., Hsu, F. F., Turk, J. W., Xu, J., Hsu, C. Y., Holtzman, D. M., et al. (2006). Matrix metalloproteinase-9 degrades amyloid- fibrils in vitro and compact plaques in situ. J. Biol. Chem. 281, 24566–24574. Ye, Z. C., Wyeth, M. S., Baltan-Tekkok, S., and Ransom, B. R. (2002). Functional hemichannels in astrocytes: A novel mechanism of glutamate release. J. Neurosci. 23, 3588–3596. Zhang, Q., Pangrsic, T., Kreft, M., Krzan, M., Li, N., Sul, J. Y., Halassa, M., Van Bockstaele, E., Zorec, R., and Haydon, P. G. (2004a). Fusion-related release of glutamate from astrocytes. J. Biol. Chem. 279, 12724–12733. Zhang, Q., Fukuda, M., Van Bockstaele, E., Pascual, O., and Haydon, P. G. (2004b). Synaptotagmin IV regulates glial glutamate release. Proc. Natl. Acad. Sci. USA 101, 9441–9446. Zhao, M., Cribbs, D. H., Anderson, A. J., Cummings, B. J., Su, J. H., Wasserman, A. J., and Cotman, C. W. (2003). The induction of the TNF- death domain signaling pathway in Alzheimer’s disease brain. Neurochem. Res. 28, 307–318. Zonta, M., Sebelin, A., Gobbo, S., Fellin, T., Pozzan, T., and Carmignoto, G. (2003). Glutamatemediated cytosolic calcium oscillations regulate a pulsatile prostaglandin release from cultured rat astrocytes. J. Physiol. 553, 407–414.